Argumentsscores, and
strwidth and strheight.
diff --git a/docs/reference/plot.cca-2.png b/docs/reference/plot.cca-2.png
index 4ddc6c14f..0eddaf4ce 100644
Binary files a/docs/reference/plot.cca-2.png and b/docs/reference/plot.cca-2.png differ
diff --git a/docs/reference/plot.cca-3.png b/docs/reference/plot.cca-3.png
new file mode 100644
index 000000000..979d6776c
Binary files /dev/null and b/docs/reference/plot.cca-3.png differ
diff --git a/docs/reference/plot.cca.html b/docs/reference/plot.cca.html
index abe8d00d1..7d11d9cb3 100644
--- a/docs/reference/plot.cca.html
+++ b/docs/reference/plot.cca.html
@@ -69,7 +69,7 @@ Usage
# S3 method for cca
plot(x, choices = c(1, 2), display = c("sp", "wa", "cn"),
scaling = "species", type, xlim, ylim, const,
- correlation = FALSE, hill = FALSE, ...)
+ correlation = FALSE, hill = FALSE, cex = 0.7, ...)
# S3 method for cca
text(x, display = "sites", labels, choices = c(1, 2),
scaling = "species", arrow.mul, head.arrow = 0.05, select, const,
@@ -199,6 +199,9 @@ ArgumentsDetails
-If you want to have a better control of plots, it is best to
- construct the plot text and points commands which
- accept graphical parameters. It is important to remember to use the
- same scaling, correlation and hill arguments
- in all calls. The plot.cca command returns invisibly an
+
If you want to have a better control of plots, it is best to construct
+ the plot text and points commands which accept graphical
+ parameters. It is important to remember to use the same
+ scaling, correlation and hill arguments in all
+ calls. The plot.cca command returns invisibly an
ordiplot result object, and this will have consistent
- scaling for all its elements. The easiest way for full control of
- graphics is to first set up the plot frame using plot with
- type = "n" and all needed scores in display and save
- this result. The points and text commands for
- ordiplot will allow full graphical control (see
- section Examples). Utility function labels returns the default
- labels in the order they are applied in text.
+ scaling for all its elements. It is also possible to use R pipes
+ (|>) which maintain consistent scaling and setting. The easiest
+ way for full control of graphics is to first set up the plot frame
+ using plot with type = "n" and all needed scores in
+ display and save this result or use it as a first command in a
+ pipe. The points and text commands for
+ ordiplot will allow full graphical control (see section
+ Examples). Utility function labels returns the default labels
+ in the order they are applied in text.
Palmer (1993) suggested using linear constraints (“LC scores”)
in ordination diagrams, because these gave better results in
simulations and site scores (“WA scores”) are a step from
@@ -305,8 +310,14 @@
Examplespl <- plot(pca, type="n", scaling="sites", correlation=TRUE)
with(dune.env, points(pl, "site", pch=21, col=1, bg=Management))
text(pl, "sp", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE)
-with(dune.env, legend("bottomleft", levels(Management), pch=21, pt.bg=1:4, bty="n"))
+with(dune.env, legend("bottomleft", levels(Management), pch=21,
+ pt.bg=1:4, bty="n"))
+## Configuration with pipes: make an arrow biplot
+plot(pca, type="n", scaling="sites", correlation=TRUE) |>
+ points("sites", pch=21, col = 1, cex=1.5, bg = dune.env$Management) |>
+ text("species", col = "blue", arrow = TRUE, xpd = TRUE)
+
## Scaling can be numeric or more user-friendly names
## e.g. Hill's scaling for (C)CA
scrs <- scores(mod, scaling = "sites", hill = TRUE)
diff --git a/docs/reference/vegan-package-3.png b/docs/reference/vegan-package-3.png
index 5b6055386..266e97ddc 100644
Binary files a/docs/reference/vegan-package-3.png and b/docs/reference/vegan-package-3.png differ
diff --git a/docs/reference/vegan-package-4.png b/docs/reference/vegan-package-4.png
index 6ed16b20b..c9cb6bd9a 100644
Binary files a/docs/reference/vegan-package-4.png and b/docs/reference/vegan-package-4.png differ
diff --git a/docs/search.json b/docs/search.json
index 8afe3f072..3d0395916 100644
--- a/docs/search.json
+++ b/docs/search.json
@@ -1 +1 @@
-[{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"vegan-faq","dir":"Articles","previous_headings":"","what":"vegan FAQ","title":"","text":"document contains answers frequently asked questions R package vegan. work licensed Creative Commons Attribution 3.0 License. view copy license, visit https://creativecommons.org/licenses//3.0/ send letter Creative Commons, 543 Howard Street, 5th Floor, San Francisco, California, 94105, USA. Copyright © 2008-2016 vegan development team","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is vegan?","title":"","text":"Vegan R package community ecologists. contains popular methods multivariate analysis needed analysing ecological communities, tools diversity analysis, potentially useful functions. Vegan self-contained must run R statistical environment, also depends many R packages. Vegan free software distributed GPL2 license.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is R?","title":"","text":"R system statistical computation graphics. consists language plus run-time environment graphics, debugger, access certain system functions, ability run programs stored script files. R home page https://www.R-project.org/. free software distributed GNU-style copyleft, official part GNU project (“GNU S”).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-obtain-vegan-and-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to obtain vegan and R?","title":"","text":"R latest release version vegan can obtained CRAN. Unstable development version vegan can obtained GitHub. github page gives instructions obtaining installing development versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-r-packages-vegan-depends-on","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What R packages vegan depends on?","title":"","text":"Vegan depends permute package provide advanced flexible permutation routines vegan. permute package developed together vegan GitHub. individual vegan functions depend packages MASS, mgcv, parallel, cluster lattice. Vegan dependence tcltk deprecated removed future releases. base recommended R packages available every R installation. Vegan declares suggested imported packages, can install vegan use functions without packages. Vegan accompanied supporting package vegan3d three-dimensional dynamic plotting. vegan3d package needs tcltk non-standard packages rgl scatterplot3d.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-packages-are-available-for-ecologists","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other packages are available for ecologists?","title":"","text":"CRAN Task Views include entries like Environmetrics, Multivariate Spatial describe several useful packages functions. install R package ctv, can inspect Task Views R session, automatically install sets important packages.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-documentation-is-available-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other documentation is available for vegan?","title":"","text":"Vegan fully documented R package standard help pages. authoritative sources documentation (last resource can use force read source, vegan open source). Vegan package ships documents can read browseVignettes(\"vegan\") command. documents included vegan package Vegan NEWS can accessed via news() command. document (FAQ-vegan). Short introduction basic ordination methods vegan (intro-vegan). Introduction diversity methods vegan (diversity-vegan). Discussion design decisions vegan (decision-vegan). Description variance partition procedures function varpart (partitioning). Web documents outside package include: https://github.com/vegandevs/vegan: development page. https://vegandevs.github.io/vegan/: vegan homepage.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-a-graphical-user-interface-gui-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is there a Graphical User Interface (GUI) for vegan?","title":"","text":"Roeland Kindt made package BiodiversityR provides GUI vegan. package available CRAN. mere GUI vegan, adds new functions complements vegan functions order provide workbench biodiversity analysis. can install BiodiversityR using install.packages(\"BiodiversityR\") graphical package management menu R. GUI works Windows, MacOS X Linux.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-cite-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to cite vegan?","title":"","text":"Use command citation(\"vegan\") R see recommended citation used publications.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-build-vegan-from-sources","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to build vegan from sources?","title":"","text":"general, need build vegan sources, binary builds release versions available CRAN Windows MacOS X. use operating systems, may use source packages. Vegan standard R package, can built like instructed R documentation. Vegan contains source files C FORTRAN, need appropriate compilers (may need work Windows MacOS X).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"are-there-binaries-for-devel-versions","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Are there binaries for devel versions?","title":"","text":"Binaries can available R Universe: see https://github.com/vegandevs/vegan instructions.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-report-a-bug-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to report a bug in vegan?","title":"","text":"think found bug vegan, report vegan maintainers developers. preferred forum report bugs GitHub. bug report detailed bug can replicated corrected. Preferably, send example causes bug. needs data set available R, send minimal data set well. also paste output error message message. also specify version vegan used. Bug reports welcome: way make vegan non-buggy. Please note shall send bug reports R mailing lists, since vegan standard R package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-a-bug-or-a-feature","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is it a bug or a feature?","title":"","text":"necessarily bug function gives different results expect: may deliberate design decision. may useful check documentation function see intended behaviour. may also happen function argument switch behaviour match expectation. instance, function vegdist always calculates quantitative indices (possible). expect calculate binary index, use argument binary = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-contribute-to-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Can I contribute to vegan?","title":"","text":"Vegan dependent user contribution. feedback welcome. problems vegan, may simple incomplete documentation, shall best improve documents. Feature requests also welcome, necessarily fulfilled. new feature added easy looks useful, submit code. can write code , best forum contribute vegan GitHub.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-only-numeric-and-positive-data-but-vegan-still-complains","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have only numeric and positive data but vegan still complains","title":"","text":"wrong! Computers painfully pedantic, find non-numeric negative data entries, really . Check data! common reasons non-numeric data row names read non-numeric variable instead used row names (check argument row.names reading data), column names interpreted data (check argument header = TRUE reading data). Another common reason empty cells input data, interpreted missing values.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-analyse-binary-or-cover-class-data","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I analyse binary or cover class data?","title":"","text":"Yes. vegan methods can handle binary data cover abundance data. statistical tests based permutation, make distributional assumptions. methods (mainly diversity analysis) need count data. methods check input data integers, may fooled cover class data.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-dissimilarities-in-vegan-differ-from-other-sources","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why dissimilarities in vegan differ from other sources?","title":"","text":"commonly reason software use presence–absence data whereas vegan used quantitative data. Usually vegan indices quantitative, can use argument binary = TRUE make presence–absence. However, index name cases, although different names usually occur literature. instance, Jaccard index actually refers binary index, vegan uses name \"jaccard\" quantitative index, . Another reason may indices indeed defined differently, people use names different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-nmds-stress-is-sometimes-0-1-and-sometimes-10","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why NMDS stress is sometimes 0.1 and sometimes 10?","title":"","text":"Stress proportional measure badness fit. proportions can expressed either parts one percents. Function isoMDS (MASS package) uses percents, function monoMDS (vegan package) uses proportions, therefore stress 100 times higher isoMDS. results goodness function also depend definition stress, goodness 100 times higher isoMDS monoMDS. conventions equally correct.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-get-zero-stress-but-no-repeated-solutions-in-metamds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I get zero stress but no repeated solutions in metaMDS","title":"","text":"first (try 0) run metaMDS starts metric scaling solution usually good, sofware return solution. However, metaMDS tries see standard solution can repeated, improved improved solution still repeated. cases, return best solution found, burning need anything get message tha solution repeated. keen know solution really global optimum, may follow instructions metaMDS help section “Results Repeated” try . common reason observations NMDS. n observations (points) k dimensions need estimate n*k parameters (ordination scores) using n*(n-1)/2 dissimilarities. k dimensions must n > 2*k + 1, two dimensions least six points. degenerate situations may need even larger number points. lower number points, can find undefined number perfect (stress zero) different solutions. Conventional wisdom due Kruskal n > 4*k + 1 points k dimensions. typical symptom insufficient data (nearly) zero stress two convergent solutions. cases reduce number dimensions (k) small data sets use NMDS, rely metric methods. seems local hybrid scaling monoMDS similar lower limits practice (although theoretically differ). However, higher number dimensions can used metric scaling, monoMDS principal coordinates analysis (cmdscale stats, wcmdscale vegan).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"zero-dissimilarities-in-isomds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Zero dissimilarities in isoMDS","title":"","text":"Function metaMDS uses function monoMDS default method NMDS, function can handle zero dissimilarities. Alternative function isoMDS handle zero dissimilarities. want use isoMDS, can use argument zerodist = \"add\" metaMDS handle zero dissimilarities. argument, zero dissimilarities replaced small positive value, can handled isoMDS. kluge, people like . principal solution remove duplicate sites using R command unique. However, standardizations dissimilarity indices, originally non-unique sites can zero dissimilarity, resort kluge (work harder data). Usually better use monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-heard-that-you-cannot-fit-environmental-vectors-or-surfaces-to-nmds-results-which-only-have-rank-order-scores","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have heard that you cannot fit environmental vectors or surfaces to NMDS results which only have rank-order scores","title":"","text":"Claims like indeed large Internet, based grave misunderstanding plainly wrong. NMDS ordination results strictly metric, vegan metaMDS monoMDS even strictly Euclidean. method called “non-metric” Euclidean distances ordination space non-metric rank-order relationship community dissimilarities. can inspect non-linear step curve using function stressplot vegan. ordination scores strictly Euclidean, correct use vegan functions envfit ordisurf NMDS results.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"where-can-i-find-numerical-scores-of-ordination-axes","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Where can I find numerical scores of ordination axes?","title":"","text":"Normally can use function scores extract ordination scores ordination method. scores function can also find ordination scores many non-vegan functions prcomp princomp ade4 functions. cases ordination result object stores raw scores, axes also scaled appropriate access scores. instance, cca rda ordination object -called normalized scores, scaled ordination plots use accessed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-the-rda-results-are-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How the RDA results are scaled?","title":"","text":"scaling RDA results indeed differ software packages. scaling RDA complicated issue explained FAQ, explained separate pdf document “Design decision implementation details vegan” can read command browseVignettes(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-cannot-print-and-plot-rda-results-properly","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I cannot print and plot RDA results properly","title":"","text":"RDA ordination results weird format plot properly, probably name clash klaR package also function rda, klaR print, plot predict functions used vegan RDA results. can choose rda functions using vegan::rda() klaR::rda(): get obscure error messages use wrong function. general, vegan able work normally vegan loaded klaR, klaR loaded later, functions take precedence vegan. Sometimes vegan namespace loaded automatically restoring previously stored workspace start-, klaR methods always take precedence vegan. check loaded packages. klaR may also loaded indirectly via packages (reported cases often loaded via agricolae package). Vegan klaR function name (rda), may possible use packages simultaneously, safest choice unload one packages possible. See discussion vegan github issues.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"ordination-fails-with-error-in-la-svd","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Ordination fails with “Error in La.svd”","title":"","text":"Constrained ordination (cca, rda, dbrda, capscale) sometimes fail error message Error La.svd(x, nu, nv): error code 1 Lapack routine 'dgesdd'. seems basic problem svd function LAPACK used numerical analysis R. LAPACK external library beyond control package developers R core team problems may unsolvable. Reducing range constraints (environmental variables) helps sometimes. instance, multiplying constraints constant < 1. rescaling influence numerical results constrained ordination, can complicate analyses values constraints needed, scaling must applied . reports problems getting rare may problem fixed R LAPACK.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"variance-explained-by-ordination-axes-","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Variance explained by ordination axes.","title":"","text":"general, vegan directly give statistics “variance explained” ordination axes constrained axes. design decision: think information normally useless often misleading. community ordination, goal typically explain variance, find “gradients” main trends data. “total variation” often meaningless, proportions meaningless values also meaningless. Often better solution explains smaller part “total variation”. instance, unstandardized principal components analysis variance generated small number abundant species, easy “explain” data really multivariate. standardize data, species equally important. first axes explains much less “total variation”, now explain species equally, results typically much useful whole community. Correspondence analysis uses another measure variation (variance), typically explains “smaller proportion” principal components better result. Detrended correspondence analysis nonmetric multidimensional scaling even try “explain” variation, use criteria. methods incommensurable, impossible compare methods using “explanation variation”. still want get “explanation variation” (deranged editor requests ), possible get information methods: Eigenvector methods: Functions rda, cca, dbrda capscale give variation conditional (partialled), constrained (canonical) residual components. Function eigenvals extracts eigenvalues, summary(eigenvals(ord)) reports proportions explained result object ord, also works decorana wcmdscale. Function RsquareAdj gives R-squared adjusted R-squared (available) constrained components. Function goodness gives statistics individual species sites. addition, special function varpart unbiased partitioning variance four separate components redundancy analysis. Nonmetric multidimensional scaling. NMDS method nonlinear mapping, concept variation explained make sense. However, 1 - stress^2 transforms nonlinear stress quantity analogous squared correlation coefficient. Function stressplot displays nonlinear fit gives statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-have-random-effects-in-constrained-ordination-or-in-adonis","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I have random effects in constrained ordination or in adonis?","title":"","text":". Strictly speaking, impossible. However, can define models respond similar goals random effects models, although strictly speaking use fixed effects. Constrained ordination functions cca, rda dbrda can Condition() terms formula. Condition() define partial terms fitted constraints can used remove effects background variables, contribution decomposing inertia (variance) reported separately. partial terms often regarded similar random effects, still fitted way terms strictly speaking fixed terms. Function adonis2 can evaluate terms sequentially. model right-hand-side ~ + B effects evaluated first, effects B removing effects . Sequential tests also available anova function constrained ordination results setting argument = \"term\". way, first terms can serve similar role random effects, although fitted way terms, strictly speaking fixed terms. permutation tests vegan based permute package allows constructing various restricted permutation schemes. instance, can set levels plots blocks factor regarded random term. major reason real random effects models impossible vegan functions tests based permutation data. data given, fixed, therefore permutation tests basically tests fixed terms fixed data. Random effect terms require permutations data random component instead given, fixed data, tests available vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-possible-to-have-passive-points-in-ordination","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Is it possible to have passive points in ordination?","title":"","text":"Vegan concept passive points, point little influence ordination results. However, can add points eigenvector methods using predict functions newdata. can first perform ordination without species sites, can find scores points using complete data newdata. predict functions available basic eigenvector methods vegan (cca, rda, decorana, --date list, use command methods(\"predict\")).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"class-variables-and-dummies","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Class variables and dummies","title":"","text":"define class variable R factor, vegan automatically handle . R (vegan) knows unordered ordered factors. Unordered factors internally coded dummy variables, one redundant level removed aliased. default contrasts, removed level first one. Ordered factors expressed polynomial contrasts. contrasts explained standard R documentation.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-are-environmental-arrows-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How are environmental arrows scaled?","title":"","text":"printed output envfit gives direction cosines coordinates unit length arrows. plotting, scaled correlation (square roots column r2). can see scaled lengths envfit arrows using command scores. scaled environmental vectors envfit arrows continuous environmental variables constrained ordination (cca, rda, dbrda) adjusted fill current graph. lengths arrows fixed meaning respect points (species, sites), can compared , therefore relative lengths important. want change scaling arrows, can use text (plotting arrows text) points (plotting arrows) functions constrained ordination. functions argument arrow.mul sets multiplier. plot function envfit also arrow.mul argument set arrow multiplier. save invisible result constrained ordination plot command, can see value currently used arrow.mul saved attribute biplot scores. Function ordiArrowMul used find scaling current plot. can use function see arrows scaled:","code":"sol <- cca(varespec) ef <- envfit(sol ~ ., varechem) plot(sol) ordiArrowMul(scores(ef, display=\"vectors\"))"},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-want-to-use-helmert-or-sum-contrasts","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I want to use Helmert or sum contrasts","title":"","text":"vegan uses standard R utilities defining contrasts. default standard installations use treatment contrasts, can change behaviour globally setting options locally using keyword contrasts. Please check R help pages user manuals details.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-are-aliased-variables-and-how-to-see-them","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"What are aliased variables and how to see them?","title":"","text":"Aliased variable information can expressed help variables. variables automatically removed constrained ordination vegan. aliased variables can redundant levels factors whole variables. Vegan function alias gives defining equations aliased variables. want see names aliased variables levels solution sol, use alias(sol, names.=TRUE).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"plotting-aliased-variables","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Plotting aliased variables","title":"","text":"can fit vectors class centroids aliased variables using envfit function. envfit function uses weighted fitting, fitted vectors identical vectors correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"restricted-permutations-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Restricted permutations in vegan","title":"","text":"Vegan uses permute package permutation tests. permute package allow restricted permutation designs time series, line transects, spatial grids blocking factors. construction restricted permutation schemes explained manual page permutations vegan documentation permute package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-use-different-plotting-symbols-in-ordination-graphics","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to use different plotting symbols in ordination graphics?","title":"","text":"default ordination plot function intended fast plotting configurable. use different plotting symbols, first create empty ordination plot plot(..., type=\"n\"), add points text created empty frame (... means arguments want give plot command). points text commands fully configurable, allow different plotting symbols characters.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-avoid-cluttered-ordination-graphs","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to avoid cluttered ordination graphs?","title":"","text":"really high number species sites, graphs often congested many labels overwritten. may impossible complete readable graphics data sets. give brief overview tricks can use. Gavin Simpson’s blog bottom heap series articles “decluttering ordination plots” detailed discussion examples. Use points, possibly different types need see labels. may need first create empty plot using plot(..., type=\"n\"), satisfied default graph. (... means arguments want give plot command.) Use points add labels desired points using interactive identify command need see labels. Add labels using function ordilabel uses non-transparent background text. labels still shadow , uppermost labels readable. Argument priority help displaying interesting labels (see Decluttering blog, part 1). Use orditorp function uses labels can added graph without overwriting labels, points otherwise, need see labels. must first create empty plot using plot(..., type=\"n\"), add labels points orditorp (see Decluttering blog). Use ordipointlabel uses points text labels points, tries optimize location text minimize overlap (see Decluttering blog). Ordination text points functions argument select can used full control selecting items plotted text points. Use interactive orditkplot function (vegan3d package) lets drag labels points better positions need see labels. one set points can used (see Decluttering blog). plot functions allow zoom part graph using xlim ylim arguments reduce clutter congested areas.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-flip-an-axis-in-ordination-diagram","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I flip an axis in ordination diagram?","title":"","text":"Use xlim ylim flipped limits. model mod <- cca(dune) can flip first axis plot(mod, xlim = c(3, -2)).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-zoom-into-an-ordination-plot","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I zoom into an ordination plot?","title":"","text":"can use xlim ylim arguments plot ordiplot zoom ordination diagrams. Normally must set xlim ylim ordination plots keep equal aspect ratio axes, fill graph longer axis fit. Dynamic zooming can done function orditkplot CRAN package vegan3d. can directly save edited orditkplot graph various graphic formats, can export graph object back R session use plot display results.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-twinspan","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Is there TWINSPAN?","title":"","text":"TWINSPAN R available github.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-restricted-permutation-does-not-influence-adonis-results","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Why restricted permutation does not influence adonis results?","title":"","text":"permutation scheme influences permutation distribution statistics probably significance levels, influence calculation statistics.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-is-deviance-calculated","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"How is deviance calculated?","title":"","text":"vegan functions, radfit use base R facility family maximum likelihood estimation. allows use several alternative error distributions, among \"poisson\" \"gaussian\". R family also defines deviance. can see equations deviance commands like poisson()$dev gaussian()$dev. general, deviance 2 times log.likelihood shifted models exact fit zero deviance.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Jari Oksanen. Author, maintainer. Gavin L. Simpson. Author. F. Guillaume Blanchet. Author. Roeland Kindt. Author. Pierre Legendre. Author. Peter R. Minchin. Author. R.B. O'Hara. Author. Peter Solymos. Author. M. Henry H. Stevens. Author. Eduard Szoecs. Author. Helene Wagner. Author. Matt Barbour. Author. Michael Bedward. Author. Ben Bolker. Author. Daniel Borcard. Author. Gustavo Carvalho. Author. Michael Chirico. Author. Miquel De Caceres. Author. Sebastien Durand. Author. Heloisa Beatriz Antoniazi Evangelista. Author. Rich FitzJohn. Author. Michael Friendly. Author. Brendan Furneaux. Author. Geoffrey Hannigan. Author. Mark O. Hill. Author. Leo Lahti. Author. Dan McGlinn. Author. Marie-Helene Ouellette. Author. Eduardo Ribeiro Cunha. Author. Tyler Smith. Author. Adrian Stier. Author. Cajo J.F. Ter Braak. Author. James Weedon. Author.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O'Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier , Ter Braak C, Weedon J (2024). vegan: Community Ecology Package. R package version 2.6-7, https://github.com/vegandevs/vegan, https://vegandevs.github.io/vegan/.","code":"@Manual{, title = {vegan: Community Ecology Package}, author = {Jari Oksanen and Gavin L. Simpson and F. Guillaume Blanchet and Roeland Kindt and Pierre Legendre and Peter R. Minchin and R.B. O'Hara and Peter Solymos and M. Henry H. Stevens and Eduard Szoecs and Helene Wagner and Matt Barbour and Michael Bedward and Ben Bolker and Daniel Borcard and Gustavo Carvalho and Michael Chirico and Miquel {De Caceres} and Sebastien Durand and Heloisa Beatriz Antoniazi Evangelista and Rich FitzJohn and Michael Friendly and Brendan Furneaux and Geoffrey Hannigan and Mark O. Hill and Leo Lahti and Dan McGlinn and Marie-Helene Ouellette and Eduardo {Ribeiro Cunha} and Tyler Smith and Adrian Stier and Cajo J.F. {Ter Braak} and James Weedon}, year = {2024}, note = {R package version 2.6-7, https://github.com/vegandevs/vegan}, url = {https://vegandevs.github.io/vegan/}, }"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"vegan-an-r-package-for-community-ecologists","dir":"","previous_headings":"","what":"vegan: an R package for community ecologists","title":"vegan: an R package for community ecologists","text":"Ordination methods, diversity analysis functions community vegetation ecologists. Website development version vegan package. Vignettes available R-universe Introduction ordination vegan Partition Variation Diversity analysis vegan Design decisions implementation","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"vegan: an R package for community ecologists","text":"install development version vegan can use usual git R CMD build -> R CMD INSTALL dance cloned repo (downloaded sources). ’ll need able install packages source work; don’t relevant developer tools, won’t able install vegan way.","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"using-remotes","dir":"","previous_headings":"","what":"Using remotes","title":"vegan: an R package for community ecologists","text":"developer tools installed don’t want hassle keeping local source code tree --date, use remotes package:","code":"install.packages(\"remotes\") remotes::install_github(\"vegandevs/vegan\")"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installing-binaries-from-r-universe","dir":"","previous_headings":"","what":"Installing binaries from R Universe","title":"vegan: an R package for community ecologists","text":"just want install binary version packages, just CRAN, can install R Universe repository. Run following R session:","code":"install.packages('vegan', repos = c('https://vegandevs.r-universe.dev','https://cloud.r-project.org'))"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":null,"dir":"Reference","previous_headings":"","what":"Barro Colorado Island Tree Counts — BCI","title":"Barro Colorado Island Tree Counts — BCI","text":"Tree counts 1-hectare plots Barro Colorado Island associated site information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI) data(BCI.env)"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Barro Colorado Island Tree Counts — BCI","text":"data frame 50 plots (rows) 1 hectare counts trees plot total 225 species (columns). Full Latin names used tree species. names updated http://www.theplantlist.org Kress et al. (2009) allows matching 207 species doi:10.5061/dryad.63q27 (Zanne et al., 2014). original species names available attribute original.names BCI. See Examples changed names. BCI.env, data frame 50 plots (rows) nine site variables derived Pyke et al. (2001) Harms et al. (2001): UTM.EW: UTM coordinates (zone 17N) East-West. UTM.NS: UTM coordinates (zone 17N) North-South. Precipitation: Precipitation mm per year. Elevation: Elevation m sea level. Age.cat: Forest age category. Geology: Underlying geological formation. Habitat: Dominant habitat type based map habitat types 25 grid cells plot (Harms et al. 2001, excluding streamside habitat). habitat types Young forests (ca. 100 years), old forests > 7 degree slopes (OldSlope), old forests 152 m elevation (OldLow) higher elevation (OldHigh) Swamp forests. River: \"Yes\" streamside habitat plot. EnvHet: Environmental Heterogeneity assessed Simpson diversity frequencies Habitat types 25 grid cells plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Barro Colorado Island Tree Counts — BCI","text":"Data give numbers trees least 10 cm diameter breast height (DBH) one hectare quadrat 1982 BCI plot. Within plot, individuals tallied recorded table. full survey included smaller trees DBH 1 cm larger, BCI dataset subset larger trees compiled Condit et al. (2002). full data thinner trees densities 4000 stems per hectare, ten times stems data. dataset BCI provided (2003) illustrate analysis methods vegan. scientific research ecological issues strongly recommend access complete modern data (Condit et al. 2019) updated taxonomy (Condit et al. 2020). data frame contains Barro Colorado Island subset full data table Condit et al. (2002). quadrats located regular grid. See BCI.env coordinates. full description site information BCI.env given Pyke et al. (2001) Harms et al. (2001). N.B. Pyke et al. (2001) Harms et al. (2001) give conflicting information forest age categories elevation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Barro Colorado Island Tree Counts — BCI","text":"https://www.science.org/doi/10.1126/science.1066854 community data References environmental data. updated complete data (incl. thinner trees 1 cm), see Condit et al. (2019).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Barro Colorado Island Tree Counts — BCI","text":"Condit, R, Pitman, N, Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., Nuñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002). Beta-diversity tropical forest trees. Science 295, 666--669. Condit R., Pérez, R., Aguilar, S., Lao, S., Foster, R. & Hubbell, S. (2019). Complete data Barro Colorado 50-ha plot: 423617 trees, 35 years [Dataset]. Dryad. doi:10.15146/5xcp-0d46 Condit, R., Aguilar, S., Lao, S., Foster, R., Hubbell, S. (2020). BCI 50-ha Plot Taxonomy [Dataset]. Dryad. doi:10.15146/R3FH61 Harms K.E., Condit R., Hubbell S.P. & Foster R.B. (2001) Habitat associations trees shrubs 50-ha neotropical forest plot. J. Ecol. 89, 947--959. Kress W.J., Erickson D.L, Jones F.., Swenson N.G, Perez R., Sanjur O. & Bermingham E. (2009) Plant DNA barcodes community phylogeny tropical forest dynamics plot Panama. PNAS 106, 18621--18626. Pyke, C. R., Condit, R., Aguilar, S., & Lao, S. (2001). Floristic composition across climatic gradient neotropical lowland forest. Journal Vegetation Science 12, 553--566. doi:10.2307/3237007 Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506, 89--92. doi:10.1038/nature12872 (published online Dec 22, 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI, BCI.env) head(BCI.env) #> UTM.EW UTM.NS Precipitation Elevation Age.cat Geology Habitat Stream EnvHet #> 1 625754 1011569 2530 120 c3 Tb OldSlope Yes 0.6272 #> 2 625754 1011669 2530 120 c3 Tb OldLow Yes 0.3936 #> 3 625754 1011769 2530 120 c3 Tb OldLow No 0.0000 #> 4 625754 1011869 2530 120 c3 Tb OldLow No 0.0000 #> 5 625754 1011969 2530 120 c3 Tb OldSlope No 0.4608 #> 6 625854 1011569 2530 120 c3 Tb OldLow No 0.0768 ## see changed species names oldnames <- attr(BCI, \"original.names\") taxa <- cbind(\"Old Names\" = oldnames, \"Current Names\" = names(BCI)) noquote(taxa[taxa[,1] != taxa[,2], ]) #> Old Names Current Names #> [1,] Abarema.macradenium Abarema.macradenia #> [2,] Acacia.melanoceras Vachellia.melanoceras #> [3,] Apeiba.aspera Apeiba.glabra #> [4,] Aspidosperma.cruenta Aspidosperma.desmanthum #> [5,] Cassipourea.elliptica Cassipourea.guianensis #> [6,] Cespedezia.macrophylla Cespedesia.spathulata #> [7,] Chlorophora.tinctoria Maclura.tinctoria #> [8,] Coccoloba.manzanillensis Coccoloba.manzinellensis #> [9,] Coussarea.curvigemmia Coussarea.curvigemma #> [10,] Cupania.sylvatica Cupania.seemannii #> [11,] Dipteryx.panamensis Dipteryx.oleifera #> [12,] Eugenia.coloradensis Eugenia.florida #> [13,] Eugenia.oerstedeana Eugenia.oerstediana #> [14,] Guapira.standleyana Guapira.myrtiflora #> [15,] Hyeronima.alchorneoides Hieronyma.alchorneoides #> [16,] Inga.marginata Inga.semialata #> [17,] Lonchocarpus.latifolius Lonchocarpus.heptaphyllus #> [18,] Maquira.costaricana Maquira.guianensis.costaricana #> [19,] Phoebe.cinnamomifolia Cinnamomum.triplinerve #> [20,] Swartzia.simplex.var.ochnacea Swartzia.simplex.continentalis #> [21,] Tabebuia.guayacan Handroanthus.guayacan"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":null,"dir":"Reference","previous_headings":"","what":"Canonical Correlation Analysis — CCorA","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis, following Brian McArdle's unpublished graduate course notes, plus improvements allow calculations case sparse collinear matrices, permutation test Pillai's trace statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Canonical Correlation Analysis — CCorA","text":"","code":"CCorA(Y, X, stand.Y=FALSE, stand.X=FALSE, permutations = 0, ...) # S3 method for CCorA biplot(x, plot.type=\"ov\", xlabs, plot.axes = 1:2, int=0.5, col.Y=\"red\", col.X=\"blue\", cex=c(0.7,0.9), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Canonical Correlation Analysis — CCorA","text":"Y Left matrix (object class: matrix data.frame). X Right matrix (object class: matrix data.frame). stand.Y Logical; Y standardized? stand.X Logical; X standardized? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x CCoaR result object. plot.type character string indicating following plots produced: \"objects\", \"variables\", \"ov\" (separate graphs objects variables), \"biplots\". unambiguous subset containing first letters names can used instead full names. xlabs Row labels. default use row names, NULL uses row numbers instead, NA suppresses plotting row names completely. plot.axes vector 2 values containing order numbers canonical axes plotted. Default: first two axes. int Radius inner circles plotted visual references plots variables. Default: int=0.5. int=0, inner circle plotted. col.Y Color used objects variables first data table (Y) plots. biplots, objects black. col.X Color used objects variables second data table (X) plots. cex vector 2 values containing size reduction factors object variable names, respectively, plots. Default values: cex=c(0.7,0.9). ... arguments passed functions. function biplot.CCorA passes graphical arguments biplot biplot.default. CCorA currently ignores extra arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis (Hotelling 1936) seeks linear combinations variables Y maximally correlated linear combinations variables X. analysis estimates relationships displays graphs. Pillai's trace statistic computed tested parametrically (F-test); permutation test also available. Algorithmic note -- blunt approach read two matrices, compute covariance matrices, matrix S12 %*% inv(S22) %*% t(S12) %*% inv(S11). trace Pillai's trace statistic. approach may fail, however, heavy multicollinearity sparse data matrices. safe approach replace data matrices PCA object scores. function can produce different types plots depending option chosen: \"objects\" produces two plots objects, one space Y, second space X; \"variables\" produces two plots variables, one variables Y space Y, second variables X space X; \"ov\" produces four plots, two objects two variables; \"biplots\" produces two biplots, one first matrix (Y) one second matrix (X) solutions. biplots, function passes arguments biplot.default; consult help page configuring biplots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Canonical Correlation Analysis — CCorA","text":"Function CCorA returns list containing following elements: Pillai Pillai's trace statistic = sum canonical eigenvalues. Eigenvalues Canonical eigenvalues. squares canonical correlations. CanCorr Canonical correlations. Mat.ranks Ranks matrices Y X. RDA.Rsquares Bimultivariate redundancy coefficients (R-squares) RDAs Y|X X|Y. RDA.adj.Rsq RDA.Rsquares adjusted n number explanatory variables. nperm Number permutations. p.Pillai Parametric probability value associated Pillai's trace. p.perm Permutational probability associated Pillai's trace. Cy Object scores Y biplot. Cx Object scores X biplot. corr.Y.Cy Scores Y variables Y biplot, computed cor(Y,Cy). corr.X.Cx Scores X variables X biplot, computed cor(X,Cx). corr.Y.Cx cor(Y,Cy) available plotting variables Y space X manually. corr.X.Cy cor(X,Cx) available plotting variables X space Y manually. control list control values permutations returned function . call Call CCorA function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Canonical Correlation Analysis — CCorA","text":"Hotelling, H. 1936. Relations two sets variates. Biometrika 28: 321-377. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Canonical Correlation Analysis — CCorA","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal. Implemented vegan help Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Canonical Correlation Analysis — CCorA","text":"","code":"# Example using two mite groups. The mite data are available in vegan data(mite) # Two mite species associations (Legendre 2005, Fig. 4) group.1 <- c(1,2,4:8,10:15,17,19:22,24,26:30) group.2 <- c(3,9,16,18,23,25,31:35) # Separate Hellinger transformations of the two groups of species mite.hel.1 <- decostand(mite[,group.1], \"hel\") mite.hel.2 <- decostand(mite[,group.2], \"hel\") rownames(mite.hel.1) = paste(\"S\",1:nrow(mite),sep=\"\") rownames(mite.hel.2) = paste(\"S\",1:nrow(mite),sep=\"\") out <- CCorA(mite.hel.1, mite.hel.2) out #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mite.hel.1, X = mite.hel.2) #> #> Y X #> Matrix Ranks 24 11 #> #> Pillai's trace: 4.573009 #> #> Significance of Pillai's trace: #> from F-distribution: 0.0032737 #> CanAxis1 CanAxis2 CanAxis3 CanAxis4 CanAxis5 CanAxis6 #> Canonical Correlations 0.92810 0.82431 0.81209 0.74981 0.70795 0.65950 #> CanAxis7 CanAxis8 CanAxis9 CanAxis10 CanAxis11 #> Canonical Correlations 0.50189 0.48179 0.41089 0.37823 0.28 #> #> Y | X X | Y #> RDA R squares 0.33224 0.5376 #> adj. RDA R squares 0.20560 0.2910 #> biplot(out, \"ob\") # Two plots of objects biplot(out, \"v\", cex=c(0.7,0.6)) # Two plots of variables biplot(out, \"ov\", cex=c(0.7,0.6)) # Four plots (2 for objects, 2 for variables) biplot(out, \"b\", cex=c(0.7,0.6)) # Two biplots biplot(out, xlabs = NA, plot.axes = c(3,5)) # Plot axes 3, 5. No object names biplot(out, plot.type=\"biplots\", xlabs = NULL) # Replace object names by numbers # Example using random numbers. No significant relationship is expected mat1 <- matrix(rnorm(60),20,3) mat2 <- matrix(rnorm(100),20,5) out2 = CCorA(mat1, mat2, permutations=99) out2 #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mat1, X = mat2, permutations = 99) #> #> Y X #> Matrix Ranks 3 5 #> #> Pillai's trace: 0.480458 #> #> Significance of Pillai's trace: #> from F-distribution: 0.90606 #> based on permutations: 0.94 #> Permutation: free #> Number of permutations: 99 #> #> CanAxis1 CanAxis2 CanAxis3 #> Canonical Correlations 0.64421 0.23458 0.1021 #> #> Y | X X | Y #> RDA R squares 0.214302 0.0839 #> adj. RDA R squares -0.066305 -0.0879 #> biplot(out2, \"b\")"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Add New Points to NMDS ordination — MDSaddpoints","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Add new points existing metaMDS monoMDS ordination.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"MDSaddpoints(nmds, dis, neighbours = 5, maxit = 200) dist2xy(dist, pick, type = c(\"xy\", \"xx\"), invert = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"nmds Result object metaMDS monoMDS. configuration points fixed, new points added. dis Rectangular non-symmetric dissimilarity matrix among new points (rows) old fixed points (columns). matrix can extracted complete dissimilarities old new points dist2xy, calculated designdist2. neighbours Number nearest points used get starting locations new points. maxit Maximum number iterations. dist Input dissimilarities. pick Indices (integers) selected observations logical vector TRUE picked items. output original order reordered argument. type \"xy\" returns rectangular data picked picked observations, \"xx\" subset symmetric dissimilarities. invert Invert pick: drop elements listed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function return list class \"nmds\" (objects type vegan) following elements points Coordinates added new points seeds Starting coordinates new points. deltastress Change stress added points. iters Number iterations. cause Cause termination iterations. Integer convergence criteria monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function provides interface monoMDS Fortran code add new points existing ordination regarded fixed. function similar role predict functions newdata Euclidean ordination (e.g. predict.cca). Input data must rectangular matrix distances among new added points (rows) fixed old points (columns). matrices can extracted complete dissimilarities helper function dist2xy. Function designdist2 can directly calculate rectangular dissimilarity matrices sampling units (rows) two matries. addition, analogue distance function can calculate dissimilarities among two matrices, including functions specified designdist2. Great care needed preparing dissimilarities input. dissimilarity index must exactly fixed ordination, columns must match old fixed points, rows added new points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"## Cross-validation: remove a point when performing NMDS and add as ## a new points data(dune) d <- vegdist(dune) ## remove point 3 from ordination mod3 <- metaMDS(dist2xy(d, 3, \"xx\", invert = TRUE), trace=0) ## add point 3 to the result MDSaddpoints(mod3, dist2xy(d, 3)) #> $points #> MDS1 MDS2 #> 3 -0.09726881 -0.4560917 #> #> $seed #> [,1] [,2] #> [1,] 0.01222493 -0.4353716 #> #> $deltastress #> [1] 0.001811377 #> #> $iters #> [1] 18 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\" ## Use designdist2 d15 <- designdist(dune[1:15,]) m15 <- metaMDS(d15, trace=0) MDSaddpoints(m15, designdist2(dune[1:15,], dune[16:20,])) #> $points #> MDS1 MDS2 #> 16 0.4296366 -0.3068902 #> 17 -0.4122463 0.4118309 #> 18 -0.1135994 0.3120708 #> 19 0.1421922 0.4869128 #> 20 0.5892538 0.1751219 #> #> $seed #> [,1] [,2] #> [1,] 0.2832566 0.019571976 #> [2,] -0.1920994 0.137270818 #> [3,] -0.1529998 0.094374124 #> [4,] -0.0244597 0.137988191 #> [5,] 0.3212258 0.009941098 #> #> $deltastress #> [1] 0.03394768 #> #> $iters #> [1] 33 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":null,"dir":"Reference","previous_headings":"","what":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function rotates multidimensional scaling result first dimension parallel external (environmental variable). function can handle results metaMDS monoMDS functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"MDSrotate(object, vec, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"object result object metaMDS monoMDS. vec environmental variable matrix variables. number variables must lower number dimensions, solution rotated variables order appear matrix. Alternatively vec can factor, solution rotated optimal separation factor levels using lda. na.rm Remove missing values continuous variable vec. ... arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"orientation rotation undefined multidimensional scaling. Functions metaMDS metaMDS can rotate solutions principal components dispersion points highest first dimension. Sometimes different rotation intuitive, MDSrotate allows rotation result first axis parallel given external variable two first variables completely two-dimensional plane etc. several external variables supplied, applied order matrix. First axis rotated first supplied variable, second axis second variable. variables usually correlated, second variable usually aligned second axis, uncorrelated later dimensions. must least one free dimension: number external variables must lower number dimensions, used environmental variables uncorrelated free dimension. Alternatively method can rotate discriminate levels factor using linear discriminant analysis (lda). hardly meaningful two-dimensional solutions, since rotations two dimensions separation cluster levels. However, function can useful finding two-dimensional projection clusters two dimensions. last dimension always show residual variation, \\(k\\) dimensions, \\(k-1\\) discrimination vectors used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function returns original ordination result, rotated scores (site species available), pc attribute scores set FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Rotation factor variable experimental feature may removed. discriminant analysis weights dimensions discriminating power, MDSrotate performs rigid rotation. Therefore solution may optimal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"data(varespec) data(varechem) mod <- monoMDS(vegdist(varespec)) mod <- with(varechem, MDSrotate(mod, pH)) plot(mod) ef <- envfit(mod ~ pH, varechem, permutations = 0) plot(ef) ordisurf(mod ~ pH, varechem, knots = 1, add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ poly(x1, 1) + poly(x2, 1) #> Total model degrees of freedom 3 #> #> REML score: -2.736059"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":null,"dir":"Reference","previous_headings":"","what":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds & Shaw test concerns location highest (hump) lowest (pit) value quadratic curve given points. Typically, used study whether quadratic hump pit located within studied interval. current test generalized applies generalized linear models (glm) link function instead simple quadratic curve. test popularized ecology analysis humped species richness patterns (Mittelbach et al. 2001), general. logarithmic link function, quadratic response defines Gaussian response model ecological gradients (ter Braak & Looman 1986), test can used inspecting location Gaussian optimum within given range gradient. can also used replace Tokeshi's test “bimodal” species frequency distribution.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"MOStest(x, y, interval, ...) # S3 method for MOStest plot(x, which = c(1,2,3,6), ...) fieller.MOStest(object, level = 0.95) # S3 method for MOStest profile(fitted, alpha = 0.01, maxsteps = 10, del = zmax/5, ...) # S3 method for MOStest confint(object, parm = 1, level = 0.95, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"x independent variable plotting object plot. y dependent variable. interval two points test statistic evaluated. missing, extremes x used. Subset plots produced. Values =1 2 define plots specific MOStest (see Details), larger values select graphs plot.lm (minus 2). object, fitted result object MOStest. level confidence level required. alpha Maximum significance level allowed. maxsteps Maximum number steps profile. del step length parameter profile (see code). parm Ignored. ... variables passed functions. Function MOStest passes glm can include family. functions pass underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function fits quadratic curve \\(\\mu = b_0 + b_1 x + b_2 x^2\\) given family link function. \\(b_2 < 0\\), defines unimodal curve highest point \\(u = -b_1/(2 b_2)\\) (ter Braak & Looman 1986). \\(b_2 > 0\\), parabola minimum \\(u\\) response sometimes called “bimodal”. null hypothesis extreme point \\(u\\) located within interval given points \\(p_1\\) \\(p_2\\). extreme point \\(u\\) exactly \\(p_1\\), \\(b_1 = 0\\) shifted axis \\(x - p_1\\). test, origin x shifted values \\(p_1\\) \\(p_2\\), test statistic based differences deviances original model model origin forced given location using standard anova.glm function (Oksanen et al. 2001). Mitchell-Olds & Shaw (1987) used first degree coefficient significance estimated summary.glm function. give identical results Normal error, error distributions preferable use test based differences deviances fitted models. test often presented general test location hump, really dependent quadratic fitted curve. hump different form quadratic, test may insignificant. strong assumptions test, use support functions inspect fit. Function plot(..., =1) displays data points, fitted quadratic model, approximate 95% confidence intervals (2 times SE). Function plot = 2 displays approximate confidence interval polynomial coefficients, together two lines indicating combinations coefficients produce evaluated points x. Moreover, cross-hair shows approximate confidence intervals polynomial coefficients ignoring correlations. Higher values produce corresponding graphs plot.lm. , must add 2 value plot.lm. Function fieller.MOStest approximates confidence limits location extreme point (hump pit) using Fieller's theorem following ter Braak & Looman (1986). test based quasideviance except family poisson binomial. Function profile evaluates profile deviance fitted model, confint finds profile based confidence limits following Oksanen et al. (2001). test typically used assessing significance diversity hump productivity gradient (Mittelbach et al. 2001). also can used location pit (deepest points) instead Tokeshi test. , can used test location Gaussian optimum ecological gradient analysis (ter Braak & Looman 1986, Oksanen et al. 2001).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function based glm, returns result object glm amended result test. new items MOStest : isHump TRUE response hump. isBracketed TRUE hump pit bracketed evaluated points. hump Sorted vector location hump pit points test evaluated. coefficients Table test statistics significances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds, T. & Shaw, R.G. 1987. Regression analysis natural selection: statistical inference biological interpretation. Evolution 41, 1149--1161. Mittelbach, G.C. Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, R.M., Dodson, S.. & Gough, L. 2001. observed relationship species richness productivity? Ecology 82, 2381--2396. Oksanen, J., Läärä, E., Tolonen, K. & Warner, B.G. 2001. Confidence intervals optimum Gaussian response function. Ecology 82, 1191--1197. ter Braak, C.J.F & Looman, C.W.N 1986. Weighted averaging, logistic regression Gaussian response model. Vegetatio 65, 3--11.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Function fieller.MOStest based package optgrad Ecological Archives (https://figshare.com/articles/dataset/Full_Archive/3521975) accompanying Oksanen et al. (2001). Ecological Archive package optgrad also contains profile deviance method location hump pit, current implementation profile confint rather follow example profile.glm confint.glm MASS package.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"## The Al-Mufti data analysed in humpfit(): mass <- c(140,230,310,310,400,510,610,670,860,900,1050,1160,1900,2480) spno <- c(1, 4, 3, 9, 18, 30, 20, 14, 3, 2, 3, 2, 5, 2) mod <- MOStest(mass, spno) ## Insignificant mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: gaussian #> Link function: identity #> #> hump min max #> 46.89749 140.00000 2480.00000 #> ***** Caution: hump/pit not bracketed by the data ****** #> #> min/max F Pr(>F) #> hump at min 140 0.0006 0.9816 #> hump at max 2480 0.3161 0.5852 #> Combined 0.9924 ## ... but inadequate shape of the curve op <- par(mfrow=c(2,2), mar=c(4,4,1,1)+.1) plot(mod) ## Looks rather like log-link with Poisson error and logarithmic biomass mod <- MOStest(log(mass), spno, family=quasipoisson) mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: quasipoisson #> Link function: log #> #> min hump max #> 4.941642 6.243371 7.816014 #> #> min/max F Pr(>F) #> hump at min 4.9416 7.1367 0.02174 * #> hump at max 7.8160 9.0487 0.01191 * #> Combined 0.03338 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod) par(op) ## Confidence Limits fieller.MOStest(mod) #> 2.5 % 97.5 % #> 5.255827 6.782979 confint(mod) #> 2.5 % 97.5 % #> 5.816021 6.574378 plot(profile(mod))"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":null,"dir":"Reference","previous_headings":"","what":"Adjusted R-square — RsquareAdj","title":"Adjusted R-square — RsquareAdj","text":"functions finds adjusted R-square.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Adjusted R-square — RsquareAdj","text":"","code":"# S3 method for default RsquareAdj(x, n, m, ...) # S3 method for rda RsquareAdj(x, ...) # S3 method for cca RsquareAdj(x, permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Adjusted R-square — RsquareAdj","text":"x Unadjusted R-squared object terms evaluation adjusted R-squared can found. n, m Number observations number degrees freedom fitted model. permutations Number permutations use computing adjusted R-squared cca. permutations can calculated parallel specifying number cores passed permutest ... arguments (ignored) except case cca arguments passed permutest.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Adjusted R-square — RsquareAdj","text":"default method finds adjusted \\(R^2\\) unadjusted \\(R^2\\), number observations, number degrees freedom fitted model. specific methods find information fitted result object. specific methods rda (also used distance-based RDA), cca, lm glm. Adjusted, even unadjusted, \\(R^2\\) may available cases, functions return NA. \\(R^2\\) values available gaussian models glm. adjusted, \\(R^2\\) cca computed using permutation approach developed Peres-Neto et al. (2006). default 1000 permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Adjusted R-square — RsquareAdj","text":"functions return list items r.squared adj.r.squared.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Adjusted R-square — RsquareAdj","text":"Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277. Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87, 2614--2625.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Adjusted R-square — RsquareAdj","text":"","code":"data(mite) data(mite.env) ## rda m <- rda(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.5265047 #> #> $adj.r.squared #> [1] 0.4367038 #> ## cca m <- cca(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.4471676 #> #> $adj.r.squared #> [1] 0.3446879 #> ## default method RsquareAdj(0.8, 20, 5) #> [1] 0.7285714"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":null,"dir":"Reference","previous_headings":"","what":"Self-Starting nls Species-Area Models — SSarrhenius","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions provide self-starting species-area models non-linear regression (nls). can also used fitting species accumulation models fitspecaccum. models (many ) reviewed Dengler (2009).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"SSarrhenius(area, k, z) SSgleason(area, k, slope) SSgitay(area, k, slope) SSlomolino(area, Asym, xmid, slope)"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"area Area size sample: independent variable. k, z, slope, Asym, xmid Estimated model parameters: see Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions assumed used species richness (number species) independent variable, area sample size independent variable. Basically, define least squares models untransformed data, differ models transformed species richness models non-Gaussian error. Arrhenius model (SSarrhenius) expression k*area^z. classical model can found textbook ecology (also Dengler 2009). Parameter z steepness species-area curve, k expected number species unit area. Gleason model (SSgleason) linear expression k + slope*log(area) (Dengler 200). linear model, starting values give final estimates; provided ease comparison models. Gitay model (SSgitay) quadratic logarithmic expression (k + slope*log(area))^2 (Gitay et al. 1991, Dengler 2009). Parameter slope steepness species-area curve, k square root expected richness unit area. Lomolino model (SSlomolino) Asym/(1 + slope^log(xmid/area)) (Lomolino 2000, Dengler 2009). Parameter Asym asymptotic maximum number species, slope maximum slope increase richness, xmid area half maximum richness achieved. addition models, several models studied Dengler (2009) available standard R self-starting models: Michaelis-Menten (SSmicmen), Gompertz (SSgompertz), logistic (SSlogis), Weibull (SSweibull), others may useful.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Numeric vector length area. value expression model. arguments names objects gradient matrix respect names attached attribute named gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Dengler, J. (2009) function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gitay, H., Roxburgh, S.H. & Wilson, J.B. (1991) Species-area relationship New Zealand tussock grassland, implications nature reserve design community structure. Journal Vegetation Science 2, 113--118. Lomolino, M. V. (2000) Ecology's general, yet protean pattern: species-area relationship. Journal Biogeography 27, 17--26.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"## Get species area data: sipoo.map gives the areas of islands data(sipoo, sipoo.map) S <- specnumber(sipoo) plot(S ~ area, sipoo.map, xlab = \"Island Area (ha)\", ylab = \"Number of Species\", ylim = c(1, max(S))) ## The Arrhenius model marr <- nls(S ~ SSarrhenius(area, k, z), data=sipoo.map) marr #> Nonlinear regression model #> model: S ~ SSarrhenius(area, k, z) #> data: sipoo.map #> k z #> 3.4062 0.4364 #> residual sum-of-squares: 78.1 #> #> Number of iterations to convergence: 5 #> Achieved convergence tolerance: 1.056e-06 ## confidence limits from profile likelihood confint(marr) #> Waiting for profiling to be done... #> 2.5% 97.5% #> k 2.6220312 4.3033906 #> z 0.3813576 0.4944693 ## draw a line xtmp <- with(sipoo.map, seq(min(area), max(area), len=51)) lines(xtmp, predict(marr, newdata=data.frame(area = xtmp)), lwd=2) ## The normal way is to use linear regression on log-log data, ## but this will be different from the previous: mloglog <- lm(log(S) ~ log(area), data=sipoo.map) mloglog #> #> Call: #> lm(formula = log(S) ~ log(area), data = sipoo.map) #> #> Coefficients: #> (Intercept) log(area) #> 1.0111 0.4925 #> lines(xtmp, exp(predict(mloglog, newdata=data.frame(area=xtmp))), lty=2) ## Gleason: log-linear mgle <- nls(S ~ SSgleason(area, k, slope), sipoo.map) lines(xtmp, predict(mgle, newdata=data.frame(area=xtmp)), lwd=2, col=2) ## Gitay: quadratic of log-linear mgit <- nls(S ~ SSgitay(area, k, slope), sipoo.map) lines(xtmp, predict(mgit, newdata=data.frame(area=xtmp)), lwd=2, col = 3) ## Lomolino: using original names of the parameters (Lomolino 2000): mlom <- nls(S ~ SSlomolino(area, Smax, A50, Hill), sipoo.map) mlom #> Nonlinear regression model #> model: S ~ SSlomolino(area, Smax, A50, Hill) #> data: sipoo.map #> Smax A50 Hill #> 53.493 94.697 2.018 #> residual sum-of-squares: 55.37 #> #> Number of iterations to convergence: 6 #> Achieved convergence tolerance: 9.715e-07 lines(xtmp, predict(mlom, newdata=data.frame(area=xtmp)), lwd=2, col = 4) ## One canned model of standard R: mmic <- nls(S ~ SSmicmen(area, Asym, slope), sipoo.map) lines(xtmp, predict(mmic, newdata = data.frame(area=xtmp)), lwd =2, col = 5) legend(\"bottomright\", c(\"Arrhenius\", \"log-log linear\", \"Gleason\", \"Gitay\", \"Lomolino\", \"Michaelis-Menten\"), col=c(1,1,2,3,4,5), lwd=c(2,1,2,2,2,2), lty=c(1,2,1,1,1,1)) ## compare models (AIC) allmods <- list(Arrhenius = marr, Gleason = mgle, Gitay = mgit, Lomolino = mlom, MicMen= mmic) sapply(allmods, AIC) #> Arrhenius Gleason Gitay Lomolino MicMen #> 83.49847 96.94018 80.54984 79.30718 83.02003"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Compute single terms can added dropped constrained ordination model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"# S3 method for cca add1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...) # S3 method for cca drop1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"object constrained ordination object cca, rda, dbrda capscale. scope formula giving terms considered adding dropping; see add1 details. test permutation test added using anova.cca. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ... arguments passed add1.default, drop1.default, anova.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"argument test = \"none\" functions call add1.default drop1.default. argument test = \"permutation\" functions add test results anova.cca. Function drop1.cca call anova.cca argument = \"margin\". Function add1.cca implement test single term additions directly available anova.cca. Functions used implicitly step, ordiR2step ordistep. deviance.cca deviance.rda used step firm basis, setting argument test = \"permutation\" may help getting useful insight validity model building. Function ordistep calls alternately drop1.cca add1.cca argument test = \"permutation\" selects variables permutation \\(P\\)-values. Meticulous use add1.cca drop1.cca allow judicious model building. default number permutations set low value, permutation tests can take long time. sufficient give impression significances terms, higher values permutations used \\(P\\) values really important.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Returns similar object add1 drop1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"data(dune) data(dune.env) ## Automatic model building based on AIC but with permutation tests step(cca(dune ~ 1, dune.env), reformulate(names(dune.env)), test=\"perm\") #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 86.608 2.2536 0.005 ** #> + Management 3 86.935 2.1307 0.005 ** #> + A1 1 87.411 2.1400 0.045 * #> 87.657 #> + Manure 4 88.832 1.5251 0.030 * #> + Use 2 89.134 1.1431 0.200 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC F Pr(>F) #> 86.608 #> + Management 3 86.813 1.4565 0.045 * #> + A1 1 86.992 1.2624 0.175 #> + Use 2 87.259 1.2760 0.085 . #> + Manure 4 87.342 1.3143 0.050 * #> - Moisture 3 87.657 2.2536 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #> ## see ?ordistep to do the same, but based on permutation P-values if (FALSE) { ordistep(cca(dune ~ 1, dune.env), reformulate(names(dune.env))) } ## Manual model building ## -- define the maximal model for scope mbig <- rda(dune ~ ., dune.env) ## -- define an empty model to start with m0 <- rda(dune ~ 1, dune.env) ## -- manual selection and updating add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 89.620 #> A1 1 89.591 1.9217 0.015 * #> Moisture 3 87.707 2.5883 0.005 ** #> Management 3 87.082 2.8400 0.005 ** #> Use 2 91.032 1.1741 0.225 #> Manure 4 89.232 1.9539 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Management) add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 87.082 #> A1 1 87.424 1.2965 0.215 #> Moisture 3 85.567 1.9764 0.010 ** #> Use 2 88.284 1.0510 0.400 #> Manure 3 87.517 1.3902 0.105 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Moisture) ## -- included variables still significant? drop1(m0, test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> Management 3 87.707 2.1769 0.010 ** #> Moisture 3 87.082 1.9764 0.015 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> A1 1 86.220 0.8359 0.605 #> Use 2 86.842 0.8027 0.765 #> Manure 3 85.762 1.1225 0.390"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"additive diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set (gamma diversity). hierarchical null model testing, statistic returned function evaluated according nested hierarchical sampling design (hiersimu).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"adipart(...) # S3 method for default adipart(y, x, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula adipart(formula, data, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) hiersimu(...) # S3 method for default hiersimu(y, x, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula hiersimu(formula, data, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, function performs overall decomposition alpha, beta gamma diversities. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variables referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, diversity index calculated (see Details). weights Character, \"unif\" uniform weights, \"prop\" weighting proportional sample abundances use weighted averaging individual alpha values within strata given level sampling hierarchy. relative Logical, TRUE alpha beta diversity values given relative value gamma function adipart. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. FUN function used hiersimu. must fully specified, currently arguments passed function via .... location Character, identifies function (mean median) used calculate location samples. drop.highest Logical, drop highest level . FUN evaluates arrays least 2 dimensions, highest level dropped, selected . ... arguments passed functions, e.g. base logarithm Shannon diversity, method, thin burnin arguments oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Additive diversity partitioning means mean alpha beta diversities add gamma diversity, thus beta diversity measured dimensions alpha gamma (Lande 1996). additive procedure extended across multiple scales hierarchical sampling design \\(= 1, 2, 3, \\ldots, m\\) levels sampling (Crist et al. 2003). Samples lower hierarchical levels nested within higher level units, thus \\(=1\\) \\(=m\\) grain size increasing constant survey extent. level \\(\\), \\(\\alpha_i\\) denotes average diversity found within samples. highest sampling level, diversity components calculated $$\\beta_m = \\gamma - \\alpha_m$$ lower sampling level $$\\beta_i = \\alpha_{+1} - \\alpha_i$$ , additive partition diversity $$\\gamma = \\alpha_1 + \\sum_{=1}^m \\beta_i$$ Average alpha components can weighted uniformly (weight=\"unif\") calculate simple average, proportionally sample abundances (weight=\"prop\") calculate weighted average follows $$\\alpha_i = \\sum_{j=1}^{n_i} D_{ij} w_{ij}$$ \\(D_{ij}\\) diversity index \\(w_{ij}\\) weight calculated \\(j\\)th sample \\(\\)th sampling level. implementation additive diversity partitioning adipart follows Crist et al. 2003. based species richness (\\(S\\), \\(S-1\\)), Shannon's Simpson's diversity indices stated index argument. expected diversity components calculated nsimul times individual based randomisation community data matrix. done \"r2dtable\" method oecosimu default. hiersimu works almost way adipart, without comparing actual statistic values returned FUN highest possible value (cf. gamma diversity). , cases, difficult ensure additive properties mean statistic values along hierarchy.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"object class \"adipart\" \"hiersimu\" structure oecosimu objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Crist, T.O., Veech, J.., Gering, J.C. Summerville, K.S. (2003). Partitioning species diversity across landscapes regions: hierarchical analysis \\(\\alpha\\), \\(\\beta\\), \\(\\gamma\\)-diversity. . Nat., 162, 734--743. Lande, R. (1996). Statistics partitioning species diversity, similarity among multiple communities. Oikos, 76, 5--13.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l1=1:nrow(mite), l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)), l4=rep(1, nrow(mite)))) ## Let's see in a map par(mfrow=c(1,3)) plot(mite.xy, main=\"l1\", col=as.numeric(levsm$l1)+1, asp = 1) plot(mite.xy, main=\"l2\", col=as.numeric(levsm$l2)+1, asp = 1) plot(mite.xy, main=\"l3\", col=as.numeric(levsm$l3)+1, asp = 1) par(mfrow=c(1,1)) ## Additive diversity partitioning adipart(mite, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -28.979 22.393 21.978 22.429 22.765 0.05 * #> gamma 35.000 0.000 35.000 35.000 35.000 35.000 1.00 #> beta.1 19.886 28.979 12.607 12.235 12.571 13.022 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## the next two define identical models adipart(mite, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, x = levsm, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -54.1889 22.35038 22.10429 22.37143 22.584 0.05 * #> alpha.2 29.750 -25.1503 34.81579 34.50000 34.75000 35.000 0.05 * #> alpha.3 33.000 0.0000 35.00000 35.00000 35.00000 35.000 0.05 * #> gamma 35.000 0.0000 35.00000 35.00000 35.00000 35.000 1.00 #> beta.1 14.636 9.9124 12.46541 12.10000 12.45000 12.779 0.05 * #> beta.2 3.250 15.2208 0.18421 0.00000 0.25000 0.500 0.05 * #> beta.3 2.000 0.0000 0.00000 0.00000 0.00000 0.000 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adipart(mite ~ l2 + l3, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -51.771 22.378947 22.100714 22.400000 22.587 0.05 * #> alpha.2 29.750 -25.578 34.736842 34.500000 34.750000 35.000 0.05 * #> alpha.3 33.000 -17.206 34.973684 34.725000 35.000000 35.000 0.05 * #> gamma 35.000 0.000 35.000000 35.000000 35.000000 35.000 1.00 #> beta.1 14.636 11.165 12.357895 12.025714 12.428571 12.686 0.05 * #> beta.2 3.250 15.455 0.236842 0.000000 0.250000 0.500 0.05 * #> beta.3 2.000 17.206 0.026316 0.000000 0.000000 0.275 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical null model testing ## diversity analysis (similar to adipart) hiersimu(mite, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(y = mite, FUN = diversity, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> level_1 0.76064 -52.201 0.93892 0.93378 0.93866 0.9444 0.05 * #> leve_2 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 hiersimu(mite ~ l2 + l3, levsm, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ l2 + l3, data = levsm, FUN = diversity, #> relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> unit 0.76064 -60.783 0.93927 0.93425 0.93883 0.9435 0.05 * #> l2 0.89736 -116.072 0.99795 0.99617 0.99808 0.9991 0.05 * #> l3 0.92791 -421.205 0.99932 0.99902 0.99935 0.9995 0.05 * #> all 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical testing with the Morisita index morfun <- function(x) dispindmorisita(x)$imst hiersimu(mite ~., levsm, morfun, drop.highest=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ ., data = levsm, FUN = morfun, #> drop.highest = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> l1 0.52070 6.4968 0.35648 0.31140 0.35449 0.3978 0.05 * #> l2 0.60234 12.5142 0.15359 0.08900 0.14525 0.2039 0.05 * #> l3 0.67509 17.6125 -0.19189 -0.26300 -0.20619 -0.1149 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Analysis variance using distance matrices --- partitioning distance matrices among sources variation fitting linear models (e.g., factors, polynomial regression) distance matrices; uses permutation test pseudo-\\(F\\) ratios.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"adonis2(formula, data, permutations = 999, method = \"bray\", sqrt.dist = FALSE, add = FALSE, by = \"terms\", parallel = getOption(\"mc.cores\"), na.action = na.fail, strata = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"formula Model formula. left-hand side (LHS) formula must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist used find dissimilarities. right-hand side (RHS) formula defines independent variables. can continuous variables factors, can transformed within formula, can interactions typical formula. data data frame independent variables, rows order community data matrix dissimilarity matrix named LHS formula. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. method name method used vegdist calculate pairwise distances left hand side formula data frame matrix. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" analyse one-degree--freedom contrasts sequentially, = NULL assess overall significance terms together. arguments passed anova.cca. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. na.action Handling missing values right-hand-side formula (see na.fail explanation alternatives). Missing values allowed left-hand-side. NB, argument subset implemented. strata Groups within constrain permutations. traditional non-movable strata set Blocks permute package, flexible alternatives may appropriate. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"adonis2 function analysis partitioning sums squares using dissimilarities. function based principles McArdle & Anderson (2001) can perform sequential, marginal overall tests. function also allows using additive constants squareroot dissimilarities avoid negative eigenvalues, can also handle semimetric indices (Bray-Curtis) produce negative eigenvalues. adonis2 tests identical anova.cca dbrda. Euclidean distances, tests also identical anova.cca rda. function partitions sums squares multivariate data set, directly analogous MANOVA (multivariate analysis variance). McArdle Anderson (2001) Anderson (2001) refer method “permutational MANOVA” (formerly “nonparametric MANOVA”). , inputs linear predictors, response matrix arbitrary number columns, robust alternative parametric MANOVA ordination methods describing variation attributed different experimental treatments uncontrolled covariates. method also analogous distance-based redundancy analysis algorithmically similar dbrda (Legendre Anderson 1999), provides alternative AMOVA (nested analysis molecular variance, Excoffier, Smouse, Quattro, 1992; amova ade4 package) crossed nested factors.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"function returns anova.cca result object new column partial \\(R^2\\): proportion sum squares total, marginal models (= \"margin\") \\(R^2\\) terms add 1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson (2001, Fig. 4) warns method may confound location dispersion effects: significant differences may caused different within-group variation (dispersion) instead different mean values groups (see Warton et al. 2012 general analysis). However, seems adonis2 less sensitive dispersion effects alternatives (anosim, mrpp). Function betadisper sister function adonis2 study differences dispersion within geometric framework.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson, M.J. 2001. new method non-parametric multivariate analysis variance. Austral Ecology, 26: 32--46. Excoffier, L., P.E. Smouse, J.M. Quattro. 1992. Analysis molecular variance inferred metric distances among DNA haplotypes: Application human mitochondrial DNA restriction data. Genetics, 131:479--491. Legendre, P. M.J. Anderson. 1999. Distance-based redundancy analysis: Testing multispecies responses multifactorial ecological experiments. Ecological Monographs, 69:1--24. McArdle, B.H. M.J. Anderson. 2001. Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology, 82: 290--297. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Martin Henry H. Stevens Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"data(dune) data(dune.env) ## default test by terms adonis2(dune ~ Management*A1, data = dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.2629 0.002 ** #> A1 1 0.4409 0.10256 2.9387 0.012 * #> Management:A1 3 0.5892 0.13705 1.3090 0.217 #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## overall tests adonis2(dune ~ Management*A1, data = dune.env, by = NULL) #> Permutation test for adonis under reduced model #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env, by = NULL) #> Df SumOfSqs R2 F Pr(>F) #> Model 7 2.4987 0.58122 2.3792 0.004 ** #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ### Example of use with strata, for nested (e.g., block) designs. dat <- expand.grid(rep=gl(2,1), NO3=factor(c(0,10)),field=gl(3,1) ) dat #> rep NO3 field #> 1 1 0 1 #> 2 2 0 1 #> 3 1 10 1 #> 4 2 10 1 #> 5 1 0 2 #> 6 2 0 2 #> 7 1 10 2 #> 8 2 10 2 #> 9 1 0 3 #> 10 2 0 3 #> 11 1 10 3 #> 12 2 10 3 Agropyron <- with(dat, as.numeric(field) + as.numeric(NO3)+2) +rnorm(12)/2 Schizachyrium <- with(dat, as.numeric(field) - as.numeric(NO3)+2) +rnorm(12)/2 total <- Agropyron + Schizachyrium dotplot(total ~ NO3, dat, jitter.x=TRUE, groups=field, type=c('p','a'), xlab=\"NO3\", auto.key=list(columns=3, lines=TRUE) ) Y <- data.frame(Agropyron, Schizachyrium) mod <- metaMDS(Y, trace = FALSE) plot(mod) ### Ellipsoid hulls show treatment with(dat, ordiellipse(mod, NO3, kind = \"ehull\", label = TRUE)) ### Spider shows fields with(dat, ordispider(mod, field, lty=3, col=\"red\", label = TRUE)) ### Incorrect (no strata) adonis2(Y ~ NO3, data = dat, permutations = 199) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.06 . #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Correct with strata with(dat, adonis2(Y ~ NO3, data = dat, permutations = 199, strata = field)) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Blocks: strata #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199, strata = field) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.005 ** #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":null,"dir":"Reference","previous_headings":"","what":"Analysis of Similarities — anosim","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Analysis of Similarities — anosim","text":"","code":"anosim(x, grouping, permutations = 999, distance = \"bray\", strata = NULL, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Analysis of Similarities — anosim","text":"x Data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. distance Choice distance metric measures dissimilarity two observations. See vegdist options. used x dissimilarity structure symmetric square matrix. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units. Function anosim operates directly dissimilarity matrix. suitable dissimilarity matrix produced functions dist vegdist. method philosophically allied NMDS ordination (monoMDS), uses rank order dissimilarity values. two groups sampling units really different species composition, compositional dissimilarities groups greater within groups. anosim statistic \\(R\\) based difference mean ranks groups (\\(r_B\\)) within groups (\\(r_W\\)): $$R = (r_B - r_W)/(N (N-1) / 4)$$ divisor chosen \\(R\\) interval \\(-1 \\dots +1\\), value \\(0\\) indicating completely random grouping. statistical significance observed \\(R\\) assessed permuting grouping vector obtain empirical distribution \\(R\\) null-model. See permutations additional details permutation tests Vegan. distribution simulated values can inspected permustats function. function summary plot methods. show valuable information assess validity method: function assumes ranked dissimilarities within groups equal median range. plot method uses boxplot options notch=TRUE varwidth=TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Analysis of Similarities — anosim","text":"function returns list class \"anosim\" following items: call Function call. statistic value ANOSIM statistic \\(R\\) signif Significance permutation. perm Permutation values \\(R\\). distribution permutation values can inspected function permustats. class.vec Factor value dissimilarities classes class name corresponding dissimilarity within class. dis.rank Rank dissimilarity entry. dissimilarity name dissimilarity index: \"method\" entry dist object. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Analysis of Similarities — anosim","text":"Clarke, K. R. (1993). Non-parametric multivariate analysis changes community structure. Australian Journal Ecology 18, 117--143. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Analysis of Similarities — anosim","text":"Jari Oksanen, help Peter R. Minchin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Analysis of Similarities — anosim","text":"anosim function can confound differences groups dispersion within groups results can difficult interpret (cf. Warton et al. 2012). function returns lot information ease studying performance. anosim models analysed adonis2 seems robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Analysis of Similarities — anosim","text":"","code":"data(dune) data(dune.env) dune.dist <- vegdist(dune) dune.ano <- with(dune.env, anosim(dune.dist, Management)) summary(dune.ano) #> #> Call: #> anosim(x = dune.dist, grouping = Management) #> Dissimilarity: bray #> #> ANOSIM statistic R: 0.2579 #> Significance: 0.009 #> #> Permutation: free #> Number of permutations: 999 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.121 0.164 0.205 0.242 #> #> Dissimilarity ranks between and within classes: #> 0% 25% 50% 75% 100% N #> Between 4 58.50 104.00 145.500 188.0 147 #> BF 5 15.25 25.50 41.250 57.0 3 #> HF 1 7.25 46.25 68.125 89.5 10 #> NM 6 64.75 124.50 156.250 181.0 15 #> SF 3 32.75 53.50 99.250 184.0 15 #> plot(dune.ano) #> Warning: some notches went outside hinges ('box'): maybe set notch=FALSE"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function performs ANOVA like permutation test Constrained Correspondence Analysis (cca), Redundancy Analysis (rda) distance-based Redundancy Analysis (dbRDA, dbrda) assess significance constraints.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"# S3 method for cca anova(object, ..., permutations = how(nperm=999), by = NULL, model = c(\"reduced\", \"direct\", \"full\"), parallel = getOption(\"mc.cores\"), strata = NULL, cutoff = 1, scope = NULL) # S3 method for cca permutest(x, permutations = how(nperm = 99), model = c(\"reduced\", \"direct\", \"full\"), by = NULL, first = FALSE, strata = NULL, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"object One several result objects cca, rda, dbrda capscale. several result objects, compared order supplied. single object, test specified overall test given. x single ordination result object. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Setting = \"axis\" assess significance constrained axis, setting = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" assess sequentially one-degree--freedom contrasts split factors. model Permutation model: model=\"direct\" permutes community data, model=\"reduced\" permutes residuals community data Conditions (partial model), model = \"full\" permutes residuals Conditions Constraints. parallel Use parallel processing given number cores. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. error use permutations matrix, defines blocks. legacy argument deprecated future: use permutations = (..., blocks) instead. cutoff effective =\"axis\" stops permutations axis equals exceeds cutoff \\(p\\)-value. scope effective =\"margin\" can used select marginal terms testing. default test marginal terms drop.scope. first Analyse significance first axis. ... Parameters passed functions. anova.cca passes arguments permutest.cca. anova = \"axis\" can use argument cutoff (defaults 1) stops permutations exceeding given level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Functions anova.cca permutest.cca implement ANOVA like permutation tests joint effect constraints cca, rda, dbrda capscale. Function anova intended user-friendly alternative permutest (real workhorse). Function anova can analyse sequence constrained ordination models. analysis based differences residual deviance permutations nested models. default test sum constrained eigenvalues. Setting first = TRUE perform test first constrained eigenvalue. Argument first can set either anova.cca permutest.cca. also possible perform significance tests axis term (constraining variable) using argument anova.cca. Setting = \"axis\" perform separate significance tests constrained axis. previous constrained axes used conditions (“partialled ”) test first constrained eigenvalues performed (Legendre et al. 2011). can stop permutation tests exceeding given significance level argument cutoff speed calculations large models. Setting = \"terms\" perform separate significance test term (constraining variable). terms assessed sequentially first last, order terms influence significances. Setting = \"onedf\" perform similar sequential test one-degree--freedom effects, multi-level factors split contrasts. Setting = \"margin\" perform separate significance test marginal term model terms. marginal test also accepts scope argument drop.scope can character vector term labels analysed, fitted model lower scope. marginal effects also known “Type III” effects, current function evaluates marginal terms. , instance, ignore main effects included interaction terms. calculating pseudo-\\(F\\), terms compared residual full model. Community data permuted choice model=\"direct\", residuals partial CCA/ RDA/ dbRDA choice model=\"reduced\" (default). partial CCA/ RDA/ dbRDA stage, model=\"reduced\" simply permutes data equivalent model=\"direct\". test statistic “pseudo-\\(F\\)”, ratio constrained unconstrained total Inertia (Chi-squares, variances something similar), divided respective degrees freedom. conditions (“partial” terms), sum eigenvalues remains constant, pseudo-\\(F\\) eigenvalues give equal results. partial CCA/ RDA/ dbRDA, effect conditioning variables (“covariables”) removed permutation, total Chi-square fixed, test based pseudo-\\(F\\) differ test based plain eigenvalues.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function anova.cca calls permutest.cca fills anova table. Additional attributes Random.seed (random seeds used), control (permutation design, see ) F.perm (permuted test statistics).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Legendre, P. Legendre, L. (2012). Numerical Ecology. 3rd English ed. Elsevier. Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"data(dune, dune.env) mod <- cca(dune ~ Moisture + Management, dune.env) ## overall test anova(mod) #> Permutation test for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Model 6 1.0024 1.9515 0.003 ** #> Residual 13 1.1129 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## tests for individual terms anova(mod, by=\"term\") #> Permutation test for cca under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.62831 2.4465 0.001 *** #> Management 3 0.37407 1.4565 0.049 * #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 anova(mod, by=\"margin\") #> Permutation test for cca under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.39854 1.5518 0.036 * #> Management 3 0.37407 1.4565 0.053 . #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential test for contrasts anova(mod, by = \"onedf\") #> Permutation test for cca under reduced model #> Sequential test for contrasts #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture.L 1 0.41081 4.7988 0.001 *** #> Moisture.Q 1 0.11261 1.3154 0.167 #> Moisture.C 1 0.10489 1.2253 0.226 #> ManagementHF 1 0.08849 1.0337 0.359 #> ManagementNM 1 0.20326 2.3744 0.011 * #> ManagementSF 1 0.08231 0.9615 0.455 #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## test for adding all environmental variables anova(mod, cca(dune ~ ., dune.env)) #> Permutation tests for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model 1: dune ~ Moisture + Management #> Model 2: dune ~ A1 + Moisture + Management + Use + Manure #> ResDf ResChiSquare Df ChiSquare F Pr(>F) #> 1 13 1.1129 #> 2 7 0.6121 6 0.50079 0.9545 0.532"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Averaged Subsampled Dissimilarity Matrices — avgdist","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function computes dissimilarity matrix dataset multiple times using vegdist randomly subsampling dataset time. subsampled iterations averaged (mean) provide distance matrix represents average multiple subsampling iterations. emulates behavior distance matrix calculator within Mothur microbial ecology toolkit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"avgdist(x, sample, distfun = vegdist, meanfun = mean, transf = NULL, iterations = 100, dmethod = \"bray\", diag = TRUE, upper = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"x Community data matrix. sample subsampling depth used iteration. Samples meet threshold removed analysis, identity returned user stdout. distfun dissimilarity matrix function used. Default vegan vegdist meanfun calculation use average (mean median). transf Option transforming count data calculating distance matrix. base transformation option can used (e.g. sqrt) iterations number random iterations perform averaging. Default 100 iterations. dmethod Dissimilarity index used specified dissimilarity matrix function. Default Bray-Curtis diag, upper Return dissimilarities diagonal upper triangle. NB. default differs vegdist returns symmetric \"dist\" structure instead lower diagonal. However, object accessed matrix indices unless cast matrix .matrix. ... additional arguments add distance function mean/median function specified.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"Geoffrey Hannigan, minor tweaks Gavin L. Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function builds function rrarefy additional distance matrix function (e.g. vegdist) add meaningful representations distances among randomly subsampled datasets presenting average multiple random iterations. function runs using vegdist. functionality utilized Mothur standalone microbial ecology toolkit .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"# Import an example count dataset data(BCI) # Test the base functionality mean.avg.dist <- avgdist(BCI, sample = 50, iterations = 10) # Test the transformation function mean.avg.dist.t <- avgdist(BCI, sample = 50, iterations = 10, transf = sqrt) # Test the median functionality median.avg.dist <- avgdist(BCI, sample = 50, iterations = 10, meanfun = median) # Print the resulting tables head(as.matrix(mean.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 #> 1 0.000 0.556 0.606 0.648 0.620 0.624 0.610 0.606 0.654 0.624 0.638 0.648 0.708 #> 2 0.556 0.000 0.568 0.568 0.632 0.568 0.532 0.554 0.576 0.578 0.588 0.560 0.678 #> 3 0.606 0.568 0.000 0.582 0.578 0.618 0.600 0.568 0.610 0.574 0.610 0.622 0.738 #> 4 0.648 0.568 0.582 0.000 0.624 0.654 0.610 0.572 0.594 0.594 0.596 0.636 0.672 #> 5 0.620 0.632 0.578 0.624 0.000 0.602 0.662 0.626 0.670 0.606 0.678 0.716 0.754 #> 6 0.624 0.568 0.618 0.654 0.602 0.000 0.598 0.566 0.616 0.644 0.576 0.568 0.706 #> 14 15 16 17 18 19 20 21 22 23 24 25 26 #> 1 0.630 0.648 0.600 0.646 0.744 0.682 0.626 0.600 0.646 0.700 0.658 0.620 0.620 #> 2 0.600 0.592 0.554 0.564 0.660 0.584 0.598 0.606 0.544 0.642 0.574 0.624 0.592 #> 3 0.628 0.612 0.616 0.646 0.730 0.660 0.646 0.568 0.636 0.722 0.652 0.672 0.660 #> 4 0.602 0.602 0.618 0.634 0.666 0.622 0.636 0.656 0.568 0.642 0.614 0.654 0.638 #> 5 0.610 0.642 0.630 0.760 0.766 0.698 0.654 0.642 0.738 0.702 0.652 0.688 0.652 #> 6 0.618 0.696 0.626 0.618 0.668 0.634 0.664 0.648 0.630 0.694 0.608 0.682 0.674 #> 27 28 29 30 31 32 33 34 35 36 37 38 39 #> 1 0.702 0.684 0.674 0.634 0.680 0.724 0.692 0.684 0.744 0.666 0.694 0.716 0.700 #> 2 0.632 0.586 0.568 0.582 0.646 0.606 0.618 0.634 0.722 0.638 0.590 0.590 0.616 #> 3 0.668 0.660 0.654 0.676 0.660 0.674 0.694 0.700 0.764 0.684 0.628 0.684 0.708 #> 4 0.644 0.596 0.612 0.608 0.662 0.594 0.610 0.632 0.730 0.648 0.562 0.588 0.636 #> 5 0.712 0.752 0.728 0.734 0.676 0.710 0.698 0.768 0.830 0.678 0.690 0.752 0.760 #> 6 0.628 0.640 0.630 0.668 0.668 0.636 0.662 0.682 0.768 0.678 0.658 0.664 0.700 #> 40 41 42 43 44 45 46 47 48 49 50 #> 1 0.714 0.686 0.624 0.688 0.654 0.668 0.668 0.650 0.676 0.694 0.676 #> 2 0.638 0.670 0.644 0.654 0.626 0.664 0.648 0.638 0.648 0.678 0.672 #> 3 0.716 0.702 0.634 0.672 0.668 0.646 0.720 0.730 0.712 0.698 0.704 #> 4 0.680 0.652 0.618 0.680 0.698 0.696 0.658 0.666 0.658 0.696 0.708 #> 5 0.828 0.688 0.654 0.658 0.690 0.654 0.772 0.720 0.704 0.692 0.656 #> 6 0.724 0.686 0.678 0.670 0.666 0.698 0.734 0.692 0.714 0.734 0.704 head(as.matrix(mean.avg.dist.t)) #> 1 2 3 4 5 6 7 #> 1 0.0000000 0.5073247 0.5451162 0.5511308 0.5863431 0.5693314 0.5561033 #> 2 0.5073247 0.0000000 0.5076177 0.5454488 0.5590474 0.5361407 0.5075526 #> 3 0.5451162 0.5076177 0.0000000 0.5169507 0.5442375 0.5536933 0.5580432 #> 4 0.5511308 0.5454488 0.5169507 0.0000000 0.5556915 0.6029668 0.5894129 #> 5 0.5863431 0.5590474 0.5442375 0.5556915 0.0000000 0.5730661 0.5904097 #> 6 0.5693314 0.5361407 0.5536933 0.6029668 0.5730661 0.0000000 0.5161496 #> 8 9 10 11 12 13 14 #> 1 0.5673559 0.5627535 0.5792425 0.5729159 0.6020539 0.7016723 0.5490011 #> 2 0.5366592 0.5380985 0.5623811 0.5321790 0.5515688 0.6779465 0.5572373 #> 3 0.5073755 0.5536056 0.5389882 0.5801096 0.5868026 0.7260347 0.5816036 #> 4 0.5259962 0.5706750 0.5613498 0.5523447 0.5753112 0.6804780 0.5838035 #> 5 0.5484180 0.5855322 0.5881431 0.6281267 0.6276212 0.7377994 0.5875698 #> 6 0.5529963 0.5552535 0.6215621 0.5618841 0.5336650 0.6716372 0.5641578 #> 15 16 17 18 19 20 21 #> 1 0.6050878 0.5605091 0.6262520 0.6790944 0.6158975 0.5894641 0.6238942 #> 2 0.5492945 0.5571279 0.5683865 0.6719160 0.5864794 0.5817164 0.5785775 #> 3 0.5651808 0.5907658 0.5986803 0.7180656 0.6208023 0.5831755 0.6141896 #> 4 0.5846365 0.5995058 0.6019378 0.6723126 0.6185437 0.5788246 0.6204008 #> 5 0.5886468 0.6313116 0.6908350 0.7310610 0.6738026 0.6056946 0.6248085 #> 6 0.6245763 0.5734551 0.5897524 0.6553604 0.5945132 0.6253476 0.6236726 #> 22 23 24 25 26 27 28 #> 1 0.6087932 0.6864876 0.5674543 0.6124061 0.6000094 0.6287013 0.6246696 #> 2 0.5851407 0.6506247 0.5554757 0.5519726 0.6143138 0.5810440 0.5887911 #> 3 0.6301892 0.7101043 0.5608707 0.5964616 0.6536689 0.6015372 0.6254962 #> 4 0.6116276 0.6570546 0.5806733 0.5942875 0.6756654 0.6170196 0.6088559 #> 5 0.6692645 0.7281868 0.6019467 0.6240673 0.6717579 0.6824310 0.6781349 #> 6 0.5779917 0.6752783 0.5474248 0.6048853 0.6297599 0.6032439 0.6025756 #> 29 30 31 32 33 34 35 #> 1 0.5877126 0.5931557 0.6122546 0.6435203 0.6114740 0.6243358 0.7012874 #> 2 0.5493477 0.5852872 0.5988185 0.6153332 0.5921568 0.6244034 0.6905708 #> 3 0.5678376 0.6210073 0.6158757 0.6377790 0.6371270 0.6555867 0.7323038 #> 4 0.5759931 0.5912841 0.5983615 0.6066556 0.6016788 0.6087876 0.6950808 #> 5 0.6276522 0.6379576 0.6034350 0.6457655 0.6563073 0.6866576 0.7235042 #> 6 0.5816252 0.6242943 0.6152092 0.6137642 0.6116231 0.6246531 0.7312678 #> 36 37 38 39 40 41 42 #> 1 0.5974168 0.6467898 0.6716840 0.6643855 0.6683070 0.6353745 0.6004874 #> 2 0.5903695 0.5752574 0.6187077 0.6192344 0.6573943 0.6571387 0.6027516 #> 3 0.5687077 0.6126409 0.6617885 0.6959650 0.6909293 0.6525653 0.6014759 #> 4 0.6194344 0.5761296 0.6102164 0.6295049 0.6560977 0.5989171 0.6281164 #> 5 0.6056532 0.6441953 0.6593381 0.7110348 0.7252049 0.6568578 0.6156351 #> 6 0.6109414 0.6067715 0.6656711 0.6870593 0.6686455 0.6769139 0.6426746 #> 43 44 45 46 47 48 49 #> 1 0.6310309 0.5900500 0.5774603 0.6474425 0.6488217 0.6157111 0.6070289 #> 2 0.6028820 0.6126007 0.6208170 0.6652427 0.6134982 0.6243098 0.6319083 #> 3 0.6162330 0.6398741 0.6039060 0.7021972 0.6677691 0.6678864 0.6696487 #> 4 0.6197236 0.6379939 0.6156272 0.6582549 0.6062615 0.6139683 0.6466872 #> 5 0.6203780 0.6542059 0.6040162 0.7417980 0.6946591 0.6954255 0.6537119 #> 6 0.6386277 0.6418793 0.6306854 0.6948586 0.6326248 0.6678749 0.6692019 #> 50 #> 1 0.5867459 #> 2 0.6227903 #> 3 0.6365131 #> 4 0.6260672 #> 5 0.6250031 #> 6 0.6423431 head(as.matrix(median.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 #> 1 0.00 0.56 0.60 0.62 0.66 0.63 0.58 0.62 0.65 0.61 0.59 0.65 0.74 0.64 0.64 #> 2 0.56 0.00 0.53 0.59 0.64 0.63 0.60 0.58 0.58 0.60 0.59 0.62 0.68 0.58 0.61 #> 3 0.60 0.53 0.00 0.61 0.63 0.58 0.56 0.57 0.60 0.55 0.60 0.64 0.76 0.66 0.61 #> 4 0.62 0.59 0.61 0.00 0.64 0.66 0.62 0.58 0.57 0.58 0.63 0.62 0.66 0.55 0.57 #> 5 0.66 0.64 0.63 0.64 0.00 0.57 0.65 0.64 0.62 0.62 0.68 0.67 0.76 0.66 0.65 #> 6 0.63 0.63 0.58 0.66 0.57 0.00 0.58 0.62 0.61 0.66 0.60 0.59 0.74 0.69 0.70 #> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 #> 1 0.63 0.65 0.76 0.67 0.65 0.65 0.66 0.68 0.66 0.66 0.62 0.66 0.65 0.66 0.66 #> 2 0.62 0.59 0.69 0.61 0.56 0.64 0.62 0.67 0.60 0.58 0.62 0.64 0.63 0.56 0.63 #> 3 0.61 0.64 0.72 0.69 0.64 0.62 0.62 0.74 0.65 0.65 0.66 0.64 0.61 0.60 0.64 #> 4 0.63 0.62 0.67 0.61 0.60 0.68 0.59 0.67 0.62 0.60 0.66 0.60 0.62 0.59 0.62 #> 5 0.65 0.73 0.81 0.70 0.65 0.64 0.72 0.69 0.71 0.67 0.71 0.68 0.72 0.70 0.67 #> 6 0.63 0.60 0.68 0.63 0.66 0.68 0.63 0.70 0.61 0.64 0.66 0.63 0.65 0.62 0.69 #> 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 #> 1 0.65 0.69 0.66 0.66 0.74 0.65 0.68 0.69 0.70 0.68 0.67 0.63 0.63 0.64 0.67 #> 2 0.62 0.61 0.61 0.63 0.75 0.66 0.60 0.60 0.67 0.68 0.68 0.62 0.64 0.64 0.69 #> 3 0.63 0.61 0.69 0.68 0.78 0.67 0.61 0.65 0.68 0.72 0.71 0.65 0.66 0.64 0.65 #> 4 0.59 0.65 0.61 0.66 0.79 0.61 0.57 0.61 0.65 0.66 0.68 0.63 0.66 0.68 0.71 #> 5 0.65 0.71 0.70 0.75 0.82 0.70 0.65 0.73 0.79 0.76 0.72 0.64 0.73 0.70 0.68 #> 6 0.68 0.68 0.67 0.68 0.82 0.72 0.71 0.69 0.71 0.75 0.74 0.69 0.67 0.69 0.72 #> 46 47 48 49 50 #> 1 0.69 0.66 0.66 0.66 0.67 #> 2 0.69 0.62 0.67 0.63 0.65 #> 3 0.73 0.70 0.70 0.68 0.72 #> 4 0.68 0.65 0.65 0.69 0.64 #> 5 0.76 0.70 0.75 0.72 0.70 #> 6 0.72 0.71 0.73 0.72 0.71 # Run example to illustrate low variance of mean, median, and stdev results # Mean and median std dev are around 0.05 sdd <- avgdist(BCI, sample = 50, iterations = 100, meanfun = sd) summary(mean.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4640 0.6180 0.6540 0.6543 0.6900 0.8360 summary(median.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4600 0.6100 0.6500 0.6501 0.6900 0.8600 summary(sdd) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.03849 0.05617 0.05921 0.05919 0.06208 0.07661 # Test for when subsampling depth excludes some samples # Return samples that are removed for not meeting depth filter depth.avg.dist <- avgdist(BCI, sample = 450, iterations = 10) #> Warning: The following sampling units were removed because they were below sampling depth: 1, 2, 6, 7, 8, 9, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 31, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 # Print the result depth.avg.dist #> 3 4 5 10 15 30 32 #> 3 0.0000000 0.3282222 0.3635556 0.2942222 0.3562222 0.4555556 0.4726667 #> 4 0.3282222 0.0000000 0.3775556 0.3173333 0.3480000 0.3908889 0.4273333 #> 5 0.3635556 0.3775556 0.0000000 0.3900000 0.3995556 0.4968889 0.5231111 #> 10 0.2942222 0.3173333 0.3900000 0.0000000 0.3135556 0.4266667 0.4200000 #> 15 0.3562222 0.3480000 0.3995556 0.3135556 0.0000000 0.4535556 0.4657778 #> 30 0.4555556 0.3908889 0.4968889 0.4266667 0.4535556 0.0000000 0.3811111 #> 32 0.4726667 0.4273333 0.5231111 0.4200000 0.4657778 0.3811111 0.0000000 #> 35 0.6591111 0.6266667 0.6966667 0.6817778 0.6573333 0.5157778 0.5893333 #> 40 0.5624444 0.5286667 0.6413333 0.5240000 0.5628889 0.4540000 0.4057778 #> 35 40 #> 3 0.6591111 0.5624444 #> 4 0.6266667 0.5286667 #> 5 0.6966667 0.6413333 #> 10 0.6817778 0.5240000 #> 15 0.6573333 0.5628889 #> 30 0.5157778 0.4540000 #> 32 0.5893333 0.4057778 #> 35 0.0000000 0.4280000 #> 40 0.4280000 0.0000000"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":null,"dir":"Reference","previous_headings":"","what":"Beals Smoothing and Degree of Absence — beals","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing replaces entry community data probability target species occurring particular site, based joint occurrences target species species actually occur site. Swan's (1970) degree absence applies Beals smoothing zero items long zeros replaced smoothed values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"beals(x, species = NA, reference = x, type = 0, include = TRUE) swan(x, maxit = Inf, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Beals Smoothing and Degree of Absence — beals","text":"x Community data frame matrix. species Column index used compute Beals function single species. default (NA) indicates function computed species. reference Community data frame matrix used compute joint occurrences. default, x used reference compute joint occurrences. type Numeric. Specifies abundance values used function beals. See details explanation. include logical flag indicates whether target species included computing mean conditioned probabilities. original Beals (1984) definition equivalent include=TRUE, formulation Münzbergová Herben equal include=FALSE. maxit Maximum number iterations. default Inf means iterations continued zeros number zeros change. Probably maxit = 1 makes sense addition default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing estimated probability \\(p_{ij}\\) species \\(j\\) occurs site \\(\\). defined \\(p_{ij} = \\frac{1}{S_i} \\sum_k \\frac{N_{jk} I_{ik}}{N_k}\\), \\(S_i\\) number species site \\(\\), \\(N_{jk}\\) number joint occurrences species \\(j\\) \\(k\\), \\(N_k\\) number occurrences species \\(k\\), \\(\\) incidence (0 1) species (last term usually omitted equation, necessary). \\(N_{jk}\\) can interpreted mean conditional probability, beals function can interpreted mean conditioned probabilities (De Cáceres & Legendre 2008). present function generalized abundance values (De Cáceres & Legendre 2008). type argument specifies abundance values used. type = 0 presence/absence mode. type = 1 abundances reference (x) used compute conditioned probabilities. type = 2 abundances x used compute weighted averages conditioned probabilities. type = 3 abundances used compute conditioned probabilities weighted averages. Beals smoothing originally suggested method data transformation remove excessive zeros (Beals 1984, McCune 1994). However, suitable method purpose since maintain information species presences: species may higher probability occurrence site occur sites occurs. Moreover, regularizes data strongly. method may useful identifying species belong species pool (Ewald 2002) identify suitable unoccupied patches metapopulation analysis (Münzbergová & Herben 2004). case, function called include=FALSE cross-validation smoothing species; argument species can used one species studied. Swan (1970) suggested replacing zero values degrees absence species community data matrix. Swan expressed method terms similarity matrix, equivalent applying Beals smoothing zero values, step shifting smallest initially non-zero item value one, repeating many times zeros left data. actually similar extended dissimilarities (implemented function stepacross), rarely used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Beals Smoothing and Degree of Absence — beals","text":"function returns transformed data matrix vector Beals smoothing requested single species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals, E.W. 1984. Bray-Curtis ordination: effective strategy analysis multivariate ecological data. Pp. 1--55 : MacFadyen, . & E.D. Ford [eds.] Advances Ecological Research, 14. Academic Press, London. De Cáceres, M. & Legendre, P. 2008. Beals smoothing revisited. Oecologia 156: 657--669. Ewald, J. 2002. probabilistic approach estimating species pools large compositional matrices. J. Veg. Sci. 13: 191--198. McCune, B. 1994. Improving community ordination Beals smoothing function. Ecoscience 1: 82--86. Münzbergová, Z. & Herben, T. 2004. Identification suitable unoccupied habitats metapopulation studies using co-occurrence species. Oikos 105: 408--414. Swan, J.M.. 1970. examination ordination problems use simulated vegetational data. Ecology 51: 89--102.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Beals Smoothing and Degree of Absence — beals","text":"Miquel De Cáceres Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"data(dune) ## Default x <- beals(dune) ## Remove target species x <- beals(dune, include = FALSE) ## Smoothed values against presence or absence of species pa <- decostand(dune, \"pa\") boxplot(as.vector(x) ~ unlist(pa), xlab=\"Presence\", ylab=\"Beals\") ## Remove the bias of tarbet species: Yields lower values. beals(dune, type =3, include = FALSE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord #> 1 0.49590853 0.38333415 0.01157407 0.4923280 0.30827883 0.4935662 0.43263047 #> 2 0.47083676 0.39501120 0.03361524 0.4718807 0.34723984 0.4917791 0.42000984 #> 3 0.34063019 0.52738394 0.01520046 0.5309152 0.21609954 0.4033301 0.33010938 #> 4 0.30816435 0.51198853 0.02876960 0.5971801 0.21542662 0.4398775 0.35732610 #> 5 0.59949785 0.27622698 0.06632771 0.3349203 0.48876285 0.4322142 0.44309579 #> 6 0.58819821 0.26299306 0.05967771 0.2700508 0.53154426 0.3696613 0.39760652 #> 7 0.56496165 0.29412293 0.05329633 0.3403047 0.48010987 0.4051777 0.40471531 #> 8 0.21230502 0.66906674 0.02588333 0.5187956 0.16247716 0.2720122 0.21219877 #> 9 0.30323659 0.59744543 0.02213662 0.5792855 0.21896113 0.3292320 0.28613526 #> 10 0.54083871 0.26902092 0.07349127 0.3372958 0.42671693 0.4705094 0.42934344 #> 11 0.40509331 0.31656550 0.10259239 0.3185489 0.38766111 0.3713794 0.31413659 #> 12 0.21008725 0.66278454 0.03625297 0.5753377 0.20078932 0.2802946 0.22974415 #> 13 0.21850759 0.68239707 0.02191119 0.6404427 0.16737280 0.2939740 0.24942466 #> 14 0.13570397 0.76284476 0.02298398 0.4107645 0.12128973 0.1682755 0.13757552 #> 15 0.09168815 0.79412733 0.02538032 0.4505613 0.10117099 0.1420251 0.09794548 #> 16 0.06335463 0.87877202 0.00742115 0.5232448 0.05538377 0.1516354 0.09458531 #> 17 0.55254140 0.07330247 0.29233391 0.1013889 0.69331132 0.3129358 0.34982363 #> 18 0.37751017 0.34451209 0.08535723 0.2838834 0.36918166 0.3676424 0.30478244 #> 19 0.29826049 0.25952255 0.35137675 0.1934048 0.51929869 0.2237843 0.18074796 #> 20 0.05429986 0.76675441 0.06144615 0.4063662 0.10738280 0.1450721 0.06706410 #> Chenalbu Cirsarve Comapalu Eleopalu Elymrepe Empenigr #> 1 0.025132275 0.09504980 0.000000000 0.05592045 0.4667439 0.00000000 #> 2 0.043866562 0.08570299 0.026548839 0.08656209 0.4407282 0.01829337 #> 3 0.065338638 0.08967477 0.031898812 0.16099072 0.4137888 0.01074444 #> 4 0.057970906 0.12920228 0.039859621 0.16112450 0.4399661 0.02527165 #> 5 0.026434737 0.05520104 0.015892090 0.05419613 0.3575948 0.03029752 #> 6 0.021256367 0.03223112 0.030347896 0.08784329 0.3138879 0.03093489 #> 7 0.038467708 0.04706743 0.017083997 0.06694311 0.3586644 0.02304603 #> 8 0.063278453 0.06688407 0.100703044 0.29777644 0.3046956 0.02102222 #> 9 0.069879277 0.07647268 0.045830682 0.19018562 0.3523460 0.01838883 #> 10 0.025686639 0.06037513 0.029746617 0.07787078 0.3736128 0.03425596 #> 11 0.021234732 0.05778318 0.035740922 0.11146095 0.2884798 0.07310076 #> 12 0.103543341 0.07799259 0.045375827 0.19518888 0.3354080 0.03413656 #> 13 0.122547745 0.07905124 0.056084315 0.22437598 0.3511708 0.01840390 #> 14 0.042990591 0.03618335 0.241811837 0.55982776 0.1428372 0.01989756 #> 15 0.035609053 0.04022968 0.198176675 0.53973883 0.1462975 0.02215971 #> 16 0.056246994 0.05184498 0.201352298 0.51523810 0.1832397 0.00742115 #> 17 0.007716049 0.01049383 0.009876543 0.02777778 0.1929470 0.21968254 #> 18 0.014640428 0.04454602 0.042890320 0.17341352 0.2651538 0.06763669 #> 19 0.019591245 0.03668466 0.031845637 0.12592768 0.1422725 0.26011417 #> 20 0.037623741 0.03453783 0.185726965 0.58476297 0.1168700 0.05905666 #> Hyporadi Juncarti Juncbufo Lolipere Planlanc Poaprat Poatriv #> 1 0.07702746 0.14794933 0.1987270 0.9226190 0.40103107 0.9863946 0.8826329 #> 2 0.07454127 0.13017869 0.2070478 0.8272395 0.40700777 0.8972046 0.8288385 #> 3 0.05562332 0.22291082 0.2544828 0.7205525 0.27933493 0.8083020 0.8383185 #> 4 0.06985986 0.21320122 0.2318440 0.7197924 0.25797285 0.7940926 0.8197302 #> 5 0.10245961 0.10406655 0.2164230 0.8380779 0.52628928 0.9035899 0.8094632 #> 6 0.11463153 0.11631772 0.2166255 0.8000021 0.58765018 0.8666677 0.7782619 #> 7 0.10837376 0.11293676 0.2110045 0.8053380 0.51905808 0.8925059 0.8018775 #> 8 0.06550319 0.33219882 0.2323566 0.5403355 0.20596764 0.6160461 0.7101299 #> 9 0.05343787 0.23134366 0.2675624 0.6874068 0.25274756 0.7523318 0.8247374 #> 10 0.13692492 0.09080902 0.1678040 0.8102783 0.52588347 0.8915882 0.7543592 #> 11 0.18108995 0.13478872 0.1656396 0.7180948 0.47012501 0.8062720 0.6404351 #> 12 0.06777311 0.27306206 0.3231724 0.5875943 0.22110550 0.6932541 0.8199960 #> 13 0.04250245 0.28204736 0.3339728 0.5714581 0.18153869 0.7063028 0.7993754 #> 14 0.04665747 0.46685537 0.1206518 0.3356311 0.14342002 0.3817081 0.5090703 #> 15 0.05040404 0.51561767 0.1370235 0.3689922 0.13523214 0.4078219 0.5263520 #> 16 0.01731602 0.54304667 0.1776781 0.3561752 0.07269979 0.4124222 0.6071083 #> 17 0.36492870 0.03333333 0.1038156 0.5858415 0.59641331 0.7434618 0.5036834 #> 18 0.17491099 0.18956922 0.1376386 0.7124388 0.45087176 0.7368632 0.5859071 #> 19 0.39145281 0.13543701 0.1127832 0.4289185 0.40784415 0.5548077 0.3605827 #> 20 0.07795311 0.53056145 0.1192488 0.3262685 0.13059496 0.3662817 0.4523029 #> Ranuflam Rumeacet Sagiproc Salirepe Scorautu Trifprat Trifrepe #> 1 0.08105273 0.3160963 0.3371121 0.02729885 0.8898317 0.21701279 0.8782576 #> 2 0.13042865 0.3031318 0.3302063 0.05983781 0.9349640 0.20673650 0.9125666 #> 3 0.22632936 0.2909068 0.4204104 0.06065155 0.9036443 0.14654749 0.8817430 #> 4 0.21909541 0.2610006 0.4191908 0.07579199 0.9204237 0.12896524 0.8943213 #> 5 0.08063087 0.3979230 0.2612828 0.07589611 0.9576838 0.34808957 0.9142360 #> 6 0.10909966 0.4330705 0.2539380 0.08921540 0.9590466 0.35423465 0.9110822 #> 7 0.10541081 0.4113622 0.2954682 0.07094548 0.9550487 0.32489503 0.9171688 #> 8 0.40134447 0.2331043 0.4009544 0.11569906 0.8755515 0.09897600 0.8002526 #> 9 0.26006489 0.3464870 0.4531178 0.07351827 0.9145996 0.16269563 0.8714833 #> 10 0.10355742 0.3226025 0.2732735 0.09037489 0.9568824 0.26807372 0.9003730 #> 11 0.13269569 0.2753878 0.3673397 0.16465286 0.9442707 0.19982976 0.8979262 #> 12 0.29873222 0.3507140 0.5122033 0.08041977 0.9377963 0.13849854 0.9079979 #> 13 0.33309468 0.3107471 0.5131337 0.06572594 0.9255312 0.11199114 0.8841739 #> 14 0.64674225 0.1241545 0.2528665 0.15917563 0.8477706 0.06176123 0.6485949 #> 15 0.64449081 0.1459458 0.3151199 0.17750323 0.8430677 0.05831084 0.7170446 #> 16 0.66893881 0.1508409 0.3480368 0.15783292 0.8131968 0.03916718 0.6776273 #> 17 0.03549383 0.2913631 0.3292030 0.25651777 0.9839744 0.27593101 0.8141660 #> 18 0.18805395 0.2668100 0.3154533 0.17191937 0.9554011 0.20461193 0.8600701 #> 19 0.14551892 0.1831168 0.4798245 0.36429493 0.9902041 0.11966159 0.8147968 #> 20 0.62796060 0.1098600 0.3105433 0.21674174 0.8457313 0.03708580 0.6350394 #> Vicilath Bracruta Callcusp #> 1 0.17244420 0.7476589 0.003527337 #> 2 0.18494940 0.7415172 0.034597921 #> 3 0.12833142 0.7666969 0.075789630 #> 4 0.12550967 0.7919786 0.081110164 #> 5 0.16693075 0.8079786 0.023129027 #> 6 0.18035860 0.8387650 0.040981168 #> 7 0.19027523 0.8089116 0.024070054 #> 8 0.10213052 0.8109194 0.201958942 #> 9 0.08630413 0.7972178 0.092982775 #> 10 0.23383453 0.7660374 0.033527777 #> 11 0.24317802 0.8182692 0.043950322 #> 12 0.08049055 0.8061715 0.100954127 #> 13 0.06604026 0.7465509 0.122856392 #> 14 0.07857237 0.7238162 0.347514804 #> 15 0.07370069 0.7997141 0.381379395 #> 16 0.03353260 0.8029953 0.364128496 #> 17 0.20728700 0.7635487 0.009876543 #> 18 0.26222869 0.8397471 0.109959916 #> 19 0.18188455 0.8161275 0.082361157 #> 20 0.09111967 0.8397124 0.397924041 ## Uses abundance information. ## Vector with beals smoothing values corresponding to the first species ## in dune. beals(dune, species=1, include=TRUE) #> 1 2 3 4 5 6 7 8 #> 0.5923077 0.5032372 0.3499038 0.3306953 0.5944041 0.5928780 0.5824352 0.2082532 #> 9 10 11 12 13 14 15 16 #> 0.2960799 0.5462492 0.3659392 0.2610043 0.1982372 0.0922619 0.1140625 0.1066506 #> 17 18 19 20 #> 0.6020408 0.3844577 0.2865741 0.0750000"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Multivariate homogeneity of groups dispersions (variances) — betadisper","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Implements Marti Anderson's PERMDISP2 procedure analysis multivariate homogeneity group dispersions (variances). betadisper multivariate analogue Levene's test homogeneity variances. Non-euclidean distances objects group centres (centroids medians) handled reducing original distances principal coordinates. procedure latterly used means assessing beta diversity. anova, scores, plot boxplot methods. TukeyHSD.betadisper creates set confidence intervals differences mean distance--centroid levels grouping factor specified family-wise probability coverage. intervals based Studentized range statistic, Tukey's 'Honest Significant Difference' method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"betadisper(d, group, type = c(\"median\",\"centroid\"), bias.adjust = FALSE, sqrt.dist = FALSE, add = FALSE) # S3 method for betadisper anova(object, ...) # S3 method for betadisper scores(x, display = c(\"sites\", \"centroids\"), choices = c(1,2), ...) # S3 method for betadisper eigenvals(x, ...) # S3 method for betadisper plot(x, axes = c(1,2), cex = 0.7, pch = seq_len(ng), col = NULL, lty = \"solid\", lwd = 1, hull = TRUE, ellipse = FALSE, conf, segments = TRUE, seg.col = \"grey\", seg.lty = lty, seg.lwd = lwd, label = TRUE, label.cex = 1, ylab, xlab, main, sub, ...) # S3 method for betadisper boxplot(x, ylab = \"Distance to centroid\", ...) # S3 method for betadisper TukeyHSD(x, which = \"group\", ordered = FALSE, conf.level = 0.95, ...) # S3 method for betadisper print(x, digits = max(3, getOption(\"digits\") - 3), neigen = 8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"d distance structure returned dist, betadiver vegdist. group vector describing group structure, usually factor object can coerced factor using .factor. Can consist factor single level (.e., one group). type type analysis perform. Use spatial median group centroid? spatial median now default. bias.adjust logical: adjust small sample bias beta diversity estimates? sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. display character; partial match access scores \"sites\" \"species\". object, x object class \"betadisper\", result call betadisper. choices, axes principal coordinate axes wanted. hull logical; convex hull group plotted? ellipse logical; standard deviation data ellipse group plotted? conf Expected fractions data coverage data ellipses, e.g. 0.95. default draw 1 standard deviation data ellipse, supplied, conf multiplied corresponding value found Chi-squared distribution 2df provide requested coverage (probability contour). pch plot symbols groups, vector length equal number groups. col colors plot symbols centroid labels groups, vector length equal number groups. lty, lwd linetype, linewidth convex hulls confidence ellipses. segments logical; segments joining points centroid drawn? seg.col colour draw segments points centroid. Can vector, case one colour per group. seg.lty, seg.lwd linetype line width segments. label logical; centroids labelled respective factor label? label.cex numeric; character expansion centroid labels. cex, ylab, xlab, main, sub graphical parameters. details, see plot.default. character vector listing terms fitted model intervals calculated. Defaults grouping factor. ordered logical; see TukeyHSD. conf.level numeric value zero one giving family-wise confidence level use. digits, neigen numeric; print method, sets number digits use (per print.default) maximum number axes display eigenvalues , repsectively. ... arguments, including graphical parameters ( plot.betadisper boxplot.betadisper), passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"One measure multivariate dispersion (variance) group samples calculate average distance group members group centroid spatial median (referred 'centroid' now unless stated otherwise) multivariate space. test dispersions (variances) one groups different, distances group members group centroid subject ANOVA. multivariate analogue Levene's test homogeneity variances distances group members group centroids Euclidean distance. However, better measures distance Euclidean distance available ecological data. can accommodated reducing distances produced using dissimilarity coefficient principal coordinates, embeds within Euclidean space. analysis proceeds calculating Euclidean distances group members group centroid basis principal coordinate axes rather original distances. Non-metric dissimilarity coefficients can produce principal coordinate axes negative Eigenvalues. correspond imaginary, non-metric part distance objects. negative Eigenvalues produced, must correct imaginary distances. distance centroid point $$z_{ij}^c = \\sqrt{\\Delta^2(u_{ij}^+, c_i^+) - \\Delta^2(u_{ij}^-, c_i^-)},$$ \\(\\Delta^2\\) squared Euclidean distance \\(u_{ij}\\), principal coordinate \\(j\\)th point \\(\\)th group, \\(c_i\\), coordinate centroid \\(\\)th group. super-scripted ‘\\(+\\)’ ‘\\(-\\)’ indicate real imaginary parts respectively. equation (3) Anderson (2006). imaginary part greater magnitude real part, taking square root negative value, resulting NaN, cases changed zero distances (warning). line behaviour Marti Anderson's PERMDISP2 programme. test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance \\(F\\). alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution \\(F\\) Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can also performed using permutest.betadisper. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper. simple wrapper TukeyHSD. user directed read help file TukeyHSD using function. particular, note statement using function unbalanced designs. results analysis can visualised using plot boxplot methods. One additional use functions assessing beta diversity (Anderson et al 2006). Function betadiver provides popular dissimilarity measures purpose. noted passing Anderson (2006) related context O'Neill (2000), estimates dispersion around central location (median centroid) calculated data biased downward. bias matters comparing diversity among treatments small, unequal numbers samples. Setting bias.adjust=TRUE using betadisper imposes \\(\\sqrt{n/(n-1)}\\) correction (Stier et al. 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"anova method returns object class \"anova\" inheriting class \"data.frame\". scores method returns list one components \"sites\" \"centroids\". plot function invisibly returns object class \"ordiplot\", plotting structure can used identify.ordiplot (identify points) functions ordiplot family. boxplot function invisibly returns list whose components documented boxplot. eigenvals.betadisper returns named vector eigenvalues. TukeyHSD.betadisper returns list. See TukeyHSD details. betadisper returns list class \"betadisper\" following components: eig numeric; eigenvalues principal coordinates analysis. vectors matrix; eigenvectors principal coordinates analysis. distances numeric; Euclidean distances principal coordinate space samples respective group centroid median. group factor; vector describing group structure centroids matrix; locations group centroids medians principal coordinates. group.distances numeric; mean distance group centroid median. call matched function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"group consists single level group, anova permutest methods appropriate used data stop error. Missing values either d group removed prior performing analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Stewart Schultz noticed permutation test type=\"centroid\" wrong type error anti-conservative. , default type changed \"median\", uses spatial median group centroid. Tests suggests permutation test type analysis gives correct error rates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62, 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. O'Neill, M.E. (2000) Weighted Least Squares Approach Levene's Test Homogeneity Variance. Australian & New Zealand Journal Statistics 42, 81-–100. Stier, .C., Geange, S.W., Hanson, K.M., & Bolker, B.M. (2013) Predator density timing arrival affect reef fish community assembly. Ecology 94, 1057--1068.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Gavin L. Simpson; bias correction Adrian Stier Ben Bolker.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F permutest(mod, pairwise = TRUE, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 99 0.07 . #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Pairwise comparisons: #> (Observed p-value below diagonal, permuted p-value above diagonal) #> grazed ungrazed #> grazed 0.08 #> ungrazed 0.04295 ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Plot the groups and distances to centroids on the ## first two PCoA axes plot(mod) ## with data ellipses instead of hulls plot(mod, ellipse = TRUE, hull = FALSE) # 1 sd data ellipse plot(mod, ellipse = TRUE, hull = FALSE, conf = 0.90) # 90% data ellipse # plot with manual colour specification my_cols <- c(\"#1b9e77\", \"#7570b3\") plot(mod, col = my_cols, pch = c(16,17), cex = 1.1) ## can also specify which axes to plot, ordering respected plot(mod, axes = c(3,1), seg.col = \"forestgreen\", seg.lty = \"dashed\") ## Draw a boxplot of the distances to centroid for each group boxplot(mod) ## `scores` and `eigenvals` also work scrs <- scores(mod) str(scrs) #> List of 2 #> $ sites : num [1:24, 1:2] 0.0946 -0.3125 -0.3511 -0.3291 -0.1926 ... #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:24] \"18\" \"15\" \"24\" \"27\" ... #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" #> $ centroids: num [1:2, 1:2] -0.1455 0.2786 0.0758 -0.2111 #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:2] \"grazed\" \"ungrazed\" #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" head(scores(mod, 1:4, display = \"sites\")) #> PCoA1 PCoA2 PCoA3 PCoA4 #> 18 0.09459373 0.15914576 0.074400844 -0.202466025 #> 15 -0.31248809 0.10032751 -0.062243360 0.110844864 #> 24 -0.35106507 -0.05954096 -0.038079447 0.095060928 #> 27 -0.32914546 -0.17019348 0.231623720 0.019110623 #> 23 -0.19259443 -0.01459250 -0.005679372 -0.209718312 #> 19 -0.06794575 -0.14501690 -0.085645653 0.002431355 # group centroids/medians scores(mod, 1:4, display = \"centroids\") #> PCoA1 PCoA2 PCoA3 PCoA4 #> grazed -0.1455200 0.07584572 -0.01366220 -0.0178990 #> ungrazed 0.2786095 -0.21114993 -0.03475586 0.0220129 # eigenvalues from the underlying principal coordinates analysis eigenvals(mod) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 #> 1.7552165 1.1334455 0.4429018 0.3698054 0.2453532 0.1960921 0.1751131 #> PCoA8 PCoA9 PCoA10 PCoA11 PCoA12 PCoA13 PCoA14 #> 0.1284467 0.0971594 0.0759601 0.0637178 0.0583225 0.0394934 0.0172699 #> PCoA15 PCoA16 PCoA17 PCoA18 PCoA19 PCoA20 PCoA21 #> 0.0051011 -0.0004131 -0.0064654 -0.0133147 -0.0253944 -0.0375105 -0.0480069 #> PCoA22 PCoA23 #> -0.0537146 -0.0741390 ## try out bias correction; compare with mod3 (mod3B <- betadisper(dis, groups, type = \"median\", bias.adjust=TRUE)) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"median\", bias.adjust #> = TRUE) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.4055 0.2893 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 anova(mod3B) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 0.06468 . #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 permutest(mod3B, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 99 0.05 * #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## should always work for a single group group <- factor(rep(\"grazed\", NROW(varespec))) (tmp <- betadisper(dis, group, type = \"median\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"median\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed #> 0.4255 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 (tmp <- betadisper(dis, group, type = \"centroid\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"centroid\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to centroid: #> grazed #> 0.4261 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## simulate missing values in 'd' and 'group' ## using spatial medians groups[c(2,20)] <- NA dis[c(2, 20)] <- NA mod2 <- betadisper(dis, groups) ## messages #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod2 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to median: #> grazed ungrazed #> 0.3984 0.3008 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod2, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 99 0.16 #> Residuals 18 0.296910 0.016495 anova(mod2) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 0.1369 #> Residuals 18 0.296910 0.016495 plot(mod2) boxplot(mod2) plot(TukeyHSD(mod2)) ## Using group centroids mod3 <- betadisper(dis, groups, type = \"centroid\") #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod3 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"centroid\") #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to centroid: #> grazed ungrazed #> 0.4001 0.3108 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod3, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 99 0.12 #> Residuals 18 0.189749 0.010542 anova(mod3) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 0.09166 . #> Residuals 18 0.189749 0.010542 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod3) boxplot(mod3) plot(TukeyHSD(mod3))"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of beta Diversity — betadiver","title":"Indices of beta Diversity — betadiver","text":"function estimates 24 indices beta diversity reviewed Koleff et al. (2003). Alternatively, finds co-occurrence frequencies triangular plots (Koleff et al. 2003).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of beta Diversity — betadiver","text":"","code":"betadiver(x, method = NA, order = FALSE, help = FALSE, ...) # S3 method for betadiver plot(x, ...) # S3 method for betadiver scores(x, triangular = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of beta Diversity — betadiver","text":"x Community data matrix, betadiver result plot scores functions. method index beta diversity defined Koleff et al. (2003), Table 1. can use either subscript \\(\\beta\\) number index. See argument help . order Order sites increasing number species. influence configuration triangular plot non-symmetric indices. help Show numbers, subscript names defining equations indices exit. triangular Return scores suitable triangular plotting proportions. FALSE, returns 3-column matrix raw counts. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of beta Diversity — betadiver","text":"commonly used index beta diversity \\(\\beta_w = S/\\alpha - 1\\), \\(S\\) total number species, \\(\\alpha\\) average number species per site (Whittaker 1960). drawback model \\(S\\) increases sample size, expectation \\(\\alpha\\) remains constant, beta diversity increases sample size. solution problem study beta diversity pairs sites (Marion et al. 2017). denote number species shared two sites \\(\\) numbers unique species (shared) \\(b\\) \\(c\\), \\(S = + b + c\\) \\(\\alpha = (2 + b + c)/2\\) \\(\\beta_w = (b+c)/(2 + b + c)\\). Sørensen dissimilarity defined vegan function vegdist argument binary = TRUE. Many indices dissimilarity indices well. Function betadiver finds indices reviewed Koleff et al. (2003). indices found function designdist, current function provides conventional shortcut. function finds indices. proper analysis must done functions betadisper, adonis2 mantel. indices directly taken Table 1 Koleff et al. (2003), can selected either index number subscript name used Koleff et al. numbers, names defining equations can seen using betadiver(help = TRUE). cases two alternative forms, one term \\(-1\\) used. several duplicate indices, number distinct alternatives much lower 24 formally provided. formulations used functions differ occasionally Koleff et al. (2003), still mathematically equivalent. method = NA, index calculated, instead object class betadiver returned. list elements , b c. Function plot can used display proportions elements triangular plot suggested Koleff et al. (2003), scores extracts triangular coordinates raw scores. Function plot returns invisibly triangular coordinates \"ordiplot\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of beta Diversity — betadiver","text":"method = NA, function returns object class \"betadisper\" elements , b, c. method specified, function returns \"dist\" object can used function analysing dissimilarities. beta diversity, particularly useful functions betadisper study betadiversity groups, adonis2 model, mantel compare beta diversities dissimilarities distances (including geographical distances). Although betadiver returns \"dist\" object, indices similarities used place dissimilarities, user error. Functions 10 (\"j\"), 11 (\"sor\") 21 (\"rlb\") similarity indices. Function sets argument \"maxdist\" similarly vegdist, using NA fixed upper limit, 0 similarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of beta Diversity — betadiver","text":"Baselga, . (2010) Partitioning turnover nestedness components beta diversity. Global Ecology Biogeography 19, 134--143. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence-absence data. Journal Animal Ecology 72, 367--382. Marion, Z.H., Fordyce, J.. Fitzpatrick, B.M. (2017) Pairwise beta diversity resolves underappreciated source confusion calculating species turnover. Ecology 98, 933--939. Whittaker, R.H. (1960) Vegetation Siskiyou mountains, Oregon California. Ecological Monographs 30, 279--338.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of beta Diversity — betadiver","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Indices of beta Diversity — betadiver","text":"indices return similarities instead dissimilarities.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of beta Diversity — betadiver","text":"","code":"## Raw data and plotting data(sipoo) m <- betadiver(sipoo) plot(m) ## The indices betadiver(help=TRUE) #> 1 \"w\" = (b+c)/(2*a+b+c) #> 2 \"-1\" = (b+c)/(2*a+b+c) #> 3 \"c\" = (b+c)/2 #> 4 \"wb\" = b+c #> 5 \"r\" = 2*b*c/((a+b+c)^2-2*b*c) #> 6 \"I\" = log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c) #> 7 \"e\" = exp(log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c))-1 #> 8 \"t\" = (b+c)/(2*a+b+c) #> 9 \"me\" = (b+c)/(2*a+b+c) #> 10 \"j\" = a/(a+b+c) #> 11 \"sor\" = 2*a/(2*a+b+c) #> 12 \"m\" = (2*a+b+c)*(b+c)/(a+b+c) #> 13 \"-2\" = pmin.int(b,c)/(pmax.int(b,c)+a) #> 14 \"co\" = (a*c+a*b+2*b*c)/(2*(a+b)*(a+c)) #> 15 \"cc\" = (b+c)/(a+b+c) #> 16 \"g\" = (b+c)/(a+b+c) #> 17 \"-3\" = pmin.int(b,c)/(a+b+c) #> 18 \"l\" = (b+c)/2 #> 19 \"19\" = 2*(b*c+1)/(a+b+c)/(a+b+c-1) #> 20 \"hk\" = (b+c)/(2*a+b+c) #> 21 \"rlb\" = a/(a+c) #> 22 \"sim\" = pmin.int(b,c)/(pmin.int(b,c)+a) #> 23 \"gl\" = 2*abs(b-c)/(2*a+b+c) #> 24 \"z\" = (log(2)-log(2*a+b+c)+log(a+b+c))/log(2) ## The basic Whittaker index d <- betadiver(sipoo, \"w\") ## This should be equal to Sorensen index (binary Bray-Curtis in ## vegan) range(d - vegdist(sipoo, binary=TRUE)) #> [1] 0 0"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":null,"dir":"Reference","previous_headings":"","what":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function computes coefficients dispersal direction geographically connected areas, defined Legendre Legendre (1984), also described Legendre Legendre (2012, section 13.3.4).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"bgdispersal(mat, PAonly = FALSE, abc = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"mat Data frame matrix containing community composition data table (species presence-absence abundance data). PAonly FALSE four types coefficients, DD1 DD4, requested; TRUE DD1 DD2 sought (see Details). abc TRUE, return tables , b c used DD1 DD2.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"signs DD coefficients indicate direction dispersal, provided asymmetry significant. positive sign indicates dispersal first (row DD tables) second region (column); negative sign indicates opposite. McNemar test asymmetry computed presence-absence data test hypothesis significant asymmetry two areas comparison. input data table, rows sites areas, columns taxa. often, taxa species, coefficients can computed genera families well. DD1 DD2 computed presence-absence data. four types coefficients computed quantitative data, converted presence-absence computation DD1 DD2. PAonly = FALSE indicates four types coefficients requested. PAonly = TRUE DD1 DD2 sought.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Function bgdispersal returns list containing following matrices: DD1 \\(DD1_{j,k} = ((b - c))/((+ b + c)^2)\\) DD2 \\(DD2_{j,k} = (2 (b - c))/((2a + b + c) (+ b + c))\\) \\(\\), \\(b\\), \\(c\\) meaning computation binary similarity coefficients. DD3 \\(DD3_{j,k} = {W(-B) / (+B-W)^2} \\) DD4 \\(DD4_{j,k} = 2W(-B) / ((+B)(+B-W))\\) W = sum(pmin(vector1, vector2)), = sum(vector1), B = sum(vector2) McNemar McNemar chi-square statistic asymmetry (Sokal Rohlf 1995): \\(2(b \\log(b) + c \\log(c) - (b+c) \\log((b+c)/2)) / q\\), \\(q = 1 + 1/(2(b+c))\\) (Williams correction continuity) prob.McNemar probabilities associated McNemar statistics, chi-square test. H0: asymmetry \\((b-c)\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Legendre, P. V. Legendre. 1984. Postglacial dispersal freshwater fishes Québec peninsula. Can. J. Fish. Aquat. Sci. 41: 1781-1802. Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Sokal, R. R. F. J. Rohlf. 1995. Biometry. principles practice statistics biological research. 3rd edn. W. H. Freeman, New York.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function uses powerful alternative McNemar test classical formula. classical formula constructed spirit Pearson's Chi-square, formula function constructed spirit Wilks Chi-square \\(G\\) statistic. Function mcnemar.test uses classical formula. new formula introduced vegan version 1.10-11, older implementations bgdispersal used classical formula.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"mat <- matrix(c(32,15,14,10,70,30,100,4,10,30,25,0,18,0,40, 0,0,20,0,0,0,0,4,0,30,20,0,0,0,0,25,74,42,1,45,89,5,16,16,20), 4, 10, byrow=TRUE) bgdispersal(mat) #> $DD1 #> [,1] [,2] [,3] [,4] #> [1,] 0.00 0.24 0.21 0.00 #> [2,] -0.24 0.00 0.08 -0.24 #> [3,] -0.21 -0.08 0.00 -0.21 #> [4,] 0.00 0.24 0.21 0.00 #> #> $DD2 #> [,1] [,2] [,3] [,4] #> [1,] 0.0000000 0.3428571 0.3230769 0.0000000 #> [2,] -0.3428571 0.0000000 0.1142857 -0.3428571 #> [3,] -0.3230769 -0.1142857 0.0000000 -0.3230769 #> [4,] 0.0000000 0.3428571 0.3230769 0.0000000 #> #> $DD3 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.1567922 0.1420408 -0.01325831 #> [2,] -0.15679216 0.0000000 0.1101196 -0.20049485 #> [3,] -0.14204082 -0.1101196 0.0000000 -0.13586560 #> [4,] 0.01325831 0.2004949 0.1358656 0.00000000 #> #> $DD4 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.2513176 0.2425087 -0.01960102 #> [2,] -0.25131757 0.0000000 0.1725441 -0.30993929 #> [3,] -0.24250871 -0.1725441 0.0000000 -0.23381521 #> [4,] 0.01960102 0.3099393 0.2338152 0.00000000 #> #> $McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 7.677938 9.0571232 0.000000 #> [2,] NA NA 0.2912555 7.677938 #> [3,] NA NA NA 9.057123 #> [4,] NA NA NA NA #> #> $prob.McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 0.005590001 0.002616734 1.000000000 #> [2,] NA NA 0.589417103 0.005590001 #> [3,] NA NA NA 0.002616734 #> [4,] NA NA NA NA #>"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":null,"dir":"Reference","previous_headings":"","what":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Function finds best subset environmental variables, Euclidean distances scaled environmental variables maximum (rank) correlation community dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# S3 method for default bioenv(comm, env, method = \"spearman\", index = \"bray\", upto = ncol(env), trace = FALSE, partial = NULL, metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), parallel = getOption(\"mc.cores\"), ...) # S3 method for formula bioenv(formula, data, ...) bioenvdist(x, which = \"best\")"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"comm Community data frame dissimilarity object square matrix can interpreted dissimilarities. env Data frame continuous environmental variables. method correlation method used cor. index dissimilarity index used community data (comm) vegdist. ignored comm dissimilarities. upto Maximum number parameters studied subsets. formula, data Model formula data. trace Trace calculations partial Dissimilarities partialled inspecting variables env. metric Metric used distances environmental distances. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. x bioenv result object. number model environmental distances evaluated, \"best\" model. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function calculates community dissimilarity matrix using vegdist. selects possible subsets environmental variables, scales variables, calculates Euclidean distances subset using dist. function finds correlation community dissimilarities environmental distances, size subsets, saves best result. \\(2^p-1\\) subsets \\(p\\) variables, exhaustive search may take , , long time (parameter upto offers partial relief). argument metric defines distances given set environmental variables. metric = \"euclidean\", variables scaled unit variance Euclidean distances calculated. metric = \"mahalanobis\", Mahalanobis distances calculated: addition scaling unit variance, matrix current set environmental variables also made orthogonal (uncorrelated). metric = \"manhattan\", variables scaled unit range Manhattan distances calculated, distances sums differences environmental variables. metric = \"gower\", Gower distances calculated using function daisy. allows also using factor variables, continuous variables results equal metric = \"manhattan\". function can called model formula LHS data matrix RHS lists environmental variables. formula interface practical selecting transforming environmental variables. argument partial can perform “partial” analysis. partializing item must dissimilarity object class dist. partial item can used correlation method, strictly correct Pearson. Function bioenvdist recalculates environmental distances used within function. default calculate distances best model, number model can given. Clarke & Ainsworth (1993) suggested method used selecting best subset environmental variables interpreting results nonmetric multidimensional scaling (NMDS). recommended parallel display NMDS community dissimilarities NMDS Euclidean distances best subset scaled environmental variables. warned use Procrustes analysis, looks like good way comparing two ordinations. Clarke & Ainsworth wrote computer program BIO-ENV giving name current function. Presumably BIO-ENV later incorporated Clarke's PRIMER software (available Windows). addition, Clarke & Ainsworth suggested novel method rank correlation available current function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function returns object class bioenv summary method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Clarke, K. R & Ainsworth, M. 1993. method linking multivariate community structure environmental variables. Marine Ecology Progress Series, 92, 205--219.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"want study ‘significance’ bioenv results, can use function mantel mantel.partial use definition correlation. However, bioenv standardizes environmental variables depending used metric, must mantel comparable results (standardized data returned item x result object). safest use bioenvdist extract environmental distances really used within bioenv. NB., bioenv selects variables maximize Mantel correlation, significance tests based priori selection variables biased.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# The method is very slow for large number of possible subsets. # Therefore only 6 variables in this example. data(varespec) data(varechem) sol <- bioenv(wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, varechem) sol #> #> Call: #> bioenv(formula = wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, data = varechem) #> #> Subset of environmental variables with best correlation to community data. #> #> Correlations: spearman #> Dissimilarities: bray #> Metric: euclidean #> #> Best model has 3 parameters (max. 6 allowed): #> P Ca Al #> with correlation 0.4004806 #> ## IGNORE_RDIFF_BEGIN summary(sol) #> size correlation #> P 1 0.2513 #> P Al 2 0.4004 #> P Ca Al 3 0.4005 #> P Ca pH Al 4 0.3619 #> log(N) P Ca pH Al 5 0.3216 #> log(N) P K Ca pH Al 6 0.2822 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA biplot — biplot.rda","title":"PCA biplot — biplot.rda","text":"Draws PCA biplot species scores indicated biplot arrows","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA biplot — biplot.rda","text":"","code":"# S3 method for rda biplot(x, choices = c(1, 2), scaling = \"species\", display = c(\"sites\", \"species\"), type, xlim, ylim, col = c(1,2), const, correlation = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA biplot — biplot.rda","text":"x rda result object. choices Axes show. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. negative scaling values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. display Scores shown. must alternatives \"species\" species scores, /\"sites\" site scores. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. Can length 2 (e.g. type = c(\"text\", \"points\")), case first element describes species scores handled, second site scores drawn. xlim, ylim x y limits (min, max) plot. col Colours used sites species (order). one colour given, used . const General scaling constant scores.rda. ... parameters plotting functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA biplot — biplot.rda","text":"Produces plot biplot results call rda. common \"species\" scores PCA drawn biplot arrows point direction increasing values variable. biplot.rda function provides wrapper plot.cca allow easy production plot. biplot.rda suitable unconstrained models. used ordination object constraints, error issued. species scores drawn using \"text\", arrows drawn origin 0.85 * species score, whilst labels drawn species score. type used \"points\", labels drawn therefore arrows drawn origin actual species score.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"PCA biplot — biplot.rda","text":"plot function returns invisibly plotting structure can used identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"PCA biplot — biplot.rda","text":"Gavin Simpson, based plot.cca Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA biplot — biplot.rda","text":"","code":"data(dune) mod <- rda(dune, scale = TRUE) biplot(mod, scaling = \"symmetric\") ## different type for species and site scores biplot(mod, scaling = \"symmetric\", type = c(\"text\", \"points\")) ## We can use ordiplot pipes in R 4.1 to build similar plots with ## flexible control if (FALSE) { if (getRversion() >= \"4.1\") { plot(mod, scaling = \"symmetric\", type=\"n\") |> text(\"sites\", cex=0.8) |> text(\"species\", arrows=TRUE, length=0.02, col=\"red\", cex=0.6) } }"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":null,"dir":"Reference","previous_headings":"","what":"K-means partitioning using a range of values of K — cascadeKM","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function wrapper kmeans function. creates several partitions forming cascade small large number groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"cascadeKM(data, inf.gr, sup.gr, iter = 100, criterion = \"calinski\", parallel = getOption(\"mc.cores\")) cIndexKM(y, x, index = \"all\") # S3 method for cascadeKM plot(x, min.g, max.g, grpmts.plot = TRUE, sortg = FALSE, gridcol = NA, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"K-means partitioning using a range of values of K — cascadeKM","text":"data data matrix. objects (samples) rows. inf.gr number groups partition smallest number groups cascade (min). sup.gr number groups partition largest \t number groups cascade (max). iter number random starting configurations value \\(K\\). criterion criterion used select best partition. default value \"calinski\", refers Calinski-Harabasz (1974) criterion. simple structure index (\"ssi\") also available. indices available package cclust. experience, two indices work best likely return maximum value near optimal number clusters \"calinski\" \"ssi\". y Object class \"kmeans\" returned clustering algorithm kmeans x Data matrix columns correspond variables rows observations, plotting object plot index available indices : \"calinski\" \"ssi\". Type \"\" obtain indices. Abbreviations names also accepted. min.g, max.g minimum maximum numbers groups displayed. grpmts.plot Show plot (TRUE FALSE). sortg Sort objects function group membership produce easily interpretable graph. See Details. original object names kept; used labels output table x, although graph. row names, sequential row numbers used keep track original order objects. gridcol colour grid lines plots. NA, default value, removes grid lines. ... parameters functions (ignored). parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function creates several partitions forming cascade small large number groups formed kmeans. work performed function cIndex based clustIndex package cclust). criteria removed version computation errors generated one object found group. default value \"calinski\", refers well-known Calinski-Harabasz (1974) criterion. available index simple structure index \"ssi\" (Dolnicar et al. 1999). case groups equal sizes, \"calinski\" generally good criterion indicate correct number groups. Users take indications literally groups equal size. Type \"\" obtain indices. indices defined : calinski: \\((SSB/(K-1))/(SSW/(n-K))\\), \\(n\\) number data points \\(K\\) number clusters. \\(SSW\\) sum squares within clusters \\(SSB\\) sum squares among clusters. index simply \\(F\\) (ANOVA) statistic. ssi: “Simple Structure Index” multiplicatively combines several elements influence interpretability partitioning solution. best partition indicated highest SSI value. simulation study, Milligan Cooper (1985) found Calinski-Harabasz criterion recovered correct number groups often. recommend criterion , groups equal sizes, maximum value \"calinski\" usually indicates correct number groups. Another available index simple structure index \"ssi\". Users take indications indices literally groups equal size explore groups corresponding values \\(K\\). Function cascadeKM plot method. Two plots produced. graph left objects abscissa number groups ordinate. groups represented colours. graph right shows values criterion (\"calinski\" \"ssi\") determining best partition. highest value criterion marked red. Points marked orange, , indicate partitions producing increase criterion value number groups increases; may represent interesting partitions. sortg=TRUE, objects reordered following procedure: (1) simple matching distance matrix computed among objects, based table K-means assignments groups, \\(K\\) = min.g \\(K\\) = max.g. (2) principal coordinate analysis (PCoA, Gower 1966) computed centred distance matrix. (3) first principal coordinate used new order objects graph. simplified algorithm used compute first principal coordinate , using iterative algorithm described Legendre & Legendre (2012). full distance matrix among objects never computed; avoids problem storing number objects large. Distance values computed needed algorithm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Function cascadeKM returns object class cascadeKM items: partition Table partitions found different numbers groups \\(K\\), \\(K\\) = inf.gr \\(K\\) = sup.gr. results Values criterion select best partition. criterion name criterion used. size number objects found group, partitions (columns). Function cIndex returns vector index values. maximum value indices supposed indicate best partition. indices work best groups equal sizes. groups equal sizes, one put much faith maximum indices, also explore groups corresponding values \\(K\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Calinski, T. J. Harabasz. 1974. dendrite method cluster analysis. Commun. Stat. 3: 1--27. Dolnicar, S., K. Grabler J. . Mazanec. 1999. tale three cities: perceptual charting analyzing destination images. Pp. 39-62 : Woodside, . et al. [eds.] Consumer psychology tourism, hospitality leisure. CAB International, New York. Gower, J. C. 1966. distance properties latent root vector methods used multivariate analysis. Biometrika 53: 325--338. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Milligan, G. W. & M. C. Cooper. 1985. examination procedures determining number clusters data set. Psychometrika 50: 159--179. Weingessel, ., Dimitriadou, . Dolnicar, S. 2002. examination indexes determining number clusters binary data sets. Psychometrika 67: 137--160.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Marie-Helene Ouellette Marie-Helene.Ouellette@UMontreal.ca, Sebastien Durand Sebastien.Durand@UMontreal.ca Pierre Legendre Pierre.Legendre@UMontreal.ca. Parallel processing Virgilio Gómez-Rubio. Edited vegan Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"# Partitioning a (10 x 10) data matrix of random numbers mat <- matrix(runif(100),10,10) res <- cascadeKM(mat, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning an autocorrelated time series vec <- sort(matrix(runif(30),30,1)) res <- cascadeKM(vec, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning a large autocorrelated time series # Note that we remove the grid lines vec <- sort(matrix(runif(1000),1000,1)) res <- cascadeKM(vec, 2, 7, iter = 10, criterion = 'calinski') toto <- plot(res, gridcol=NA)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca performs correspondence analysis, optionally constrained correspondence analysis (.k.. canonical correspondence analysis), optionally partial constrained correspondence analysis. Function rda performs redundancy analysis, optionally principal components analysis. popular ordination techniques community ecology.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"# S3 method for formula cca(formula, data, na.action = na.fail, subset = NULL, ...) # S3 method for formula rda(formula, data, scale=FALSE, na.action = na.fail, subset = NULL, ...) # S3 method for default cca(X, Y, Z, ...) # S3 method for default rda(X, Y, Z, scale=FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"formula Model formula, left hand side gives community data matrix, right hand side gives constraining variables, conditioning variables can given within special function Condition. data Data frame containing variables right hand side model formula. X Community data matrix. Y Constraining matrix, typically environmental variables. Can missing. data.frame, expanded model.matrix factors expanded contrasts (“dummy variables”). better use formula instead argument, analyses work formula used. Z Conditioning matrix, effect removed (“partialled ”) next step. Can missing. data.frame, expanded similarly constraining matrix. scale Scale species unit variance (like correlations). na.action Handling missing values constraints conditions. default (na.fail) stop missing value. Choice na.omit removes rows missing values. Choice na.exclude keeps observations gives NA results calculated. WA scores rows may found also missing values constraints. Missing values never allowed dependent community data. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data. ... arguments print plot functions (ignored functions).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Since introduction (ter Braak 1986), constrained, canonical, correspondence analysis spin-, redundancy analysis, popular ordination methods community ecology. Functions cca rda similar popular proprietary software Canoco, although implementation completely different. functions based Legendre & Legendre's (2012) algorithm: cca Chi-square transformed data matrix subjected weighted linear regression constraining variables, fitted values submitted correspondence analysis performed via singular value decomposition (svd). Function rda similar, uses ordinary, unweighted linear regression unweighted SVD. Legendre & Legendre (2012), Table 11.5 (p. 650) give skeleton RDA algorithm vegan. algorithm CCA similar, involves standardization row column weights. functions can called either matrix-like entries community data constraints, formula interface. general, formula interface preferred, allows better control model allows factor constraints. analyses ordination results possible model fitted formula (e.g., cases anova.cca, automatic model building). following sections, X, Y Z, although referred matrices, commonly data frames. matrix interface, community data matrix X must given, data matrices may omitted, corresponding stage analysis skipped. matrix Z supplied, effects removed community matrix, residual matrix submitted next stage. called partial correspondence redundancy analysis. matrix Y supplied, used constrain ordination, resulting constrained canonical correspondence analysis, redundancy analysis. Finally, residual submitted ordinary correspondence analysis (principal components analysis). matrices Z Y missing, data matrix analysed ordinary correspondence analysis ( principal components analysis). Instead separate matrices, model can defined using model formula. left hand side must community data matrix (X). right hand side defines constraining model. constraints can contain ordered unordered factors, interactions among variables functions variables. defined contrasts honoured factor variables. constraints can also matrices (data frames). formula can include special term Condition conditioning variables (“covariables”) partialled analysis. following commands equivalent: cca(X, Y, Z), cca(X ~ Y + Condition(Z)), Y Z refer constraints conditions matrices respectively. Constrained correspondence analysis indeed constrained method: CCA try display variation data, part can explained used constraints. Consequently, results strongly dependent set constraints transformations interactions among constraints. shotgun method use environmental variables constraints. However, exploratory problems better analysed unconstrained methods correspondence analysis (decorana, corresp) non-metric multidimensional scaling (metaMDS) environmental interpretation analysis (envfit, ordisurf). CCA good choice user clear strong priori hypotheses constraints interested major structure data set. CCA able correct curve artefact commonly found correspondence analysis forcing configuration linear constraints. However, curve artefact can avoided low number constraints curvilinear relation . curve can reappear even two badly chosen constraints single factor. Although formula interface makes easy include polynomial interaction terms, terms often produce curved artefacts (difficult interpret), probably avoided. According folklore, rda used ``short gradients'' rather cca. However, based research finds methods based Euclidean metric uniformly weaker based Chi-squared metric. However, standardized Euclidean distance may appropriate measures (see Hellinger standardization decostand particular). Partial CCA (pCCA; alternatively partial RDA) can used remove effect conditioning background random variables covariables CCA proper. fact, pCCA compares models cca(X ~ Z) cca(X ~ Y + Z) attributes difference effect Y cleansed effect Z. people used method extracting “components variance” CCA. However, effect variables together stronger sum separately, can increase total Chi-square partialling variation, give negative “components variance”. general, components “variance” trusted due interactions two sets variables. functions summary plot methods documented separately (see plot.cca, summary.cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca returns huge object class cca, described separately cca.object. Function rda returns object class rda inherits class cca described cca.object. scaling used rda scores described separate vignette package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"original method ter Braak, current implementation follows Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. McCune, B. (1997) Influence noisy environmental data canonical correspondence analysis. Ecology 78, 2617-2623. Palmer, M. W. (1993) Putting things even better order: advantages canonical correspondence analysis. Ecology 74,2215-2230. Ter Braak, C. J. F. (1986) Canonical Correspondence Analysis: new eigenvector technique multivariate direct gradient analysis. Ecology 67, 1167-1179.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"responsible author Jari Oksanen, code borrows heavily Dave Roberts (Montana State University, USA).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"data(varespec) data(varechem) ## Common but bad way: use all variables you happen to have in your ## environmental data matrix vare.cca <- cca(varespec, varechem) vare.cca #> Call: cca(X = varespec, Y = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Constrained 1.4415 0.6920 14 #> Unconstrained 0.6417 0.3080 9 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4389 0.2918 0.1628 0.1421 0.1180 0.0890 0.0703 0.0584 0.0311 0.0133 0.0084 #> CCA12 CCA13 CCA14 #> 0.0065 0.0062 0.0047 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 #> 0.19776 0.14193 0.10117 0.07079 0.05330 0.03330 0.01887 0.01510 0.00949 #> plot(vare.cca) ## Formula interface and a better model vare.cca <- cca(varespec ~ Al + P*(K + Baresoil), data=varechem) vare.cca #> Call: cca(formula = varespec ~ Al + P * (K + Baresoil), data = #> varechem) #> #> Inertia Proportion Rank #> Total 2.083 1.000 #> Constrained 1.046 0.502 6 #> Unconstrained 1.038 0.498 17 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 0.3756 0.2342 0.1407 0.1323 0.1068 0.0561 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.27577 0.15411 0.13536 0.11803 0.08887 0.05511 0.04919 0.03781 #> (Showing 8 of 17 unconstrained eigenvalues) #> plot(vare.cca) ## Partialling out and negative components of variance cca(varespec ~ Ca, varechem) #> Call: cca(formula = varespec ~ Ca, data = varechem) #> #> Inertia Proportion Rank #> Total 2.08320 1.00000 #> Constrained 0.15722 0.07547 1 #> Unconstrained 1.92598 0.92453 22 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.15722 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.4745 0.2939 0.2140 0.1954 0.1748 0.1171 0.1121 0.0880 #> (Showing 8 of 22 unconstrained eigenvalues) #> cca(varespec ~ Ca + Condition(pH), varechem) #> Call: cca(formula = varespec ~ Ca + Condition(pH), data = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Conditional 0.1458 0.0700 1 #> Constrained 0.1827 0.0877 1 #> Unconstrained 1.7547 0.8423 21 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.18269 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3834 0.2749 0.2123 0.1760 0.1701 0.1161 0.1089 0.0880 #> (Showing 8 of 21 unconstrained eigenvalues) #> ## RDA data(dune) data(dune.env) dune.Manure <- rda(dune ~ Manure, dune.env) plot(dune.Manure)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":null,"dir":"Reference","previous_headings":"","what":"Result Object from Constrained Ordination — cca.object","title":"Result Object from Constrained Ordination — cca.object","text":"Ordination methods cca, rda, dbrda capscale return similar result objects. methods use internal function ordConstrained. differ (1) initial transformation data defining inertia, (2) weighting, (3) use rectangular rows \\(\\times\\) columns data symmetric rows \\(\\times\\) rows dissimilarities: rda initializes data give variance correlations inertia, cca based double-standardized data give Chi-square inertia uses row column weights, capscale maps real part dissimilarities rectangular data performs RDA, dbrda performs RDA-like analysis directly symmetric dissimilarities. Function ordConstrained returns result components methods, calling function may add components final result. However, access result components directly (using $): internal structure regarded stable application interface (API), can change release. access results components directly, take risk breakage vegan release. vegan provides wide set accessor functions components, functions updated result object changes. documentation gives overview accessor functions cca result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Result Object from Constrained Ordination — cca.object","text":"","code":"ordiYbar(x, model = c(\"CCA\", \"CA\", \"pCCA\", \"partial\", \"initial\")) # S3 method for cca model.frame(formula, ...) # S3 method for cca model.matrix(object, ...) # S3 method for cca weights(object, display = \"sites\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Result Object from Constrained Ordination — cca.object","text":"object, x, formula result object cca, rda, dbrda, capscale. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. ordiYbar value can also \"initial\" internal working input data, \"partial\" internal working input data removing partial effects. display Display either \"sites\" \"species\". ... arguments passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Result Object from Constrained Ordination — cca.object","text":"internal (“working”) form dependent (community) data can accessed function ordiYbar. form depends ordination method: instance, cca data weighted Chi-square transformed, dbrda Gower-centred dissimilarities. input data original (“response”) form can accessed fitted.cca residuals.cca. Function predict.cca can return either working response data, also lower-rank approximations. model matrix independent data (“Constraints” “Conditions”) can extracted model.matrix. partial analysis, function returns list design matrices called Conditions Constraints. either component missing, single matrix returned. redundant (aliased) terms appear model matrix. terms can found alias.cca. Function model.frame tries reconstruct data frame model matrices derived. possible original model fitted formula data arguments, still fails data unavailable. number observations can accessed nobs.cca, residual degrees freedom df.residual.cca. information observations missing values can accessed na.action. terms formula fitted model can accessed formula terms. weights used cca can accessed weights. unweighted methods (rda) weights equal. ordination results saved separate components partial terms, constraints residual unconstrained ordination. guarantee components internal names currently, cautious developing scripts functions directly access components. constrained ordination algorithm based QR decomposition constraints conditions (environmental data), QR component saved separately partial constrained components. QR decomposition constraints can accessed qr.cca. also include residual effects partial terms (Conditions), used together ordiYbar(x, \"partial\"). environmental data first centred rda weighted centred cca. QR decomposition used many functions access cca results, can used find many items directly stored object. examples, see coef.cca, coef.rda, vif.cca, permutest.cca, predict.cca, predict.rda, calibrate.cca. See qr possible uses component. instance, rank constraints can found QR decomposition. eigenvalues solution can accessed eigenvals.cca. Eigenvalues evaluated partial component, available constrained residual components. ordination scores internally stored (weighted) orthonormal scores matrices. results can accessed scores.cca scores.rda functions. ordination scores scaled accessed scores functions, internal (weighted) orthonormal scores can accessed setting scaling = FALSE. Unconstrained residual component species site scores, constrained component also fitted site scores linear combination scores sites biplot scores centroids constraint variables. biplot scores correspond model.matrix, centroids calculated factor variables used. scores can selected defining axes, direct way accessing scores certain component. number dimensions can assessed eigenvals. addition, types can derived results although saved results. instance, regression scores model coefficients can accessed scores coef functions. Partial component scores. Distance-based methods (dbrda, capscale) can negative eigenvalues associated imaginary axis scores. addition, species scores initially missing dbrda accessory found analysis capscale (may misleading). Function sppscores can used add species scores replace meaningful ones.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Result Object from Constrained Ordination — cca.object","text":"latest large change result object made release 2.5-1 2016. can modernize ancient stray results modernobject <- update(ancientobject).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Result Object from Constrained Ordination — cca.object","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Result Object from Constrained Ordination — cca.object","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":null,"dir":"Reference","previous_headings":"","what":"Multinomial Species Classification Method (CLAM) — clamtest","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"CLAM statistical approach classifying generalists specialists two distinct habitats described Chazdon et al. (2011).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"clamtest(comm, groups, coverage.limit = 10, specialization = 2/3, npoints = 20, alpha = 0.05/20) # S3 method for clamtest summary(object, ...) # S3 method for clamtest plot(x, xlab, ylab, main, pch = 21:24, col.points = 1:4, col.lines = 2:4, lty = 1:3, position = \"bottomright\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"comm Community matrix, consisting counts. groups vector identifying two habitats. Must exactly two unique values levels. Habitat IDs grouping vector must match corresponding rows community matrix comm. coverage.limit Integer, sample coverage based correction applied rare species counts limit. Sample coverage calculated separately two habitats. Sample relative abundances used species higher equal coverage.limit total counts per habitat. specialization Numeric, specialization threshold value 0 1. value \\(2/3\\) represents ‘supermajority’ rule, value \\(1/2\\) represents ‘simple majority’ rule assign shared species habitat specialists. npoints Integer, number points used determine boundary lines plots. alpha Numeric, nominal significance level individual tests. default value reduces conventional limit \\(0.05\\) account overdispersion multiple testing several species simultaneously. However, firm reason exactly limit. x, object Fitted model object class \"clamtest\". xlab, ylab Labels plot axes. main Main title plot. pch, col.points Symbols colors used plotting species groups. lty, col.lines Line types colors boundary lines plot separate species groups. position Position figure legend, see legend specification details. Legend shown position = NULL. ... Additional arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"method uses multinomial model based estimated species relative abundance two habitats (, B). minimizes bias due differences sampling intensities two habitat types well bias due insufficient sampling within habitat. method permits robust statistical classification habitat specialists generalists, without excluding rare species priori (Chazdon et al. 2011). Based user-defined specialization threshold, model classifies species one four groups: (1) generalists; (2) habitat specialists; (3) habitat B specialists; (4) rare classify confidence.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"data frame (class attribute \"clamtest\"), columns: Species: species name (column names comm), Total_**: total count habitat , Total_*B*: total count habitat B, Classes: species classification, factor levels Generalist, Specialist_**, Specialist_*B*, Too_rare. ** *B* placeholders habitat names/labels found data. summary method returns descriptive statistics results. plot method returns values invisibly produces bivariate scatterplot species total abundances two habitats. Symbols boundary lines shown species groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Chazdon, R. L., Chao, ., Colwell, R. K., Lin, S.-Y., Norden, N., Letcher, S. G., Clark, D. B., Finegan, B. Arroyo J. P.(2011). novel statistical method classifying habitat generalists specialists. Ecology 92, 1332--1343.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Peter Solymos solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"code tested standalone CLAM software provided website Anne Chao (http://chao.stat.nthu.edu.tw/wordpress); minor inconsistencies found, especially finding threshold 'rare' species. inconsistencies probably due numerical differences two implementation. current R implementation uses root finding iso-lines instead iterative search. original method (Chazdon et al. 2011) two major problems: assumes error distribution multinomial. justified choice individuals freely distributed, -dispersion clustering individuals. ecological data, variance much higher multinomial assumption, therefore test statistic optimistic. original authors suggest multiple testing adjustment multiple testing based number points (npoints) used draw critical lines plot, whereas adjustment based number tests (.e., tested species). function uses numerical values original paper, automatic connection npoints alpha arguments, must work adjustment .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"data(mite) data(mite.env) sol <- with(mite.env, clamtest(mite, Shrub==\"None\", alpha=0.005)) summary(sol) #> Two Groups Species Classification Method (CLAM) #> #> Specialization threshold = 0.6666667 #> Alpha level = 0.005 #> #> Estimated sample coverage: #> FALSE TRUE #> 1.0000 0.9996 #> #> Minimum abundance for classification: #> FALSE TRUE #> 27 9 #> #> Species Proportion #> Generalist 10 0.286 #> Specialist_FALSE 14 0.400 #> Specialist_TRUE 4 0.114 #> Too_rare 7 0.200 head(sol) #> Species Total_FALSE Total_TRUE Classes #> 1 Brachy 534 77 Generalist #> 2 PHTH 89 0 Specialist_FALSE #> 3 HPAV 389 207 Generalist #> 4 RARD 85 0 Specialist_FALSE #> 5 SSTR 22 0 Too_rare #> 6 Protopl 26 0 Too_rare plot(sol)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an Object for Null Model Algorithms — commsim","title":"Create an Object for Null Model Algorithms — commsim","text":"commsim function can used feed Null Model algorithms nullmodel analysis. make.commsim function returns various predefined algorithm types (see Details). functions represent low level interface community null model infrastructure vegan intent extensibility, less emphasis direct use users.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"commsim(method, fun, binary, isSeq, mode) make.commsim(method) # S3 method for commsim print(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create an Object for Null Model Algorithms — commsim","text":"method Character, name algorithm. fun function. possible formal arguments function see Details. binary Logical, algorithm applies presence-absence count matrices. isSeq Logical, algorithm sequential (needs burnin thinning) . mode Character, storage mode community matrix, either \"integer\" \"double\". x object class commsim. ... Additional arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create an Object for Null Model Algorithms — commsim","text":"function fun must return array dim(nr, nc, n), must take following arguments: x: input matrix, n: number permuted matrices output, nr: number rows, nc: number columns, rs: vector row sums, cs: vector column sums, rf: vector row frequencies (non-zero cells), cf: vector column frequencies (non-zero cells), s: total sum x, fill: matrix fill (non-zero cells), thin: thinning value sequential algorithms, ...: additional arguments. can define null model, several null model algorithm pre-defined can called name. predefined algorithms described detail following chapters. binary null models produce matrices zeros (absences) ones (presences) also input matrix quantitative. two types quantitative data: Counts integers natural unit individuals can shuffled, abundances can real (floating point) values natural subunit shuffling. quantitative models can handle counts, able handle real values. null models sequential next matrix derived current one. makes models dependent previous models, usually must thin matrices study sequences stability: see oecosimu details instructions. See Examples structural constraints imposed algorithm defining null model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"binary-null-models","dir":"Reference","previous_headings":"","what":"Binary null models","title":"Create an Object for Null Model Algorithms — commsim","text":"binary null models preserve fill: number presences conversely number absences. classic models may also preserve column (species) frequencies (c0) row frequencies species richness site (r0) take account commonness rarity species (r1, r2). Algorithms swap, tswap, curveball, quasiswap backtracking preserve row column frequencies. Three first ones sequential two latter non-sequential produce independent matrices. Basic algorithms reviewed Wright et al. (1998). \"r00\": non-sequential algorithm binary matrices preserves number presences (fill). \"r0\": non-sequential algorithm binary matrices preserves site (row) frequencies. \"r1\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses column marginal frequencies probabilities selecting species. \"r2\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses squared column marginal frequencies probabilities selecting species. \"c0\": non-sequential algorithm binary matrices preserves species frequencies (Jonsson 2001). \"swap\": sequential algorithm binary matrices changes matrix structure, influence marginal sums (Gotelli & Entsminger 2003). inspects \\(2 \\times 2\\) submatrices long swap can done. \"tswap\": sequential algorithm binary matrices. \"swap\" algorithm, tries fixed number times performs zero many swaps one step (according thin argument call). approach suggested Miklós & Podani (2004) found ordinary swap may lead biased sequences, since columns rows easily swapped. \"curveball\": sequential method binary matrices implements ‘Curveball’ algorithm Strona et al. (2014). algorithm selects two random rows finds set unique species occur one rows. algorithm distributes set unique species rows preserving original row frequencies. Zero several species swapped one step, usually matrix perturbed strongly sequential methods. \"quasiswap\": non-sequential algorithm binary matrices implements method matrix first filled honouring row column totals, integers may larger one. method inspects random \\(2 \\times 2\\) matrices performs quasiswap . addition ordinary swaps, quasiswap can reduce numbers one ones preserving marginal totals (Miklós & Podani 2004). method non-sequential, accepts thin argument: convergence checked every thin steps. allows performing several ordinary swaps addition fill changing swaps helps reducing removing bias. \"greedyqswap\": greedy variant quasiswap. greedy step, one element \\(2 \\times 2\\) matrix taken \\(> 1\\) elements. greedy steps biased, method can thinned, first thin steps greedy. Even modest thinning (say thin = 20) removes reduces bias, thin = 100 (1% greedy steps) looks completely safe still speeds simulation. code experimental provided scrutiny, tested bias use. \"backtracking\": non-sequential algorithm binary matrices implements filling method constraints row column frequencies (Gotelli & Entsminger 2001). matrix first filled randomly, typically row column sums reached incidences filled . begins \"backtracking\", incidences removed, filling started , backtracking done many times incidences filled matrix. results may biased inspected carefully use.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-models-for-counts-with-fixed-marginal-sums","dir":"Reference","previous_headings":"","what":"Quantitative Models for Counts with Fixed Marginal Sums","title":"Create an Object for Null Model Algorithms — commsim","text":"models shuffle individuals counts keep marginal sums fixed, marginal frequencies preserved. Algorithm r2dtable uses standard R function r2dtable also used simulated \\(P\\)-values chisq.test. Algorithm quasiswap_count uses , preserves original fill. Typically means increasing numbers zero cells result zero-inflated respect r2dtable. \"r2dtable\": non-sequential algorithm count matrices. algorithm keeps matrix sum row/column sums constant. Based r2dtable. \"quasiswap_count\": non-sequential algorithm count matrices. algorithm similar Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function preserving row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping \\(2 \\times 2\\) submatrices (see \"swap_count\" algorithm details) maintain row column totals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative swap models similar binary swap, swap largest permissible value. models section maintain fill perform quantitative swap can done without changing fill. Single step swap often changes matrix little. particular, cell counts variable, high values change slowly. Checking chain stability independence even crucial binary swap, strong thinning often needed. models never used without inspecting properties current data. null models can also defined using permatswap function. \"swap_count\": sequential algorithm count matrices. algorithm find \\(2 \\times 2\\) submatrices can swapped leaving column row totals fill unchanged. algorithm finds largest value submatrix can swapped (\\(d\\)). Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. \"abuswap_r\": sequential algorithm count nonnegative real valued matrices fixed row frequencies (see also permatswap). algorithm similar swap_count, uses different swap value row \\(2 \\times 2\\) submatrix. step changes corresponding column sums, honours matrix fill, row sums, row/column frequencies (Hardy 2008; randomization scheme 2x). \"abuswap_c\": sequential algorithm count nonnegative real valued matrices fixed column frequencies (see also permatswap). algorithm similar previous one, operates columns. step changes corresponding row sums, honours matrix fill, column sums, row/column frequencies (Hardy 2008; randomization scheme 3x).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-and-shuffle-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap and Shuffle Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative Swap Shuffle methods (swsh methods) preserve fill column row frequencies, also either row column sums. methods first perform binary quasiswap shuffle original quantitative data non-zero cells. samp methods shuffle original non-zero cell values can used also non-integer data. methods redistribute individuals randomly among non-zero cells can used integer data. shuffling either free whole matrix, within rows (r methods) within columns (c methods). Shuffling within row preserves row sums, shuffling within column preserves column sums. models can also defined permatswap. \"swsh_samp\": non-sequential algorithm quantitative data (either integer counts non-integer values). Original non-zero values values shuffled. \"swsh_both\": non-sequential algorithm count data. Individuals shuffled freely non-zero cells. \"swsh_samp_r\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately row. \"swsh_samp_c\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately column. \"swsh_both_r\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values within row. \"swsh_both_c\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values column.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-shuffle-methods","dir":"Reference","previous_headings":"","what":"Quantitative Shuffle Methods","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative shuffle methods generalizations binary models r00, r0 c0. _ind methods shuffle individuals grand sum, row sum column sums preserved. methods similar r2dtable still slacker constraints marginal sums. _samp _both methods first apply corresponding binary model similar restriction marginal frequencies distribute quantitative values non-zero cells. _samp models shuffle original cell values can therefore handle also non-count real values. _both models shuffle individuals among non-zero values. shuffling whole matrix r00_, within row r0_ within column c0_ cases. \"r00_ind\": non-sequential algorithm count matrices. algorithm preserves grand sum individuals shuffled among cells matrix. \"r0_ind\": non-sequential algorithm count matrices. algorithm preserves row sums individuals shuffled among cells row matrix. \"c0_ind\": non-sequential algorithm count matrices. algorithm preserves column sums individuals shuffled among cells column matrix. \"r00_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves grand sum cells matrix shuffled. \"r0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves row sums cells within row shuffled. \"c0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves column sums constant cells within column shuffled. \"r00_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells matrix shuffled. \"r0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells row shuffled. \"c0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells column shuffled.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create an Object for Null Model Algorithms — commsim","text":"object class commsim elements corresponding arguments (method, binary, isSeq, mode, fun). input make.comsimm commsim object, returned without evaluation. case, character method argument matched predefined algorithm names. error message issued none found. method argument missing, function returns names currently available null model algorithms character vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Create an Object for Null Model Algorithms — commsim","text":"Gotelli, N.J. & Entsminger, N.J. (2001). Swap fill algorithms null model analysis: rethinking knight's tour. Oecologia 129, 281--291. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97. Strona, G., Nappo, D., Boccacci, F., Fattorini, S. & San-Miguel-Ayanz, J. (2014). fast unbiased procedure randomize ecological binary matrices fixed row column totals. Nature Communications 5:4114 doi:10.1038/ncomms5114 . Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an Object for Null Model Algorithms — commsim","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"## write the r00 algorithm f <- function(x, n, ...) array(replicate(n, sample(x)), c(dim(x), n)) (cs <- commsim(\"r00\", fun=f, binary=TRUE, isSeq=FALSE, mode=\"integer\")) #> An object of class “commsim” #> ‘r00’ method (binary, non-sequential, integer mode) #> ## retrieving the sequential swap algorithm (cs <- make.commsim(\"swap\")) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## feeding a commsim object as argument make.commsim(cs) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## making the missing c1 model using r1 as a template ## non-sequential algorithm for binary matrices ## that preserves the species (column) frequencies, ## but uses row marginal frequencies ## as probabilities of selecting sites f <- function (x, n, nr, nc, rs, cs, ...) { out <- array(0L, c(nr, nc, n)) J <- seq_len(nc) storage.mode(rs) <- \"double\" for (k in seq_len(n)) for (j in J) out[sample.int(nr, cs[j], prob = rs), j, k] <- 1L out } cs <- make.commsim(\"r1\") cs$method <- \"c1\" cs$fun <- f ## structural constraints diagfun <- function(x, y) { c(sum = sum(y) == sum(x), fill = sum(y > 0) == sum(x > 0), rowSums = all(rowSums(y) == rowSums(x)), colSums = all(colSums(y) == colSums(x)), rowFreq = all(rowSums(y > 0) == rowSums(x > 0)), colFreq = all(colSums(y > 0) == colSums(x > 0))) } evalfun <- function(meth, x, n) { m <- nullmodel(x, meth) y <- simulate(m, nsim=n) out <- rowMeans(sapply(1:dim(y)[3], function(i) diagfun(attr(y, \"data\"), y[,,i]))) z <- as.numeric(c(attr(y, \"binary\"), attr(y, \"isSeq\"), attr(y, \"mode\") == \"double\")) names(z) <- c(\"binary\", \"isSeq\", \"double\") c(z, out) } x <- matrix(rbinom(10*12, 1, 0.5)*rpois(10*12, 3), 12, 10) algos <- make.commsim() a <- t(sapply(algos, evalfun, x=x, n=10)) print(as.table(ifelse(a==1,1,0)), zero.print = \".\") #> binary isSeq double sum fill rowSums colSums rowFreq colFreq #> r00 1 . . 1 1 . . . . #> c0 1 . . 1 1 . 1 . 1 #> r0 1 . . 1 1 1 . 1 . #> r1 1 . . 1 1 1 . 1 . #> r2 1 . . 1 1 1 . 1 . #> quasiswap 1 . . 1 1 1 1 1 1 #> greedyqswap 1 . . 1 1 1 1 1 1 #> swap 1 1 . 1 1 1 1 1 1 #> tswap 1 1 . 1 1 1 1 1 1 #> curveball 1 1 . 1 1 1 1 1 1 #> backtrack 1 . . 1 1 1 1 1 1 #> r2dtable . . . 1 . 1 1 . . #> swap_count . 1 . 1 1 1 1 . . #> quasiswap_count . . . 1 1 1 1 . . #> swsh_samp . . 1 1 1 . . 1 1 #> swsh_both . . . 1 1 . . 1 1 #> swsh_samp_r . . 1 1 1 1 . 1 1 #> swsh_samp_c . . 1 1 1 . 1 1 1 #> swsh_both_r . . . 1 1 1 . 1 1 #> swsh_both_c . . . 1 1 . 1 1 1 #> abuswap_r . 1 1 1 1 1 . 1 1 #> abuswap_c . 1 1 1 1 . 1 1 1 #> r00_samp . . 1 1 1 . . . . #> c0_samp . . 1 1 1 . 1 . 1 #> r0_samp . . 1 1 1 1 . 1 . #> r00_ind . . . 1 . . . . . #> c0_ind . . . 1 . . 1 . . #> r0_ind . . . 1 . 1 . . . #> r00_both . . . 1 1 . . . . #> c0_both . . . 1 1 . 1 . 1 #> r0_both . . . 1 1 1 . 1 ."},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":null,"dir":"Reference","previous_headings":"","what":"Contribution Diversity Approach — contribdiv","title":"Contribution Diversity Approach — contribdiv","text":"contribution diversity approach based differentiation within-unit among-unit diversity using additive diversity partitioning unit distinctiveness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Contribution Diversity Approach — contribdiv","text":"","code":"contribdiv(comm, index = c(\"richness\", \"simpson\"), relative = FALSE, scaled = TRUE, drop.zero = FALSE) # S3 method for contribdiv plot(x, sub, xlab, ylab, ylim, col, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Contribution Diversity Approach — contribdiv","text":"comm community data matrix samples rows species column. index Character, diversity index calculated. relative Logical, TRUE contribution diversity values expressed signed deviation mean. See details. scaled Logical, TRUE relative contribution diversity values scaled sum gamma values (index = \"richness\") sum gamma values times number rows comm (index = \"simpson\"). See details. drop.zero Logical, empty rows dropped result? empty rows dropped, corresponding results NAs. x object class \"contribdiv\". sub, xlab, ylab, ylim, col Graphical arguments passed plot. ... arguments passed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Contribution Diversity Approach — contribdiv","text":"approach proposed Lu et al. (2007). Additive diversity partitioning (see adipart references) deals relation mean alpha total (gamma) diversity. Although alpha diversity values often vary considerably. Thus, contributions sites total diversity uneven. site specific contribution measured contribution diversity components. unit e.g. many unique species contribute higher level (gamma) diversity another unit number species, common. Distinctiveness species \\(j\\) can defined number sites occurs (\\(n_j\\)), sum relative frequencies (\\(p_j\\)). Relative frequencies computed sitewise \\(sum_j{p_ij}\\)s site \\(\\) sum \\(1\\). contribution site \\(\\) total diversity given \\(alpha_i = sum_j(1 / n_ij)\\) dealing richness \\(alpha_i = sum(p_{ij} * (1 - p_{ij}))\\) Simpson index. unit distinctiveness site \\(\\) average species distinctiveness, averaging species occur site \\(\\). species richness: \\(alpha_i = mean(n_i)\\) (paper, second equation contains typo, \\(n\\) without index). Simpson index: \\(alpha_i = mean(n_i)\\). Lu et al. (2007) gives -depth description different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Contribution Diversity Approach — contribdiv","text":"object class \"contribdiv\" inheriting data frame. Returned values alpha, beta gamma components sites (rows) community matrix. \"diff.coef\" attribute gives differentiation coefficient (see Examples).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Contribution Diversity Approach — contribdiv","text":"Lu, H. P., Wagner, H. H. Chen, X. Y. 2007. contribution diversity approach evaluate species diversity. Basic Applied Ecology, 8, 1--12.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Contribution Diversity Approach — contribdiv","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Contribution Diversity Approach — contribdiv","text":"","code":"## Artificial example given in ## Table 2 in Lu et al. 2007 x <- matrix(c( 1/3,1/3,1/3,0,0,0, 0,0,1/3,1/3,1/3,0, 0,0,0,1/3,1/3,1/3), 3, 6, byrow = TRUE, dimnames = list(LETTERS[1:3],letters[1:6])) x #> a b c d e f #> A 0.3333333 0.3333333 0.3333333 0.0000000 0.0000000 0.0000000 #> B 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 0.0000000 #> C 0.0000000 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 ## Compare results with Table 2 contribdiv(x, \"richness\") #> alpha beta gamma #> A 1 1.5 2.5 #> B 1 0.5 1.5 #> C 1 1.0 2.0 contribdiv(x, \"simpson\") #> alpha beta gamma #> A 0.6666667 0.1851852 0.8518519 #> B 0.6666667 0.1111111 0.7777778 #> C 0.6666667 0.1481481 0.8148148 ## Relative contribution (C values), compare with Table 2 (cd1 <- contribdiv(x, \"richness\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.5 0.5 #> B 0 -0.5 -0.5 #> C 0 0.0 0.0 (cd2 <- contribdiv(x, \"simpson\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.03703704 0.03703704 #> B 0 -0.03703704 -0.03703704 #> C 0 0.00000000 0.00000000 ## Differentiation coefficients attr(cd1, \"diff.coef\") # D_ST #> [1] 0.5 attr(cd2, \"diff.coef\") # D_DT #> [1] 0.1818182 ## BCI data set data(BCI) opar <- par(mfrow=c(2,2)) plot(contribdiv(BCI, \"richness\"), main = \"Absolute\") plot(contribdiv(BCI, \"richness\", relative = TRUE), main = \"Relative\") plot(contribdiv(BCI, \"simpson\")) plot(contribdiv(BCI, \"simpson\", relative = TRUE)) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] Distance-based Redundancy Analysis — dbrda","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Distance-based redundancy analysis (dbRDA) ordination method similar Redundancy Analysis (rda), allows non-Euclidean dissimilarity indices, Manhattan Bray-Curtis distance. Despite non-Euclidean feature, analysis strictly linear metric. called Euclidean distance, results identical rda, dbRDA less efficient. Functions dbrda constrained versions metric scaling, .k.. principal coordinates analysis, based Euclidean distance can used, useful, dissimilarity measures. Function capscale simplified version based Euclidean approximation dissimilarities. functions can also perform unconstrained principal coordinates analysis, optionally using extended dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"dbrda(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...) capscale(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, comm = NULL, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"formula Model formula. function can called formula interface. usual features formula hold, especially defined cca rda. LHS must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist function given dfun used find dissimilarities. RHS defines constraints. constraints can continuous variables factors, can transformed within formula, can interactions typical formula. RHS can special term Condition defines variables “partialled ” constraints, just like rda cca. allows use partial dbRDA. data Data frame containing variables right hand side model formula. distance name dissimilarity (distance) index LHS formula data frame instead dissimilarity matrix. sqrt.dist Take square roots dissimilarities. See section Details . comm Community data frame used finding species scores LHS formula dissimilarity matrix. used LHS data frame. supplied, “species scores” unavailable dissimilarities supplied. N.B., available capscale: dbrda return species scores. Function sppscores can used add species scores missing. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). \"lingoes\" (TRUE) uses recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. latter one cmdscale. dfun Distance dissimilarity function used. function returning standard \"dist\" taking index name first argument can used. metaMDSdist Use metaMDSdist similarly metaMDS. means automatic data transformation using extended flexible shortest path dissimilarities (function stepacross) many dissimilarities based shared species. na.action Handling missing values constraints conditions. default (na.fail) stop missing values. Choices na.omit na.exclude delete rows missing values, differ representation results. na.omit non-missing site scores shown, na.exclude gives NA scores missing observations. Unlike rda, WA scores available missing constraints conditions. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data (given formula comm argument). ... parameters passed underlying functions (e.g., metaMDSdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Functions dbrda capscale provide two alternative implementations dbRDA. Function dbrda based McArdle & Anderson (2001) directly decomposes dissimilarities. Euclidean distances results identical rda. Non-Euclidean dissimilarities may give negative eigenvalues associated imaginary axes. Function capscale based Legendre & Anderson (1999): dissimilarity data first ordinated using metric scaling, ordination results analysed rda. capscale ignores imaginary component give negative eigenvalues (report magnitude imaginary component). user supplied community data frame instead dissimilarities, functions find dissimilarities using vegdist distance function given dfun specified distance. functions accept distance objects vegdist, dist, method producing compatible objects. constraining variables can continuous factors , can interaction terms, can transformed call. Moreover, can special term Condition just like rda cca “partial” analysis can performed. Function dbrda return species scores, can also missing capscale, can added analysis using function sppscores. Non-Euclidean dissimilarities can produce negative eigenvalues (Legendre & Anderson 1999, McArdle & Anderson 2001). negative eigenvalues, printed output capscale add column sums positive eigenvalues item sum negative eigenvalues, dbrda add column giving number real dimensions positive eigenvalues. negative eigenvalues disturbing, functions let distort dissimilarities non-negative eigenvalues produced argument add = TRUE. Alternatively, sqrt.dist = TRUE, square roots dissimilarities can used may help avoiding negative eigenvalues (Legendre & Anderson 1999). functions can also used perform ordinary metric scaling .k.. principal coordinates analysis using formula constant right hand side, comm ~ 1. metaMDSdist = TRUE, function can automatic data standardization use extended dissimilarities using function stepacross similarly non-metric multidimensional scaling metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"functions return object class dbrda capscale inherit rda. See cca.object description result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Anderson, M.J. & Willis, T.J. (2003). Canonical analysis principal coordinates: useful method constrained ordination ecology. Ecology 84, 511--525. Gower, J.C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. Legendre, P. & Legendre, L. (2012). Numerical Ecology. 3rd English Edition. Elsevier. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290--297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Function dbrda implements real distance-based RDA preferred capscale. capscale originally developed variant constrained analysis proximities (Anderson & Willis 2003), developments made similar dbRDA. However, discards imaginary dimensions negative eigenvalues ordination significance tests area based real dimensions positive eigenvalues. capscale may removed vegan future. vegan since 2003 (CRAN release 1.6-0) dbrda first released 2016 (version 2.4-0), removal capscale may disruptive historical examples scripts, modern times dbrda used. inertia named dissimilarity index defined dissimilarity data, unknown distance information missing. largest original dissimilarity larger 4, capscale handles input similarly rda bases analysis variance instead sum squares. Keyword mean added inertia cases, e.g. Euclidean Manhattan distances. Inertia based squared index, keyword squared added name distance, unless data square root transformed (argument sqrt.dist=TRUE). additive constant used argument add, Lingoes Cailliez adjusted added name inertia, value constant printed.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"data(varespec, varechem) ## dbrda dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\") #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\") #> #> Inertia Proportion Rank RealDims #> Total 4.5444 1.0000 #> Conditional 0.9726 0.2140 1 #> Constrained 0.9731 0.2141 3 3 #> Unconstrained 2.5987 0.5718 19 13 #> Inertia is squared Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5362 0.3198 0.1171 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9054 0.5070 0.3336 0.2581 0.2027 0.1605 0.1221 0.0825 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues with sqrt distances dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\", sqrt.dist = TRUE) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\", sqrt.dist = TRUE) #> #> Inertia Proportion Rank #> Total 6.9500 1.0000 #> Conditional 0.9535 0.1372 1 #> Constrained 1.2267 0.1765 3 #> Unconstrained 4.7698 0.6863 19 #> Inertia is Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5817 0.4086 0.2365 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9680 0.6100 0.4469 0.3837 0.3371 0.3012 0.2558 0.2010 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues also with Jaccard distances (m <- dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"jaccard\")) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"jaccard\") #> #> Inertia Proportion Rank #> Total 6.5044 1.0000 #> Conditional 1.0330 0.1588 1 #> Constrained 1.2068 0.1855 3 #> Unconstrained 4.2646 0.6557 19 #> Inertia is squared Jaccard distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5992 0.3994 0.2082 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.0388 0.6441 0.4518 0.3759 0.3239 0.2785 0.2279 0.1644 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## add species scores sppscores(m) <- wisconsin(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":null,"dir":"Reference","previous_headings":"","what":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Performs detrended correspondence analysis basic reciprocal averaging orthogonal correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"decorana(veg, iweigh=0, iresc=4, ira=0, mk=26, short=0, before=NULL, after=NULL) # S3 method for decorana plot(x, choices=c(1,2), origin=TRUE, display=c(\"both\",\"sites\",\"species\",\"none\"), cex = 0.8, cols = c(1,2), type, xlim, ylim, ...) # S3 method for decorana text(x, display = c(\"sites\", \"species\"), labels, choices = 1:2, origin = TRUE, select, ...) # S3 method for decorana points(x, display = c(\"sites\", \"species\"), choices=1:2, origin = TRUE, select, ...) # S3 method for decorana scores(x, display=\"sites\", choices=1:4, origin=TRUE, tidy=FALSE, ...) downweight(veg, fraction = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"veg Community data, matrix-like object. iweigh Downweighting rare species (0: ). iresc Number rescaling cycles (0: rescaling). ira Type analysis (0: detrended, 1: basic reciprocal averaging). mk Number segments rescaling. short Shortest gradient rescaled. Hill's piecewise transformation: values transformation. Hill's piecewise transformation: values transformation -- must correspond values . x decorana result object. choices Axes shown. origin Use true origin even detrended correspondence analysis. display Display sites, species, neither. cex Plot character size. cols Colours used sites species. type Type plots, partial match \"text\", \"points\" \"none\". labels Optional text used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. xlim, ylim x y limits (min,max) plot. fraction Abundance fraction downweighting begins. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score (\"sites\", \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. ... arguments plot function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"late 1970s, correspondence analysis became method choice ordination vegetation science, since seemed better able cope non-linear species responses principal components analysis. However, even correspondence analysis can produce arc-shaped configuration single gradient. Mark Hill developed detrended correspondence analysis correct two assumed ‘faults’ correspondence analysis: curvature straight gradients packing sites ends gradient. curvature removed replacing orthogonalization axes detrending. orthogonalization successive axes made non-correlated, detrending remove systematic dependence axes. Detrending performed using smoothing window mk segments. packing sites ends gradient undone rescaling axes extraction. rescaling, axis supposed scaled ‘SD’ units, average width Gaussian species responses supposed one whole axis. innovations piecewise linear transformation species abundances downweighting rare species regarded unduly high influence ordination axes. seems detrending actually works twisting ordination space, results look non-curved two-dimensional projections (‘lolly paper effect’). result, points usually easily recognized triangular diamond shaped pattern, obviously artefact detrending. Rescaling works differently commonly presented, . decorana use, even evaluate, widths species responses. Instead, tries equalize weighted standard deviation species scores axis segments (parameter mk effect, since decorana finds segments internally). Function tolerance returns internal criterion can used assess success rescaling. plot method plots species site scores. Classical decorana scaled axes smallest site score 0 ( smallest species score negative), summary, plot scores use true origin, unless origin = FALSE. addition proper eigenvalues, function reports ‘decorana values’ detrended analysis. ‘decorana values’ values legacy code decorana returns eigenvalues. estimated iteration, describe joint effects axes detrending. ‘decorana values’ estimated rescaling show effect eigenvalues. proper eigenvalues estimated extraction axes ratio weighted sum squares site species scores even detrended rescaled solutions. eigenvalues estimated axis separately, additive, higher decorana axes can show effects already explained prior axes. ‘Additive eigenvalues’ cleansed effects prior axes, can assumed add total inertia (scaled Chi-square). proportions cumulative proportions explained can use eigenvals.decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana returns object class \"decorana\", print, summary, scores, plot, points text methods, support functions eigenvals, bstick, screeplot, predict tolerance. downweight independent function can also used methods decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Hill, M.O. Gauch, H.G. (1980). Detrended correspondence analysis: improved ordination technique. Vegetatio 42, 47--58. Oksanen, J. Minchin, P.R. (1997). Instability ordination results changes input data order: explanations remedies. Journal Vegetation Science 8, 447--454.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Mark O. Hill wrote original Fortran code, R port Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana uses central numerical engine original Fortran code (public domain), 1/3 original program. tried implement original behaviour, although great part preparatory steps written R language, may differ somewhat original code. However, well-known bugs corrected strict criteria used (Oksanen & Minchin 1997). Please note really need piecewise transformation even downweighting within decorana, since powerful extensive alternatives R, options included compliance original software. different fraction abundance needed downweighting, function downweight must applied decorana. Function downweight indeed can applied prior correspondence analysis, can used together cca, . Github package natto R implementation decorana allows easier inspection algorithm also easier development function. vegan 2.6-6 earlier summary method, nothing useful now defunct. former information can extracted scores weights.decorana.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"data(varespec) vare.dca <- decorana(varespec) vare.dca #> #> Call: #> decorana(veg = varespec) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.0832 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.5235 0.3253 0.20010 0.19176 #> Additive Eigenvalues 0.5235 0.3217 0.17919 0.11922 #> Decorana values 0.5249 0.1572 0.09669 0.06075 #> Axis lengths 2.8161 2.2054 1.54650 1.64864 #> plot(vare.dca) ### the detrending rationale: gaussresp <- function(x,u) exp(-(x-u)^2/2) x <- seq(0,6,length=15) ## The gradient u <- seq(-2,8,len=23) ## The optima pack <- outer(x,u,gaussresp) matplot(x, pack, type=\"l\", main=\"Species packing\") opar <- par(mfrow=c(2,2)) plot(scores(prcomp(pack)), asp=1, type=\"b\", main=\"PCA\") plot(scores(decorana(pack, ira=1)), asp=1, type=\"b\", main=\"CA\") plot(scores(decorana(pack)), asp=1, type=\"b\", main=\"DCA\") plot(scores(cca(pack ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") ### Let's add some noise: noisy <- (0.5 + runif(length(pack)))*pack par(mfrow=c(2,1)) matplot(x, pack, type=\"l\", main=\"Ideal model\") matplot(x, noisy, type=\"l\", main=\"Noisy model\") par(mfrow=c(2,2)) plot(scores(prcomp(noisy)), type=\"b\", main=\"PCA\", asp=1) plot(scores(decorana(noisy, ira=1)), type=\"b\", main=\"CA\", asp=1) plot(scores(decorana(noisy)), type=\"b\", main=\"DCA\", asp=1) plot(scores(cca(noisy ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":null,"dir":"Reference","previous_headings":"","what":"Standardization Methods for Community Ecology — decostand","title":"Standardization Methods for Community Ecology — decostand","text":"function provides popular (effective) standardization methods community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"decostand(x, method, MARGIN, range.global, logbase = 2, na.rm=FALSE, ...) wisconsin(x) decobackstand(x, zap = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Standardization Methods for Community Ecology — decostand","text":"x Community data, matrix-like object. decobackstand standardized data. method Standardization method. See Details available options. MARGIN Margin, default acceptable. 1 = rows, 2 = columns x. range.global Matrix range found method = \"range\". allows using ranges across subsets data. dimensions MARGIN must match x. logbase logarithm base used method = \"log\". na.rm Ignore missing values row column standardizations. zap Make near-zero values exact zeros avoid negative values exaggerated estimates species richness. ... arguments function (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Standardization Methods for Community Ecology — decostand","text":"function offers following standardization methods community data: total: divide margin total (default MARGIN = 1). max: divide margin maximum (default MARGIN = 2). frequency: divide margin total multiply number non-zero items, average non-zero entries one (Oksanen 1983; default MARGIN = 2). normalize: make margin sum squares equal one (default MARGIN = 1). range: standardize values range 0 ... 1 (default MARGIN = 2). values constant, transformed 0. rank, rrank: rank replaces abundance values increasing ranks leaving zeros unchanged, rrank similar uses relative ranks maximum 1 (default MARGIN = 1). Average ranks used tied values. standardize: scale x zero mean unit variance (default MARGIN = 2). pa: scale x presence/absence scale (0/1). chi.square: divide row sums square root column sums, adjust square root matrix total (Legendre & Gallagher 2001). used Euclidean distance, distances similar Chi-square distance used correspondence analysis. However, results cmdscale still differ, since CA weighted ordination method (default MARGIN = 1). hellinger: square root method = \"total\" (Legendre & Gallagher 2001). log: logarithmic transformation suggested Anderson et al. (2006): \\(\\log_b (x) + 1\\) \\(x > 0\\), \\(b\\) base logarithm; zeros left zeros. Higher bases give less weight quantities presences, logbase = Inf gives presence/absence scaling. Please note \\(\\log(x+1)\\). Anderson et al. (2006) suggested (strongly) modified Gower distance (implemented method = \"altGower\" vegdist), standardization can used independently distance indices. alr: Additive log ratio (\"alr\") transformation (Aitchison 1986) reduces data skewness compositionality bias. transformation assumes positive values, pseudocounts can added argument pseudocount. One rows/columns reference can given reference (name index). first row/column used default (reference = 1). Note transformation drops one row column transformed output data. alr transformation defined formally follows: $$alr = [log\\frac{x_1}{x_D}, ..., log\\frac{x_{D-1}}{x_D}]$$ denominator sample \\(x_D\\) can chosen arbitrarily. transformation often used pH chemistry measurenments. also commonly used multinomial logistic regression. Default MARGIN = 1 uses row reference. clr: centered log ratio (\"clr\") transformation proposed Aitchison (1986) used reduce data skewness compositionality bias. transformation frequent applications microbial ecology (see e.g. Gloor et al., 2017). clr transformation defined : $$clr = log\\frac{x}{g(x)} = log x - log g(x)$$ \\(x\\) single value, g(x) geometric mean \\(x\\). method can operate positive data; common way deal zeroes add pseudocount (e.g. smallest positive value data), either adding manually input data, using argument pseudocount decostand(x, method = \"clr\", pseudocount = 1). Adding pseudocount inevitably introduce bias; see rclr method one available solution. rclr: robust clr (\"rclr\") similar regular clr (see ) allows data contains zeroes. method use pseudocounts, unlike standard clr. robust clr (rclr) divides values geometric mean observed features; zero values kept zeroes, taken account. high dimensional data, geometric mean rclr approximates true geometric mean; see e.g. Martino et al. (2019) rclr transformation defined formally follows: $$rclr = log\\frac{x}{g(x > 0)}$$ \\(x\\) single value, \\(g(x > 0)\\) geometric mean sample-wide values \\(x\\) positive (> 0). Standardization, contrasted transformation, means entries transformed relative entries. methods default margin. MARGIN=1 means rows (sites normal data set) MARGIN=2 means columns (species normal data set). Command wisconsin shortcut common Wisconsin double standardization species (MARGIN=2) first standardized maxima (max) sites (MARGIN=1) site totals (tot). standardization methods give nonsense results negative data entries normally occur community data. empty sites species (constant method = \"range\"), many standardization change NaN. Function decobackstand can used transform standardized data back original. possible standardization may implemented cases possible. round-errors back-transformation exact, wise overwrite original data. zap=TRUE original zeros exact.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Standardization Methods for Community Ecology — decostand","text":"Returns standardized data frame, adds attribute \"decostand\" giving name applied standardization \"method\" attribute \"parameters\" appropriate transformation parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Standardization Methods for Community Ecology — decostand","text":"Jari Oksanen, Etienne Laliberté (method = \"log\"), Leo Lahti (alr, \"clr\" \"rclr\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Standardization Methods for Community Ecology — decostand","text":"Common transformations can made standard R functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Standardization Methods for Community Ecology — decostand","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Egozcue, J.J., Pawlowsky-Glahn, V., Mateu-Figueras, G., Barcel'o-Vidal, C. (2003) Isometric logratio transformations compositional data analysis. Mathematical Geology 35, 279--300. Gloor, G.B., Macklaim, J.M., Pawlowsky-Glahn, V. & Egozcue, J.J. (2017) Microbiome Datasets Compositional: Optional. Frontiers Microbiology 8, 2224. Legendre, P. & Gallagher, E.D. (2001) Ecologically meaningful transformations ordination species data. Oecologia 129, 271--280. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Oksanen, J. (1983) Ordination boreal heath-like vegetation principal component analysis, correspondence analysis multidimensional scaling. Vegetatio 52, 181--189.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"data(varespec) sptrans <- decostand(varespec, \"max\") apply(sptrans, 2, max) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv Descflex Betupube #> 1 1 1 1 1 1 1 1 #> Vacculig Diphcomp Dicrsp Dicrfusc Dicrpoly Hylosple Pleuschr Polypili #> 1 1 1 1 1 1 1 1 #> Polyjuni Polycomm Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 1 1 1 1 1 1 1 1 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris Cladchlo Cladbotr #> 1 1 1 1 1 1 1 1 #> Cladamau Cladsp Cetreric Cetrisla Flavniva Nepharct Stersp Peltapht #> 1 1 1 1 1 1 1 1 #> Icmaeric Cladcerv Claddefo Cladphyl #> 1 1 1 1 sptrans <- wisconsin(varespec) # CLR transformation for rows, with pseudocount varespec.clr <- decostand(varespec, \"clr\", pseudocount=1) # ALR transformation for rows, with pseudocount and reference sample varespec.alr <- decostand(varespec, \"alr\", pseudocount=1, reference=1) ## Chi-square: PCA similar but not identical to CA. ## Use wcmdscale for weighted analysis and identical results. sptrans <- decostand(varespec, \"chi.square\") plot(procrustes(rda(sptrans), cca(varespec)))"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Design your own Dissimilarities — designdist","title":"Design your own Dissimilarities — designdist","text":"Function designdist lets define dissimilarities using terms shared total quantities, number rows number columns. shared total quantities can binary, quadratic minimum terms. binary terms, shared component number shared species, totals numbers species sites. quadratic terms cross-products sums squares, minimum terms sums parallel minima row totals. Function designdist2 similar, finds dissimilarities among two data sets. Function chaodist lets define dissimilarities using terms supposed take account “unseen species” (see Chao et al., 2005 Details vegdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Design your own Dissimilarities — designdist","text":"","code":"designdist(x, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) designdist2(x, y, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) chaodist(x, method = \"1 - 2*U*V/(U+V)\", name)"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Design your own Dissimilarities — designdist","text":"x Input data. y Another input data set: dissimilarities calculated among rows x rows y. method Equation dissimilarities. can use terms J shared quantity, B totals, N number rows (sites) P number columns (species) chaodist can use terms U V. equation can also contain R functions accepts vector arguments returns vectors length. terms shared total components found. vectors x y \"quadratic\" terms J = sum(x*y), = sum(x^2), B = sum(y^2), \"minimum\" terms J = sum(pmin(x,y)), = sum(x) B = sum(y), \"binary\" terms either transforming data binary form (shared number species, number species row). abcd Use 2x2 contingency table notation binary data: \\(\\) number shared species, \\(b\\) \\(c\\) numbers species occurring one sites , \\(d\\) number species occur neither sites. alphagamma Use beta diversity notation terms alpha average alpha diversity compared sites, gamma diversity pooled sites, delta absolute value difference average alpha alpha diversities compared sites. Terms B refer alpha diversities compared sites. name name want use index. default combine method equation terms argument. maxdist Theoretical maximum dissimilarity, NA index open absolute maximum. necessary argument, used vegan functions, certain maximum, better supply value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Design your own Dissimilarities — designdist","text":"popular dissimilarity measures ecology can expressed help terms J, B, also involve matrix dimensions N P. examples can define designdist : function designdist can implement dissimilarity indices vegdist elsewhere, can also used implement many indices, amongst , described Legendre & Legendre (2012). can also used implement indices beta diversity described Koleff et al. (2003), also specific function betadiver purpose. want implement binary dissimilarities based 2x2 contingency table notation, can set abcd = TRUE. notation = J, b = -J, c = B-J, d = P--B+J. notation often used instead tangible default notation reasons opaque . alphagamma = TRUE possible use beta diversity notation terms alpha average alpha diversity gamma gamma diversity two compared sites. terms calculated alpha = (+B)/2, gamma = +B-J delta = abs(-B)/2. Terms B also available give alpha diversities individual compared sites. beta diversity terms may make sense binary terms (diversities expressed numbers species), calculated quadratic minimum terms well ( warning). Function chaodist similar designgist, uses terms U V Chao et al. (2005). terms supposed take account effects unseen species. U V scaled range \\(0 \\dots 1\\). take place B product U*V used place J designdist. Function chaodist can implement commonly used Chao et al. (2005) style dissimilarity: Function vegdist implements Jaccard-type Chao distance, documentation contains complete discussion calculation terms.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Design your own Dissimilarities — designdist","text":"designdist returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Design your own Dissimilarities — designdist","text":"Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005) new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence--absence data. J. Animal Ecol. 72, 367--382. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Design your own Dissimilarities — designdist","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Design your own Dissimilarities — designdist","text":"designdist use compiled code, based vectorized R code. designdist function can much faster vegdist, although latter uses compiled code. However, designdist skip missing values uses much memory calculations. use sum terms can numerically unstable. particularly, terms large, precision may lost. risk large number columns high, particularly large quadratic terms. precise calculations better use functions like dist vegdist robust numerical problems.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Design your own Dissimilarities — designdist","text":"","code":"data(BCI) ## Four ways of calculating the same Sørensen dissimilarity d0 <- vegdist(BCI, \"bray\", binary = TRUE) d1 <- designdist(BCI, \"(A+B-2*J)/(A+B)\") d2 <- designdist(BCI, \"(b+c)/(2*a+b+c)\", abcd = TRUE) d3 <- designdist(BCI, \"gamma/alpha - 1\", alphagamma = TRUE) ## Arrhenius dissimilarity: the value of z in the species-area model ## S = c*A^z when combining two sites of equal areas, where S is the ## number of species, A is the area, and c and z are model parameters. ## The A below is not the area (which cancels out), but number of ## species in one of the sites, as defined in designdist(). dis <- designdist(BCI, \"(log(A+B-J)-log(A+B)+log(2))/log(2)\") ## This can be used in clustering or ordination... ordiplot(cmdscale(dis)) #> species scores not available ## ... or in analysing beta diversity (without gradients) summary(dis) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2733 0.3895 0.4192 0.4213 0.4537 0.5906"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions extract statistics resemble deviance AIC result constrained correspondence analysis cca redundancy analysis rda. functions rarely needed directly, called step automatic model building. Actually, cca rda AIC functions certainly wrong.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# S3 method for cca deviance(object, ...) # S3 method for cca extractAIC(fit, scale = 0, k = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"object result constrained ordination (cca rda). fit fitted model constrained ordination. scale optional numeric specifying scale parameter model, see scale step. k numeric specifying \"weight\" equivalent degrees freedom (=:edf) part AIC formula. ... arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions find statistics resemble deviance AIC constrained ordination. Actually, constrained ordination methods log-Likelihood, means AIC deviance. Therefore use functions, use , trust . use functions, remains responsibility check adequacy result. deviance cca equal Chi-square residual data matrix fitting constraints. deviance rda defined residual sum squares. deviance function rda also used distance-based RDA dbrda. Function extractAIC mimics extractAIC.lm translating deviance AIC. little need call functions directly. However, called implicitly step function used automatic selection constraining variables. check resulting model criteria, statistics used unfounded. particular, penalty k properly defined, default k = 2 justified theoretically. continuous covariates, step function base model building magnitude eigenvalues, value k influences stopping point ( variables highest eigenvalues necessarily significant permutation tests anova.cca). also multi-class factors, value k capricious effect model building. step function pass arguments add1.cca drop1.cca, setting test = \"permutation\" provide permutation tests deletion addition can help judging validity model building.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"deviance functions return “deviance”, extractAIC returns effective degrees freedom “AIC”.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Godínez-Domínguez, E. & Freire, J. (2003) Information-theoretic approach selection spatial temporal models community organization. Marine Ecology Progress Series 253, 17--24.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions unfounded untested used directly implicitly. Moreover, usual caveats using step valid.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# The deviance of correspondence analysis equals Chi-square data(dune) data(dune.env) chisq.test(dune) #> Warning: Chi-squared approximation may be incorrect #> #> \tPearson's Chi-squared test #> #> data: dune #> X-squared = 1449, df = 551, p-value < 2.2e-16 #> deviance(cca(dune)) #> [1] 1448.956 # Stepwise selection (forward from an empty model \"dune ~ 1\") ord <- cca(dune ~ ., dune.env) step(cca(dune ~ 1, dune.env), scope = formula(ord)) #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC #> + Moisture 3 86.608 #> + Management 3 86.935 #> + A1 1 87.411 #> 87.657 #> + Manure 4 88.832 #> + Use 2 89.134 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC #> 86.608 #> + Management 3 86.813 #> + A1 1 86.992 #> + Use 2 87.259 #> + Manure 4 87.342 #> - Moisture 3 87.657 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #>"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":null,"dir":"Reference","previous_headings":"","what":"Morisita index of intraspecific aggregation — dispindmorisita","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Calculates Morisita index dispersion, standardized index values, called clumpedness uniform indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"dispindmorisita(x, unique.rm = FALSE, crit = 0.05, na.rm = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"x community data matrix, sites (samples) rows species columns. unique.rm logical, TRUE, unique species (occurring one sample) removed result. crit two-sided p-value used calculate critical Chi-squared values. na.rm logical. missing values (including NaN) omitted calculations?","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita index dispersion defined (Morisita 1959, 1962): Imor = n * (sum(xi^2) - sum(xi)) / (sum(xi)^2 - sum(xi)) \\(xi\\) count individuals sample \\(\\), \\(n\\) number samples (\\(= 1, 2, \\ldots, n\\)). \\(Imor\\) values 0 \\(n\\). uniform (hyperdispersed) patterns value falls 0 1, clumped patterns falls 1 \\(n\\). increasing sample sizes (.e. joining neighbouring quadrats), \\(Imor\\) goes \\(n\\) quadrat size approaches clump size. random patterns, \\(Imor = 1\\) counts samples follow Poisson frequency distribution. deviation random expectation (null hypothesis) can tested using critical values Chi-squared distribution \\(n-1\\) degrees freedom. Confidence intervals around 1 can calculated clumped \\(Mclu\\) uniform \\(Muni\\) indices (Hairston et al. 1971, Krebs 1999) (Chi2Lower Chi2Upper refers e.g. 0.025 0.975 quantile values Chi-squared distribution \\(n-1\\) degrees freedom, respectively, crit = 0.05): Mclu = (Chi2Lower - n + sum(xi)) / (sum(xi) - 1) Muni = (Chi2Upper - n + sum(xi)) / (sum(xi) - 1) Smith-Gill (1975) proposed scaling Morisita index [0, n] interval [-1, 1], setting -0.5 0.5 values confidence limits around random distribution rescaled value 0. rescale Morisita index, one following four equations apply calculate standardized index \\(Imst\\): () Imor >= Mclu > 1: Imst = 0.5 + 0.5 (Imor - Mclu) / (n - Mclu), (b) Mclu > Imor >= 1: Imst = 0.5 (Imor - 1) / (Mclu - 1), (c) 1 > Imor > Muni: Imst = -0.5 (Imor - 1) / (Muni - 1), (d) 1 > Muni > Imor: Imst = -0.5 + 0.5 (Imor - Muni) / Muni.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Returns data frame many rows number columns input data, four columns. Columns : imor unstandardized Morisita index, mclu clumpedness index, muni uniform index, imst standardized Morisita index, pchisq Chi-squared based probability null hypothesis random expectation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita, M. 1959. Measuring dispersion individuals analysis distributional patterns. Mem. Fac. Sci. Kyushu Univ. Ser. E 2, 215--235. Morisita, M. 1962. Id-index, measure dispersion individuals. Res. Popul. Ecol. 4, 1--7. Smith-Gill, S. J. 1975. Cytophysiological basis disruptive pigmentary patterns leopard frog, Rana pipiens. II. Wild type mutant cell specific patterns. J. Morphol. 146, 35--54. Hairston, N. G., Hill, R. Ritte, U. 1971. interpretation aggregation patterns. : Patil, G. P., Pileou, E. C. Waters, W. E. eds. Statistical Ecology 1: Spatial Patterns Statistical Distributions. Penn. State Univ. Press, University Park. Krebs, C. J. 1999. Ecological Methodology. 2nd ed. Benjamin Cummings Publishers.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"common error found several papers standardizing case (b), denominator given Muni - 1. results hiatus [0, 0.5] interval standardized index. root typo book Krebs (1999), see Errata book (Page 217, https://www.zoology.ubc.ca/~krebs/downloads/errors_2nd_printing.pdf).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"data(dune) x <- dispindmorisita(dune) x #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Chenalbu NaN Inf -Inf NaN NaN #> Cirsarve 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Empenigr 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 y <- dispindmorisita(dune, unique.rm = TRUE) y #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 dim(x) ## with unique species #> [1] 30 5 dim(y) ## unique species removed #> [1] 27 5"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":null,"dir":"Reference","previous_headings":"","what":"Dispersion-based weighting of species counts — dispweight","title":"Dispersion-based weighting of species counts — dispweight","text":"Transform abundance data downweighting species overdispersed Poisson error.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"dispweight(comm, groups, nsimul = 999, nullmodel = \"c0_ind\", plimit = 0.05) gdispweight(formula, data, plimit = 0.05) # S3 method for dispweight summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dispersion-based weighting of species counts — dispweight","text":"comm Community data matrix. groups Factor describing group structure. missing, sites regarded belonging one group. NA values allowed. nsimul Number simulations. nullmodel nullmodel used commsim within groups. default follows Clarke et al. (2006). plimit Downweight species \\(p\\)-value limit. formula, data Formula left-hand side community data frame right-hand side gives explanatory variables. explanatory variables found data frame given data parent frame. object Result object dispweight gdispweight. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dispersion-based weighting of species counts — dispweight","text":"dispersion index (\\(D\\)) calculated ratio variance expected value species. species abundances follow Poisson distribution, expected dispersion \\(E(D) = 1\\), \\(D > 1\\), species overdispersed. inverse \\(1/D\\) can used downweight species abundances. Species downweighted overdispersion judged statistically significant (Clarke et al. 2006). Function dispweight implements original procedure Clarke et al. (2006). one factor can used group sites find species means. significance overdispersion assessed freely distributing individuals species within factor levels. achieved using nullmodel \"c0_ind\" (accords Clarke et al. 2006), nullmodels can used, though may meaningful (see commsim alternatives). species absent factor level, whole level ignored calculation overdispersion, number degrees freedom can vary among species. reduced number degrees freedom used divisor overdispersion \\(D\\), species higher dispersion hence lower weights transformation. Function gdispweight generalized parametric version dispweight. function based glm quasipoisson error family. glm model can used, including several factors continuous covariates. Function gdispweight uses test statistic dispweight (Pearson Chi-square), ignore factor levels species absent, number degrees freedom equal species. Therefore transformation weights can higher dispweight. gdispweight function evaluates significance overdispersion parametrically Chi-square distribution (pchisq). Functions dispweight gdispweight transform data, add information overdispersion weights attributes result. summary can used extract print information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dispersion-based weighting of species counts — dispweight","text":"Function returns transformed data following new attributes: D Dispersion statistic. df Degrees freedom species. p \\(p\\)-value Dispersion statistic \\(D\\). weights weights applied community data. nsimul Number simulations used assess \\(p\\)-value, NA simulations performed. nullmodel name commsim null model, NA simulations performed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dispersion-based weighting of species counts — dispweight","text":"Clarke, K. R., M. G. Chapman, P. J. Somerfield, H. R. Needham. 2006. Dispersion-based weighting species counts assemblage analyses. Marine Ecology Progress Series, 320, 11–27.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dispersion-based weighting of species counts — dispweight","text":"Eduard Szöcs eduardszoesc@gmail.com wrote original dispweight, Jari Oksanen significantly modified code, provided support functions developed gdispweight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"data(mite, mite.env) ## dispweight and its summary mite.dw <- with(mite.env, dispweight(mite, Shrub, nsimul = 99)) ## IGNORE_RDIFF_BEGIN summary(mite.dw) #> Dispersion Weight Df Pr(Disp.) #> Brachy 9.6908 0.1031909 67 0.01 ** #> PHTH 3.2809 0.3047900 49 0.01 ** #> HPAV 6.5263 0.1532264 67 0.01 ** #> RARD 6.0477 0.1653525 49 0.01 ** #> SSTR 2.2619 0.4421053 49 0.01 ** #> Protopl 5.4229 0.1844031 49 0.01 ** #> MEGR 4.5354 0.2204860 67 0.01 ** #> MPRO 1.2687 0.7882353 67 0.04 * #> TVIE 2.5956 0.3852706 67 0.01 ** #> HMIN 10.0714 0.0992906 67 0.01 ** #> HMIN2 7.5674 0.1321466 49 0.01 ** #> NPRA 2.6743 0.3739344 67 0.01 ** #> TVEL 9.6295 0.1038474 49 0.01 ** #> ONOV 11.3628 0.0880064 67 0.01 ** #> SUCT 8.7372 0.1144533 67 0.01 ** #> LCIL 129.4436 0.0077254 67 0.01 ** #> Oribatl1 4.1250 0.2424248 67 0.01 ** #> Ceratoz1 1.7150 0.5830768 67 0.01 ** #> PWIL 2.2943 0.4358538 67 0.01 ** #> Galumna1 2.8777 0.3474943 49 0.01 ** #> Stgncrs2 3.8242 0.2614953 49 0.01 ** #> HRUF 1.7575 0.5690021 67 0.01 ** #> Trhypch1 14.9225 0.0670128 67 0.01 ** #> PPEL 1.3628 1.0000000 49 0.07 . #> NCOR 2.5875 0.3864771 67 0.01 ** #> SLAT 2.7857 0.3589744 49 0.01 ** #> FSET 4.8901 0.2044944 49 0.01 ** #> Lepidzts 1.6577 0.6032360 49 0.02 * #> Eupelops 1.4611 0.6844033 67 0.04 * #> Miniglmn 1.6505 0.6058733 49 0.01 ** #> LRUG 12.0658 0.0828790 67 0.01 ** #> PLAG2 3.2403 0.3086090 67 0.01 ** #> Ceratoz3 3.5125 0.2846947 67 0.01 ** #> Oppiminu 3.1680 0.3156525 67 0.01 ** #> Trimalc2 10.5927 0.0944046 67 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Based on 99 simulations on 'c0_ind' nullmodel ## IGNORE_RDIFF_END ## generalized dispersion weighting mite.dw <- gdispweight(mite ~ Shrub + WatrCont, data = mite.env) rda(mite.dw ~ Shrub + WatrCont, data = mite.env) #> Call: rda(formula = mite.dw ~ Shrub + WatrCont, data = mite.env) #> #> Inertia Proportion Rank #> Total 38.1640 1.0000 #> Constrained 9.2129 0.2414 3 #> Unconstrained 28.9511 0.7586 35 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 7.986 0.748 0.480 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 5.886 3.634 2.791 2.592 1.932 1.573 1.210 1.078 #> (Showing 8 of 35 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":null,"dir":"Reference","previous_headings":"","what":"Connectedness of Dissimilarities — distconnected","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected finds groups connected disregarding dissimilarities threshold NA. function can used find groups can ordinated together transformed stepacross. Function .shared returns logical dissimilarity object, TRUE means sites species common. minimal structure distconnected can used set missing values dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"distconnected(dis, toolong = 1, trace = TRUE) no.shared(x)"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Connectedness of Dissimilarities — distconnected","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. trace Summarize results distconnected x Community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Connectedness of Dissimilarities — distconnected","text":"Data sets disconnected sample plots groups sample plots share species sites groups sites. data sets sensibly ordinated unconstrained method subsets related . instance, correspondence analysis polarize subsets eigenvalue 1. Neither can dissimilarities transformed stepacross, path points, result contain NAs. Function distconnected find subsets dissimilarity matrices. function return grouping vector can used sub-setting data. data connected, result vector \\(1\\)s. connectedness two points can defined either threshold toolong using input dissimilarities NAs. Function .shared returns dist structure value TRUE two sites nothing common, value FALSE least one shared species. minimal structure can analysed distconnected. function can used select dissimilarities shared species indices fixed upper limit. Function distconnected uses depth-first search (Sedgewick 1990).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected returns vector observations using integers identify connected groups. data connected, values 1. Function .shared returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Connectedness of Dissimilarities — distconnected","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Connectedness of Dissimilarities — distconnected","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"## There are no disconnected data in vegan, and the following uses an ## extremely low threshold limit for connectedness. This is for ## illustration only, and not a recommended practice. data(dune) dis <- vegdist(dune) gr <- distconnected(dis, toolong=0.4) #> Connectivity of distance matrix with threshold dissimilarity 0.4 #> Data are disconnected: 6 groups #> Groups sizes #> 1 2 3 4 5 6 #> 1 11 2 4 1 1 # Make sites with no shared species as NA in Manhattan dissimilarities dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":null,"dir":"Reference","previous_headings":"","what":"Ecological Diversity Indices — diversity","title":"Ecological Diversity Indices — diversity","text":"Shannon, Simpson, Fisher diversity indices species richness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ecological Diversity Indices — diversity","text":"","code":"diversity(x, index = \"shannon\", groups, equalize.groups = FALSE, MARGIN = 1, base = exp(1)) simpson.unb(x, inverse = FALSE) fisher.alpha(x, MARGIN = 1, ...) specnumber(x, groups, MARGIN = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ecological Diversity Indices — diversity","text":"x Community data, matrix-like object vector. index Diversity index, one \"shannon\", \"simpson\" \"invsimpson\". MARGIN Margin index computed. base logarithm base used shannon. inverse Use inverse Simpson similarly diversity(x, \"invsimpson\"). groups grouping factor: given, finds diversity communities pooled groups. equalize.groups Instead observed abundances, standardize communities unit total. ... Parameters passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ecological Diversity Indices — diversity","text":"Shannon Shannon--Weaver (Shannon--Wiener) index defined \\(H' = -\\sum_i p_i \\log_{b} p_i\\), \\(p_i\\) proportional abundance species \\(\\) \\(b\\) base logarithm. popular use natural logarithms, argue base \\(b = 2\\) (makes sense, real difference). variants Simpson's index based \\(D = \\sum p_i^2\\). Choice simpson returns \\(1-D\\) invsimpson returns \\(1/D\\). simpson.unb finds unbiased Simpson indices discrete samples (Hurlbert 1971, eq. 5). less sensitive sample size basic Simpson indices. unbiased indices can calculated data integer counts. diversity function can find total (gamma) diversity pooled communities argument groups. average alpha diversity can found mean diversities groups, difference ratio estimate beta diversity (see Examples). pooling can based either observed abundancies, communities can equalized unit total pooling; see Jost (2007) discussion. Functions adipart multipart provide canned alternatives estimating alpha, beta gamma diversities hierarchical settings. fisher.alpha estimates \\(\\alpha\\) parameter Fisher's logarithmic series (see fisherfit). estimation possible genuine counts individuals. None diversity indices usable empty sampling units without species, indices can give numeric value. Filtering cases left user. Function specnumber finds number species. MARGIN = 2, finds frequencies species. groups given, finds total number species group (see example finding one kind beta diversity option). Better stories can told Simpson's index Shannon's index, still grander narratives rarefaction (Hurlbert 1971). However, indices closely related (Hill 1973), reason despise one others (graduate student, drag , obey Professor's orders). particular, exponent Shannon index linearly related inverse Simpson (Hill 1973) although former may sensitive rare species. Moreover, inverse Simpson asymptotically equal rarefied species richness sample two individuals, Fisher's \\(\\alpha\\) similar inverse Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ecological Diversity Indices — diversity","text":"vector diversity indices numbers species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ecological Diversity Indices — diversity","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12, 42--58. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ecological Diversity Indices — diversity","text":"Jari Oksanen Bob O'Hara (fisher.alpha).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ecological Diversity Indices — diversity","text":"","code":"data(BCI, BCI.env) H <- diversity(BCI) simp <- diversity(BCI, \"simpson\") invsimp <- diversity(BCI, \"inv\") ## Unbiased Simpson unbias.simp <- simpson.unb(BCI) ## Fisher alpha alpha <- fisher.alpha(BCI) ## Plot all pairs(cbind(H, simp, invsimp, unbias.simp, alpha), pch=\"+\", col=\"blue\") ## Species richness (S) and Pielou's evenness (J): S <- specnumber(BCI) ## rowSums(BCI > 0) does the same... J <- H/log(S) ## beta diversity defined as gamma/alpha - 1: ## alpha is the average no. of species in a group, and gamma is the ## total number of species in the group (alpha <- with(BCI.env, tapply(specnumber(BCI), Habitat, mean))) #> OldHigh OldLow OldSlope Swamp Young #> 85.75000 91.76923 91.58333 94.00000 90.00000 (gamma <- with(BCI.env, specnumber(BCI, Habitat))) #> OldHigh OldLow OldSlope Swamp Young #> 158 210 183 128 117 gamma/alpha - 1 #> OldHigh OldLow OldSlope Swamp Young #> 0.8425656 1.2883487 0.9981802 0.3617021 0.3000000 ## similar calculations with Shannon diversity (alpha <- with(BCI.env, tapply(diversity(BCI), Habitat, mean))) # average #> OldHigh OldLow OldSlope Swamp Young #> 3.638598 3.876413 3.887122 4.003780 3.246729 (gamma <- with(BCI.env, diversity(BCI, groups=Habitat))) # pooled #> OldHigh OldLow OldSlope Swamp Young #> 3.873186 4.284972 4.212098 4.164335 3.387536 ## additive beta diversity based on Shannon index gamma-alpha #> OldHigh OldLow OldSlope Swamp Young #> 0.2345878 0.4085595 0.3249760 0.1605548 0.1408068"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and Environment in Dutch Dune Meadows. — dune","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune meadow vegetation data, dune, cover class values 30 species 20 sites. corresponding environmental data frame dune.env following entries:","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune data frame observations 30 species 20 sites. species names abbreviated 4+4 letters (see make.cepnames). following names changed original source (Jongman et al. 1987): Leontodon autumnalis Scorzoneroides, Potentilla palustris Comarum. dune.env data frame 20 observations following 5 variables: A1: numeric vector thickness soil A1 horizon. Moisture: ordered factor levels: 1 < 2 < 4 < 5. Management: factor levels: BF (Biological farming), HF (Hobby farming), NM (Nature Conservation Management), SF (Standard Farming). Use: ordered factor land-use levels: Hayfield < Haypastu < Pasture. Manure: ordered factor levels: 0 < 1 < 2 < 3 < 4.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"Jongman, R.H.G, ter Braak, C.J.F & van Tongeren, O.F.R. (1987). Data Analysis Community Landscape Ecology. Pudoc, Wageningen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":null,"dir":"Reference","previous_headings":"","what":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"Classification table species dune data set.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"dune.taxon data frame 30 species (rows) classified five taxonomic levels (columns). dune.phylodis dist object estimated coalescence ages extracted doi:10.5061/dryad.63q27 (Zanne et al. 2014) using tools packages ape phylobase.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"families orders based APG IV (2016) vascular plants Hill et al. (2006) mosses. higher levels (superorder subclass) based Chase & Reveal (2009). Chase & Reveal (2009) treat Angiosperms mosses subclasses class Equisetopsida (land plants), brylogists traditionally used much inflated levels adjusted match Angiosperm classification.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"APG IV [Angiosperm Phylogeny Group] (2016) update Angiosperm Phylogeny Group classification orders families flowering plants: APG IV. Bot. J. Linnean Soc. 181: 1--20. Chase, M.W. & Reveal, J. L. (2009) phylogenetic classification land plants accompany APG III. Bot. J. Linnean Soc. 161: 122--127. Hill, M.O et al. (2006) annotated checklist mosses Europe Macaronesia. J. Bryology 28: 198--267. Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506: 89--92.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Eigenvalues from an Ordination Object — eigenvals","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Function extracts eigenvalues object . Many multivariate methods return objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"eigenvals(x, ...) # S3 method for cca eigenvals(x, model = c(\"all\", \"unconstrained\", \"constrained\"), constrained = NULL, ...) # S3 method for decorana eigenvals(x, kind = c(\"additive\", \"axiswise\", \"decorana\"), ...) # S3 method for eigenvals summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"x object extract eigenvalues. object eigenvals result object. model eigenvalues return objects inherit class \"cca\" . constrained Return constrained eigenvalues. Deprecated vegan 2.5-0. Use model instead. kind Kind eigenvalues returned decorana. \"additive\" eigenvalues can used reporting importances components summary. \"axiswise\" gives non-additive eigenvalues, \"decorana\" decorana values (see decorana details). ... arguments functions (usually ignored)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"generic function methods cca, wcmdscale, pcnm, prcomp, princomp, dudi (ade4), pca pco ( labdsv) result objects. default method also extracts eigenvalues result looks like eigen svd. Functions prcomp princomp contain square roots eigenvalues called standard deviations, eigenvals function returns squares. Function svd contains singular values, function eigenvals returns squares. constrained ordination methods cca, rda capscale function returns constrained unconstrained eigenvalues concatenated one vector, partial component ignored. However, argument constrained = TRUE constrained eigenvalues returned. summary eigenvals result returns eigenvalues, proportion explained cumulative proportion explained. result object can negative eigenvalues (wcmdscale, dbrda, pcnm) correspond imaginary axes Euclidean mapping non-Euclidean distances (Gower 1985). case real axes (corresponding positive eigenvalues) \"explain\" proportion >1 total variation, negative eigenvalues bring cumulative proportion 1. capscale find positive eigenvalues used finding proportions. decorana importances cumulative proportions reported kind = \"additive\", alternatives add total inertia input data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"object class \"eigenvals\", vector eigenvalues. summary method returns object class \"summary.eigenvals\", matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Gower, J. C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ev <- eigenvals(mod) ev #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 CA5 #> 0.3615566 0.1699600 0.1126167 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 #> CA6 CA7 CA8 CA9 CA10 CA11 CA12 CA13 #> 0.1002670 0.0772991 0.0536938 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 #> CA14 CA15 CA16 CA17 CA18 CA19 CA20 #> 0.0101910 0.0094701 0.0055090 0.0030529 0.0025118 0.0019485 0.0005178 summary(ev) #> Importance of components: #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 #> Eigenvalue 0.3616 0.16996 0.11262 0.3500 0.2201 0.18507 0.15512 #> Proportion Explained 0.1736 0.08159 0.05406 0.1680 0.1056 0.08884 0.07446 #> Cumulative Proportion 0.1736 0.25514 0.30920 0.4772 0.5829 0.67172 0.74618 #> CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> Eigenvalue 0.13511 0.10027 0.07730 0.05369 0.03656 0.03509 0.02823 #> Proportion Explained 0.06485 0.04813 0.03711 0.02577 0.01755 0.01684 0.01355 #> Cumulative Proportion 0.81104 0.85917 0.89627 0.92205 0.93960 0.95644 0.96999 #> CA12 CA13 CA14 CA15 CA16 CA17 #> Eigenvalue 0.017065 0.012247 0.010191 0.009470 0.005509 0.003053 #> Proportion Explained 0.008192 0.005879 0.004892 0.004546 0.002644 0.001465 #> Cumulative Proportion 0.978183 0.984062 0.988954 0.993500 0.996145 0.997610 #> CA18 CA19 CA20 #> Eigenvalue 0.002512 0.0019485 0.0005178 #> Proportion Explained 0.001206 0.0009353 0.0002486 #> Cumulative Proportion 0.998816 0.9997514 1.0000000 ## choose which eignevalues to return eigenvals(mod, model = \"unconstrained\") #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 0.1002670 0.0772991 0.0536938 #> CA9 CA10 CA11 CA12 CA13 CA14 CA15 CA16 #> 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 0.0101910 0.0094701 0.0055090 #> CA17 CA18 CA19 CA20 #> 0.0030529 0.0025118 0.0019485 0.0005178"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fits an Environmental Vector or Factor onto an Ordination — envfit","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"function fits environmental vectors factors onto ordination. projections points onto vectors maximum correlation corresponding environmental variables, factors show averages factor levels. continuous varaibles equal fitting linear trend surface (plane 2D) variable (see ordisurf); trend surface can presented showing gradient (direction steepest increase) using arrow. environmental variables dependent variables explained ordination scores, dependent variable analysed separately.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"# S3 method for default envfit(ord, env, permutations = 999, strata = NULL, choices=c(1,2), display = \"sites\", w = weights(ord, display), na.rm = FALSE, ...) # S3 method for formula envfit(formula, data, ...) # S3 method for envfit plot(x, choices = c(1,2), labels, arrow.mul, at = c(0,0), axis = FALSE, p.max = NULL, col = \"blue\", bg, add = TRUE, ...) # S3 method for envfit scores(x, display, choices, arrow.mul=1, tidy = FALSE, ...) vectorfit(X, P, permutations = 0, strata = NULL, w, ...) factorfit(X, P, permutations = 0, strata = NULL, w, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"ord ordination object structure ordination scores can extracted (including data frame matrix scores). env Data frame, matrix vector environmental variables. variables can mixed type (factors, continuous variables) data frames. X Matrix data frame ordination scores. P Data frame, matrix vector environmental variable(s). must continuous vectorfit factors characters factorfit. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Set permutations = 0 skip permutations. formula, data Model formula data. na.rm Remove points missing values ordination scores environmental variables. operation casewise: whole row data removed missing value na.rm = TRUE. x result object envfit. ordiArrowMul ordiArrowTextXY must two-column matrix ( matrix-like object) containing coordinates arrow heads two plot axes, methods extract structure envfit results. choices Axes plotted. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable scores (\"vectors\" \"factors\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Change plotting labels. argument list elements vectors factors give new plotting labels. either elements omitted, default labels used. one type elements (vectors factors), labels can given vector. default labels can displayed labels command. arrow.mul Multiplier vector lengths. arrows automatically scaled similarly plot.cca given plot add = TRUE. However, scores can used adjust arrow lengths plot function used. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. axis Plot axis showing scaling fitted arrows. p.max Maximum estimated \\(P\\) value displayed variables. must calculate \\(P\\) values setting permutations use option. col Colour plotting. bg Background colour labels. bg set, labels displayed ordilabel instead text. See Examples using semitransparent background. add Results added existing ordination plot. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. display fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). scores function either \"vectors\" \"factors\" (synonyms \"bp\" \"cn\", resp.). w Weights used fitting (concerns mainly cca decorana results nonconstant weights). ... Parameters passed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Function envfit finds vectors factor averages environmental variables. Function plot.envfit adds ordination diagram. X data.frame, envfit uses factorfit factor variables vectorfit variables. X matrix vector, envfit uses vectorfit. Alternatively, model can defined simplified model formula, left hand side must ordination result object matrix ordination scores, right hand side lists environmental variables. formula interface can used easier selection /transformation environmental variables. main effects analysed even interaction terms defined formula. ordination results extracted scores extra arguments passed scores. fitted models apply results defined extracting scores using envfit. instance, scaling constrained ordination (see scores.rda, scores.cca) must set way envfit plot ordination results (see Examples). printed output continuous variables (vectors) gives direction cosines coordinates heads unit length vectors. plot scaled correlation (square root column r2) “weak” predictors shorter arrows “strong” predictors. can see scaled relative lengths using command scores. plotted (scaled) arrows adjusted current graph using constant multiplier: keep relative r2-scaled lengths arrows tries fill current plot. can see multiplier using ordiArrowMul(result_of_envfit), set argument arrow.mul. Functions vectorfit factorfit can called directly. Function vectorfit finds directions ordination space towards environmental vectors change rapidly maximal correlations ordination configuration. Function factorfit finds averages ordination scores factor levels. Function factorfit treats ordered unordered factors similarly. permutations \\(> 0\\), significance fitted vectors factors assessed using permutation environmental variables. goodness fit statistic squared correlation coefficient (\\(r^2\\)). factors defined \\(r^2 = 1 - ss_w/ss_t\\), \\(ss_w\\) \\(ss_t\\) within-group total sums squares. See permutations additional details permutation tests Vegan. User can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. fitted vectors similar biplot arrows constrained ordination fitted LC scores (display = \"lc\") set scaling = \"species\" (see scores.cca). weighted fitting gives similar results biplot arrows class centroids cca. lengths arrows fitted vectors automatically adjusted physical size plot, arrow lengths compared across plots. similar scaling arrows, must explicitly set arrow.mul argument plot command; see ordiArrowMul ordiArrowTextXY. results can accessed scores.envfit function returns either fitted vectors scaled correlation coefficient centroids fitted environmental variables, named list .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Functions vectorfit factorfit return lists classes vectorfit factorfit print method. result object following items: arrows Arrow endpoints vectorfit. arrows scaled unit length. centroids Class centroids factorfit. r Goodness fit statistic: Squared correlation coefficient permutations Number permutations. control list control values permutations returned function . pvals Empirical P-values variable. Function envfit returns list class envfit results vectorfit envfit items. Function plot.envfit scales vectors correlation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Jari Oksanen. permutation test derives code suggested Michael Scroggie.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Fitted vectors become method choice displaying environmental variables ordination. Indeed, optimal way presenting environmental variables Constrained Correspondence Analysis cca, since linear constraints. unconstrained ordination relation external variables ordination configuration may less linear, therefore methods arrows may useful. simplest adjust plotting symbol sizes (cex, symbols) environmental variables. Fancier methods involve smoothing regression methods abound R, ordisurf provides wrapper .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"data(varespec, varechem) library(MASS) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2276827 #> Run 2 stress 0.2204286 #> Run 3 stress 0.1967393 #> Run 4 stress 0.2066032 #> Run 5 stress 0.2234314 #> Run 6 stress 0.1869637 #> Run 7 stress 0.18584 #> Run 8 stress 0.18458 #> ... Procrustes: rmse 0.04935318 max resid 0.1575045 #> Run 9 stress 0.2178486 #> Run 10 stress 0.195049 #> Run 11 stress 0.2300388 #> Run 12 stress 0.1948413 #> Run 13 stress 0.1955836 #> Run 14 stress 0.2390981 #> Run 15 stress 0.2229031 #> Run 16 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04162605 max resid 0.1518104 #> Run 17 stress 0.2390411 #> Run 18 stress 0.1825658 #> ... Procrustes: rmse 1.358437e-05 max resid 2.641802e-05 #> ... Similar to previous best #> Run 19 stress 0.2092456 #> Run 20 stress 0.1982376 #> *** Best solution repeated 1 times (fit <- envfit(ord, varechem, perm = 999)) #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05732 -0.99836 0.2536 0.046 * #> P 0.61973 0.78481 0.1938 0.093 . #> K 0.76646 0.64229 0.1809 0.121 #> Ca 0.68520 0.72835 0.4119 0.008 ** #> Mg 0.63253 0.77454 0.4270 0.004 ** #> S 0.19140 0.98151 0.1752 0.123 #> Al -0.87159 0.49023 0.5269 0.001 *** #> Fe -0.93602 0.35195 0.4451 0.002 ** #> Mn 0.79871 -0.60172 0.5231 0.002 ** #> Zn 0.61756 0.78652 0.1879 0.107 #> Mo -0.90306 0.42951 0.0609 0.539 #> Baresoil 0.92488 -0.38025 0.2508 0.041 * #> Humdepth 0.93283 -0.36032 0.5201 0.002 ** #> pH -0.64798 0.76165 0.2308 0.070 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> scores(fit, \"vectors\") #> NMDS1 NMDS2 #> N -0.02886657 -0.5027884 #> P 0.27284738 0.3455254 #> K 0.32602389 0.2732093 #> Ca 0.43973450 0.4674260 #> Mg 0.41334179 0.5061403 #> S 0.08011287 0.4108193 #> Al -0.63269483 0.3558584 #> Fe -0.62444176 0.2347947 #> Mn 0.57765881 -0.4351900 #> Zn 0.26771371 0.3409563 #> Mo -0.22293623 0.1060315 #> Baresoil 0.46316234 -0.1904201 #> Humdepth 0.67271099 -0.2598473 #> pH -0.31130599 0.3659163 plot(ord) plot(fit) plot(fit, p.max = 0.05, col = \"red\") ## Adding fitted arrows to CCA. We use \"lc\" scores, and hope ## that arrows are scaled similarly in cca and envfit plots ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, type=\"p\") fit <- envfit(ord, varechem, perm = 999, display = \"lc\") plot(fit, p.max = 0.05, col = \"red\") ## 'scaling' must be set similarly in envfit and in ordination plot plot(ord, type = \"p\", scaling = \"sites\") fit <- envfit(ord, varechem, perm = 0, display = \"lc\", scaling = \"sites\") plot(fit, col = \"red\") ## Class variables, formula interface, and displaying the ## inter-class variability with ordispider, and semitransparent ## white background for labels (semitransparent colours are not ## supported by all graphics devices) data(dune) data(dune.env) ord <- cca(dune) fit <- envfit(ord ~ Moisture + A1, dune.env, perm = 0) plot(ord, type = \"n\") with(dune.env, ordispider(ord, Moisture, col=\"skyblue\")) with(dune.env, points(ord, display = \"sites\", col = as.numeric(Moisture), pch=16)) plot(fit, cex=1.2, axis=TRUE, bg = rgb(1, 1, 1, 0.5)) ## Use shorter labels for factor centroids labels(fit) #> $vectors #> [1] \"A1\" #> #> $factors #> [1] \"Moisture1\" \"Moisture2\" \"Moisture4\" \"Moisture5\" #> plot(ord) plot(fit, labels=list(factors = paste(\"M\", c(1,2,4,5), sep = \"\")), bg = rgb(1,1,0,0.5))"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds minimum (\\(q^*\\)) evenness profile based Tsallis entropy. scale factor entropy represents specific weighting species relative frequencies leads minimum evenness community (Mendes et al. 2008).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"eventstar(x, qmax = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"x community matrix numeric vector. qmax Maximum scale parameter Tsallis entropy used finding minimum Tsallis based evenness range c(0, qmax).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds characteristic value scale parameter \\(q\\) Tsallis entropy corresponding minimum evenness (equitability) profile based Tsallis entropy. value proposed Mendes et al. (2008) \\(q^*\\). \\(q^\\ast\\) index represents scale parameter one parameter Tsallis diversity family leads greatest deviation maximum equitability given relative abundance vector community. value \\(q^\\ast\\) found identifying minimum evenness profile scaling factor \\(q\\) one-dimensional minimization. evenness profile known convex function, guaranteed underlying optimize function find unique solution range c(0, qmax). scale parameter value \\(q^\\ast\\) used find corresponding values diversity (\\(H_{q^\\ast}\\)), evenness (\\(H_{q^\\ast}(\\max)\\)), numbers equivalent (\\(D_{q^\\ast}\\)). calculation details, see tsallis Examples . Mendes et al. (2008) advocated use \\(q^\\ast\\) corresponding diversity, evenness, Hill numbers, unique value representing diversity profile, positively associated rare species community, thus potentially useful indicator certain relative abundance distributions communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"data frame columns: qstar scale parameter value \\(q\\ast\\) corresponding minimum value Tsallis based evenness profile. Estar Value evenness based normalized Tsallis entropy \\(q^\\ast\\). Hstar Value Tsallis entropy \\(q^\\ast\\). Dstar Value Tsallis entropy \\(q^\\ast\\) converted numbers equivalents (also called Hill numbers, effective number species, ‘true’ diversity; cf. Jost 2007). See tsallis calculation details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Mendes, R.S., Evangelista, L.R., Thomaz, S.M., Agostinho, .. Gomes, L.C. (2008) unified index measure ecological diversity species rarity. Ecography 31, 450--456. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439. Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Values \\(q^\\ast\\) found Mendes et al. (2008) ranged 0.56 1.12 presenting low variability, interval 0 5 safely encompass possibly expected \\(q^\\ast\\) values practice, profiling evenness changing value qmax argument advised output values near range limits found.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Eduardo Ribeiro Cunha edurcunha@gmail.com Heloisa Beatriz Antoniazi Evangelista helobeatriz@gmail.com, technical input Péter Sólymos.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"data(BCI) (x <- eventstar(BCI[1:5,])) #> qstar Estar Hstar Dstar #> 1 0.6146389 0.4263687 10.524584 67.03551 #> 2 0.6249249 0.4080263 9.534034 57.66840 #> 3 0.6380858 0.4062032 9.225458 57.69174 #> 4 0.6245808 0.4062213 10.140189 65.50247 #> 5 0.6404825 0.4219957 9.828138 66.96440 ## profiling y <- as.numeric(BCI[10,]) (z <- eventstar(y)) #> qstar Estar Hstar Dstar #> 1 0.6372529 0.4117557 9.546332 61.77715 q <- seq(0, 2, 0.05) Eprof <- tsallis(y, scales=q, norm=TRUE) Hprof <- tsallis(y, scales=q) Dprof <- tsallis(y, scales=q, hill=TRUE) opar <- par(mfrow=c(3,1)) plot(q, Eprof, type=\"l\", main=\"Evenness\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, norm=TRUE), col=2) plot(q, Hprof, type=\"l\", main=\"Diversity\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar), col=2) plot(q, Dprof, type=\"l\", main=\"Effective number of species\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, hill=TRUE), col=2) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Function fisherfit fits Fisher's logseries abundance data. Function prestonfit groups species frequencies doubling octave classes fits Preston's lognormal model, function prestondistr fits truncated lognormal model without pooling data octaves.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"fisherfit(x, ...) prestonfit(x, tiesplit = TRUE, ...) prestondistr(x, truncate = -1, ...) # S3 method for prestonfit plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", line.col = \"red\", lwd = 2, ...) # S3 method for prestonfit lines(x, line.col = \"red\", lwd = 2, ...) veiledspec(x, ...) as.fisher(x, ...) # S3 method for fisher plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", kind = c(\"bar\", \"hiplot\", \"points\", \"lines\"), add = FALSE, ...) as.preston(x, tiesplit = TRUE, ...) # S3 method for preston plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", ...) # S3 method for preston lines(x, xadjust = 0.5, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"x Community data vector fitting functions result object plot functions. tiesplit Split frequencies \\(1, 2, 4, 8\\) etc adjacent octaves. truncate Truncation point log-Normal model, log2 units. Default value \\(-1\\) corresponds left border zero Octave. choice strongly influences fitting results. xlab, ylab Labels x y axes. bar.col Colour data bars. line.col Colour fitted line. lwd Width fitted line. kind Kind plot drawn: \"bar\" similar bar plot plot.fisherfit, \"hiplot\" draws vertical lines plot(..., type=\"h\"), \"points\" \"lines\" obvious. add Add existing plot. xadjust Adjustment horizontal positions octaves. ... parameters passed functions. Ignored prestonfit tiesplit passed .preston prestondistr.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher's logarithmic series expected number species \\(f\\) \\(n\\) observed individuals \\(f_n = \\alpha x^n / n\\) (Fisher et al. 1943). estimation possible genuine counts individuals. parameter \\(\\alpha\\) used diversity index can estimated separate function fisher.alpha. parameter \\(x\\) taken nuisance parameter estimated separately taken \\(n/(n+\\alpha)\\). Helper function .fisher transforms abundance data Fisher frequency table. Diversity given NA communities one (zero) species: reliable way estimating diversity, even equations return bogus numeric value cases. Preston (1948) satisfied Fisher's model seemed imply infinite species richness, postulated rare species diminishing class species middle frequency scale. achieved collapsing higher frequency classes wider wider “octaves” doubling class limits: 1, 2, 3--4, 5--8, 9--16 etc. occurrences. seems Preston regarded frequencies 1, 2, 4, etc.. “tied” octaves (Williamson & Gaston 2005). means half species frequency 1 shown lowest octave, rest transferred second octave. Half species second octave transferred higher one well, usually large number species. practise makes data look lognormal reducing usually high lowest octaves. can achieved setting argument tiesplit = TRUE. tiesplit = FALSE frequencies split, ones lowest octave, twos second, etc. Williamson & Gaston (2005) discuss alternative definitions detail, consulted critical review log-Normal model. logseries data look like lognormal plotted Preston's way. expected frequency \\(f\\) abundance octave \\(o\\) defined \\(f_o = S_0 \\exp(-(\\log_2(o) - \\mu)^2/2/\\sigma^2)\\), \\(\\mu\\) location mode \\(\\sigma\\) width, \\(\\log_2\\) scale, \\(S_0\\) expected number species mode. lognormal model usually truncated left rare species observed. Function prestonfit fits truncated lognormal model second degree log-polynomial octave pooled data using Poisson ( tiesplit = FALSE) quasi-Poisson (tiesplit = TRUE) error. Function prestondistr fits left-truncated Normal distribution \\(\\log_2\\) transformed non-pooled observations direct maximization log-likelihood. Function prestondistr modelled function fitdistr can used alternative distribution models. functions common print, plot lines methods. lines function adds fitted curve octave range line segments showing location mode width (sd) response. Function .preston transforms abundance data octaves. Argument tiesplit influence fit prestondistr, influence barplot octaves. total extrapolated richness fitted Preston model can found function veiledspec. function accepts results prestonfit prestondistr. veiledspec called species count vector, internally use prestonfit. Function specpool provides alternative ways estimating number unseen species. fact, Preston's lognormal model seems truncated ends, may main reason result differ lognormal models fitted Rank--Abundance diagrams functions rad.lognormal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"function prestonfit returns object fitted coefficients, observed (freq) fitted (fitted) frequencies, string describing fitting method. Function prestondistr omits entry fitted. function fisherfit returns result nlm, item estimate \\(\\alpha\\). result object amended nuisance parameter item fisher observed data .fisher","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12: 42--58. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Williamson, M. & Gaston, K.J. (2005). lognormal distribution appropriate null hypothesis species--abundance distribution. Journal Animal Ecology 74, 409--422.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Bob O'Hara Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"data(BCI) mod <- fisherfit(BCI[5,]) mod #> #> Fisher log series model #> No. of species: 101 #> Fisher alpha: 37.96423 #> # prestonfit seems to need large samples mod.oct <- prestonfit(colSums(BCI)) mod.ll <- prestondistr(colSums(BCI)) mod.oct #> #> Preston lognormal model #> Method: Quasi-Poisson fit to octaves #> No. of species: 225 #> #> mode width S0 #> 4.885798 2.932690 32.022923 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.500000 16.00000 18.00000 19.000 30.00000 35.00000 31.00000 26.50000 #> Fitted 7.994154 13.31175 19.73342 26.042 30.59502 31.99865 29.79321 24.69491 #> 8 9 10 11 #> Observed 18.00000 13.00000 7.000000 2.0000 #> Fitted 18.22226 11.97021 7.000122 3.6443 #> mod.ll #> #> Preston lognormal model #> Method: maximized likelihood to log2 abundances #> No. of species: 225 #> #> mode width S0 #> 4.365002 2.753531 33.458185 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.50000 16.00000 18.00000 19.00000 30.00000 35.00000 31.00000 26.50000 #> Fitted 9.52392 15.85637 23.13724 29.58961 33.16552 32.58022 28.05054 21.16645 #> 8 9 10 11 #> Observed 18.00000 13.000000 7.000000 2.00000 #> Fitted 13.99829 8.113746 4.121808 1.83516 #> plot(mod.oct) lines(mod.ll, line.col=\"blue3\") # Different ## Smoothed density den <- density(log2(colSums(BCI))) lines(den$x, ncol(BCI)*den$y, lwd=2) # Fairly similar to mod.oct ## Extrapolated richness veiledspec(mod.oct) #> Extrapolated Observed Veiled #> 235.40577 225.00000 10.40577 veiledspec(mod.ll) #> Extrapolated Observed Veiled #> 230.931018 225.000000 5.931018"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Functions goodness inertcomp can used assess goodness fit individual sites species. Function vif.cca alias.cca can used analyse linear dependencies among constraints conditions. addition, diagnostic tools (see 'Details').","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"# S3 method for cca goodness(object, choices, display = c(\"species\", \"sites\"), model = c(\"CCA\", \"CA\"), summarize = FALSE, addprevious = FALSE, ...) inertcomp(object, display = c(\"species\", \"sites\"), unity = FALSE, proportional = FALSE) spenvcor(object) intersetcor(object) vif.cca(object) # S3 method for cca alias(object, names.only = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"object result object cca, rda, dbrda capscale. display Display \"species\" \"sites\". Species available dbrda capscale. choices Axes shown. Default show axes \"model\". model Show constrained (\"CCA\") unconstrained (\"CA\") results. summarize Show accumulated total. addprevious Add variation explained previous components statistic=\"explained\". model = \"CCA\" add conditioned (partialled ) variation, model = \"CA\" add conditioned constrained variation. give cumulative explanation previous components. unity Scale inertia components unit sum (sum items 1). proportional Give inertia components proportional corresponding total item (sum row 1). option takes precedence unity. names.Return names aliased variable(s) instead defining equations. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Function goodness gives cumulative proportion inertia accounted species chosen axes. proportions can assessed either species sites depending argument display, species available distance-based dbrda. function implemented capscale. Function inertcomp decomposes inertia partial, constrained unconstrained components site species. Legendre & De Cáceres (2012) called inertia components local contributions beta-diversity (LCBD) species contributions beta-diversity (SCBD), give relative contributions summing unity (argument unity = TRUE). interpretation, appropriate dissimilarity measures used dbrda appropriate standardization rda (Legendre & De Cáceres 2012). function implemented capscale. Function spenvcor finds -called “species -- environment correlation” (weighted) correlation weighted average scores linear combination scores. bad measure goodness ordination, sensitive extreme scores (like correlations ), sensitive overfitting using many constraints. Better models often poorer correlations. Function ordispider can show graphically. Function intersetcor finds -called “interset correlation” (weighted) correlation weighted averages scores constraints. defined contrasts used factor variables. bad measure since correlation. , focuses correlations single contrasts single axes instead looking multivariate relationship. Fitted vectors (envfit) provide better alternative. Biplot scores (see scores.cca) multivariate alternative (weighted) correlation linear combination scores constraints. Function vif.cca gives variance inflation factors constraint contrast factor constraints. partial ordination, conditioning variables analysed together constraints. Variance inflation diagnostic tool identify useless constraints. common rule values 10 indicate redundant constraints. later constraints complete linear combinations conditions previous constraints, completely removed estimation, biplot scores centroids calculated aliased constraints. note printed default output aliased constraints. Function alias give linear coefficients defining aliased constraints, names argument names.= TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"functions return matrices vectors appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London. Gross, J. (2003). Variance inflation factors. R News 3(1), 13--15. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Jari Oksanen. vif.cca relies heavily code W. N. Venables. alias.cca simplified version alias.lm.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) goodness(mod, addprevious = TRUE) #> CCA1 CCA2 CCA3 CCA4 #> Achimill 0.36630013 0.3822685 0.3838616 0.4934158 #> Agrostol 0.67247051 0.6724758 0.6779597 0.7773267 #> Airaprae 0.36213737 0.3698100 0.3816619 0.3908018 #> Alopgeni 0.61547145 0.6966105 0.7042650 0.7212918 #> Anthodor 0.24619147 0.2795001 0.3509172 0.3609709 #> Bellpere 0.41185412 0.4179432 0.4847618 0.4849622 #> Bromhord 0.33487622 0.3397416 0.3870032 0.5505037 #> Chenalbu 0.23594716 0.2684323 0.2828928 0.2885321 #> Cirsarve 0.29041563 0.3013655 0.3080671 0.3591280 #> Comapalu 0.16338257 0.6836790 0.7390659 0.7963425 #> Eleopalu 0.55132024 0.6099415 0.6193301 0.6259818 #> Elymrepe 0.25239595 0.2710266 0.2761491 0.2882666 #> Empenigr 0.27089495 0.3132399 0.3153052 0.3154203 #> Hyporadi 0.31349648 0.3371809 0.3387669 0.3388716 #> Juncarti 0.43923609 0.4492937 0.4871043 0.5224072 #> Juncbufo 0.70439967 0.7226263 0.7228786 0.7257471 #> Lolipere 0.48141171 0.5720410 0.5727299 0.6034007 #> Planlanc 0.54969676 0.6084389 0.6802195 0.6826265 #> Poaprat 0.40267189 0.4944813 0.5014516 0.5326546 #> Poatriv 0.49694972 0.5409439 0.5468830 0.5594817 #> Ranuflam 0.68677962 0.6983001 0.7020461 0.7064850 #> Rumeacet 0.44788204 0.5211145 0.7673956 0.7691199 #> Sagiproc 0.27039747 0.3497634 0.3553109 0.3613746 #> Salirepe 0.64788354 0.7264891 0.7276110 0.7639711 #> Scorautu 0.54312496 0.5510319 0.6078931 0.6140593 #> Trifprat 0.37328840 0.4101104 0.6624199 0.6625703 #> Trifrepe 0.03048149 0.2115857 0.3300132 0.4207437 #> Vicilath 0.17824132 0.1784611 0.3762406 0.4279428 #> Bracruta 0.15585567 0.1641095 0.1672797 0.2449864 #> Callcusp 0.30771429 0.3143582 0.3308502 0.3518027 goodness(mod, addprevious = TRUE, summ = TRUE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 0.4934158 0.7773267 0.3908018 0.7212918 0.3609709 0.4849622 0.5505037 0.2885321 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 0.3591280 0.7963425 0.6259818 0.2882666 0.3154203 0.3388716 0.5224072 0.7257471 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 0.6034007 0.6826265 0.5326546 0.5594817 0.7064850 0.7691199 0.3613746 0.7639711 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 0.6140593 0.6625703 0.4207437 0.4279428 0.2449864 0.3518027 # Inertia components inertcomp(mod, prop = TRUE) #> pCCA CCA CA #> Achimill 0.34271900 0.15069678 0.5065842 #> Agrostol 0.55602406 0.22130269 0.2226733 #> Airaprae 0.06404726 0.32675457 0.6091982 #> Alopgeni 0.34238968 0.37890210 0.2787082 #> Anthodor 0.10259139 0.25837947 0.6390291 #> Bellpere 0.40972447 0.07523776 0.5150378 #> Bromhord 0.33046684 0.22003683 0.4494963 #> Chenalbu 0.11064346 0.17788865 0.7114679 #> Cirsarve 0.26649913 0.09262886 0.6408720 #> Comapalu 0.16096277 0.63537969 0.2036575 #> Eleopalu 0.53954819 0.08643366 0.3740182 #> Elymrepe 0.22234322 0.06592337 0.7117334 #> Empenigr 0.10361994 0.21180040 0.6845797 #> Hyporadi 0.03889627 0.29997533 0.6611284 #> Juncarti 0.43439190 0.08801527 0.4775928 #> Juncbufo 0.66622672 0.05952038 0.2742529 #> Lolipere 0.46273045 0.14067027 0.3965993 #> Planlanc 0.51993753 0.16268893 0.3173735 #> Poaprat 0.39408053 0.13857406 0.4673454 #> Poatriv 0.05598349 0.50349824 0.4405183 #> Ranuflam 0.68509904 0.02138594 0.2935150 #> Rumeacet 0.40125987 0.36786003 0.2308801 #> Sagiproc 0.26050435 0.10087025 0.6386254 #> Salirepe 0.12527838 0.63869277 0.2360289 #> Scorautu 0.10895437 0.50510492 0.3859407 #> Trifprat 0.34544815 0.31712212 0.3374297 #> Trifrepe 0.02132183 0.39942191 0.5792563 #> Vicilath 0.12125433 0.30668844 0.5720572 #> Bracruta 0.07222706 0.17275938 0.7550136 #> Callcusp 0.29447422 0.05732850 0.6481973 inertcomp(mod) #> pCCA CCA CA #> Achimill 0.0173766015 0.007640656 0.02568493 #> Agrostol 0.0456558521 0.018171449 0.01828399 #> Airaprae 0.0066672285 0.034014687 0.06341666 #> Alopgeni 0.0325977567 0.036073980 0.02653486 #> Anthodor 0.0096274015 0.024246897 0.05996790 #> Bellpere 0.0154640710 0.002839669 0.01943887 #> Bromhord 0.0180126793 0.011993496 0.02450059 #> Chenalbu 0.0031913088 0.005130874 0.02052099 #> Cirsarve 0.0110663060 0.003846389 0.02661204 #> Comapalu 0.0127652351 0.050389111 0.01615116 #> Eleopalu 0.0797827194 0.012780901 0.05530588 #> Elymrepe 0.0193932154 0.005749967 0.06207879 #> Empenigr 0.0063826176 0.013046147 0.04216766 #> Hyporadi 0.0046669914 0.035992710 0.07932587 #> Juncarti 0.0359126341 0.007276518 0.03948420 #> Juncbufo 0.0494087668 0.004414156 0.02033917 #> Lolipere 0.0368344271 0.011197683 0.03157023 #> Planlanc 0.0366139947 0.011456552 0.02234944 #> Poaprat 0.0142991623 0.005028142 0.01695757 #> Poatriv 0.0028845344 0.025942611 0.02269759 #> Ranuflam 0.0446783229 0.001394671 0.01914141 #> Rumeacet 0.0288221948 0.026423110 0.01658394 #> Sagiproc 0.0151161507 0.005853146 0.03705718 #> Salirepe 0.0142756439 0.072779924 0.02689581 #> Scorautu 0.0030643984 0.014206339 0.01085478 #> Trifprat 0.0228613139 0.020986733 0.02233067 #> Trifrepe 0.0008339368 0.015622139 0.02265580 #> Vicilath 0.0049088357 0.012415912 0.02315905 #> Bracruta 0.0032317812 0.007730074 0.03378289 #> Callcusp 0.0319130878 0.006212868 0.07024716 # vif.cca vif.cca(mod) #> Moisture.L Moisture.Q Moisture.C A1 ManagementHF ManagementNM #> 1.504327 1.284489 1.347660 1.367328 2.238653 2.570972 #> ManagementSF #> 2.424444 # Aliased constraints mod <- cca(dune ~ ., dune.env) mod #> Call: cca(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 1.5032 0.7106 12 #> Unconstrained 0.6121 0.2894 7 #> Inertia is scaled Chi-square #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4671 0.3410 0.1761 0.1532 0.0953 0.0703 0.0589 0.0499 0.0318 0.0260 0.0228 #> CCA12 #> 0.0108 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 #> 0.27237 0.10876 0.08975 0.06305 0.03489 0.02529 0.01798 #> vif.cca(mod) #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> 2.208249 2.858927 3.072715 3.587087 6.608315 142.359372 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> 12.862713 2.642718 3.007238 80.828330 49.294455 21.433337 #> Manure^4 #> NA alias(mod) #> Model : #> dune ~ A1 + Moisture + Management + Use + Manure #> #> Complete : #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> Manure^4 8.366600 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> Manure^4 5.291503 -4.472136 2.645751 #> with(dune.env, table(Management, Manure)) #> Manure #> Management 0 1 2 3 4 #> BF 0 2 1 0 0 #> HF 0 1 2 2 0 #> NM 6 0 0 0 0 #> SF 0 0 1 2 3 # The standard correlations (not recommended) ## IGNORE_RDIFF_BEGIN spenvcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 #> 0.9636709 0.9487249 0.9330741 0.8734876 0.9373716 0.8362687 0.9748793 0.8392720 #> CCA9 CCA10 CCA11 CCA12 #> 0.8748741 0.6087512 0.6633248 0.7581210 intersetcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 #> A1 -0.5332506 0.13691202 -0.47996401 -0.259859587 -0.09894964 #> Moisture.L -0.8785505 0.17867589 0.03714134 0.181952935 -0.09826534 #> Moisture.Q -0.1956664 -0.33044917 -0.27321286 -0.180333890 0.26609291 #> Moisture.C -0.2023782 -0.09698397 0.28596824 -0.261712720 -0.49103002 #> ManagementHF 0.3473460 0.01680324 -0.51205769 0.194144965 0.30752664 #> ManagementNM -0.5699549 -0.61111645 0.14751127 -0.013777789 0.04571982 #> ManagementSF -0.1197499 0.64084416 0.19780650 0.134892908 -0.09679992 #> Use.L -0.1871999 0.32990444 -0.30941161 -0.372747011 0.09586963 #> Use.Q -0.1820298 -0.48874152 -0.01997442 -0.009812946 0.04812588 #> Manure.L 0.3175126 0.65945634 0.03724864 -0.025383543 -0.04077470 #> Manure.Q -0.4075615 -0.21149073 0.49297244 -0.176686201 0.11973190 #> Manure.C 0.4676279 0.11376054 0.29132473 -0.173382982 0.14219924 #> Manure^4 0.2222349 -0.12789494 -0.12921227 0.108367170 -0.02559567 #> CCA6 CCA7 CCA8 CCA9 CCA10 #> A1 -0.15225816 0.25788462 0.19247720 -0.27694466 -0.1158449480 #> Moisture.L -0.02923342 0.07858647 -0.10772510 0.07101300 0.0952517164 #> Moisture.Q -0.11211675 0.05062810 -0.48302647 0.06138704 -0.2053304965 #> Moisture.C -0.23581275 -0.38693407 -0.10144580 -0.21907160 0.1875632770 #> ManagementHF -0.24278705 0.16364055 -0.14053438 0.31066725 0.1310215145 #> ManagementNM -0.06430101 0.23917584 0.14375754 -0.27103732 0.0002768613 #> ManagementSF -0.01611984 -0.49726250 0.08073472 -0.30235728 -0.1381281272 #> Use.L 0.19127262 -0.44624831 -0.18450714 0.12950951 0.0452826749 #> Use.Q 0.13485545 0.10367354 -0.11020112 0.41245485 -0.0766932005 #> Manure.L -0.22265819 -0.49627772 -0.16971786 -0.03943343 -0.0045229147 #> Manure.Q -0.19402211 -0.11937394 0.17611673 -0.44002593 0.0903998202 #> Manure.C 0.14760330 0.07842345 0.37774417 0.10181374 0.1055057288 #> Manure^4 -0.36683782 0.05953330 0.40927409 -0.06054381 -0.1500198368 #> CCA11 CCA12 #> A1 -0.03550223 -0.08881387 #> Moisture.L 0.06404776 -0.08587882 #> Moisture.Q -0.21810558 0.16917878 #> Moisture.C 0.13701079 -0.14260914 #> ManagementHF 0.17283125 0.13296499 #> ManagementNM -0.01358436 0.09533598 #> ManagementSF -0.01468592 -0.06614834 #> Use.L -0.08584883 0.32559307 #> Use.Q 0.41893616 0.04881247 #> Manure.L 0.02396993 0.13049087 #> Manure.Q 0.12987366 0.07137031 #> Manure.C 0.05176927 -0.41550238 #> Manure^4 -0.41603287 0.01661279 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS find goodness fit measure points nonmetric multidimensional scaling, function stressplot makes Shepard diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"# S3 method for metaMDS goodness(object, dis, ...) # S3 method for default stressplot(object, dis, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"object result object metaMDS, monoMDS isoMDS. dis Dissimilarities. used metaMDS monoMDS, must used isoMDS. pch Plotting character points. Default dependent number points. p.col, l.col Point line colours. lwd Line width. monoMDS default lwd = 1 two lines drawn, lwd = 2 otherwise. ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS finds goodness fit statistic observations (points). defined sum squared values equal squared stress. Large values indicate poor fit. absolute values goodness statistic depend definition stress: isoMDS expresses stress percents, therefore goodness values 100 times higher monoMDS expresses stress proportion. Function stressplot draws Shepard diagram plot ordination distances monotone linear fit line original dissimilarities. addition, displays two correlation-like statistics goodness fit graph. nonmetric fit based stress \\(S\\) defined \\(R^2 = 1-S^2\\). “linear fit” squared correlation fitted values ordination distances. monoMDS, “linear fit” \\(R^2\\) “stress type 2” equal. functions can used metaMDS, monoMDS isoMDS. original dissimilarities given monoMDS metaMDS results (latter tries reconstruct dissimilarities using metaMDSredist isoMDS used engine). isoMDS dissimilarities must given. either case, functions inspect dissimilarities consistent current ordination, refuse analyse inconsistent dissimilarities. Function goodness.metaMDS generic vegan, must spell name completely isoMDS class.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness returns vector values. Function stressplot returns invisibly object items original dissimilarities, ordination distances fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"data(varespec) mod <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2410643 #> Run 2 stress 0.1948413 #> Run 3 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.0416189 max resid 0.1517616 #> Run 4 stress 0.2383056 #> Run 5 stress 0.2096851 #> Run 6 stress 0.2329653 #> Run 7 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 2.732084e-05 max resid 8.096006e-05 #> ... Similar to previous best #> Run 8 stress 0.1825658 #> ... Procrustes: rmse 1.167373e-05 max resid 3.019892e-05 #> ... Similar to previous best #> Run 9 stress 0.2092456 #> Run 10 stress 0.1976154 #> Run 11 stress 0.2331577 #> Run 12 stress 0.2330486 #> Run 13 stress 0.1948413 #> Run 14 stress 0.2356726 #> Run 15 stress 0.1955838 #> Run 16 stress 0.2339637 #> Run 17 stress 0.2088293 #> Run 18 stress 0.2138946 #> Run 19 stress 0.2136761 #> Run 20 stress 0.2352729 #> *** Best solution repeated 2 times stressplot(mod) gof <- goodness(mod) gof #> [1] 0.02984516 0.03513715 0.04189194 0.04598243 0.04003134 0.03441442 #> [7] 0.03294898 0.03050145 0.03060754 0.02994058 0.03526332 0.02621426 #> [13] 0.03830984 0.02980913 0.03369611 0.02225897 0.03561569 0.03505254 #> [19] 0.06577496 0.03268352 0.03503121 0.02956628 0.05168013 0.04601957 plot(mod, display = \"sites\", type = \"n\") points(mod, display = \"sites\", cex = 2*gof/mean(gof))"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":null,"dir":"Reference","previous_headings":"","what":"Indicator Power of Species — indpower","title":"Indicator Power of Species — indpower","text":"Indicator power calculation Halme et al. (2009) congruence indicator target species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indicator Power of Species — indpower","text":"","code":"indpower(x, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indicator Power of Species — indpower","text":"x Community data frame matrix. type type statistic returned. See Details explanation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indicator Power of Species — indpower","text":"Halme et al. (2009) described index indicator power defined \\(IP_I = \\sqrt{\\times b}\\), \\(= S / O_I\\) \\(b = 1 - (O_T - S) / (N - O_I)\\). \\(N\\) number sites, \\(S\\) number shared occurrences indicator (\\(\\)) target (\\(T\\)) species. \\(O_I\\) \\(O_T\\) number occurrences indicator target species. type argument function call enables choose statistic return. type = 0 returns \\(IP_I\\), type = 1 returns \\(\\), type = 2 returns \\(b\\). Total indicator power (TIP) indicator species column mean (without value, see examples). Halme et al. (2009) explain calculate confidence intervals statistics, see Examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indicator Power of Species — indpower","text":"matrix indicator species rows target species columns (indicated first letters row/column names).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indicator Power of Species — indpower","text":"Halme, P., Mönkkönen, M., Kotiaho, J. S, Ylisirniö, -L. 2009. Quantifying indicator power indicator species. Conservation Biology 23: 1008--1016.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indicator Power of Species — indpower","text":"Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indicator Power of Species — indpower","text":"","code":"data(dune) ## IP values ip <- indpower(dune) ## and TIP values diag(ip) <- NA (TIP <- rowMeans(ip, na.rm=TRUE)) #> i.Achimill i.Agrostol i.Airaprae i.Alopgeni i.Anthodor i.Bellpere i.Bromhord #> 0.3186250 0.3342800 0.2168133 0.3416198 0.3567884 0.3432281 0.3665632 #> i.Chenalbu i.Cirsarve i.Comapalu i.Eleopalu i.Elymrepe i.Empenigr i.Hyporadi #> 0.2095044 0.2781640 0.1713273 0.2414787 0.3263516 0.2016196 0.2378197 #> i.Juncarti i.Juncbufo i.Lolipere i.Planlanc i.Poaprat i.Poatriv i.Ranuflam #> 0.2915850 0.3331330 0.3998442 0.3426064 0.4094319 0.3929520 0.2663080 #> i.Rumeacet i.Sagiproc i.Salirepe i.Scorautu i.Trifprat i.Trifrepe i.Vicilath #> 0.3484684 0.3788905 0.2898512 0.4362493 0.3145854 0.4503764 0.2605349 #> i.Bracruta i.Callcusp #> 0.4252676 0.2070766 ## p value calculation for a species ## from Halme et al. 2009 ## i is ID for the species i <- 1 fun <- function(x, i) indpower(x)[i,-i] ## 'c0' randomizes species occurrences os <- oecosimu(dune, fun, \"c0\", i=i, nsimul=99) #> Warning: nullmodel transformed 'comm' to binary data ## get z values from oecosimu output z <- os$oecosimu$z ## p-value (p <- sum(z) / sqrt(length(z))) #> [1] -1.471628 ## 'heterogeneity' measure (chi2 <- sum((z - mean(z))^2)) #> [1] 86.33867 pchisq(chi2, df=length(z)-1) #> [1] 0.9999999 ## Halme et al.'s suggested output out <- c(TIP=TIP[i], significance=p, heterogeneity=chi2, minIP=min(fun(dune, i=i)), varIP=sd(fun(dune, i=i)^2)) out #> TIP.i.Achimill significance heterogeneity minIP varIP #> 0.3186250 -1.4716280 86.3386705 0.0000000 0.2142097"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Linear Model Diagnostics for Constrained Ordination — influence.cca","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"set function extracts influence statistics linear model statistics directly constrained ordination result object cca, rda, capscale dbrda. constraints linear model functions support functions return identical results corresponding linear models (lm), can use documentation. main functions normal usage leverage values (hatvalues), standardized residuals (rstandard), studentized leave-one-residuals (rstudent), Cook's distance (cooks.distance). addition, vcov returns variance-covariance matrix coefficients, diagonal values variances coefficients. functions mainly support functions , can used directly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"# S3 method for cca hatvalues(model, ...) # S3 method for cca rstandard(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca rstudent(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca cooks.distance(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca sigma(object, type = c(\"response\", \"canoco\"), ...) # S3 method for cca vcov(object, type = \"canoco\", ...) # S3 method for cca SSD(object, type = \"canoco\", ...) # S3 method for cca qr(x, ...) # S3 method for cca df.residual(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"model, object, x constrained ordination result object. type Type statistics used extracting raw residuals residual standard deviation (sigma). Either \"response\" species data difference WA LC scores \"canoco\". ... arguments functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"vegan algorithm constrained ordination uses linear model (weighted linear model cca) find fitted values dependent community data, constrained ordination based fitted response (Legendre & Legendre 2012). hatvalues give leverage values constraints, leverage independent response data. influence statistics (rstandard, rstudent, cooks.distance) based leverage, raw residuals residual standard deviation (sigma). type = \"response\" raw residuals given unconstrained component constrained ordination, influence statistics matrix dimensions . observations times . species. cca statistics obtained lm model using Chi-square standardized species data (see decostand) dependent variable, row sums community data weights, rda lm model uses non-modified community data weights. algorithm CANOCO software constraints results iteration performing linear regression weighted averages (WA) scores constraints taking fitted values regression linear combination (LC) scores (ter Braak 1984). WA scores directly found species scores, LC scores linear combinations constraints regression. type = \"canoco\" raw residuals differences WA LC scores, residual standard deviation (sigma) taken axis sum squared WA scores minus one. quantities relationship residual component ordination, rather methodological artefacts algorithm used vegan. result matrix dimensions . observations times . constrained axes. Function vcov returns matrix variances covariances regression coefficients. diagonal values matrix variances, square roots give standard errors regression coefficients. function based SSD extracts sum squares crossproducts residuals. residuals defined similarly influence measures type similar properties limitations, define dimensions result matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Function .mlm casts ordination object multiple linear model class \"mlm\" (see lm), similar statistics can derived modified object set functions. However, problems R implementation analysis multiple linear model objects. results differ, current set functions probable correct. use .mlm objects avoided.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"data(varespec, varechem) mod <- cca(varespec ~ Al + P + K, varechem) ## leverage hatvalues(mod) #> 18 15 24 27 23 19 22 #> 0.06904416 0.06666628 0.15245083 0.18944882 0.09291510 0.05122338 0.15309307 #> 16 28 13 14 20 25 7 #> 0.09605909 0.27139695 0.75889765 0.04958141 0.06582891 0.10590183 0.20630888 #> 5 6 3 4 2 9 12 #> 0.19797654 0.16280522 0.22738889 0.30915530 0.15557066 0.14855598 0.09046701 #> 10 11 21 #> 0.12745850 0.10984996 0.14195559 plot(hatvalues(mod), type = \"h\") ## ordination plot with leverages: points with high leverage have ## similar LC and WA scores plot(mod, type = \"n\") ordispider(mod) # segment from LC to WA scores points(mod, dis=\"si\", cex=5*hatvalues(mod), pch=21, bg=2) # WA scores text(mod, dis=\"bp\", col=4) ## deviation and influence head(rstandard(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.4084518 0.9442480 -0.68178124 -0.798241724 0.9883838 -0.3086868 #> 15 -1.3902462 -1.5717947 -0.70784872 -0.645563228 0.2353736 -0.1679226 #> 24 0.9622453 -0.9520875 -0.08884556 -0.654099911 0.2420416 0.4832198 #> 27 -1.1080099 1.0938951 1.70146427 -0.196668562 -0.3937467 -0.7424140 #> 23 0.3979939 1.3218254 -0.63872221 -1.003315524 1.8996365 -0.4495408 #> 19 -1.5874575 0.7894087 -0.59609083 -0.006142973 0.1334143 -0.1060450 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 -0.5785258 -0.4585683 0.7640788 4.3748349 -0.39301720 -0.656213958 #> 15 -0.5416812 -0.4594716 -0.3410155 -0.2704388 -0.05769657 0.406022095 #> 24 -0.6409619 0.1908003 0.0198320 -0.2175720 3.75416938 -0.009140093 #> 27 4.2976822 -0.2704153 0.9211453 -0.2037075 -0.84410200 -0.712968237 #> 23 -0.8356637 -0.2779318 -0.0206861 -0.2899895 -0.67773316 -0.294674408 #> 19 -0.4453843 -0.3739569 -0.3174207 -0.2429740 -0.14281646 -0.819230368 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 -0.4933634 -0.6030042 -1.3812122 -0.07430854 -0.5332731 -0.7140513 #> 15 -0.3630434 -0.4055353 1.9930791 0.02126658 -0.2274864 -0.6730241 #> 24 2.2312025 -1.3907968 0.5792314 -0.45358405 -0.4628096 -0.2153846 #> 27 -0.5715008 1.6310289 0.8124329 -0.34628172 -0.8630615 0.9423113 #> 23 -0.4684368 -1.1995321 -0.8241477 -0.04953929 0.7890327 -0.6774675 #> 19 -0.3186606 -0.3948224 0.6180114 0.05035990 0.8260214 2.3626483 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 18 0.2918311 -0.42027512 -0.40191333 1.5303054 0.85056948 -0.2218833 #> 15 -0.6168627 -0.47096984 -0.42715787 0.1758993 -0.45409579 -0.3739920 #> 24 1.9274063 0.09984384 0.21066100 0.1668363 -0.08508535 -1.2835676 #> 27 -1.5729552 -0.31944598 -0.09167396 -0.1864165 0.51080623 -0.1083797 #> 23 0.4875575 0.61798970 -0.18848086 0.5574309 0.28079468 -0.4026464 #> 19 -0.1141521 -0.25344373 -0.31179839 -0.2391669 -0.65577452 0.6584127 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris #> 18 -0.39796095 0.93836573 -0.2564543 0.33864378 1.1572584 -0.2083305 #> 15 0.06761526 0.24340663 -0.1767166 0.27491203 1.0721632 1.9724410 #> 24 1.24902375 -0.98471253 -0.4801382 2.51184311 -1.4063518 -0.3084304 #> 27 -0.59021669 -1.25354423 -0.2460447 -1.09351514 -1.1681499 -1.0337232 #> 23 -0.34604539 -0.10730202 3.9477300 2.51924664 0.3536280 3.3882402 #> 19 -0.33866721 0.02698153 0.1776632 0.03968833 -0.7512944 -0.4763562 #> Cladchlo Cladbotr Cladamau Cladsp Cetreric Cetrisla #> 18 -0.5711604 -0.4914716 4.0852019 0.2489284 -0.4428064 -0.5834462 #> 15 -0.4347061 -0.6517740 -0.3155708 -0.2568784 0.2369559 -0.3468005 #> 24 0.6013607 0.4603779 -0.1576900 -0.5995616 2.7826114 0.3931826 #> 27 -0.5436659 -0.2788962 -0.1606997 0.1557879 -0.7298364 -0.5867612 #> 23 0.1029999 0.6494142 -0.3415411 -0.2330698 -0.6309632 -0.5301977 #> 19 0.2793712 -0.1315438 -0.2705486 -0.2865837 -0.4664078 -0.5056846 #> Flavniva Nepharct Stersp Peltapht Icmaeric Cladcerv #> 18 0.30428187 -0.3624631 -0.23665431 -0.1571633 -0.630333375 0.12058739 #> 15 0.17345018 -0.1919943 0.05469573 -0.3233311 -0.561177494 0.08518455 #> 24 -0.74154401 -0.4146848 -0.05500461 -0.7609417 0.255058737 -0.92761801 #> 27 0.11301489 -0.5064006 -0.08681568 -0.1247151 -0.001277338 -0.04038189 #> 23 0.09411988 -0.4627811 0.47668055 3.5826478 -0.274664798 -0.05517988 #> 19 0.07211309 -0.1693122 -0.17244475 -0.3155345 -0.461532920 -0.02709075 #> Claddefo Cladphyl #> 18 -0.43581630 -0.2098378 #> 15 0.94176661 -0.1028102 #> 24 -0.07508682 -1.0479632 #> 27 -1.06110299 -0.4908554 #> 23 2.66430575 -0.4244333 #> 19 -0.08427954 -0.1692474 head(cooks.distance(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.003093283 0.01653142 0.0086184263 1.181427e-02 0.0181129462 0.0017667454 #> 15 0.034513793 0.04411649 0.0089472619 7.441951e-03 0.0009892926 0.0005035324 #> 24 0.041636714 0.04076229 0.0003549575 1.923947e-02 0.0026344196 0.0105001237 #> 27 0.071736260 0.06992022 0.1691597848 2.260067e-03 0.0090591037 0.0322065174 #> 23 0.004056312 0.04474315 0.0104472601 2.577825e-02 0.0924100906 0.0051750754 #> 19 0.034013281 0.00841101 0.0047958896 5.093326e-07 0.0002402422 0.0001517834 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 0.006205594 0.003898934 1.082466e-02 0.3548634058 2.863921e-03 7.984152e-03 #> 15 0.005239584 0.003769873 2.076622e-03 0.0013060122 5.944416e-05 2.943802e-03 #> 24 0.018474359 0.001637053 1.768633e-05 0.0021286828 6.337714e-01 3.756697e-06 #> 27 1.079245024 0.004272814 4.958014e-02 0.0024247421 4.163335e-02 2.970242e-02 #> 23 0.017883042 0.001978130 1.095811e-05 0.0021534900 1.176240e-02 2.223634e-03 #> 19 0.002677405 0.001887502 1.359924e-03 0.0007968265 2.752966e-04 9.058502e-03 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 0.004513066 0.006741841 0.035371924 1.023801e-04 0.0052727512 0.009453588 #> 15 0.002353565 0.002936747 0.070934636 8.076156e-06 0.0009241023 0.008088546 #> 24 0.223863289 0.086982577 0.015087217 9.251676e-03 0.0096318471 0.002086096 #> 27 0.019084693 0.155444288 0.038567940 7.006653e-03 0.0435246182 0.051884812 #> 23 0.005619275 0.036846999 0.017393559 6.284604e-05 0.0159429407 0.011753178 #> 19 0.001370570 0.002104010 0.005155103 3.423056e-05 0.0092093061 0.075342921 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang #> 18 0.0015790689 0.0032749544 0.0029950405 0.0434204290 0.0134139742 #> 15 0.0067949512 0.0039609143 0.0032582625 0.0005525065 0.0036821731 #> 24 0.1670519112 0.0004482780 0.0019955986 0.0012516596 0.0003255476 #> 27 0.1445719869 0.0059627459 0.0004910701 0.0020305802 0.0152462776 #> 23 0.0060873754 0.0097800450 0.0009097308 0.0079572035 0.0020190921 #> 19 0.0001758783 0.0008669766 0.0013121758 0.0007720518 0.0058043470 #> Cladstel Cladunci Cladcocc Cladcorn Cladgrac Cladfimb #> 18 0.0009128242 2.936424e-03 1.632609e-02 0.0012194323 2.126298e-03 0.024831249 #> 15 0.0024976637 8.163919e-05 1.057972e-03 0.0005576530 1.349574e-03 0.020527262 #> 24 0.0740870642 7.015301e-02 4.360375e-02 0.0103666250 2.837200e-01 0.088939118 #> 27 0.0006863532 2.035516e-02 9.181862e-02 0.0035373629 6.987166e-02 0.079734929 #> 23 0.0041517010 3.066511e-03 2.948453e-04 0.3990922140 1.625248e-01 0.003202372 #> 19 0.0058511426 1.548070e-03 9.826017e-06 0.0004260291 2.126033e-05 0.007618415 #> Cladcris Cladchlo Cladbotr Cladamau Cladsp Cetreric #> 18 0.0008047178 0.0060485887 0.0044785217 0.3094317912 0.001148912 0.003635513 #> 15 0.0694732020 0.0033744318 0.0075858327 0.0017782910 0.001178323 0.001002639 #> 24 0.0042777886 0.0162620357 0.0095308986 0.0011181833 0.016164882 0.348184969 #> 27 0.0624395810 0.0172709295 0.0045450274 0.0015089732 0.001418140 0.031124511 #> 23 0.2939860714 0.0002716767 0.0107999564 0.0029872002 0.001391075 0.010194981 #> 19 0.0030627243 0.0010534342 0.0002335528 0.0009879496 0.001108529 0.002936134 #> Cetrisla Flavniva Nepharct Stersp Peltapht Icmaeric #> 18 0.006311601 1.716683e-03 0.0024359336 1.038405e-03 0.0004579733 7.366793e-03 #> 15 0.002147676 5.372282e-04 0.0006582422 5.342149e-05 0.0018668272 5.623539e-03 #> 24 0.006951741 2.472742e-02 0.0077328775 1.360514e-04 0.0260380002 2.925400e-03 #> 27 0.020117510 7.463161e-04 0.0149844191 4.404005e-04 0.0009088445 9.533741e-08 #> 23 0.007198701 2.268512e-04 0.0054844061 5.818798e-03 0.3286900860 1.931899e-03 #> 19 0.003451467 7.018956e-05 0.0003869197 4.013694e-04 0.0013438101 2.875078e-03 #> Cladcerv Claddefo Cladphyl #> 18 2.696135e-04 3.521639e-03 0.0008164040 #> 15 1.295779e-04 1.583784e-02 0.0001887477 #> 24 3.869397e-02 2.535317e-04 0.0493852158 #> 27 9.528504e-05 6.579101e-02 0.0140785723 #> 23 7.797221e-05 1.817802e-01 0.0046131462 #> 19 9.905729e-06 9.587131e-05 0.0003866237 ## Influence measures from lm y <- decostand(varespec, \"chi.square\") # needed in cca y1 <- with(y, Cladstel) # take one species for lm lmod1 <- lm(y1 ~ Al + P + K, varechem, weights = rowSums(varespec)) ## numerically identical within 2e-15 all(abs(cooks.distance(lmod1) - cooks.distance(mod)[, \"Cladstel\"]) < 1e-8) #> [1] TRUE ## t-values of regression coefficients based on type = \"canoco\" ## residuals coef(mod) #> CCA1 CCA2 CCA3 #> Al 0.007478556 -0.001883637 0.003380774 #> P -0.006491081 -0.102189737 -0.022306682 #> K -0.006755568 0.015343662 0.017067351 coef(mod)/sqrt(diag(vcov(mod, type = \"canoco\"))) #> CCA1 CCA2 CCA3 #> Al 6.5615451 -1.397643 3.313629 #> P -0.4576132 -6.092557 -1.756774 #> K -2.0862129 4.007159 5.887926"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":null,"dir":"Reference","previous_headings":"","what":"Isometric Feature Mapping Ordination — isomap","title":"Isometric Feature Mapping Ordination — isomap","text":"function performs isometric feature mapping consists three simple steps: (1) retain shortest dissimilarities among objects, (2) estimate dissimilarities shortest path distances, (3) perform metric scaling (Tenenbaum et al. 2000).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"isomap(dist, ndim=10, ...) isomapdist(dist, epsilon, k, path = \"shortest\", fragmentedOK =FALSE, ...) # S3 method for isomap summary(object, ...) # S3 method for isomap plot(x, net = TRUE, n.col = \"gray\", type = \"points\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Isometric Feature Mapping Ordination — isomap","text":"dist Dissimilarities. ndim Number axes metric scaling (argument k cmdscale). epsilon Shortest dissimilarity retained. k Number shortest dissimilarities retained point. epsilon k given, epsilon used. path Method used stepacross estimate shortest path, alternatives \"shortest\" \"extended\". fragmentedOK dissimilarity matrix fragmented. TRUE, analyse largest connected group, otherwise stop error. x, object isomap result object. net Draw net retained dissimilarities. n.col Colour drawn net segments. can also vector recycled points, colour net segment mixture joined points. type Plot observations either \"points\", \"text\" use \"none\" plot observations. \"text\" use ordilabel net = TRUE ordiplot net = FALSE, pass extra arguments functions. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Isometric Feature Mapping Ordination — isomap","text":"function isomap first calls function isomapdist dissimilarity transformation, performs metric scaling result. arguments isomap passed isomapdist. functions separate isompadist transformation easily used functions simple linear mapping cmdscale. Function isomapdist retains either dissimilarities equal shorter epsilon, epsilon given, least k shortest dissimilarities point. complete dissimilarity matrix reconstructed using stepacross using either flexible shortest paths extended dissimilarities (details, see stepacross). De'ath (1999) actually published essentially method Tenenbaum et al. (2000), De'ath's function available function xdiss non-CRAN package mvpart. differences isomap introduced k criterion, whereas De'ath used epsilon criterion. practice, De'ath also retains higher proportion dissimilarities typical isomap. plot function uses internally ordiplot, except adds text net using ordilabel. plot function passes extra arguments functions. addition, vegan3d package function rgl.isomap make dynamic 3D plots can rotated screen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Isometric Feature Mapping Ordination — isomap","text":"Function isomapdist returns dissimilarity object similar dist. Function isomap returns object class isomap plot summary methods. plot function returns invisibly object class ordiplot. Function scores can extract ordination scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Isometric Feature Mapping Ordination — isomap","text":"De'ath, G. (1999) Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecology 144, 191--199 Tenenbaum, J.B., de Silva, V. & Langford, J.C. (2000) global network framework nonlinear dimensionality reduction. Science 290, 2319--2323.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Isometric Feature Mapping Ordination — isomap","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Isometric Feature Mapping Ordination — isomap","text":"Tenenbaum et al. (2000) justify isomap tool unfolding manifold (e.g. 'Swiss Roll'). Even manifold structure, sampling must even dense dissimilarities along manifold shorter across folds. data manifold structure, results sensitive parameter values.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"## The following examples also overlay minimum spanning tree to ## the graphics in red. op <- par(mar=c(4,4,1,1)+0.2, mfrow=c(2,2)) data(BCI) dis <- vegdist(BCI) tr <- spantree(dis) pl <- ordiplot(cmdscale(dis), main=\"cmdscale\") #> species scores not available lines(tr, pl, col=\"red\") ord <- isomap(dis, k=3) ord #> #> Isometric Feature Mapping (isomap) #> #> Call: #> isomap(dist = dis, k = 3) #> #> Distance method: bray shortest isomap #> Criterion: k = 3 pl <- plot(ord, main=\"isomap k=3\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, k=5), main=\"isomap k=5\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, epsilon=0.45), main=\"isomap epsilon=0.45\") lines(tr, pl, col=\"red\") par(op) ## colour points and web by the dominant species dom <- apply(BCI, 1, which.max) ## need nine colours, but default palette has only eight op <- palette(c(palette(\"default\"), \"sienna\")) plot(ord, pch = 16, col = dom, n.col = dom) palette(op)"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":null,"dir":"Reference","previous_headings":"","what":"Kendall coefficient of concordance — kendall.global","title":"Kendall coefficient of concordance — kendall.global","text":"Function kendall.global computes tests coefficient concordance among several judges (variables, species) permutation test. Function kendall.post carries posteriori tests contributions individual judges (variables, species) overall concordance group permutation tests. several groups judges identified data table, coefficients concordance (kendall.global) posteriori tests (kendall.post) computed group separately. Use ecology: identify significant species associations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"kendall.global(Y, group, nperm = 999, mult = \"holm\") kendall.post(Y, group, nperm = 999, mult = \"holm\")"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kendall coefficient of concordance — kendall.global","text":"Y Data file (data frame matrix) containing quantitative semiquantitative data. Rows objects columns judges (variables). community ecology, table often site--species table. group vector defining judges divided groups. See example . groups explicitly defined, judges data file considered forming single group. nperm Number permutations performed. Default 999. mult Correct P-values multiple testing using alternatives described p.adjust addition \"sidak\" (see Details). Bonferroni correction overly conservative; recommended. included allow comparisons methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Kendall coefficient of concordance — kendall.global","text":"Y must contain quantitative data. transformed ranks within column computation coefficient concordance. search species associations described Legendre (2005) proceeds 3 steps: (1) Correlation analysis species. possible method compute Ward's agglomerative clustering matrix correlations among species. detail: (1.1) compute Pearson Spearman correlation matrix (correl.matrix) among species; (1.2) turn distance matrix: mat.D = .dist(1-correl.matrix); (1.3) carry Ward's hierarchical clustering matrix using hclust: clust.ward = hclust(mat.D, \"ward\"); (1.4) plot dendrogram: plot(clust.ward, hang=-1); (1.5) cut dendrogram two groups, retrieve vector species membership: group.2 = cutree(clust.ward, k=2). (1.6) steps 2 3 , may come back try divisions species k = \\(3, 4, 5, \\dots\\) groups. (2) Compute global tests significance 2 () groups using function kendall.global vector defining groups. Groups globally significant must refined abandoned. (3) Compute posteriori tests contribution individual species concordance group using function kendall.post vector defining groups. species negative values \"Spearman.mean\", means species clearly belong group, hence group inclusive. Go back (1.5) cut dendrogram finely. left right groups can cut separately, independently levels along dendrogram; write vector group membership cutree produce desired groups. corrections used multiple testing applied list P-values (P); take account number tests (k) carried simultaneously (number groups kendall.global, number species kendall.post). corrections performed using function p.adjust; see function description correction methods. addition, Šidák correction defined \\(P_{corr} = 1 -(1 - P)^k\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Kendall coefficient of concordance — kendall.global","text":"table containing following information rows. columns correspond groups \"judges\" defined vector \"group\". function Kendall.post used, many tables number predefined groups. W Kendall's coefficient concordance, W. F F statistic. F = W*(m-1)/(1-W) m number judges. Prob.F Probability associated F statistic, computed F distribution nu1 = n-1-(2/m) nu2 = nu1*(m-1); n number objects. Corrected prob.F Probabilities associated F, corrected using method selected parameter mult. Shown one group. Chi2 Friedman's chi-square statistic (Friedman 1937) used permutation test W. Prob.perm Permutational probabilities, uncorrected. Corrected prob.perm Permutational probabilities corrected using method selected parameter mult. Shown one group. Spearman.mean Mean Spearman correlations judge test judges group. W.per.species Contribution judge test overall concordance statistic group.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Kendall coefficient of concordance — kendall.global","text":"Friedman, M. 1937. use ranks avoid assumption normality implicit analysis variance. Journal American Statistical Association 32: 675-701. Kendall, M. G. B. Babington Smith. 1939. problem m rankings. Annals Mathematical Statistics 10: 275-287. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245. Legendre, P. 2009. Coefficient concordance. : Encyclopedia Research Design. SAGE Publications (press). Siegel, S. N. J. Castellan, Jr. 1988. Nonparametric statistics behavioral sciences. 2nd edition. McGraw-Hill, New York.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Kendall coefficient of concordance — kendall.global","text":"F. Guillaume Blanchet, University Alberta, Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"data(mite) mite.hel <- decostand(mite, \"hel\") # Reproduce the results shown in Table 2 of Legendre (2005), a single group mite.small <- mite.hel[c(4,9,14,22,31,34,45,53,61,69),c(13:15,23)] kendall.global(mite.small, nperm=49) #> $Concordance_analysis #> Group.1 #> W 0.44160305 #> F 2.37252221 #> Prob.F 0.04403791 #> Chi2 15.89770992 #> Prob.perm 0.12000000 #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.small, mult=\"holm\", nperm=49) #> $A_posteriori_tests #> TVEL ONOV SUCT Trhypch1 #> Spearman.mean 0.3265678 0.3965503 0.4570402 -0.1681251 #> W.per.species 0.4949258 0.5474127 0.5927802 0.1239061 #> Prob 0.0800000 0.0200000 0.0200000 0.7200000 #> Corrected prob 0.1600000 0.0800000 0.0800000 0.7200000 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # Reproduce the results shown in Tables 3 and 4 of Legendre (2005), 2 groups group <-c(1,1,2,1,1,1,1,1,2,1,1,1,1,1,1,2,1,2,1,1,1,1,2,1,2,1,1,1,1,1,2,2,2,2,2) kendall.global(mite.hel, group=group, nperm=49) #> $Concordance_analysis #> Group.1 Group.2 #> W 3.097870e-01 2.911888e-01 #> F 1.032305e+01 4.108130e+00 #> Prob.F 1.177138e-85 4.676566e-22 #> Corrected prob.F 2.354275e-85 4.676566e-22 #> Chi2 5.130073e+02 2.210123e+02 #> Prob.perm 2.000000e-02 2.000000e-02 #> Corrected prob.perm 4.000000e-02 4.000000e-02 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.hel, group=group, mult=\"holm\", nperm=49) #> $A_posteriori_tests_Group #> $A_posteriori_tests_Group[[1]] #> Brachy PHTH RARD SSTR Protopl MEGR #> Spearman.mean 0.1851177 0.4258111 0.359058 0.2505486 0.1802160 0.2833298 #> W.per.species 0.2190711 0.4497357 0.385764 0.2817757 0.2143736 0.3131911 #> Prob 0.0200000 0.0200000 0.020000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 #> MPRO HMIN HMIN2 NPRA TVEL ONOV #> Spearman.mean 0.09248024 0.2444656 0.4138494 0.1263751 0.4177343 0.3301159 #> W.per.species 0.13029357 0.2759462 0.4382723 0.1627761 0.4419954 0.3580278 #> Prob 0.14000000 0.0200000 0.0200000 0.0400000 0.0200000 0.0200000 #> Corrected prob 0.70000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> SUCT Oribatl1 PWIL Galumna1 Stgncrs2 HRUF #> Spearman.mean 0.2185421 0.421216 0.2574779 0.4180699 0.3623428 0.1250230 #> W.per.species 0.2511028 0.445332 0.2884163 0.4423170 0.3889118 0.1614804 #> Prob 0.0200000 0.020000 0.0200000 0.0200000 0.0200000 0.0600000 #> Corrected prob 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 0.7000000 #> PPEL SLAT FSET Lepidzts Eupelops Miniglmn #> Spearman.mean 0.2188216 0.3016159 0.4217606 0.2577037 0.1108022 0.2301430 #> W.per.species 0.2513707 0.3307153 0.4458539 0.2886327 0.1478521 0.2622203 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0600000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> $A_posteriori_tests_Group[[2]] #> HPAV TVIE LCIL Ceratoz1 Trhypch1 NCOR #> Spearman.mean 0.1222579 0.2712078 0.1906408 0.1375601 0.1342409 0.3342345 #> W.per.species 0.2020527 0.3374616 0.2642189 0.2159637 0.2129463 0.3947586 #> Prob 0.0400000 0.0200000 0.0200000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> LRUG PLAG2 Ceratoz3 Oppiminu Trimalc2 #> Spearman.mean 0.3446561 0.1833099 0.3188922 0.1764232 0.2498877 #> W.per.species 0.4042328 0.2575544 0.3808111 0.2512938 0.3180797 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Function linestack plots vertical one-dimensional plots numeric vectors. plots always labelled, labels moved vertically avoid overwriting.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"linestack(x, labels, cex = 0.8, side = \"right\", hoff = 2, air = 1.1, at = 0, add = FALSE, axis = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"x Numeric vector plotted. labels Labels used instead default (names x). May expressions drawn plotmath. cex Size labels. side Put labels \"right\" \"left\" axis. hoff Distance vertical axis label units width letter “m”. air Multiplier string height leave empty space labels. Position plot horizontal axis. add Add existing plot. axis Add axis plot. ... graphical parameters labels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function returns invisibly shifted positions labels user coordinates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Jari Oksanen modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function always draws labelled diagrams. want unlabelled diagrams, can use, e.g., plot, stripchart rug.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"## First DCA axis data(dune) ord <- decorana(dune) linestack(scores(ord, choices=1, display=\"sp\")) linestack(scores(ord, choices=1, display=\"si\"), side=\"left\", add=TRUE) title(main=\"DCA axis 1\") ## Expressions as labels N <- 10 # Number of sites df <- data.frame(Ca = rlnorm(N, 2), NO3 = rlnorm(N, 4), SO4 = rlnorm(N, 10), K = rlnorm(N, 3)) ord <- rda(df, scale = TRUE) ### vector of expressions for labels labs <- expression(Ca^{2+phantom()}, NO[3]^{-phantom()}, SO[4]^{2-phantom()}, K^{+phantom()}) scl <- \"sites\" linestack(scores(ord, choices = 1, display = \"species\", scaling = scl), labels = labs, air = 2) linestack(scores(ord, choices = 1, display = \"site\", scaling = scl), side = \"left\", add = TRUE) title(main = \"PCA axis 1\")"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":null,"dir":"Reference","previous_headings":"","what":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"standard CEP name four first letters generic name four first letters specific epithet Latin name. last epithet, may subspecific name, used current function. name one component, abbreviated eight characters (see abbreviate). returned names made unique function make.unique adds numbers end CEP names needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(names, seconditem = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"names names formatted CEP names. seconditem Take always second item original name abbreviated name instead last original item (default).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Cornell Ecology Programs (CEP) used eight-letter abbreviations species site names. species, names formed taking four first letters generic name four first letters specific subspecific epithet. current function first makes valid R names using make.names, splits elements. CEP name made taking four first letters first element, four first letters last (default) second element ( seconditem = TRUE). one name element, abbreviated eight letters. Finally, names made unique may add numbers duplicated names. CEP names originally used, old FORTRAN IV CHARACTER data type, text stored numerical variables, popular computers hold four characters. modern times, reason limitation, ecologists used names, may practical avoid congestion ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Function returns CEP names.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"function simpleminded rigid. must write better one need.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(c(\"Aa maderoi\", \"Poa sp.\", \"Cladina rangiferina\", \"Cladonia cornuta\", \"Cladonia cornuta var. groenlandica\", \"Cladonia rangiformis\", \"Bryoerythrophyllum\")) #> [1] \"Aamade\" \"Poasp\" \"Cladrang\" \"Cladcorn\" \"Cladgroe\" #> [6] \"Cladrang.1\" \"Bryrythr\" data(BCI) colnames(BCI) <- make.cepnames(colnames(BCI))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel Correlogram — mantel.correlog","title":"Mantel Correlogram — mantel.correlog","text":"Function mantel.correlog computes multivariate Mantel correlogram. Proposed Sokal (1986) Oden Sokal (1986), method also described Legendre Legendre (2012, pp. 819--821).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel Correlogram — mantel.correlog","text":"","code":"mantel.correlog(D.eco, D.geo=NULL, XY=NULL, n.class=0, break.pts=NULL, cutoff=TRUE, r.type=\"pearson\", nperm=999, mult=\"holm\", progressive=TRUE) # S3 method for mantel.correlog plot(x, alpha=0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel Correlogram — mantel.correlog","text":"D.eco ecological distance matrix, class either dist matrix. D.geo geographic distance matrix, class either dist matrix. Provide either D.geo XY. Default: D.geo=NULL. XY file Cartesian geographic coordinates points. Default: XY=NULL. n.class Number classes. n.class=0, Sturges equation used unless break points provided. break.pts Vector containing break points distance distribution. Provide (n.class+1) breakpoints, , list beginning ending point. Default: break.pts=NULL. cutoff second half distance classes, cutoff = TRUE limits correlogram distance classes include points. cutoff = FALSE, correlogram includes distance classes. r.type Type correlation calculation Mantel statistic. Default: r.type=\"pearson\". choices r.type=\"spearman\" r.type=\"kendall\", functions cor mantel. nperm Number permutations tests significance. Default: nperm=999. large data files, permutation tests rather slow. mult Correct P-values multiple testing. correction methods \"holm\" (default), \"hochberg\", \"sidak\", methods available p.adjust function: \"bonferroni\" (best known, recommended overly conservative), \"hommel\", \"BH\", \"\", \"fdr\", \"none\". progressive Default: progressive=TRUE progressive correction multiple-testing, described Legendre Legendre (1998, p. 721). Test first distance class: correction; second distance class: correct 2 simultaneous tests; distance class k: correct k simultaneous tests. progressive=FALSE: correct tests n.class simultaneous tests. x Output mantel.correlog. alpha Significance level points drawn black symbols correlogram. Default: alpha=0.05. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel Correlogram — mantel.correlog","text":"correlogram graph spatial correlation values plotted, ordinate, function geographic distance classes among study sites along abscissa. Mantel correlogram, Mantel correlation (Mantel 1967) computed multivariate (e.g. multi-species) distance matrix user's choice design matrix representing geographic distance classes turn. Mantel statistic tested permutational Mantel test performed vegan's mantel function. correction multiple testing applied, permutations necessary -correction case, obtain significant p-values higher correlogram classes. print.mantel.correlog function prints correlogram. See examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel Correlogram — mantel.correlog","text":"mantel.res table distance classes rows class indices, number distances per class, Mantel statistics (computed using Pearson's r, Spearman's r, Kendall's tau), p-values columns. positive Mantel statistic indicates positive spatial correlation. additional column p-values corrected multiple testing added unless mult=\"none\". n.class n umber distance classes. break.pts break points provided user computed program. mult name correction multiple testing. correction: mult=\"none\". progressive logical (TRUE, FALSE) value indicating whether progressive correction multiple testing requested. n.tests number distance classes Mantel tests computed tested significance. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel Correlogram — mantel.correlog","text":"Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel Correlogram — mantel.correlog","text":"Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Mantel, N. 1967. detection disease clustering generalized regression approach. Cancer Res. 27: 209-220. Oden, N. L. R. R. Sokal. 1986. Directional autocorrelation: extension spatial correlograms two dimensions. Syst. Zool. 35: 608-617. Sokal, R. R. 1986. Spatial data analysis historical processes. 29-43 : E. Diday et al. [eds.] Data analysis informatics, IV. North-Holland, Amsterdam. Sturges, H. . 1926. choice class interval. Journal American Statistical Association 21: 65–66.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel Correlogram — mantel.correlog","text":"","code":"# Mite data available in \"vegan\" data(mite) data(mite.xy) mite.hel <- decostand(mite, \"hellinger\") # Detrend the species data by regression on the site coordinates mite.hel.resid <- resid(lm(as.matrix(mite.hel) ~ ., data=mite.xy)) # Compute the detrended species distance matrix mite.hel.D <- dist(mite.hel.resid) # Compute Mantel correlogram with cutoff, Pearson statistic mite.correlog <- mantel.correlog(mite.hel.D, XY=mite.xy, nperm=49) summary(mite.correlog) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 4 -none- call mite.correlog #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.135713 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.118174 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.037820 0.04 0.06 . #> D.cl.4 2.699274 696.000000 -0.098605 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.112682 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.107603 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.022264 0.12 0.14 #> D.cl.8 5.612730 326.000000 NA NA NA #> D.cl.9 6.341094 260.000000 NA NA NA #> D.cl.10 7.069458 184.000000 NA NA NA #> D.cl.11 7.797822 130.000000 NA NA NA #> D.cl.12 8.526186 66.000000 NA NA NA #> D.cl.13 9.254550 32.000000 NA NA NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # or: print(mite.correlog) # or: print.mantel.correlog(mite.correlog) plot(mite.correlog) # Compute Mantel correlogram without cutoff, Spearman statistic mite.correlog2 <- mantel.correlog(mite.hel.D, XY=mite.xy, cutoff=FALSE, r.type=\"spearman\", nperm=49) summary(mite.correlog2) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 6 -none- call mite.correlog2 #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, cutoff = FALSE, r.type = \"spearman\", nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.134229 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.121270 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.035413 0.02 0.06 . #> D.cl.4 2.699274 696.000000 -0.095899 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.118692 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.117148 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.031123 0.08 0.14 #> D.cl.8 5.612730 326.000000 0.026064 0.14 0.16 #> D.cl.9 6.341094 260.000000 0.050573 0.08 0.24 #> D.cl.10 7.069458 184.000000 0.057017 0.04 0.20 #> D.cl.11 7.797822 130.000000 0.036195 0.10 0.32 #> D.cl.12 8.526186 66.000000 -0.054242 0.02 0.24 #> D.cl.13 9.254550 32.000000 -0.066677 0.04 0.26 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mite.correlog2) # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Function mantel finds Mantel statistic matrix correlation two dissimilarity matrices, function mantel.partial finds partial Mantel statistic partial matrix correlation three dissimilarity matrices. significance statistic evaluated permuting rows columns first dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"mantel(xdis, ydis, method=\"pearson\", permutations=999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\")) mantel.partial(xdis, ydis, zdis, method = \"pearson\", permutations = 999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"xdis, ydis, zdis Dissimilarity matrices ordist objects. first object xdis permuted permutation tests. method Correlation method, accepted cor: \"pearson\", \"spearman\" \"kendall\". permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. na.rm Remove missing values calculation Mantel correlation. Use option care: Permutation tests can biased, particular two matrices missing values matching positions. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Mantel statistic simply correlation entries two dissimilarity matrices (use cross products, linearly related). However, significance directly assessed, \\(N(N-1)/2\\) entries just \\(N\\) observations. Mantel developed asymptotic test, use permutations \\(N\\) rows columns dissimilarity matrix. first matrix (xdist) permuted, second kept constant. See permutations additional details permutation tests Vegan. Partial Mantel statistic uses partial correlation conditioned third matrix. first matrix permuted correlation structure second first matrices kept constant. Although mantel.partial silently accepts methods \"pearson\", partial correlations probably wrong methods. function uses cor, accept alternatives pearson product moment correlations spearman kendall rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"function returns list class mantel following components: Call Function call. method Correlation method used, returned cor.test. statistic Mantel statistic. signif Empirical significance level permutations. perm vector permuted values. distribution permuted values can inspected permustats function. permutations Number permutations. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"test due Mantel, course, current implementation based Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English Edition. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Legendre & Legendre (2012, Box 10.4) warn using partial Mantel correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"## Is vegetation related to environment? data(varespec) data(varechem) veg.dist <- vegdist(varespec) # Bray-Curtis env.dist <- vegdist(scale(varechem), \"euclid\") mantel(veg.dist, env.dist) #> #> Mantel statistic based on Pearson's product-moment correlation #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist) #> #> Mantel statistic r: 0.3047 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.116 0.147 0.176 0.210 #> Permutation: free #> Number of permutations: 999 #> mantel(veg.dist, env.dist, method=\"spear\") #> #> Mantel statistic based on Spearman's rank correlation rho #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist, method = \"spear\") #> #> Mantel statistic r: 0.2838 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.123 0.162 0.182 0.208 #> Permutation: free #> Number of permutations: 999 #>"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS performs Nonmetric Multidimensional Scaling (NMDS), tries find stable solution using several random starts. addition, standardizes scaling result, configurations easier interpret, adds species scores site ordination. metaMDS function provide actual NMDS, calls another function purpose. Currently monoMDS default choice, also possible call isoMDS (MASS package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"metaMDS(comm, distance = \"bray\", k = 2, try = 20, trymax = 20, engine = c(\"monoMDS\", \"isoMDS\"), autotransform =TRUE, noshare = (engine == \"isoMDS\"), wascores = TRUE, expand = TRUE, trace = 1, plot = FALSE, previous.best, ...) # S3 method for metaMDS plot(x, display = c(\"sites\", \"species\"), choices = c(1, 2), type = \"p\", shrink = FALSE, ...) # S3 method for metaMDS points(x, display = c(\"sites\", \"species\"), choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS text(x, display = c(\"sites\", \"species\"), labels, choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS scores(x, display = c(\"sites\", \"species\"), shrink = FALSE, choices, tidy = FALSE, ...) metaMDSdist(comm, distance = \"bray\", autotransform = TRUE, noshare = TRUE, trace = 1, commname, zerodist = \"ignore\", distfun = vegdist, ...) metaMDSiter(dist, k = 2, try = 20, trymax = 20, trace = 1, plot = FALSE, previous.best, engine = \"monoMDS\", maxit = 200, parallel = getOption(\"mc.cores\"), ...) initMDS(x, k=2) postMDS(X, dist, pc=TRUE, center=TRUE, halfchange, threshold=0.8, nthreshold=10, plot=FALSE, ...) metaMDSredist(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"comm Community data. Alternatively, dissimilarities either dist structure symmetric square matrix. latter case stages skipped except random starts centring pc rotation axes. distance Dissimilarity index used vegdist. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher global non-metric MDS, still higher local NMDS. try, trymax Minimum maximum numbers random starts search stable solution. try reached, iteration stop similar solutions repeated trymax reached. engine function used MDS. default use monoMDS function vegan, backward compatibility also possible use isoMDS MASS. autotransform Use simple heuristics possible data transformation typical community data (see ). community data, probably set autotransform = FALSE. noshare Triggering calculation step-across extended dissimilarities function stepacross. argument can logical numerical value greater zero less one. TRUE, extended dissimilarities used always shared species sites, FALSE, never used. noshare numerical value, stepacross used proportion site pairs shared species exceeds noshare. number pairs shared species found .shared function, noshare effect input data dissimilarities instead community data. wascores Calculate species scores using function wascores. expand Expand weighted averages species wascores. trace Trace function; trace = 2 higher voluminous. plot Graphical tracing: plot interim results. may want set par(ask = TRUE) option. previous.best Start searches previous solution. x metaMDS result (dissimilarity structure initMDS). choices Axes shown. type Plot type: \"p\" points, \"t\" text, \"n\" axes . display Display \"sites\" \"species\". shrink Shrink back species scores expanded originally. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable code (\"sites\" \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Optional test used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. X Configuration multidimensional scaling. commname name comm: given function called directly. zerodist Handling zero dissimilarities: either \"fail\" \"add\" small positive value, \"ignore\". monoMDS accepts zero dissimilarities default zerodist = \"ignore\", isoMDS may need set zerodist = \"add\". distfun Dissimilarity function. function returning dist object accepting argument method can used (extra arguments may cause name conflicts). maxit Maximum number iterations single NMDS run; passed engine function monoMDS isoMDS. parallel Number parallel processes predefined socket cluster. use pre-defined socket clusters (say, clus), must issue clusterEvalQ(clus, library(vegan)) make available internal vegan functions. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist Dissimilarity matrix used multidimensional scaling. pc Rotate principal components. center Centre configuration. halfchange Scale axes half-change units. defaults TRUE dissimilarities known theoretical maximum value (ceiling). Function vegdist information attribute maxdist, distfun interpreted simple test (can fail), information may available input data distances. FALSE, ordination dissimilarities scaled range input dissimilarities. threshold Largest dissimilarity used half-change scaling. dissimilarities known (inferred) ceiling, threshold relative ceiling (see halfchange). nthreshold Minimum number points half-change scaling. object result object metaMDS. ... parameters passed functions. Function metaMDS passes arguments component functions metaMDSdist, metaMDSiter, postMDS, distfun engine.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-metric Multidimensional Scaling (NMDS) commonly regarded robust unconstrained ordination method community ecology (Minchin 1987). Function metaMDS wrapper function calls several functions combine Minchin's (1987) recommendations one command. complete steps metaMDS : Transformation: data values larger common abundance class scales, function performs Wisconsin double standardization (wisconsin). values look large, function also performs sqrt transformation. standardizations generally found improve results. However, limits completely arbitrary (present, data maximum 50 triggers sqrt \\(>9\\) triggers wisconsin). want full control analysis, set autotransform = FALSE standardize transform data independently. autotransform intended community data, data types, set autotransform = FALSE. step perfomed using metaMDSdist, step skipped input dissimilarities. Choice dissimilarity: good result, use dissimilarity indices good rank order relation ordering sites along gradients (Faith et al. 1987). default Bray-Curtis dissimilarity, often test winner. However, dissimilarity index vegdist can used. Function rankindex can used finding test winner data gradients. default choice may bad analyse community data, probably select appropriate index using argument distance. step performed using metaMDSdist, step skipped input dissimilarities. Step-across dissimilarities: Ordination may difficult large proportion sites shared species. case, results may improved stepacross dissimilarities, flexible shortest paths among sites. default NMDS engine monoMDS able break tied values maximum dissimilarity, often sufficient handle cases shared species, therefore default use stepacross monoMDS. Function isoMDS handle tied values adequately, therefore default use stepacross always sites shared species engine = \"isoMDS\". stepacross triggered option noshare. like manipulation original distances, set noshare = FALSE. step skipped input data dissimilarities instead community data. step performed using metaMDSdist, step skipped always input dissimilarities. NMDS random starts: NMDS easily gets trapped local optima, must start NMDS several times random starts confident found global solution. strategy metaMDS first run NMDS starting metric scaling (cmdscale usually finds good solution often close local optimum), use previous.best solution supplied, take solution standard (Run 0). metaMDS starts NMDS several random starts (minimum number given try maximum number trymax). random starts generated initMDS. solution better (lower stress) previous standard, taken new standard. solution better close standard, metaMDS compares two solutions using Procrustes analysis (function procrustes option symmetric = TRUE). solutions similar Procrustes rmse largest residual small, solutions regarded repeated better one taken new standard. conditions stringent, may found good relatively similar solutions although function yet satisfied. Setting trace = TRUE monitor final stresses, plot = TRUE display Procrustes overlay plots comparison. step performed using metaMDSiter. first step performed input data (comm) dissimilarities. Random starts can run parallel processing (argument parallel). Scaling results: metaMDS run postMDS final result. Function postMDS provides following ways “fixing” indeterminacy scaling orientation axes NMDS: Centring moves origin average axes; Principal components rotate configuration variance points maximized first dimension (function MDSrotate can alternatively rotate configuration first axis parallel environmental variable); Half-change scaling scales configuration one unit means halving community similarity replicate similarity. Half-change scaling based closer dissimilarities relation ordination distance community dissimilarity rather linear (limit set argument threshold). enough points threshold (controlled parameter nthreshold), dissimilarities regressed distances. intercept regression taken replicate dissimilarity, half-change distance similarity halves according linear regression. Obviously method applicable dissimilarity indices scaled \\(0 \\ldots 1\\), Kulczynski, Bray-Curtis Canberra indices. half-change scaling used, ordination scaled range original dissimilarities. Half-change scaling skipped default input dissimilarities, can turned argument halfchange = TRUE. NB., PC rotation changes directions reference axes, influence configuration solution general. Species scores: Function adds species scores final solution weighted averages using function wascores given value parameter expand. expansion weighted averages can undone shrink = TRUE plot scores functions, calculation species scores can suppressed wascores = FALSE. step skipped input dissimilarities community data unavailable. However, species scores can added replaced sppscores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"results-could-not-be-repeated","dir":"Reference","previous_headings":"","what":"Results Could Not Be Repeated","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-linear optimization hard task, best possible solution (“global optimum”) may found random starting configuration. software solve starting result metric scaling (cmdscale). probably give good result, necessarily “global optimum”. Vegan , metaMDS tries verify improve first solution (“try 0”) using several random starts seeing result can repeated improved improved solution repeated. succeed, get message result repeated. However, result least good usual standard strategy starting metric scaling may improved. may need anything message, can satisfied result. want sure probably “global optimum” may try following instructions. default engine = \"monoMDS\" function tabulate stopping criteria used, can see criterion made stringent. criteria can given arguments metaMDS current values described monoMDS. particular, reach maximum number iterations, increase value maxit. may ask larger number random starts without losing old ones giving previous solution argument previous.best. addition slack convergence criteria low number random starts, wrong number dimensions (argument k) common reason able repeat similar solutions. NMDS usually run low number dimensions (k=2 k=3), complex data increasing k one may help. run NMDS much higher number dimensions (say, k=10 ), reconsider drastically reduce k. heterogeneous data sets partial disjunctions, may help set stepacross, data sets default weakties = TRUE sufficient. Please note can give arguments metaMDS* functions NMDS engine (default monoMDS) metaMDS command, check documentation functions details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"common-wrong-claims","dir":"Reference","previous_headings":"","what":"Common Wrong Claims","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"NMDS often misunderstood wrong claims properties common Web even publications. often claimed NMDS configuration non-metric means fit environmental variables species onto space. false statement. fact, result configuration NMDS metric, can used like ordination result. NMDS rank orders Euclidean distances among points ordination non-metric monotone relationship observed dissimilarities. transfer function observed dissimilarities ordination distances non-metric (Kruskal 1964a, 1964b), ordination result configuration metric observed dissimilarities can kind (metric non-metric). ordination configuration usually rotated principal components metaMDS. rotation performed finding result, changes direction reference axes. important feature NMDS solution ordination distances, change rotation. Similarly, rank order distances change uniform scaling centring configuration points. can also rotate NMDS solution external environmental variables MDSrotate. rotation also change orientation axes, change configuration points distances points ordination space. Function stressplot displays method graphically: plots observed dissimilarities distances ordination space, also shows non-metric monotone regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS returns object class metaMDS. final site ordination stored item points, species ordination item species, stress item stress (NB, scaling stress depends engine: isoMDS uses percents, monoMDS proportions range \\(0 \\ldots 1\\)). items store information steps taken items returned engine function. object print, plot, points text methods. Functions metaMDSdist metaMDSredist return vegdist objects. Function initMDS returns random configuration intended used within isoMDS . Functions metaMDSiter postMDS returns result NMDS updated configuration.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Kruskal, J.B. (1964a). Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. (1964b). Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. (1987). evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS simple wrapper NMDS engine (either monoMDS isoMDS) support functions (metaMDSdist, stepacross, metaMDSiter, initMDS, postMDS, wascores). can call support functions separately better control results. Data transformation, dissimilarities possible stepacross made function metaMDSdist returns dissimilarity result. Iterative search ( starting values initMDS monoMDS) made metaMDSiter. Processing result configuration done postMDS, species scores added wascores. want certain reaching global solution, can compare results several independent runs. can also continue analysis previous results configuration. Function may save used dissimilarity matrix (monoMDS ), metaMDSredist tries reconstruct used dissimilarities original data transformation possible stepacross. metaMDS function designed used community data. type data, probably set arguments non-default values: probably least wascores, autotransform noshare FALSE. negative data entries, metaMDS set previous FALSE warning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"metaMDS uses monoMDS NMDS engine vegan version 2.0-0, replaced isoMDS function. can set argument engine select old engine.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"## The recommended way of running NMDS (Minchin 1987) ## data(dune) ## IGNORE_RDIFF_BEGIN ## Global NMDS using monoMDS sol <- metaMDS(dune) #> Run 0 stress 0.1192678 #> Run 1 stress 0.3016052 #> Run 2 stress 0.1192678 #> ... New best solution #> ... Procrustes: rmse 9.804396e-06 max resid 2.993113e-05 #> ... Similar to previous best #> Run 3 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 0.02027462 max resid 0.06499128 #> Run 4 stress 0.1192679 #> Run 5 stress 0.1183186 #> ... Procrustes: rmse 3.964007e-06 max resid 1.2573e-05 #> ... Similar to previous best #> Run 6 stress 0.1192679 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 6.675204e-06 max resid 1.892118e-05 #> ... Similar to previous best #> Run 8 stress 0.1192678 #> Run 9 stress 0.1183186 #> ... Procrustes: rmse 2.31839e-06 max resid 6.973206e-06 #> ... Similar to previous best #> Run 10 stress 0.1889645 #> Run 11 stress 0.1183186 #> ... Procrustes: rmse 4.550156e-06 max resid 1.4618e-05 #> ... Similar to previous best #> Run 12 stress 0.1192678 #> Run 13 stress 0.1192678 #> Run 14 stress 0.1192678 #> Run 15 stress 0.1192678 #> Run 16 stress 0.1192679 #> Run 17 stress 0.1808912 #> Run 18 stress 0.1183186 #> ... Procrustes: rmse 1.954831e-05 max resid 6.115607e-05 #> ... Similar to previous best #> Run 19 stress 0.1808911 #> Run 20 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 1.3197e-06 max resid 3.321941e-06 #> ... Similar to previous best #> *** Best solution repeated 1 times sol #> #> Call: #> metaMDS(comm = dune) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1183186 #> Stress type 1, weak ties #> Best solution was repeated 1 time in 20 tries #> The best solution was from try 20 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> plot(sol, type=\"t\") ## Start from previous best solution sol <- metaMDS(dune, previous.best = sol) #> Starting from 2-dimensional configuration #> Run 0 stress 0.1183186 #> Run 1 stress 0.1183186 #> ... Procrustes: rmse 5.279158e-06 max resid 1.797511e-05 #> ... Similar to previous best #> Run 2 stress 0.1183186 #> ... Procrustes: rmse 2.460953e-06 max resid 5.877957e-06 #> ... Similar to previous best #> Run 3 stress 0.1192678 #> Run 4 stress 0.1886532 #> Run 5 stress 0.1192678 #> Run 6 stress 0.1808911 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 7.828309e-06 max resid 2.405585e-05 #> ... Similar to previous best #> Run 8 stress 0.1183186 #> ... Procrustes: rmse 6.32253e-06 max resid 2.037397e-05 #> ... Similar to previous best #> Run 9 stress 0.1808911 #> Run 10 stress 0.1183186 #> ... Procrustes: rmse 1.809694e-06 max resid 5.978548e-06 #> ... Similar to previous best #> Run 11 stress 0.1901491 #> Run 12 stress 0.1183186 #> ... Procrustes: rmse 1.546268e-05 max resid 4.957148e-05 #> ... Similar to previous best #> Run 13 stress 0.1812932 #> Run 14 stress 0.1183186 #> ... Procrustes: rmse 5.112711e-06 max resid 1.098817e-05 #> ... Similar to previous best #> Run 15 stress 0.1183186 #> ... Procrustes: rmse 6.246131e-06 max resid 2.000824e-05 #> ... Similar to previous best #> Run 16 stress 0.1192679 #> Run 17 stress 0.1192678 #> Run 18 stress 0.1886532 #> Run 19 stress 0.1192679 #> Run 20 stress 0.1192678 #> *** Best solution repeated 9 times ## Local NMDS and stress 2 of monoMDS sol2 <- metaMDS(dune, model = \"local\", stress=2) #> Run 0 stress 0.1928478 #> Run 1 stress 0.1928482 #> ... Procrustes: rmse 0.0006537695 max resid 0.001899655 #> ... Similar to previous best #> Run 2 stress 0.1928476 #> ... New best solution #> ... Procrustes: rmse 0.0004471422 max resid 0.001294366 #> ... Similar to previous best #> Run 3 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 0.0002885977 max resid 0.0008320155 #> ... Similar to previous best #> Run 4 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 5.59458e-05 max resid 0.0001616763 #> ... Similar to previous best #> Run 5 stress 0.1928475 #> ... Procrustes: rmse 4.864069e-05 max resid 0.0001571056 #> ... Similar to previous best #> Run 6 stress 0.1928477 #> ... Procrustes: rmse 0.0002420844 max resid 0.0006909829 #> ... Similar to previous best #> Run 7 stress 0.1928475 #> ... Procrustes: rmse 0.0001346077 max resid 0.0003887498 #> ... Similar to previous best #> Run 8 stress 0.1928475 #> ... Procrustes: rmse 0.0001625818 max resid 0.0004729866 #> ... Similar to previous best #> Run 9 stress 0.1928475 #> ... Procrustes: rmse 8.000053e-05 max resid 0.0002277773 #> ... Similar to previous best #> Run 10 stress 0.192849 #> ... Procrustes: rmse 0.0006301494 max resid 0.001813666 #> ... Similar to previous best #> Run 11 stress 0.1928475 #> ... Procrustes: rmse 0.0001014964 max resid 0.0002751379 #> ... Similar to previous best #> Run 12 stress 0.192848 #> ... Procrustes: rmse 0.0002964116 max resid 0.0008527077 #> ... Similar to previous best #> Run 13 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 6.461832e-05 max resid 0.0001901839 #> ... Similar to previous best #> Run 14 stress 0.1928482 #> ... Procrustes: rmse 0.0003439501 max resid 0.001017732 #> ... Similar to previous best #> Run 15 stress 0.1928479 #> ... Procrustes: rmse 0.0002821687 max resid 0.0008172892 #> ... Similar to previous best #> Run 16 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 3.13617e-05 max resid 8.909983e-05 #> ... Similar to previous best #> Run 17 stress 0.1928475 #> ... Procrustes: rmse 4.709197e-05 max resid 0.0001417594 #> ... Similar to previous best #> Run 18 stress 0.1928477 #> ... Procrustes: rmse 0.0002414341 max resid 0.0007100825 #> ... Similar to previous best #> Run 19 stress 0.1928479 #> ... Procrustes: rmse 0.0003023992 max resid 0.0008897382 #> ... Similar to previous best #> Run 20 stress 0.1928476 #> ... Procrustes: rmse 0.0002063278 max resid 0.0006048887 #> ... Similar to previous best #> *** Best solution repeated 5 times sol2 #> #> Call: #> metaMDS(comm = dune, model = \"local\", stress = 2) #> #> local Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1928475 #> Stress type 2, weak ties #> Best solution was repeated 5 times in 20 tries #> The best solution was from try 16 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## Use Arrhenius exponent 'z' as a binary dissimilarity measure sol <- metaMDS(dune, distfun = betadiver, distance = \"z\") #> Run 0 stress 0.1067169 #> Run 1 stress 0.1073148 #> Run 2 stress 0.1073148 #> Run 3 stress 0.1073148 #> Run 4 stress 0.1073148 #> Run 5 stress 0.1073148 #> Run 6 stress 0.1067169 #> ... New best solution #> ... Procrustes: rmse 2.925047e-06 max resid 7.049929e-06 #> ... Similar to previous best #> Run 7 stress 0.1069781 #> ... Procrustes: rmse 0.006690411 max resid 0.02349817 #> Run 8 stress 0.1067169 #> ... Procrustes: rmse 6.432458e-06 max resid 1.886945e-05 #> ... Similar to previous best #> Run 9 stress 0.1069786 #> ... Procrustes: rmse 0.006774522 max resid 0.02386629 #> Run 10 stress 0.1067169 #> ... Procrustes: rmse 2.346391e-06 max resid 5.306599e-06 #> ... Similar to previous best #> Run 11 stress 0.1067169 #> ... Procrustes: rmse 2.816635e-06 max resid 7.563776e-06 #> ... Similar to previous best #> Run 12 stress 0.1069784 #> ... Procrustes: rmse 0.006735205 max resid 0.02369698 #> Run 13 stress 0.1067169 #> ... Procrustes: rmse 3.36442e-06 max resid 1.127286e-05 #> ... Similar to previous best #> Run 14 stress 0.2355185 #> Run 15 stress 0.1644741 #> Run 16 stress 0.1067169 #> ... Procrustes: rmse 8.389288e-06 max resid 2.174836e-05 #> ... Similar to previous best #> Run 17 stress 0.1073148 #> Run 18 stress 0.1067169 #> ... Procrustes: rmse 5.169725e-06 max resid 1.633663e-05 #> ... Similar to previous best #> Run 19 stress 0.1067169 #> ... Procrustes: rmse 3.667758e-06 max resid 9.091241e-06 #> ... Similar to previous best #> Run 20 stress 0.1073148 #> *** Best solution repeated 8 times sol #> #> Call: #> metaMDS(comm = dune, distance = \"z\", distfun = betadiver) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: beta.z #> #> Dimensions: 2 #> Stress: 0.1067169 #> Stress type 1, weak ties #> Best solution was repeated 8 times in 20 tries #> The best solution was from try 6 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":null,"dir":"Reference","previous_headings":"","what":"Oribatid Mite Data with Explanatory Variables — mite","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Oribatid mite data. 70 soil cores collected Daniel Borcard 1989. See Borcard et al. (1992, 1994) details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) data(mite.xy)"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"three linked data sets: mite contains data 35 species Oribatid mites, mite.env contains environmental data sampling sites, mite.xy contains geographic coordinates, mite.pcnm contains 22 PCNM base functions (columns) computed geographic coordinates 70 sampling sites (Borcard & Legendre 2002). whole sampling area 2.5 m x 10 m size. fields environmental data : SubsDens Substrate density (g/L) WatrCont Water content substrate (g/L) Substrate Substrate type, factor levels Sphagn1, \tSphagn2 Sphagn3 Sphagn Litter Barepeat Interface Shrub Shrub density, ordered factor levels 1 < 2 < 3 Topo Microtopography, factor levels Blanket Hummock","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Pierre Legendre","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Borcard, D., P. Legendre P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045-1055. Borcard, D. P. Legendre. 1994. Environmental control spatial structure ecological communities: example using Oribatid mites (Acari, Oribatei). Environmental Ecological Statistics 1: 37-61. Borcard, D. P. Legendre. 2002. -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153: 51-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Function implements Kruskal's (1964a,b) non-metric multidimensional scaling (NMDS) using monotone regression primary (“weak”) treatment ties. addition traditional global NMDS, function implements local NMDS, linear hybrid multidimensional scaling.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"monoMDS(dist, y, k = 2, model = c(\"global\", \"local\", \"linear\", \"hybrid\"), threshold = 0.8, maxit = 200, weakties = TRUE, stress = 1, scaling = TRUE, pc = TRUE, smin = 1e-4, sfgrmin = 1e-7, sratmax=0.999999, ...) # S3 method for monoMDS scores(x, choices = NA, ...) # S3 method for monoMDS plot(x, choices = c(1,2), type = \"t\", ...) # S3 method for monoMDS points(x, choices = c(1,2), select, ...) # S3 method for monoMDS text(x, labels, choices = c(1,2), select, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"dist Input dissimilarities. y Starting configuration. random configuration generated missing. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher non-metric MDS. model MDS model: \"global\" normal non-metric MDS monotone regression, \"local\" non-metric MDS separate regressions point, \"linear\" uses linear regression, \"hybrid\" uses linear regression dissimilarities threshold addition monotone regression. See Details. threshold Dissimilarity linear regression used alternately monotone regression. maxit Maximum number iterations. weakties Use primary weak tie treatment, equal observed dissimilarities allowed different fitted values. FALSE, secondary (strong) tie treatment used, tied values broken. stress Use stress type 1 2 (see Details). scaling Scale final scores unit root mean squares. pc Rotate final scores principal components. smin, sfgrmin, sratmax Convergence criteria: iterations stop stress drops smin, scale factor gradient drops sfgrmin, stress ratio two iterations goes sratmax (still \\(< 1\\)). x monoMDS result. choices Dimensions returned plotted. default NA returns dimensions. type type plot: \"t\" text, \"p\" points, \"n\" none. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. labels Labels use used instead row names. ... parameters functions (ignored monoMDS, passed graphical functions plot.).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"several versions non-metric multidimensional scaling R, monoMDS offers following unique combination features: “Weak” treatment ties (Kruskal 1964a,b), tied dissimilarities can broken monotone regression. especially important cases compared sites share species dissimilarities tied maximum value one. Breaking ties allows points different distances can help recovering long coenoclines (gradients). Functions smacof package also hav adequate tie treatment. Handles missing values meaningful way. Offers “local” “hybrid” scaling addition usual “global” NMDS (see ). Uses fast compiled code (isoMDS MASS package also uses compiled code). Function monoMDS uses Kruskal's (1964b) original monotone regression minimize stress. two alternatives stress: Kruskal's (1964a,b) original “stress 1” alternative version “stress 2” (Sibson 1972). stresses can expressed general formula $$s^2 = \\frac{\\sum (d - \\hat d)^2}{\\sum(d - d_0)^2}$$ \\(d\\) distances among points ordination configuration, \\(\\hat d\\) fitted ordination distances, \\(d_0\\) ordination distances null model. “stress 1” \\(d_0 = 0\\), “stress 2” \\(d_0 = \\bar{d}\\) mean distances. “Stress 2” can expressed \\(s^2 = 1 - R^2\\), \\(R^2\\) squared correlation fitted values ordination distances, related “linear fit” stressplot. Function monoMDS can fit several alternative NMDS variants can selected argument model. default model = \"global\" fits global NMDS, Kruskal's (1964a,b) original NMDS similar isoMDS (MASS). Alternative model = \"local\" fits local NMDS independent monotone regression used point (Sibson 1972). Alternative model = \"linear\" fits linear MDS. fits linear regression instead monotone, identical metric scaling principal coordinates analysis (cmdscale) performs eigenvector decomposition dissimilarities (Gower 1966). Alternative model = \"hybrid\" implements hybrid MDS uses monotone regression points linear regression dissimilarities threshold dissimilarity alternating steps (Faith et al. 1987). Function stressplot can used display kind regression model. Scaling, orientation direction axes arbitrary. However, function always centres axes, default scaling scale configuration unit root mean square rotate axes (argument pc) principal components first dimension shows major variation. possible rotate solution first axis parallel given environmental variable using function MDSrotate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"convergence-criteria","dir":"Reference","previous_headings":"","what":"Convergence Criteria","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"NMDS iterative, function stops convergence criteria met. actually criterion assured convergence, solution can local optimum. compare several random starts (use monoMDS via metaMDS) assess solutions likely global optimum. stopping criteria : maxit: Maximum number iterations. Reaching criterion means solutions almost certainly found, maxit increased. smin: Minimum stress. stress nearly zero, fit almost perfect. Usually means data set small requested analysis, may several different solutions almost perfect. reduce number dimensions (k), get data ( observations) use method, metric scaling (cmdscale, wcmdscale). sratmax: Change stress. Values close one mean almost unchanged stress. may mean solution, can also signal stranding suboptimal solution flat stress surface. sfgrmin: Minimum scale factor. Values close zero mean almost unchanged configuration. may mean solution, also happen local optima.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Returns object class \"monoMDS\". final scores returned item points (function scores extracts results), stress item stress. addition, large number items (may change without notice future releases).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Faith, D.P., Minchin, P.R Belbin, L. 1987. Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J.C. (1966). distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Kruskal, J.B. 1964a. Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. 1964b. Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. 1987. evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107. Sibson, R. 1972. Order invariant methods data analysis. Journal Royal Statistical Society B 34, 311--349.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Peter R. Michin (Fortran core) Jari Oksanen (R interface).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"default NMDS function used metaMDS. Function metaMDS adds support functions NMDS can run like recommended Minchin (1987).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"data(dune) dis <- vegdist(dune) m <- monoMDS(dis, model = \"loc\") m #> #> Call: #> monoMDS(dist = dis, model = \"loc\") #> #> Local non-metric Multidimensional Scaling #> #> 20 points, dissimilarity ‘bray’, call ‘vegdist(x = dune)’ #> #> Dimensions: 2 #> Stress: 0.07596745 #> Stress type 1, weak ties #> Scores scaled to unit root mean square, rotated to principal components #> Stopped after 62 iterations: Stress nearly unchanged (ratio > sratmax) plot(m)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":null,"dir":"Reference","previous_headings":"","what":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. Function meandist finds mean within block dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"mrpp(dat, grouping, permutations = 999, distance = \"euclidean\", weight.type = 1, strata = NULL, parallel = getOption(\"mc.cores\")) meandist(dist, grouping, ...) # S3 method for meandist summary(object, ...) # S3 method for meandist plot(x, kind = c(\"dendrogram\", \"histogram\"), cluster = \"average\", ylim, axes = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"dat data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor numeric index grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. used assess significance MRPP statistic, \\(delta\\). distance Choice distance metric measures dissimilarity two observations . See vegdist options. used dat dissimilarity structure symmetric square matrix. weight.type choice group weights. See Details options. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist dist object dissimilarities, produced functions dist, vegdist designdist. . object, x meandist result object. kind Draw dendrogram histogram; see Details. cluster clustering method hclust function kind = \"dendrogram\". hclust method can used, perhaps \"average\" \"single\" make sense. ylim Limits vertical axes (optional). axes Draw scale vertical axis. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. difference may one location (differences mean) one spread (differences within-group distance; cf. Warton et al. 2012). Function mrpp operates data.frame matrix rows observations responses data matrix. response(s) may uni- multivariate. method philosophically mathematically allied analysis variance, compares dissimilarities within among groups. two groups sampling units really different (e.g. species composition), average within-group compositional dissimilarities less average dissimilarities two random collection sampling units drawn entire population. mrpp statistic \\(\\delta\\) overall weighted mean within-group means pairwise dissimilarities among sampling units. choice group weights currently clear. mrpp function offers three choices: (1) group size (\\(n\\)), (2) degrees--freedom analogue (\\(n-1\\)), (3) weight number unique distances calculated among \\(n\\) sampling units (\\(n(n-1)/2\\)). mrpp algorithm first calculates pairwise distances entire dataset, calculates \\(\\delta\\). permutes sampling units associated pairwise distances, recalculates \\(\\delta\\) based permuted data. repeats permutation step permutations times. significance test fraction permuted deltas less observed delta, small sample correction. function also calculates change-corrected within-group agreement \\(= 1 -\\delta/E(\\delta)\\), \\(E(\\delta)\\) expected \\(\\delta\\) assessed average dissimilarities. first argument dat can interpreted dissimilarities, used directly. cases function treats dat observations, uses vegdist find dissimilarities. default distance Euclidean traditional use method, dissimilarities vegdist also available. Function meandist calculates matrix mean within-cluster dissimilarities (diagonal) -cluster dissimilarities (-diagonal elements), attribute n grouping counts. Function summary finds within-class, -class overall means dissimilarities, MRPP statistics weight.type options Classification Strength, CS (Van Sickle Hughes, 2000). CS defined dissimilarities \\(\\bar{B} - \\bar{W}\\), \\(\\bar{B}\\) mean cluster dissimilarity \\(\\bar{W}\\) mean within cluster dissimilarity weight.type = 1. function perform significance tests statistics, must use mrpp appropriate weight.type. currently significance test CS, mrpp weight.type = 1 gives correct test \\(\\bar{W}\\) good approximation CS. Function plot draws dendrogram histogram result matrix based within-group group dissimilarities. dendrogram found method given cluster argument using function hclust. terminal segments hang within-cluster dissimilarity. clusters heterogeneous combined class, leaf segment reversed. histograms based dissimilarities, ore otherwise similar Van Sickle Hughes (2000): horizontal line drawn level mean -cluster dissimilarity vertical lines connect within-cluster dissimilarities line.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"function returns list class mrpp following items: call Function call. delta overall weighted mean group mean distances. E.delta expected delta, null hypothesis group structure. mean original dissimilarities. CS Classification strength (Van Sickle Hughes, 2000). Currently implemented always NA. n Number observations class. classdelta Mean dissimilarities within classes. overall \\(\\delta\\) weighted average values given weight.type . Pvalue Significance test. chance-corrected estimate proportion distances explained group identity; value analogous coefficient determination linear model. distance Choice distance metric used; \"method\" entry dist object. weight.type choice group weights used. boot.deltas vector \"permuted deltas,\" deltas calculated permuted datasets. distribution item can inspected permustats function. permutations number permutations used. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"B. McCune J. B. Grace. 2002. Analysis Ecological Communities. MjM Software Design, Gleneden Beach, Oregon, USA. P. W. Mielke K. J. Berry. 2001. Permutation Methods: Distance Function Approach. Springer Series Statistics. Springer. J. Van Sickle R. M. Hughes 2000. Classification strengths ecoregions, catchments, geographic clusters aquatic vertebrates Oregon. J. N. . Benthol. Soc. 19:370--384. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"M. Henry H. Stevens HStevens@muohio.edu Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"difference may one location (differences mean) one spread (differences within-group distance). , may find significant difference two groups simply one groups greater dissimilarities among sampling units. mrpp models can analysed adonis2 seems suffer problems mrpp robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"data(dune) data(dune.env) dune.mrpp <- with(dune.env, mrpp(dune, Management)) dune.mrpp #> #> Call: #> mrpp(dat = dune, grouping = Management) #> #> Dissimilarity index: euclidean #> Weights for groups: n #> #> Class means and counts: #> #> BF HF NM SF #> delta 10.03 11.08 10.66 12.27 #> n 3 5 6 6 #> #> Chance corrected within-group agreement A: 0.1246 #> Based on observed delta 11.15 and expected delta 12.74 #> #> Significance of delta: 0.001 #> Permutation: free #> Number of permutations: 999 #> # Save and change plotting parameters def.par <- par(no.readonly = TRUE) layout(matrix(1:2,nr=1)) plot(dune.ord <- metaMDS(dune, trace=0), type=\"text\", display=\"sites\" ) with(dune.env, ordihull(dune.ord, Management)) with(dune.mrpp, { fig.dist <- hist(boot.deltas, xlim=range(c(delta,boot.deltas)), main=\"Test of Differences Among Groups\") abline(v=delta); text(delta, 2*mean(fig.dist$counts), adj = -0.5, expression(bold(delta)), cex=1.5 ) } ) par(def.par) ## meandist dune.md <- with(dune.env, meandist(vegdist(dune), Management)) dune.md #> BF HF NM SF #> BF 0.4159972 0.4736637 0.7296979 0.6247169 #> HF 0.4736637 0.4418115 0.7217933 0.5673664 #> NM 0.7296979 0.7217933 0.6882438 0.7723367 #> SF 0.6247169 0.5673664 0.7723367 0.5813015 #> attr(,\"class\") #> [1] \"meandist\" \"matrix\" #> attr(,\"n\") #> grouping #> BF HF NM SF #> 3 5 6 6 summary(dune.md) #> #> Mean distances: #> Average #> within groups 0.5746346 #> between groups 0.6664172 #> overall 0.6456454 #> #> Summary statistics: #> Statistic #> MRPP A weights n 0.1423836 #> MRPP A weights n-1 0.1339124 #> MRPP A weights n(n-1) 0.1099842 #> Classification strength 0.1127012 plot(dune.md) plot(dune.md, kind=\"histogram\")"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":null,"dir":"Reference","previous_headings":"","what":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso adds attribute vario object class \"cca\" describes spatial partitioning cca object performs optional permutation test spatial independence residuals. function plot.mso creates diagnostic plot spatial partitioning \"cca\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"mso(object.cca, object.xy, grain = 1, round.up = FALSE, permutations = 0) msoplot(x, alpha = 0.05, explained = FALSE, ylim = NULL, legend = \"topleft\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"object.cca object class cca, created cca rda function. object.xy vector, matrix data frame spatial coordinates data represented object.cca. number rows must match number observations (given nobs) cca.object. Alternatively, interpoint distances can supplied dist object. grain Interval size distance classes. round.Determines choice breaks. false, distances rounded nearest multiple grain. true, distances rounded upper multiple grain. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x result object mso. alpha Significance level two-sided permutation test Mantel statistic spatial independence residual inertia point-wise envelope variogram total variance. Bonferroni-type correction can achieved dividing overall significance value (e.g. 0.05) number distance classes. explained false, suppresses plotting variogram explained variance. ylim Limits y-axis. legend x y co-ordinates used position legend. can specified keyword way accepted legend. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Mantel test adaptation function mantel vegan package parallel testing several distance classes. compares mean inertia distance class pooled mean inertia distance classes. explanatory variables (RDA, CCA, pRDA, pCCA) significance test residual autocorrelation performed running function mso, function plot.mso print estimate much autocorrelation (based significant distance classes) causes global error variance regression analysis underestimated","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso returns amended cca rda object additional attributes grain, H, H.test vario. grain grain attribute defines interval size distance classes . H H object class 'dist' contains geographic distances observations. H.test H.test contains set dummy variables describe pairs observations (rows = elements object$H) fall distance class (columns). vario vario attribute data frame contains following components rda case (cca case brackets): H Distance class multiples grain. Dist Average distance pairs observations distance class H. n Number unique pairs observations distance class \tH. Empirical (chi-square) variogram total variance \t(inertia). Sum Sum empirical (chi-square) variograms explained \tresidual variance (inertia). CA Empirical (chi-square) variogram residual variance \t(inertia). CCA Empirical (chi-square) variogram explained variance \t(inertia). pCCA Empirical (chi-square) variogram conditioned \tvariance (inertia). se Standard error empirical (chi-square) variogram \ttotal variance (inertia). CA.signif P-value permutation test spatial \tindependence residual variance (inertia).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Wagner, H.H. 2004. Direct multi-scale ordination canonical correspondence analysis. Ecology 85: 342--351.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"responsible author Helene Wagner.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function based code published Ecological Archives E085-006 (doi:10.1890/02-0738 ).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"## Reconstruct worked example of Wagner (submitted): X <- matrix(c(1, 2, 3, 2, 1, 0), 3, 2) Y <- c(3, -1, -2) tmat <- c(1:3) ## Canonical correspondence analysis (cca): Example.cca <- cca(X, Y) Example.cca <- mso(Example.cca, tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.cca) Example.cca$vario #> H Dist n All Sum CA CCA se #> 1 1 1 2 0.25 0.3456633 0.07461735 0.2710459 0 #> 2 2 2 1 1.00 0.8086735 0.01147959 0.7971939 NA ## Correspondence analysis (ca): Example.ca <- mso(cca(X), tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.ca) ## Unconstrained ordination with test for autocorrelation ## using oribatid mite data set as in Wagner (2004) data(mite) data(mite.env) data(mite.xy) mite.cca <- cca(log(mite + 1)) mite.cca <- mso(mite.cca, mite.xy, grain = 1, permutations = 99) msoplot(mite.cca) mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, grain = 1, #> permutations = 99) #> #> Inertia Rank #> Total 1.164 #> Unconstrained 1.164 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3662 0.1328 0.0723 0.0658 0.0559 0.0481 0.0418 0.0391 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All CA CA.signif #> 0 0 0.3555 63 0.6250 0.6250 0.01 #> 1 1 1.0659 393 0.7556 0.7556 0.01 #> 2 2 2.0089 534 0.8931 0.8931 0.01 #> 3 3 2.9786 417 1.0988 1.0988 0.01 #> 4 4 3.9817 322 1.3321 1.3321 0.01 #> 5 5 5.0204 245 1.5109 1.5109 0.01 #> 10 10 6.8069 441 1.7466 1.7466 0.01 #> #> Permutation: free #> Number of permutations: 99 #> ## Constrained ordination with test for residual autocorrelation ## and scale-invariance of species-environment relationships mite.cca <- cca(log(mite + 1) ~ SubsDens + WatrCont + Substrate + Shrub + Topo, mite.env) mite.cca <- mso(mite.cca, mite.xy, permutations = 99) msoplot(mite.cca) #> Error variance of regression model underestimated by 0.4 percent mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, permutations = #> 99) #> #> Inertia Proportion Rank #> Total 1.1638 1.0000 #> Constrained 0.5211 0.4478 11 #> Unconstrained 0.6427 0.5522 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 #> 0.31207 0.06601 0.04117 0.02938 0.02438 0.01591 0.01201 0.00752 0.00612 0.00373 #> CCA11 #> 0.00284 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.07888 0.06752 0.05457 0.04023 0.03855 0.03491 0.03233 0.02692 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All Sum CA CCA se CA.signif #> 0 0 0.3555 63 0.6250 0.7479 0.5512 0.1967 0.03506 0.01 #> 1 1 1.0659 393 0.7556 0.8820 0.6339 0.2482 0.01573 0.18 #> 2 2 2.0089 534 0.8931 0.9573 0.6473 0.3100 0.01487 0.68 #> 3 3 2.9786 417 1.0988 1.1010 0.6403 0.4607 0.01858 0.46 #> 4 4 3.9817 322 1.3321 1.2548 0.6521 0.6027 0.02439 0.98 #> 5 5 5.0204 245 1.5109 1.4564 0.6636 0.7928 0.02801 0.41 #> 10 10 6.8069 441 1.7466 1.6266 0.6914 0.9351 0.02052 0.19 #> #> Permutation: free #> Number of permutations: 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Multiplicative Diversity Partitioning — multipart","title":"Multiplicative Diversity Partitioning — multipart","text":"multiplicative diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set pooled samples (gamma diversity).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"multipart(...) # S3 method for default multipart(y, x, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula multipart(formula, data, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multiplicative Diversity Partitioning — multipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, two levels assumed: row group first level, rows group second level. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variable(s) referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, entropy index calculated (see Details). relative Logical, TRUE beta diversity standardized maximum (see Details). scales Numeric, length 1, order generalized diversity index used. global Logical, indicates calculation beta diversity values, see Details. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. ... arguments passed oecosimu, .e. method, thin burnin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multiplicative Diversity Partitioning — multipart","text":"Multiplicative diversity partitioning based Whittaker's (1972) ideas, recently generalised one parametric diversity families (.e. Rényi Tsallis) Jost (2006, 2007). Jost recommends use numbers equivalents (Hill numbers), instead pure diversities, proofs, satisfies multiplicative partitioning requirements. current implementation multipart calculates Hill numbers based functions renyi tsallis (provided index argument). values one scales desired, done separate runs, adds extra dimensionality implementation, resolved efficiently. Alpha diversities averages Hill numbers hierarchy levels, global gamma diversity alpha value calculated highest hierarchy level. global = TRUE, beta calculated relative global gamma value: $$\\beta_i = \\gamma / \\alpha_{}$$ global = FALSE, beta calculated relative local gamma values (local gamma means diversity calculated particular cluster based pooled abundance vector): $$\\beta_ij = \\alpha_{(+1)j} / mean(\\alpha_{ij})$$ \\(j\\) particular cluster hierarchy level \\(\\). beta diversity value level \\(\\) mean beta values clusters level, \\(\\beta_{} = mean(\\beta_{ij})\\). relative = TRUE, respective beta diversity values standardized maximum possible values (\\(mean(\\beta_{ij}) / \\beta_{max,ij}\\)) given \\(\\beta_{max,ij} = n_{j}\\) (number lower level units given cluster \\(j\\)). expected diversity components calculated nsimul times individual based randomization community data matrix. done \"r2dtable\" method oecosimu default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multiplicative Diversity Partitioning — multipart","text":"object class \"multipart\" structure \"oecosimu\" objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multiplicative Diversity Partitioning — multipart","text":"Jost, L. (2006). Entropy diversity. Oikos, 113, 363--375. Jost, L. (2007). Partitioning diversity independent alpha beta components. Ecology, 88, 2427--2439. Whittaker, R. (1972). Evolution measurement species diversity. Taxon, 21, 213--251.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multiplicative Diversity Partitioning — multipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)))) ## Multiplicative diversity partitioning multipart(mite, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(y = mite, x = levsm, index = \"renyi\", scales = 1, #> nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 11.235 -81.754 14.07974 14.02539 14.07261 14.148 0.05 * #> gamma 12.006 -256.950 14.13219 14.11654 14.13205 14.146 0.05 * #> beta.1 1.071 29.363 1.00373 0.99935 1.00399 1.008 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ l2 + l3, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"renyi\", scales = 1, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -46.004 12.17819 12.02119 12.15824 12.3415 0.05 * #> alpha.2 11.2353 -87.290 14.08463 14.03637 14.07817 14.1383 0.05 * #> alpha.3 12.0064 -490.481 14.13774 14.13229 14.13634 14.1450 0.05 * #> gamma 14.1603 0.000 14.16027 14.16027 14.16027 14.1603 1.00 #> beta.1 1.3568 20.034 1.16000 1.14399 1.16150 1.1779 0.05 * #> beta.2 1.0710 29.394 1.00379 0.99989 1.00410 1.0073 0.05 * #> beta.3 1.1794 577.616 1.00159 1.00108 1.00169 1.0020 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, relative=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.055481 -46.949 12.221716 12.073934 12.244586 12.3654 0.05 * #> alpha.2 11.235261 -123.496 14.087872 14.043345 14.090654 14.1218 0.05 * #> alpha.3 12.006443 -323.402 14.135613 14.124838 14.135306 14.1443 0.05 * #> gamma 14.160271 0.000 14.160271 14.160271 14.160271 14.1603 1.00 #> beta.1 0.078594 17.624 0.068099 0.067172 0.068222 0.0691 0.05 * #> beta.2 0.535514 42.418 0.501697 0.500646 0.501814 0.5031 0.05 * #> beta.3 0.589695 380.700 0.500872 0.500566 0.500883 0.5013 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, global=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, global = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global TRUE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -74.528 12.1813 12.0860 12.1748 12.2676 0.05 * #> alpha.2 11.2353 -92.796 14.0859 14.0459 14.0812 14.1484 0.05 * #> alpha.3 12.0064 -331.084 14.1370 14.1256 14.1378 14.1455 0.05 * #> gamma 14.1603 0.000 14.1603 14.1603 14.1603 14.1603 1.00 #> beta.1 1.7578 112.658 1.1625 1.1543 1.1631 1.1716 0.05 * #> beta.2 1.2603 116.533 1.0053 1.0008 1.0056 1.0081 0.05 * #> beta.3 1.1794 389.764 1.0016 1.0010 1.0016 1.0025 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":null,"dir":"Reference","previous_headings":"","what":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Patches local communities regarded nested subsets community. general, species poor communities subsets species rich communities, rare species occur species rich communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"nestedchecker(comm) nestedn0(comm) nesteddisc(comm, niter = 200) nestedtemp(comm, ...) nestednodf(comm, order = TRUE, weighted = FALSE, wbinary = FALSE) nestedbetasor(comm) nestedbetajac(comm) # S3 method for nestedtemp plot(x, kind = c(\"temperature\", \"incidence\"), col=rev(heat.colors(100)), names = FALSE, ...) # S3 method for nestednodf plot(x, col = \"red\", names = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"comm Community data. niter Number iterations reorder tied columns. x Result object plot. col Colour scheme matrix temperatures. kind kind plot produced. names Label columns rows plot using names comm. logical vector length 2, row column labels returned accordingly. order Order rows columns frequencies. weighted Use species abundances weights interactions. wbinary Modify original method binary data give result weighted unweighted analysis. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"nestedness functions evaluate alternative indices nestedness. functions intended used together Null model communities used argument oecosimu analyse non-randomness results. Function nestedchecker gives number checkerboard units, 2x2 submatrices species occur different sites (Stone & Roberts 1990). Function nestedn0 implements nestedness measure N0 number absences sites richer pauperate site species occurs (Patterson & Atmar 1986). Function nesteddisc implements discrepancy index number ones shifted fill row ones table arranged species frequencies (Brualdi & Sanderson 1999). original definition arranges species (columns) frequencies, method handling tied frequencies. nesteddisc function tries order tied columns minimize discrepancy statistic rather slow, large number tied columns guarantee best ordering found (argument niter gives maximum number tried orders). case warning tied columns issued. Function nestedtemp finds matrix temperature defined sum “surprises” arranged matrix. arranged unsurprising matrix species within proportion given matrix fill upper left corner matrix, surprise absence presences diagonal distance fill line (Atmar & Patterson 1993). Function tries pack species sites low temperature (Rodríguez-Gironés & Santamaria 2006), iterative procedure, temperatures usually vary among runs. Function nestedtemp also plot method can display either incidences temperatures surprises. Matrix temperature rather vaguely described (Atmar & Patterson 1993), Rodríguez-Gironés & Santamaria (2006) explicit description used . However, results probably differ implementations, users cautious interpreting results. details calculations explained vignette Design decisions implementation can read using functions browseVignettes. Function nestedness bipartite package direct port BINMATNEST programme Rodríguez-Gironés & Santamaria (2006). Function nestednodf implements nestedness metric based overlap decreasing fill (Almeida-Neto et al., 2008). Two basic properties required matrix maximum degree nestedness according metric: (1) complete overlap 1's right left columns rows, (2) decreasing marginal totals pairs columns pairs rows. nestedness statistic evaluated separately columns (N columns) rows (N rows) combined whole matrix (NODF). set order = FALSE, statistic evaluated current matrix ordering allowing tests meaningful hypothesis matrix structure default ordering row column totals (breaking ties total abundances weighted = TRUE) (see Almeida-Neto et al. 2008). weighted = TRUE, function finds weighted version index (Almeida-Neto & Ulrich, 2011). However, requires quantitative null models adequate testing. Almeida-Neto & Ulrich (2011) say positive nestedness values first row/column higher second. condition, weighted analysis binary data always give zero nestedness. argument wbinary = TRUE, equality rows/columns also indicates nestedness, binary data give identical results weighted unweighted analysis. However, can also influence results weighted analysis results may differ Almeida-Neto & Ulrich (2011). Functions nestedbetasor nestedbetajac find multiple-site dissimilarities decompose components turnover nestedness following Baselga (2012); pairwise dissimilarities can found designdist. can seen decomposition beta diversity (see betadiver). Function nestedbetasor uses Sørensen dissimilarity turnover component Simpson dissimilarity (Baselga 2012), nestedbetajac uses analogous methods Jaccard index. functions return vector three items: turnover, nestedness sum multiple Sørensen Jaccard dissimilarity. last one total beta diversity (Baselga 2012). functions treat data presence/absence (binary) can used binary nullmodel. overall dissimilarity constant nullmodels fix species (column) frequencies (\"c0\"), components constant row columns also fixed (e.g., model \"quasiswap\"), functions meaningful null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"result returned nestedness function contains item called statistic, components differ among functions. functions constructed can handled oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Almeida-Neto, M., Guimarães, P., Guimarães, P.R., Loyola, R.D. & Ulrich, W. (2008). consistent metric nestedness analysis ecological systems: reconciling concept measurement. Oikos 117, 1227--1239. Almeida-Neto, M. & Ulrich, W. (2011). straightforward computational approach measuring nestedness using quantitative matrices. Env. Mod. Software 26, 173--178. Atmar, W. & Patterson, B.D. (1993). measurement order disorder distribution species fragmented habitat. Oecologia 96, 373--382. Baselga, . (2012). relationship species replacement, dissimilarity derived nestedness, nestedness. Global Ecol. Biogeogr. 21, 1223--1232. Brualdi, R.. & Sanderson, J.G. (1999). Nested species subsets, gaps, discrepancy. Oecologia 119, 256--264. Patterson, B.D. & Atmar, W. (1986). Nested subsets structure insular mammalian faunas archipelagos. Biol. J. Linnean Soc. 28, 65--82. Rodríguez-Gironés, M.. & Santamaria, L. (2006). new algorithm calculate nestedness temperature presence-absence matrices. J. Biogeogr. 33, 924--935. Stone, L. & Roberts, . (1990). checkerboard score species distributions. Oecologia 85, 74--79. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Jari Oksanen Gustavo Carvalho (nestednodf).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"data(sipoo) ## Matrix temperature out <- nestedtemp(sipoo) out #> nestedness temperature: 10.23506 #> with matrix fill 0.2233333 plot(out) plot(out, kind=\"incid\") ## Use oecosimu to assess the non-randomness of checker board units nestedchecker(sipoo) #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 oecosimu(sipoo, nestedchecker, \"quasiswap\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = #> \"quasiswap\") #> #> nullmodel method ‘quasiswap’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.55254 2715.3 2564.9 2713.0 2911.2 0.57 ## Another Null model and standardized checkerboard score oecosimu(sipoo, nestedchecker, \"r00\", statistic = \"C.score\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r00\", #> statistic = \"C.score\") #> #> nullmodel method ‘r00’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> C.score 2.2588 -28.215 9.2285 8.7036 9.2294 9.6735 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract the Number of Observations from a vegan Fit. — nobs.cca","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Extract number ‘observations’ vegan model fit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"","code":"# S3 method for cca nobs(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"object fitted model object. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Function nobs generic R, vegan provides methods objects betadisper, cca related methods, CCorA, decorana, isomap, metaMDS, pcnm, procrustes, radfit, varpart wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"single number, normally integer, giving number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Null Model and Simulation — nullmodel","title":"Null Model and Simulation — nullmodel","text":"nullmodel function creates object, can serve basis Null Model simulation via simulate method. update method updates nullmodel object without sampling (effective sequential algorithms). smbind binds together multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Null Model and Simulation — nullmodel","text":"","code":"nullmodel(x, method) # S3 method for nullmodel print(x, ...) # S3 method for nullmodel simulate(object, nsim = 1, seed = NULL, burnin = 0, thin = 1, ...) # S3 method for nullmodel update(object, nsim = 1, seed = NULL, ...) # S3 method for simmat print(x, ...) smbind(object, ..., MARGIN, strict = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Null Model and Simulation — nullmodel","text":"x community matrix. print method, object printed. method Character, specifying one null model algorithms listed help page commsim. can user supplied object class commsim. object object class nullmodel returned function nullmodel. case smbind simmat object returned update simulate methods. nsim Positive integer, number simulated matrices return. update method, number burnin steps made sequential algorithms update status input model object. seed object specifying random number generator initialized (\"seeded\"). Either NULL integer used call set.seed simulating matrices. set, value saved \"seed\" attribute returned value. default, NULL change random generator state, return .Random.seed \"seed\" attribute, see Value. burnin Nonnegative integer, specifying number steps discarded starting simulation. Active sequential null model algorithms. Ignored non-sequential null model algorithms. thin Positive integer, number simulation steps made returned matrix. Active sequential null model algorithms. Ignored non-sequential null model algorithms. MARGIN Integer, indicating dimension multiple simmat objects bound together smbind. 1: matrices stacked (row bound), 2: matrices column bound, 3: iterations combined. Needs length 1. dimensions expected match across objects. strict Logical, consistency time series attributes (\"start\", \"end\", \"thin\", number simulated matrices) simmat objects strictly enforced binding multiple objects together using smbind. Applies input objects based sequential null model algorithms. ... Additional arguments supplied algorithms. case smbind can contain multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Null Model and Simulation — nullmodel","text":"purpose nullmodel function create object, necessary statistics input matrix calculated . information reused, recalculated step simulation process done simulate method. simulate method carries simulation, simulated matrices stored array. sequential algorithms, method updates state input nullmodel object. Therefore, possible diagnostic tests returned simmat object, make simulations, use increased thinning value desired. update method makes burnin steps case sequential algorithms update status input model without attempt return matrices. non-sequential algorithms method nothing. update preferred way making burnin iterations without sampling. Alternatively, burnin can done via simulate method. convergence diagnostics, recommended use simulate method without burnin. input nullmodel object updated, samples can simulated desired without start process . See Examples. smbind function can used combine multiple simmat objects. comes handy null model simulations stratified sites (MARGIN = 1) species (MARGIN = 2), case multiple objects returned identical/consistent settings e.g. parallel computations (MARGIN = 3). Sanity checks made ensure combining multiple objects sensible, user's responsibility check independence simulated matrices null distribution converged case sequential null model algorithms. strict = FALSE setting can relax checks regarding start, end, thinning values sequential null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Null Model and Simulation — nullmodel","text":"function nullmodel returns object class nullmodel. set objects sharing environment: data: original matrix integer mode. nrow: number rows. ncol: number columns. rowSums: row sums. colSums: column sums. rowFreq: row frequencies (number nonzero cells). colFreq: column frequencies (number nonzero cells). totalSum: total sum. fill: number nonzero cells matrix. commsim: commsim object result method argument. state: current state permutations, matrix similar original. NULL non-sequential algorithms. iter: current number iterations sequential algorithms. NULL non-sequential algorithms. simulate method returns object class simmat. array simulated matrices (third dimension corresponding nsim argument). update method returns current state (last updated matrix) invisibly, update input object sequential algorithms. non sequential algorithms, returns NULL. smbind function returns object class simmat.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Null Model and Simulation — nullmodel","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Null Model and Simulation — nullmodel","text":"Care must taken input matrix contains single row column. input invalid swapping hypergeometric distribution (calling r2dtable) based algorithms. also applies cases input stratified subsets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Null Model and Simulation — nullmodel","text":"","code":"data(mite) x <- as.matrix(mite)[1:12, 21:30] ## non-sequential nullmodel (nm <- nullmodel(x, \"r00\")) #> An object of class “nullmodel” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> (sm <- simulate(nm, nsim=10)) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## sequential nullmodel (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 55, End = 100, Thin = 5 #> ## sequential nullmodel with burnin and extra updating (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, burnin=10, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> ## sequential nullmodel with separate initial burnin (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> nm <- update(nm, nsim=10) (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> ## combining multiple simmat objects ## stratification nm1 <- nullmodel(x[1:6,], \"r00\") sm1 <- simulate(nm1, nsim=10) nm2 <- nullmodel(x[7:12,], \"r00\") sm2 <- simulate(nm2, nsim=10) smbind(sm1, sm2, MARGIN=1) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## binding subsequent samples from sequential algorithms ## start, end, thin retained nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=10) sm1 <- simulate(nm, nsim=10, thin=5) sm2 <- simulate(nm, nsim=20, thin=5) sm3 <- simulate(nm, nsim=10, thin=5) smbind(sm3, sm2, sm1, MARGIN=3) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 40 #> Start = 15, End = 210, Thin = 5 #> ## 'replicate' based usage which is similar to the output ## of 'parLapply' or 'mclapply' in the 'parallel' package ## start, end, thin are set, also noting number of chains smfun <- function(x, burnin, nsim, thin) { nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=burnin) simulate(nm, nsim=nsim, thin=thin) } smlist <- replicate(3, smfun(x, burnin=50, nsim=10, thin=5), simplify=FALSE) smbind(smlist, MARGIN=3) # Number of permuted matrices = 30 #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 30 #> Start = 55, End = 100, Thin = 5 (3 chains) #> if (FALSE) { ## parallel null model calculations library(parallel) if (.Platform$OS.type == \"unix\") { ## forking on Unix systems smlist <- mclapply(1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) smbind(smlist, MARGIN=3) } ## socket type cluster, works on all platforms cl <- makeCluster(3) clusterEvalQ(cl, library(vegan)) clusterExport(cl, c(\"smfun\", \"x\")) smlist <- parLapply(cl, 1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) stopCluster(cl) smbind(smlist, MARGIN=3) }"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function evaluates statistic vector statistics community evaluates significance series simulated random communities. approach used traditionally analysis nestedness, function general can used statistics evaluated simulated communities. Function oecosimu collects evaluates statistics. Null model communities described make.commsim permatfull/ permatswap, definition Null models nullmodel, nestedness statistics nestednodf (describes several alternative statistics, including nestedness temperature, \\(N0\\), checker board units, nestedness discrepancy NODF).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"oecosimu(comm, nestfun, method, nsimul = 99, burnin = 0, thin = 1, statistic = \"statistic\", alternative = c(\"two.sided\", \"less\", \"greater\"), batchsize = NA, parallel = getOption(\"mc.cores\"), ...) # S3 method for oecosimu as.ts(x, ...) # S3 method for oecosimu toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"comm Community data, Null model object generated nullmodel object class simmat (array permuted matrices simulate.nullmodel). comm community data, null model simulation method must specified. comm nullmodel, simulation method ignored, comm simmat object, arguments ignored except nestfun, statistic alternative. nestfun Function analysed. nestedness functions provided vegan (see nestedtemp), function can used accepts community first argument, returns either plain number vector result list item name defined argument statistic. See Examples defining functions. method Null model method: either name (character string) method defined make.commsim commsim function. argument ignored comm nullmodel simmat object. See Details Examples. nsimul Number simulated null communities (ignored comm simmat object). burnin Number null communities discarded proper analysis sequential methods (\"tswap\") (ignored non-sequential methods comm simmat object). thin Number discarded null communities two evaluations nestedness statistic sequential methods (ignored non-sequential methods comm simmat object). statistic name statistic returned nestfun. alternative character string specifying alternative hypothesis, must one \"two.sided\" (default), \"greater\" \"less\". Please note \\(p\\)-value two-sided test approximately two times higher corresponding one-sided test (\"greater\" \"less\" depending sign difference). batchsize Size Megabytes largest simulation object. larger structure produced, analysis broken internally batches. default NA analysis broken batches. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. define nestfun Windows needs R packages vegan permute, must set socket cluster call. x oecosimu result object. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu wrapper evaluates statistic using function given nestfun, simulates series null models based nullmodel, evaluates statistic null models. vegan packages contains nestedness functions described separately (nestedchecker, nesteddisc, nestedn0, nestedtemp, nestednodf), many functions can used long meaningful simulated communities. applicable function must return either statistic plain number vector, list element \"statistic\" (like chisq.test), item whose name given argument statistic. statistic can single number (like typical nestedness index), can vector. vector indices can used analyse site (row) species (column) properties, see treedive example. Raup-Crick index (raupcrick) gives example using dissimilarities. Null model type can given name (quoted character string) used define Null model make.commsim. include binary models described Wright et al. (1998), Jonsson (2001), Gotelli & Entsminger (2003), Miklós & Podani (2004), others. several quantitative Null models, discussed Hardy (2008), several unpublished (see make.commsim, permatfull, permatswap discussion). user can also define commsim function (see Examples). Function works first defining nullmodel given commsim, generating series simulated communities simulate.nullmodel. shortcut can used stages input can Community data (comm), Null model function (nestfun) number simulations (nsimul). nullmodel object number simulations, argument method ignored. three-dimensional array simulated communities generated simulate.nullmodel, arguments method nsimul ignored. last case allows analysing several statistics simulations. function first generates simulations given nullmodel analyses using nestfun. large data sets /large number simulations, generated objects can large, memory exhausted, analysis can become slow system can become unresponsive. simulation broken several smaller batches simulated nullmodel objective set batchsize avoid memory problems (see object.size estimating size current data set). parallel processing still increases memory needs. parallel processing used evaluating nestfun. main load may simulation nullmodel, parallel argument help . Function .ts transforms simulated results sequential methods time series ts object. allows using analytic tools time series studying sequences (see examples). Function toCoda transforms simulated results sequential methods \"mcmc\" object coda package. coda package provides functions analysis stationarity, adequacy sample size, autocorrelation, need burn-much sequential methods, summary results. Please consult documentation coda package. Function permustats provides support standard density, densityplot, qqnorm qqmath functions simulated values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu returns object class \"oecosimu\". result object items statistic oecosimu. statistic contains complete object returned nestfun original data. oecosimu component contains following items: statistic Observed values statistic. simulated Simulated values statistic. means Mean values statistic simulations. z Standardized effect sizes (SES, .k.. \\(z\\)-values) observed statistic based simulations. pval \\(P\\)-values statistic based simulations. alternative type testing given argument alternative. method method used nullmodel. isSeq TRUE method sequential.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Jari Oksanen Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"wonder name oecosimu, look journal names References (nestedtemp). internal structure function radically changed vegan 2.2-0 introduction commsim nullmodel deprecation commsimulator.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"## Use the first eigenvalue of correspondence analysis as an index ## of structure: a model for making your own functions. data(sipoo) ## Traditional nestedness statistics (number of checkerboard units) oecosimu(sipoo, nestedchecker, \"r0\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r0\") #> #> nullmodel method ‘r0’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 -18.742 8017.0 7469.0 8021.0 8450.5 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential model, one-sided test, a vector statistic out <- oecosimu(sipoo, decorana, \"swap\", burnin=100, thin=10, statistic=\"evals\", alt = \"greater\") out #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = decorana, method = \"swap\", #> burnin = 100, thin = 10, statistic = \"evals\", alternative = \"greater\") #> #> nullmodel method ‘swap’ with 99 simulations #> options: thin 10, burnin 100 #> alternative hypothesis: statistic is greater than simulated values #> #> #> Call: #> nestfun(veg = comm) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.4436 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.3822 0.2612 0.1668 0.08723 #> Additive Eigenvalues 0.3822 0.2609 0.1631 0.07650 #> Decorana values 0.4154 0.2465 0.1391 0.04992 #> Axis lengths 2.9197 2.5442 2.7546 1.78074 #> #> #> statistic SES mean 50% 95% Pr(sim.) #> DCA1 0.382249 2.04544 0.32960 0.33249 0.3677 0.01 ** #> DCA2 0.261208 1.77368 0.21549 0.21404 0.2587 0.05 * #> DCA3 0.166788 0.63257 0.15363 0.15405 0.1907 0.22 #> DCA4 0.087226 -1.69622 0.12533 0.12788 0.1636 0.96 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Inspect the swap sequence as a time series object plot(as.ts(out)) lag.plot(as.ts(out)) acf(as.ts(out)) ## Density plot densityplot(permustats(out), as.table = TRUE, layout = c(1,4)) ## Use quantitative null models to compare ## mean Bray-Curtis dissimilarities data(dune) meandist <- function(x) mean(vegdist(x, \"bray\")) mbc1 <- oecosimu(dune, meandist, \"r2dtable\") mbc1 #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = \"r2dtable\") #> #> nullmodel method ‘r2dtable’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 13.84 0.46601 0.44155 0.46756 0.4928 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Define your own null model as a 'commsim' function: shuffle cells ## in each row foo <- function(x, n, nr, nc, ...) { out <- array(0, c(nr, nc, n)) for (k in seq_len(n)) out[,,k] <- apply(x, 2, function(z) sample(z, length(z))) out } cf <- commsim(\"myshuffle\", foo, isSeq = FALSE, binary = FALSE, mode = \"double\") oecosimu(dune, meandist, cf) #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = cf) #> #> nullmodel method ‘myshuffle’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 3.8862 0.63505 0.62995 0.63508 0.64 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Use pre-built null model nm <- simulate(nullmodel(sipoo, \"curveball\"), 99) oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘curveball’ with 99 simulations #> options: thin 1, burnin 0 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 1.4459 2710.6 2635.0 2723.0 2762.7 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Several chains of a sequential model -- this can be generalized ## for parallel processing (see ?smbind) nm <- replicate(5, simulate(nullmodel(sipoo, \"swap\"), 99, thin=10, burnin=100), simplify = FALSE) ## nm is now a list of nullmodels: use smbind to combine these into one ## nullmodel with several chains ## IGNORE_RDIFF_BEGIN nm <- smbind(nm, MARGIN = 3) nm #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 18 x 50 matrix #> Number of permuted matrices = 495 #> Start = 110, End = 1090, Thin = 10 (5 chains) #> oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘swap’ with 495 simulations #> options: thin 10, burnin 100, chains 5 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.71842 2698.1 2572.0 2676.0 2904.7 0.4577 ## IGNORE_RDIFF_END ## After this you can use toCoda() and tools in the coda package to ## analyse the chains (these will show that thin, burnin and nsimul are ## all too low for real analysis)."},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":null,"dir":"Reference","previous_headings":"","what":"Support Functions for Drawing Vectors — ordiArrowTextXY","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Support functions assist drawing vectors (arrows) ordination plots. ordiArrowMul finds multiplier coordinates head vector occupy fill proportion plot region. ordiArrowTextXY finds coordinates locations labels drawn just beyond head vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"ordiArrowTextXY(x, labels, display, choices = c(1,2), rescale = TRUE, fill = 0.75, at = c(0,0), ...) ordiArrowMul(x, at = c(0,0), fill = 0.75, display, choices = c(1,2), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"x R object, scores can determine suitable ordination scores object created envfit, two-column matrix coordinates arrow heads two plot axes. labels Change plotting labels. character vector labels label coordinates sought. supplied, determined row names x, scores(x, ...) required. either defined, suitable labels generated. display character string known scores one methods indicates type scores extract. fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). x created envfit display can set user takes value \"vectors\". Ignored x matrix. choices Axes plotted. rescale logical; coordinates extracted x rescaled fill fill proportion plot region? default always rescale coordinates usually desired objects x coordinates retrieved. supplying x 2-column matrix already rescaled, set FALSE. fill numeric; proportion plot fill span arrows. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. ... Parameters passed scores, strwidth strheight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowMul finds multiplier scale bunch arrows fill ordination plot, ordiArrowTextXY finds coordinates labels arrows. NB., ordiArrowTextXY draw labels; simply returns coordinates labels drawn use another function, text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowTextXY, 2-column matrix coordinates label centres coordinate system currently active plotting device. ordiArrowMul, length-1 vector containing scaling factor.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Jari Oksanen, modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"## Scale arrows by hand to fill 80% of the plot ## Biplot arrows by hand data(varespec, varechem) ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, display = c(\"species\",\"sites\")) ## biplot scores bip <- scores(ord, choices = 1:2, display = \"bp\") ## scaling factor for arrows to fill 80% of plot (mul <- ordiArrowMul(bip, fill = 0.8)) #> [1] 2.092173 bip.scl <- bip * mul # Scale the biplot scores labs <- rownames(bip) # Arrow labels ## calculate coordinate of labels for arrows (bip.lab <- ordiArrowTextXY(bip.scl, rescale = FALSE, labels = labs)) #> [,1] [,2] #> Al 1.9098765 -0.3562415 #> P -0.9298005 -1.6652122 #> K -1.0069931 -0.3764923 ## draw arrows and text labels arrows(0, 0, bip.scl[,1], bip.scl[,2], length = 0.1) text(bip.lab, labels = labs) ## Handling of ordination objects directly mul2 <- ordiArrowMul(ord, display = \"bp\", fill = 0.8) stopifnot(all.equal(mul, mul2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Functions add arrows, line segments, regular grids points. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"ordiarrows(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, startmark, label = FALSE, length = 0.1, ...) ordisegments(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, label = FALSE, ...) ordigrid(ord, levels, replicates, display = \"sites\", lty = c(1,1), col = c(1,1), lwd = c(1,1), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. levels, replicates Alternatively, regular groups can defined arguments levels replicates, levels gives number groups, replicates number successive items group. order.Order points increasing order variable within groups. Reverse sign variable decreasing ordering. display Item displayed. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names. ordiellipse, ordihull ordispider group name centroid object, ordiarrows start arrow, ordisegments ends. ordiellipse ordihull use standard text, others use ordilabel. startmark plotting character used mark first item. default use mark, instance, startmark = 1 draw circle. plotting characters, see pch points. col Colour lines, label borders startmark ordiarrows ordisegments. can vector recycled groups. ordigrid can vector length 2 used levels replicates. length Length edges arrow head (inches). lty, lwd Line type, line width used levels replicates ordigrid. ... Parameters passed graphical functions lines, segments, arrows, scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Function ordiarrows draws arrows ordisegments draws line segments successive items groups. Function ordigrid draws line segments within groups corresponding items among groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"functions add graphical items ordination graph: must draw graph first.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"example(pyrifos) #> #> pyrifs> data(pyrifos) #> #> pyrifs> ditch <- gl(12, 1, length=132) #> #> pyrifs> week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) #> #> pyrifs> dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) mod <- rda(pyrifos) plot(mod, type = \"n\") ## Annual succession by ditches, colour by dose ordiarrows(mod, ditch, label = TRUE, col = as.numeric(dose)) legend(\"topright\", levels(dose), lty=1, col=1:5, title=\"Dose\") ## Show only control and highest Pyrifos treatment plot(mod, type = \"n\") ordiarrows(mod, ditch, label = TRUE, show.groups = c(\"2\", \"3\", \"5\", \"11\")) ordiarrows(mod, ditch, label = TRUE, show = c(\"6\", \"9\"), col = 2) legend(\"topright\", c(\"Control\", \"Pyrifos 44\"), lty = 1, col = c(1,2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions add convex hulls, “spider” graphs, ellipses cluster dendrogram ordination diagrams. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"ordihull(ord, groups, display = \"sites\", draw = c(\"lines\",\"polygon\", \"none\"), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd = NULL, ...) ordiellipse(ord, groups, display=\"sites\", kind = c(\"sd\",\"se\", \"ehull\"), conf, draw = c(\"lines\",\"polygon\", \"none\"), w = weights(ord, display), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd=NULL, ...) ordibar(ord, groups, display = \"sites\", kind = c(\"sd\", \"se\"), conf, w = weights(ord, display), col = 1, show.groups, label = FALSE, lwd = NULL, length = 0, ...) ordispider(ord, groups, display=\"sites\", w = weights(ord, display), spiders = c(\"centroid\", \"median\"), show.groups, label = FALSE, col = NULL, lty = NULL, lwd = NULL, ...) ordicluster(ord, cluster, prune = 0, display = \"sites\", w = weights(ord, display), col = 1, draw = c(\"segments\", \"none\"), ...) # S3 method for ordihull summary(object, ...) # S3 method for ordiellipse summary(object, ...) ordiareatest(ord, groups, area = c(\"hull\", \"ellipse\"), kind = \"sd\", permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. display Item displayed. draw character; objects represented plot? ordihull ordiellipse use either lines polygon draw lines. ordicluster, line segments drawn using segments. suppress plotting, use \"none\". Graphical parameters passed . main difference polygons may filled non-transparent. none nothing drawn, function returns invisible plotting. col Colour hull ellipse lines (draw = \"lines\") fills (draw = \"polygon\") ordihull ordiellipse. draw = \"polygon\", colour bordering lines can set argument border polygon function. functions effect depends underlining functions argument passed . multiple values col specified used element names(table(groups)) (order), shorter vectors recycled. Function ordicluster groups, argument recycled points, colour connecting lines mixture point s cluster. alpha Transparency fill colour draw = \"polygon\" ordihull ordiellipse. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names centroid object. ordiellipse ordihull use standard text, others use ordilabel. w Weights used find average within group. Weights used automatically cca decorana results, unless undone user. w=NULL sets equal weights points. kind Draw standard deviations points (sd), standard errors (se) ellipsoid hulls enclose points group (ehull). conf Confidence limit ellipses, e.g. 0.95. given, corresponding sd se multiplied corresponding value found Chi-squared distribution 2df. spiders centres spider bodies calculated either centroids (averages) spatial medians. cluster Result hierarchic cluster analysis, hclust agnes. prune Number upper level hierarchies removed dendrogram. prune \\(>0\\), dendrogram disconnected. object result object ordihull ordiellipse. result invisible, can saved, used summaries (areas etc. hulls ellipses). area Evaluate area convex hulls ordihull, ellipses ordiellipse. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. lty, lwd, border Vectors parameters can supplied applied (appropriate) element names(table(groups)) (order). Shorter vectors recycled. length Width (inches) small (“caps”) ends bar segment (passed arrows). ... Parameters passed graphical functions scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Function ordihull draws lines polygons convex hulls found function chull encircling items groups. Function ordiellipse draws lines polygons ellipses groups. function can either draw standard deviation points (kind=\"sd\") standard error (weighted) centroids (kind=\"se\"), (weighted) correlation defines direction principal axis ellipse. kind = \"se\" used together argument conf, ellipses show confidence regions locations group centroids. kind=\"ehull\" function draws ellipse encloses points group using ellipsoidhull (cluster package). Function ordibar draws crossed “error bars” using either either standard deviation point scores standard error (weighted) average scores. principal axes corresponding ordiellipse, found principal component analysis (weighted) covariance matrix. Functions ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups. Argument kind can used select kind ellipse, effect convex hulls. Function ordispider draws ‘spider’ diagram point connected group centroid segments. Weighted centroids used correspondence analysis methods cca decorana user gives weights call. ordispider called cca rda result without groups argument, function connects ‘WA’ scores corresponding ‘LC’ score. argument (invisible) ordihull object, function connect points hull centroid. Function ordicluster overlays cluster dendrogram onto ordination. needs result hierarchic clustering hclust agnes, similar structure. Function ordicluster connects cluster centroids line segments. Function uses centroids points clusters, therefore similar average linkage methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"functions add graphical items ordination graph: must draw graph first. draw line segments, grids arrows, see ordisegments, ordigrid andordiarrows.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions ordihull, ordiellipse ordispider return invisible plotting structure. Function ordispider return coordinates point connected (centroids ‘LC’ scores). Function ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ Management, dune.env) plot(mod, type=\"n\", scaling = \"symmetric\") ## Catch the invisible result of ordihull... pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", label = TRUE)) ## ... and find centres and areas of the hulls summary(pl) #> BF HF NM SF #> CCA1 0.2917476 0.36826105 -1.3505642 0.2762936 #> CCA2 0.8632208 0.09419919 0.2681515 -0.8139398 #> Area 0.1951715 0.59943363 1.7398193 1.0144372 ## use more colours and add ellipsoid hulls plot(mod, type = \"n\") pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", col = 1:4, draw=\"polygon\", label =TRUE)) with(dune.env, ordiellipse(mod, Management, scaling = \"symmetric\", kind = \"ehull\", col = 1:4, lwd=3)) ## ordispider to connect WA and LC scores plot(mod, dis=c(\"wa\",\"lc\"), type=\"p\") ordispider(mod) ## Other types of plots plot(mod, type = \"p\", display=\"sites\") cl <- hclust(vegdist(dune)) ordicluster(mod, cl, prune=3, col = cutree(cl, 4)) ## confidence ellipse: location of the class centroids plot(mod, type=\"n\", display = \"sites\") with(dune.env, text(mod, display=\"sites\", labels = as.character(Management), col=as.numeric(Management))) pl <- with(dune.env, ordiellipse(mod, Management, kind=\"se\", conf=0.95, lwd=2, draw = \"polygon\", col=1:4, border=1:4, alpha=63)) summary(pl) #> BF HF NM SF #> CCA1 0.4312652 0.5583211 -1.87848340 0.5601499 #> CCA2 1.3273917 0.6373120 -0.05503211 -1.3859924 #> Area 1.4559842 1.3806668 2.73667419 1.5559135 ## add confidence bars with(dune.env, ordibar(mod, Management, kind=\"se\", conf=0.95, lwd=2, col=1:4, label=TRUE))"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Function ordilabel similar text, text opaque label. can help crowded ordination plots: still see text labels, least uppermost readable. Argument priority helps make important labels visible.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"ordilabel(x, display, labels, choices = c(1, 2), priority, select, cex = 0.8, fill = \"white\", border = NULL, col = NULL, xpd = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"x ordination object object known scores. display Kind scores displayed (passed scores). labels Optional text used plots. given, text found ordination object. choices Axes shown (passed scores). priority Vector length number labels. items high priority plotted uppermost. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. cex Character expansion text (passed text). fill Background colour labels (col argument polygon). border colour visibility border label defined polygon. col Text colour. Default NULL give value border par(\"fg\") border NULL. xpd Draw labels also outside plot region (see par). ... arguments (passed text).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"function may useful crowded ordination plots, particular together argument priority. see text labels, least readable. alternatives crowded plots identify.ordiplot, orditorp orditkplot ( vegan3d package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"data(dune) ord <- cca(dune) plot(ord, type = \"n\") ordilabel(ord, dis=\"sites\", cex=1.2, font=3, fill=\"hotpink\", col=\"blue\") ## You may prefer separate plots, but here species as well ordilabel(ord, dis=\"sp\", font=2, priority=colSums(dune))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Alternative plot and identify Functions for Ordination — ordiplot","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot alternative plotting function can worked vegan ordination result many non-vegan results. addition, plot functions vegan ordinations return invisibly \"ordiplot\" result object, allows using ordiplot support functions result: identify can used add labels selected site, species constraint points, points text can add elements plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"ordiplot(ord, choices = c(1, 2), type=\"points\", display, xlim, ylim, cex = 0.7, ...) # S3 method for ordiplot identify(x, what, labels, ...) # S3 method for ordiplot points(x, what, select, arrows = FALSE, ...) # S3 method for ordiplot text(x, what, labels, select, arrows = FALSE, length = 0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"ord result ordination. choices Axes shown. type type graph may \"points\", \"text\" \"none\" ordination method. display Display \"sites\" \"species\". default methods display , cca, rda, dbrda capscale plot.cca. xlim, ylim x y limits (min,max) plot. cex Character expansion factor points text. ... graphical parameters. x result object ordiplot. Items identified ordination plot. types depend kind plot used. methods know sites species, functions cca rda know addition constraints (LC scores), centroids biplot, plot.procrustes ordination plot heads points. labels Optional text used labels. Row names used missing. arrows Draw arrows origin. always TRUE biplot scores value ignored. Setting TRUE draw arrows type scores. allows, e.g, using biplot arrows species. arrow head value scores, possible text moved outwards. length Length arrow heads (see arrows). select Items displayed. can either logical vector TRUE displayed items vector indices displayed items.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot draws ordination diagram using black circles sites red crosses species. returns invisibly object class ordiplot can used identify.ordiplot label selected sites species, constraints cca rda. function can handle output several alternative ordination methods. cca, rda decorana uses plot method option type = \"points\". addition, plot functions methods return invisibly ordiplot object can used identify.ordiplot label points. ordinations relies scores extract scores. full user control plots, best call ordiplot type = \"none\" save result, add sites species using points.ordiplot text.ordiplot pass arguments corresponding default graphical functions. functions can chained pipes allows alternative intuitive way building plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot returns invisibly object class ordiplot used scores. general, vegan plot functions ordination results also return invisible ordiplot object. plot(..., type = \"n\") used originally, plot empty, items can added invisible object. Functions points text return input object without modification, allows chaining commands pipes. Function identify.ordiplot uses object label point.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"purpose functions provide similar functionality plot, plotid specid methods library labdsv. functions somewhat limited parametrization, can call directly standard identify plot functions better user control.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"## Draw a plot for a non-vegan ordination (cmdscale). data(dune) dune.dis <- vegdist(wisconsin(dune)) dune.mds <- cmdscale(dune.dis, eig = TRUE) dune.mds$species <- wascores(dune.mds$points, dune, expand = TRUE) pl <- ordiplot(dune.mds, type = \"none\") points(pl, \"sites\", pch=21, col=\"red\", bg=\"yellow\") text(pl, \"species\", col=\"blue\", cex=0.9) if (FALSE) { ## same plot using pipes (|>) ordiplot(dune.mds, type=\"n\") |> points(\"sites\", pch=21, col=\"red\", bg=\"yellow\") |> text(\"species\", col=\"blue\", cex=0.9) ## Some people think that species should be shown with arrows in PCA. ## Other ordination methods also return an invisible ordiplot object and ## we can use pipes to draw those arrows. mod <- rda(dune) plot(mod, type=\"n\") |> points(\"sites\", pch=16, col=\"red\") |> text(\"species\", arrows = TRUE, length=0.05, col=\"blue\") } ## Default plot of the previous using identify to label selected points if (FALSE) { pl <- ordiplot(dune.mds) identify(pl, \"spec\")}"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function ordipointlabel produces ordination plots points text label points. points exact location given ordination, function tries optimize location text labels minimize overplotting text. function may useful moderately crowded ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"ordipointlabel(x, display = c(\"sites\", \"species\"), choices = c(1, 2), col = c(1, 2), pch = c(\"o\", \"+\"), font = c(1, 1), cex = c(0.8, 0.8), add = FALSE, select, ...) # S3 method for ordipointlabel plot(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"x ordipointlabel() result object ordination function. plot.ordipointlabel object resulting call ordipointlabel(). display Scores displayed plot. choices Axes shown. col, pch, font, cex Colours, point types, font style character expansion kind scores displayed plot. vectors length number items display. add Add existing plot. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. select used single set scores plotted (.e. length(display) == 1), otherwise ignored warning issued. logical vector used, must length scores plotted. ... arguments passed points text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function uses simulated annealing (optim, method = \"SANN\") optimize location text labels points. eight possible locations: , , sides corners. weak preference text right point, weak avoidance corner positions. exact locations goodness solution varies runs, guarantee finding global optimum. optimization can take long time difficult cases high number potential overlaps. Several sets scores can displayed one plot. function modelled pointLabel maptools package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function returns invisibly object class ordipointlabel items xy coordinates points, labels coordinates labels, items pch, cex font graphical parameters point label. addition, returns result optim attribute \"optim\". unit overlap area character \"m\", variable cex smallest alternative. plot method based orditkplot alter reset graphical parameters via par. result object ordipointlabel inherits orditkplot vegan3d package, may possible edit result object orditkplot, good results necessary points span whole horizontal axis without empty margins.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function designed ordination graphics, optimization works properly plots isometric aspect ratio.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"data(dune) ord <- cca(dune) plt <- ordipointlabel(ord) ## set scaling - should be no warnings! ordipointlabel(ord, scaling = \"sites\") ## plot then add plot(ord, scaling = \"symmetric\", type = \"n\") ordipointlabel(ord, display = \"species\", scaling = \"symm\", add = TRUE) ordipointlabel(ord, display = \"sites\", scaling = \"symm\", add = TRUE) ## redraw plot without rerunning SANN optimisation plot(plt)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function provides plot.lm style diagnostic plots results constrained ordination cca, rda, dbrda capscale. Normally need plots, ordination descriptive make assumptions distribution residuals. However, permute residuals significance tests (anova.cca), may interested inspecting residuals really exchangeable independent fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"ordiresids(x, kind = c(\"residuals\", \"scale\", \"qqmath\"), residuals = \"working\", type = c(\"p\", \"smooth\", \"g\"), formula, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"x Ordination result cca, rda, dbrda, capscale. kind type plot: \"residuals\" plot residuals fitted values, \"scale\" square root absolute residuals fitted values, \"qqmath\" residuals expected distribution (defaults qnorm), unless defined differently formula argument. residuals kind residuals fitted values, alternatives \"working\", \"response\", \"standardized\" \"studentized\" (see Details). type type plot. argument passed lattice functions. formula Formula override default plot. formula can contain items Fitted, Residuals, Species Sites (provided names species sites available ordination result). ... arguments passed lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"default plots similar plot.lm, use Lattice functions xyplot qqmath. alternatives default formulae can replaced user. elements available formula groups argument Fitted, Residuals, Species Sites. residuals = \"response\" residuals = \"working\" fitted values residuals found functions fitted.cca residuals.cca. residuals = \"standardized\" residuals found rstandard.cca, residuals = \"studentized\" found rstudent.cca, cases fitted values standardized sigma.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function returns Lattice object can displayed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ordiresids(mod) ordiresids(mod, formula = Residuals ~ Fitted | Species, residuals=\"standard\", cex = 0.5)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":null,"dir":"Reference","previous_headings":"","what":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Automatic stepwise model building constrained ordination methods (cca, rda, dbrda, capscale). function ordistep modelled step can forward, backward stepwise model selection using permutation tests. Function ordiR2step performs forward model choice solely adjusted \\(R^2\\) \\(P\\)-value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"ordistep(object, scope, direction = c(\"both\", \"backward\", \"forward\"), Pin = 0.05, Pout = 0.1, permutations = how(nperm = 199), steps = 50, trace = TRUE, ...) ordiR2step(object, scope, Pin = 0.05, R2scope = TRUE, permutations = how(nperm = 499), trace = TRUE, R2permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"object ordistep, ordination object inheriting cca rda. scope Defines range models examined stepwise search. can list containing components upper lower, formulae. single item, interpreted target scope, depending direction. direction \"forward\", single item interpreted upper scope formula input object lower scope. See step details. ordiR2step, defines upper scope; can also ordination object model extracted. direction mode stepwise search, can one \"\", \"backward\", \"forward\", default \"\". scope argument missing, default direction \"backward\" ordistep ( ordiR2step argument, works forward). Pin, Pout Limits permutation \\(P\\)-values adding (Pin) term model, dropping (Pout) model. Term added \\(P \\le\\) Pin, removed \\(P >\\) Pout. R2scope Use adjusted \\(R^2\\) stopping criterion: models lower adjusted \\(R^2\\) scope accepted. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. passed anova.cca: see details. steps Maximum number iteration steps dropping adding terms. trace positive, information printed model building. Larger values may give information. R2permutations Number permutations used estimation adjusted \\(R^2\\) cca using RsquareAdj. ... additional arguments add1.cca drop1.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"basic functions model choice constrained ordination add1.cca drop1.cca. functions, ordination models can chosen standard R function step bases term choice AIC. AIC-like statistics ordination provided functions deviance.cca extractAIC.cca ( similar functions rda). Actually, constrained ordination methods AIC, therefore step may trusted. function provides alternative using permutation \\(P\\)-values. Function ordistep defines model, scope models considered, direction procedure similarly step. function alternates drop add steps stops model changed one step. - + signs summary table indicate stage performed. often sensible Pout \\(>\\) Pin stepwise models avoid cyclic adds drops single terms. Function ordiR2step builds model forward maximizes adjusted \\(R^2\\) (function RsquareAdj) every step, stopping adjusted \\(R^2\\) starts decrease, adjusted \\(R^2\\) scope exceeded, selected permutation \\(P\\)-value exceeded (Blanchet et al. 2008). second criterion ignored option R2scope = FALSE, third criterion can ignored setting Pin = 1 (higher). function used adjusted \\(R^2\\) calculated. number predictors higher number observations, adjusted \\(R^2\\) also unavailable. models can analysed R2scope = FALSE, variable selection stop models become overfitted adjusted \\(R^2\\) calculated, adjusted \\(R^2\\) reported zero. \\(R^2\\) cca based simulations (see RsquareAdj) different runs ordiR2step can give different results. Functions ordistep (based \\(P\\) values) ordiR2step (based adjusted \\(R^2\\) hence eigenvalues) can select variables different order.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Functions return selected model one additional component, anova, contains brief information steps taken. can suppress voluminous output model building setting trace = FALSE, find summary model history anova item.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Blanchet, F. G., Legendre, P. & Borcard, D. (2008) Forward selection explanatory variables. Ecology 89, 2623--2632.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"## See add1.cca for another example ### Dune data data(dune) data(dune.env) mod0 <- rda(dune ~ 1, dune.env) # Model with intercept only mod1 <- rda(dune ~ ., dune.env) # Model with all explanatory variables ## With scope present, the default direction is \"both\" mod <- ordistep(mod0, scope = formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.035 * #> + Use 2 91.032 1.1741 0.300 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.085 . #> + A1 1 87.424 1.2965 0.170 #> + Use 2 88.284 1.0510 0.355 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.010 ** #> - Management 3 87.707 2.1769 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.535 #> + Use 2 86.842 0.8027 0.785 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## summary table of steps mod$anova #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 85.567 1.9764 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Example of ordistep, forward ordistep(mod0, scope = formula(mod1), direction=\"forward\") #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.020 * #> + Use 2 91.032 1.1741 0.290 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.055 . #> + A1 1 87.424 1.2965 0.225 #> + Use 2 88.284 1.0510 0.410 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.595 #> + Use 2 86.842 0.8027 0.730 #> #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## Example of ordiR2step (always forward) ## stops because R2 of 'mod1' exceeded ordiR2step(mod0, mod1) #> Step: R2.adj= 0 #> Call: dune ~ 1 #> #> R2.adjusted #> 0.32508817 #> + Management 0.22512409 #> + Moisture 0.20050225 #> + Manure 0.16723149 #> + A1 0.04626579 #> + Use 0.01799755 #> 0.00000000 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.84 0.002 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: R2.adj= 0.2251241 #> Call: dune ~ Management #> #> R2.adjusted #> + Moisture 0.3450334 #> 0.3250882 #> + Manure 0.2779515 #> + A1 0.2392216 #> + Use 0.2300349 #> 0.2251241 #> #> Call: rda(formula = dune ~ Management, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 29.2307 0.3475 3 #> Unconstrained 54.8930 0.6525 16 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 14.865 10.690 3.675 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 #> 15.270 8.428 6.899 5.675 3.988 3.121 2.588 2.380 1.818 1.376 0.995 #> PC12 PC13 PC14 PC15 PC16 #> 0.785 0.661 0.467 0.283 0.159 #>"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface given variable plots result ordination diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"# S3 method for default ordisurf(x, y, choices = c(1, 2), knots = 10, family = \"gaussian\", col = \"red\", isotropic = TRUE, thinplate = TRUE, bs = \"tp\", fx = FALSE, add = FALSE, display = \"sites\", w = weights(x, display), main, nlevels = 10, levels, npoints = 31, labcex = 0.6, bubble = FALSE, cex = 1, select = TRUE, method = \"REML\", gamma = 1, plot = TRUE, lwd.cl = par(\"lwd\"), ...) # S3 method for formula ordisurf(formula, data, ...) # S3 method for ordisurf calibrate(object, newdata, ...) # S3 method for ordisurf plot(x, what = c(\"contour\",\"persp\",\"gam\"), add = FALSE, bubble = FALSE, col = \"red\", cex = 1, nlevels = 10, levels, labcex = 0.6, lwd.cl = par(\"lwd\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"x ordisurf ordination configuration, either matrix result known scores. plot.ordisurf object class \"ordisurf\" returned ordisurf. y Variable plotted / modelled function ordination scores. choices Ordination axes. knots Number initial knots gam (one degrees freedom). knots = 0 knots = 1 function fit linear trend surface, knots = 2 function fit quadratic trend surface instead smooth surface. vector length 2 allowed isotropic = FALSE, first second elements knots referring first second ordination dimensions (indicated choices) respectively. family Error distribution gam. col Colour contours. isotropic, thinplate Fit isotropic smooth surface (.e. smoothness ordination dimensions) via gam. Use thinplate deprecated removed future version package. bs two letter character string indicating smoothing basis use. (e.g. \"tp\" thin plate regression spline, \"cr\" cubic regression spline). One c(\"tp\", \"ts\", \"cr\", \"cs\", \"ds\", \"ps\", \"ad\"). See smooth.terms view refer . default use thin plate splines: bs = \"tp\". fx indicates whether smoothers fixed degree freedom regression splines (fx = FALSE) penalised regression splines (fx = TRUE). Can vector length 2 anisotropic surfaces (isotropic = FALSE). make sense use fx = TRUE select = TRUE error . warning issued specify fx = TRUE forget use select = FALSE though fitting continues using select = FALSE. add Add contours existing diagram draw new plot? display Type scores known scores: typically \"sites\" ordinary site scores \"lc\" linear combination scores. w Prior weights data. Concerns mainly cca decorana results nonconstant weights. main main title plot, default name plotted variable new plot. nlevels, levels Either vector levels contours drawn, suggested number contours nlevels levels supplied. npoints numeric; number locations evaluate fitted surface. represents number locations dimension. labcex Label size contours. Setting zero suppress labels. bubble Use “bubble plot” points, vary point diameter value plotted variable. bubble numeric, value used maximum symbol size ( cex), bubble = TRUE, value cex gives maximum. minimum size always cex = 0.4. option effect add = FALSE. cex Character expansion plotting symbols. select Logical; specify gam argument \"select\". TRUE gam can add extra penalty term can penalized zero. means smoothing parameter estimation part fitting can completely remove terms model. corresponding smoothing parameter estimated zero extra penalty effect. method character; smoothing parameter estimation method. Options allowed : \"GCV.Cp\" uses GCV models unknown scale parameter Mallows' Cp/UBRE/AIC models known scale; \"GACV.Cp\" \"GCV.Cp\" uses GACV (Generalised Approximate CV) instead GCV; \"REML\" \"ML\" use restricted maximum likelihood maximum likelihood estimation known unknown scale; \"P-REML\" \"P-ML\" use REML ML estimation use Pearson estimate scale. gamma Multiplier inflate model degrees freedom GCV UBRE/AIC score . effectively places extra penalty complex models. oft-used value gamma = 1.4. plot logical; plotting done ordisurf? Useful want fitted response surface model. lwd.cl numeric; lwd (line width) parameter use drawing contour lines. formula, data Alternative definition fitted model x ~ y, left-hand side ordination x right-hand side single fitted continuous variable y. variable y must working environment data frame environment given data. arguments passed default method. object ordisurf result object. newdata Coordinates two-dimensional ordination new points. character; type plot produce. \"contour\" produces contour plot response surface, see contour details. \"persp\" produces perspective plot , see persp details. \"gam\" plots fitted GAM model, object inherits class \"gam\" returned ordisurf, see plot.gam. ... parameters passed scores, graphical functions. See Note exceptions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface using penalised splines (Wood 2003) gam, uses predict.gam find fitted values regular grid. smooth surface can fitted extra penalty allows entire smoother penalized back 0 degrees freedom, effectively removing term model (see Marra & Wood, 2011). addition extra penalty invoked setting argument select TRUE. alternative use spline basis includes shrinkage (bs = \"ts\" bs = \"cs\"). ordisurf() exposes large number options gam specifying basis functions used surface. stray defaults, read Notes section relevant documentation s smooth.terms. function plots fitted contours convex hull data points either existing ordination diagram draws new plot. select = TRUE smooth effectively penalised model, contours plotted. gam determines degree smoothness fitted response surface model fitting, unless fx = TRUE. Argument method controls gam performs smoothness selection. See gam details available options. Using \"REML\" \"ML\" yields p-values smooths best coverage properties things matter . function uses scores extract ordination scores, x can result object known function. user can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. behaviour consistent envfit. complete accordance constrained cca, set display = \"lc\". Function calibrate returns fitted values response variable. newdata must coordinates points fitted values desired. function based predict.gam pass extra arguments function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"ordisurf usually called side effect drawing contour plot. function returns result object class \"ordisurf\" inherits gam used internally fit surface, adds item grid contains data grid surface. item grid elements x y vectors axis coordinates, element z matrix fitted values contour. values outside convex hull observed points indicated NA z. gam component result can used analysis like predicting new values (see predict.gam).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Dave Roberts, Jari Oksanen Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"default use isotropic smoother via s employing thin plate regression splines (bs = \"tp\"). make sense ordination equal smoothing directions rotation invariant. However, different degrees smoothness along dimensions required, anisotropic smooth surface may applicable. can achieved use isotropic = FALSE, wherein surface fitted via tensor product smoother via te (unless bs = \"ad\", case separate splines dimension fitted using s). Cubic regression splines P splines can used isotropic = FALSE. Adaptive smooths (bs = \"ad\"), especially two dimensions, require large number observations; without many hundreds observations, default complexities smoother exceed number observations fitting fail. get old behaviour ordisurf use select = FALSE, method = \"GCV.Cp\", fx = FALSE, bs = \"tp\". latter two options current defaults. Graphical arguments supplied plot.ordisurf passed underlying plotting functions, contour, persp, plot.gam. exception arguments col cex can currently passed plot.gam bug way function evaluates arguments arranging plot. work-around call plot.gam directly result call ordisurf. See Examples illustration .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"fitted GAM regression model usual assumptions models. particular note assumption independence residuals. observations independent (e.g. repeat measures set objects, experimental design, inter alia) trust p-values GAM output. need control (.e. add additional fixed effects model, use complex smoothers), extract ordination scores using scores function generate gam call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Marra, G.P & Wood, S.N. (2011) Practical variable selection generalized additive models. Comput. Stat. Data Analysis 55, 2372--2387. Wood, S.N. (2003) Thin plate regression splines. J. R. Statist. Soc. B 65, 95--114.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(varespec) vare.mds <- monoMDS(vare.dist) ## IGNORE_RDIFF_BEGIN ordisurf(vare.mds ~ Baresoil, varechem, bubble = 5) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## as above but without the extra penalties on smooth terms, ## and using GCV smoothness selection (old behaviour of `ordisurf()`): ordisurf(vare.mds ~ Baresoil, varechem, col = \"blue\", add = TRUE, select = FALSE, method = \"GCV.Cp\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.34 total = 8.34 #> #> GCV score: 154.2897 ## Cover of Cladina arbuscula fit <- ordisurf(vare.mds ~ Cladarbu, varespec, family=quasipoisson) ## Get fitted values calibrate(fit) #> 18 15 24 27 23 19 22 #> 21.6245333 8.0344376 3.8614798 2.4595115 6.3958275 5.4260377 6.7158154 #> 16 28 13 14 20 25 7 #> 11.6759768 0.8041457 31.1726268 16.1942820 9.5897933 5.4219853 29.8214589 #> 5 6 3 4 2 9 12 #> 22.8500283 29.9936019 7.1653707 15.5309226 2.8467735 0.9051372 3.5541639 #> 10 11 21 #> 1.3099294 10.7783687 0.9177922 ## Variable selection via additional shrinkage penalties ## This allows non-significant smooths to be selected out ## of the model not just to a linear surface. There are 2 ## options available: ## - option 1: `select = TRUE` --- the *default* ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", select = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## - option 2: use a basis with shrinkage ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", bs = \"ts\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ts\", fx = FALSE) #> #> Estimated degrees of freedom: #> 6.26 total = 7.26 #> #> REML score: 98.83943 ## or bs = \"cs\" with `isotropic = FALSE` ## IGNORE_RDIFF_END ## Plot method plot(fit, what = \"contour\") ## Plotting the \"gam\" object plot(fit, what = \"gam\") ## 'col' and 'cex' not passed on ## or via plot.gam directly library(mgcv) #> Loading required package: nlme #> This is mgcv 1.9-1. For overview type 'help(\"mgcv-package\")'. plot.gam(fit, cex = 2, pch = 1, col = \"blue\") ## 'col' effects all objects drawn... ### controlling the basis functions used ## Use Duchon splines ordisurf(vare.mds ~ Baresoil, varechem, bs = \"ds\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ds\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94117 ## A fixed degrees of freedom smooth, must use 'select = FALSE' ordisurf(vare.mds ~ Baresoil, varechem, knots = 4, fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 4, bs = \"tp\", fx = TRUE) #> #> Estimated degrees of freedom: #> 3 total = 4 #> #> REML score: 85.55126 ## An anisotropic smoother with cubic regression spline bases ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cr\", knots = 4) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(4, 4), bs = c(\"cr\", \"cr\"), fx = c(FALSE, #> FALSE)) #> #> Estimated degrees of freedom: #> 3.08 total = 4.08 #> #> REML score: 93.02782 ## An anisotropic smoother with cubic regression spline with ## shrinkage bases & different degrees of freedom in each dimension ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cs\", knots = c(3,4), fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(3, 4), bs = c(\"cs\", \"cs\"), fx = c(TRUE, #> TRUE)) #> #> Estimated degrees of freedom: #> 11 total = 12 #> #> REML score: 42.9834"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text or Points to Ordination Plots — orditorp","title":"Add Text or Points to Ordination Plots — orditorp","text":"function adds text points ordination plots. Text used can done without overwriting text labels, points used otherwise. function can help reducing clutter ordination graphics, manual editing may still necessary.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"orditorp(x, display, labels, choices = c(1, 2), priority, select, cex = 0.7, pcex, col = par(\"col\"), pcol, pch = par(\"pch\"), air = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text or Points to Ordination Plots — orditorp","text":"x result object ordination ordiplot result. display Items displayed plot. one alternative allowed. Typically \"sites\" \"species\". labels Optional text used labels. Row names used missing. choices Axes shown. priority Text used items higher priority labels overlap. vector length number items plotted. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. logical vector used, must length scores plotted. cex, pcex Text point sizes, see plot.default.. col, pcol Text point colours, see plot.default. pch Plotting character, see points. air Amount empty space text labels. Values <1 allow overlapping text. ... arguments scores (various methods), text points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text or Points to Ordination Plots — orditorp","text":"Function orditorp add either text points existing plot. items high priority added first text used can done without overwriting previous labels,points used otherwise. priority missing, labels added outskirts centre. Function orditorp can used ordination results, plotting results ordiplot ordination plot functions (plot.cca, plot.decorana, plot.metaMDS). Arguments can passed relevant scores method ordination object (x) drawn. See relevant scores help page arguments can used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add Text or Points to Ordination Plots — orditorp","text":"function returns invisibly logical vector TRUE means item labelled text FALSE means marked point. returned vector can used select argument ordination text points functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text or Points to Ordination Plots — orditorp","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"## A cluttered ordination plot : data(BCI) mod <- cca(BCI) plot(mod, dis=\"sp\", type=\"t\") # Now with orditorp and abbreviated species names cnam <- make.cepnames(names(BCI)) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, pch=\"+\", pcol=\"grey\") ## show select in action set.seed(1) take <- sample(ncol(BCI), 50) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, select = take, pch=\"+\", pcol=\"grey\") # \\dontshow{ ## example(orditorp) should not set random seed in the user session rm(.Random.seed) #> Warning: object '.Random.seed' not found # }"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Trellis (Lattice) Plots for Ordination — ordixyplot","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Functions ordicloud, ordisplom ordixyplot provide interface plot ordination results using Trellis functions cloud, splom xyplot package lattice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"ordixyplot(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", aspect = \"iso\", envfit, type = c(\"p\", \"biplot\"), ...) ordisplom(x, data=NULL, formula = NULL, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", type = \"p\", ...) ordicloud(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi3d\", prepanel = \"prepanel.ordi3d\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"x ordination result scores knows: ordination result vegan many others. data Optional data amend ordination results. ordination results found x, may give data variables needed plots. Typically environmental data. formula Formula define plots. default formula used omitted. ordination axes must called names ordination results (names vary among methods). ordisplom, special character . refers ordination result. display kind scores: argument passed scores. choices axes selected: argument passed scores. panel, prepanel names panel prepanel functions. aspect aspect plot (passed lattice function). envfit Result envfit function displayed ordixyplot. Please note needs choices ordixyplot. type type plot. knows alternatives panel.xyplot. addition ordixyplot alternatives \"biplot\", \"arrows\" \"polygon\". first displays fitted vectors factor centroids envfit, constrained ordination, biplot arrows factor centroids envfit given. second (type = \"arrows\") trellis variant ordiarrows draws arrows groups. line parameters controlled trellis.par.set superpose.line, user can set length, angle ends parameters panel.arrows. last one (type = \"polygon\") draws polygon enclosing points panel polygon enclosing points data. overall polygon controlled trellis.par.set plot.polygon, panel polygon controlled superpose.polygon. ... Arguments passed scores methods lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"functions provide interface corresponding lattice functions. graphical parameters passed lattice function graphs extremely configurable. See Lattice xyplot, splom cloud details, usage possibilities. argument x must always ordination result. scores extracted vegan function scores functions work vegan ordinations many others. formula used define models. functions simple default formulae used formula missing. formula omitted ordisplom produces pairs plot ordination axes variables data. formula given, ordination results must referred . variables names. functions, formula must use names ordination scores names data. ordination scores found x, data optional. data contain variables ordination scores used plots. Typically, environmental variables (typically factors) define panels plot symbols. proper work done panel function. layout can changed defining panel functions. See panel.xyplot, panel.splom panel.cloud details survey possibilities. Ordination graphics always isometric: scale used axes. controlled (can changed) argument aspect ordixyplot. ordicloud isometric scaling defined panel prepanel functions. must replace functions want non-isometric scaling graphs. select isometric scaling ordisplom.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"function return Lattice objects class \"trellis\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"data(dune, dune.env) ord <- cca(dune) ## Pairs plots ordisplom(ord) ordisplom(ord, data=dune.env, choices=1:2) ordisplom(ord, data=dune.env, form = ~ . | Management, groups=Manure) ## Scatter plot with polygons ordixyplot(ord, data=dune.env, form = CA1 ~ CA2 | Management, groups=Manure, type = c(\"p\",\"polygon\")) ## Choose a different scaling ordixyplot(ord, scaling = \"symmetric\") ## ... Slices of third axis ordixyplot(ord, form = CA1 ~ CA2 | equal.count(CA3, 4), type = c(\"g\",\"p\", \"polygon\")) ## Display environmental variables ordixyplot(ord, envfit = envfit(ord ~ Management + A1, dune.env, choices=1:3)) ## 3D Scatter plots ordicloud(ord, form = CA2 ~ CA3*CA1, groups = Manure, data = dune.env) ordicloud(ord, form = CA2 ~ CA3*CA1 | Management, groups = Manure, data = dune.env, auto.key = TRUE, type = c(\"p\",\"h\"))"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Coordinates of Neighbourhood Matrix — pcnm","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"function computed classical PCNM principal coordinate analysis truncated distance matrix. commonly used transform (spatial) distances rectangular data suitable constrained ordination regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"pcnm(dis, threshold, w, dist.ret = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"dis distance matrix. threshold threshold value truncation distance. missing, minimum distance giving connected network used. found longest distance minimum spanning tree dis. w Prior weights rows. dist.ret Return distances used calculate PCNMs.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Principal Coordinates Neighbourhood Matrix (PCNM) map distances rows onto rectangular matrix rows using truncation threshold long distances (Borcard & Legendre 2002). original distances Euclidean distances two dimensions (like normal spatial distances), mapped onto two dimensions truncation distances. truncation, higher number principal coordinates. selection truncation distance huge influence PCNM vectors. default use longest distance keep data connected. distances truncation threshold given arbitrary value 4 times threshold. regular data, first PCNM vectors show wide scale variation later PCNM vectors show smaller scale variation (Borcard & Legendre 2002), irregular data interpretation clear. PCNM functions used express distances rectangular form similar normal explanatory variables used , e.g., constrained ordination (rda, cca dbrda) univariate regression (lm) together environmental variables (row weights supplied cca; see Examples). regarded powerful method forcing rectangular environmental data distances using partial mantel analysis (mantel.partial) together geographic distances (Legendre et al. 2008, see Tuomisto & Ruokolainen 2008). function based pcnm function Dray's unreleased spacemakeR package. differences current function uses spantree internal support function. current function also can use prior weights rows using weighted metric scaling wcmdscale. use row weights allows finding orthonormal PCNMs also correspondence analysis (e.g., cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"list following elements: values Eigenvalues obtained principal coordinates analysis. vectors Eigenvectors obtained principal coordinates analysis. scaled unit norm. vectors can extracted scores function. default return PCNM vectors, argument choices selects given vectors. threshold Truncation distance. dist distance matrix values threshold replaced arbitrary value four times threshold. String \"pcnm\" added method attribute, new attribute threshold added distances. returned dist.ret = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Borcard D. Legendre P. (2002) -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153, 51--68. Legendre, P., Borcard, D Peres-Neto, P. (2008) Analyzing explaining beta diversity? Comment. Ecology 89, 3238--3244. Tuomisto, H. & Ruokolainen, K. (2008) Analyzing explaining beta diversity? reply. Ecology 89, 3244--3256.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Jari Oksanen, based code Stephane Dray.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"## Example from Borcard & Legendre (2002) data(mite.xy) pcnm1 <- pcnm(dist(mite.xy)) op <- par(mfrow=c(1,3)) ## Map of PCNMs in the sample plot ordisurf(mite.xy, scores(pcnm1, choi=1), bubble = 4, main = \"PCNM 1\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.71 total = 9.71 #> #> REML score: -120.7705 ordisurf(mite.xy, scores(pcnm1, choi=2), bubble = 4, main = \"PCNM 2\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.18 total = 8.18 #> #> REML score: -103.4662 ordisurf(mite.xy, scores(pcnm1, choi=3), bubble = 4, main = \"PCNM 3\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.32 total = 9.32 #> #> REML score: -94.19053 par(op) ## Plot first PCNMs against each other ordisplom(pcnm1, choices=1:4) ## Weighted PCNM for CCA data(mite) rs <- rowSums(mite)/sum(mite) pcnmw <- pcnm(dist(mite.xy), w = rs) ord <- cca(mite ~ scores(pcnmw)) ## Multiscale ordination: residual variance should have no distance ## trend msoplot(mso(ord, mite.xy))"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":null,"dir":"Reference","previous_headings":"","what":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Individual (count data) incidence (presence-absence data) based null models can generated community level simulations. Options preserving characteristics original matrix (rows/columns sums, matrix fill) restricted permutations (based strata) discussed Details section.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"permatfull(m, fixedmar = \"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, ...) permatswap(m, method = \"quasiswap\", fixedmar=\"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, burnin = 0, thin = 1, ...) # S3 method for permat print(x, digits = 3, ...) # S3 method for permat summary(object, ...) # S3 method for summary.permat print(x, digits = 2, ...) # S3 method for permat plot(x, type = \"bray\", ylab, xlab, col, lty, lowess = TRUE, plot = TRUE, text = TRUE, ...) # S3 method for permat lines(x, type = \"bray\", ...) # S3 method for permat as.ts(x, type = \"bray\", ...) # S3 method for permat toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"m community data matrix plots (samples) rows species (taxa) columns. fixedmar character, stating row/column sums preserved (\"none\", \"rows\", \"columns\", \"\"). strata Numeric vector factor length nrow(m) grouping rows within strata restricted permutations. Unique values levels used. mtype Matrix data type, either \"count\" count data, \"prab\" presence-absence type incidence data. times Number permuted matrices. method Character method used swap algorithm (\"swap\", \"tswap\", \"quasiswap\", \"backtrack\") described function make.commsim. mtype=\"count\" \"quasiswap\", \"swap\", \"swsh\" \"abuswap\" methods available (see details). shuffle Character, indicating whether individuals (\"ind\"), samples (\"samp\") (\"\") shuffled, see details. burnin Number null communities discarded proper analysis sequential (\"swap\", \"tswap\") methods. thin Number discarded permuted matrices two evaluations sequential (\"swap\", \"tswap\") methods. x, object Object class \"permat\" digits Number digits used rounding. ylab, xlab, col, lty graphical parameters plot method. type Character, type plot displayed: \"bray\" Bray-Curtis dissimilarities, \"chisq\" Chi-squared values. lowess, plot, text Logical arguments plot method, whether locally weighted regression curve drawn, plot drawn, statistic values printed plot. ... arguments passed simulate.nullmodel methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"function permatfull useful matrix fill allowed vary, matrix type count. fixedmar argument used set constraints permutation. none margins fixed, cells randomised within matrix. rows columns fixed, cells within rows columns randomised, respectively. margins fixed, r2dtable function used based Patefield's (1981) algorithm. presence absence data, matrix fill necessarily fixed, permatfull wrapper function make.commsim. r00, r0, c0, quasiswap algorithms make.commsim used \"none\", \"rows\", \"columns\", \"\" values fixedmar argument, respectively shuffle argument effect mtype = \"count\" permatfull function used \"none\", \"rows\", \"columns\" values fixedmar. cases count data individual based randomisations. \"samp\" \"\" options result fixed matrix fill. \"\" option means individuals shuffled among non zero cells ensuring cell zeros result, cell (zero new valued cells) shuffled. function permatswap useful matrix fill (.e. proportion empty cells) row/columns sums kept constant. permatswap uses different kinds swap algorithms, row columns sums fixed cases. presence-absence data, swap tswap methods make.commsim can used. count data, special swap algorithm ('swapcount') implemented results permuted matrices fixed marginals matrix fill time. 'quasiswapcount' algorithm (method=\"quasiswap\" mtype=\"count\") uses trick Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function retaining row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping 2x2 submatrices (see 'swapcount' algorithm details) maintain row column totals. algorithm generates independent matrices step, burnin thin arguments considered. default method, sequential (swapcount ) independence subsequent matrices checked. swapcount algorithm (method=\"swap\" mtype=\"count\") tries find 2x2 submatrices (identified 2 random row 2 random column indices), can swapped order leave column row totals fill unchanged. First, algorithm finds largest value submatrix can swapped (\\(d\\)) whether diagonal antidiagonal way. Submatrices contain values larger zero either diagonal antidiagonal position can swapped. Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. algorithm sequential, subsequent matrices independent, swaps modify little matrix large. cases many burnin steps thinning needed get independent random matrices. Although algorithm implemented C, large burnin thin values can slow considerably. WARNING: according simulations, algorithm seems biased non random, thus use avoided! algorithm \"swsh\" function permatswap hybrid algorithm. First, makes binary quasiswaps keep row column incidences constant, non-zero values modified according shuffle argument (\"samp\" \"\" available case, applied non-zero values). also recognizes fixedmar argument \"\" (vegan versions <= 2.0 algorithm fixedmar = \"none\"). algorithm \"abuswap\" produces two kinds null models (based fixedmar=\"columns\" fixedmar=\"rows\") described Hardy (2008; randomization scheme 2x 3x, respectively). preserve column row occurrences, column row sums time. (Note similar constraints can achieved non sequential \"swsh\" algorithm fixedmar argument set \"columns\" \"rows\", respectively.) Constraints row/column sums, matrix fill, total sum sums within strata can checked summary method. plot method visually testing randomness permuted matrices, especially sequential swap algorithms. tendency graph, higher burnin thin values can help sequential methods. New lines can added existing plot lines method. Unrestricted restricted permutations: strata NULL, functions perform unrestricted permutations. Otherwise, used restricted permutations. strata contain least 2 rows order perform randomization (case low row numbers, swap algorithms can rather slow). design well balanced (.e. number observations within stratum), permuted matrices may biased constraints forced submatrices different dimensions. often means, number potential permutations decrease dimensions. constraints put, less randomness can expected. plot method useful graphically testing trend independence permuted matrices. especially important using sequential algorithms (\"swap\", \"tswap\", \"abuswap\"). .ts method can used extract Bray-Curtis dissimilarities Chi-squared values time series. can used testing independence (see Examples). method toCoda useful accessing diagnostic tools available coda package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Functions permatfull permatswap return object class \"permat\" containing function call (call), original data matrix used permutations (orig) list permuted matrices length times (perm). summary method returns various statistics list (including mean Bray-Curtis dissimilarities calculated pairwise among original permuted matrices, Chi-square statistics, check results constraints; see Examples). Note strata used original call, summary calculation may take longer. plot creates plot side effect. .ts method returns object class \"ts\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Original references presence-absence algorithms given help page make.commsim. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Péter Sólymos, solymos@ualberta.ca Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"## A simple artificial community data matrix. m <- matrix(c( 1,3,2,0,3,1, 0,2,1,0,2,1, 0,0,1,2,0,3, 0,0,0,1,4,3 ), 4, 6, byrow=TRUE) ## Using the quasiswap algorithm to create a ## list of permuted matrices, where ## row/columns sums and matrix fill are preserved: x1 <- permatswap(m, \"quasiswap\") summary(x1) #> Summary of object of class 'permat' #> #> Call: permatswap(m = m, method = \"quasiswap\") #> #> Matrix type: count #> Permutation type: swap #> Method: quasiswap_count, burnin: 0, thin: 1 #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 13.13 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3667 0.4333 0.4145 0.4667 0.6000 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 16.27 19.59 21.10 21.51 23.36 31.69 ## Unrestricted permutation retaining ## row/columns sums but not matrix fill: x2 <- permatfull(m) summary(x2) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 16.16 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 0 % #> Column incidences retained: 1.01 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.3667 0.3865 0.4333 0.6333 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 7.824 12.046 15.660 16.015 19.413 28.132 ## Unrestricted permutation of presence-absence type ## not retaining row/columns sums: x3 <- permatfull(m, \"none\", mtype=\"prab\") x3$orig ## note: original matrix is binarized! #> [,1] [,2] [,3] [,4] [,5] [,6] #> [1,] 1 1 1 0 1 1 #> [2,] 0 1 1 0 1 1 #> [3,] 0 0 1 1 0 1 #> [4,] 0 0 0 1 1 1 summary(x3) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, fixedmar = \"none\", mtype = \"prab\") #> #> Matrix type: prab #> Permutation type: full #> Method: r00 #> Restricted: FALSE #> Fixed margins: none #> Individuals and samples are shuffled #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 15 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 4.04 % #> Column sums retained: 0 % #> Row incidences retained: 4.04 % #> Column incidences retained: 0 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.4000 0.3852 0.4000 0.5333 #> #> Chi-squared for original matrix: 8.4 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 8.812 13.583 15.208 15.295 17.083 21.896 ## Restricted permutation, ## check sums within strata: x4 <- permatfull(m, strata=c(1,1,2,2)) summary(x4) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, strata = c(1, 1, 2, 2)) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: TRUE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 38.38 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 2.02 % #> Sums within strata retained: 100 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.06667 0.20000 0.26667 0.23502 0.26667 0.46667 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 14.21 18.68 21.05 22.09 25.26 36.50 ## NOTE: 'times' argument usually needs to be >= 99 ## here much lower value is used for demonstration ## Not sequential algorithm data(BCI) a <- permatswap(BCI, \"quasiswap\", times=19) ## Sequential algorithm b <- permatswap(BCI, \"abuswap\", fixedmar=\"col\", burnin=0, thin=100, times=19) opar <- par(mfrow=c(2,2)) plot(a, main=\"Not sequential\") plot(b, main=\"Sequential\") plot(a, \"chisq\") plot(b, \"chisq\") par(opar) ## Extract Bray-Curtis dissimilarities ## as time series bc <- as.ts(b) ## Lag plot lag.plot(bc) ## First order autoregressive model mar <- arima(bc, c(1,0,0)) mar #> #> Call: #> arima(x = bc, order = c(1, 0, 0)) #> #> Coefficients: #> ar1 intercept #> 0.9915 0.1850 #> s.e. 0.0120 0.1374 #> #> sigma^2 estimated as 0.000346: log likelihood = 46.71, aic = -87.42 ## Ljung-Box test of residuals Box.test(residuals(mar)) #> #> \tBox-Pierce test #> #> data: residuals(mar) #> X-squared = 0.35725, df = 1, p-value = 0.55 #> ## Graphical diagnostics tsdiag(mar)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract, Analyse and Display Permutation Results — permustats","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results vegan functions. support functions can find quantiles standardized effect sizes, plot densities Q-Q plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"permustats(x, ...) # S3 method for permustats summary(object, interval = 0.95, alternative, ...) # S3 method for permustats densityplot(x, data, xlab = \"Permutations\", ...) # S3 method for permustats density(x, observed = TRUE, ...) # S3 method for permustats qqnorm(y, observed = TRUE, ...) # S3 method for permustats qqmath(x, data, observed = TRUE, sd.scale = FALSE, ylab = \"Permutations\", ...) # S3 method for permustats boxplot(x, scale = FALSE, names, ...) # S3 method for permustats pairs(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract, Analyse and Display Permutation Results — permustats","text":"object, x, y object handled. interval numeric; coverage interval reported. alternative character string specifying limits used interval direction test evaluating \\(p\\)-values. Must one \"two.sided\" (upper lower limit), \"greater\" (upper limit), \"less\" (lower limit). Usually alternative given result object, can specified argument. xlab, ylab Arguments densityplot qqmath functions. observed Add observed statistic among permutations. sd.scale Scale permutations unit standard deviation observed statistic standardized effect size. data Ignored. scale Use standardized effect size (SES). names Names boxes (default: names statistics). ... arguments passed function. density passed density.default, boxplot boxplot.default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results observed statistics several vegan functions perform permutations simulations. summary method permustats estimates standardized effect sizes (SES) difference observed statistic mean permutations divided standard deviation permutations (also known \\(z\\)-values). also prints mean, median, limits contain interval percent permuted values. default (interval = 0.95), two-sided test (2.5%, 97.5%) one-sided tests either 5% 95% quantile \\(p\\)-value depending test direction. mean, quantiles \\(z\\) values evaluated permuted values without observed statistic, \\(p\\)-value evaluated observed statistic. intervals \\(p\\)-value evaluated test direction original test, can changed argument alternative. Several permustats objects can combined c function. c function checks statistics equal, performs sanity tests. density densityplot methods display kernel density estimates permuted values. observed value statistic included permuted values, densityplot method marks observed statistic vertical line. However density method uses standard plot method mark observed value. qqnorm qqmath display Q-Q plots permutations, optionally together observed value (default) shown horizontal line plots. qqnorm plots permutation values standard Normal variate. qqmath defaults standard Normal well, can accept alternatives (see standard qqmath). qqmath function can also plot observed statistic standardized effect size (SES) standandized permutations (argument sd.scale). permutations standardized without observed statistic, similarly summary. Functions density qqnorm based standard R methods accept arguments. handle one statistic, used several test statistic evaluated. densityplot qqmath lattice graphics, can used either one several statistics. functions pass arguments underlying functions; see documentation. Functions qqmath densityplot default use axis scaling subplots lattice. can use argument scales set independent scaling subplots appropriate (see xyplot exhaustive list arguments). Function boxplot draws box--whiskers plots effect size, difference permutations observed statistic. scale = TRUE, permutations standardized unit standard deviation, plot show standardized effect sizes. Function pairs plots permutation values statistics . function passes extra arguments pairs. permustats can extract permutation statistics results adonis2, anosim, anova.cca, mantel, mantel.partial, mrpp, oecosimu, ordiareatest, permutest.cca, protest, permutest.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function returns object class \"permustats\". list items \"statistic\" observed statistics, permutations contains permuted values, alternative contains text defining character test (\"two.sided\", \"less\" \"greater\"). qqnorm density methods return standard result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract, Analyse and Display Permutation Results — permustats","text":"Jari Oksanen contributions Gavin L. Simpson (permustats.permutest.betadisper method related modifications summary.permustats print method) Eduard Szöcs (permustats.anova.cca).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"data(dune, dune.env) mod <- adonis2(dune ~ Management + A1, data = dune.env) ## use permustats perm <- permustats(mod) summary(perm) #> #> statistic SES mean lower median upper Pr(perm) #> Management 3.0730 4.6870 1.0387 0.9565 1.8217 0.004 ** #> A1 2.7676 2.7175 1.0022 0.8484 2.2436 0.028 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.95) densityplot(perm) qqmath(perm) boxplot(perm, scale=TRUE, lty=1, pch=16, cex=0.6, col=\"hotpink\", ylab=\"SES\") abline(h=0, col=\"skyblue\") ## example of multiple types of statistic mod <- with(dune.env, betadisper(vegdist(dune), Management)) pmod <- permutest(mod, nperm = 99, pairwise = TRUE) perm <- permustats(pmod) summary(perm, interval = 0.90) #> #> statistic SES mean lower median upper Pr(perm) #> Overall (F) 1.9506 0.7173 1.1427 0.8211 2.4909 0.154 #> BF-HF (t) -0.5634 -0.4124 -0.0443 -2.0202 -0.0293 1.8851 0.591 #> BF-NM (t) -2.2387 -1.8672 -0.0045 -1.8423 0.0074 2.0628 0.067 . #> BF-SF (t) -1.1675 -0.9341 -0.0086 -1.9337 -0.0450 1.9486 0.283 #> HF-NM (t) -2.1017 -1.9328 0.0277 -1.6716 0.0346 1.7582 0.067 . #> HF-SF (t) -0.8789 -0.7872 0.0321 -1.8598 0.0284 1.8394 0.379 #> NM-SF (t) 0.9485 0.8265 0.0121 -1.9118 0.0690 1.7827 0.379 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.9)"},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation tests in Vegan — permutations","title":"Permutation tests in Vegan — permutations","text":"version 2.2-0, vegan significantly improved access restricted permutations brings line offered Canoco. permutation designs modelled permutation schemes Canoco 3.1 (ter Braak, 1990). vegan currently provides following features within permutation tests: Free permutation DATA, also known randomisation, Free permutation DATA within levels grouping variable, Restricted permutations line transects time series, Permutation groups samples whilst retaining within-group ordering, Restricted permutations spatial grids, Blocking, samples never permuted blocks, Split-plot designs, permutation whole plots, split plots, . , use DATA mean either observed data function data, example residuals ordination model presence covariables. capabilities provided functions permute package. user can request particular type permutation supplying permutations argument function object returned , defines samples permuted. Alternatively, user can simply specify required number permutations simple randomisation procedure performed. Finally, user can supply matrix permutations ( number rows equal number permutations number columns equal number observations data) vegan use permutations instead generating new permutations. majority functions vegan allow full range possibilities outlined . Exceptions include kendall.post kendall.global. Null hypothesis first two types permutation test listed assumes free exchangeability DATA (within levels grouping variable, specified). Dependence observations, arises due spatial temporal autocorrelation, -complicated experimental designs, split-plot designs, violates fundamental assumption test requires complex restricted permutation test designs. designs available via permute package vegan provides access version 2.2-0 onwards. Unless otherwise stated help pages specific functions, permutation tests vegan follow format/structure: appropriate test statistic chosen. statistic chosen described help pages individual functions. value test statistic evaluate observed data analysis/model recorded. Denote value \\(x_0\\). DATA randomly permuted according one schemes, value test statistic permutation evaluated recorded. Step 3 repeated total \\(n\\) times, \\(n\\) number permutations requested. Denote values \\(x_i\\), \\(= 1, ..., n\\) Count number values test statistic, \\(x_i\\), Null distribution extreme test statistic observed data \\(x_0\\). Denote count \\(N\\). use phrase extreme include cases two-sided test performed large negative values test statistic considered. permutation p-value computed $$p = \\frac{N + 1}{n + 1}$$ description illustrates default number permutations specified vegan functions takes values 199 999 example. Pretty p values achieved \\(+ 1\\) denominator results division 200 1000, 199 999 random permutations used test. simple intuition behind presence \\(+ 1\\) numerator denominator represent inclusion observed value statistic Null distribution (e.g. Manly 2006). Phipson & Smyth (2010) present compelling explanation inclusion \\(+ 1\\) numerator denominator p value calculation. Fisher (1935) mind permutation test involve enumeration possible permutations data yielding exact test. However, complete enumeration may feasible practice owing potentially vast number arrangements data, even modestly-sized data sets free permutation samples. result evaluate p value tail probability Null distribution test statistic directly random sample possible permutations. Phipson & Smyth (2010) show naive calculation permutation p value $$p = \\frac{N}{n}$$ leads invalid test incorrect type error rate. go show replacing unknown tail probability ( p value) Null distribution biased estimator $$p = \\frac{N + 1}{n + 1}$$ positive bias induced just right size account uncertainty estimation tail probability set randomly sampled permutations yield test correct type error rate. estimator described correct situation permutations data samples randomly without replacement. strictly happens vegan permutations drawn pseudo-randomly independent one another. Note actual chance happening practice small functions permute guarantee generate unique set permutations unless complete enumeration permutations requested. feasible smallest data sets restrictive permutation designs, cases chance drawing set permutations repeats lessened sample size, thence size set possible permutations, increases. situation sampling permutations replacement , tail probability \\(p\\) calculated biased estimator described somewhat conservative, large amount depends number possible values test statistic can take permutation data (Phipson & Smyth, 2010). represents slight loss statistical power conservative p value calculation used . However, unless sample sizes small permutation design set values test statistic can take also small, loss power unlikely critical. minimum achievable p-value $$p_{\\mathrm{min}} = \\frac{1}{n + 1}$$ hence depends number permutations evaluated. However, one simply increase number permutations (\\(n\\)) achieve potentially lower p-value unless number observations available permits number permutations. unlikely problem smallest data sets free permutation (randomisation) valid, restricted permutation designs low number observations, may many unique permutations data might desire reach required level significance. currently responsibility user determine total number possible permutations DATA. number possible permutations allowed specified design can calculated using numPerms permute package. Heuristics employed within shuffleSet function used vegan can triggered generate entire set permutations instead random set. settings controlling triggering complete enumeration step contained within permutation design created using link[permute]{} can set user. See details. Limits total number permutations DATA severe temporally spatially ordered data experimental designs low replication. example, time series \\(n = 100\\) observations just 100 possible permutations including observed ordering. situations low number permutations possible due nature DATA experimental design, enumeration permutations becomes important achievable computationally. , provided brief overview capabilities vegan permute. get best new functionality details set permutation designs using , consult vignette Restricted permutations; using permute package supplied permute accessible via vignette(\"permutations\", package = \"permute\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"random-number-generation","dir":"Reference","previous_headings":"","what":"Random Number Generation","title":"Permutation tests in Vegan — permutations","text":"permutations based random number generator provided R. may change R releases change permutations vegan test results. One change R release 3.6.0. new version clearly better permutation tests use . However, need reproduce old results, can set R random number generator previous version RNGversion.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation tests in Vegan — permutations","text":"Manly, B. F. J. (2006). Randomization, Bootstrap Monte Carlo Methods Biology, Third Edition. Chapman Hall/CRC. Phipson, B., & Smyth, G. K. (2010). Permutation P-values never zero: calculating exact P-values permutations randomly drawn. Statistical Applications Genetics Molecular Biology, 9, Article 39. DOI: 10.2202/1544-6115.1585 ter Braak, C. J. F. (1990). Update notes: CANOCO version 3.1. Wageningen: Agricultural Mathematics Group. (UR). See also: Davison, . C., & Hinkley, D. V. (1997). Bootstrap Methods Application. Cambridge University Press.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation tests in Vegan — permutations","text":"Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Implements permutation-based test multivariate homogeneity group dispersions (variances) results call betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"# S3 method for betadisper permutest(x, pairwise = FALSE, permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"x object class \"betadisper\", result call betadisper. pairwise logical; perform pairwise comparisons group means? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. ... Arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance F. alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution F Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can performed setting argument pairwise TRUE. classical t test performed pairwise group dispersions. combined permutation test based t statistic calculated pairwise group dispersions. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"permutest.betadisper returns list class \"permutest.betadisper\" following components: tab ANOVA table object inheriting class \"data.frame\". pairwise list components observed permuted containing observed permuted p-values pairwise comparisons group mean distances (dispersions variances). groups character; levels grouping factor. control list, result call .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62(1), 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9(6), 683--693.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F pmod <- permutest(mod, permutations = 99, pairwise = TRUE) ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Has permustats() method pstat <- permustats(pmod) densityplot(pstat, scales = list(x = list(relation = \"free\"))) qqmath(pstat, scales = list(relation = \"free\"))"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Functions plot extract results constrained correspondence analysis (cca), redundancy analysis (rda), distance-based redundancy analysis (dbrda) constrained analysis principal coordinates (capscale).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"# S3 method for cca plot(x, choices = c(1, 2), display = c(\"sp\", \"wa\", \"cn\"), scaling = \"species\", type, xlim, ylim, const, correlation = FALSE, hill = FALSE, ...) # S3 method for cca text(x, display = \"sites\", labels, choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca points(x, display = \"sites\", choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", hill = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for rda scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", const, correlation = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for cca summary(object, digits = max(3, getOption(\"digits\") - 3), ...) # S3 method for cca labels(object, display, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"x, object cca result object. choices Axes shown. display Scores shown. must include alternatives \"species\" \"sp\" species scores, sites \"wa\" site scores, \"lc\" linear constraints LC scores, \"bp\" biplot arrows \"cn\" centroids factor constraints instead arrow, \"reg\" regression coefficients (.k.. canonical coefficients). alternative \"\" selects available scores. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. Corresponding negative values can used cca additionally multiply results \\(\\sqrt(1/(1-\\lambda))\\). scaling know Hill scaling (although nothing Hill's rescaling decorana). corresponding negative values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Arguments correlation hill scores.rda scores.cca respectively can used combination character descriptions get corresponding negative value. correlation, hill logical; scaling character description scaling type, correlation hill used select corresponding negative scaling type; either correlation-like scores Hill's scaling PCA/RDA CA/CCA respectively. See argument scaling details. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score, names variable label, weights (CCA) variable weight. possible values score species, sites (WA scores), constraints (LC scores sites calculated directly constraining variables), biplot (biplot arrows), centroids (levels factor variables), factorbiplot (biplot arrows model centroids), regression (regression coefficients find LC scores constraints). scores used conventional plot, directly suitable used ggplot2 package. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. xlim, ylim x y limits (min,max) plot. labels Optional text used instead row names. use , good check default labels order using labels command. arrow.mul Factor expand arrows graph. Arrows scaled automatically fit graph missing. head.arrow Default length arrow heads. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. const General scaling constant rda scores. default use constant gives biplot scores, , scores approximate original data (see vignette ‘Design Decisions’ browseVignettes(\"vegan\") details discussion). const vector two items, first used species, second item site scores. droplist Return matrix instead named list one kind scores requested. axis.bp Draw axis biplot arrows. digits Number digits output. ... Parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function used cca rda. produces quick, standard plot current scaling. plot function sets colours (col), plotting characters (pch) character sizes (cex) certain standard values. fuller control produced plot, best call plot type=\"none\" first, add plotting item separately using text.cca points.cca functions. use default settings standard text points functions accept parameters, allowing full user control produced plots. Environmental variables receive special treatment. display=\"bp\", arrows drawn. labelled text unlabelled points. arrows basically unit scaling, sites scaled (scaling \"sites\" \"symmetric\"), scores requested axes adjusted relative axis highest eigenvalue. scaling = \"species\" scaling = \"none\", arrows consistent vectors fitted linear combination scores (display = \"lc\" function envfit), scaling alternatives differ. basic plot function uses simple heuristics adjusting unit-length arrows current plot area, user can give expansion factor mul.arrow. display=\"cn\" centroids levels factor variables displayed (available factors formula interface used cca rda). option continuous variables still presented arrows ordered factors arrows centroids. display = \"reg\" arrows drawn regression coefficients (.k.. canonical coefficients) constraints conditions. Biplot arrows can interpreted individually, regression coefficients must interpreted together: LC score site sum regressions displayed arrows. partialled conditions zero shown biplot arrows, shown regressions, show effect must partialled get LC scores. biplot arrows standard easily interpreted, regression arrows used know need . want better control plots, best construct plot text points commands accept graphical parameters. important remember use scaling, correlation hill arguments calls. plot.cca command returns invisibly ordiplot result object, consistent scaling elements. easiest way full control graphics first set plot frame using plot type = \"n\" needed scores display save result. points text commands ordiplot allow full graphical control (see section Examples). Utility function labels returns default labels order applied text. Palmer (1993) suggested using linear constraints (“LC scores”) ordination diagrams, gave better results simulations site scores (“WA scores”) step constrained unconstrained analysis. However, McCune (1997) showed noisy environmental variables (environmental measurements noisy) destroy “LC scores” whereas “WA scores” little affected. Therefore plot function uses site scores (“WA scores”) default. consistent usage statistics functions R (lda, cancor).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function returns invisibly plotting structure can used function identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Moisture + Management, dune.env) ## better control -- remember to set scaling etc identically plot(mod, type=\"n\", scaling=\"sites\") text(mod, dis=\"cn\", scaling=\"sites\") points(mod, pch=21, col=\"red\", bg=\"yellow\", cex=1.2, scaling=\"sites\") text(mod, \"species\", col=\"blue\", cex=0.8, scaling=\"sites\") ## catch the invisible result and use ordiplot support - the example ## will make a biplot with arrows for species and correlation scaling pca <- rda(dune) pl <- plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) with(dune.env, points(pl, \"site\", pch=21, col=1, bg=Management)) text(pl, \"sp\", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE) with(dune.env, legend(\"bottomleft\", levels(Management), pch=21, pt.bg=1:4, bty=\"n\")) ## Scaling can be numeric or more user-friendly names ## e.g. Hill's scaling for (C)CA scrs <- scores(mod, scaling = \"sites\", hill = TRUE) ## or correlation-based scores in PCA/RDA scrs <- scores(rda(dune ~ A1 + Moisture + Management, dune.env), scaling = \"sites\", correlation = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Response Curves for Treatments with Repeated Observations — prc","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Principal Response Curves (PRC) special case Redundancy Analysis (rda) multivariate responses repeated observation design. originally suggested ecological communities. easier interpret traditional constrained ordination. can also used study effects factor depend levels factor B, + :B, multivariate response experiment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"prc(response, treatment, time, ...) # S3 method for prc summary(object, axis = 1, scaling = \"sites\", const, digits = 4, correlation = FALSE, ...) # S3 method for prc plot(x, species = TRUE, select, scaling = \"symmetric\", axis = 1, correlation = FALSE, const, type = \"l\", xlab, ylab, ylim, lty = 1:5, col = 1:6, pch, legpos, cex = 0.8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"response Multivariate response data. Typically community (species) data. data counts, probably log transformed prior analysis. treatment factor treatments. time unordered factor defining observations times repeated design. object, x prc result object. axis Axis shown (one axis can selected). scaling Scaling species scores, identical scaling scores.rda. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. const General scaling constant species scores (see scores.rda details). Lower values reduce range species scores, influence regression coefficients. digits Number significant digits displayed. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. species Display species scores. select Vector select displayed species. can vector indices logical vector TRUE selected species type Type plot: \"l\" lines, \"p\" points \"b\" . xlab, ylab Text replace default axis labels. ylim Limits vertical axis. lty, col, pch Line type, colour plotting characters (defaults supplied). legpos position legend. guess made supplied, NA suppress legend. cex Character expansion symbols species labels. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"PRC special case rda single factor treatment single factor time points repeated observations. vegan, corresponding rda model defined rda(response ~ treatment * time + Condition(time)). Since time appears twice model formula, main effects aliased, main effect treatment interaction terms available, used PRC. Instead usual multivariate ordination diagrams, PRC uses canonical (regression) coefficients species scores single axis. current functions provide special summary plot methods display rda results PRC fashion. current version works default contrasts (contr.treatment) coefficients contrasts first level, levels must arranged first level control ( baseline). necessary, must change baseline level function relevel. Function summary prints species scores coefficients. Function plot plots coefficients time using matplot, similar defaults. graph (PRC) meaningful first treatment level control, results contrasts first level unordered factors used. plot also displays species scores right vertical axis using function linestack. Typically number species high can displayed default settings, users can reduce character size padding (air) linestack, select subset species. legend displayed unless suppressed legpos = NA, functions tries guess put legend legpos supplied.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"function special case rda returns result object (see cca.object). However, special summary plot methods display returns differently rda.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Jari Oksanen Cajo ter Braak","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"first level treatment must control: use function relevel guarantee correct reference level. current version ignore user setting contrasts always use treatment contrasts (contr.treatment). time must unordered factor.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"## Chlorpyrifos experiment and experimental design: Pesticide ## treatment in ditches (replicated) and followed over from 4 weeks ## before to 24 weeks after exposure data(pyrifos) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) ditch <- gl(12, 1, length=132) ## IGNORE_RDIFF_BEGIN ## PRC mod <- prc(pyrifos, dose, week) mod # RDA #> Call: prc(response = pyrifos, treatment = dose, time = week) #> #> Inertia Proportion Rank #> Total 288.9920 1.0000 #> Conditional 63.3493 0.2192 10 #> Constrained 96.6837 0.3346 44 #> Unconstrained 128.9589 0.4462 77 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 25.282 8.297 6.044 4.766 4.148 3.857 3.587 3.334 3.087 2.551 2.466 #> RDA12 RDA13 RDA14 RDA15 RDA16 RDA17 RDA18 RDA19 RDA20 RDA21 RDA22 #> 2.209 2.129 1.941 1.799 1.622 1.579 1.440 1.398 1.284 1.211 1.133 #> RDA23 RDA24 RDA25 RDA26 RDA27 RDA28 RDA29 RDA30 RDA31 RDA32 RDA33 #> 1.001 0.923 0.862 0.788 0.750 0.712 0.685 0.611 0.584 0.537 0.516 #> RDA34 RDA35 RDA36 RDA37 RDA38 RDA39 RDA40 RDA41 RDA42 RDA43 RDA44 #> 0.442 0.417 0.404 0.368 0.340 0.339 0.306 0.279 0.271 0.205 0.179 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 17.156 9.189 7.585 6.064 5.730 4.843 4.518 4.105 #> (Showing 8 of 77 unconstrained eigenvalues) #> summary(mod) # PRC #> #> Call: #> prc(response = pyrifos, treatment = dose, time = week) #> Species scores: #> Simve Daplo Cerpu Alogu Aloco Alore Aloaf Copsp #> 2.688099 1.464566 0.542739 0.280040 0.177019 0.315038 0.426524 1.169368 #> Ostsp Slyla Acrha Aloex Chysp Alona Plead Oxyte #> 2.312186 -0.556899 0.105535 0.228092 0.095042 0.063689 0.138397 0.025401 #> Grate Copdi NauLa CilHa Strvi amosp Ascmo Synsp #> 0.096840 1.428854 4.847070 0.895241 3.069709 -1.357663 0.069736 -0.026494 #> Squro Squmu Polar Kerqu Anufi Mytve Mytvi Mytmu #> 0.264390 -0.452667 0.461989 0.495348 0.432767 0.074372 0.090928 0.105891 #> Lepsp Leppa Colob Colbi Colun Lecsp Lecqu Lecco #> 0.998286 0.084809 -0.723051 -0.139569 -0.828338 0.472866 -0.088860 -0.290531 #> Leclu Lecfl Tripo Cepsp Monlo Monae Scalo Trilo #> 0.049788 -0.408035 0.215234 -0.809597 -0.527913 -0.089948 -0.077192 -0.039086 #> Tripo.1 Tricy Trisp Tepat Rotne Notla Filsp Lopox #> 0.246435 0.335400 0.078423 0.007719 0.143730 -0.114301 -0.168356 -0.030990 #> hydrspec bothrosp olchaeta erpoocto glsicomp alglhete hebdstag sphidae #> -0.048698 0.398665 -1.165154 -0.901030 -0.144389 -0.073049 0.902223 1.463655 #> ansuvote armicris bathcont binitent gyraalbu hippcomp lymnstag lymnaes7 #> 0.140685 1.680010 0.073282 -1.950500 0.033051 0.404473 -0.263679 0.135150 #> physfont plbacorn popyanti radiovat radipere valvcris valvpisc hycarina #> -0.026383 0.084761 1.272223 -0.019815 -0.625468 0.010579 -0.267577 1.044034 #> gammpule aselaqua proameri collembo caenhora caenluct caenrobu cloedipt #> 1.526450 1.578743 0.116577 0.029906 5.767844 2.376188 0.126181 4.734035 #> cloesimi aeshniae libellae conagrae corident coripanz coripunc cymabons #> 1.242207 0.212699 -0.081867 1.630574 0.013761 0.120439 -0.176746 0.046360 #> hesplinn hespsahl notoglau notomacu notoobli notoviri pacoconc pleaminu #> -0.069465 0.033240 0.555201 0.050060 0.082356 0.215863 0.016709 0.071168 #> sigadist sigafall sigastri sigarasp colyfusc donacis6 gyrimari haliconf #> 0.076479 -0.018364 0.060206 -0.277312 0.036493 -0.078608 0.010579 0.446911 #> haliflav haligruf haliobli herubrev hya_herm hyglpusi hyhyovat hypoplan #> 0.044538 0.450146 0.131472 -0.128486 -0.322379 -0.011775 0.024196 0.014976 #> hyporusp hytuinae hytuvers laphminu noteclav rhantusp sialluta ablalong #> 0.232042 2.316179 1.772351 0.632897 0.007912 0.067618 1.109341 0.014976 #> ablaphmo cltanerv malopisp mopetenu prdiussp pstavari chironsp crchirsp #> 2.992695 0.075631 0.047658 0.008716 0.554911 0.082842 1.889916 0.016709 #> crclglat ditendsp mitegchl pachgarc pachgvit popegnub popedisp acriluce #> 0.028953 0.083481 0.230629 -0.012187 -0.029907 0.224272 -0.069649 -0.007950 #> chclpige conescut cricotsp liesspec psclbarb psclgsli psclobvi psclplat #> -0.008744 0.821036 0.121530 0.107387 -0.028639 0.601568 -0.362378 -0.052054 #> psclpsil pscladsp cladotsp laa_spec patanysp tatarssp zaa_spec anopmacu #> 0.007339 0.005674 0.539894 0.034105 0.146807 0.669430 0.049949 -0.163731 #> cepogoae chaoobsc cucidae4 tabanusp agdasphr athrater cyrncren holodubi #> 2.555403 2.442310 0.033240 -0.011601 0.271815 0.067618 0.071168 0.094754 #> holopici leceriae lilurhom monaangu mystazur mystloni oecefurv oecelacu #> 0.611618 0.298633 0.009063 0.644295 0.033240 2.998460 0.536628 0.259064 #> triabico paponysp #> 0.088915 0.097788 #> #> Coefficients for dose + week:dose interaction #> which are contrasts to dose 0 #> rows are dose, columns are week #> -4 -1 0.1 1 2 4 8 12 15 #> 0.1 -0.07218 -0.1375 -0.1020 -0.04068 -0.2101 -0.1364 -0.08077 -0.1536 -0.1123 #> 0.9 -0.08106 -0.1935 -0.1936 -0.47699 -0.4977 -0.4306 -0.13532 -0.3548 -0.2408 #> 6 -0.16616 -0.1232 -0.4539 -1.15638 -1.0835 -1.1511 -0.56112 -0.4698 -0.3078 #> 44 -0.13979 -0.1958 -0.7308 -1.26088 -1.2978 -1.4627 -1.29139 -1.0081 -0.7819 #> 19 24 #> 0.1 -0.2163 -0.07835 #> 0.9 -0.1756 -0.15442 #> 6 -0.3293 -0.18227 #> 44 -0.5768 -0.31022 logabu <- colSums(pyrifos) plot(mod, select = logabu > 100) ## IGNORE_RDIFF_END ## Ditches are randomized, we have a time series, and are only ## interested in the first axis ctrl <- how(plots = Plots(strata = ditch,type = \"free\"), within = Within(type = \"series\"), nperm = 99) anova(mod, permutations = ctrl, first=TRUE) #> Permutation test for rda under reduced model #> Plots: ditch, plot permutation: free #> Permutation: series #> Number of permutations: 99 #> #> Model: prc(response = pyrifos, treatment = dose, time = week) #> Df Variance F Pr(>F) #> RDA1 1 25.282 15.096 0.01 ** #> Residual 77 128.959 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function predict can used find site species scores estimates response data new data sets, Function calibrate estimates values constraints new data set. Functions fitted residuals return estimates response data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"# S3 method for cca fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\"), type = c(\"response\", \"working\"), ...) # S3 method for capscale fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\", \"Imaginary\"), type = c(\"response\", \"working\"), ...) # S3 method for cca residuals(object, ...) # S3 method for cca predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", hill = FALSE, ...) # S3 method for rda predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", correlation = FALSE, const, ...) # S3 method for dbrda predict(object, newdata, type = c(\"response\", \"lc\", \"wa\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", const, ...) # S3 method for cca calibrate(object, newdata, rank = \"full\", ...) # S3 method for cca coef(object, norm = FALSE, ...) # S3 method for decorana predict(object, newdata, type = c(\"response\", \"sites\", \"species\"), rank = 4, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"object result object cca, rda, dbrda, capscale decorana. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. fitted method capscale can also \"Imaginary\" imaginary components negative eigenvalues newdata New data frame used prediction calibration. Usually new community data frame, type = \"lc\" constrained component type = \"response\" type = \"working\" must data frame constraints. newdata must number rows original community data cca result type = \"response\" type = \"working\". original model row column names, new data must contain rows columns names (row names species scores, column names \"wa\" scores constraint names \"lc\" scores). cases rows columns must match directly. argument implemented \"wa\" scores dbrda. type type prediction, fitted values residuals: \"response\" scales results ordination gives results, \"working\" gives values used internally, Chi-square standardization cca scaling centring rda. capscale dbrda \"response\" gives dissimilarities, \"working\" internal data structure analysed ordination. Alternative \"wa\" gives site scores weighted averages community data, \"lc\" site scores linear combinations environmental data, \"sp\" species scores. predict.decorana alternatives scores \"sites\" \"species\". rank rank number axes used approximation. default use axes (full rank) \"model\" available four axes predict.decorana. scaling logical, character, numeric; Scaling predicted scores meaning cca, rda, dbrda, capscale. See scores.cca details acceptable values. correlation, hill logical; correlation-like scores Hill's scaling appropriate RDA CCA respectively. See scores.cca additional details. const Constant multiplier RDA scores. used scaling FALSE, default value give similar scaling scores.rda. norm Coefficients variables centred scaled unit norm. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function fitted gives approximation original data matrix dissimilarities ordination result either scale response scaled internally function. Function residuals gives approximation original data unconstrained ordination. argument type = \"response\" fitted.cca residuals.cca function give marginal totals original data matrix, fitted residuals add original data. Functions fitted residuals dbrda capscale give dissimilarities type = \"response\", additive. However, \"working\" scores additive capscale ( dbrda). fitted residuals capscale dbrda include additive constant requested function call. variants fitted residuals defined model mod <- cca(y ~ x), cca(fitted(mod)) equal constrained ordination, cca(residuals(mod)) equal unconstrained part ordination. Function predict can find estimate original data matrix dissimilarities (type = \"response\") rank. rank = \"full\" identical fitted. addition, function can find species scores site scores community data matrix cca rda. function can used new data, can used add new species site scores existing ordinations. function returns (weighted) orthonormal scores default, must specify explicit scaling add scores ordination diagrams. type = \"wa\" function finds site scores species scores. case, new data can contain new sites, species must match original new data. type=\"sp\" function finds species scores site constraints (linear combination scores). case new data can contain new species, sites must match original new data. type = \"lc\" function finds linear combination scores sites environmental data. case new data frame must contain constraining conditioning environmental variables model formula. type = \"response\" type = \"working\" new data must contain environmental variables constrained component desired, community data matrix residual unconstrained component desired. types, function uses newdata find new \"lc\" (constrained) \"wa\" scores (unconstrained) finds response working data new row scores species scores. original site (row) species (column) weights used type = \"response\" type = \"working\" correspondence analysis (cca) therefore number rows must match original data newdata. completely new data frame created, extreme care needed defining variables similarly original model, particular (ordered) factors. ordination performed formula interface, newdata can data frame matrix, extreme care needed columns match original newdata. Function calibrate.cca finds estimates constraints community ordination \"wa\" scores cca, rda capscale. often known calibration, bioindication environmental reconstruction. Basically, method similar projecting site scores onto biplot arrows, uses regression coefficients. function can called newdata cross-validation possible. newdata may contain new sites, species must match original new data. function work ‘partial’ models Condition term, used newdata capscale dbrda results. results may interpretable continuous variables. Function coef give regression coefficients centred environmental variables (constraints conditions) linear combination scores. coefficients unstandardized environmental variables. coefficients NA aliased effects. Function predict.decorana similar predict.cca. However, type = \"species\" available detrended correspondence analysis (DCA), detrending destroys mutual reciprocal averaging (except first axis rescaling used). Detrended CA attempt approximate original data matrix, type = \"response\" meaning detrended analysis (except rank = 1).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"functions return matrices, vectors dissimilarities appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) # Definition of the concepts 'fitted' and 'residuals' mod #> Call: cca(formula = dune ~ A1 + Management + Condition(Moisture), data #> = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Conditional 0.6283 0.2970 3 #> Constrained 0.5109 0.2415 4 #> Unconstrained 0.9761 0.4615 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 #> 0.24932 0.12090 0.08160 0.05904 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> cca(fitted(mod)) #> Call: cca(X = fitted(mod)) #> #> Inertia Rank #> Total 0.5109 #> Unconstrained 0.5109 4 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 #> 0.24932 0.12090 0.08160 0.05904 #> cca(residuals(mod)) #> Call: cca(X = residuals(mod)) #> #> Inertia Rank #> Total 0.9761 #> Unconstrained 0.9761 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> # Remove rare species (freq==1) from 'cca' and find their scores # 'passively'. freq <- specnumber(dune, MARGIN=2) freq #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 7 10 2 8 6 6 5 1 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 1 2 5 6 1 3 5 4 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 12 7 14 13 6 5 7 3 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 18 3 16 3 15 3 mod <- cca(dune[, freq>1] ~ A1 + Management + Condition(Moisture), dune.env) ## IGNORE_RDIFF_BEGIN predict(mod, type=\"sp\", newdata=dune[, freq==1], scaling=\"species\") #> CCA1 CCA2 CCA3 CCA4 #> Chenalbu 1.5737337 0.7842538 0.5503660 -0.35108333 #> Cirsarve 0.5945146 0.3714228 -0.2862647 -0.88373727 #> Empenigr -1.8771953 0.9904299 -0.2446222 -0.04858656 # New sites predict(mod, type=\"lc\", new=data.frame(A1 = 3, Management=\"NM\", Moisture=\"2\"), scal=2) #> CCA1 CCA2 CCA3 CCA4 #> 1 -2.38829 1.230652 -0.2363485 0.3338258 # Calibration and residual plot mod <- cca(dune ~ A1 + Moisture, dune.env) pred <- calibrate(mod) pred #> A1 Moisture.L Moisture.Q Moisture.C #> 1 2.2630533 -0.62633470 -0.20456759 0.220761764 #> 2 4.0510042 -0.47341146 -0.36986691 0.474939409 #> 3 4.2752294 -0.07214500 -0.60797514 0.303213289 #> 4 4.5398659 0.03192745 -1.12417368 0.932223234 #> 5 5.0409406 -0.84235946 0.43000738 -0.291599200 #> 6 5.1962100 -0.91316862 1.11354235 -0.804453944 #> 7 4.2452549 -0.76452556 0.60464291 -0.484842066 #> 8 5.0208369 0.43886340 0.08169514 0.132995916 #> 9 4.2663219 0.10720486 -0.34067849 -0.675151598 #> 10 4.0411356 -0.65472729 0.02832164 0.558402684 #> 11 2.8280051 -0.45762457 0.63079135 -0.089977975 #> 12 5.1204137 0.36328912 -0.69118581 -0.665622948 #> 13 4.9034218 0.47069541 -0.54378271 -0.118643453 #> 14 11.6455841 0.60920550 0.78341426 0.532852308 #> 15 10.7829689 0.69208513 0.82190786 0.237311062 #> 16 7.9892176 0.96421599 0.46793089 0.373647014 #> 17 0.9218684 -0.15822891 0.14593271 1.189161582 #> 18 3.1680733 -0.41737900 1.03352732 -0.236938282 #> 19 -1.2003506 0.57033354 0.72777285 0.509955590 #> 20 4.7876770 1.00324330 1.49898460 0.009202396 ## IGNORE_RDIFF_END with(dune.env, plot(A1, pred[,\"A1\"] - A1, ylab=\"Prediction Error\")) abline(h=0)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":null,"dir":"Reference","previous_headings":"","what":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes rotates configuration maximum similarity another configuration. Function protest tests non-randomness (significance) two configurations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"procrustes(X, Y, scale = TRUE, symmetric = FALSE, scores = \"sites\", ...) # S3 method for procrustes summary(object, digits = getOption(\"digits\"), ...) # S3 method for procrustes plot(x, kind=1, choices=c(1,2), to.target = TRUE, type = \"p\", xlab, ylab, main, ar.col = \"blue\", length=0.05, cex = 0.7, ...) # S3 method for procrustes points(x, display = c(\"target\", \"rotated\"), choices = c(1,2), truemean = FALSE, ...) # S3 method for procrustes text(x, display = c(\"target\", \"rotated\"), choices = c(1,2), labels, truemean = FALSE, ...) # S3 method for procrustes lines(x, type = c(\"segments\", \"arrows\"), choices = c(1, 2), truemean = FALSE, ...) # S3 method for procrustes residuals(object, ...) # S3 method for procrustes fitted(object, truemean = TRUE, ...) # S3 method for procrustes predict(object, newdata, truemean = TRUE, ...) protest(X, Y, scores = \"sites\", permutations = how(nperm = 999), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"X Target matrix Y Matrix rotated. scale Allow scaling axes Y. symmetric Use symmetric Procrustes statistic (rotation still non-symmetric). scores Kind scores used. display argument used corresponding scores function: see scores, scores.cca scores.cca alternatives. x, object object class procrustes. digits Number digits output. kind plot function, kind plot produced: kind = 1 plots shifts two configurations, kind = 0 draws corresponding empty plot, kind = 2 plots impulse diagram residuals. choices Axes (dimensions) plotted. xlab, ylab Axis labels, defaults unacceptable. main Plot title, default unacceptable. display Show \"target\" \"rotated\" matrix points. .target Draw arrows point target. type type plot drawn. plot, type can \"points\" \"text\" select marker tail arrow, \"none\" drawing empty plot. lines type selects either arrows line segments connect target rotated configuration. truemean Use original range target matrix instead centring fitted values. Function plot.procrustes needs truemean = FALSE, adding graphical items plots original results may need truemean = TRUE. newdata Matrix coordinates rotated translated target. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ar.col Arrow colour. length Width arrow head. labels Character vector text labels. Rownames result object used default. cex Character expansion points text. ... parameters passed functions. procrustes protest parameters passed scores, graphical functions underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Procrustes rotation rotates matrix maximum similarity target matrix minimizing sum squared differences. Procrustes rotation typically used comparison ordination results. particularly useful comparing alternative solutions multidimensional scaling. scale=FALSE, function rotates matrix Y. scale=TRUE, scales linearly configuration Y maximum similarity. Since Y scaled fit X, scaling non-symmetric. However, symmetric=TRUE, configurations scaled equal dispersions symmetric version Procrustes statistic computed. Instead matrix, X Y can results ordination scores can extract results. Function procrustes passes extra arguments scores, scores.cca etc. can specify arguments scaling. Function plot plots procrustes object returns invisibly ordiplot object function identify.ordiplot can used identifying points. items ordiplot object called heads points kind=1 (ordination diagram) sites kind=2 (residuals). ordination diagrams, arrow heads point target configuration .target = TRUE, rotated configuration .target = FALSE. Target original rotated axes shown cross hairs two-dimensional Procrustes analysis, higher number dimensions, rotated axes projected onto plot scaled centred range. Function plot passes parameters underlying plotting functions. full control plots, can draw axes using plot kind = 0, add items points lines. functions pass parameters underlying functions can select plotting characters, size, colours etc., can select width, colour type line segments arrows, can select orientation head width arrows. Function residuals returns pointwise residuals, fitted fitted values, either centred zero mean (truemean=FALSE) original scale ( hardly make sense symmetric = TRUE). addition, summary print methods. matrix X lower number columns matrix Y, matrix X filled zero columns match dimensions. means function can used rotate ordination configuration environmental variable ( practically extracting result fitted function). Function predict can used add new rotated coordinates target. predict function always translate coordinates original non-centred matrix. function used newdata symmetric analysis. Function protest performs symmetric Procrustes analysis repeatedly estimate significance Procrustes statistic. Function protest uses correlation-like statistic derived symmetric Procrustes sum squares \\(ss\\) \\(r =\\sqrt{1-ss}\\), also prints sum squares symmetric analysis, sometimes called \\(m_{12}^2\\). Function protest print method, otherwise uses procrustes methods. Thus plot protest object yields Procrustean superimposition plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes returns object class procrustes items. Function protest inherits procrustes, amends new items: Yrot Rotated matrix Y. X Target matrix. ss Sum squared differences X Yrot. rotation Orthogonal rotation matrix. translation Translation origin. scale Scaling factor. xmean centroid target. symmetric Type ss statistic. call Function call. t0 following items class protest: Procrustes correlation non-permuted solution. t Procrustes correlations permutations. distribution correlations can inspected permustats function. signif Significance t permutations Number permutations. control list control values permutations returned function . control list passed argument control describing permutation design.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate Analysis. Academic Press. Peres-Neto, P.R. Jackson, D.. (2001). well multivariate data sets match? advantages Procrustean superimposition approach Mantel test. Oecologia 129: 169-178.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"function protest follows Peres-Neto & Jackson (2001), implementation still Mardia et al. (1979).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"## IGNORE_RDIFF_BEGIN data(varespec) vare.dist <- vegdist(wisconsin(varespec)) mds.null <- monoMDS(vare.dist, y = cmdscale(vare.dist)) mds.alt <- monoMDS(vare.dist) vare.proc <- procrustes(mds.alt, mds.null) vare.proc #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Procrustes sum of squares: #> 11.17 #> summary(vare.proc) #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Number of objects: 24 Number of dimensions: 2 #> #> Procrustes sum of squares: #> 11.17448 #> Procrustes root mean squared error: #> 0.6823512 #> Quantiles of Procrustes errors: #> Min 1Q Median 3Q Max #> 0.1642438 0.2425785 0.2783603 0.4983976 2.4447632 #> #> Rotation matrix: #> [,1] [,2] #> [1,] 0.99937107 -0.03546065 #> [2,] 0.03546065 0.99937107 #> #> Translation of averages: #> [,1] [,2] #> [1,] -1.713781e-17 1.56769e-17 #> #> Scaling of target: #> [1] 0.7310245 #> plot(vare.proc) plot(vare.proc, kind=2) residuals(vare.proc) #> 18 15 24 27 23 19 22 16 #> 0.2734040 0.2032392 0.4708118 0.4420710 0.3547337 0.1642438 0.2515286 0.2611623 #> 28 13 14 20 25 7 5 6 #> 0.7773604 0.3075051 0.2833167 0.1749943 0.2684784 0.5167965 0.9747233 0.2437827 #> 3 4 2 9 12 10 11 21 #> 0.2252071 0.7586954 2.4447632 0.2389659 0.2093104 0.2597721 0.4922646 1.0884414 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":null,"dir":"Reference","previous_headings":"","what":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data log transformed abundances aquatic invertebrate twelve ditches studied eleven times insecticide treatment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos)"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data frame 132 observations log-transformed (log(10*x + 1)) abundances 178 species. twelve sites (ditches, mesocosms), studied repeatedly eleven occasions. treatment levels, treatment times, ditch ID's data frame, data regular, example shows obtain external variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data set obtained experiment outdoor experimental ditches. Twelve mesocosms allocated random treatments; four served controls, remaining eight treated insecticide chlorpyrifos, nominal dose levels 0.1, 0.9, 6, 44 \\(\\mu\\)g/ L two mesocosms . example data set invertebrates. Sampling done 11 times, week -4 pre-treatment week 24 post-treatment, giving total 132 samples (12 mesocosms times 11 sampling dates), see van den Brink & ter Braak (1999) details. data set contains species data, example shows obtain treatment, time ditch ID variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"CANOCO 4 example data, permission Cajo J. F. ter Braak.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos) ditch <- gl(12, 1, length=132) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank -- Abundance or Dominance / Diversity Models — radfit","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions construct rank -- abundance dominance / diversity Whittaker plots fit brokenstick, preemption, log-Normal, Zipf Zipf-Mandelbrot models species abundance.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"# S3 method for default radfit(x, ...) rad.null(x, family=poisson, ...) rad.preempt(x, family = poisson, ...) rad.lognormal(x, family = poisson, ...) rad.zipf(x, family = poisson, ...) rad.zipfbrot(x, family = poisson, ...) # S3 method for radline predict(object, newdata, total, ...) # S3 method for radfit plot(x, BIC = FALSE, legend = TRUE, ...) # S3 method for radfit.frame plot(x, order.by, BIC = FALSE, model, legend = TRUE, as.table = TRUE, ...) # S3 method for radline plot(x, xlab = \"Rank\", ylab = \"Abundance\", type = \"b\", ...) radlattice(x, BIC = FALSE, ...) # S3 method for radfit lines(x, ...) # S3 method for radfit points(x, ...) as.rad(x) # S3 method for rad plot(x, xlab = \"Rank\", ylab = \"Abundance\", log = \"y\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"x Data frame, matrix vector giving species abundances, object plotted. family Error distribution (passed glm). alternatives accepting link = \"log\" family can used, although make sense. object fitted result object. newdata Ranks used ordinations. models can interpolate non-integer “ranks” (although may approximate), extrapolation may fail total new total used predicting abundance. Observed total count used omitted. order.vector used ordering sites plots. BIC Use Bayesian Information Criterion, BIC, instead Akaike's AIC. penalty BIC \\(k = \\log(S)\\) \\(S\\) number species, whereas AIC uses \\(k = 2\\). model Show specified model. missing, AIC used select model. model names (can abbreviated) Null, Preemption, Lognormal, Zipf, Mandelbrot. legend Add legend line colours. .table Arrange panels starting upper left corner (passed xyplot). xlab,ylab Labels x y axes. type Type plot, \"b\" plotting observed points fitted lines, \"p\" points, \"l\" fitted lines, \"n\" setting frame. log Use logarithmic scale given axis. default log = \"y\" gives traditional plot community ecology preemption model straight line, log = \"xy\" Zipf model straight line. log = \"\" axes original arithmetic scale. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Rank--Abundance Dominance (RAD) Dominance/Diversity plots (Whittaker 1965) display logarithmic species abundances species rank order. plots supposed effective analysing types abundance distributions communities. functions fit popular models mainly following Wilson (1991). Functions rad.null, rad.preempt, rad.lognormal, rad.zipf zipfbrot fit individual models (described ) single vector (row data frame), function radfit fits models. argument function radfit can either vector single community data frame row represents distinct community. Function rad.null fits brokenstick model expected abundance species rank \\(r\\) \\(a_r = (J/S) \\sum_{x=r}^S (1/x)\\) (Pielou 1975), \\(J\\) total number individuals (site total) \\(S\\) total number species community. gives Null model individuals randomly distributed among observed species, fitted parameters. Function rad.preempt fits niche preemption model, .k.. geometric series Motomura model, expected abundance \\(\\) species rank \\(r\\) \\(a_r = J \\alpha (1 - \\alpha)^{r-1}\\). estimated parameter preemption coefficient \\(\\alpha\\) gives decay rate abundance per rank. niche preemption model straight line RAD plot. Function rad.lognormal fits log-Normal model assumes logarithmic abundances distributed Normally, \\(a_r = \\exp( \\log \\mu + \\log \\sigma N)\\), \\(N\\) Normal deviate. Function rad.zipf fits Zipf model \\(a_r = J p_1 r^\\gamma\\) \\(p_1\\) fitted proportion abundant species, \\(\\gamma\\) decay coefficient. Zipf--Mandelbrot model (rad.zipfbrot) adds one parameter: \\(a_r = J c (r + \\beta)^\\gamma\\) \\(p_1\\) Zipf model changes meaningless scaling constant \\(c\\). Log-Normal Zipf models generalized linear models (glm) logarithmic link function. Zipf--Mandelbrot adds one nonlinear parameter Zipf model, fitted using nlm nonlinear parameter estimating parameters log-Likelihood glm. Preemption model fitted purely nonlinear model. estimated parameters Null model. default family poisson appropriate genuine counts (integers), families accept link = \"log\" can used. Families Gamma gaussian may appropriate abundance data, cover. best model selected AIC. Therefore ‘quasi’ families quasipoisson used: AIC log-Likelihood needed non-linear models. functions plot functions. radfit applied data frame, plot uses Lattice graphics, plot functions use ordinary graphics. ordinary graphics functions return invisibly ordiplot object observed points, function identify.ordiplot can used label selected species. Alternatively, radlattice uses Lattice graphics display radfit model single site separate panel together AIC BIC values. Function .rad base function construct ordered RAD data. plot used RAD plot functions pass extra arguments (xlab log) function. function returns ordered vector taxa occurring site, corresponding attribute \"index\" included taxa.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions rad.null, rad.preempt, rad.lognormal, zipf zipfbrot fit single RAD model single site. result object class \"radline\" inherits glm, can handled ( ) glm methods. Function radfit fits models either single site rows data frame matrix. fitted single site, function returns object class \"radfit\" items y (observed values), family, models list fitted \"radline\" models. applied data frame matrix, radfit function returns object class \"radfit.frame\" list \"radfit\" objects, item names corresponding row name. result objects (\"radline\", \"radfit\", \"radfit.frame\") can accessed method functions. following methods available: AIC, coef, deviance, logLik. addition fit results can accessed fitted, predict residuals (inheriting residuals.glm). graphical functions discussed Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Pielou, E.C. (1975) Ecological Diversity. Wiley & Sons. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Whittaker, R. H. (1965) Dominance diversity plant communities. Science 147, 250--260. Wilson, J. B. (1991) Methods fitting dominance/diversity curves. Journal Vegetation Science 2, 35--46.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"RAD models usually fitted proportions instead original abundances. However, nothing models seems require division abundances site totals, original observations used functions. wish use proportions, must standardize data site totals, e.g. decostand use appropriate family Gamma. lognormal model fitted standard way, think quite correct -- least equivalent fitting Normal density log abundances like originally suggested (Preston 1948). models may fail. particular, estimation Zipf-Mandelbrot model difficult. fitting fails, NA returned. Wilson (1991) defined preemption model \\(a_r = J p_1 (1 - \\alpha)^{r-1}\\), \\(p_1\\) fitted proportion first species. However, parameter \\(p_1\\) completely defined \\(\\alpha\\) since fitted proportions must add one, therefore handle preemption one-parameter model. Veiled log-Normal model included earlier releases function, removed flawed: implicit veil line also appears ordinary log-Normal. latest release version rad.veil 1.6-10.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"data(BCI) mod <- rad.lognormal(BCI[5,]) mod #> #> RAD model: Log-Normal #> Family: poisson #> No. of species: 101 #> Total abundance: 505 #> #> log.mu log.sigma Deviance AIC BIC #> 0.951926 1.165929 17.077549 317.656487 322.886728 plot(mod) mod <- radfit(BCI[1,]) ## Standard plot overlaid for all models ## Preemption model is a line plot(mod) ## log for both axes: Zipf model is a line plot(mod, log = \"xy\") ## Lattice graphics separately for each model radlattice(mod) # Take a subset of BCI to save time and nerves mod <- radfit(BCI[3:5,]) mod #> #> Deviance for RAD models: #> #> 3 4 5 #> Null 86.1127 49.8111 80.855 #> Preemption 58.9295 39.7817 76.311 #> Lognormal 29.2719 16.6588 17.078 #> Zipf 50.1262 47.9108 30.936 #> Mandelbrot 5.7342 5.5665 10.573 plot(mod, pch=\".\")"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":null,"dir":"Reference","previous_headings":"","what":"Compares Dissimilarity Indices for Gradient Detection — rankindex","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Rank correlations dissimilarity indices gradient separation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"rankindex(grad, veg, indices = c(\"euc\", \"man\", \"gow\", \"bra\", \"kul\"), stepacross = FALSE, method = \"spearman\", metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"grad gradient variable matrix. veg community data matrix. indices Dissimilarity indices compared, partial matches alternatives vegdist. Alternatively, can (named) list functions returning objects class 'dist'. stepacross Use stepacross find shorter path dissimilarity. dissimilarities site pairs shared species set NA using .shared indices fixed upper limit can also analysed. method Correlation method used. metric Metric evaluate gradient separation. See Details. ... parameters stepacross.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"good dissimilarity index multidimensional scaling high rank-order similarity gradient separation. function compares indices vegdist gradient separation using rank correlation coefficients cor. gradient separation point assessed using given metric. default use Euclidean distance continuous variables scaled unit variance, use Gower metric mixed data using function daisy grad factors. alternatives Mahalanabis distances based grad matrix scaled columns orthogonal (uncorrelated) unit variance, Manhattan distances grad variables scaled unit range. indices argument can accept dissimilarity indices besides ones calculated vegdist function. , argument value (possibly named) list functions. function must return valid 'dist' object dissimilarities, similarities accepted converted dissimilarities beforehand.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Returns named vector rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Faith, F.P., Minchin, P.R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Jari Oksanen, additions Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"several problems using rank correlation coefficients. Typically many ties \\(n(n-1)/2\\) gradient separation values derived just \\(n\\) observations. Due floating point arithmetics, many tied values differ machine epsilon arbitrarily ranked differently rank used cor.test. Two indices identical certain transformation standardization may differ slightly (magnitude \\(10^{-15}\\)) may lead third fourth decimal instability rank correlations. Small differences rank correlations taken seriously. Probably method replaced sounder method, yet know ... may experiment mantel, anosim even protest. Earlier version function used method = \"kendall\", far slow large data sets. functions returning dissimilarity objects self contained, ... argument passes additional parameters stepacross functions supplied via indices argument.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"data(varespec) data(varechem) ## The variables are automatically scaled rankindex(varechem, varespec) #> euc man gow bra kul #> 0.2396330 0.2735087 0.2288358 0.2837910 0.2839834 rankindex(varechem, wisconsin(varespec)) #> euc man gow bra kul #> 0.4200990 0.4215642 0.3708606 0.4215642 0.4215642 ## Using non vegdist indices as functions funs <- list(Manhattan=function(x) dist(x, \"manhattan\"), Gower=function(x) cluster:::daisy(x, \"gower\"), Ochiai=function(x) designdist(x, \"1-J/sqrt(A*B)\")) rankindex(scale(varechem), varespec, funs) #> Manhattan Gower Ochiai #> 0.2735087 0.2288358 0.1696862"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":null,"dir":"Reference","previous_headings":"","what":"Rarefaction Species Richness — rarefy","title":"Rarefaction Species Richness — rarefy","text":"Rarefied species richness community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rarefaction Species Richness — rarefy","text":"","code":"rarefy(x, sample, se = FALSE, MARGIN = 1) rrarefy(x, sample) drarefy(x, sample) rarecurve(x, step = 1, sample, xlab = \"Sample Size\", ylab = \"Species\", label = TRUE, col, lty, tidy = FALSE, ...) rareslope(x, sample)"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rarefaction Species Richness — rarefy","text":"x Community data, matrix-like object vector. MARGIN Margin index computed. sample Subsample size rarefying community, either single value vector. se Estimate standard errors. step Step size sample sizes rarefaction curves. xlab, ylab Axis labels plots rarefaction curves. label Label rarefaction curves rownames x (logical). col, lty plotting colour line type, see par. Can vector length nrow(x), one per sample, extended length internally. tidy Instead drawing plot, return “tidy” data frame can used ggplot2 graphics. data frame variables Site (factor), Sample Species. ... Parameters passed nlm, plot, lines ordilabel rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rarefaction Species Richness — rarefy","text":"Function rarefy gives expected species richness random subsamples size sample community. size sample smaller total community size, function work larger sample well (warning) return non-rarefied species richness (standard error = 0). sample vector, rarefaction observations performed sample size separately. Rarefaction can performed genuine counts individuals. function rarefy based Hurlbert's (1971) formulation, standard errors Heck et al. (1975). Function rrarefy generates one randomly rarefied community data frame vector given sample size. sample can vector giving sample sizes row. sample size equal larger observed number individuals, non-rarefied community returned. random rarefaction made without replacement variance rarefied communities rather related rarefaction proportion size sample. Random rarefaction sometimes used remove effects different sample sizes. usually bad idea: random rarefaction discards valid data, introduces random error reduces quality data (McMurdie & Holmes 2014). better use normalizing transformations (decostand vegan) possible variance stabilization (decostand dispweight vegan) methods sensitive sample sizes. Function drarefy returns probabilities species occur rarefied community size sample. sample can vector giving sample sizes row. sample equal larger observed number individuals, observed species sampling probability 1. Function rarecurve draws rarefaction curve row input data. rarefaction curves evaluated using interval step sample sizes, always including 1 total sample size. sample specified, vertical line drawn sample horizontal lines rarefied species richnesses. Function rareslope calculates slope rarecurve (derivative rarefy) given sample size; sample need integer. Rarefaction functions used observed counts. think necessary use multiplier data, rarefy first multiply. Removing rare species rarefaction can also give biased results. Observed count data normally include singletons (species count 1), missing, functions issue warnings. may false positives, recommended check observed counts multiplied rare taxa removed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rarefaction Species Richness — rarefy","text":"vector rarefied species richness values. single sample se = TRUE, function rarefy returns 2-row matrix rarefied richness (S) standard error (se). sample vector rarefy, function returns matrix column sample size, se = TRUE, rarefied richness standard error consecutive lines. Function rarecurve returns invisible list rarefy results corresponding drawn curve. Alternatively, tidy = TRUE returns data frame can used ggplot2 graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rarefaction Species Richness — rarefy","text":"Heck, K.L., van Belle, G. & Simberloff, D. (1975). Explicit calculation rarefaction diversity measurement determination sufficient sample size. Ecology 56, 1459--1461. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. McMurdie, P.J. & Holmes, S. (2014). Waste , want : rarefying microbiome data inadmissible. PLoS Comput Biol 10(4): e1003531. doi:10.1371/journal.pcbi.1003531","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rarefaction Species Richness — rarefy","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rarefaction Species Richness — rarefy","text":"","code":"data(BCI) S <- specnumber(BCI) # observed number of species (raremax <- min(rowSums(BCI))) #> [1] 340 Srare <- rarefy(BCI, raremax) plot(S, Srare, xlab = \"Observed No. of Species\", ylab = \"Rarefied No. of Species\") abline(0, 1) rarecurve(BCI, step = 20, sample = raremax, col = \"blue\", cex = 0.6)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":null,"dir":"Reference","previous_headings":"","what":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Function finds Raup-Crick dissimilarity probability number co-occurring species species occurrence probabilities proportional species frequencies.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"raupcrick(comm, null = \"r1\", nsimul = 999, chase = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"comm Community data treated presence/absence data. null Null model used method oecosimu. nsimul Number null communities assessing dissimilarity index. chase Use Chase et al. (2011) method tie handling ( recommended except comparing results Chase script). ... parameters passed oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Raup-Crick index probability compared sampling units non-identical species composition. probability can regarded dissimilarity, although metric: identical sampling units can dissimilarity slightly \\(0\\), dissimilarity can nearly zero range shared species, sampling units shared species can dissimilarity slightly \\(1\\). Moreover, communities sharing rare species appear similar (lower probability finding rare species together), communities sharing number common species. function always treat data binary (presence/ absence). probability assessed using simulation oecosimu test statistic observed number shared species sampling units evaluated community null model (see Examples). default null model \"r1\" probability selecting species proportional species frequencies. vegdist function implements variant Raup-Crick index equal sampling probabilities species using exact analytic equations without simulation. corresponds null model \"r0\" also can used current function. null model methods oecosimu can used current function, new unpublished methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function returns object inheriting dist can interpreted dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function developed Brian Inouye contacted us informed us method Chase et al. (2011), function takes idea code published paper. current function written Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"test statistic number shared species, typically tied large number simulation results. tied values handled differently current function function published Chase et al. (2011). vegan, index number simulated values smaller equal observed value, smaller observed value used Chase et al. (2011) option split = FALSE script; can achieved chase = TRUE vegan. Chase et al. (2011) script split = TRUE uses half tied simulation values calculate distance measure, choice directly reproduced vegan ( average vegan raupcrick results chase = TRUE chase = FALSE).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"## data set with variable species richness data(sipoo) ## default raupcrick dr1 <- raupcrick(sipoo) ## use null model \"r0\" of oecosimu dr0 <- raupcrick(sipoo, null = \"r0\") ## vegdist(..., method = \"raup\") corresponds to 'null = \"r0\"' d <- vegdist(sipoo, \"raup\") op <- par(mfrow=c(2,1), mar=c(4,4,1,1)+.1) plot(dr1 ~ d, xlab = \"Raup-Crick with Null R1\", ylab=\"vegdist\") plot(dr0 ~ d, xlab = \"Raup-Crick with Null R0\", ylab=\"vegdist\") par(op) ## The calculation is essentially as in the following oecosimu() call, ## except that designdist() is replaced with faster code if (FALSE) oecosimu(sipoo, function(x) designdist(x, \"J\", \"binary\"), method = \"r1\")"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":null,"dir":"Reference","previous_headings":"","what":"Reads a CEP (Canoco) data file — read.cep","title":"Reads a CEP (Canoco) data file — read.cep","text":"read.cep reads file formatted relaxed strict CEP format used Canoco software, among others.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"read.cep(file, positive=TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reads a CEP (Canoco) data file — read.cep","text":"file File name (character variable). positive positive entries, like community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reads a CEP (Canoco) data file — read.cep","text":"Cornell Ecology Programs (CEP) introduced several data formats designed punched cards. One ‘condensed strict’ format adopted popular software DECORANA TWINSPAN. relaxed variant format later adopted Canoco software (ter Braak 1984). Function read.cep reads legacy files written format. condensed CEP CANOCO formats : Two three title cards, importantly specifying format number items per record. Data condensed format: First number line site identifier (integer), followed pairs (‘couplets’) numbers identifying species abundance (integer floating point number). Species site names, given Fortran format (10A8): Ten names per line, eight columns . option positive = TRUE function removes rows columns zero negative marginal sums. community data positive entries, removes empty sites species. data entries can negative, ruins data, data sets read option positive = FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reads a CEP (Canoco) data file — read.cep","text":"Returns data frame, columns species rows sites. Column row names taken CEP file, changed unique R names make.names stripping blanks.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Reads a CEP (Canoco) data file — read.cep","text":"ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reads a CEP (Canoco) data file — read.cep","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reads a CEP (Canoco) data file — read.cep","text":"Function read.cep used Fortran read data vegan 2.4-5 earlier, Fortran /O longer allowed CRAN packages, function re-written R. original Fortran code robust, several legacy data sets may fail current version, read previous Fortran version. CRAN package cepreader makes available original Fortran-based code run separate subprocess. cepreader package can also read ‘free’ ‘open’ Canoco formats handled function. function based read.fortran. REAL format defines decimal part species abundances (F5.1), read.fortran divides input corresponding power 10 even input data explicit decimal separator. F5.1, 100 become 10, 0.1 become 0.01. Function read.cep tries undo division, check scaling results reading data, necessary, multiply results original scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"## Provided that you have the file \"dune.spe\" if (FALSE) { theclassic <- read.cep(\"dune.spe\")}"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":null,"dir":"Reference","previous_headings":"","what":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi find Rényi diversities scale corresponding Hill number (Hill 1973). Function renyiaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"renyi(x, scales = c(0, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64, Inf), hill = FALSE) # S3 method for renyi plot(x, ...) renyiaccum(x, scales = c(0, 0.5, 1, 2, 4, Inf), permutations = 100, raw = FALSE, collector = FALSE, subset, ...) # S3 method for renyiaccum plot(x, what = c(\"Collector\", \"mean\", \"Qnt 0.025\", \"Qnt 0.975\"), type = \"l\", ...) # S3 method for renyiaccum persp(x, theta = 220, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"x Community data matrix plotting object. scales Scales Rényi diversity. hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. collector Accumulate diversities order sites data set, collector curve can plotted summary permutations. argument ignored raw = TRUE. subset logical expression indicating sites (rows) keep: missing values taken FALSE. Items plotted. type Type plot, type = \"l\" means lines. theta Angle defining viewing direction (azimuthal) persp. col Colours used surface. Single colour passed , vector colours selected midpoint rectangle persp. zlim Limits vertical axis. ... arguments passed renyi graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Common diversity indices special cases Rényi diversity $$H_a = \\frac{1}{1-} \\log \\sum p_i^$$ \\(\\) scale parameter, Hill (1975) suggested use -called ‘Hill numbers’ defined \\(N_a = \\exp(H_a)\\). Hill numbers number species \\(= 0\\), \\(\\exp(H')\\) exponent Shannon diversity \\(= 1\\), inverse Simpson \\(= 2\\) \\(1/ \\max(p_i)\\) \\(= \\infty\\). According theory diversity ordering, one community can regarded diverse another Rényi diversities higher (Tóthmérész 1995). plot method renyi uses lattice graphics, displays diversity values scale separate panel site together minimum, maximum median values complete data. Function renyiaccum similar specaccum finds Rényi Hill diversities given scales random permutations accumulated sites. plot function uses lattice function xyplot display accumulation curves value scales separate panel. addition, persp method plot diversity surface scale number sites. Similar dynamic graphics can made rgl.renyiaccum vegan3d package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi returns data frame selected indices. Function renyiaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Hill, M.O. (1973). Diversity evenness: unifying notation consequences. Ecology 54, 427--473. Kindt, R., Van Damme, P., Simons, .J. (2006). Tree diversity western Kenya: using profiles characterise richness evenness. Biodiversity Conservation 15, 1253--1270. Tóthmérész, B. (1995). Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Roeland Kindt Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) mod <- renyi(BCI[i,]) plot(mod) mod <- renyiaccum(BCI[i,]) plot(mod, as.table=TRUE, col = c(1, 2, 2)) persp(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":null,"dir":"Reference","previous_headings":"","what":"Reorder a Hierarchical Clustering Tree — reorder.hclust","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Function takes hierarchical clustering tree hclust vector values reorders clustering tree order supplied vector, maintaining constraints tree. method generic function reorder alternative reordering \"dendrogram\" object reorder.dendrogram","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"# S3 method for hclust reorder(x, wts, agglo.FUN = c(\"mean\", \"min\", \"max\", \"sum\", \"uwmean\"), ...) # S3 method for hclust rev(x) # S3 method for hclust scores(x, display = \"internal\", ...) cutreeord(tree, k = NULL, h = NULL)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"x, tree hierarchical clustering hclust. wts numeric vector reordering. agglo.FUN function weights agglomeration, see . display return \"internal\" nodes \"terminal\" nodes (also called \"leaves\"). k, h scalars vectors giving numbers desired groups heights tree cut (passed function cutree). ... additional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Dendrograms can ordered many ways. reorder function reorders hclust tree provides alternative reorder.dendrogram can reorder dendrogram. current function also work differently agglo.FUN \"mean\": reorder.dendrogram always take direct mean member groups ignoring sizes, function used weighted.mean weighted group sizes, group mean always mean member leaves (terminal nodes). want ignore group sizes, can use unweighted mean \"uwmean\". function accepts limited list agglo.FUN functions assessing value wts groups. ordering always ascending, order leaves can reversed rev. Function scores finds coordinates nodes two-column matrix. terminal nodes (leaves) value item merged tree, labels can still hang level (see plot.hclust). Function cutreeord cuts tree groups numbered left right tree. based standard function cutree numbers groups order appear input data instead order tree.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Reordered hclust result object added item value gives value statistic merge level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"functions really base R.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"## reorder by water content of soil data(mite, mite.env) hc <- hclust(vegdist(wisconsin(sqrt(mite)))) ohc <- with(mite.env, reorder(hc, WatrCont)) plot(hc) plot(ohc) ## label leaves by the observed value, and each branching point ## (internal node) by the cluster mean with(mite.env, plot(ohc, labels=round(WatrCont), cex=0.7)) ordilabel(scores(ohc), label=round(ohc$value), cex=0.7) ## Slightly different from reordered 'dendrogram' which ignores group ## sizes in assessing means. den <- as.dendrogram(hc) den <- with(mite.env, reorder(den, WatrCont, agglo.FUN = mean)) plot(den)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Species or Site Scores from an Ordination — scores","title":"Get Species or Site Scores from an Ordination — scores","text":"Function access either species site scores specified axes ordination methods. scores function generic vegan, vegan ordination functions scores functions documented separately method (see e.g. scores.cca, scores.metaMDS, scores.decorana). help file documents default scores method used non-vegan ordination objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"# S3 method for default scores(x, choices, display=c(\"sites\", \"species\", \"both\"), tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Species or Site Scores from an Ordination — scores","text":"x ordination result. choices Ordination axes. missing, default method returns axes. display Partial match access scores \"sites\" \"species\" \"\". tidy Return \"\" scores data frame compatible ggplot2, variable score labelling scores \"sites\" \"species\". ... parameters (unused).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Species or Site Scores from an Ordination — scores","text":"Function scores generic method vegan. Several vegan functions scores methods defaults new arguments. help page describes default method. methods, see, e.g., scores.cca, scores.rda, scores.decorana. vegan ordination functions scores method used extract scores instead directly accessing . Scaling transformation scores also happen scores function. scores function available, results can plotted using ordiplot, ordixyplot etc., ordination results can compared procrustes analysis. scores.default function used extract scores non-vegan ordination results. Many standard ordination methods libraries specific class, specific method can written . However, scores.default guesses commonly used functions keep site scores possible species scores. x matrix, scores.default returns chosen columns matrix, ignoring whether species sites requested (regard bug feature, please). Currently function seems work least isoMDS, prcomp, princomp ade4 objects. may work cases fail mysteriously.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Species or Site Scores from an Ordination — scores","text":"function returns matrix scores one type requested, named list matrices display = \"\", ggplot2 compatible data frame tidy = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get Species or Site Scores from an Ordination — scores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"data(varespec) vare.pca <- prcomp(varespec) scores(vare.pca, choices=c(1,2)) #> PC1 PC2 #> 18 -10.7847878 18.7094315 #> 15 -27.8036826 -11.7414745 #> 24 -25.6919559 -14.5399684 #> 27 -31.7820166 -31.2216800 #> 23 -19.6315869 -2.5541193 #> 19 -0.2413294 -11.4974077 #> 22 -26.6771373 -12.3140897 #> 16 -21.9230366 0.4449159 #> 28 -39.6083051 -41.8877392 #> 13 -4.0664328 20.4191153 #> 14 -18.4416245 5.4406988 #> 20 -17.3999191 2.3653380 #> 25 -25.1673547 -13.2508067 #> 7 -11.4065430 41.7356300 #> 5 -8.4243752 45.3805255 #> 6 -2.0759474 36.9311222 #> 3 39.8617580 8.0590041 #> 4 13.1065901 12.8377217 #> 2 57.6827011 -4.8983565 #> 9 63.3138332 -22.4481549 #> 12 44.1073111 -10.1653935 #> 10 64.9418975 -16.7633564 #> 11 11.5313633 3.9720890 #> 21 -3.4194194 -3.0130455"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Screeplot methods plotting variances ordination axes/components overlaying broken stick distributions. Also, provides alternative screeplot methods princomp prcomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"# S3 method for cca screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, if (is.null(x$CCA) || x$CCA$rank == 0) x$CA$rank else x$CCA$rank), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for decorana screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = 4, ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for prcomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for princomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) bstick(n, ...) # S3 method for default bstick(n, tot.var = 1, ...) # S3 method for cca bstick(n, ...) # S3 method for prcomp bstick(n, ...) # S3 method for princomp bstick(n, ...) # S3 method for decorana bstick(n, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"x object component variances can determined. bstick logical; broken stick distribution drawn? npcs number components plotted. type type plot. ptype type == \"lines\" bstick = TRUE, character indicating type plotting used lines; actually types plot.default. bst.col, bst.lty colour line type used draw broken stick distribution. xlab, ylab, main graphics parameters. legend logical; draw legend? n object variances can extracted number variances (components) case bstick.default. tot.var total variance split. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"functions provide screeplots ordination methods vegan enhanced versions broken stick prcomp princomp. Function bstick gives brokenstick values ordered random proportions, defined \\(p_i = (tot/n) \\sum_{x=}^n (1/x)\\) (Legendre & Legendre 2012), \\(tot\\) total \\(n\\) number brokenstick components (cf. radfit). Broken stick recommended stopping rule principal component analysis (Jackson 1993): principal components retained long observed eigenvalues higher corresponding random broken stick components. bstick function generic. default needs number components total, specific methods extract information ordination results. also bstick method cca. However, broken stick model strictly valid correspondence analysis (CA), eigenvalues CA defined \\(\\leq 1\\), whereas brokenstick components restrictions. brokenstick components detrended correspondence analysis (DCA) assume input data full rank, additive eigenvalues used screeplot (see decorana).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Function screeplot draws plot currently active device, returns invisibly xy.coords points bars eigenvalues. Function bstick returns numeric vector broken stick components.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Jackson, D. . (1993). Stopping rules principal components analysis: comparison heuristical statistical approaches. Ecology 74, 2204--2214. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"data(varespec) vare.pca <- rda(varespec, scale = TRUE) bstick(vare.pca) #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 7.1438620 5.2308185 4.2742968 3.6366156 3.1583548 2.7757461 2.4569055 2.1836136 #> PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 #> 1.9444831 1.7319228 1.5406184 1.3667054 1.2072851 1.0601279 0.9234819 0.7959457 #> PC17 PC18 PC19 PC20 PC21 PC22 PC23 #> 0.6763805 0.5638485 0.4575683 0.3568818 0.2612296 0.1701323 0.0831758 screeplot(vare.pca, bstick = TRUE, type = \"lines\")"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":null,"dir":"Reference","previous_headings":"","what":"Similarity Percentages — simper","title":"Similarity Percentages — simper","text":"Discriminating species two groups using Bray-Curtis dissimilarities","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Similarity Percentages — simper","text":"","code":"simper(comm, group, permutations = 999, parallel = 1, ...) # S3 method for simper summary(object, ordered = TRUE, digits = max(3,getOption(\"digits\") - 3), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Similarity Percentages — simper","text":"comm Community data. group Factor describing group structure. missing one level, contributions estimated non-grouped data dissimilarities show overall heterogeneity species abundances. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. object object returned simper. ordered Logical; species ordered average contribution? digits Number digits output. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. (yet implemented). ... Parameters passed functions. simper extra parameters passed shuffleSet permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Similarity Percentages — simper","text":"Similarity percentage, simper (Clarke 1993) based decomposition Bray-Curtis dissimilarity index (see vegdist, designdist). contribution individual species \\(\\) overall Bray-Curtis dissimilarity \\(d_{jk}\\) given $$d_{ijk} = \\frac{|x_{ij}-x_{ik}|}{\\sum_{=1}^S (x_{ij}+x_{ik})}$$ \\(x\\) abundance species \\(\\) sampling units \\(j\\) \\(k\\). overall index sum individual contributions \\(S\\) species \\(d_{jk}=\\sum_{=1}^S d_{ijk}\\). simper functions performs pairwise comparisons groups sampling units finds contribution species average -group Bray-Curtis dissimilarity. Although method called “Similarity Percentages”, really studied dissimilarities instead similarities (Clarke 1993). function displays important species pair groups. species contribute least 70 % differences groups. function returns much extensive results (including species) can accessed directly result object (see section Value). Function summary transforms result list data frames. argument ordered = TRUE data frames also include cumulative contributions ordered species contribution. results simper can difficult interpret often misunderstood even publications. method gives contribution species overall dissimilarities, caused variation species abundances, partly differences among groups. Even make groups copies , method single species high contribution, contributions non-existing -group differences random noise variation species abundances. abundant species usually highest variances, high contributions even differ among groups. Permutation tests study differences among groups, can used find species differences among groups important component contribution dissimilarities. Analysis without group argument find species contributions average overall dissimilarity among sampling units. non-grouped contributions can compared grouped contributions see much added value grouping species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Similarity Percentages — simper","text":"list class \"simper\" following items: species species names. average Species contribution average -group dissimilarity. overall average -group dissimilarity. sum item average. sd Standard deviation contribution. ratio Average sd ratio. ava, avb Average abundances per group. ord index vector order vectors contribution order cusum back original data order. cusum Ordered cumulative contribution. based item average, sum total 1. p Permutation \\(p\\)-value. Probability getting larger equal average contribution random permutation group factor. area available permutations used (default: calculated).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Similarity Percentages — simper","text":"Eduard Szöcs Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Similarity Percentages — simper","text":"Clarke, K.R. 1993. Non-parametric multivariate analyses changes community structure. Australian Journal Ecology, 18, 117–143.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Similarity Percentages — simper","text":"","code":"data(dune) data(dune.env) (sim <- with(dune.env, simper(dune, Management, permutations = 99))) #> cumulative contributions of most influential species: #> #> $SF_BF #> Agrostol Alopgeni Lolipere Trifrepe Poatriv Scorautu Bromhord #> 0.09824271 0.18254830 0.25956958 0.33367870 0.40734444 0.47729205 0.53120026 #> Achimill Planlanc Elymrepe Bracruta #> 0.57946526 0.62522255 0.67016196 0.71098133 #> #> $SF_HF #> Agrostol Alopgeni Lolipere Planlanc Rumeacet Elymrepe Poatriv #> 0.08350879 0.16534834 0.23934930 0.30843624 0.37716139 0.43334492 0.48351753 #> Bracruta Eleopalu Poaprat Anthodor Sagiproc Trifprat #> 0.52804045 0.57205850 0.61423981 0.65549838 0.69628951 0.73696831 #> #> $SF_NM #> Poatriv Alopgeni Agrostol Lolipere Eleopalu Poaprat Bracruta Elymrepe #> 0.1013601 0.1935731 0.2667383 0.3377578 0.3999419 0.4526707 0.5044725 0.5505643 #> Scorautu Trifrepe Sagiproc Salirepe #> 0.5926117 0.6320111 0.6712478 0.7091528 #> #> $BF_HF #> Rumeacet Poatriv Planlanc Bromhord Lolipere Elymrepe Trifrepe #> 0.08163219 0.15193797 0.21918333 0.27967181 0.33969561 0.39843338 0.45298204 #> Anthodor Achimill Bracruta Alopgeni Trifprat Juncarti #> 0.50276849 0.55222648 0.60021994 0.64584333 0.69126471 0.73366621 #> #> $BF_NM #> Lolipere Poatriv Poaprat Trifrepe Bromhord Bracruta Eleopalu Agrostol #> 0.1242718 0.1992126 0.2711756 0.3414609 0.3958520 0.4448077 0.4910724 0.5369083 #> Achimill Scorautu Anthodor Planlanc #> 0.5823926 0.6253645 0.6638182 0.7012577 #> #> $HF_NM #> Poatriv Lolipere Rumeacet Poaprat Planlanc Bracruta Eleopalu #> 0.09913221 0.17468460 0.23917190 0.29701331 0.35469313 0.40365488 0.44804851 #> Agrostol Trifrepe Elymrepe Anthodor Juncarti Trifprat Salirepe #> 0.49226546 0.53434466 0.57564661 0.61543243 0.65341300 0.68921695 0.72432408 #> ## IGNORE_RDIFF_BEGIN summary(sim) #> #> Contrast: SF_BF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.06137 0.03419 1.79490 4.66700 0.00000 0.098 0.05 * #> Alopgeni 0.05267 0.03648 1.44390 4.33300 0.66700 0.182 0.14 #> Lolipere 0.04812 0.03945 1.21980 3.00000 6.00000 0.260 0.39 #> Trifrepe 0.04630 0.02553 1.81380 1.33300 4.66700 0.334 0.09 . #> Poatriv 0.04602 0.03380 1.36150 4.66700 3.66700 0.407 0.46 #> Scorautu 0.04370 0.02492 1.75340 1.33300 4.33300 0.477 0.04 * #> Bromhord 0.03368 0.02586 1.30230 0.50000 2.66700 0.531 0.02 * #> Achimill 0.03015 0.02082 1.44820 0.16700 2.33300 0.580 0.04 * #> Planlanc 0.02859 0.02155 1.32650 0.00000 2.00000 0.625 0.49 #> Elymrepe 0.02807 0.02978 0.94280 2.00000 1.33300 0.670 0.52 #> Bracruta 0.02550 0.02390 1.06690 2.00000 2.00000 0.711 0.83 #> Poaprat 0.02513 0.02397 1.04850 2.50000 4.00000 0.751 0.82 #> Sagiproc 0.02433 0.02215 1.09830 1.83300 0.66700 0.790 0.39 #> Bellpere 0.01986 0.01709 1.16220 0.66700 1.66700 0.822 0.10 . #> Eleopalu 0.01861 0.04296 0.43330 1.33300 0.00000 0.852 0.82 #> Anthodor 0.01754 0.02580 0.67980 0.00000 1.33300 0.880 0.75 #> Juncbufo 0.01603 0.02371 0.67620 1.16700 0.00000 0.905 0.57 #> Vicilath 0.01467 0.01331 1.10260 0.00000 1.00000 0.929 0.04 * #> Hyporadi 0.01029 0.01520 0.67680 0.00000 0.66700 0.945 0.62 #> Ranuflam 0.00931 0.01360 0.68450 0.66700 0.00000 0.960 0.93 #> Juncarti 0.00698 0.01611 0.43330 0.50000 0.00000 0.972 0.95 #> Callcusp 0.00698 0.01611 0.43330 0.50000 0.00000 0.983 0.79 #> Rumeacet 0.00453 0.01044 0.43330 0.33300 0.00000 0.990 0.95 #> Cirsarve 0.00398 0.00918 0.43360 0.33300 0.00000 0.996 0.37 #> Chenalbu 0.00233 0.00537 0.43330 0.16700 0.00000 1.000 0.41 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_HF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.04738 0.03127 1.51510 4.66700 1.40000 0.084 0.35 #> Alopgeni 0.04643 0.03290 1.41150 4.33300 1.60000 0.165 0.20 #> Lolipere 0.04199 0.02701 1.55460 3.00000 4.00000 0.239 0.75 #> Planlanc 0.03920 0.03321 1.18040 0.00000 3.00000 0.308 0.02 * #> Rumeacet 0.03899 0.02737 1.42470 0.33300 3.20000 0.377 0.01 ** #> Elymrepe 0.03188 0.02955 1.07870 2.00000 2.00000 0.433 0.30 #> Poatriv 0.02847 0.02152 1.32270 4.66700 4.80000 0.484 1.00 #> Bracruta 0.02526 0.02104 1.20040 2.00000 2.80000 0.528 0.92 #> Eleopalu 0.02497 0.03888 0.64240 1.33300 0.80000 0.572 0.77 #> Poaprat 0.02393 0.01918 1.24780 2.50000 3.40000 0.614 0.98 #> Anthodor 0.02341 0.02143 1.09230 0.00000 1.80000 0.655 0.67 #> Sagiproc 0.02314 0.02048 1.13010 1.83300 0.80000 0.696 0.40 #> Trifprat 0.02308 0.02343 0.98500 0.00000 1.80000 0.737 0.01 ** #> Juncarti 0.02285 0.02568 0.88990 0.50000 1.60000 0.777 0.51 #> Trifrepe 0.02238 0.01949 1.14860 1.33300 2.80000 0.817 0.94 #> Juncbufo 0.02164 0.02224 0.97330 1.16700 1.20000 0.855 0.24 #> Scorautu 0.02051 0.01642 1.24890 1.33300 2.80000 0.891 0.79 #> Achimill 0.01518 0.01139 1.33260 0.16700 1.20000 0.918 0.75 #> Bromhord 0.01338 0.01450 0.92220 0.50000 0.80000 0.941 0.79 #> Ranuflam 0.01066 0.01339 0.79640 0.66700 0.40000 0.960 0.86 #> Bellpere 0.00999 0.01257 0.79480 0.66700 0.40000 0.978 0.84 #> Callcusp 0.00662 0.01508 0.43930 0.50000 0.00000 0.989 0.92 #> Cirsarve 0.00381 0.00867 0.43940 0.33300 0.00000 0.996 0.54 #> Chenalbu 0.00221 0.00503 0.43930 0.16700 0.00000 1.000 0.51 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Hyporadi 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Vicilath 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07828 0.04095 1.91180 4.66700 0.00000 0.101 0.01 ** #> Alopgeni 0.07122 0.04696 1.51670 4.33300 0.00000 0.194 0.01 ** #> Agrostol 0.05651 0.04418 1.27920 4.66700 2.16700 0.267 0.05 * #> Lolipere 0.05485 0.05991 0.91550 3.00000 0.33300 0.338 0.15 #> Eleopalu 0.04803 0.04717 1.01820 1.33300 2.16700 0.400 0.04 * #> Poaprat 0.04072 0.03179 1.28100 2.50000 0.66700 0.453 0.06 . #> Bracruta 0.04001 0.03440 1.16310 2.00000 2.83300 0.504 0.08 . #> Elymrepe 0.03560 0.03852 0.92430 2.00000 0.00000 0.551 0.12 #> Scorautu 0.03247 0.03481 0.93280 1.33300 3.16700 0.593 0.13 #> Trifrepe 0.03043 0.03163 0.96190 1.33300 1.83300 0.632 0.59 #> Sagiproc 0.03030 0.03048 0.99430 1.83300 0.50000 0.671 0.02 * #> Salirepe 0.02928 0.03201 0.91440 0.00000 1.83300 0.709 0.02 * #> Anthodor 0.02454 0.03669 0.66880 0.00000 1.33300 0.741 0.56 #> Callcusp 0.02276 0.02944 0.77310 0.50000 1.16700 0.770 0.08 . #> Ranuflam 0.02257 0.02282 0.98890 0.66700 1.33300 0.800 0.08 . #> Juncarti 0.02254 0.02860 0.78830 0.50000 1.16700 0.829 0.53 #> Hyporadi 0.02011 0.03129 0.64260 0.00000 1.16700 0.855 0.21 #> Juncbufo 0.01986 0.02903 0.68400 1.16700 0.00000 0.881 0.21 #> Planlanc 0.01542 0.02277 0.67720 0.00000 0.83300 0.900 0.98 #> Airaprae 0.01488 0.02188 0.68020 0.00000 0.83300 0.920 0.06 . #> Bellpere 0.01232 0.01592 0.77370 0.66700 0.33300 0.936 0.72 #> Comapalu 0.01188 0.01741 0.68260 0.00000 0.66700 0.951 0.05 * #> Achimill 0.00929 0.01493 0.62240 0.16700 0.33300 0.963 0.98 #> Bromhord 0.00717 0.01633 0.43910 0.50000 0.00000 0.972 0.98 #> Rumeacet 0.00559 0.01275 0.43840 0.33300 0.00000 0.980 0.98 #> Empenigr 0.00523 0.01200 0.43540 0.00000 0.33300 0.986 0.29 #> Cirsarve 0.00478 0.01089 0.43910 0.33300 0.00000 0.993 0.02 * #> Chenalbu 0.00289 0.00660 0.43820 0.16700 0.00000 0.996 0.02 * #> Vicilath 0.00279 0.00642 0.43450 0.00000 0.16700 1.000 0.81 #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_HF #> #> average sd ratio ava avb cumsum p #> Rumeacet 0.03867 0.02606 1.48380 0.00000 3.20000 0.082 0.03 * #> Poatriv 0.03330 0.02579 1.29110 3.66700 4.80000 0.152 0.96 #> Planlanc 0.03185 0.01830 1.74010 2.00000 3.00000 0.219 0.34 #> Bromhord 0.02865 0.01799 1.59260 2.66700 0.80000 0.280 0.07 . #> Lolipere 0.02843 0.02215 1.28340 6.00000 4.00000 0.340 1.00 #> Elymrepe 0.02782 0.02959 0.94040 1.33300 2.00000 0.398 0.56 #> Trifrepe 0.02584 0.01656 1.56030 4.66700 2.80000 0.453 0.78 #> Anthodor 0.02358 0.02042 1.15470 1.33300 1.80000 0.503 0.55 #> Achimill 0.02343 0.01474 1.58930 2.33300 1.20000 0.552 0.24 #> Bracruta 0.02273 0.01802 1.26170 2.00000 2.80000 0.600 0.89 #> Alopgeni 0.02161 0.02308 0.93630 0.66700 1.60000 0.646 0.91 #> Trifprat 0.02151 0.02207 0.97470 0.00000 1.80000 0.691 0.12 #> Juncarti 0.02008 0.02555 0.78600 0.00000 1.60000 0.734 0.59 #> Scorautu 0.01932 0.01357 1.42410 4.33300 2.80000 0.774 0.78 #> Bellpere 0.01829 0.01486 1.23050 1.66700 0.40000 0.813 0.22 #> Agrostol 0.01761 0.02284 0.77080 0.00000 1.40000 0.850 1.00 #> Juncbufo 0.01500 0.02066 0.72600 0.00000 1.20000 0.882 0.67 #> Vicilath 0.01285 0.01140 1.12740 1.00000 0.00000 0.909 0.04 * #> Sagiproc 0.01168 0.01297 0.90080 0.66700 0.80000 0.934 0.92 #> Eleopalu 0.01017 0.02111 0.48170 0.00000 0.80000 0.955 0.93 #> Hyporadi 0.00895 0.01312 0.68240 0.66700 0.00000 0.974 0.64 #> Poaprat 0.00720 0.01010 0.71330 4.00000 3.40000 0.989 1.00 #> Ranuflam 0.00508 0.01055 0.48170 0.00000 0.40000 1.000 0.97 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Callcusp 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_NM #> #> average sd ratio ava avb cumsum p #> Lolipere 0.09068 0.02644 3.42900 6.00000 0.33300 0.124 0.04 * #> Poatriv 0.05468 0.04465 1.22500 3.66700 0.00000 0.199 0.28 #> Poaprat 0.05251 0.01813 2.89700 4.00000 0.66700 0.271 0.02 * #> Trifrepe 0.05129 0.02756 1.86100 4.66700 1.83300 0.342 0.04 * #> Bromhord 0.03969 0.02920 1.35900 2.66700 0.00000 0.396 0.01 ** #> Bracruta 0.03572 0.02869 1.24500 2.00000 2.83300 0.445 0.40 #> Eleopalu 0.03376 0.03573 0.94500 0.00000 2.16700 0.491 0.43 #> Agrostol 0.03345 0.03473 0.96300 0.00000 2.16700 0.537 0.87 #> Achimill 0.03319 0.02338 1.42000 2.33300 0.33300 0.582 0.02 * #> Scorautu 0.03136 0.02026 1.54800 4.33300 3.16700 0.625 0.29 #> Anthodor 0.02806 0.03295 0.85200 1.33300 1.33300 0.664 0.37 #> Planlanc 0.02732 0.02193 1.24600 2.00000 0.83300 0.701 0.62 #> Salirepe 0.02677 0.02927 0.91400 0.00000 1.83300 0.738 0.11 #> Bellpere 0.02353 0.01909 1.23200 1.66700 0.33300 0.770 0.04 * #> Hyporadi 0.02172 0.02450 0.88600 0.66700 1.16700 0.800 0.26 #> Ranuflam 0.02031 0.02275 0.89300 0.00000 1.33300 0.828 0.27 #> Elymrepe 0.01999 0.02926 0.68300 1.33300 0.00000 0.855 0.79 #> Callcusp 0.01783 0.02681 0.66500 0.00000 1.16700 0.880 0.44 #> Juncarti 0.01769 0.02600 0.68100 0.00000 1.16700 0.904 0.73 #> Vicilath 0.01577 0.01447 1.09000 1.00000 0.16700 0.925 0.01 ** #> Sagiproc 0.01543 0.01857 0.83100 0.66700 0.50000 0.947 0.79 #> Airaprae 0.01341 0.01969 0.68100 0.00000 0.83300 0.965 0.30 #> Comapalu 0.01074 0.01571 0.68400 0.00000 0.66700 0.980 0.42 #> Alopgeni 0.01000 0.01463 0.68300 0.66700 0.00000 0.993 0.99 #> Empenigr 0.00479 0.01105 0.43300 0.00000 0.33300 1.000 0.41 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Juncbufo 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Rumeacet 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: HF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07155 0.01368 5.23000 4.80000 0.00000 0.099 0.01 ** #> Lolipere 0.05453 0.02962 1.84100 4.00000 0.33300 0.175 0.22 #> Rumeacet 0.04655 0.03081 1.51100 3.20000 0.00000 0.239 0.01 ** #> Poaprat 0.04175 0.01885 2.21500 3.40000 0.66700 0.297 0.06 . #> Planlanc 0.04163 0.02956 1.40800 3.00000 0.83300 0.355 0.03 * #> Bracruta 0.03534 0.02010 1.75800 2.80000 2.83300 0.404 0.40 #> Eleopalu 0.03204 0.03231 0.99200 0.80000 2.16700 0.448 0.51 #> Agrostol 0.03192 0.02889 1.10500 1.40000 2.16700 0.492 0.99 #> Trifrepe 0.03037 0.02287 1.32800 2.80000 1.83300 0.534 0.71 #> Elymrepe 0.02981 0.03868 0.77100 2.00000 0.00000 0.576 0.53 #> Anthodor 0.02872 0.02480 1.15800 1.80000 1.33300 0.615 0.24 #> Juncarti 0.02741 0.02854 0.96100 1.60000 1.16700 0.653 0.19 #> Trifprat 0.02584 0.02597 0.99500 1.80000 0.00000 0.689 0.01 ** #> Salirepe 0.02534 0.02729 0.92900 0.00000 1.83300 0.724 0.18 #> Alopgeni 0.02446 0.03240 0.75500 1.60000 0.00000 0.758 0.92 #> Scorautu 0.02070 0.01412 1.46600 2.80000 3.16700 0.787 0.85 #> Ranuflam 0.01928 0.01994 0.96700 0.40000 1.33300 0.814 0.36 #> Juncbufo 0.01818 0.02465 0.73800 1.20000 0.00000 0.839 0.43 #> Hyporadi 0.01714 0.02655 0.64600 0.00000 1.16700 0.863 0.42 #> Callcusp 0.01683 0.02490 0.67600 0.00000 1.16700 0.886 0.42 #> Achimill 0.01656 0.01490 1.11100 1.20000 0.33300 0.909 0.59 #> Sagiproc 0.01528 0.01653 0.92400 0.80000 0.50000 0.930 0.89 #> Airaprae 0.01261 0.01824 0.69100 0.00000 0.83300 0.947 0.30 #> Bromhord 0.01209 0.01517 0.79700 0.80000 0.00000 0.964 0.84 #> Comapalu 0.01011 0.01456 0.69400 0.00000 0.66700 0.978 0.25 #> Bellpere 0.00880 0.01373 0.64100 0.40000 0.33300 0.990 0.94 #> Empenigr 0.00454 0.01033 0.43900 0.00000 0.33300 0.997 0.60 #> Vicilath 0.00240 0.00546 0.43900 0.00000 0.16700 1.000 0.86 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 99 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Function simulates response data frame adds Gaussian error fitted responses Redundancy Analysis (rda), Constrained Correspondence Analysis (cca) distance-based RDA (capscale). function special case generic simulate, works similarly simulate.lm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"# S3 method for rda simulate(object, nsim = 1, seed = NULL, indx = NULL, rank = \"full\", correlated = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"object object representing fitted rda, cca capscale model. nsim number response matrices simulated. one dissimilarity matrix returned capscale, larger nsim error. seed object specifying random number generator initialized (‘seeded’). See simulate details. indx Index residuals added fitted values, produced shuffleSet sample. index can duplicate entries bootstrapping allowed. nsim \\(>1\\), output compliant shuffleSet one line simulation. nsim missing, number rows indx used define number simulations, nsim given, match number rows indx. null, parametric simulation used Gaussian error added fitted values. rank rank constrained component: passed predict.rda predict.cca. correlated species regarded correlated parametric simulation indx given? correlated = TRUE, multivariate Gaussian random error generated, FALSE, Gaussian random error generated separately species. argument effect capscale information species. ... additional optional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"implementation follows \"lm\" method simulate, adds Gaussian (Normal) error fitted values (fitted.rda) using function rnorm correlated = FALSE mvrnorm correlated = TRUE. standard deviations (rnorm) covariance matrices species (mvrnorm) estimated residuals fitting constraints. Alternatively, function can take permutation index used add permuted residuals (unconstrained component) fitted values. Raw data used rda. Internal Chi-square transformed data used cca within function, returned matrix similar original input data. simulation performed internal metric scaling data capscale, function returns Euclidean distances calculated simulated data. simulation uses real components, imaginary dimensions ignored.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"nsim = 1, returns matrix dissimilarities ( capscale) similar additional arguments random number seed simulate. nsim > 1, returns similar array returned simulate.nullmodel similar attributes.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"data(dune) data(dune.env) mod <- rda(dune ~ Moisture + Management, dune.env) ## One simulation update(mod, simulate(mod) ~ .) #> Call: rda(formula = simulate(mod) ~ Moisture + Management, data = #> dune.env) #> #> Inertia Proportion Rank #> Total 79.3906 1.0000 #> Constrained 52.2955 0.6587 6 #> Unconstrained 27.0951 0.3413 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 24.007 14.238 5.712 3.314 2.879 2.145 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 6.128 4.391 3.331 3.143 2.121 1.757 1.488 1.152 1.115 1.044 0.682 0.539 0.205 #> ## An impression of confidence regions of site scores plot(mod, display=\"sites\") for (i in 1:5) lines(procrustes(mod, update(mod, simulate(mod) ~ .)), col=\"blue\") ## Simulate a set of null communities with permutation of residuals simulate(mod, indx = shuffleSet(nrow(dune), 99)) #> An object of class “simulate.rda” #> ‘simulate index’ method (abundance, non-sequential) #> 20 x 30 matrix #> Number of permuted matrices = 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":null,"dir":"Reference","previous_headings":"","what":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Land birds islands covered coniferous forest Sipoo Archipelago, southern Finland.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map)"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"sipoo data frame contains data occurrences 50 land bird species 18 islands Sipoo Archipelago (Simberloff & Martin, 1991, Appendix 3). species referred 4+4 letter abbreviation Latin names (using five letters two species names make unique). sipoo.map data contains geographic coordinates islands ETRS89-TM35FIN coordinate system (EPSG:3067) areas islands hectares.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Simberloff, D. & Martin, J.-L. (1991). Nestedness insular avifaunas: simple summary statistics masking complex species patterns. Ornis Fennica 68:178--192.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map) plot(N ~ E, data=sipoo.map, asp = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":null,"dir":"Reference","previous_headings":"","what":"Minimum Spanning Tree — spantree","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree connecting points, disregarding dissimilarities threshold NA.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Minimum Spanning Tree — spantree","text":"","code":"spantree(d, toolong = 0) # S3 method for spantree as.hclust(x, ...) # S3 method for spantree cophenetic(x) spandepth(x) # S3 method for spantree plot(x, ord, cex = 0.7, type = \"p\", labels, dlim, FUN = sammon, ...) # S3 method for spantree lines(x, ord, display=\"sites\", col = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Minimum Spanning Tree — spantree","text":"d Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. x spantree result object. ord ordination configuration, ordination result known scores. cex Character expansion factor. type Observations plotted points type=\"p\" type=\"b\", text label type=\"t\". tree (lines) always plotted. labels Text used type=\"t\" node names missing. dlim ceiling value used highest cophenetic dissimilarity. FUN Ordination function find configuration cophenetic dissimilarities. supplied FUN work, supply ordination result argument ord. display Type scores used ord. col Colour line segments. can vector recycled points, line colour mixture two joined points. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree dissimilarities (may several minimum spanning trees, function finds one). Dissimilarities threshold toolong NAs disregarded, spanning tree found dissimilarities. data disconnected, function return disconnected tree (forest), corresponding link NA. Connected subtrees can identified using distconnected. Minimum spanning tree closely related single linkage clustering, .k.. nearest neighbour clustering, genetics neighbour joining tree available hclust agnes functions. important practical difference minimum spanning tree concept cluster membership, always joins individual points . Function .hclust can change spantree result corresponding hclust object. Function cophenetic finds distances points along tree segments. Function spandepth returns depth node. nodes tree either leaves (one link) internal nodes (one link). leaves recursively removed tree, depth layer leaf removed. disconnected spantree object (forest) tree analysed separately disconnected nodes tree depth zero. Function plot displays tree supplied ordination configuration, lines adds spanning tree ordination graph. configuration supplied plot, function ordinates cophenetic dissimilarities spanning tree overlays tree result. default ordination function sammon (package MASS), Sammon scaling emphasizes structure neighbourhood nodes may able beautifully represent tree (may need set dlim, sometimes results remain twisted). ordination methods work disconnected trees, must supply ordination configuration. Function lines overlay tree existing plot. Function spantree uses Prim's method implemented priority-first search dense graphs (Sedgewick 1990). Function cophenetic uses function stepacross option path = \"extended\". spantree fast, cophenetic slow large data sets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Minimum Spanning Tree — spantree","text":"Function spantree returns object class spantree list two vectors, length \\(n-1\\). number links tree one less number observations, first item omitted. items kid child node parent, starting parent number two. link parent, value NA tree disconnected node. dist Corresponding distance. kid = NA, dist = 0. labels Names nodes found input dissimilarities. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Minimum Spanning Tree — spantree","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Minimum Spanning Tree — spantree","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Minimum Spanning Tree — spantree","text":"principle, minimum spanning tree equivalent single linkage clustering can performed using hclust agnes. However, functions combine clusters information actually connected points (“single link”) recovered result. graphical output single linkage clustering plotted ordicluster look different equivalent spanning tree plotted lines.spantree.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Minimum Spanning Tree — spantree","text":"","code":"data(dune) dis <- vegdist(dune) tr <- spantree(dis) ## Add tree to a metric scaling plot(tr, cmdscale(dis), type = \"t\") ## Find a configuration to display the tree neatly plot(tr, type = \"t\") #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Depths of nodes depths <- spandepth(tr) plot(tr, type = \"t\", label = depths) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Plot as a dendrogram cl <- as.hclust(tr) plot(cl) ## cut hclust tree to classes and show in colours in spantree plot(tr, col = cutree(cl, 5), pch=16) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":null,"dir":"Reference","previous_headings":"","what":"Species Accumulation Curves — specaccum","title":"Species Accumulation Curves — specaccum","text":"Function specaccum finds species accumulation curves number species certain number sampled sites individuals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species Accumulation Curves — specaccum","text":"","code":"specaccum(comm, method = \"exact\", permutations = 100, conditioned =TRUE, gamma = \"jack1\", w = NULL, subset, ...) # S3 method for specaccum plot(x, add = FALSE, random = FALSE, ci = 2, ci.type = c(\"bar\", \"line\", \"polygon\"), col = par(\"fg\"), lty = 1, ci.col = col, ci.lty = 1, ci.length = 0, xlab, ylab = x$method, ylim, xvar = c(\"sites\", \"individuals\", \"effort\"), ...) # S3 method for specaccum boxplot(x, add = FALSE, ...) fitspecaccum(object, model, method = \"random\", ...) # S3 method for fitspecaccum plot(x, col = par(\"fg\"), lty = 1, xlab = \"Sites\", ylab = x$method, ...) # S3 method for specaccum predict(object, newdata, interpolation = c(\"linear\", \"spline\"), ...) # S3 method for fitspecaccum predict(object, newdata, ...) specslope(object, at)"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species Accumulation Curves — specaccum","text":"comm Community data set. method Species accumulation method (partial match). Method \"collector\" adds sites order happen data, \"random\" adds sites random order, \"exact\" finds expected (mean) species richness, \"coleman\" finds expected richness following Coleman et al. 1982, \"rarefaction\" finds mean accumulating individuals instead sites. permutations Number permutations method = \"random\". Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. conditioned Estimation standard deviation conditional empirical dataset exact SAC gamma Method estimating total extrapolated number species survey area function specpool w Weights giving sampling effort. subset logical expression indicating sites (rows) keep: missing values taken FALSE. x specaccum result object add Add existing graph. random Draw random simulation separately instead drawing average confidence intervals. ci Multiplier used get confidence intervals standard deviation (standard error estimate). Value ci = 0 suppresses drawing confidence intervals. ci.type Type confidence intervals graph: \"bar\" draws vertical bars, \"line\" draws lines, \"polygon\" draws shaded area. col Colour drawing lines. lty line type (see par). ci.col Colour drawing lines filling \"polygon\". ci.lty Line type confidence intervals border \"polygon\". ci.length Length horizontal bars (inches) end vertical bars ci.type = \"bar\". xlab,ylab Labels x (defaults xvar) y axis. ylim y limits plot. xvar Variable used horizontal axis: \"individuals\" can used method = \"rarefaction\". object Either community data set fitted specaccum model. model Nonlinear regression model (nls). See Details. newdata Optional data used prediction interpreted number sampling units (sites). missing, fitted values returned. interpolation Interpolation method used newdata. Number plots slope evaluated. Can real number. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species Accumulation Curves — specaccum","text":"Species accumulation curves (SAC) used compare diversity properties community data sets using different accumulator functions. classic method \"random\" finds mean SAC standard deviation random permutations data, subsampling without replacement (Gotelli & Colwell 2001). \"exact\" method finds expected SAC using sample-based rarefaction method independently developed numerous times (Chiarucci et al. 2008) often known Mao Tau estimate (Colwell et al. 2012). unconditional standard deviation exact SAC represents moment-based estimation conditioned empirical data set (sd samples > 0). unconditional standard deviation based estimation extrapolated number species survey area (.k.. gamma diversity), estimated function specpool. conditional standard deviation developed Jari Oksanen ( published, sd=0 samples). Method \"coleman\" finds expected SAC standard deviation following Coleman et al. (1982). methods based sampling sites without replacement. contrast, method = \"rarefaction\" finds expected species richness standard deviation sampling individuals instead sites. achieves applying function rarefy number individuals corresponding average number individuals per site. Methods \"random\" \"collector\" can take weights (w) give sampling effort site. weights w influence order sites accumulated, value sampling effort sites equal. summary results expressed sites even accumulation uses weights (methods \"random\", \"collector\"), based individuals (\"rarefaction\"). actual sampling effort given item Effort Individuals printed result. weighted \"random\" method effort refers average effort per site, sum weights per number sites. weighted method = \"random\", averaged species richness found linear interpolation single random permutations. Therefore least first value (often several first) NA richness, values interpolated cases extrapolated. plot function defaults display results scaled sites, can changed selecting xvar = \"effort\" (weighted methods) xvar = \"individuals\" (method = \"rarefaction\"). summary boxplot methods available method = \"random\". Function predict specaccum can return values corresponding newdata. method \"exact\", \"rarefaction\" \"coleman\" function uses analytic equations interpolated non-integer values, methods linear (approx) spline (spline) interpolation. newdata given, function returns values corresponding data. NB., fitted values method=\"rarefaction\" based rounded integer counts, predict can use fractional non-integer counts newdata give slightly different results. Function fitspecaccum fits nonlinear (nls) self-starting species accumulation model. input object can result specaccum community data frame. latter case function first fits specaccum model proceeds fitting nonlinear model. function can apply limited set nonlinear regression models suggested species-area relationship (Dengler 2009). selfStart models. permissible alternatives \"arrhenius\" (SSarrhenius), \"gleason\" (SSgleason), \"gitay\" (SSgitay), \"lomolino\" (SSlomolino) vegan package. addition following standard R models available: \"asymp\" (SSasymp), \"gompertz\" (SSgompertz), \"michaelis-menten\" (SSmicmen), \"logis\" (SSlogis), \"weibull\" (SSweibull). See functions model specification details. weights w used fit based accumulated effort model = \"rarefaction\" accumulated number individuals. plot still based sites, unless alternative selected xvar. Function predict fitspecaccum uses predict.nls, can pass arguments function. addition, fitted, residuals, nobs, coef, AIC, logLik deviance work result object. Function specslope evaluates derivative species accumulation curve given number sample plots, gives rate increase number species. function works specaccum result object based analytic models \"exact\", \"rarefaction\" \"coleman\", non-linear regression results fitspecaccum. Nonlinear regression may fail reason, fitspecaccum models fragile may succeed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species Accumulation Curves — specaccum","text":"Function specaccum returns object class \"specaccum\", fitspecaccum model class \"fitspecaccum\" adds items \"specaccum\" (see end list ): call Function call. method Accumulator method. sites Number sites. method = \"rarefaction\" number sites corresponding certain number individuals generally integer, average number individuals also returned item individuals. effort Average sum weights corresponding number sites model fitted argument w richness number species corresponding number sites. method = \"collector\" observed richness, methods average expected richness. sd standard deviation SAC (standard error). NULL method = \"collector\", estimated permutations method = \"random\", analytic equations methods. perm Permutation results method = \"random\" NULL cases. column perm holds one permutation. weights Matrix accumulated weights corresponding columns perm matrix model fitted argument w. fitted, residuals, coefficients fitspecacum: fitted values, residuals nonlinear model coefficients. method = \"random\" matrices column random accumulation. models fitspecaccum: list fitted nls models (see Examples accessing models).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Species Accumulation Curves — specaccum","text":"Chiarucci, ., Bacaro, G., Rocchini, D. & Fattorini, L. (2008). Discovering rediscovering sample-based rarefaction formula ecological literature. Commun. Ecol. 9: 121--123. Coleman, B.D, Mares, M.., Willis, M.R. & Hsieh, Y. (1982). Randomness, area species richness. Ecology 63: 1121--1133. Colwell, R.K., Chao, ., Gotelli, N.J., Lin, S.Y., Mao, C.X., Chazdon, R.L. & Longino, J.T. (2012). Models estimators linking individual-based sample-based rarefaction, extrapolation comparison assemblages. J. Plant Ecol. 5: 3--21. Dengler, J. (2009). function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity: procedures pitfalls measurement comparison species richness. Ecol. Lett. 4, 379--391.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species Accumulation Curves — specaccum","text":"Roeland Kindt r.kindt@cgiar.org Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Species Accumulation Curves — specaccum","text":"SAC method = \"exact\" developed Roeland Kindt, standard deviation Jari Oksanen (unpublished). method = \"coleman\" underestimates SAC handle properly sampling without replacement. , standard deviation take account species correlations, generally low.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species Accumulation Curves — specaccum","text":"","code":"data(BCI) sp1 <- specaccum(BCI) #> Warning: the standard deviation is zero sp2 <- specaccum(BCI, \"random\") sp2 #> Species Accumulation Curve #> Accumulation method: random, with 100 permutations #> Call: specaccum(comm = BCI, method = \"random\") #> #> #> Sites 1.00000 2.00000 3.00000 4.00000 5.00000 6.00000 7.00000 #> Richness 90.42000 121.29000 138.34000 150.46000 158.46000 165.38000 170.34000 #> sd 7.23931 7.43361 7.38305 6.80377 5.92959 5.07475 4.58196 #> #> Sites 8.00000 9.00000 10.00000 11.0000 12.00000 13.00000 14.0000 #> Richness 175.02000 178.93000 182.15000 184.8500 187.47000 189.88000 191.9000 #> sd 4.43808 4.74662 4.58671 4.4093 4.23419 4.08565 4.1451 #> #> Sites 15.00000 16.00000 17.00000 18.00000 19.00000 20.00000 21.00000 #> Richness 193.75000 195.50000 197.32000 198.94000 200.32000 201.68000 202.95000 #> sd 4.13503 4.01889 3.54447 3.53859 3.62617 3.62895 3.41528 #> #> Sites 22.00000 23.00000 24.00000 25.00000 26.00000 27.000 28.00000 #> Richness 204.17000 205.39000 206.65000 207.73000 208.91000 209.880 210.90000 #> sd 3.47009 3.28417 3.22357 3.06448 3.06856 3.217 3.14787 #> #> Sites 29.00000 30.00000 31.00000 32.00000 33.00000 34.00000 35.00000 #> Richness 211.64000 212.74000 213.57000 214.43000 215.15000 215.93000 216.75000 #> sd 3.12862 3.01719 3.01931 2.92069 2.76111 2.77163 2.69446 #> #> Sites 36.00000 37.00000 38.00000 39.00000 40.00000 41.00000 42.00000 #> Richness 217.54000 218.20000 219.00000 219.61000 220.14000 220.77000 221.28000 #> sd 2.62629 2.64384 2.45361 2.22427 2.11307 1.93769 1.82618 #> #> Sites 43.00000 44.00000 45.00000 46.00000 47.00000 48.0000 49.00000 #> Richness 221.86000 222.33000 222.85000 223.28000 223.69000 224.1400 224.60000 #> sd 1.79235 1.49784 1.45904 1.36389 1.07961 0.8764 0.58603 #> #> Sites 50 #> Richness 225 #> sd 0 summary(sp2) #> 1 sites 2 sites 3 sites 4 sites #> Min. : 77.00 Min. :105.0 Min. :119.0 Min. :132.0 #> 1st Qu.: 85.00 1st Qu.:115.8 1st Qu.:134.0 1st Qu.:146.0 #> Median : 88.00 Median :121.5 Median :139.0 Median :151.0 #> Mean : 90.42 Mean :121.3 Mean :138.3 Mean :150.5 #> 3rd Qu.: 94.25 3rd Qu.:126.0 3rd Qu.:143.2 3rd Qu.:155.0 #> Max. :109.00 Max. :144.0 Max. :155.0 Max. :166.0 #> 5 sites 6 sites 7 sites 8 sites 9 sites #> Min. :140.0 Min. :149.0 Min. :157.0 Min. :161 Min. :166.0 #> 1st Qu.:155.0 1st Qu.:162.0 1st Qu.:168.0 1st Qu.:173 1st Qu.:176.0 #> Median :159.0 Median :165.0 Median :171.0 Median :175 Median :179.0 #> Mean :158.5 Mean :165.4 Mean :170.3 Mean :175 Mean :178.9 #> 3rd Qu.:163.0 3rd Qu.:170.0 3rd Qu.:174.0 3rd Qu.:178 3rd Qu.:182.0 #> Max. :172.0 Max. :176.0 Max. :181.0 Max. :186 Max. :190.0 #> 10 sites 11 sites 12 sites 13 sites #> Min. :170.0 Min. :174.0 Min. :174.0 Min. :179.0 #> 1st Qu.:179.0 1st Qu.:182.0 1st Qu.:184.0 1st Qu.:187.0 #> Median :183.0 Median :185.0 Median :188.0 Median :190.5 #> Mean :182.2 Mean :184.8 Mean :187.5 Mean :189.9 #> 3rd Qu.:185.0 3rd Qu.:188.0 3rd Qu.:191.0 3rd Qu.:193.0 #> Max. :191.0 Max. :193.0 Max. :200.0 Max. :202.0 #> 14 sites 15 sites 16 sites 17 sites #> Min. :181.0 Min. :183.0 Min. :185.0 Min. :188.0 #> 1st Qu.:189.0 1st Qu.:191.0 1st Qu.:193.0 1st Qu.:195.0 #> Median :193.0 Median :194.0 Median :196.0 Median :197.5 #> Mean :191.9 Mean :193.8 Mean :195.5 Mean :197.3 #> 3rd Qu.:194.0 3rd Qu.:196.0 3rd Qu.:198.0 3rd Qu.:200.0 #> Max. :203.0 Max. :204.0 Max. :205.0 Max. :205.0 #> 18 sites 19 sites 20 sites 21 sites #> Min. :188.0 Min. :188.0 Min. :190.0 Min. :195.0 #> 1st Qu.:196.8 1st Qu.:198.0 1st Qu.:199.0 1st Qu.:201.0 #> Median :199.0 Median :200.0 Median :202.0 Median :203.0 #> Mean :198.9 Mean :200.3 Mean :201.7 Mean :202.9 #> 3rd Qu.:201.0 3rd Qu.:203.0 3rd Qu.:204.0 3rd Qu.:205.0 #> Max. :208.0 Max. :209.0 Max. :210.0 Max. :211.0 #> 22 sites 23 sites 24 sites 25 sites #> Min. :195.0 Min. :199.0 Min. :199.0 Min. :200.0 #> 1st Qu.:202.0 1st Qu.:203.0 1st Qu.:204.0 1st Qu.:206.0 #> Median :204.0 Median :205.0 Median :207.0 Median :207.5 #> Mean :204.2 Mean :205.4 Mean :206.7 Mean :207.7 #> 3rd Qu.:206.2 3rd Qu.:207.0 3rd Qu.:209.0 3rd Qu.:210.0 #> Max. :214.0 Max. :214.0 Max. :214.0 Max. :215.0 #> 26 sites 27 sites 28 sites 29 sites #> Min. :203.0 Min. :203.0 Min. :204.0 Min. :204.0 #> 1st Qu.:207.0 1st Qu.:208.0 1st Qu.:209.0 1st Qu.:210.0 #> Median :209.0 Median :210.0 Median :211.0 Median :212.0 #> Mean :208.9 Mean :209.9 Mean :210.9 Mean :211.6 #> 3rd Qu.:211.0 3rd Qu.:212.0 3rd Qu.:213.0 3rd Qu.:214.0 #> Max. :217.0 Max. :218.0 Max. :220.0 Max. :220.0 #> 30 sites 31 sites 32 sites 33 sites #> Min. :206.0 Min. :206.0 Min. :206.0 Min. :207.0 #> 1st Qu.:211.0 1st Qu.:212.0 1st Qu.:213.0 1st Qu.:214.0 #> Median :212.5 Median :213.0 Median :214.0 Median :215.0 #> Mean :212.7 Mean :213.6 Mean :214.4 Mean :215.2 #> 3rd Qu.:215.0 3rd Qu.:215.0 3rd Qu.:216.0 3rd Qu.:216.0 #> Max. :221.0 Max. :221.0 Max. :222.0 Max. :222.0 #> 34 sites 35 sites 36 sites 37 sites #> Min. :208.0 Min. :208.0 Min. :208.0 Min. :208.0 #> 1st Qu.:214.0 1st Qu.:215.0 1st Qu.:216.0 1st Qu.:217.0 #> Median :216.0 Median :217.0 Median :218.0 Median :218.0 #> Mean :215.9 Mean :216.8 Mean :217.5 Mean :218.2 #> 3rd Qu.:218.0 3rd Qu.:219.0 3rd Qu.:219.0 3rd Qu.:220.0 #> Max. :222.0 Max. :222.0 Max. :223.0 Max. :224.0 #> 38 sites 39 sites 40 sites 41 sites #> Min. :212.0 Min. :212.0 Min. :212.0 Min. :212.0 #> 1st Qu.:217.0 1st Qu.:218.0 1st Qu.:219.0 1st Qu.:220.0 #> Median :219.0 Median :220.0 Median :220.0 Median :221.0 #> Mean :219.0 Mean :219.6 Mean :220.1 Mean :220.8 #> 3rd Qu.:220.2 3rd Qu.:221.0 3rd Qu.:221.0 3rd Qu.:222.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 42 sites 43 sites 44 sites 45 sites #> Min. :215.0 Min. :215.0 Min. :218.0 Min. :219.0 #> 1st Qu.:220.0 1st Qu.:221.0 1st Qu.:221.0 1st Qu.:222.0 #> Median :221.0 Median :222.0 Median :222.0 Median :223.0 #> Mean :221.3 Mean :221.9 Mean :222.3 Mean :222.8 #> 3rd Qu.:222.2 3rd Qu.:223.0 3rd Qu.:223.2 3rd Qu.:224.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 46 sites 47 sites 48 sites 49 sites 50 sites #> Min. :220.0 Min. :221.0 Min. :222.0 Min. :223.0 Min. :225 #> 1st Qu.:222.8 1st Qu.:223.0 1st Qu.:224.0 1st Qu.:224.0 1st Qu.:225 #> Median :223.0 Median :224.0 Median :224.0 Median :225.0 Median :225 #> Mean :223.3 Mean :223.7 Mean :224.1 Mean :224.6 Mean :225 #> 3rd Qu.:224.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 Max. :225 plot(sp1, ci.type=\"poly\", col=\"blue\", lwd=2, ci.lty=0, ci.col=\"lightblue\") boxplot(sp2, col=\"yellow\", add=TRUE, pch=\"+\") ## Fit Lomolino model to the exact accumulation mod1 <- fitspecaccum(sp1, \"lomolino\") coef(mod1) #> Asym xmid slope #> 258.440682 2.442061 1.858694 fitted(mod1) #> [1] 94.34749 121.23271 137.45031 148.83053 157.45735 164.31866 169.95946 #> [8] 174.71115 178.78954 182.34254 185.47566 188.26658 190.77402 193.04337 #> [15] 195.11033 197.00350 198.74606 200.35705 201.85227 203.24499 204.54643 #> [22] 205.76612 206.91229 207.99203 209.01150 209.97609 210.89054 211.75903 #> [29] 212.58527 213.37256 214.12386 214.84180 215.52877 216.18692 216.81820 #> [36] 217.42437 218.00703 218.56767 219.10762 219.62811 220.13027 220.61514 #> [43] 221.08369 221.53679 221.97528 222.39991 222.81138 223.21037 223.59747 #> [50] 223.97327 plot(sp1) ## Add Lomolino model using argument 'add' plot(mod1, add = TRUE, col=2, lwd=2) ## Fit Arrhenius models to all random accumulations mods <- fitspecaccum(sp2, \"arrh\") plot(mods, col=\"hotpink\") boxplot(sp2, col = \"yellow\", border = \"blue\", lty=1, cex=0.3, add= TRUE) ## Use nls() methods to the list of models sapply(mods$models, AIC) #> [1] 329.2456 321.8614 331.1568 352.1309 354.8606 323.3767 322.0897 347.7218 #> [9] 327.6283 327.3510 335.4843 308.5300 373.4959 334.3691 320.6975 338.1084 #> [17] 330.2018 343.5560 348.5827 300.2513 315.7909 310.7837 305.5217 295.1005 #> [25] 359.2639 342.7111 339.7456 328.4175 338.4335 368.4988 354.7097 328.8704 #> [33] 342.1528 292.9875 328.3752 299.7201 336.5796 364.3153 336.4822 339.6666 #> [41] 354.7445 320.7841 304.1628 314.9640 349.6753 350.7169 307.6840 296.7971 #> [49] 359.6018 302.9863 319.6832 339.5021 328.8450 329.0553 298.0970 334.0182 #> [57] 291.9027 349.7161 325.1376 347.7797 340.6111 338.6148 352.0292 308.8538 #> [65] 315.1890 289.7947 351.6732 339.3587 320.3515 300.9377 323.3054 290.9419 #> [73] 350.8587 322.6970 340.7948 334.1965 358.2890 300.7034 338.5079 338.8941 #> [81] 341.7687 331.4553 337.5975 349.7867 331.8581 270.3068 354.5083 326.0026 #> [89] 337.8555 322.7217 302.5019 324.7469 322.1415 324.6922 324.5427 347.2050 #> [97] 309.1766 316.1194 343.9958 350.5935"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":null,"dir":"Reference","previous_headings":"","what":"Extrapolated Species Richness in a Species Pool — specpool","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions estimate extrapolated species richness species pool, number unobserved species. Function specpool based incidences sample sites, gives single estimate collection sample sites (matrix). Function estimateR based abundances (counts) single sample site.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"specpool(x, pool, smallsample = TRUE) estimateR(x, ...) specpool2vect(X, index = c(\"jack1\",\"jack2\", \"chao\", \"boot\",\"Species\")) poolaccum(x, permutations = 100, minsize = 3) estaccumR(x, permutations = 100, parallel = getOption(\"mc.cores\")) # S3 method for poolaccum summary(object, display, alpha = 0.05, ...) # S3 method for poolaccum plot(x, alpha = 0.05, type = c(\"l\",\"g\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"x Data frame matrix species data analysis result plot function. pool vector giving classification pooling sites species data. missing, sites pooled together. smallsample Use small sample correction \\((N-1)/N\\), \\(N\\) number sites within pool. X, object specpool result object. index selected index extrapolated richness. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. minsize Smallest number sampling units reported. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. display Indices displayed. alpha Level quantiles shown. proportion left outside symmetric limits. type Type graph produced xyplot. ... parameters (used).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Many species always remain unseen undetected collection sample plots. function uses popular ways estimating number unseen species adding observed species richness (Palmer 1990, Colwell & Coddington 1994). incidence-based estimates specpool use frequencies species collection sites. following, \\(S_P\\) extrapolated richness pool, \\(S_0\\) observed number species collection, \\(a_1\\) \\(a_2\\) number species occurring one two sites collection, \\(p_i\\) frequency species \\(\\), \\(N\\) number sites collection. variants extrapolated richness specpool : specpool normally uses basic Chao equation, doubletons (\\(a2=0\\)) switches bias-corrected version. case Chao equation simplifies \\(S_0 + \\frac{1}{2} a_1 (a_1-1) \\frac{N-1}{N}\\). abundance-based estimates estimateR use counts (numbers individuals) species single site. called matrix data frame, function give separate estimates site. two variants extrapolated richness estimateR bias-corrected Chao ACE (O'Hara 2005, Chiu et al. 2014). Chao estimate similar bias corrected one , \\(a_i\\) refers number species abundance \\(\\) instead number sites, small-sample correction used. ACE estimate defined : \\(a_i\\) refers number species abundance \\(\\) \\(S_{rare}\\) number rare species, \\(S_{abund}\\) number abundant species, arbitrary threshold abundance 10 rare species, \\(N_{rare}\\) number individuals rare species. Functions estimate standard errors estimates. concern number added species, assume variance observed richness. equations standard errors complicated reproduced help page, can studied R source code function discussed vignette can read browseVignettes(\"vegan\"). standard error based following sources: Chiu et al. (2014) Chao estimates Smith van Belle (1984) first-order Jackknife bootstrap (second-order jackknife still missing). variance estimator \\(S_{ace}\\) see O'Hara (2005). Functions poolaccum estaccumR similar specaccum, estimate extrapolated richness indices specpool estimateR addition number species random ordering sampling units. Function specpool uses presence data estaccumR count data. functions share summary plot methods. summary returns quantile envelopes permutations corresponding given level alpha standard deviation permutations sample size. NB., based standard deviations estimated within specpool estimateR, based permutations. plot function shows mean envelope permutations given alpha models. selection models can restricted order changes using display argument summary plot. configuration plot command, see xyplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Function specpool returns data frame entries observed richness indices class pool vector. utility function specpool2vect maps pooled values vector giving value selected index original site. Function estimateR returns estimates standard errors site. Functions poolaccum estimateR return matrices permutation results richness estimator, vector sample sizes table means permutations estimator.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Chao, . (1987). Estimating population size capture-recapture data unequal catchability. Biometrics 43, 783--791. Chiu, C.H., Wang, Y.T., Walther, B.. & Chao, . (2014). Improved nonparametric lower bound species richness via modified Good-Turing frequency formula. Biometrics 70, 671--682. Colwell, R.K. & Coddington, J.. (1994). Estimating terrestrial biodiversity extrapolation. Phil. Trans. Roy. Soc. London B 345, 101--118. O'Hara, R.B. (2005). Species richness estimators: many species can dance head pin? J. Anim. Ecol. 74, 375--386. Palmer, M.W. (1990). estimation species richness extrapolation. Ecology 71, 1195--1198. Smith, E.P & van Belle, G. (1984). Nonparametric estimation species richness. Biometrics 40, 119--129.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Bob O'Hara (estimateR) Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions based assumption species pool: community closed fixed pool size \\(S_P\\). general, functions give lower limit species richness: real richness \\(S >= S_P\\), consistent bias estimates. Even bias-correction Chao reduces bias, remove completely (Chiu et al. 2014). Optional small sample correction added specpool vegan 2.2-0. used older literature (Chao 1987), recommended recently (Chiu et al. 2014).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"data(dune) data(dune.env) pool <- with(dune.env, specpool(dune, Management)) pool #> Species chao chao.se jack1 jack1.se jack2 boot boot.se n #> BF 16 17.19048 1.5895675 19.33333 2.211083 19.83333 17.74074 1.646379 3 #> HF 21 21.51429 0.9511693 23.40000 1.876166 22.05000 22.56864 1.821518 5 #> NM 21 22.87500 2.1582871 26.00000 3.291403 25.73333 23.77696 2.300982 6 #> SF 21 29.88889 8.6447967 27.66667 3.496029 31.40000 23.99496 1.850288 6 op <- par(mfrow=c(1,2)) boxplot(specnumber(dune) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") boxplot(specnumber(dune)/specpool2vect(pool) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") par(op) data(BCI) ## Accumulation model pool <- poolaccum(BCI) summary(pool, display = \"chao\") #> $chao #> N Chao 2.5% 97.5% Std.Dev #> [1,] 3 162.3374 142.5272 186.2227 11.217082 #> [2,] 4 176.1243 156.4704 206.3243 12.158304 #> [3,] 5 183.8507 162.9868 209.2821 12.488675 #> [4,] 6 188.9018 165.6732 214.2779 12.888148 #> [5,] 7 193.8679 175.8712 216.1384 11.602711 #> [6,] 8 199.0315 180.2949 227.8918 12.604562 #> [7,] 9 202.0946 183.3317 228.8381 11.958432 #> [8,] 10 204.9843 185.3556 227.5898 12.637762 #> [9,] 11 206.6160 186.7205 234.5167 11.679759 #> [10,] 12 209.1286 189.2824 232.6359 12.351073 #> [11,] 13 211.3655 191.7918 232.0423 11.991967 #> [12,] 14 212.8128 195.8592 239.3318 11.537817 #> [13,] 15 215.5275 198.2739 239.7978 11.422201 #> [14,] 16 218.7094 198.3891 243.3826 12.165145 #> [15,] 17 221.6814 197.6029 256.5328 14.464302 #> [16,] 18 222.7418 201.4964 257.9494 13.227892 #> [17,] 19 224.8619 206.6236 251.9979 13.165360 #> [18,] 20 226.3324 210.2832 249.5616 12.858116 #> [19,] 21 228.4284 210.9920 257.9636 13.784663 #> [20,] 22 228.8425 211.1903 254.3500 11.629250 #> [21,] 23 230.6429 214.1737 251.1681 10.510649 #> [22,] 24 231.9109 213.0874 252.7553 10.925851 #> [23,] 25 232.9972 214.3610 258.4917 11.207306 #> [24,] 26 233.8067 216.9697 263.2733 11.546339 #> [25,] 27 235.6347 217.3831 266.9230 12.800245 #> [26,] 28 235.8942 218.6913 261.4115 11.761873 #> [27,] 29 235.9612 218.6591 258.5682 11.133700 #> [28,] 30 236.6706 219.4315 259.9315 10.291499 #> [29,] 31 236.8530 220.2338 259.2903 9.810390 #> [30,] 32 237.6018 221.4868 262.0864 10.195825 #> [31,] 33 237.4291 222.2990 256.2778 9.227891 #> [32,] 34 237.6712 222.7844 258.2268 8.979536 #> [33,] 35 237.7856 222.2750 257.2540 8.537867 #> [34,] 36 238.4325 223.6473 255.5610 8.566487 #> [35,] 37 238.1834 224.9892 253.5811 7.776992 #> [36,] 38 238.4685 225.2802 254.0155 7.626532 #> [37,] 39 238.2082 226.0390 252.7975 7.642443 #> [38,] 40 238.7991 227.1971 254.1619 7.155802 #> [39,] 41 239.2246 227.3368 258.5178 7.926574 #> [40,] 42 238.4935 228.5195 256.2137 7.287824 #> [41,] 43 238.3856 227.8327 255.8470 7.300153 #> [42,] 44 238.0092 228.5279 252.9488 6.520961 #> [43,] 45 238.1537 228.7980 253.3757 6.443894 #> [44,] 46 238.0532 230.5990 250.9539 5.466587 #> [45,] 47 237.7542 231.3089 251.0192 4.821564 #> [46,] 48 237.2842 231.9015 248.6399 4.038862 #> [47,] 49 236.8407 233.3115 245.4082 2.803888 #> [48,] 50 236.3732 236.3732 236.3732 0.000000 #> #> attr(,\"class\") #> [1] \"summary.poolaccum\" plot(pool) ## Quantitative model estimateR(BCI[1:5,]) #> 1 2 3 4 5 #> S.obs 93.000000 84.000000 90.000000 94.000000 101.000000 #> S.chao1 117.473684 117.214286 141.230769 111.550000 136.000000 #> se.chao1 11.583785 15.918953 23.001405 8.919663 15.467344 #> S.ACE 122.848959 117.317307 134.669844 118.729941 137.114088 #> se.ACE 5.736054 5.571998 6.191618 5.367571 5.848474"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distance-based ordination (dbrda, capscale, metaMDS) information species, methods may add species scores community data available. However, species scores may missing ( always dbrda), may close relation used dissimilarity index. function add species scores replace existing species scores distance-based methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"sppscores(object) <- value"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"object Ordination result. value Community data find species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distances information species (columns, variables), hence distance-based ordination information species scores. However, species scores can added supplementary information analysis help interpretation results. ordination methods (capscale, metaMDS) can supplement species scores analysis community data available analysis. capscale species scores found projecting community data site ordination (linear combination scores), scores accurate analysis used Euclidean distances. dissimilarity index can expressed Euclidean distances transformed data (instance, Chord Hellinger Distances), species scores based transformed data accurate, function still finds dissimilarities untransformed data. Usually community dissimilarities differ two significant ways Euclidean distances: bound maximum 1, use absolute differences instead squared differences. cases, may better use species scores transformed Euclidean distances good linear relation used dissimilarities. often useful standardize data row unit total, perform squareroot transformation damp effect squared differences (see Examples). Function dbrda never finds species scores, mathematically similar capscale, similar rules followed supplementing species scores. Function metaMDS uses weighted averages (wascores) find species scores. better relationship dissimilarities projection scores used metric ordination, similar transformation community data used dissimilarities species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Replacement function adds species scores replaces old scores ordination object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"data(BCI, BCI.env) mod <- dbrda(vegdist(BCI) ~ Habitat, BCI.env) ## add species scores sppscores(mod) <- BCI ## Euclidean distances of BCI differ from used dissimilarity plot(vegdist(BCI), dist(BCI)) ## more linear relationship plot(vegdist(BCI), dist(sqrt(decostand(BCI, \"total\")))) ## better species scores sppscores(mod) <- sqrt(decostand(BCI, \"total\"))"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":null,"dir":"Reference","previous_headings":"","what":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function stepacross tries replace dissimilarities shortest paths stepping across intermediate sites regarding dissimilarities threshold missing data (NA). path = \"shortest\" flexible shortest path (Williamson 1978, Bradfield & Kenkel 1987), path = \"extended\" approximation known extended dissimilarities (De'ath 1999). use stepacross improve ordination high beta diversity, many sites species common.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"stepacross(dis, path = \"shortest\", toolong = 1, trace = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. path method stepping across (partial match) Alternative \"shortest\" finds shortest paths, \"extended\" approximation known extended dissimilarities. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . trace Trace calculations. ... parameters (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Williamson (1978) suggested using flexible shortest paths estimate dissimilarities sites nothing common, shared species. path = \"shortest\" function stepacross replaces dissimilarities toolong longer NA, tries find shortest paths sites using remaining dissimilarities. Several dissimilarity indices semi-metric means obey triangle inequality \\(d_{ij} \\leq d_{ik} + d_{kj}\\), shortest path algorithm can replace dissimilarities well, even shorter toolong. De'ath (1999) suggested simplified method known extended dissimilarities, calculated path = \"extended\". method, dissimilarities toolong longer first made NA, function tries replace NA dissimilarities path single stepping stone points. NA replaced one pass, function make new passes updated dissimilarities long NA replaced extended dissimilarities. mean second passes, remaining NA dissimilarities allowed one stepping stone site, previously replaced dissimilarities updated. , function consider dissimilarities shorter toolong, although replaced shorter path semi-metric indices, used part paths. optimal cases, extended dissimilarities equal shortest paths, may longer. alternative defining long dissimilarities parameter toolong, input dissimilarities can contain NAs. toolong zero negative, function make dissimilarities NA. NAs input toolong = 0, path = \"shortest\" find shorter paths semi-metric indices, path = \"extended\" nothing. Function .shared can used set dissimilarities NA. data disconnected path points, result contain NAs warning issued. Several methods handle NA dissimilarities, warning taken seriously. Function distconnected can used find connected groups remove rare outlier observations groups observations. Alternative path = \"shortest\" uses Dijkstra's method finding flexible shortest paths, implemented priority-first search dense graphs (Sedgewick 1990). Alternative path = \"extended\" follows De'ath (1999), implementation simpler code.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function returns object class dist extended dissimilarities (see functions vegdist dist). value path appended method attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Bradfield, G.E. & Kenkel, N.C. (1987). Nonlinear ordination using flexible shortest path adjustment ecological distances. Ecology 68, 750--753. De'ath, G. (1999). Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecol. 144, 191--199. Sedgewick, R. (1990). Algorithms C. Addison Wesley. Williamson, M.H. (1978). ordination incidence data. J. Ecol. 66, 911-920.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"function changes original dissimilarities, like . may best use function really must: extremely high beta diversity large proportion dissimilarities upper limit (species common). Semi-metric indices vary degree violating triangle inequality. Morisita Horn--Morisita indices vegdist may strongly semi-metric, shortest paths can change indices much. Mountford index violates basic rules dissimilarities: non-identical sites zero dissimilarity species composition poorer site subset richer. Mountford index, can find three sites \\(, j, k\\) \\(d_{ik} = 0\\) \\(d_{jk} = 0\\), \\(d_{ij} > 0\\). results stepacross Mountford index can weird. stepacross needed, best try use metric indices .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"# There are no data sets with high beta diversity in vegan, but this # should give an idea. data(dune) dis <- vegdist(dune) edis <- stepacross(dis) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities... plot(edis, dis, xlab = \"Shortest path\", ylab = \"Original\") ## Manhattan distance have no fixed upper limit. dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune) dis <- stepacross(dis, toolong=0) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities..."},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions plot ordination distances given number dimensions observed distances distances full space eigenvector methods. display similar Shepard diagram (stressplot non-metric multidimensional scaling metaMDS monoMDS), shows linear relationship eigenvector ordinations. stressplot methods available wcmdscale, rda, cca, capscale, dbrda, prcomp princomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"# S3 method for wcmdscale stressplot(object, k = 2, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"object Result object eigenvector ordination (wcmdscale, rda, cca, dbrda, capscale) k Number dimensions ordination distances displayed. pch, p.col, l.col, lwd Plotting character, point colour line colour like default stressplot ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"functions offer similar display eigenvector ordinations standard Shepard diagram (stressplot) non-metric multidimensional scaling. ordination distances given number dimensions plotted observed distances. metric distances, ordination distances full space (ordination axes) equal observed distances, fit line shows equality. general, fit line go points, points observed distances approach fit line . However, non-Euclidean distances ( wcmdscale, dbrda capscale) negative eigenvalues ordination distances can exceed observed distances real dimensions; imaginary dimensions negative eigenvalues correct excess distances. used dbrda, capscale wcmdscale argument add avoid negative eigenvalues, ordination distances exceed observed dissimilarities. partial ordination (cca, rda, capscale Condition formula), distances partial component included observed distances ordination distances. k=0, ordination distances refer partial ordination. exception dbrda distances partial, constrained residual components additive, first components can shown, partial models shown .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions draw graph return invisibly ordination distances ordination distances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"data(dune, dune.env) mod <- rda(dune) stressplot(mod) mod <- rda(dune ~ Management, dune.env) stressplot(mod, k=3)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of Taxonomic Diversity and Distinctness — taxondive","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function finds indices taxonomic diversity distinctness, averaged taxonomic distances among species individuals community (Clarke & Warwick 1998, 2001)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"taxondive(comm, dis, match.force = FALSE) taxa2dist(x, varstep = FALSE, check = TRUE, labels)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"comm Community data. dis Taxonomic distances among taxa comm. dist object symmetric square matrix. match.force Force matching column names comm labels dis. FALSE, matching happens dimensions differ, case species must identical order . x Classification table row species basic taxon, columns identifiers classification higher levels. varstep Vary step lengths successive levels relative proportional loss number distinct classes. check TRUE, remove redundant levels different rows constant rows regard row different basal taxon (species). FALSE levels retained basal taxa (species) also must coded variables (columns). get warning species coded, can ignore intention. labels labels attribute taxonomic distances. Row names used given. Species matched labels comm dis taxondive different dimensions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke & Warwick (1998, 2001) suggested several alternative indices taxonomic diversity distinctness. Two basic indices called taxonomic diversity (\\(\\Delta\\)) distinctness (\\(\\Delta^*\\)): equations give index value single site, summation goes species \\(\\) \\(j\\). \\(\\omega\\) taxonomic distances among taxa, \\(x\\) species abundances, \\(n\\) total abundance site. presence/absence data indices reduce index \\(\\Delta^+\\), index Clarke & Warwick (1998) also estimate standard deviation. Clarke & Warwick (2001) presented two new indices: \\(s\\Delta^+\\) product species richness \\(\\Delta^+\\), index variation taxonomic distinctness (\\(\\Lambda^+\\)) defined dis argument must species dissimilarities. must similar dissimilarities produced dist. customary integer steps taxonomic hierarchies, kind dissimilarities can used, phylogenetic trees genetic differences. , dis need taxonomic, species classifications can used. Function taxa2dist can produce suitable dist object classification table. species (basic taxon) corresponds row classification table, columns give classification different levels. varstep = FALSE successive levels separated equal steps, varstep = TRUE step length relative proportional decrease number classes (Clarke & Warwick 1999). check = TRUE, function removes classes distinct species combine species one class, assumes row presents distinct basic taxon. function scales distances longest path length taxa 100 (necessarily check = FALSE). Function plot.taxondive plots \\(\\Delta^+\\) Number species, together expectation approximate 2*sd limits. Function summary.taxondive finds \\(z\\) values significances Normal distribution \\(\\Delta^+\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function returns object class taxondive following items: Species Number species site. D, Dstar, Dplus, SDplus, Lambda \\(\\Delta\\), \\(\\Delta^*\\), \\(\\Delta^+\\), \\(s\\Delta^+\\) \\(\\Lambda^+\\) site. sd.Dplus Standard deviation \\(\\Delta^+\\). ED, EDstar, EDplus Expected values corresponding statistics. Function taxa2dist returns object class \"dist\", attribute \"steps\" step lengths successive levels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke, K.R & Warwick, R.M. (1998) taxonomic distinctness index statistical properties. Journal Applied Ecology 35, 523--531. Clarke, K.R. & Warwick, R.M. (1999) taxonomic distinctness measure biodiversity: weighting step lengths hierarchical levels. Marine Ecology Progress Series 184: 21--29. Clarke, K.R. & Warwick, R.M. (2001) biodiversity index applicable species lists: variation taxonomic distinctness. Marine Ecology Progress Series 216, 265--278.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"function still preliminary may change. scaling taxonomic dissimilarities influences results. multiply taxonomic distances (step lengths) constant, values Deltas multiplied constant, value \\(\\Lambda^+\\) square constant.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"## Preliminary: needs better data and some support functions data(dune) data(dune.taxon) # Taxonomic distances from a classification table with variable step lengths. taxdis <- taxa2dist(dune.taxon, varstep=TRUE) plot(hclust(taxdis), hang = -1) # Indices mod <- taxondive(dune, taxdis) mod #> Species Delta Delta* Lambda+ Delta+ S Delta+ #> 1 5.000 22.736 29.232 900.298 43.364 216.82 #> 2 10.000 51.046 55.988 822.191 56.232 562.32 #> 3 10.000 41.633 46.194 1025.471 62.869 628.69 #> 4 13.000 50.795 55.140 888.244 64.837 842.88 #> 5 14.000 63.498 67.856 715.393 69.211 968.95 #> 6 11.000 70.201 76.361 628.743 73.281 806.09 #> 7 13.000 61.605 66.187 679.337 69.918 908.94 #> 8 12.000 52.544 56.374 756.375 66.729 800.74 #> 9 13.000 50.526 54.108 849.448 63.205 821.67 #> 10 12.000 60.068 64.960 730.736 69.291 831.49 #> 11 9.000 69.589 77.740 404.609 77.803 700.23 #> 12 9.000 62.405 69.795 552.129 74.470 670.23 #> 13 10.000 47.316 53.842 536.429 66.657 666.57 #> 14 7.000 71.383 82.091 239.543 82.013 574.09 #> 15 8.000 68.564 77.097 334.889 79.010 632.08 #> 16 8.000 55.984 64.400 978.014 69.708 557.66 #> 17 7.000 53.913 60.222 632.990 59.286 415.00 #> 18 9.000 73.235 81.865 438.355 76.288 686.59 #> 19 9.000 68.727 76.091 336.364 78.636 707.73 #> 20 8.000 72.343 80.670 444.915 82.078 656.62 #> Expected 65.330 62.560 71.031 summary(mod) #> Delta Delta* Delta+ sd(Delta+) z(Delta+) Pr(>|z|) #> 1 22.7362 29.2322 43.3636 10.0499 -2.7530 0.005905 ** #> 2 51.0458 55.9878 56.2323 5.7727 -2.5636 0.010359 * #> 3 41.6334 46.1936 62.8687 5.7727 -1.4140 0.157360 #> 4 50.7952 55.1396 64.8368 4.5677 -1.3562 0.175047 #> 5 63.4979 67.8564 69.2108 4.2482 -0.4285 0.668251 #> 6 70.2011 76.3614 73.2810 5.3189 0.4229 0.672332 #> 7 61.6049 66.1871 69.9184 4.5677 -0.2437 0.807499 #> 8 52.5437 56.3743 66.7287 4.9215 -0.8743 0.381975 #> 9 50.5258 54.1079 63.2051 4.5677 -1.7134 0.086640 . #> 10 60.0680 64.9597 69.2906 4.9215 -0.3537 0.723567 #> 11 69.5894 77.7396 77.8030 6.3011 1.0747 0.282517 #> 12 62.4049 69.7949 74.4697 6.3011 0.5457 0.585290 #> 13 47.3158 53.8421 66.6566 5.7727 -0.7578 0.448548 #> 14 71.3834 82.0909 82.0130 7.7069 1.4249 0.154186 #> 15 68.5645 77.0970 79.0097 6.9314 1.1510 0.249714 #> 16 55.9840 64.3999 69.7078 6.9314 -0.1909 0.848565 #> 17 53.9134 60.2224 59.2857 7.7069 -1.5240 0.127501 #> 18 73.2349 81.8645 76.2879 6.3011 0.8342 0.404155 #> 19 68.7273 76.0909 78.6364 6.3011 1.2069 0.227457 #> 20 72.3431 80.6704 82.0779 6.9314 1.5937 0.111005 #> Expected 65.3302 62.5603 71.0313 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":null,"dir":"Reference","previous_headings":"","what":"Species tolerances and sample heterogeneities — tolerance","title":"Species tolerances and sample heterogeneities — tolerance","text":"Species tolerances sample heterogeneities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"tolerance(x, ...) # S3 method for cca tolerance(x, choices = 1:2, which = c(\"species\",\"sites\"), scaling = \"species\", useN2 = TRUE, hill = FALSE, ...) # S3 method for decorana tolerance(x, data, choices = 1:4, which = c(\"sites\", \"species\"), useN2 = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species tolerances and sample heterogeneities — tolerance","text":"Function compute species tolerances site heterogeneity measures unimodal ordinations (CCA & CA). Implements Eq 6.47 6.48 Canoco 4.5 Reference Manual (pages 178--179).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species tolerances and sample heterogeneities — tolerance","text":"Matrix tolerances/heterogeneities additional attributes: , scaling, N2, latter NA useN2 = FALSE N2 estimated.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species tolerances and sample heterogeneities — tolerance","text":"Gavin L. Simpson Jari Oksanen (decorana method).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species tolerances and sample heterogeneities — tolerance","text":"x object class \"cca\". choices numeric; ordination axes compute tolerances heterogeneities . Defaults axes 1 2. character; one \"species\" \"sites\", indicating whether species tolerances sample heterogeneities respectively computed. scaling character numeric; ordination scaling use. See scores.cca details. hill logical; scaling character, control whether Hill's scaling used (C)CA respectively. See scores.cca details. useN2 logical; bias tolerances / heterogeneities reduced via scaling Hill's N2? data Original input data used decorana. missing, function tries get data used decorana call. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ ., data = dune.env) ## defaults to species tolerances tolerance(mod) #> #> Species Tolerance #> #> Scaling: 2 #> #> CCA1 CCA2 #> Achimill 0.32968099 0.9241988 #> Agrostol 0.93670069 0.9238455 #> Airaprae 1.04694096 0.5889849 #> Alopgeni 0.72227472 0.3760138 #> Anthodor 1.00596787 0.8338212 #> Bellpere 0.32891011 0.9962790 #> Bromhord 0.27740999 0.6236199 #> Chenalbu 0.00000000 0.0000000 #> Cirsarve 0.00000000 0.0000000 #> Comapalu 0.47185632 0.8029414 #> Eleopalu 0.50344134 0.9384960 #> Elymrepe 0.35119963 0.5642491 #> Empenigr 0.00000000 0.0000000 #> Hyporadi 1.05840696 0.7523003 #> Juncarti 0.78397702 1.0686743 #> Juncbufo 0.69275956 0.6180830 #> Lolipere 0.51006235 0.8278177 #> Planlanc 0.36040676 0.6962294 #> Poaprat 0.58184277 0.9547104 #> Poatriv 0.78695928 0.7433503 #> Ranuflam 0.56576326 1.1725628 #> Rumeacet 0.58715663 0.8751491 #> Sagiproc 0.70922180 1.1153129 #> Salirepe 0.98530179 0.1077917 #> Scorautu 1.04355761 1.0724439 #> Trifprat 0.03045846 0.3651949 #> Trifrepe 1.21543364 0.9115613 #> Vicilath 0.24853962 0.6194084 #> Bracruta 1.03787313 1.0958331 #> Callcusp 0.57882025 1.0418623 #> ## sample heterogeneities for CCA axes 1:6 tolerance(mod, which = \"sites\", choices = 1:6) #> #> Sample Heterogeneity #> #> Scaling: 2 #> #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 1 0.2350112 0.8611530 1.7964571 0.4445499 2.4235732 0.5496289 #> 2 0.7100754 0.4136311 0.8151643 0.6311751 1.0467901 0.2514646 #> 3 0.5076492 0.7279717 0.8306874 0.5590739 0.3904998 0.9162012 #> 4 0.5955037 0.6901907 0.7931255 0.4873638 0.3966068 0.8700581 #> 5 0.6001048 0.5614830 1.1481560 0.3569604 0.4423909 1.9420043 #> 6 0.7272637 0.6867342 1.6068628 0.7778498 0.9187843 0.4938865 #> 7 0.6478967 0.4993262 0.7207318 0.3817131 0.4130713 0.7228173 #> 8 0.8563491 0.5498552 0.4217718 0.3370226 0.3013276 0.9535190 #> 9 0.5599722 0.7399384 0.4170304 1.0535541 1.4612437 0.7626183 #> 10 0.5210280 0.5806978 0.5856634 0.4174860 1.8559344 0.8890262 #> 11 0.4489323 0.6016877 0.3317371 1.8780211 1.2965939 2.1953737 #> 12 0.4948094 1.1084494 0.5226746 1.5064446 0.5703077 1.1561020 #> 13 0.6998985 0.8859365 0.4215474 0.8582272 0.5673698 0.5186678 #> 14 1.5925779 0.6747926 0.8927360 1.6798300 0.3480218 0.1575892 #> 15 1.0107648 0.5294221 1.0975629 1.7632888 0.2240900 0.3727240 #> 16 0.8031479 0.6058313 0.4871527 0.4227451 0.5341256 0.6990815 #> 17 0.5936276 1.5142792 0.5137979 1.0224938 1.7931775 0.6261853 #> 18 0.5689409 1.4067575 0.6398557 0.4983399 0.4364791 0.6590394 #> 19 1.1330387 0.9816332 1.1242398 0.7238920 0.5577662 0.7036044 #> 20 0.6737757 1.4458326 1.4380928 1.0959027 0.4142423 0.5332460 #> ## average should be 1 with scaling = \"sites\", hill = TRUE tol <- tolerance(mod, which = \"sites\", scaling = \"sites\", hill = TRUE, choices = 1:4) colMeans(tol) #> CCA1 CCA2 CCA3 CCA4 #> 1.059199 1.048823 1.000551 1.077612 apply(tol, 2, sd) #> CCA1 CCA2 CCA3 CCA4 #> 0.3174462 0.2793521 0.3714540 0.2681931 ## Rescaling tries to set all tolerances to 1 tol <- tolerance(decorana(dune)) colMeans(tol) #> DCA1 DCA2 DCA3 DCA4 #> 0.9817657 0.9249544 0.9444811 0.9821666 apply(tol, 2, sd) #> DCA1 DCA2 DCA3 DCA4 #> 0.1977777 0.3204058 0.2646872 0.1210543"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":null,"dir":"Reference","previous_headings":"","what":"Functional Diversity and Community Distances from Species Trees — treedive","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Functional diversity defined total branch length trait dendrogram connecting species, excluding unnecessary root segments tree (Petchey Gaston 2006). Tree distance increase total branch length combining two sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"treedive(comm, tree, match.force = TRUE, verbose = TRUE) treeheight(tree) treedist(x, tree, relative = TRUE, match.force = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"comm, x Community data frame matrix. tree dendrogram treedive must species (columns). match.force Force matching column names data (comm, x) labels tree. FALSE, matching happens dimensions differ (warning message). order data must match order tree matching names done. verbose Print diagnostic messages warnings. relative Use distances relative height combined tree. ... arguments passed functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Function treeheight finds sum lengths connecting segments dendrogram produced hclust, dendrogram can coerced correct type using .hclust. applied clustering species traits, measure functional diversity (Petchey Gaston 2002, 2006), applied phylogenetic trees phylogenetic diversity. Function treedive finds treeheight site (row) community matrix. function uses subset dendrogram species occur site, excludes tree root needed connect species (Petchey Gaston 2006). subset dendrogram found first calculating cophenetic distances input dendrogram, reconstructing dendrogram subset cophenetic distance matrix species occurring site. Diversity 0 one species, NA empty communities. Function treedist finds dissimilarities among trees. Pairwise dissimilarity two trees found combining species common tree seeing much tree height shared much unique. relative = FALSE dissimilarity defined \\(2 (\\cup B) - - B\\), \\(\\) \\(B\\) heights component trees \\(\\cup B\\) height combined tree. relative = TRUE dissimilarity \\((2(\\cup B)--B)/(\\cup B)\\). Although latter formula similar Jaccard dissimilarity (see vegdist, designdist), range \\(0 \\ldots 1\\), since combined tree can add new root. two zero-height trees combined tree zero height, relative index attains maximum value \\(2\\). dissimilarity zero combined zero-height tree. functions need dendrogram species traits phylogenies input. species traits contain factor ordered factor variables, recommended use Gower distances mixed data (function daisy package cluster), usually recommended clustering method UPGMA (method = \"average\" function hclust) (Podani Schmera 2006). Phylogenetic trees can changed dendrograms using function .hclust.phylo ape package. possible analyse non-randomness tree diversity using oecosimu. needs specifying adequate Null model, results change choice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"vector diversity values single tree height, dissimilarity structure inherits dist can used similarly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Lozupone, C. Knight, R. 2005. UniFrac: new phylogenetic method comparing microbial communities. Applied Environmental Microbiology 71, 8228--8235. Petchey, O.L. Gaston, K.J. 2002. Functional diversity (FD), species richness community composition. Ecology Letters 5, 402--411. Petchey, O.L. Gaston, K.J. 2006. Functional diversity: back basics looking forward. Ecology Letters 9, 741--758. Podani J. Schmera, D. 2006. dendrogram-based methods functional diversity. Oikos 115, 179--185.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"## There is no data set on species properties yet, and we demonstrate ## the methods using phylogenetic trees data(dune) data(dune.phylodis) cl <- hclust(dune.phylodis) treedive(dune, cl) #> forced matching of 'tree' labels and 'comm' names #> 1 2 3 4 5 6 7 8 #> 384.0913 568.8791 1172.9455 1327.9317 1426.9067 1391.1628 1479.5062 1523.0792 #> 9 10 11 12 13 14 15 16 #> 1460.0423 1316.4832 1366.9960 1423.5582 895.1120 1457.2705 1505.9501 1187.5165 #> 17 18 19 20 #> 517.6920 1394.5162 1470.4671 1439.5571 ## Significance test using Null model communities. ## The current choice fixes numbers of species and picks species ## proportionally to their overall frequency oecosimu(dune, treedive, \"r1\", tree = cl, verbose = FALSE) #> Warning: nullmodel transformed 'comm' to binary data #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = treedive, method = \"r1\", tree = #> cl, verbose = FALSE) #> #> nullmodel method ‘r1’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> 1 384.09 -1.238698 773.72 383.47 628.63 1237.7 0.11 #> 2 568.88 -2.353877 1222.61 663.93 1337.50 1561.4 0.01 ** #> 3 1172.95 -0.136353 1210.76 679.53 1327.94 1581.5 0.63 #> 4 1327.93 -0.418624 1427.25 886.63 1505.15 1733.1 0.45 #> 5 1426.91 -0.322462 1496.97 926.98 1559.25 1734.4 0.43 #> 6 1391.16 0.254066 1326.70 753.19 1408.30 1634.7 0.87 #> 7 1479.51 0.280433 1407.90 880.79 1476.23 1735.1 0.99 #> 8 1523.08 0.567886 1389.63 792.74 1428.01 1655.3 0.65 #> 9 1460.04 0.065245 1448.21 899.67 1472.92 1654.4 0.87 #> 10 1316.48 -0.192836 1362.43 708.60 1441.63 1608.0 0.47 #> 11 1367.00 0.768160 1150.05 621.65 1266.73 1486.4 0.43 #> 12 1423.56 1.101039 1088.13 622.24 1208.95 1474.2 0.25 #> 13 895.11 -0.972053 1188.10 656.34 1325.53 1558.5 0.61 #> 14 1457.27 1.477151 980.91 491.11 1101.65 1459.3 0.07 . #> 15 1505.95 1.532058 1052.35 575.91 1171.88 1460.7 0.03 * #> 16 1187.52 0.306608 1089.58 543.75 1238.26 1490.7 0.89 #> 17 517.69 -1.446211 959.50 504.26 1101.16 1367.8 0.09 . #> 18 1394.52 0.961778 1099.50 618.22 1249.49 1493.8 0.33 #> 19 1470.47 1.134795 1121.18 564.70 1240.26 1562.8 0.15 #> 20 1439.56 1.160566 1104.59 629.08 1244.20 1467.4 0.17 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Phylogenetically ordered community table dtree <- treedist(dune, cl) tabasco(dune, hclust(dtree), cl) ## Use tree distances in distance-based RDA dbrda(dtree ~ 1) #> Call: dbrda(formula = dtree ~ 1) #> #> Inertia Rank RealDims #> Total 2.183 #> Unconstrained 2.183 19 10 #> Inertia is squared Treedist distance #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.1971 0.4546 0.2967 0.1346 0.1067 0.0912 0.0391 0.0190 #> (Showing 8 of 19 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":null,"dir":"Reference","previous_headings":"","what":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis find Tsallis diversities scale corresponding evenness measures. Function tsallisaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"tsallis(x, scales = seq(0, 2, 0.2), norm = FALSE, hill = FALSE) tsallisaccum(x, scales = seq(0, 2, 0.2), permutations = 100, raw = FALSE, subset, ...) # S3 method for tsallisaccum persp(x, theta = 220, phi = 15, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"x Community data matrix plotting object. scales Scales Tsallis diversity. norm Logical, TRUE diversity values normalized maximum (diversity value equiprobability conditions). hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. subset logical expression indicating sites (rows) keep: missing values taken FALSE. theta, phi angles defining viewing direction. theta gives azimuthal direction phi colatitude. col Colours used surface. zlim Limits vertical axis. ... arguments passed tsallis graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis diversity (also equivalent Patil Taillie diversity) one-parametric generalised entropy function, defined : $$H_q = \\frac{1}{q-1} (1-\\sum_{=1}^S p_i^q)$$ \\(q\\) scale parameter, \\(S\\) number species sample (Tsallis 1988, Tothmeresz 1995). diversity concave \\(q>0\\), non-additive (Keylock 2005). \\(q=0\\) gives number species minus one, \\(q\\) tends 1 gives Shannon diversity, \\(q=2\\) gives Simpson index (see function diversity). norm = TRUE, tsallis gives values normalized maximum: $$H_q(max) = \\frac{S^{1-q}-1}{1-q}$$ \\(S\\) number species. \\(q\\) tends 1, maximum defined \\(ln(S)\\). hill = TRUE, tsallis gives Hill numbers (numbers equivalents, see Jost 2007): $$D_q = (1-(q-1) H)^{1/(1-q)}$$ Details plotting methods accumulating values can found help pages functions renyi renyiaccum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis returns data frame selected indices. Function tsallisaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487. Tothmeresz, B. (1995) Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290. Patil, G. P. Taillie, C. (1982) Diversity concept measurement. J. . Stat. Ass. 77, 548--567. Keylock, C. J. (2005) Simpson diversity Shannon-Wiener index special cases generalized entropy. Oikos 109, 203--207. Jost, L (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Péter Sólymos, solymos@ualberta.ca, based code Roeland Kindt Jari Oksanen written renyi","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) x1 <- tsallis(BCI[i,]) x1 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 #> 8 87 39.98200 19.77567 10.60391 6.181934 3.908381 2.659907 1.928610 1.473408 #> 10 93 41.50369 20.10059 10.63988 6.164157 3.889803 2.648176 1.922207 1.470121 #> 5 100 44.20598 21.17686 11.08126 6.350040 3.969940 2.683412 1.937940 1.477217 #> 17 92 40.88788 19.62643 10.29183 5.926929 3.736897 2.553271 1.864739 1.435868 #> 19 108 47.27018 22.32602 11.49839 6.492802 4.013094 2.692466 1.936696 1.473828 #> 13 92 41.98171 20.60707 10.96297 6.342501 3.982373 2.694847 1.945425 1.481616 #> 46 85 38.71044 19.06049 10.22455 5.987209 3.810489 2.611431 1.904913 1.461976 #> 16 92 41.45601 20.19256 10.71573 6.210870 3.916821 2.663870 1.931567 1.475861 #> 36 91 40.76595 19.77916 10.48193 6.081803 3.846109 2.625076 1.910150 1.463936 #> 7 81 37.47840 18.73414 10.17269 6.004908 3.836811 2.631757 1.918091 1.469869 #> 24 94 42.55654 20.75558 10.99284 6.343926 3.979427 2.692702 1.944478 1.481404 #> 22 90 40.16368 19.41416 10.25630 5.939677 3.755413 2.566788 1.872586 1.439722 #> 1.8 2 #> 8 1.173986 0.9671998 #> 10 1.172341 0.9663808 #> 5 1.175553 0.9678267 #> 17 1.152122 0.9545126 #> 19 1.172481 0.9655820 #> 13 1.178033 0.9692075 #> 46 1.168556 0.9646728 #> 16 1.175935 0.9686598 #> 36 1.169232 0.9648567 #> 7 1.173083 0.9672014 #> 24 1.178152 0.9694268 #> 22 1.153640 0.9548316 diversity(BCI[i,],\"simpson\") == x1[[\"2\"]] #> 8 10 5 17 19 13 46 16 36 7 24 22 #> TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE plot(x1) x2 <- tsallis(BCI[i,],norm=TRUE) x2 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 #> 8 1 0.9154260 0.8674322 0.8491531 0.8535754 0.8729254 0.8992501 0.9258854 #> 10 1 0.9001004 0.8450759 0.8255446 0.8324355 0.8561634 0.8872568 0.9180261 #> 5 1 0.9038026 0.8502581 0.8308757 0.8372445 0.8602030 0.8904867 0.9204832 #> 17 1 0.8945733 0.8308281 0.8026305 0.8032747 0.8244489 0.8566981 0.8913214 #> 19 1 0.9079031 0.8537683 0.8315799 0.8347890 0.8554245 0.8846675 0.9147428 #> 13 1 0.9185048 0.8723405 0.8549707 0.8595970 0.8786069 0.9042012 0.9298883 #> 46 1 0.9032439 0.8485357 0.8278693 0.8331492 0.8554539 0.8856795 0.9162035 #> 16 1 0.9070033 0.8547937 0.8356888 0.8417570 0.8641446 0.8938075 0.9232641 #> 36 1 0.8998669 0.8431295 0.8216927 0.8272622 0.8505725 0.8820846 0.9137994 #> 7 1 0.9094899 0.8600339 0.8427936 0.8492653 0.8706729 0.8985570 0.9261475 #> 24 1 0.9149445 0.8667059 0.8486326 0.8536824 0.8738548 0.9008891 0.9279024 #> 22 1 0.8945746 0.8334045 0.8082215 0.8109045 0.8325271 0.8637842 0.8965990 #> 1.6 1.8 2 #> 8 0.9486740 0.9660687 0.9783170 #> 10 0.9438798 0.9632975 0.9767719 #> 5 0.9456409 0.9644767 0.9775050 #> 17 0.9223038 0.9469046 0.9648877 #> 19 0.9406572 0.9605041 0.9745226 #> 13 0.9516896 0.9681999 0.9797424 #> 46 0.9422698 0.9621112 0.9760219 #> 16 0.9479928 0.9664763 0.9791887 #> 36 0.9407648 0.9611954 0.9754595 #> 7 0.9493990 0.9669335 0.9791421 #> 24 0.9507028 0.9678509 0.9797399 #> 22 0.9256369 0.9486075 0.9654409 plot(x2) mod1 <- tsallisaccum(BCI[i,]) plot(mod1, as.table=TRUE, col = c(1, 2, 2)) persp(mod1) mod2 <- tsallisaccum(BCI[i,], norm=TRUE) persp(mod2,theta=100,phi=30)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and environment in lichen pastures — varespec","title":"Vegetation and environment in lichen pastures — varespec","text":"varespec data frame 24 rows 44 columns. Columns estimated cover values 44 species. variable names formed scientific names, self explanatory anybody familiar vegetation type. varechem data frame 24 rows 14 columns, giving soil characteristics sites varespec data frame. chemical measurements obvious names. Baresoil gives estimated cover bare soil, Humdepth thickness humus layer.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varechem) data(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Vegetation and environment in lichen pastures — varespec","text":"Väre, H., Ohtonen, R. Oksanen, J. (1995) Effects reindeer grazing understorey vegetation dry Pinus sylvestris forests. Journal Vegetation Science 6, 523--530.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varespec) data(varechem)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":null,"dir":"Reference","previous_headings":"","what":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"function partitions variation community data community dissimilarities respect two, three, four explanatory tables, using adjusted \\(R^2\\) redundancy analysis ordination (RDA) distance-based redundancy analysis. response single vector, partitioning partial regression. Collinear variables explanatory tables removed prior partitioning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"varpart(Y, X, ..., data, chisquare = FALSE, transfo, scale = FALSE, add = FALSE, sqrt.dist = FALSE, permutations) # S3 method for varpart summary(object, ...) showvarparts(parts, labels, bg = NULL, alpha = 63, Xnames, id.size = 1.2, ...) # S3 method for varpart234 plot(x, cutoff = 0, digits = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Y Data frame matrix containing response data table dissimilarity structure inheriting dist. community ecology, table often site--species table dissimilarity object. X Two four explanatory models, variables tables. can defined three alternative ways: (1) one-sided model formulae beginning ~ defining model, (2) name single numeric factor variable, (3) name matrix numeric data frame numeric factor variables. model formulae can factors, interaction terms transformations variables. names variables model formula found data frame given data argument, found , user environment. Single variables, data frames matrices found user environment. entries till next argument (data transfo) interpreted explanatory models, names extra arguments abbreviated omitted. ... parameters passed functions. NB, arguments dots abbreviated must spelt completely. data data frame variables used formulae X. chisquare Partition Chi-square inertia Correspondence Analysis (cca). transfo Transformation Y (community data) using decostand. alternatives decostand can used, preserving Euclidean metric include \"hellinger\", \"chi.square\", \"total\", \"norm\". Ignored Y dissimilarities. scale columns Y standardized unit variance. Ignored Y dissimilarities. add Add constant non-diagonal values euclidify dissimilarities (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. argument effect Y dissimilarities. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. NB., argument name abbreviated. argument effect Y dissimilarities. permutations chisquare = TRUE, adjusted \\(R^2\\) estimated permutations, paramater can list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parts Number explanatory tables (circles) displayed. labels Labels used displayed fractions. Default use letters printed output. bg Fill colours circles ellipses. alpha Transparency fill colour. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. Xnames Names sources variation. Default names X1, X2, X3 X4. Xnames=NA, Xnames=NULL Xnames=\"\" produce names. names can changed names. often best use short names. id.size numerical value giving character expansion factor names circles ellipses. x, object varpart result. cutoff values cutoff displayed. digits number significant digits; number decimal places least one higher.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"functions partition variation Y components accounted two four explanatory tables combined effects. Y multicolumn data frame matrix, partitioning based redundancy analysis (RDA, see rda) constrained correspondence analysis chisquare = TRUE (CCA, see cca). Y single variable, partitioning based linear regression. Y dissimilarities, decomposition based distance-based redundancy analysis (db-RDA, see dbrda) following McArdle & Anderson (2001). input dissimilarities must compatible results dist. Vegan functions vegdist, designdist, raupcrick betadiver produce objects, many dissimilarity functions R packages. Partitioning made squared dissimilarities analogously using variance rectangular data -- unless sqrt.dist = TRUE specified. function primarily uses adjusted \\(R^2\\) assess partitions explained explanatory tables combinations (see RsquareAdj), unbiased method (Peres-Neto et al., 2006). raw \\(R^2\\) basic fractions also displayed, biased estimates variation explained explanatory table. correspondence analysis (chisquare = TRUE), adjusted \\(R^2\\) found permutation vary repeated analyses. identifiable fractions designated lower case alphabets. meaning symbols can found separate document (use browseVignettes(\"vegan\")), can displayed graphically using function showvarparts. fraction testable can directly expressed RDA db-RDA model. cases printed output also displays corresponding RDA model using notation explanatory tables | conditions (partialled ; see rda details). Although single fractions can testable, mean fractions simultaneously can tested, since number testable fractions higher number estimated models. non-testable components found differences testable components. testable components permutation variance correspondence analysis (chisquare = TRUE), non-testable components even higher variance. abridged explanation alphabetic symbols individual fractions follows, computational details checked vignette (readable browseVignettes(\"vegan\")) source code. two explanatory tables, fractions explained uniquely two tables [] [b], joint effect [c]. three explanatory tables, fractions explained uniquely three tables [] [c], joint fractions two tables [d] [f], joint fraction three tables [g]. four explanatory tables, fractions explained uniquely four tables [] [d], joint fractions two tables [e] [j], joint fractions three variables [k] [n], joint fraction four tables [o]. summary give overview unique overall contribution group variables. overall contribution (labelled “Contributed”) consists unique contribution variable equal shares fraction variable contributes. summary tabulates fraction divided variables, contributed component sum divided fractions. summary based idea Lai et al. (2022), similar output rdacca.hp package. plot function displays Venn diagram labels intersection (individual fraction) adjusted R squared higher cutoff. helper function showvarpart displays fraction labels. circles ellipses labelled short default names names defined user argument Xnames. Longer explanatory file names can written varpart output plot follows: use option Xnames=NA, add new names using text function. bit fiddling coordinates (see locator) character size allow users place names reasonably short lengths varpart plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Function varpart returns object class \"varpart\" items scale transfo (can missing) hold information standardizations, tables contains names explanatory tables, call function call. function varpart calls function varpart2, varpart3 varpart4 return object class \"varpart234\" saves result item part. items object : SS.Y Sum squares matrix Y. n Number observations (rows). nsets Number explanatory tables bigwarning Warnings collinearity. fract Basic fractions estimated constrained models. indfract Individual fractions possible subsections Venn diagram (see showvarparts). contr1 Fractions can found conditioning single explanatory table models three four explanatory tables. contr2 Fractions can found conditioning two explanatory tables models four explanatory tables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"fraction-data-frames","dir":"Reference","previous_headings":"","what":"Fraction Data Frames","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Items fract, indfract, contr1 contr2 data frames items: Df: Degrees freedom numerator \\(F\\)-statistic fraction. R.square: Raw \\(R^2\\). calculated fract NA items. Adj.R.square: Adjusted \\(R^2\\). Testable: fraction can expressed (partial) RDA model, directly Testable, field TRUE. case fraction label also gives specification testable RDA model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"() References variation partitioning Borcard, D., P. Legendre & P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045--1055. Lai J., Y. Zou, J. Zhang & P. Peres-Neto. 2022. Generalizing hierarchical variation partitioning multiple regression canonical analysis using rdacca.hp R package. Methods Ecology Evolution, 13: 782--788. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. (b) Reference transformations species data Legendre, P. E. D. Gallagher. 2001. Ecologically meaningful transformations ordination species data. Oecologia 129: 271--280. (c) Reference adjustment bimultivariate redundancy statistic Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87: 2614--2625. (d) References partitioning dissimilarities Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290-297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal, Canada. developed Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"can use command browseVignettes(\"vegan\") display document presents Venn diagrams showing fraction names partitioning variation Y respect 2, 3, 4 tables explanatory variables, well equations used variation partitioning. functions frequently give negative estimates variation. Adjusted \\(R^2\\) can negative fraction; unadjusted \\(R^2\\) testable fractions variances non-negative. Non-testable fractions found directly, subtracting different models, subtraction results can negative. fractions orthogonal, linearly independent, complicated nonlinear dependencies can cause negative non-testable fractions. fraction can negative non-Euclidean dissimilarities underlying db-RDA model can yield negative eigenvalues (see dbrda). negative eigenvalues underlying analysis can avoided arguments sqrt.dist add similar effect dbrda: square roots several dissimilarities negative eigenvalues, negative eigenvalues produced Lingoes Cailliez adjustment, effect add random variation dissimilarities. simplified, fast version RDA, CCA adn dbRDA used (functions simpleRDA2, simpleCCA simpleDBRDA). actual calculations done functions varpart2 varpart4, intended called directly user.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) # Two explanatory data frames -- Hellinger-transform Y mod <- varpart(mite, mite.env, mite.pcnm, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = mite.env, mite.pcnm, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.52650 0.43670 TRUE #> [b+c] = X2 22 0.62300 0.44653 TRUE #> [a+b+c] = X1+X2 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09141 TRUE #> [b] = X2|X1 22 0.10124 TRUE #> [c] 0 0.34530 FALSE #> [d] = Residuals 0.46206 FALSE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0914 0.264 mite.env #> X2 0.1012 0.274 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 #> [a] 0.0914 #> [b] 0.1012 #> [c] 0.1726 0.1726 ## Use fill colours showvarparts(2, bg = c(\"hotpink\",\"skyblue\")) plot(mod, bg = c(\"hotpink\",\"skyblue\")) ## Test fraction [a] using partial RDA, '~ .' in formula tells to use ## all variables of data mite.env. aFrac <- rda(decostand(mite, \"hel\"), mite.env, mite.pcnm) anova(aFrac) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(X = decostand(mite, \"hel\"), Y = mite.env, Z = mite.pcnm) #> Df Variance F Pr(>F) #> Model 11 0.053592 1.8453 0.001 *** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## RsquareAdj gives the same result as component [a] of varpart RsquareAdj(aFrac) #> $r.squared #> [1] 0.1359251 #> #> $adj.r.squared #> [1] 0.09140797 #> ## Partition Bray-Curtis dissimilarities varpart(vegdist(mite), mite.env, mite.pcnm) #> #> Partition of squared Bray distance in dbRDA #> #> Call: varpart(Y = vegdist(mite), X = mite.env, mite.pcnm) #> #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 14.696 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.50512 0.41127 TRUE #> [b+c] = X2 22 0.60144 0.41489 TRUE #> [a+b+c] = X1+X2 33 0.74631 0.51375 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09887 TRUE #> [b] = X2|X1 22 0.10249 TRUE #> [c] 0 0.31240 FALSE #> [d] = Residuals 0.48625 FALSE #> --- #> Use function ‘dbrda’ to test significance of fractions of interest ## Three explanatory tables with formula interface mod <- varpart(mite, ~ SubsDens + WatrCont, ~ Substrate + Shrub + Topo, mite.pcnm, data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm, data = mite.env, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm #> #> No. of explanatory tables: 3 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [a+d+f+g] = X1 2 0.32677 0.30667 TRUE #> [b+d+e+g] = X2 9 0.40395 0.31454 TRUE #> [c+e+f+g] = X3 22 0.62300 0.44653 TRUE #> [a+b+d+e+f+g] = X1+X2 11 0.52650 0.43670 TRUE #> [a+c+d+e+f+g] = X1+X3 24 0.67372 0.49970 TRUE #> [b+c+d+e+f+g] = X2+X3 31 0.72400 0.49884 TRUE #> [a+b+c+d+e+f+g] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3 2 0.03910 TRUE #> [b] = X2 | X1+X3 9 0.03824 TRUE #> [c] = X3 | X1+X2 22 0.10124 TRUE #> [d] 0 0.01407 FALSE #> [e] 0 0.09179 FALSE #> [f] 0 0.08306 FALSE #> [g] 0 0.17045 FALSE #> [h] = Residuals 0.46206 FALSE #> Controlling 1 table X #> [a+d] = X1 | X3 2 0.05317 TRUE #> [a+f] = X1 | X2 2 0.12216 TRUE #> [b+d] = X2 | X3 9 0.05231 TRUE #> [b+e] = X2 | X1 9 0.13003 TRUE #> [c+e] = X3 | X1 22 0.19303 TRUE #> [c+f] = X3 | X2 22 0.18429 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.144 ~SubsDens + WatrCont #> X2 0.0382 0.148 ~Substrate + Shrub + Topo #> X3 0.1012 0.245 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 X3 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10124 #> [d] 0.00703 0.00703 #> [e] 0.04590 0.04590 #> [f] 0.04153 0.04153 #> [g] 0.05682 0.05682 0.05682 showvarparts(3, bg=2:4) plot(mod, bg=2:4) ## Use RDA to test fraction [a] ## Matrix can be an argument in formula rda.result <- rda(decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) anova(rda.result) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) #> Df Variance F Pr(>F) #> Model 2 0.013771 2.6079 0.005 ** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Four explanatory tables mod <- varpart(mite, ~ SubsDens + WatrCont, ~Substrate + Shrub + Topo, mite.pcnm[,1:11], mite.pcnm[,12:22], data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm[, 1:11], mite.pcnm[, 12:22], data = mite.env, transfo = #> \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm[, 1:11] #> X4: mite.pcnm[, 12:22] #> #> No. of explanatory tables: 4 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [aeghklno] = X1 2 0.32677 0.30667 TRUE #> [befiklmo] = X2 9 0.40395 0.31454 TRUE #> [cfgjlmno] = X3 11 0.53231 0.44361 TRUE #> [dhijkmno] = X4 11 0.09069 -0.08176 TRUE #> [abefghiklmno] = X1+X2 11 0.52650 0.43670 TRUE #> [acefghjklmno] = X1+X3 13 0.59150 0.49667 TRUE #> [adeghijklmno] = X1+X4 13 0.40374 0.26533 TRUE #> [bcefgijklmno] = X2+X3 20 0.63650 0.48813 TRUE #> [bdefhijklmno] = X2+X4 20 0.53338 0.34292 TRUE #> [cdfghijklmno] = X3+X4 22 0.62300 0.44653 TRUE #> [abcefghijklmno] = X1+X2+X3 22 0.67947 0.52944 TRUE #> [abdefghijklmno] = X1+X2+X4 22 0.61553 0.43557 TRUE #> [acdefghijklmno] = X1+X3+X4 24 0.67372 0.49970 TRUE #> [bcdefghijklmno] = X2+X3+X4 31 0.72400 0.49884 TRUE #> [abcdefghijklmno] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3+X4 2 0.03910 TRUE #> [b] = X2 | X1+X3+X4 9 0.03824 TRUE #> [c] = X3 | X1+X2+X4 11 0.10237 TRUE #> [d] = X4 | X1+X2+X3 11 0.00850 TRUE #> [e] 0 0.01407 FALSE #> [f] 0 0.13200 FALSE #> [g] 0 0.05355 FALSE #> [h] 0 0.00220 FALSE #> [i] 0 -0.00547 FALSE #> [j] 0 -0.00963 FALSE #> [k] 0 -0.00231 FALSE #> [l] 0 0.24037 FALSE #> [m] 0 -0.03474 FALSE #> [n] 0 0.02730 FALSE #> [o] 0 -0.06761 FALSE #> [p] = Residuals 0 0.46206 FALSE #> Controlling 2 tables X #> [ae] = X1 | X3+X4 2 0.05317 TRUE #> [ag] = X1 | X2+X4 2 0.09265 TRUE #> [ah] = X1 | X2+X3 2 0.04131 TRUE #> [be] = X2 | X3+X4 9 0.05231 TRUE #> [bf] = X2 | X1+X4 9 0.17024 TRUE #> [bi] = X2 | X1+X3 9 0.03277 TRUE #> [cf] = X3 | X1+X4 11 0.23437 TRUE #> [cg] = X3 | X2+X4 11 0.15592 TRUE #> [cj] = X3 | X1+X2 11 0.09274 TRUE #> [dh] = X4 | X2+X3 11 0.01071 TRUE #> [di] = X4 | X1+X3 11 0.00303 TRUE #> [dj] = X4 | X1+X2 11 -0.00113 TRUE #> Controlling 1 table X #> [aghn] = X1 | X2 2 0.12216 TRUE #> [aehk] = X1 | X3 2 0.05306 TRUE #> [aegl] = X1 | X4 2 0.34709 TRUE #> [bfim] = X2 | X1 9 0.13003 TRUE #> [beik] = X2 | X3 9 0.04452 TRUE #> [befl] = X2 | X4 9 0.42468 TRUE #> [cfjm] = X3 | X1 11 0.19000 TRUE #> [cgjn] = X3 | X2 11 0.17359 TRUE #> [cfgl] = X3 | X4 11 0.52830 TRUE #> [dijm] = X4 | X1 11 -0.04134 TRUE #> [dhjn] = X4 | X2 11 0.02837 TRUE #> [dhik] = X4 | X3 11 0.00292 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.1456 ~SubsDens + WatrCont #> X2 0.0382 0.1594 ~Substrate + Shrub + Topo #> X3 0.1024 0.2511 mite.pcnm[, 1:11] #> X4 0.0085 -0.0181 mite.pcnm[, 12:22] #> #> Contributions of fractions to sets: #> #> X1 X2 X3 X4 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10237 #> [d] 0.00850 #> [e] 0.00703 0.00703 #> [f] 0.06600 0.06600 #> [g] 0.02678 0.02678 #> [h] 0.00110 0.00110 #> [i] -0.00274 -0.00274 #> [j] -0.00482 -0.00482 #> [k] -0.00077 -0.00077 -0.00077 #> [l] 0.08012 0.08012 0.08012 #> [m] -0.01158 -0.01158 -0.01158 #> [n] 0.00910 0.00910 0.00910 #> [o] -0.01690 -0.01690 -0.01690 -0.01690 plot(mod, bg=2:5) ## Show values for all partitions by putting 'cutoff' low enough: plot(mod, cutoff = -Inf, cex = 0.7, bg=2:5)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":null,"dir":"Reference","previous_headings":"","what":"Defunct Functions in Package vegan — vegan-defunct","title":"Defunct Functions in Package vegan — vegan-defunct","text":"functions variables listed longer part vegan longer needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Defunct Functions in Package vegan — vegan-defunct","text":"","code":"## defunct in vegan 2.7-0 (development 2.6-7) adonis(formula, data, permutations = 999, method = \"bray\", strata = NULL, contr.unordered = \"contr.sum\", contr.ordered = \"contr.poly\", parallel = getOption(\"mc.cores\"), ...) ## defunct in vegan 2.6-0 as.mlm(x) humpfit(mass, spno, family = poisson, start) vegandocs(doc = c(\"NEWS\", \"ONEWS\", \"FAQ-vegan\", \"intro-vegan\", \"diversity-vegan\", \"decision-vegan\", \"partitioning\", \"permutations\")) ## defunct in vegan 2.5-0 commsimulator(x, method, thin=1) ## defunct in vegan 2.4-0 # S3 method for adonis density(x, ...) # S3 method for vegandensity plot(x, main = NULL, xlab = NULL, ylab = \"Density\", type = \"l\", zero.line = TRUE, obs.line = TRUE, ...) # S3 method for adonis densityplot(x, data, xlab = \"Null\", ...) ## defunct in vegan 2.2-0 metaMDSrotate(object, vec, na.rm = FALSE, ...) ## defunct in vegan 2.0-0 getNumObs(object, ...) permuted.index2(n, control = permControl())"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Defunct Functions in Package vegan — vegan-defunct","text":"adonis2 replaces adonis extended functionality completely new internal design. shared arguments adonis similar adonis2, arguments contr.unordered contr.ordered can set contrasts within adonis. .mlm function replaced set functions can find statistics directly ordination result object: see hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca, vcov.cca. Function humpfit transferred natto package still available https://github.com/jarioksa/natto/. R functions news used read vegan NEWS (news(package = \"vegan\")), browseVignettes better tool reading vignettes vegandocs. Function commsimulator replaced make.commsim defines Null models, functions nullmodel simulate.nullmodel check input data generate Null model communities. deprecated density densityplot methods replaced similar methods permustats. permustats offers powerful analysis tools permutations, including summary.permustats giving \\(z\\) values (.k.. standardized effect sizes, SES), Q-Q plots (qqnorm.permustats, qqmath.permustats). Function metaMDSrotate replaced MDSrotate can handle monoMDS results addition metaMDS. permutation functions moved permute package, documented . permute package replaces permuted.index permuted.index2 shuffle getNumObs specific nobs-methods.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions in vegan package — vegan-deprecated","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"functions provided compatibility older versions vegan , may defunct soon next release.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"","code":"## moved to vegan3d package: install from CRAN orditkplot(...) ## use toCoda instead as.mcmc.oecosimu(x) as.mcmc.permat(x)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"x object transformed. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"orditkplot moved vegan3d (version 1.3-0). Install package CRAN use old way. .mcmc functions replaced toCoda.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":null,"dir":"Reference","previous_headings":"","what":"Internal vegan functions — vegan-internal","title":"Internal vegan functions — vegan-internal","text":"Internal vegan functions intended called directly, within functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Internal vegan functions — vegan-internal","text":"","code":"ordiParseFormula(formula, data, xlev = NULL, na.action = na.fail, subset = NULL, X) ordiTerminfo(d, data) ordiNAexclude(x, excluded) ordiNApredict(omit, x) ordiArgAbsorber(..., shrink, origin, scaling, triangular, display, choices, const, truemean, FUN) centroids.cca(x, mf, wt) getPermuteMatrix(perm, N, strata = NULL) howHead(x, ...) pasteCall(call, prefix = \"Call:\") veganCovEllipse(cov, center = c(0, 0), scale = 1, npoints = 100) veganMahatrans(x, s2, tol = sqrt(.Machine$double.eps)) hierParseFormula(formula, data) GowerDblcen(x, na.rm = TRUE) addLingoes(d) addCailliez(d)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Internal vegan functions — vegan-internal","text":"description intended vegan developers: functions intended users, used within functions. general, functions exported namespace, must use get ::: directly call functions. ordiParseFormula returns list three matrices (dependent variables, model.matrix constraints conditions, possibly NULL) needed constrained ordination. Argument xlev passed model.frame. left-hand-side already evaluated calling code, can given argument X re-evaluated. ordiTermInfo finds term information constrained ordination described cca.object. ordiNAexclude implements na.action = na.exclude constrained ordination finding WA scores CCA components site scores unconstrained component excluded rows observations. Function ordiNApredict pads result object WA scores similarly napredict. ordiArgAbsorber absorbs arguments scores function vegan cause superfluous warnings graphical function FUN. implement scores functions new arguments, update ordiArgAbsorber. centroids.cca finds weighted centroids variables. getPermuteMatrix interprets user input returns permutation matrix row gives indices observations permutation. input perm can single number number simple permutations, result defining permutation scheme permutation matrix. howHead formats permutation scheme display. formatting compact one used print permute package, shows non-default choices. output normally used printing results vegan permutations. pasteCall prints function call nicely wrapped Sweave output. veganCovEllipse finds coordinates drawing covariance ellipse. veganMahatrans transforms data matrix Euclidean distances Mahalanobis distances. input data x must matrix centred columns, s2 covariance matrix. s2 given, covariance matrix found x within function. hierParseFormula returns list one matrix (left hand side) model frame factors representing hierarchy levels (right hand side) used adipart, multipart hiersimu. GowerDblcen performs Gower double centring matrix dissimilarities. Similar function earlier available compiled code stats, part official API, therefore poorer replacement. addLingoes addCailliez find constant added non-diagonal (squared) dissimilarities make eigenvalues non-negative Principal Co-ordinates Analysis (wcmdscale, dbrda, capscale). Function cmdscale implements Cailliez method. argument matrix dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":null,"dir":"Reference","previous_headings":"","what":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan package provides tools descriptive community ecology. basic functions diversity analysis, community ordination dissimilarity analysis. multivariate tools can used data types well.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"functions vegan package contain tools diversity analysis, ordination methods tools analysis dissimilarities. Together labdsv package, vegan package provides standard tools descriptive community analysis. Package ade4 provides alternative comprehensive package, several packages complement vegan provide tools deeper analysis specific fields. Package BiodiversityR provides GUI large subset vegan functionality. vegan package developed GitHub (https://github.com/vegandevs/vegan/). GitHub provides --date information forums bug reports. important changes vegan documents can read news(package=\"vegan\") vignettes can browsed browseVignettes(\"vegan\"). vignettes include vegan FAQ, discussion design decisions, short introduction ordination discussion diversity methods. see preferable citation package, type citation(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan development team Jari Oksanen, F. Guillaume Blanchet, Roeland Kindt, Pierre Legendre, Peter R. Minchin, R. B. O'Hara, Gavin L. Simpson, Peter Solymos, M. Henry H. Stevens, Helene Wagner. Many people contributed individual functions: see credits function help pages.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"","code":"### Example 1: Unconstrained ordination ## NMDS data(varespec) data(varechem) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.184583 #> ... Procrustes: rmse 0.04902295 max resid 0.1551746 #> Run 2 stress 0.1948413 #> Run 3 stress 0.2136761 #> Run 4 stress 0.2079056 #> Run 5 stress 0.2087937 #> Run 6 stress 0.196245 #> Run 7 stress 0.2069725 #> Run 8 stress 0.1948413 #> Run 9 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04165046 max resid 0.1519318 #> Run 10 stress 0.1843196 #> Run 11 stress 0.2109853 #> Run 12 stress 0.2177543 #> Run 13 stress 0.2224267 #> Run 14 stress 0.2044974 #> Run 15 stress 0.195049 #> Run 16 stress 0.1969805 #> Run 17 stress 0.22911 #> Run 18 stress 0.2397062 #> Run 19 stress 0.18584 #> Run 20 stress 0.2295494 #> *** Best solution was not repeated -- monoMDS stopping criteria: #> 1: no. of iterations >= maxit #> 19: stress ratio > sratmax plot(ord, type = \"t\") ## Fit environmental variables ef <- envfit(ord, varechem) ef #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05721 -0.99836 0.2537 0.038 * #> P 0.61964 0.78489 0.1938 0.102 #> K 0.76634 0.64244 0.1809 0.127 #> Ca 0.68509 0.72846 0.4119 0.005 ** #> Mg 0.63243 0.77462 0.4271 0.003 ** #> S 0.19127 0.98154 0.1752 0.110 #> Al -0.87166 0.49011 0.5269 0.001 *** #> Fe -0.93612 0.35169 0.4451 0.001 *** #> Mn 0.79870 -0.60173 0.5231 0.001 *** #> Zn 0.61752 0.78655 0.1879 0.102 #> Mo -0.90302 0.42960 0.0609 0.521 #> Baresoil 0.92500 -0.37996 0.2508 0.050 * #> Humdepth 0.93288 -0.36018 0.5200 0.002 ** #> pH -0.64804 0.76161 0.2308 0.074 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> plot(ef, p.max = 0.05) ### Example 2: Constrained ordination (RDA) ## The example uses formula interface to define the model data(dune) data(dune.env) ## No constraints: PCA mod0 <- rda(dune ~ 1, dune.env) mod0 #> Call: rda(formula = dune ~ 1, data = dune.env) #> #> Inertia Rank #> Total 84.12 #> Unconstrained 84.12 19 #> Inertia is variance #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 24.795 18.147 7.629 7.153 5.695 4.333 3.199 2.782 #> (Showing 8 of 19 unconstrained eigenvalues) #> plot(mod0) ## All environmental variables: Full model mod1 <- rda(dune ~ ., dune.env) mod1 #> Call: rda(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 63.2062 0.7513 12 #> Unconstrained 20.9175 0.2487 7 #> Inertia is variance #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 22.396 16.208 7.039 4.038 3.760 2.609 2.167 1.803 1.404 0.917 0.582 #> RDA12 #> 0.284 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 #> 6.627 4.309 3.549 2.546 2.340 0.934 0.612 #> plot(mod1) ## Automatic selection of variables by permutation P-values mod <- ordistep(mod0, scope=formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.015 * #> + A1 1 89.591 1.9217 0.065 . #> + Use 2 91.032 1.1741 0.240 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.095 . #> + A1 1 87.424 1.2965 0.200 #> + Use 2 88.284 1.0510 0.360 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.015 * #> - Management 3 87.707 2.1769 0.010 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.325 #> + A1 1 86.220 0.8359 0.585 #> + Use 2 86.842 0.8027 0.770 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> plot(mod) ## Permutation test for all variables anova(mod) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Model 6 46.425 2.6682 0.001 *** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test of \"type III\" effects, or significance when a term ## is added to the model after all other terms anova(mod, by = \"margin\") #> Permutation test for rda under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Management 3 18.938 2.1769 0.005 ** #> Moisture 3 17.194 1.9764 0.008 ** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Plot only sample plots, use different symbols and draw SD ellipses ## for Managemenet classes plot(mod, display = \"sites\", type = \"n\") with(dune.env, points(mod, disp = \"si\", pch = as.numeric(Management))) with(dune.env, legend(\"topleft\", levels(Management), pch = 1:4, title = \"Management\")) with(dune.env, ordiellipse(mod, Management, label = TRUE)) ## add fitted surface of diversity to the model ordisurf(mod, diversity(dune), add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 1.28 total = 2.28 #> #> REML score: 3.00623 ### Example 3: analysis of dissimilarites a.k.a. non-parametric ### permutational anova adonis2(dune ~ ., dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ ., data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> A1 1 0.7230 0.16817 5.2038 0.002 ** #> Moisture 3 1.1871 0.27613 2.8482 0.006 ** #> Management 3 0.9036 0.21019 2.1681 0.028 * #> Use 2 0.0921 0.02143 0.3315 0.983 #> Manure 3 0.4208 0.09787 1.0096 0.439 #> Residual 7 0.9725 0.22621 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adonis2(dune ~ Management + Moisture, dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management + Moisture, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.7907 0.001 *** #> Moisture 3 1.1516 0.26788 2.9726 0.002 ** #> Residual 13 1.6788 0.39051 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Dissimilarity Indices for Community Ecologists — vegdist","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function computes dissimilarity indices useful popular community ecologists. indices use quantitative data, although named corresponding binary index, can calculate binary index using appropriate argument. find favourite index , can see can implemented using designdist. Gower, Bray--Curtis, Jaccard Kulczynski indices good detecting underlying ecological gradients (Faith et al. 1987). Morisita, Horn--Morisita, Binomial, Cao Chao indices able handle different sample sizes (Wolda 1981, Krebs 1999, Anderson & Millar 2004), Mountford (1962) Raup-Crick indices presence--absence data able handle unknown (variable) sample sizes. indices discussed Krebs (1999) Legendre & Legendre (2012), properties compared Wolda (1981) Legendre & De Cáceres (2012). Aitchison (1986) distance equivalent Euclidean distance CLR-transformed samples (\"clr\") deals positive compositional data. Robust Aitchison distance Martino et al. (2019) uses robust CLR (\"rlcr\"), making applicable non-negative data including zeroes (unlike standard Aitchison).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"vegdist(x, method=\"bray\", binary=FALSE, diag=FALSE, upper=FALSE, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"x Community data matrix. method Dissimilarity index, partial match \"manhattan\", \"euclidean\", \"canberra\", \"clark\", \"bray\", \"kulczynski\", \"jaccard\", \"gower\", \"altGower\", \"morisita\", \"horn\", \"mountford\", \"raup\", \"binomial\", \"chao\", \"cao\", \"mahalanobis\", \"chisq\", \"chord\", \"hellinger\", \"aitchison\", \"robust.aitchison\". binary Perform presence/absence standardization analysis using decostand. diag Compute diagonals. upper Return upper diagonal. na.rm Pairwise deletion missing observations computing dissimilarities. ... parameters. ignored, except method =\"gower\" accepts range.global parameter decostand, method=\"aitchison\", accepts pseudocount parameter decostand used clr transformation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jaccard (\"jaccard\"), Mountford (\"mountford\"), Raup--Crick (\"raup\"), Binomial Chao indices discussed later section. function also finds indices presence/ absence data setting binary = TRUE. following overview gives first quantitative version, \\(x_{ij}\\) \\(x_{ik}\\) refer quantity species (column) \\(\\) sites (rows) \\(j\\) \\(k\\). binary versions \\(\\) \\(B\\) numbers species compared sites, \\(J\\) number species occur compared sites similarly designdist (many indices produce identical binary versions): Jaccard index computed \\(2B/(1+B)\\), \\(B\\) Bray--Curtis dissimilarity. Binomial index derived Binomial deviance null hypothesis two compared communities equal. able handle variable sample sizes. index fixed upper limit, can vary among sites shared species. discussion, see Anderson & Millar (2004). Cao index CYd index (Cao et al. 1997) suggested minimally biased index high beta diversity variable sampling intensity. Cao index fixed upper limit, can vary among sites shared species. index intended count (integer) data, undefined zero abundances; replaced arbitrary value \\(0.1\\) following Cao et al. (1997). Cao et al. (1997) used \\(\\log_{10}\\), current function uses natural logarithms values approximately \\(2.30\\) times higher 10-based logarithms. Anderson & Thompson (2004) give alternative formulation Cao index highlight relationship Binomial index (). Mountford index defined \\(M = 1/\\alpha\\) \\(\\alpha\\) parameter Fisher's logseries assuming compared communities samples community (cf. fisherfit, fisher.alpha). index \\(M\\) found positive root equation \\(\\exp() + \\exp(bM) = 1 + \\exp[(+b-j)M]\\), \\(j\\) number species occurring communities, \\(\\) \\(b\\) number species separate community (index uses presence--absence information). Mountford index usually misrepresented literature: indeed Mountford (1962) suggested approximation used starting value iterations, proper index defined root equation . function vegdist solves \\(M\\) Newton method. Please note either \\(\\) \\(b\\) equal \\(j\\), one communities subset , dissimilarity \\(0\\) meaning non-identical objects may regarded similar index non-metric. Mountford index range \\(0 \\dots \\log(2)\\). Raup--Crick dissimilarity (method = \"raup\") probabilistic index based presence/absence data. defined \\(1 - prob(j)\\), based probability observing least \\(j\\) species shared compared communities. current function uses analytic result hypergeometric distribution (phyper) find probabilities. probability (index) dependent number species missing sites, adding -zero species data removing missing species data influence index. probability ( index) may almost zero almost one wide range parameter values. index nonmetric: two communities shared species may dissimilarity slightly one, two identical communities may dissimilarity slightly zero. index uses equal occurrence probabilities species, Raup Crick originally suggested sampling probabilities proportional species frequencies (Chase et al. 2011). simulation approach unequal species sampling probabilities implemented raupcrick function following Chase et al. (2011). index can also used transposed data give probabilistic dissimilarity index species co-occurrence (identical Veech 2013). Chao index tries take account number unseen species pairs, similarly method = \"chao\" specpool. Function vegdist implements Jaccard, index defined \\(1-\\frac{U \\times V}{U + V - U \\times V}\\); types can defined function chaodist. Chao equation, \\(U = C_j/N_j + (N_k - 1)/N_k \\times a_1/(2 a_2) \\times S_j/N_j\\), \\(V\\) similar except site index \\(k\\). \\(C_j\\) total number individuals species site \\(j\\) shared site \\(k\\), \\(N_j\\) total number individuals site \\(j\\), \\(a_1\\) (\\(a_2\\)) number species occurring site \\(j\\) one (two) individuals site \\(k\\), \\(S_j\\) total number individuals species present site \\(j\\) occur one individual site \\(k\\) (Chao et al. 2005). Morisita index can used genuine count data (integers) . Horn--Morisita variant able handle abundance data. Mahalanobis distances Euclidean distances matrix columns centred, unit variance, uncorrelated. index commonly used community data, sometimes used environmental variables. calculation based transforming data matrix using Euclidean distances following Mardia et al. (1979). Mahalanobis transformation usually fails number columns larger number rows (sampling units). transformation fails, distances nearly constant except small numeric noise. Users must check returned Mahalanobis distances meaningful. Euclidean Manhattan dissimilarities good gradient separation without proper standardization still included comparison special needs. Chi-square distances (\"chisq\") Euclidean distances Chi-square transformed data (see decostand). internal standardization used correspondence analysis (cca, decorana). Weighted principal coordinates analysis distances row sums weights equal correspondence analysis (see Example wcmdscale). Chi-square distance intended non-negative data, typical community data. However, can calculated long margin sums positive, warning issued negative data entries. Chord distances (\"chord\") Euclidean distance matrix rows standardized unit norm (sums squares 1) using decostand. Geometrically standardization moves row points surface multidimensional unit sphere, distances chords across hypersphere. Hellinger distances (\"hellinger\") related Chord distances, data standardized unit total (row sums 1) using decostand, square root transformed. distances upper limit \\(\\sqrt{2}\\). Bray--Curtis Jaccard indices rank-order similar, indices become identical rank-order similar standardizations, especially presence/absence transformation equalizing site totals decostand. Jaccard index metric, probably preferred instead default Bray-Curtis semimetric. Aitchison distance (1986) robust Aitchison distance (Martino et al. 2019) metrics deal compositional data. Aitchison distance said outperform Jensen-Shannon divergence Bray-Curtis dissimilarity, due better stability subsetting aggregation, proper distance (Aitchison et al., 2000). naming conventions vary. one adopted traditional rather truthful priority. function finds either quantitative binary variants indices name, correctly may refer one alternatives instance, Bray index known also Steinhaus, Czekanowski Sørensen index. quantitative version Jaccard probably called Ružička index. abbreviation \"horn\" Horn--Morisita index misleading, since separate Horn index. abbreviation changed index implemented vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Function drop-replacement dist function returns distance object type. result object adds attribute maxdist gives theoretical maximum index sampling units share species, NA maximum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Aitchison, J., Barceló-Vidal, C., Martín-Fernández, J.., Pawlowsky-Glahn, V. (2000). Logratio analysis compositional distance. Math. Geol. 32, 271–275. Anderson, M.J. Millar, R.B. (2004). Spatial variation effects habitat temperate reef fish assemblages northeastern New Zealand. Journal Experimental Marine Biology Ecology 305, 191--221. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006). Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Anderson, M.J & Thompson, .. (2004). Multivariate control charts ecological environmental monitoring. Ecological Applications 14, 1921--1935. Cao, Y., Williams, W.P. & Bark, .W. (1997). Similarity measure bias river benthic Auswuchs community analysis. Water Environment Research 69, 95--106. Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005). new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1 Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J. C. (1971). general coefficient similarity properties. Biometrics 27, 623--637. Krebs, C. J. (1999). Ecological Methodology. Addison Wesley Longman. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141 Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate analysis. Academic Press. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Mountford, M. D. (1962). index similarity application classification problems. : P.W.Murphy (ed.), Progress Soil Zoology, 43--50. Butterworths. Veech, J. . (2013). probabilistic model analysing species co-occurrence. Global Ecology Biogeography 22, 252--260. Wolda, H. (1981). Similarity indices, sample size diversity. Oecologia 50, 296--302.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jari Oksanen, contributions Tyler Smith (Gower index), Michael Bedward (Raup--Crick index), Leo Lahti (Aitchison robust Aitchison distance).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function alternative dist adding ecologically meaningful indices. methods produce similar types objects can interchanged method accepting either. Manhattan Euclidean dissimilarities identical methods. Canberra index divided number variables vegdist, dist. differ constant multiplier, alternative vegdist range (0,1). Function daisy (package cluster) provides alternative implementation Gower index also can handle mixed data numeric class variables. two versions Gower distance (\"gower\", \"altGower\") differ scaling: \"gower\" divides distances number observations (rows) scales column unit range, \"altGower\" omits double-zeros divides number pairs least one -zero value, scale columns (Anderson et al. 2006). can use decostand add range standardization \"altGower\" (see Examples). Gower (1971) suggested omitting double zeros presences, often taken general feature Gower distances. See Examples implementing Anderson et al. (2006) variant Gower index. dissimilarity indices vegdist designed community data, give misleading values negative data entries. results may also misleading NA NaN empty sites. principle, study species composition without species remove empty sites community data.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"data(varespec) vare.dist <- vegdist(varespec) # Orlóci's Chord distance: range 0 .. sqrt(2) vare.dist <- vegdist(decostand(varespec, \"norm\"), \"euclidean\") # Anderson et al. (2006) version of Gower vare.dist <- vegdist(decostand(varespec, \"log\"), \"altGower\") #> Warning: non-integer data: divided by smallest positive value # Range standardization with \"altGower\" (that excludes double-zeros) vare.dist <- vegdist(decostand(varespec, \"range\"), \"altGower\")"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Compact Ordered Community Tables — vegemite","title":"Display Compact Ordered Community Tables — vegemite","text":"Functions vegemite tabasco display compact community tables. Function vegemite prints text tables species rows, site takes one column without spaces. Function tabasco provides interface heatmap colour image data. community table can ordered explicit indexing, environmental variables results ordination cluster analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"vegemite(x, use, scale, sp.ind, site.ind, zero=\".\", select ,...) tabasco(x, use, sp.ind = NULL, site.ind = NULL, select, Rowv = TRUE, Colv = TRUE, labRow = NULL, labCol = NULL, scale, col = heat.colors(12), ...) coverscale(x, scale=c(\"Braun.Blanquet\", \"Domin\", \"Hult\", \"Hill\", \"fix\",\"log\"), maxabund, character = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Compact Ordered Community Tables — vegemite","text":"x Community data. use Either vector, object cca, decorana etc. hclust dendrogram ordering sites species. sp.ind, site.ind Species site indices. tabasco, can also hclust tree, agnes clusterings dendrograms. zero Character used zeros. select Select subset sites. can logical vector (TRUE selected sites), vector indices selected sites. order indices influence results, must specify use site.ind reorder sites. Rowv, Colv Re-order dendrograms rows (sites) columns (species) x. Rowv = TRUE, row dendrograms ordered first axis correspondence analysis, Colv = TRUE column dendrograms weighted average (wascores) row order. Alternatively, arguments can vectors used reorder dendrogram. labRow, labCol character vectors row column labels used heatmap instead default. NB., input matrix transposed row labels used data columns. scale vegemite coverscale: cover scale used (can abbreviated). tabasco: scaling colours heatmap. alternatives coverscale can used tabasco, addition \"column\" \"row\" scale columns rows equal maxima (NB., refer transposed data heatmap), \"none\" uses original values. col vector colours used -zero abundance values. maxabund Maximum abundance used scale = \"log\". Data maximum selected subset used missing. character Return character codes suitable vegemite. FALSE, returns corresponding integers. ... Arguments passed coverscale (.e., maxabund) vegemite heatmap tabasco.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Compact Ordered Community Tables — vegemite","text":"function vegemite prints traditional community table. display transposed, species rows sites columns. table printed compact form: one character can used abundance, spaces columns. Species occurrences dropped table. Function tabasco produces similar table vegemite using heatmap, abundances coded colours. function scales abundances equal intervals colour palette, either rows columns can scaled equal maxima, coverscale class systems can used. function can also display dendrograms sites (columns) species given argument (use sites, sp.ind species). parameter use used re-order output. use can vector object hclust agnes, dendrogram ordination result recognized scores (ordination methods vegan vegan). hclust, agnes dendrogram must sites. dendrogram displayed sites tabasco, shown vegemite. dendrogram species displayed, except given sp.ind. use vector, used ordering sites. use object ordination, sites species arranged first axis (provided results available also species). use object hclust, agnes dendrogram, sites ordered similarly cluster dendrogram. Function tabasco re-orders dendrogram Rowv = TRUE Rowv vector. re-ordering available vegemite, can done hand using reorder.dendrogram reorder.hclust. Please note dendrogram hclust reordering can differ: unweighted means merged branches used dendrogram, weighted means (= means leaves cluster) used reorder.hclust. cases species scores missing, species ordered weighted averages (wascores) site order. Species sites can ordered explicitly giving indices names parameters sp.ind site.ind. given, take precedence use. subset sites can displayed using argument select, used order sites, still must give use site.ind. However, tabasco makes two exceptions: site.ind select used use dendrogram (clustering result). addition, sp.ind can hclust tree, agnes clustering dendrogram, case dendrogram plotted left side heatmap. Phylogenetic trees directly used, package ape tools transform hclust trees. scale given, vegemite calls coverscale transform percent cover scale scales traditional class scales used vegetation science (coverscale can called directly, ). Function tabasco can also use traditional class scales, treats transformed values corresponding integers. Braun-Blanquet Domin scales actually strict cover scales, limits used codes r + arbitrary. Scale Hill may inappropriately named, since Mark O. Hill probably never intended cover scale. However, used default “cut levels” TWINSPAN, surprisingly many users stick default, de facto standard publications. traditional scales assume values cover percentages maximum 100. However, non-traditional alternative log can used scale range. class limits integer powers 1/2 maximum (argument maxabund), + used non-zero entries less 1/512 maximum (log stands alternatively logarithmic logical). Scale fix intended “fixing” 10-point scales: truncates scale values integers, replaces 10 X positive values 1 +.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Compact Ordered Community Tables — vegemite","text":"functions used mainly display table, return (invisibly) list items species ordered species index, sites ordered site index, table final ordered community table. items can used arguments sp.ind site.ind reproduce table, table can edited. addition table, vegemite prints numbers species sites name used cover scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Display Compact Ordered Community Tables — vegemite","text":"cover scales presented many textbooks vegetation science; used: Shimwell, D.W. (1971) Description Classification Vegetation. Sidgwick & Jackson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Compact Ordered Community Tables — vegemite","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Compact Ordered Community Tables — vegemite","text":"name vegemite chosen output compact, tabasco just compact, uses heat colours.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"data(varespec) ## Print only more common species freq <- apply(varespec > 0, 2, sum) vegemite(varespec, scale=\"Hult\", sp.ind = freq > 10) #> #> 1122212121122 1112 #> 854739268340575634292011 #> Callvulg 111...11.311...1111.111. #> Empenigr 211332121112211111212213 #> Vaccviti 323332212113221211233234 #> Pinusylv 111.111111111.11.1111111 #> Dicrfusc 12.111441121211111111111 #> Dicrpoly .11.11.11..1.11....1.111 #> Pleuschr 144533435123411111111131 #> Polyjuni 111.11.111112111.111.111 #> Pohlnuta 111111111111.1..11.11111 #> Ptilcili 1111111111..1..11.11..12 #> Cladarbu 321122121332143423111121 #> Cladrang 321221131312145443313241 #> Cladstel 11111311.211.11254555542 #> Cladunci 112111111131111111111111 #> Cladcocc 11..11111111111111.1.11. #> Cladcorn 111111111111111111111111 #> Cladgrac 111111111111111111111.11 #> Cladfimb 11.1111111111111111111.1 #> Cladcris 111111111111111111111111 #> Cladchlo ..1.111..1...1.11.11.1.1 #> Cetreric 111...11.111111111.1.1.. #> Cetrisla .11....1111.1....1.11111 #> Stersp 111.11.1111.112111...11. #> Claddefo 1111111111111111111111.1 #> 24 sites, 24 species #> scale: Hult ## Order by correspondence analysis, use Hill scaling and layout: dca <- decorana(varespec) vegemite(varespec, dca, \"Hill\", zero=\"-\") #> #> 1 1 1 11122211122222 #> 203942561738913046572458 #> Flavniva -1114-11-1-11-1--------- #> Cladstel 5555551451425411111211-- #> Cladphyl -1-1----1-------1------- #> Cladcerv 1---1-----------------1- #> Cladsp ---11--1--11---1-1-11-1- #> Cladamau --1---1----1------------ #> Cladchlo 1111---1-11-111-----11-- #> Cladrang 535254555555223414332321 #> Diphcomp --11-----112----------1- #> Stersp -11-1-4111111-1-111--111 #> Pinusylv 11-111111-111111111-1111 #> Polypili ------111--11-11--1----- #> Cetrisla -1-111--1-1--1--111--111 #> Cladcocc -1111-1111111-11111-1-11 #> Cladarbu 113142453555313343413231 #> Vacculig --1-1----3-1---1---12-11 #> Pohlnuta -11111--11111111111111-1 #> Cladfimb 11111111-111111111111-11 #> Callvulg 111-21-12-51---1221-21-- #> Icmaeric ------1--1------11------ #> Empenigr 342214131314344333143131 #> Vaccviti 342514244334455432444443 #> Cladgrac 1-1111111111111111111111 #> Cetreric -1111-11-111---1111-111- #> Cladcorn 111111111111111111111111 #> Cladcris 111111111111111111111111 #> Peltapht --------1-11--1----1--1- #> Ptilcili 1-11---11-11141--1111111 #> Barbhatc ----------1-121----1---- #> Claddefo 11111111-111111111111111 #> Cladbotr ----------1-1111---1-1-1 #> Betupube -------------1------1-1- #> Dicrpoly -1-1--1-11--1211-11--1-1 #> Cladunci 111111122111111251212311 #> Polycomm ------------11--1--1--1- #> Polyjuni 11-11-1111111121111--131 #> Rhodtome ----------1--2-----21--1 #> Dicrfusc 111111111121112145425-41 #> Pleuschr 111213114132524434555555 #> Vaccmyrt ---1------1-24--12133--4 #> Descflex ----1----1--11-----21-11 #> Nepharct --1------1-1---1------2- #> Dicrsp -----1----1--1-1-11-1541 #> Hylosple ---------------1---3--13 #> 24 sites, 44 species #> scale: Hill ## Show one class from cluster analysis, but retain the ordering above clus <- hclust(vegdist(varespec)) cl <- cutree(clus, 3) sel <- vegemite(varespec, use=dca, select = cl == 3, scale=\"Br\") #> #> 1 12 #> 20921 #> Flavniva .++.. #> Cladstel 55542 #> Cladphyl .++.. #> Cladcerv r.... #> Cladsp ..+.. #> Cladchlo r++.+ #> Cladrang 22121 #> Diphcomp ..+.. #> Stersp .+... #> Pinusylv r++1+ #> Cetrisla .++++ #> Cladcocc .++.. #> Cladarbu +1+1+ #> Pohlnuta .++++ #> Cladfimb r++++ #> Callvulg r+.+. #> Empenigr 22122 #> Vaccviti 22223 #> Cladgrac r.+++ #> Cetreric .++.. #> Cladcorn r+++r #> Cladcris r++++ #> Ptilcili r.+.2 #> Barbhatc ....1 #> Claddefo r++++ #> Cladbotr ....+ #> Betupube ....+ #> Dicrpoly .++.1 #> Cladunci +++1+ #> Polycomm ....+ #> Polyjuni r++.+ #> Rhodtome ....1 #> Dicrfusc r++++ #> Pleuschr ++121 #> Vaccmyrt ..+.2 #> Descflex ....+ #> Dicrsp ...++ #> 5 sites, 37 species #> scale: Braun.Blanquet ## Re-create previous vegemite(varespec, sp=sel$sp, site=sel$site, scale=\"Hult\") #> #> 1 12 #> 20921 #> Flavniva .11.. #> Cladstel 55552 #> Cladphyl .11.. #> Cladcerv 1.... #> Cladsp ..1.. #> Cladchlo 111.1 #> Cladrang 32131 #> Diphcomp ..1.. #> Stersp .1... #> Pinusylv 11111 #> Cetrisla .1111 #> Cladcocc .11.. #> Cladarbu 11111 #> Pohlnuta .1111 #> Cladfimb 11111 #> Callvulg 11.1. #> Empenigr 22123 #> Vaccviti 22334 #> Cladgrac 1.111 #> Cetreric .11.. #> Cladcorn 11111 #> Cladcris 11111 #> Ptilcili 1.1.2 #> Barbhatc ....1 #> Claddefo 11111 #> Cladbotr ....1 #> Betupube ....1 #> Dicrpoly .11.1 #> Cladunci 11111 #> Polycomm ....1 #> Polyjuni 111.1 #> Rhodtome ....1 #> Dicrfusc 11111 #> Pleuschr 11111 #> Vaccmyrt ..1.3 #> Descflex ....1 #> Dicrsp ...11 #> 5 sites, 37 species #> scale: Hult ## Re-order clusters by ordination clus <- as.dendrogram(clus) clus <- reorder(clus, scores(dca, choices=1, display=\"sites\"), agglo.FUN = mean) vegemite(varespec, clus, scale = \"Hult\") #> #> 1 111 1211221212222 #> 431567380922149306254578 #> Flavniva 21.111.111....11........ #> Cladamau .1.1...1................ #> Stersp 111211111....111.1.111.1 #> Polypili ..111..1......111..1.... #> Diphcomp .1...111.1...........1.. #> Cladphyl ..1.....11...1.......... #> Cladrang 344544332133111223121121 #> Cladcerv 1.........1..........1.. #> Cladstel 454121215555213111111.1. #> Cladarbu 322344331111132222121111 #> Vacculig 11...2.1........1.1..111 #> Callvulg 111.1.311.11.1..11111... #> Icmaeric ...1.1.......1...1...... #> Cladsp 1...1.11.1......111..11. #> Cladcocc 1111111111...1111111.1.1 #> Pinusylv 1.111.1111111111111111.1 #> Cladchlo .1..111.111.1.11..1.1... #> Cetrisla 1.1...1.11.111...1.111.1 #> Cladfimb 11.11111111111111111.111 #> Peltapht ..1...11.......1.....11. #> Cetreric 11.1111111...1..111111.. #> Cladgrac 11111111.111111111111111 #> Pohlnuta 111..11111.1111111111.11 #> Ptilcili .11.1.11.11.2.11.1111111 #> Barbhatc ......1.....1.11......1. #> Cladcorn 111111111111111111111111 #> Vaccviti 113122132323412331223232 #> Cladcris 111111111111111111111111 #> Empenigr 111111122122312322111231 #> Cladbotr ......1.....1.111...1.11 #> Betupube ............1.....1..1.. #> Cladunci 111111111111131111112111 #> Claddefo 11.111111111111111111111 #> Dicrpoly ..11.1..11..1.1111.11..1 #> Polycomm ............111......11. #> Rhodtome ......1.....1.....1...11 #> Polyjuni 1.111111111.111111.112.1 #> Dicrfusc 11111111111112111442.211 #> Pleuschr 113111111111123333444455 #> Vaccmyrt ......1..1..311..111..13 #> Nepharct .1...1.1........1....1.. #> Dicrsp ......1....11...111133.1 #> Descflex 1....1......1.1...1..111 #> Hylosple ................1....122 #> 24 sites, 44 species #> scale: Hult ## Abundance values have such a wide range that they must be rescaled tabasco(varespec, dca, scale=\"Braun\") ## Classification trees for species data(dune, dune.taxon) taxontree <- hclust(taxa2dist(dune.taxon)) plotree <- hclust(vegdist(dune), \"average\") ## Automatic reordering of clusters tabasco(dune, plotree, sp.ind = taxontree) ## No reordering of taxonomy tabasco(dune, plotree, sp.ind = taxontree, Colv = FALSE) ## Species cluster: most dissimilarity indices do a bad job when ## comparing rare and common species, but Raup-Crick makes sense sptree <- hclust(vegdist(t(dune), \"raup\"), \"average\") tabasco(dune, plotree, sptree)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Averages Scores for Species — wascores","title":"Weighted Averages Scores for Species — wascores","text":"Computes Weighted Averages scores species ordination configuration environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Averages Scores for Species — wascores","text":"","code":"wascores(x, w, expand=FALSE) eigengrad(x, w)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Averages Scores for Species — wascores","text":"x Environmental variables ordination scores. w Weights: species abundances. expand Expand weighted averages weighted variance corresponding environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores computes weighted averages. Weighted averages “shrink”: extreme values used calculating averages. expand = TRUE, function “deshrinks” weighted averages making biased weighted variance equal biased weighted variance corresponding environmental variable. Function eigengrad returns inverses squared expansion factors attribute shrinkage wascores result environmental gradient. equal constrained eigenvalue cca one gradient used constraint, describes strength gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores returns matrix species define rows ordination axes environmental variables define columns. expand = TRUE, attribute shrinkage inverses squared expansion factors cca eigenvalues variable. Function eigengrad returns shrinkage attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Weighted Averages Scores for Species — wascores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Averages Scores for Species — wascores","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(wisconsin(varespec)) vare.mds <- monoMDS(vare.dist) vare.points <- postMDS(vare.mds$points, vare.dist) vare.wa <- wascores(vare.points, varespec) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa, rownames(vare.wa), cex=0.8, col=\"blue\") ## Omit rare species (frequency <= 4) freq <- apply(varespec>0, 2, sum) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa[freq > 4,], rownames(vare.wa)[freq > 4],cex=0.8,col=\"blue\") ## Works for environmental variables, too. wascores(varechem, varespec) #> N P K Ca Mg S Al #> Callvulg 25.12401 41.66188 246.92383 572.3431 99.40828 49.10552 245.15751 #> Empenigr 21.61371 44.17350 158.92517 580.8403 89.20628 35.75327 107.05670 #> Rhodtome 22.34553 40.35530 162.31108 643.8819 95.07712 30.84303 28.83159 #> Vaccmyrt 24.96352 49.80649 189.70177 656.4179 96.75529 34.55190 32.75705 #> Vaccviti 21.14028 45.86984 162.12372 613.2126 94.30084 37.26975 116.14656 #> Pinusylv 18.37299 44.24818 163.60195 670.9387 93.52214 37.02238 150.45523 #> Descflex 22.24089 54.74804 212.20357 771.0159 114.15179 38.00000 24.64143 #> Betupube 21.51034 28.86552 112.28276 513.5586 75.15172 23.55172 33.12414 #> Vacculig 28.00729 33.48758 114.90230 372.9653 70.89954 29.32378 202.89008 #> Diphcomp 22.34228 39.71049 127.44259 446.1565 80.16173 32.32963 122.31574 #> Dicrsp 21.33007 60.03758 185.04563 828.0544 148.67509 46.75427 90.42294 #> Dicrfusc 23.45681 39.14575 162.91954 578.2309 77.81648 33.48086 60.66890 #> Dicrpoly 20.65446 43.87409 150.51485 665.5845 115.22112 36.17079 90.16733 #> Hylosple 26.10599 67.88980 245.78681 779.6520 111.96685 42.27433 24.92738 #> Pleuschr 22.60476 54.22534 199.96241 712.6278 109.23425 40.01132 70.43900 #> Polypili 23.17377 43.75902 144.82623 724.9738 85.42623 30.58525 145.73115 #> Polyjuni 22.89480 47.98022 154.81906 643.3864 87.27819 33.63863 53.24888 #> Polycomm 21.73521 41.17042 154.91549 631.8704 101.84789 32.17324 46.80986 #> Pohlnuta 19.99885 48.59198 169.68855 678.3813 104.31641 39.94427 132.05458 #> Ptilcili 21.27880 33.44211 127.08522 564.5652 85.96417 27.11720 56.60692 #> Barbhatc 21.17461 27.93323 113.13542 497.9138 77.50564 23.72288 42.09749 #> Cladarbu 23.56127 38.04952 142.03073 454.9019 74.00779 33.81002 173.12698 #> Cladrang 24.28421 38.60534 135.31177 463.2750 70.54209 32.53349 183.79979 #> Cladstel 19.28049 46.71060 158.00576 540.4904 80.19153 40.29106 225.89526 #> Cladunci 21.41240 45.49844 163.40402 621.9100 98.35538 40.00734 119.59481 #> Cladcocc 21.72473 42.80681 156.32330 557.9007 80.95448 36.25161 149.82616 #> Cladcorn 22.11640 47.06656 160.36881 623.5185 95.17781 36.75273 104.71463 #> Cladgrac 22.51887 44.06576 156.50214 583.1558 94.10292 36.93930 134.13424 #> Cladfimb 21.77980 41.82652 153.29444 512.4646 78.28232 35.62323 128.96061 #> Cladcris 20.88795 44.12262 171.04016 574.5672 92.52169 38.24003 116.03507 #> Cladchlo 19.51207 45.39655 150.93190 571.0233 95.77586 39.50862 156.81983 #> Cladbotr 22.97660 38.89574 167.20000 590.8021 99.57234 34.79362 87.75957 #> Cladamau 25.07143 35.84286 105.07857 395.2214 68.18571 27.11429 95.91429 #> Cladsp 19.21923 47.37308 168.49231 526.7654 79.54423 45.15385 215.33846 #> Cetreric 21.00944 47.76972 165.07972 579.6322 99.14944 42.25472 163.46000 #> Cetrisla 18.36552 42.73695 151.78374 626.2813 89.77833 35.33498 132.68227 #> Flavniva 18.56110 61.18194 207.67705 502.9203 60.91755 50.22532 396.82405 #> Nepharct 23.33099 49.10019 146.84715 618.1601 64.27319 29.95760 31.72300 #> Stersp 28.19743 32.84800 94.33459 389.5143 53.25377 24.22175 95.39326 #> Peltapht 21.08553 54.45395 193.38816 886.5487 119.35132 37.92500 106.16447 #> Icmaeric 28.88636 27.00000 87.86818 307.0500 40.48182 22.17273 89.94091 #> Cladcerv 20.25000 56.79000 192.36000 519.2300 62.10000 45.18000 314.92000 #> Claddefo 22.19198 45.22981 167.73069 583.7983 92.01320 38.51369 100.46139 #> Cladphyl 15.73750 54.56875 180.39375 775.4500 99.65625 43.35000 208.55000 #> Fe Mn Zn Mo Baresoil Humdepth pH #> Callvulg 75.457843 52.38247 8.281074 0.4734635 27.241036 2.187819 2.845108 #> Empenigr 38.146102 53.49357 7.159938 0.3289657 27.324317 2.367439 2.888078 #> Rhodtome 5.560906 70.48260 7.444100 0.2251490 37.325030 2.689154 2.895352 #> Vaccmyrt 5.589213 75.17221 7.838533 0.2666732 31.404171 2.798935 2.855216 #> Vaccviti 37.586067 51.81515 7.617213 0.3680289 26.307701 2.307879 2.923128 #> Pinusylv 39.121898 35.22311 7.733333 0.3485401 17.762968 1.996350 3.049148 #> Descflex 6.066429 110.87232 9.526607 0.2316071 22.740179 2.834821 2.822857 #> Betupube 5.417241 37.53448 5.637931 0.2068966 51.496552 2.527586 2.979310 #> Vacculig 93.963929 37.73062 4.593824 0.3780552 21.410710 2.041196 3.006965 #> Diphcomp 73.281173 46.88025 4.593827 0.3725309 31.836574 2.103704 2.856790 #> Dicrsp 22.504617 65.35020 13.060765 0.5610370 23.182889 2.232247 2.954272 #> Dicrfusc 13.922252 61.40958 6.922859 0.3218816 26.918674 2.484399 2.806431 #> Dicrpoly 20.973927 33.28779 9.110561 0.3892739 37.304043 2.228713 3.015842 #> Hylosple 4.729157 115.14606 9.885976 0.2851996 20.956264 2.925000 2.807594 #> Pleuschr 19.113811 77.12277 9.007860 0.3405945 24.584979 2.596881 2.858446 #> Polypili 51.993443 36.74754 8.045902 0.2204918 17.368852 1.493443 3.227869 #> Polyjuni 12.885704 82.56274 7.945126 0.2760289 28.116303 2.615523 2.874729 #> Polycomm 7.895775 68.22535 7.843662 0.2591549 38.687324 2.926761 2.860563 #> Pohlnuta 33.089313 42.02290 8.452290 0.3935115 24.709351 2.147328 2.985496 #> Ptilcili 14.036188 33.93547 5.906924 0.2303712 48.941884 2.502498 2.973376 #> Barbhatc 8.199687 31.11379 5.550784 0.2084639 54.331975 2.514734 2.986834 #> Cladarbu 65.470394 38.28429 6.387500 0.4464046 22.592997 2.048540 2.937879 #> Cladrang 76.612752 34.86010 6.616452 0.3903501 17.270158 1.799128 3.022346 #> Cladstel 84.639467 33.27903 7.287216 0.4054879 9.854042 1.851973 3.052167 #> Cladunci 27.463504 40.10322 9.108102 0.5120114 28.312376 2.362687 2.858564 #> Cladcocc 46.653763 43.53584 7.269176 0.3698925 19.972222 2.025090 2.974194 #> Cladcorn 32.916238 53.38441 7.518489 0.3639871 26.620868 2.399518 2.890997 #> Cladgrac 45.758366 44.63911 7.702724 0.4244163 25.546654 2.230350 2.933658 #> Cladfimb 41.710354 47.83687 6.815152 0.3887626 24.763889 2.239394 2.902778 #> Cladcris 34.729585 44.75984 6.933735 0.3746988 29.711352 2.400000 2.841633 #> Cladchlo 42.089655 35.13276 7.908621 0.3814655 22.319655 2.075000 3.022414 #> Cladbotr 23.374468 46.54255 7.263830 0.3106383 45.725532 2.580851 2.904255 #> Cladamau 68.971429 41.98571 4.928571 0.3214286 27.592857 1.857143 2.914286 #> Cladsp 47.913462 49.61923 8.421154 0.6173077 16.864423 2.213462 2.921154 #> Cetreric 42.614167 36.96694 9.516389 0.5687500 21.452639 2.058056 2.923889 #> Cetrisla 29.617734 33.24138 7.518227 0.3192118 26.417980 2.027586 3.065025 #> Flavniva 94.339916 37.03232 9.116371 0.9987764 19.692312 1.799241 2.923629 #> Nepharct 12.910837 115.15684 7.743536 0.2180608 23.135932 2.541065 2.918251 #> Stersp 30.226998 31.97061 7.635502 0.2828767 15.844007 1.477740 3.038756 #> Peltapht 37.598684 56.64079 7.652632 0.2046053 28.321053 2.286842 3.026316 #> Icmaeric 24.236364 23.95909 6.618182 0.2863636 18.727273 1.568182 2.968182 #> Cladcerv 111.090000 52.04000 8.530000 0.6800000 15.393000 1.870000 2.900000 #> Claddefo 25.116325 48.81105 7.599609 0.4096285 33.814545 2.468133 2.823656 #> Cladphyl 50.475000 35.28125 8.568750 0.2812500 7.728125 1.575000 3.231250 ## And the strengths of these variables are: eigengrad(varechem, varespec) #> N P K Ca Mg S Al #> 0.13000842 0.18880078 0.16246365 0.15722067 0.16359171 0.13391967 0.29817406 #> Fe Mn Zn Mo Baresoil Humdepth pH #> 0.20766831 0.27254480 0.16783834 0.09542514 0.20931501 0.25051326 0.14583161"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Weighted classical multidimensional scaling, also known weighted principal coordinates analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"wcmdscale(d, k, eig = FALSE, add = FALSE, x.ret = FALSE, w) # S3 method for wcmdscale plot(x, choices = c(1, 2), type = \"t\", ...) # S3 method for wcmdscale scores(x, choices = NA, tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"d distance structure returned dist full symmetric matrix containing dissimilarities. k dimension space data represented ; must \\(\\{1,2,\\ldots,n-1\\}\\). missing, dimensions zero eigenvalue. eig indicates whether eigenvalues returned. add additive constant \\(c\\) added non-diagonal dissimilarities \\(n-1\\) eigenvalues non-negative. Alternatives \"lingoes\" (default, also used TRUE) \"cailliez\" ( alternative cmdscale). See Legendre & Anderson (1999). x.ret indicates whether doubly centred symmetric distance matrix returned. w Weights points. x wcmdscale result object function called options eig = TRUE x.ret = TRUE (See Details). choices Axes returned; NA returns real axes. type Type graph may \"t\"ext, \"p\"oints \"n\"one. tidy Return scores compatible ggplot2: scores data.frame, score type variable score labelled \"sites\", weights variable weigth, names variable label. ... arguments passed graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Function wcmdscale based function cmdscale (package stats base R), uses point weights. Points high weights stronger influence result low weights. Setting equal weights w = 1 give ordinary multidimensional scaling. default options, function returns matrix scores scaled eigenvalues real axes. function called eig = TRUE x.ret = TRUE, function returns object class \"wcmdscale\" print, plot, scores, eigenvals stressplot methods, described section Value. method Euclidean, non-Euclidean dissimilarities eigenvalues can negative. disturbs , can avoided adding constant non-diagonal dissimilarities making eigenvalues non-negative. function implements methods discussed Legendre & Anderson (1999): method Lingoes (add=\"lingoes\") adds constant \\(c\\) squared dissimilarities \\(d\\) using \\(\\sqrt{d^2 + 2 c}\\) method Cailliez (add=\"cailliez\") dissimilarities using \\(d + c\\). Legendre & Anderson (1999) recommend method Lingoes, base R function cmdscale implements method Cailliez.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"eig = FALSE x.ret = FALSE (default), matrix k columns whose rows give coordinates points corresponding positive eigenvalues. Otherwise, object class wcmdscale containing components mostly similar cmdscale: points matrix k columns whose rows give coordinates points chosen represent dissimilarities. eig \\(n-1\\) eigenvalues computed scaling process eig true. x doubly centred weighted distance matrix x.ret true. ac, add additive constant adjustment method used avoid negative eigenvalues. NA FALSE adjustment done. GOF Goodness fit statistics k axes. first value based sum absolute values eigenvalues, second value based sum positive eigenvalues weights Weights. negaxes matrix scores axes negative eigenvalues scaled absolute eigenvalues similarly points. NULL negative eigenvalues k specified, include negative eigenvalues. call Function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Gower, J. C. (1966) distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecology 69, 1--24. Mardia, K. V., Kent, J. T. Bibby, J. M. (1979). Chapter 14 Multivariate Analysis, London: Academic Press.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"## Correspondence analysis as a weighted principal coordinates ## analysis of Euclidean distances of Chi-square transformed data data(dune) rs <- rowSums(dune)/sum(dune) d <- dist(decostand(dune, \"chi\")) ord <- wcmdscale(d, w = rs, eig = TRUE) ## Ordinary CA ca <- cca(dune) ## IGNORE_RDIFF_BEGIN ## Eigevalues are numerically similar ca$CA$eig - ord$eig #> CA1 CA2 CA3 CA4 CA5 #> 6.661338e-16 4.440892e-16 0.000000e+00 1.942890e-16 4.163336e-16 #> CA6 CA7 CA8 CA9 CA10 #> -2.359224e-16 -2.081668e-16 1.110223e-16 -1.387779e-17 -2.775558e-17 #> CA11 CA12 CA13 CA14 CA15 #> 7.632783e-17 1.387779e-17 1.387779e-17 -3.122502e-17 6.938894e-18 #> CA16 CA17 CA18 CA19 #> -3.295975e-17 -1.214306e-17 -1.647987e-17 6.028164e-17 ## Configurations are similar when site scores are scaled by ## eigenvalues in CA procrustes(ord, ca, choices=1:19, scaling = \"sites\") #> #> Call: #> procrustes(X = ord, Y = ca, choices = 1:19, scaling = \"sites\") #> #> Procrustes sum of squares: #> -4.263e-14 #> ## IGNORE_RDIFF_END plot(procrustes(ord, ca, choices=1:2, scaling=\"sites\")) ## Reconstruction of non-Euclidean distances with negative eigenvalues d <- vegdist(dune) ord <- wcmdscale(d, eig = TRUE) ## Only positive eigenvalues: cor(d, dist(ord$points)) #> [1] 0.9975185 ## Correction with negative eigenvalues: cor(d, sqrt(dist(ord$points)^2 - dist(ord$negaxes)^2)) #> [1] 1"},{"path":[]},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-7","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-7 (in development)","text":"vegan longer suggests tcltk, suggests vegan3d (version 1.3-0). See orditkplot section Deprecated Defunct.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-6-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.6-7 (in development)","text":"set functions add new points existing NMDS ordination metaMDS monoMDS. serves purpose adding new points existing eigenvector ordination (instance, predict.rda). main function MDSaddpoints. needs input rectangular matrix dissimilarities new points (rows) old points (columns). Support function dist2xy can extract needed matrices dissimilarities (old new) points, function designdist2 can directly find needed dissimilarities two data matrices. addition, analogue package can calculate rectangular dissimilarities, including many indices defined designdist2. function still experimental. particular user interface may need development. Comments welcome.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-7","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-7 (in development)","text":"adonis defunct: use adonis2 improved functionality. Disabled use summary get ordination scores: use scores! summary.cca see #644. summary.decorana defunct. nothing useful, can extract information scores weights. orditkplot code removed, try launch vegan3d::orditkplot loudly (deprecation message go vegan3d). experimental, may go direct error, depending user experience: please comment!","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-61","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6.1","title":"vegan 2.6-6.1","text":"CRAN release: 2024-05-21","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6.1","text":"C function do_wcentre (weighted centring) can segfault due protection error. problem found automatic CRAN checks. do_wcentre internal function called envfit (vectorfit), wcmdscale varpart (simpleCCA) Fixes bug #653.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-6","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6","title":"vegan 2.6-6","text":"CRAN release: 2024-05-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-6","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-6","text":"vegan depends R version 4.1.0. possible build vegan webR/wasm Fortran compiler. Issue #623.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-6","text":"Permutation tests CCA completely redesigned follow C.J.F ter Braak & D.E. te Beest: Environ Ecol Stat 29, 849–868 (2022) (https://doi.org/10.1007/s10651-022-00545-4). constraints now re-weighted permuted response data, partial model also residualized conditions (partial terms). vegan (release 2.4-6) tests identical Canoco, ter Braak & te Beest demonstrated results biased. old vegan (release 2.4-2 earlier) predictors re-weighted residualized. Re-weighting sufficient remove bias moderate variation weights, residualizing predictors necessary strongly varying weights. See discussion issue #542. new scheme concerns CCA weighted method, RDA dbRDA permutation unchanged. summary ordination results longer prints ordination scores often voluminous hide real summary; see issue #203. Ordination scores extracted scores function. breaks CRAN packages use summary.cca extract scores. switch use scores. maintainers contacted patch files suggested adapt change. See instructions fix packages. scores function constrained ordination (CCA, RDA,dbRDA) default return types scores (display = \"\"). Function can optionally return single type scores list one matrix instead returning matrix (new argument droplist). Constrained ordination objects (cca, rda, dbrda) fitted without formula interface can permutation tests (anova) \"axis\" \"onedf\". Models \"terms\" \"margin\" possible formula interface. Permutation tests constrained ordination objects (cca, rda, dbrda) = \"axis\" stop permutations later axis cutoff limit reached. Earlier cutoff exceeded. default stop permutations P-value 1 reached. analysis takes care P-values axes non-decreasing similarly Canoco. Coefficients effects prc models scaled similarly scaled vegan pre 2.5-1. change suggested Cajo ter Braak. Handling negative eigenvalues changed summary eigenvals. Negative eigenvalues given negative “explanation”, accumulated proportions add 1 last non-negative eigenvalue, 1 last negative eigenvalue. printed output capscale shows proportions real components ignores imaginary dimensions. consistent summary support methods. Issue #636. RsquareAdj capscale based positive eigenvalues, imaginary components ignored. stressplot.dbrda refuses handle partial models. first component variation can displayed dbrda internal (“working”) data structures additive. unconstrained model \"CA\", constrained \"CCA\" partial none. predict dbrda return actual type = \"working\". Earlier returned \"lc\" scores weighted eigenvalues. generated distances eigenvalues, though.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6","text":"Parallel processing inefficiently implemented slower non-parallel permutation tests constrained ordination adonis2. plot scores cca rda family methods gave error non-existing axes requested. Now ignores requests axes numbers higher result object. summary prc ignored extra parameters (const). -fitted models high number aliased variables caused rare failure adonis2 permutation tests constrained ordination methods (cca, rda, dbrda, capscale) arguments = \"margin\" = \"axis\". also concerned vif.cca intersetcor. Typically occurred high-order interactions factor variables. See issues #452 #622 methods accept rectangular raw data input alternative distances, pass arguments distance functions. arguments vegdist binary = TRUE pseudocount Aitchison distance. concerns dbrda, capscale bioenv. See issue #631 simper gave arbitrary p-values species occur subset. Now given NA. See https://stackoverflow.com/questions/77881877/ Rsquare.adj gave arbitrary p-values -fitted models residual variation. Now returns NA R2 adjusted. Automatic model building proceed cases, fixed ordiR2step returns R2 = 0 overfitted cases. constrained ordination methods issue warning model residual component. See issue #610 inertcomp(..., display = \"sites\", proportional = TRUE) gave wrong values.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"data-sets-2-6-6","dir":"Changelog","previous_headings":"","what":"Data Sets","title":"vegan 2.6-6","text":"Extended description BCI data sets avoid confusion. complete BCI survey includes stems 1 cm DBH, BCI data set vegan subset stems DBH 10 cm published Science 295, 666—669, 2002. data set intended demonstrate methods vegan ecological research suggest contacting BCI team using complete surveys made available Dryad.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-6","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-6","text":"adonis deprecated: use adonis2. several CRAN packages still use adonis although contacted authors June 2022 April 2024, printed message forthcoming deprecation since vegan 2.6-2. See issue #523. See instructions adapt packages functions use adonis2. orditkplot moved CRAN package vegan3d deprecated vegan. See issue #585 announcement #632 use summary extract ordination scores deprecated: use scores extract scores. version still allows extracting scores summary, fail next versions. summary.cca see instructions change package. Support removed ancient cca objects (results cca, rda, dbrda capscale) generated CRAN release 2.5 (2016). still stray relics, use newobject <- update(ancientobject) modernize result. .mcmc.oecosimu .mcmc.permat defunct: use toCoda. Code defunct functions completely removed.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-4","dir":"Changelog","previous_headings":"","what":"vegan 2.6-4","title":"vegan 2.6-4","text":"CRAN release: 2022-10-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-4","text":"Support scores ggplot2 graphics improved extended ordination functions. Suitable scores can requested argument tidy = TRUE, general available types scores returned data frame variable score labelling type. option implemented default method scores structured wcmdscale objects, glitches fixed rda family decorana. Previously tidy scores implemented cca, rda, dbrda family methods, metaMDS, envfit rarecurve. adonis2 anova constrained ordination results can perform sequential test one-degree--freedom effects multi-level factors split contrasts. Previously test available permutest. New summary function varpart brief overview. summary shows unique overall contributed variation set variables. fractions shared several sets variables divided equally contributing sets following Lai J, Zou Y, Zhang J, Peres-Neto P (2022) Methods Ecology Evolution, 13: 782–788. decorana estimates orthogonalized eigenvalues total inertia (scaled Chi-square). Orthogonalized eigenvalues can add total inertia. Together enabled implementing eigenvals, bstick screeplot methods decorana. Axis lengths reported decorana methods. Implemented tolerance method decorana. returns criterion used rescaling DCA, can used inspect success rescaling: constant 1 whole axis. New toCoda function transform sequential null model results oecosimu object can analysed coda convergence independence MCMC model. Function replaces .mcmc.oecosimu .mcmc.permat. metaMDS informative finding similar repeated results random starts uses less confusing language reporting results. Hellinger distance directly available vegdist. vegdist, betadiver raupcrick set attribute maxdist giving numeric value theoretical maximum dissimilarity index. many dissimilarities 1, √2 Chord Hellinger distances, instance. attribute NA open indices ceiling. betadiver three similarity indices set maxdist 0. metaMDS defaults halfchange scaling dissimilarities numeric maxdist attribute, adapt threshold ceiling value. open indices without ceiling, threshold scale dissimilarities. metaMDS used simple test detect index ceiling 1, test now robust can also find maximum values. inference made, function broadcast message assumed value ceiling. Mountford index vegdist now scaled maximum value log(2). Earlier Mountford distances scaled maximum 1. hatvalues constrained ordination objects can sometimes practically 1 1, now cases exactly 1. cases rstandard, rstudent cooks.distance NaN. behaviour similar stats::lm.influence functions. .rad can handle multi-row data frames matrices return list Rank-Abundance data row. Earlier one site handled. decostand returns attribute parameters settings variables used standardization. New function decobackstand can use parameters reconstruct original non-standardized data. Back-transformation exact round-errors, although attempt keep original zeros exact. Back-transformation possible methods pa, rank rrank implemented alr. Back-transformation queried https://stackoverflow.com/questions/73263526/ Rarefaction rarefaction-based methods make sense original observed counts give misleading results data multiplied rare species removed. Observed counts usually singletons (species count one), method issue warning minimum count higher one (may false positive, inspect data). Concerns functions rarefy, drarefy, rrarefy, rarecurve, specaccum(..., method=\"rarefy\"), rareslope avgdist. See github discussion #537. avgdist exposes .dist arguments can return \"dist\"ance objects appear lower triangles instead appearing symmetric matrices. betadisper plots accept col argument (PR #300).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-4","text":"decorana returned wrong results Hill’s piecewise transformation (arguments /) used, unless downweighting also used. scores failed metaMDS result species scores. Bug introduced release 2.6-2. Issue raised https://stackoverflow.com/questions/72483924/ tolerance.cca failed one axis (choice) requested. decostand(..., method=\"alr\") accept name reference, fail cases. CRAN package proxy interfered simper caused obscure error (github issue #528).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-4","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-4","text":"adonis way deprecation. Use adonis2 instead. .mcmc.oecosimu .mcmc.permat deprecated: used S3 methods without depending coda package. Use toCoda instead.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-2","dir":"Changelog","previous_headings":"","what":"vegan 2.6-2","title":"vegan 2.6-2","text":"CRAN release: 2022-04-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-2","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-2","text":"Compiled code adapted changes R 4.2.0. See issues #447, #507. Cross-references function packages adapted stringent tests CRAN","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-2","text":"Aitchison robust Aitchison distances added vegdist. Similar data transformations added decostand. Several functions can return “tidy” data structures can used ggplot2 graphics: rarecurve, scores functions constrained ordination (cca etc.), decorana, envfit, metaMDS. scores.envfit gained argument arrow.mul. vegan plot functions used automatically, now easier use envfit non-vegan plotting. Added function simpson.unb unbiased Simpson diversity robust variation sample sizes. diversity gained argument group calculate indices pooled data. Discussed issue #393. simper much faster even though parallel processing implemented new code. pairs function added plot permustats variables . varpart accepts dissimilarities given symmetric square matrix instead \"dist\" object per wish issue #497. metaMDS adopted user-friendly policy, trymax always maximum number tries. See dicussion https://stackoverflow.com/questions/66748605/. adonis2 accepts strata. adonis2 new main function replaces old adonis. See issue #427. Fisher alpha (fisherfit) badly suited extreme communities follow Fisher’s model. Now fisherfit returns NA communities 0 1 species, issues warning communities consisting singletons extreme Fisher alpha. adipart multipart formulae automatically add unique id constant. always sandwich requested grouping alpha gamma diversities, change results requested groupings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-2","text":"anova function failed marginal tests constrained partial ordination model (cca, rda etc.) interaction terms. Issue #463. Constrained ordination (cca etc.) gave misleading results external variables (constraints, condition) constant explained nothing. decorana fail axes zero eigenvalues. Issue #401. Species accumulation (specaccum) failed one species, several “communities”. Issue #501. Parallel processing failed Windows socket clusters permutest betadisper. Issue #369. orditorp failed numeric labels supplied. Reported https://stackoverflow.com/questions/69272366/. Argument summarize accidentally dropped goodness.cca 2017. taxa2dist failed one usable taxonomic level. See https://stackoverflow.com/questions/67231431/.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-2","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-2","text":"Function adonis2 replace adonis. humpfit functions defunct removed. available non-CRAN package natto https://github.com/jarioksa/natto. commisimulator defunct. Use simulate nullmodel objects. permuted.index finally defunct (deprecated vegan 2.2-0). .mlm defunct. Use functions documented influence.cca, hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca others.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-7","dir":"Changelog","previous_headings":"","what":"vegan 2.5-7","title":"vegan 2.5-7","text":"CRAN release: 2020-11-28","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-7","text":"Several distance-based functions failed distances zero (betadisper, capscale, isomap, monoMDS, pcnm, wcmdscale). Reported github issue #372. Non-linear self-starting regression models SSarrhenius, SSgitay, SSgleason SSlomolino failed future R. failure caused internal changes R-devel. Github issue #382. Arrow labels wrong position plot.envfit(..., add = FALSE). rarecurve added unnecessary names results. Github issue #352. permutest betadisper failed parallel processing Windows systems socket clusters used. Github issue #369.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-7","text":"Chi-square Chord distances added vegdist. distances can calculated Euclidean distances transformed data, actually available earlier, many users notice . monoMDS (hence metaMDS) uses stricter convergence criteria. improves possibilities find stable solutions. However, users may still need tweak convergence criteria data. See discussion Github issue #354. text functions constrained ordination plots (cca, rda, dbrda, capscale) accept now expression labels. allows using subscripts, superscripts mathematical expressions. New support function labels.cca returns current text labels authors can change desired ones. See github issue #374. vegemite returns invisibly final formatted table allowing processing. ordiplot passes cex argument linestack decorana plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-6","dir":"Changelog","previous_headings":"","what":"vegan 2.5-6","title":"vegan 2.5-6","text":"CRAN release: 2019-09-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-6","text":"vegdist silently accepted missing values (NA) removed analysis also option na.rm = FALSE. behaviour introduced vegan version 2.5-1. See GitHub issue #319. labels displaced bunch arrows drawn origin ordination graph envfit. See GitHub issue #315. Hill scale coverscale open-ended limited percent data, unlike traditional cover class scales undefined 100% cover.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-6","text":"results .rad longer print index attribute: attribute still object, printing made output messy.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-5","dir":"Changelog","previous_headings":"","what":"vegan 2.5-5","title":"vegan 2.5-5","text":"CRAN release: 2019-05-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-5","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-5","text":"vegan depends R 3.4.0 higher. next vegan release may increase dependence R 3.6.0. R 3.6.0 improved method find random indices permuting sampling data. Vegan relies now R functions ecological null models (functions nullmodel, oecosimu, commsim, permatfull, permatswap others). Technically change compatible R 3.4.0 later, can gain benefits improved code current release R. null models may change due change, certainly change R 3.6.0. See NEWS R 3.6.0 release discussion github issue #312. vegan permutation routines rely permute, gain similar benefits improved randomness upgrade R. Thanks new R dependence, sigma constrained ordination results works without workarounds vegan 2.5-2. fixes completely issue discussed #274. Vegan test results reproduced older versions R 3.6.0. worried , upgrade R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-5","text":"metaMDS failed scaling results engine monoMDS used. However, recommend use monoMDS. See github issue #310.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-5","text":"betadisper changed interpretation negative squared distances give complex-valued distances. Now regarded zero-distances whereas earlier used modulus. change results cases negative squared distances. discussion, see github issue #306.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-4","dir":"Changelog","previous_headings":"","what":"vegan 2.5-4","title":"vegan 2.5-4","text":"CRAN release: 2019-02-04","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-testing-2-5-4","dir":"Changelog","previous_headings":"","what":"Installation and Testing","title":"vegan 2.5-4","text":"code interpreting formula change R 3.6.0, makes constrained ordination methods (cca, rda, dbrda, capscale) fail. See github issue #299. R 3.6.0 introduces new environment variable _R_CHECK_LENGTH_1_LOGIC2_, several functions fail variable set. Changes concern ordiplot, plot summary constrained ordination objects, ordixyplot. See github issue #305.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-4","text":"decorana gave incorrect results downweighting used (argument iweigh = 1). bug introduced vegan 2.5-1 reported github issue #303. goodness constrained ordination methods failed constraints rank = 1 (one constraining variable). Reported Pierre Legendre.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-4","text":"Adjusted R2 enabled partial RDA models (functions rda dbrda) partial CCA models (function cca) function RsquareAdj. feature disabled vegan 2.5-1 . RDA, calculation similar vegan 2.4-6 earlier. Partial CCA now consistent RDA differs earlier implementation. methods, partial models consistent varpart. See github issue #295.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-3","dir":"Changelog","previous_headings":"","what":"vegan 2.5-3","title":"vegan 2.5-3","text":"CRAN release: 2018-10-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-3","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-3","text":"Tests numerical analysis written robustly give similar results alternative platforms versions R BLAS/Lapack libraries. See github issue #282.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-3","text":"Constrained ordination gave misleading results constraints conditions data NULL variables. rarely happens normal usage, happen marginal anova reported github issue #291. Several functions numerical analysis wrongly accepted non-numeric data (instance, factors) gave either meaningless results confusing error messages. Fixed functions include beals, designdist, diversity, gdispweight, indpower, spantree, specpool, tsallis, tsallisaccum vegdist. See github issue #292. envfit vectors fail missing data. original data scaled centred similarly simulations simulate.rda several simulations returned simmat object (compatible nullmodel simulations can used oecosimu).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-3","text":"anosim checks input avoid confusing error messages like reported StackOverflow question 52082743. Broken-stick distribution (function bstick) longer calculated distance-based Redundancy Analysis (dbrda) negative eigenvalues, clear done. Now dbrda capscale similar respect. print function betadisper results gained new argument neigen select number eigenvalues shown. print robust number eigenvalues lower requested neigen.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-5-3","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.5-3","text":"Function humpfit moved natto package still available https://github.com/jarioksa/natto. scheduled complete removal vegan 2.6-0.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-2","dir":"Changelog","previous_headings":"","what":"vegan 2.5-2","title":"vegan 2.5-2","text":"CRAN release: 2018-05-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-compatibility-2-5-2","dir":"Changelog","previous_headings":"","what":"Installation and Compatibility","title":"vegan 2.5-2","text":"Vegan declares dependence R version 3.2.0. dependence yet noticed previous vegan release. However, generic sigma function defined R-3.3.0, therefore sigma.cca vegan must spelt completely using R-3.2.x. See discussion issue #274. CRAN package klaR function rda, loaded together vegan clashes vegan rda Redundancy Analysis. Vegan tries mitigate problem. cases vegan functions used vegan loaded klaR, error message issued klaR objects handled vegan functions. klaR also tricked print informative message handles vegan objects. However, vegan namespace can attached automatically start-klaR functions take precedence. reported issue #277. Bioconductor package phyloseq problem vegdist function dissimilarities. problem can fixed re-installing phyloseq source package. , must either downgrade vegan version 2.4-6 wait till Bioconductor binary packages upgraded. reported Stackoverflow, vegan issue #272, phyloseq issues #918 #921.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-2","text":"Plotting betadisper failed groups one member. Reported Stackoverflow “Error: Incorrect .dimensions” plotting multivariate data Vegan. Permutation tests constrained ordination (anova.cca, permutest.cca) fail parallel processing socket clusters. Socket clusters always used Windows can also used operating systems created makeCluster. See issue #276.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-1","dir":"Changelog","previous_headings":"","what":"vegan 2.5-1","title":"vegan 2.5-1","text":"CRAN release: 2018-04-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-5-1","dir":"Changelog","previous_headings":"","what":"GENERAL","title":"vegan 2.5-1","text":"major new release changes package: Nearly 40% program files changed previous release. Please report regressions issues https://github.com/vegandevs/vegan/issues/. Compiled code used much extensively, compiled functions use .Call interface. gives smaller memory footprint also faster. wall clock time, greatest gains permutation tests constrained ordination methods (anova.cca) binary null models (nullmodel). Constrained ordination functions (cca, rda, dbrda, capscale) completely rewritten share code. makes consistent robust. internal structure changed constrained ordination objects, scripts may fail try access result object directly. never guarantee unchanged internal structure, scripts changed use provided support functions access result object (see documentation cca.object github issue #262). support analysis functions may longer work result objects created previous vegan versions. use update(old.result.object) fix old result objects. See github issues #218, #227. vegan includes tests run checking package installation. See github issues #181, #271. informative messages (warnings, notes error messages) cleaned unified also makes possible provide translations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.5-1","text":"avgdist: new function find averaged dissimilarities several random rarefactions communities. Code Geoffrey Hannigan. See github issues #242, #243, #246. chaodist: new function similar designdist, uses Chao terms supposed take account effects unseen species (Chao et al., Ecology Letters 8, 148-159; 2005). Earlier Jaccard-type Chao dissimilarity vegdist, new code allows defining kind Chao dissimilarity. New functions find influence statistics constrained ordination objects: hatvalues, sigma, rstandard, rstudent, cooks.distance, SSD, vcov, df.residual. earlier found via .mlm function deprecated. See github issue #234. boxplot added permustats results display (standardized) effect sizes. sppscores: new function add replace species scores distance-based ordination dbrda, capscale metaMDS. Earlier dbrda species scores, species scores capscale metaMDS based raw input data may consistent used dissimilarity measure. See github issue #254. cutreeord: new function similar stats::cutree, numbers cluster order appear dendrogram (left right) instead labelling order appeared data. sipoo.map: new data set locations sizes islands Sipoo archipelago bird data set sipoo.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-constrained-ordination-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Constrained Ordination","title":"vegan 2.5-1","text":"inertia Correspondence Analysis (cca) called “scaled Chi-square” instead using name little known statistic. elements Constraints Conditions data frames non-formula call rda cca, automatically expanded model matrices can contain factor variables. Earlier numerical model matrices factors used formula interface. Regression scores constraints can extracted plotted constrained ordination methods. See github issue #226. Full model (model = \"full\") enabled permutations tests constrained ordination results anova.cca permutest.cca. permutest.cca gained new option = \"onedf\" perform tests sequential one degree--freedom contrasts factors. option (yet) enabled anova.cca. permutation tests robust, scoping issues fixed. Permutation tests use compiled C code much faster. See github issue #211. permutest printed layout similar anova.cca. eigenvals gained new argument (model) select either constrained unconstrained scores. old argument (constrained) deprecated. See github issue #207. summary.eigenvals returns matrix instead list containing matrix. Adjusted R2 calculated partial ordination, unclear done (function RsquareAdj). ordiresids can display standardized studentized residuals. Function construct model.frame model.matrix constrained ordination robust fail fewer cases. goodness inertcomp constrained ordination result object longer option find distances: explained variation available. inertcomp gained argument unity. give “local contributions beta-diversity” (LCBD) “species contribution beta-diversity” (SCBD) Legendre & De Cáceres (Ecology Letters 16, 951-963; 2012). goodness disabled capscale. prc gained argument const general scaling results similarly rda. prc uses regression scores Canoco-compatibility.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-null-model-communities-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Null Model Communities","title":"vegan 2.5-1","text":"C code swap-based binary null models made efficients, models faster. Many models selected 2 times 2 submatrix, generated four random numbers (two rows, two columns). Now skip selecting third fourth random number obvious matrix swapped. Since time used generating random numbers functions, candidates rejected, speeds functions. However, also means random number sequences change previous vegan versions, old binary model results replicated exactly. See github issues #197, #255 details timing. Ecological null models (nullmodel, simulate, make.commsim, oecosimu) gained new null model \"greedyqswap\" can radically speed quasi-swap models minimal risk introducing bias. Backtracking written C much faster. However, backtracking models biased, provided classic legacy models.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-other-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Other Functions","title":"vegan 2.5-1","text":"adonis2 gained column R2 similarly old adonis. Great part R code decorana written C makes faster reduces memory footprint. metaMDS results gained new points text methods. ordiplot ordination plot functions can chained points text functions allowing use magrittr pipes. points text functions gained argument draw arrows allowing use drawing biplots adding vectors environmental variables ordiplot. Since many ordination plot methods return invisible \"ordiplot\" object, points text methods also work . See github issue #257. Lattice graphics (ordixyplot) ordination can add polygons enclose points panel complete data. ordicluster gained option suppress drawing plots can easily embedded functions calculations. .rad returns index included taxa attribute. Random rarefaction (function rrarefy) uses compiled C code much faster. plot specaccum can draw short horizontal bars vertical error bars. See StackOverflow question 45378751. decostand gained new standardization methods rank rrank replace abundance values ranks relative ranks. See github issue #225. Clark dissimilarity added vegdist (calculated designdist). designdist evaluates minimum terms compiled code, function faster vegdist also dissimilarities using minimum terms. Although designdist usually faster vegdist, numerically less stable, particular large data sets. swan passes type argument beals. tabasco can use traditional cover scale values function coverscale. Function coverscale can return scaled values integers numerical analysis instead returning characters printing. varpart can partition Chi-squared inertia correspondence analysis new argument chisquare. adjusted R2 based permutation tests, replicate analysis random variation. explanatory tables can data frames factors single factors varpart automatically expanded model matrices. Earlier factors used one-sided model formulae. Based code suggested Daniel Borcard, Univ. Montréal.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-1","text":"long Condition() statements (> 500 characters) failed partial constrained ordination models (cca, rda, dbrda, capscale). problem detected StackOverflow question 49249816. Labels adjusted arrows rescaled envfit plots. See StackOverflow question 49259747. ordiArrowMul failed one arrow plotted envfit.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-5-1","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.5-1","text":".mlm function constrained correspondence analysis deprecated favour new functions directly give influence statistics. See github issue #234. commsimulator now defunct: use simulate nullmodel objects. ade4 cca objects longer handled vegan: ade4 cca since version 1.7-8 (August 9, 2017).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-6","dir":"Changelog","previous_headings":"","what":"vegan 2.4-6","title":"vegan 2.4-6","text":"CRAN release: 2018-01-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-6","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-6","text":"CRAN packages longer allowed use FORTRAN input, read.cep function used FORTRAN format read legacy CEP Canoco files. avoid NOTEs WARNINGs, function re-written R. new read.cep less powerful fragile, can read data “condensed” format, can fail several cases successful old code. old FORTRAN-based function still available cepreader. See github issue #263. cepreader package developed https://github.com/vegandevs/cepreader.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-6","text":"functions rarefaction (rrarefy), species abundance distribution (preston) species pool (estimateR) need exact integer data, test allowed small fuzz. functions worked correctly original data, data transformed back-transformed, pass integer test fuzz give wrong results. instance, sqrt(3)^2 pass test 3, interpreted strictly integer 2. See github issue #259.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-6","text":"ordiresids uses now weighted residuals cca results.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-5","dir":"Changelog","previous_headings":"","what":"vegan 2.4-5","title":"vegan 2.4-5","text":"CRAN release: 2017-12-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-5","text":"Several “Swap & Shuffle” null models generated wrong number initial matrices. Usually generated many, dangerous, slow. However, random sequences change fix. Lattice graphics ordination (ordixyplot friends) colour arrows groups instead randomly mixed colours. Information constant mirrored permutations omitted reporting permutation tests (e.g., anova constrained ordination).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-5","text":"ordistep improved interpretation scope: lower scope missing, formula starting solution taken lower scope instead using empty model. See Stackoverflow question 46985029. fitspecaccum gained new support functions nobs logLik allow better co-operation packages functions. See GitHub issue #250. “backtracking” null model community simulation faster. However, “backtracking” biased legacy model used except comparative studies.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-4","dir":"Changelog","previous_headings":"","what":"vegan 2.4-4","title":"vegan 2.4-4","text":"CRAN release: 2017-08-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-4","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-4","text":"orditkplot longer give warnings CRAN tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-4","text":"anova(..., = \"axis\") constrained ordination (cca, rda, dbrda) ignored partial terms Condition(). inertcomp summary.cca failed constrained component defined, explained nothing zero rank. See StackOverflow: R - Error message RDA analysis - vegan package. Labels longer cropped meandist plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-4","text":"significance tests axes constrained ordination use now forward testing strategy. extensive analysis indicated previous marginal tests biased. conflict Legendre, Oksanen & ter Braak, Methods Ecol Evol 2, 269–277 (2011) regarded marginal tests unbiased. Canberra distance vegdist can now handle negative input entries similarly latest versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-3","dir":"Changelog","previous_headings":"","what":"vegan 2.4-3","title":"vegan 2.4-3","text":"CRAN release: 2017-04-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-3","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-3","text":"vegan registers native C Fortran routines. avoids warnings model checking, may also give small gain speed. Future versions vegan deprecate remove elements pCCA$Fit, CCA$Xbar, CA$Xbar cca result objects. release provides new function ordiYbar able construct elements current future releases. Scripts functions directly accessing elements switch ordiYbar smooth transition.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-3","text":".mlm methods constrained ordination include zero intercept give correct residual degrees freedom derived statistics. biplot method rda passes correlation argument scaling algorithm. Biplot scores wrongly centred cca caused small error values. Weighting centring corrected intersetcor spenvcor. fix can make small difference analysing cca results. Partial models correctly handled intersetcor. envfit ordisurf functions failed applied species scores. Non-standard variable names can used within Condition() partial ordination. Partial models used internally within several functions, problem reported Albin Meyer (Univ Lorraine, Metz, France) ordiR2step using variable name contained hyphen (wrongly interpreted minus sign partial ordination). ordispider pass graphical arguments used show difference LC WA scores constrained ordination. ordiR2step uses forward selection avoid several problems model evaluation. tolerance function return NaN cases returned 0. Partial models correctly analysed. Misleading (non-zero) tolerances sometimes given species occurred sampling units one species.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-2","dir":"Changelog","previous_headings":"","what":"vegan 2.4-2","title":"vegan 2.4-2","text":"CRAN release: 2017-01-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-2","text":"Permutation tests (permutests, anova) first axis failed constrained distance-based ordination (dbrda, capscale). Now capscale also throw away negative eigenvalues first eigenvalues tested. permutation tests first axis now faster. problem reported Cleo Tebby fixes discussed GitHub issue #198 pull request #199. support functions dbrda capscale gave results components wrong scale. Fixes stressplot, simulate, predict fitted functions. intersetcor use correct weighting cca results slightly . anova permutest failed betadisper fitted argument bias.adjust = TRUE. Fixes Github issue #219 reported Ross Cunning, O’ahu, Hawaii. ordicluster return invisibly coordinates internal points (clusters points joined), last rows contained coordinates external points (ordination scores points). cca method tolerance returning incorrect values second axis sample heterogeneities species tolerances. See issue #216 details.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-2","text":"Biplot scores scaled similarly site scores constrained ordination methods cca, rda, capscale dbrda. Earlier unscaled (technically, equal scaling axes). tabasco adds argument scale colours rows columns addition old equal scale whole plot. New arguments labRow labCex can used change column row labels. Function also takes care -zero observations coloured: earlier tiny observed values merged zeros distinct plots. Sequential null models somewhat faster (10%). Non-sequential null models may marginally faster. null models generated function nullmodel also used oecosimu. vegdist much faster. used clearly slower stats::dist, now nearly equally fast dissimilarity measure. Handling data= formula interface robust, messages user errors improved. fixes points raised Github issue #200. families orders dune.taxon updated APG IV (Bot J Linnean Soc 181, 1–20; 2016) corresponding classification higher levels (Chase & Reveal, Bot J Linnean Soc 161, 122-127; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-1","dir":"Changelog","previous_headings":"","what":"vegan 2.4-1","title":"vegan 2.4-1","text":"CRAN release: 2016-09-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-4-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.4-1","text":"Fortran code modernized avoid warnings latest R. modernization visible effect functions. Please report suspect cases vegan issues.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-1","text":"Several support functions ordination methods failed solution one ordination axis, instance, one constraining variable CCA, RDA friends. concerned goodness constrained ordination, inertcomp, fitted capscale, stressplot RDA, CCA (GitHub issue #189). goodness CCA & friends ignored choices argument (GitHub issue #190). goodness function consider negative eigenvalues db-RDA (function dbrda). Function meandist failed cases one groups one observation. linestack handle expressions labels. regression discussed GitHub issue #195. Nestedness measures nestedbetajac nestedbetasor expecting binary data cope quantitative input evaluating Baselga’s matrix-wide Jaccard Sørensen dissimilarity indices. Function .mcmc cast oecosimu result MCMC object (coda package) failed one chain.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-1","text":"diversity function returns now NA observation NA values instead returning 0. function also checks input refuses handle data negative values. GitHub issue #187. rarefy function work robustly marginal case user asks one individual can one species zero variance. Several functions robust factor arguments contain missing values (NA): betadisper, adipart, multipart, hiersimu, envfit constrained ordination methods cca, rda, capscale dbrda. GitHub issues #192 #193.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-0","dir":"Changelog","previous_headings":"","what":"vegan 2.4-0","title":"vegan 2.4-0","text":"CRAN release: 2016-06-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"distance-based-analysis-2-4-0","dir":"Changelog","previous_headings":"","what":"Distance-based Analysis","title":"vegan 2.4-0","text":"Distance-based methods redesigned made consistent ordination (capscale, new dbrda), permutational ANOVA (adonis, new adonis2), multivariate dispersion (betadisper) variation partitioning (varpart). methods can produce negative eigenvalues several popular semimetric dissimilarity indices, handled similarly functions. Now functions designed McArdle & Anderson (Ecology 82, 290–297; 2001). dbrda new function distance-based Redundancy Analysis following McArdle & Anderson (Ecology 82, 290–297; 2001). metric dissimilarities, function equivalent old capscale, negative eigenvalues semimetric indices handled differently. dbrda dissimilarities decomposed directly conditions, constraints residuals negative eigenvalues, components can imaginary dimensions. Function mostly compatible capscale constrained ordination methods, full compatibility achieved (see issue #140 Github). function based code Pierre Legendre. old capscale function constrained ordination still based real components, total inertia components assessed similarly dbrda. significance tests differ previous version, function oldCapscale cast capscale result similar form previously. adonis2 new function permutational ANOVA dissimilarities. based algorithm dbrda. function can perform overall tests independent variables well sequential marginal tests term. old adonis still available, can perform sequential tests. settings, adonis adonis2 give identical results (see Github issue #156 differences). Function varpart can partition dissimilarities using algorithm dbrda. Argument sqrt.dist takes square roots dissimilarities can change many popular semimetric indices metric distances capscale, dbrda, wcmdscale, adonis2, varpart betadisper (issue #179 Github). Lingoes Cailliez adjustments change dissimilarity metric distance capscale, dbrda, adonis2, varpart, betadisper wcmdscale. Earlier Cailliez adjustment capscale (issue #179 Github). RsquareAdj works capscale dbrda allows using ordiR2step model building.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-0","text":"specaccum: plot failed line type (lty) given. Reported Lila Nath Sharma (Univ Bergen, Norway)","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-4-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.4-0","text":"ordibar new function draw crosses standard deviations standard errors ordination diagrams instead corresponding ellipses. Several permustats results can combined new c() function. New function smbind binds together null models row, column replication. sequential models bound together, can treated parallel chains subsequent analysis (e.g., .mcmc). See issue #164 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-0","text":"Null model analysis upgraded: New \"curveball\" algorithm provides fast null model fixed row column sums binary matrices Strona et al. (Nature Commun. 5: 4114; 2014). \"quasiswap\" algorithm gained argument thin can reduce bias null models. \"backtracking\" now much faster, still slow, provided mainly allow comparison better faster methods. Compiled code can now interrupted null model simulations. designdist can now use beta diversity notation (gamma, alpha) easier definition beta diversity indices. metaMDS new iteration strategy: Argument try gives minimum number random starts, trymax maximum number. Earlier hand try gave maximum number, now run least try times. reduces risk trapped local optimum (issue #154 Github). convergent solutions, metaMDS now tabulate stopping criteria (trace = TRUE). can help deciding criteria made stringent number iterations increased. documentation monoMDS metaMDS give detailed information convergence criteria. summary permustats prints now P-values, test direction (alternative) can changed. qqmath function permustats can now plot standardized statistics. partial solution issue #172 Github. MDSrotate can rotate ordination show maximum separation factor levels (classes) using linear discriminant analysis (lda MASS package). adipart, hiersimu multipart expose argument method specify null model. RsquareAdj works cca allows using ordiR2step model building. code developed Dan McGlinn (issue #161 Github). However, cca still used varpart. ordiellipse ordihull allow setting colours, line types graphical parameters. alpha channel can now given also real number 0 … 1 addition integer 0 … 255. ordiellipse can now draw ellipsoid hulls enclose points group. ordicluster, ordisegments, ordispider lines plot functions isomap spantree can use mixture colours connected points. behaviour similar analogous functions vegan3d package. plot betadisper configurable. See issues #128 #166 Github details. text points methods orditkplot respect stored graphical parameters. Environmental data Barro Colorado Island forest plots gained new variables Harms et al. (J. Ecol. 89, 947–959; 2001). Issue #178 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-4-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.4-0","text":"Function metaMDSrotate removed replaced MDSrotate. density densityplot methods various vegan objects deprecated replaced density densityplot permustats. Function permustats can extract permutation simulation results vegan result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-5","dir":"Changelog","previous_headings":"","what":"vegan 2.3-5","title":"vegan 2.3-5","text":"CRAN release: 2016-04-09","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-5","text":"eigenvals fails prcomp results R-devel. next version prcomp argument limit number eigenvalues shown (rank.), breaks eigenvals vegan. calibrate failed cca friends rank given.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-4","dir":"Changelog","previous_headings":"","what":"vegan 2.3-4","title":"vegan 2.3-4","text":"CRAN release: 2016-02-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-4","text":"betadiver index 19 wrong sign one terms. linestack failed labels given, input scores names. Reported Jeff Wood (ANU, Canberra, ACT).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-3-4","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.3-4","text":"vegandocs deprecated. Current R provides better tools seeing extra documentation (news() browseVignettes()).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vignettes-2-3-4","dir":"Changelog","previous_headings":"","what":"Vignettes","title":"vegan 2.3-4","text":"vignettes built standard R tools can browsed browseVignettes. FAQ-vegan partitioning accessible vegandocs function.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"building-2-3-4","dir":"Changelog","previous_headings":"","what":"Building","title":"vegan 2.3-4","text":"Dependence external software texi2dvi removed. Version 6.1 texi2dvi incompatible R prevented building vegan. FAQ-vegan earlier built texi2dvi uses now knitr. , vegan now dependent R-3.0.0. Fixes issue #158 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-3","dir":"Changelog","previous_headings":"","what":"vegan 2.3-3","title":"vegan 2.3-3","text":"CRAN release: 2016-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-3","text":"metaMDS monoMDS fail input dissimilarities huge: reported case magnitude 1E85. Fixes issue #152 Github. Permutations failed defined permute control structures estaccum, ordiareatest, renyiaccum tsallisaccum. Reported Dan Gafta (Cluj-Napoca) renyiaccum. rarefy gave false warnings input vector single sampling unit. extrapolated richness indices specpool needed number doubletons (= number species occurring two sampling units), failed one sampling unit supplied. extrapolated richness estimated single sampling unit, now cases handled smoothly instead failing: observed non-extrapolated richness zero standard error reported. issue reported StackOverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-3","text":"treedist treedive refuse handle trees reversals, .e, higher levels homogeneous lower levels. Function treeheight estimate total height absolute values branch lengths. Function treedive refuses handle trees negative branch heights indicating negative dissimilarities. Function treedive faster. gdispweight works input data matrix instead data frame. Input dissimilarities supplied symmetric matrices data frames robustly recognized anosim, bioenv mrpp.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-2","dir":"Changelog","previous_headings":"","what":"vegan 2.3-2","title":"vegan 2.3-2","text":"CRAN release: 2015-11-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-2","text":"Printing details gridded permutation design fail grid within-plot level. ordicluster joined branches wrong coordinates cases. ordiellipse ignored weights calculating standard errors (kind = \"se\"). influenced plots cca, also influenced ordiareatest.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-2","text":"adonis capscale functions recognize symmetric square matrices dissimilarities. Formerly dissimilarities given \"dist\" objects produced dist vegdist functions, data frames matrices regarded observations x variables data confuse users (e.g., issue #147). mso accepts \"dist\" objects distances among locations alternative coordinates locations. text, points lines functions procrustes analysis gained new argument truemean allows adding procrustes items plots original analysis. rrarefy returns observed non-rarefied communities (warning) users request subsamples larger observed community instead failing. Function drarefy similar returned sampling probabilities 1, now also issues warning. Fixes issue #144 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-1","dir":"Changelog","previous_headings":"","what":"vegan 2.3-1","title":"vegan 2.3-1","text":"CRAN release: 2015-09-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-1","text":"Permutation tests always correctly recognize ties observed statistic result low P-values. happen particular predictor variables factors (classes). changes concern functions adonis, anosim, anova permutest functions cca, rda capscale, permutest betadisper, envfit, mantel mantel.partial, mrpp, mso, oecosimu, ordiareatest, protest simper. also fixes issues #120 #132 GitHub. Automated model building constrained ordination (cca, rda, capscale) step, ordistep ordiR2step fail aliased candidate variables, constraints completely explained variables already model. regression introduced vegan 2.2-0. Constrained ordination methods cca, rda capscale treat character variables factors analysis, return centroids plotting. Recovery original data metaMDS computing WA scores species fail expression supplied argument comm long & got deparsed multiple strings. metaMDSdist now returns (possibly modified) data frame community data comm attribute \"comm\" returned dist object. metaMDS now uses compute WA species scores NMDS. addition, deparsed expression comm now robust long expressions. Reported Richard Telford. metaMDS monoMDS rejected dissimilarities missing values. Function rarecurve check input cause confusing error messages. Now function checks input data integers can interpreted counts individuals sampling units species. Unchecked bad inputs reason problems reported Stackoverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-1","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-1","text":"Scaling ordination axes cca, rda capscale can now expressed descriptive strings \"none\", \"sites\", \"species\" \"symmetric\" tell kind scores scaled eigenvalues. can modified arguments hill cca correlation rda. old numeric scaling can still used. permutation data can extracted anova results constrained ordination (cca, rda, capscale) analysed permustats function. New data set BCI.env site information Barro Colorado Island tree community data. useful variables UTM coordinates sample plots. variables constant nearly constant little use normal analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-0","dir":"Changelog","previous_headings":"","what":"vegan 2.3-0","title":"vegan 2.3-0","text":"CRAN release: 2015-05-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-0","text":"Constrained ordination functions cca, rda capscale now robust. Scoping data set names variable names much improved. fix numerous long-standing problems, instance reported Benedicte Bachelot (email) Richard Telford (Twitter), well issues #16 #100 GitHub. Ordination functions cca rda silently accepted dissimilarities input although analysis makes sense methods. Dissimilarities analysed distance-based redundancy analysis (capscale). variance conditional component -estimated goodness rda results, results wrong partial RDA. problems reported R-sig-ecology message Christoph von Redwitz.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"windows-2-3-0","dir":"Changelog","previous_headings":"","what":"Windows","title":"vegan 2.3-0","text":"orditkplot add file type identifier saved graphics Windows although required. problem concerned Windows OS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-0","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-0","text":"goodness function constrained ordination (cca, rda, capscale) redesigned. Function gained argument addprevious add variation explained previous ordination components axes statistic = \"explained\". option, model = \"CCA\" include variation explained partialled-conditions, model = \"CA\" include accumulated variation explained conditions constraints. former behaviour addprevious = TRUE model = \"CCA\", addprevious = FALSE model = \"CA\". argument effect statistic = \"distance\", always show residual distance previous components. Formerly displayed residual distance currently analysed model. Functions ordiArrowMul ordiArrowTextXY exported can used normal interactive sessions. functions used scale bunch arrows fit ordination graphics, formerly internal functions used within vegan functions. orditkplot can export graphics SVG format. SVG vector graphics format can edited several external programs, Illustrator Inkscape. Rarefaction curve (rarecurve) species accumulation models (specaccum, fitspecaccum) gained new functions estimate slope curve given location. Originally based response R-SIG-ecology query. rarefaction curves, function rareslope, species accumulation models specslope. functions based analytic equations, can also evaluated interpolated non-integer values. specaccum models functions can evaluated analytic models \"exact\", \"rarefaction\" \"coleman\". \"random\" \"collector\" methods can use finite differences (diff(fitted())). Analytic functions slope used non-linear regression models known fitspecaccum. Species accumulation models (specaccum) non-liner regression models species accumulation (fitspecaccum) work consistently weights. cases, models defined using number sites independent variable, weights means observations can non-integer numbers virtual sites. predict models also use number sites newdata, analytic models can estimate expected values non-integer number sites, non-analytic randomized collector models can interpolate non-integer values. fitspecaccum gained support functions AIC deviance. varpart plots four-component models redesigned following Legendre, Borcard & Roberts Ecology 93, 1234–1240 (2012), use now four ellipses instead three circles two rectangles. components now labelled plots, circles ellipses can easily filled transparent background colour.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-1","dir":"Changelog","previous_headings":"","what":"vegan 2.2-1","title":"vegan 2.2-1","text":"CRAN release: 2015-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-1","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-1","text":"maintenance release avoid warning messages caused changes CRAN repository. namespace usage also stringent avoid warnings notes development versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-1","text":"vegan can installed loaded without tcltk package. tcltk package needed orditkplot function interactive editing ordination graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-1","text":"ordisurf failed gam package loaded due namespace issues: support functions gam used instead mgcv functions. tolerance function failed unconstrained correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-1","text":"estimateR uses exact variance formula bias-corrected Chao estimate extrapolated number species. new formula may unpublished, derived following guidelines Chiu, Wang, Walther & Chao, Biometrics 70, 671–682 (2014), doi:10.1111/biom.12200, online supplementary material. Diversity accumulation functions specaccum, renyiaccum, tsallisaccum, poolaccum estaccumR use now permute package permutations order sampling sites. Normally functions need simple random permutation sites, restricted permutation permute package user-supplied permutation matrices can used. estaccumR function can use parallel processing. linestack accepts now expressions labels. allows using mathematical symbols formula given mathematical expressions.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-0","dir":"Changelog","previous_headings":"","what":"vegan 2.2-0","title":"vegan 2.2-0","text":"CRAN release: 2014-11-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-0","text":"Several vegan functions can now use parallel processing slow repeating calculations. functions argument parallel. argument can integer giving number parallel processes. unix-alikes (Mac OS, Linux) launch \"multicore\" processing Windows set \"snow\" clusters desribed documentation parallel package. option \"mc.cores\" set integer > 1, used automatically start parallel processing. Finally, argument can also previously set \"snow\" cluster used Windows unix-alikes. Vegan vignette Design decision explains implementation (use vegandocs(\"decission\"), parallel package extensive documentation parallel processing R. following function use parallel processing analysing permutation statistics: adonis, anosim, anova.cca (permutest.cca), mantel (mantel.partial), mrpp, ordiareatest, permutest.betadisper simper. addition, bioenv can compare several candidate sets models paralle, metaMDS can launch several random starts parallel, oecosimu can evaluate test statistics several null models parallel. permutation tests based permute package offers strong tools restricted permutation. functions argument permutations. default usage simple non-restricted permutations achieved giving single integer number. Restricted permutations can defined using function permute package. Finally, argument can permutation matrix rows define permutations. possible use external user constructed permutations. See help(permutations) brief introduction permutations vegan, permute package full documention. vignette permute package can read vegan command vegandocs(\"permutations\"). following functions use permute package: CCorA, adonis, anosim, anova.cca (plus associated permutest.cca, add1.cca, drop1.cca, ordistep, ordiR2step), envfit (plus associated factorfit vectorfit), mantel (mantel.partial), mrpp, mso, ordiareatest, permutest.betadisper, protest simper. Community null model generation completely redesigned rewritten. communities constructed new nullmodel function defined low level commsim function. actual null models generated simulate function builds array null models. new null models include wide array quantitative models addition old binary models, users can plug generating functions. basic tool invoking analysing null models oecosimu. null models often used analysis nestedness, implementation oecosimu allows analysing statistic, null models better seen alternative permutation tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-0","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-0","text":"vegan package dependencies namespace imports adapted changes R, trigger warnings notes package tests. Three-dimensional ordination graphics using scatterplot3d static plots rgl dynamic plots removed vegan moved companion package vegan3d. package available CRAN.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-2-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.2-0","text":"Function dispweight implements dispersion weighting Clarke et al. (Marine Ecology Progress Series, 320, 11–27). addition, implemented new method generalized dispersion weighting gdispweight. methods downweight species significantly -dispersed. New hclust support functions reorder, rev scores. Functions reorder rev similar functions dendrogram objects base R. However, reorder can use (defaults ) weighted mean. weighted mean node average always mean member leaves, whereas dendrogram uses always unweighted means joined branches. Function ordiareatest supplements ordihull ordiellipse provides randomization test one-sided alternative hypothesis convex hulls ellipses two-dimensional ordination space smaller areas randomized groups. Function permustats extracts inspects permutation results support functions summary, density, densityplot, qqnorm qqmath. density qqnorm standard R tools work one statistic, densityplot qqmath lattice graphics work univariate multivariate statistics. results following functions can extracted: anosim, adonis, mantel (mantel.partial), mrpp, oecosimu, permustest.cca (corresponding anova methods), permutest.betadisper, protest. stressplot functions display ordination distances given number dimensions original distances. method functins similar stressplot metaMDS, always use inherent distances ordination method. functions available results capscale, cca, princomp, prcomp, rda, wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-0","text":"cascadeKM one group NA instead random value. ordiellipse can handle points exactly line, including two points (warning). plotting radfit results several species failed communities species one species. RsquareAdj capscale negative eigenvalues now report NA instead using biased method rda results. simper failed group single member.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-0","text":"anova.cca functions re-written use permute package. Old results may exactly reproduced, models missing data may fail several cases. new option analysing sequence models . simulate functions cca rda can return several simulations nullmodel compatible object. functions can produce simulations correlated errors (also capscale) parametric simulation Gaussian error. bioenv can use Manhattan, Gower Mahalanobis distances addition default Euclidean. New helper function bioenvdist can extract dissimilarities applied best model model. metaMDS(..., trace = 2) show convergence information default monoMDS engine. Function MDSrotate can rotate k-dimensional ordination k-1 variables. variables correlated (like usually case), vectors can also correlated previously rotated dimensions, uncorrelated later ones. vegan 2.0-10 changed weighted nestednodf weighted analysis binary data equivalent binary analysis. However, broke equivalence original method. Now function argument wbinary select method analysis. problem reported fix submitted Vanderlei Debastiani (Universidade Federal Rio Grande Sul, Brasil). ordiellipse, ordihull ordiellipse can handle missing values groups. ordispider can now use spatial medians instead means. rankindex can use Manhattan, Gower Mahalanobis distance addition default Euclidean. User can set colours line types function rarecurve plotting rarefaction curves. spantree gained support function .hclust change minimum spanning tree hclust tree. fitspecaccum can weighted analysis. Gained lines method. Functions extrapolated number species size species pool using Chao method modified following Chiu et al., Biometrics 70, 671–682 (2014). Incidence based specpool can now use (defaults ) small sample correction number sites sample size. Function uses basic Chao extrapolation based ratio singletons doubletons, switches now bias corrected Chao extrapolation doubletons (species found twice). variance formula bias corrected Chao derived following supporting line material doi:10.1111/biom.12200 differs slightly Chiu et al. (2014). poolaccum function changed similarly, small sample correction used always. abundance based estimateR uses bias corrected Chao extrapolation, earlier estimated variance classic Chao model. Now use widespread approximate estimate EstimateS variance. changes functions similar EstimateS tabasco uses now reorder.hclust hclust object better ordering previously cast trees dendrogram objects. treedive treedist default now match.force = TRUE can silenced verbose = FALSE. vegdist gained Mahalanobis distance. Nomenclature updated plant community data help Taxonstand taxize packages. taxonomy dune data adapted sources APG III. varespec dune use 8-character names (4 genus + 4 species epithet). New data set phylogenetic distances dune extracted Zanne et al. (Nature 506, 89–92; 2014). User configurable plots rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-2-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.2-0","text":"strata deprecated permutations. still accepted phased next releases. Use permute package. cca, rda capscale return scores scaled eigenvalues: use scores function extract scaled results. commsimulator deprecated. Replace commsimulator(x, method) simulate(nullmodel(x, method)). density densityplot permutation results deprecated: use permustats density densityplot method.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-10","dir":"Changelog","previous_headings":"","what":"vegan 2.0-10","title":"vegan 2.0-10","text":"CRAN release: 2013-12-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-10","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-10","text":"version adapted changes permute package version 0.8-0 triggers NOTEs package checks. release may last 2.0 series, next vegan release scheduled major release newly designed oecosimu community pattern simulation, support parallel processing, full support permute package. interested developments, may try development versions vegan GitHub report problems user experience us.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-10","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-10","text":"envfit function assumed external variables either numeric factors, failed , say, character strings. Now numeric variables taken continuous vectors, variables (character strings, logical) coerced factors possible. function also work degenerate data, like one level factor constant value continuous environmental variable. ties wrongly assessing permutation P-values vectorfit. nestednodf quantitative data consistent binary models, fill wrongly calculated quantitative data. oecosimu now correctly adapts displayed quantiles simulated values alternative test direction. renyiaccum plotting failed one level diversity scale used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-10","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-10","text":"Kempton Taylor algorithm found unreliable fisherfit fisher.alpha, now estimation Fisher α based number species number individuals. estimation standard errors profile confidence intervals also scrapped. renyiaccum, specaccum tsallisaccum functions gained subset argument. renyiaccum can now add collector curve analysis. collector curve diversity accumulation order sampling units. interesting ordering sampling units allows comparing actual species accumulations expected randomized accumulation. specaccum can now perform weighted accumulation using sampling effort weights.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-9","dir":"Changelog","previous_headings":"","what":"vegan 2.0-9","title":"vegan 2.0-9","text":"CRAN release: 2013-09-25 version released due changes programming interface testing procedures R 3.0.2. using older version R, need upgrade vegan. new features bug fixes. user-visible changes documentation output messages formatting. R changes, version dependent R version 2.14.0 newer lattice package.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-8","dir":"Changelog","previous_headings":"","what":"vegan 2.0-8","title":"vegan 2.0-8","text":"CRAN release: 2013-07-10","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-8","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-8","text":"maintenance release fixes issues raised changed R toolset processing vignettes. also fix typographic issues vignettes.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-8","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-8","text":"ordisurf gained new arguments flexible definition fitted models better utilize mgcv::gam function. linewidth contours can now set argument lwd. Labels arrows positioned better way plot functions results envfit, cca, rda capscale. labels longer overlap arrow tips. setting test direction clearer oecosimu. ordipointlabel gained plot method can used replot saved result.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-7","dir":"Changelog","previous_headings":"","what":"vegan 2.0-7","title":"vegan 2.0-7","text":"CRAN release: 2013-03-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-7","text":"tabasco() new function graphical display community data matrix. Technically interface R heatmap, use closer vegan function vegemite. function can reorder community data matrix similarly vegemite, instance, ordination results. Unlike heatmap, displays dendrograms supplied user, defaults re-order dendrograms correspondence analysis. Species ordered match site ordering like determined user.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-7","text":"Function fitspecaccum(..., model = \"asymp\") fitted logistic model instead asymptotic model (model = \"logis\"). nestedtemp() failed sparse data (fill < 0.38%).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-7","text":"plot function constrained ordination results (cca, rda, capscale) gained argument axis.bp (defaults TRUE) can used suppress axis scale biplot arrays. Number iterations nonmetric multidimensional scaling (NMDS) can set keyword maxit (defaults 200) metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-0-7","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.0-7","text":"result objects cca, rda capscale longer scores u.eig, v.eig wa.eig future versions vegan. change influence normal usage, vegan functions need items. However, external scripts packages may need changes future versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-6","dir":"Changelog","previous_headings":"","what":"vegan 2.0-6","title":"vegan 2.0-6","text":"CRAN release: 2013-02-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-6","text":"species scores scaled wrongly capscale(). scaled correctly Euclidean distances used, usually capscale() used non-Euclidean distances. graphics change redone. change scaling mainly influences spread species scores respect site scores. Function clamtest() failed set minimum abundance threshold cases. addition, output wrong possible species groups missing. problems reported Richard Telford (Bergen, Norway). Plotting object fitted envfit() fail p.max used unused levels one factors. unused levels result deletion observations missing values simply result supplying subset larger data set envfit(). multipart() printed wrong information analysis type (analysis correctly). Reported Valerie Coudrain. oecosimu() failed nestedfun returned data frame. fundamental fix vegan 2.2-0, structure oecosimu() result change. plot two-dimensional procrustes() solutions often draw original axes wrong angle. problem reported Elizabeth Ottesen (MIT). Function treedive() functional phylogenetic diversity correctly match species names community data species tree tree contained species occur data. Related function treedist() phylogenetic distances try match names .","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-6","text":"output capscale() displays value additive constant argument add = TRUE used. fitted() functions cca(), rda() capscale() can now return conditioned (partial) component response: Argument model gained new alternative model = \"pCCA\". dispindmorisita() output gained new column Chi-squared based probabilities null hypothesis (random distribution) true. metaMDS() monoMDS() new default convergence criteria. importantly, scale factor gradient (sfgrmin) stricter. former limit slack large data sets iterations stopped early without getting close solution. addition, scores() ignore now requests dimensions beyond calculated instead failing, scores() metaMDS() results drop dimensions. msoplot() gained legend argument positioning legend. Nestedness function nestednodf() gained plot method. ordiR2step() gained new argument R2scope (defaults TRUE) can used turn criterion stopping adjusted R2 current model exceeds scope. option allows model building scope overdetermined (number predictors higher number observations). ordiR2step() now handles partial redundancy analysis (pRDA). orditorp() gained argument select select rows columns results display. protest() prints standardized residual statistic squared m12 addition squared Procrustes correlation R2. calculated, latter displayed. Permutation tests much faster protest(). Instead calling repeatedly procrustes(), goodness fit statistic evaluated within function. wcmdscale() gained methods print, plot etc. results. methods used full wcmdscale result returned , e.g., argument eig = TRUE. default still return matrix scores similarly standard R function cmdscale(), case new methods used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-5","dir":"Changelog","previous_headings":"","what":"vegan 2.0-5","title":"vegan 2.0-5","text":"CRAN release: 2012-10-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-5","text":"anova(, ...) failed = \"axis\" = \"term\". bug reported Dr Sven Neulinger (Christian Albrecht University, Kiel, Germany). radlattice honour argument BIC = TRUE, always displayed AIC.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-5","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-5","text":"vegan functions permutation tests now density method can used find empirical probability distributions permutations. new plot method functions displays density observed statistic. density function available adonis, anosim, mantel, mantel.partial, mrpp, permutest.cca procrustes. Function adonis can return several statistics, now densityplot method (based lattice). Function oecosimu already density densityplot, now similar vegan methods, also work adipart, hiersimu multipart. radfit functions got predict method also accepts arguments newdata total new ranks site totals prediction. functions can also interpolate non-integer “ranks”, models also extrapolate.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-5","text":"Labels can now set plot envfit results. labels must given order function uses internally, new support function labels can used display default labels correct order. Mantel tests (functions mantel mantel.partial) gained argument na.rm can used remove missing values. options used care: Permutation tests can biased missing values originally matching fixed positions. radfit results can consistently accessed methods whether single model single site, models single site models sites data. functions now methods AIC, coef, deviance, logLik, fitted, predict residuals.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-0-5","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.0-5","text":"Building vegan vignettes failed latest version LaTeX (TeXLive 2012). R versions later 2.15-1 (including development version) report warnings errors installing checking vegan, must upgrade vegan version. warnings concern functions cIndexKM betadisper, error occurs betadisper. errors warnings triggered internal changes R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-4","dir":"Changelog","previous_headings":"","what":"vegan 2.0-4","title":"vegan 2.0-4","text":"CRAN release: 2012-06-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-4","text":"adipart assumed constant gamma diversity simulations assessing P-value. give biased results null model produces variable gamma diversities option weights = \"prop\" used. default null model (\"r2dtable\") default option (weights = \"unif\") analysed correctly. anova(, = \"axis\") cases failed due ‘NAMESPACE’ issues. clamtest wrongly used frequencies instead counts calculating sample coverage. detectable differences produced rerunning examples Chazdon et al. 2011 vegan help page. envfit failed unused factor levels. predict cca results type = \"response\" type = \"working\" failed newdata number rows match original data. Now newdata ignored wrong number rows. number rows must match results cca must weighted original row totals. problem concern rda capscale results need row weights. Reported Glenn De’ath.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-4","text":"Functions diversity partitioning (adipart, hiersimu multipart) now formula default methods. formula method identical previous functions, default method can take two matrices input. Functions adipart multipart can used fast easy overall partitioning alpha, beta gamma diversities omitting argument describing hierarchy. method betadisper biased small sample sizes. effects bias strongest unequal sample sizes. bias adjusted version developed Adrian Stier Ben Bolker, can invoked argument bias.adjust (defaults FALSE). bioenv accepts dissimilarities (square matrices can interpreted dissimilarities) alternative community data. allows using dissimilarities available vegdist. plot function envfit results gained new argument bg can used set background colour plotted labels. msoplot configurable, allows, instance, setting y-axis limits. Hulls ellipses now filled using semitransparent colours ordihull ordiellipse, user can set degree transparency new argument alpha. filled shapes used functions called argument draw = \"polygon\". Function ordihull puts labels (argument label = TRUE) now real polygon centre. ordiplot3d returns function envfit.convert projected location origin. Together can used add envfit results existing ordiplot3d plots. Equal aspect ratio set exactly ordiplot3d underlying core routines allow . Now ordiplot3d sets equal axis ranges, documents urge users verify aspect ratio reasonably equal graph looks like cube. problems solved future, ordiplot3d may removed next releases vegan. Function ordipointlabel gained argument select items plotting. argument can used one set points.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-3","dir":"Changelog","previous_headings":"","what":"vegan 2.0-3","title":"vegan 2.0-3","text":"CRAN release: 2012-03-03","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-3","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-3","text":"Added new nestedness functions nestedbetasor nestedbetajac implement multiple-site dissimilarity indices decomposition turnover nestedness components following Baselga (Global Ecology Biogeography 19, 134–143; 2010). Added function rarecurve draw rarefaction curves row (sampling unit) input data, optionally lines showing rarefied species richness given sample size curve. Added function simper implements “similarity percentages” Clarke (Australian Journal Ecology 18, 117–143; 1993). method compares two groups decomposes average -group Bray-Curtis dissimilarity index contributions individual species. code developed GitHub Eduard Szöcs (Uni Landau, Germany).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-3","text":"betadisper() failed groups factor empty levels. constrained ordination methods support functions robust border cases (completely aliased effects, saturated models, user requests non-existng scores etc). Concerns capscale, ordistep, varpart, plot function constrained ordination, anova(, = \"margin\"). scores function monoMDS honour choices argument hence dimensions chosen plot. default scores method failed number requested axes higher ordination object . reported error ordiplot R-sig-ecology mailing list.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-3","text":"metaMDS argument noshare = 0 now regarded numeric threshold always triggers extended dissimilarities (stepacross), instead treated synonymous noshare = FALSE always suppresses extended dissimilarities. Nestedness discrepancy index nesteddisc gained new argument allows user set number iterations optimizing index. oecosimu displays mean simulations describes alternative hypothesis clearly printed output. Implemented adjusted R2 partial RDA. partial model rda(Y ~ X1 + Condition(X2)) component [] = X1|X2 variance partition varpart describes marginal (unique) effect constraining term adjusted R2. Added Cao dissimilarity (CYd) new dissimilarity method vegdist following Cao et al., Water Envir Res 69, 95–106 (1997). index good data high beta diversity variable sampling intensity. Thanks consultation Yong Cao (Univ Illinois, USA).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-2","dir":"Changelog","previous_headings":"","what":"vegan 2.0-2","title":"vegan 2.0-2","text":"CRAN release: 2011-11-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-2","text":"Function capscale failed constrained component zero rank. happened likely partial models conditions aliased constraints. problem observed anova(..., =\"margin\") uses partial models analyses marginal effects, reported email message R-News mailing list. stressplot goodness sometimes failed metaMDS based isoMDS (MASS package) metaMDSdist use defaults step-across (extended) dissimilarities metaMDS(..., engine = \"isoMDS\"). change defaults can also influence triggering step-across capscale(..., metaMDSdist = TRUE). adonis contained minor bug resulting incomplete implementation speed-affect results. fixing bug, bug identified transposing hat matrices. second bug active following fixing first bug. fixing bugs, speed-internal f.test() function fully realised. Reported Nicholas Lewin-Koh.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-2","text":"ordiarrows ordisegments gained argument order.gives variable sort points within groups. Earlier points assumed order. Function ordispider invisibly returns coordinates points connected. Typically class centroids point, constrained ordination groups LC scores.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-1","dir":"Changelog","previous_headings":"","what":"vegan 2.0-1","title":"vegan 2.0-1","text":"CRAN release: 2011-10-20","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-1","text":"clamtest: new function classify species generalists specialists two distinct habitats (CLAM test Chazdon et al., Ecology 92, 1332–1343; 2011). test based multinomial distribution individuals two habitat types sampling units, applicable count data -dispersion. .preston gained plot lines methods, .fisher gained plot method (also can add items existing plots). similar plot lines prestonfit fisherfit, display data without fitted lines. raupcrick: new function implement Raup-Crick dissimilarity probability number co-occurring species occurrence probabilities proportional species frequencies. Vegan Raup-Crick index choice vegdist, uses equal sampling probabilities species analytic equations. new raupcrick function uses simulation oecosimu. function follows Chase et al. (2011) Ecosphere 2:art24 [doi:10.1890/ES10-00117.1], developed consultation Brian Inouye.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-1","text":"Function meandist scramble items give wrong results, especially grouping numerical. problem reported Dr Miguel Alvarez (Univ. Bonn). metaMDS reset tries new model started previous.best solution different model. Function permatswap community null models using quantitative swap never swapped items 2x2 submatrix cells filled. result permutest.cca updated ‘NAMESPACE’ issue. R 2.14.0 changed accept using sd() function matrices (behaviour least since R 1.0-0), several vegan functions changed adapt change (rda, capscale, simulate methods rda, cca capscale). change R 2.14.0 influence results probably wish upgrade vegan avoid annoying warnings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-1","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-1","text":"nesteddisc slacker hence faster trying optimize statistic tied column frequencies. Tracing showed cases improved ordering found rather early tries, results equally good cases.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-0","dir":"Changelog","previous_headings":"","what":"vegan 2.0-0","title":"vegan 2.0-0","text":"CRAN release: 2011-09-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-0","text":"Peter Minchin joins vegan team. vegan implements standard R ‘NAMESPACE’. general, S3 methods exported means directly use see contents functions like cca.default, plot.cca anova.ccabyterm. use functions rely R delegation simply use cca result objects use plot anova without suffix .cca. see contents function can use :::, vegan:::cca.default. change may break packages, documents scripts rely non-exported names. vegan depends permute package. package provides powerful tools restricted permutation schemes. vegan permutation gradually move use permute, currently betadisper uses new feature.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-0","text":"monoMDS: new function non-metric multidimensional scaling (NMDS). function replaces MASS::isoMDS default method metaMDS. Major advantages monoMDS ‘weak’ (‘primary’) tie treatment means can split tied observed dissimilarities. ‘Weak’ tie treatment improves ordination heterogeneous data sets, maximum dissimilarities 1 can split. addition global NMDS, monoMDS can perform local hybrid NMDS metric MDS. can also handle missing zero dissimilarities. Moreover, monoMDS faster previous alternatives. function uses Fortran code written Peter Minchin. MDSrotate new function replace metaMDSrotate. function can rotate metaMDS monoMDS results first axis parallel environmental vector. eventstar finds minimum evenness profile Tsallis entropy, uses find corresponding values diversity, evenness numbers equivalent following Mendes et al. (Ecography 31, 450-456; 2008). code contributed Eduardo Ribeira Cunha Heloisa Beatriz Antoniazi Evangelista adapted vegan Peter Solymos. fitspecaccum fits non-linear regression models species accumulation results specaccum. function can use new self-starting species accumulation models vegan self-starting non-linear regression models R. function can fit Arrhenius, Gleason, Gitay, Lomolino (vegan), asymptotic, Gompertz, Michaelis-Menten, logistic Weibull (base R) models. function plot predict methods. Self-starting non-linear species accumulation models SSarrhenius, SSgleason, SSgitay SSlomolino. can used fitspecaccum directly non-linear regression nls. functions implemented found good species-area models Dengler (J. Biogeogr. 36, 728-744; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-0","text":"adonis, anosim, meandist mrpp warn negative dissimilarities, betadisper refuses analyse . functions expect dissimilarities, giving something else (like correlations) probably user error. betadisper uses restricted permutation permute package. metaMDS uses monoMDS default ordination engine. Function gains new argument engine can used alternatively select MASS::isoMDS. default use stepacross monoMDS ‘weak’ tie treatment can cope tied maximum dissimilarities one. However, stepacross default isoMDS handle adequately tied maximum dissimilarities. specaccum gained predict method uses either linear spline interpolation data observed points. Extrapolation possible spline interpolation, may make little sense. specpool can handle missing values empty factor levels grouping factor pool. Now also checks length pool matches number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-0-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.0-0","text":"metaMDSrotate replaced MDSrotate can also handle results monoMDS. permuted.index2 “new” permutation code removed favour permute package. code intended normal use, packages depending code vegan instead depend permute.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-0","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-0","text":"treeheight uses much snappier code. results unchanged.","code":""}]
+[{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"vegan-faq","dir":"Articles","previous_headings":"","what":"vegan FAQ","title":"","text":"document contains answers frequently asked questions R package vegan. work licensed Creative Commons Attribution 3.0 License. view copy license, visit https://creativecommons.org/licenses//3.0/ send letter Creative Commons, 543 Howard Street, 5th Floor, San Francisco, California, 94105, USA. Copyright © 2008-2016 vegan development team","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is vegan?","title":"","text":"Vegan R package community ecologists. contains popular methods multivariate analysis needed analysing ecological communities, tools diversity analysis, potentially useful functions. Vegan self-contained must run R statistical environment, also depends many R packages. Vegan free software distributed GPL2 license.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is R?","title":"","text":"R system statistical computation graphics. consists language plus run-time environment graphics, debugger, access certain system functions, ability run programs stored script files. R home page https://www.R-project.org/. free software distributed GNU-style copyleft, official part GNU project (“GNU S”).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-obtain-vegan-and-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to obtain vegan and R?","title":"","text":"R latest release version vegan can obtained CRAN. Unstable development version vegan can obtained GitHub. github page gives instructions obtaining installing development versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-r-packages-vegan-depends-on","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What R packages vegan depends on?","title":"","text":"Vegan depends permute package provide advanced flexible permutation routines vegan. permute package developed together vegan GitHub. individual vegan functions depend packages MASS, mgcv, parallel, cluster lattice. Vegan dependence tcltk deprecated removed future releases. base recommended R packages available every R installation. Vegan declares suggested imported packages, can install vegan use functions without packages. Vegan accompanied supporting package vegan3d three-dimensional dynamic plotting. vegan3d package needs tcltk non-standard packages rgl scatterplot3d.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-packages-are-available-for-ecologists","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other packages are available for ecologists?","title":"","text":"CRAN Task Views include entries like Environmetrics, Multivariate Spatial describe several useful packages functions. install R package ctv, can inspect Task Views R session, automatically install sets important packages.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-documentation-is-available-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other documentation is available for vegan?","title":"","text":"Vegan fully documented R package standard help pages. authoritative sources documentation (last resource can use force read source, vegan open source). Vegan package ships documents can read browseVignettes(\"vegan\") command. documents included vegan package Vegan NEWS can accessed via news() command. document (FAQ-vegan). Short introduction basic ordination methods vegan (intro-vegan). Introduction diversity methods vegan (diversity-vegan). Discussion design decisions vegan (decision-vegan). Description variance partition procedures function varpart (partitioning). Web documents outside package include: https://github.com/vegandevs/vegan: development page. https://vegandevs.github.io/vegan/: vegan homepage.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-a-graphical-user-interface-gui-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is there a Graphical User Interface (GUI) for vegan?","title":"","text":"Roeland Kindt made package BiodiversityR provides GUI vegan. package available CRAN. mere GUI vegan, adds new functions complements vegan functions order provide workbench biodiversity analysis. can install BiodiversityR using install.packages(\"BiodiversityR\") graphical package management menu R. GUI works Windows, MacOS X Linux.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-cite-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to cite vegan?","title":"","text":"Use command citation(\"vegan\") R see recommended citation used publications.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-build-vegan-from-sources","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to build vegan from sources?","title":"","text":"general, need build vegan sources, binary builds release versions available CRAN Windows MacOS X. use operating systems, may use source packages. Vegan standard R package, can built like instructed R documentation. Vegan contains source files C FORTRAN, need appropriate compilers (may need work Windows MacOS X).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"are-there-binaries-for-devel-versions","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Are there binaries for devel versions?","title":"","text":"Binaries can available R Universe: see https://github.com/vegandevs/vegan instructions.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-report-a-bug-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to report a bug in vegan?","title":"","text":"think found bug vegan, report vegan maintainers developers. preferred forum report bugs GitHub. bug report detailed bug can replicated corrected. Preferably, send example causes bug. needs data set available R, send minimal data set well. also paste output error message message. also specify version vegan used. Bug reports welcome: way make vegan non-buggy. Please note shall send bug reports R mailing lists, since vegan standard R package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-a-bug-or-a-feature","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is it a bug or a feature?","title":"","text":"necessarily bug function gives different results expect: may deliberate design decision. may useful check documentation function see intended behaviour. may also happen function argument switch behaviour match expectation. instance, function vegdist always calculates quantitative indices (possible). expect calculate binary index, use argument binary = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-contribute-to-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Can I contribute to vegan?","title":"","text":"Vegan dependent user contribution. feedback welcome. problems vegan, may simple incomplete documentation, shall best improve documents. Feature requests also welcome, necessarily fulfilled. new feature added easy looks useful, submit code. can write code , best forum contribute vegan GitHub.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-only-numeric-and-positive-data-but-vegan-still-complains","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have only numeric and positive data but vegan still complains","title":"","text":"wrong! Computers painfully pedantic, find non-numeric negative data entries, really . Check data! common reasons non-numeric data row names read non-numeric variable instead used row names (check argument row.names reading data), column names interpreted data (check argument header = TRUE reading data). Another common reason empty cells input data, interpreted missing values.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-analyse-binary-or-cover-class-data","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I analyse binary or cover class data?","title":"","text":"Yes. vegan methods can handle binary data cover abundance data. statistical tests based permutation, make distributional assumptions. methods (mainly diversity analysis) need count data. methods check input data integers, may fooled cover class data.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-dissimilarities-in-vegan-differ-from-other-sources","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why dissimilarities in vegan differ from other sources?","title":"","text":"commonly reason software use presence–absence data whereas vegan used quantitative data. Usually vegan indices quantitative, can use argument binary = TRUE make presence–absence. However, index name cases, although different names usually occur literature. instance, Jaccard index actually refers binary index, vegan uses name \"jaccard\" quantitative index, . Another reason may indices indeed defined differently, people use names different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-nmds-stress-is-sometimes-0-1-and-sometimes-10","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why NMDS stress is sometimes 0.1 and sometimes 10?","title":"","text":"Stress proportional measure badness fit. proportions can expressed either parts one percents. Function isoMDS (MASS package) uses percents, function monoMDS (vegan package) uses proportions, therefore stress 100 times higher isoMDS. results goodness function also depend definition stress, goodness 100 times higher isoMDS monoMDS. conventions equally correct.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-get-zero-stress-but-no-repeated-solutions-in-metamds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I get zero stress but no repeated solutions in metaMDS","title":"","text":"first (try 0) run metaMDS starts metric scaling solution usually good, sofware return solution. However, metaMDS tries see standard solution can repeated, improved improved solution still repeated. cases, return best solution found, burning need anything get message tha solution repeated. keen know solution really global optimum, may follow instructions metaMDS help section “Results Repeated” try . common reason observations NMDS. n observations (points) k dimensions need estimate n*k parameters (ordination scores) using n*(n-1)/2 dissimilarities. k dimensions must n > 2*k + 1, two dimensions least six points. degenerate situations may need even larger number points. lower number points, can find undefined number perfect (stress zero) different solutions. Conventional wisdom due Kruskal n > 4*k + 1 points k dimensions. typical symptom insufficient data (nearly) zero stress two convergent solutions. cases reduce number dimensions (k) small data sets use NMDS, rely metric methods. seems local hybrid scaling monoMDS similar lower limits practice (although theoretically differ). However, higher number dimensions can used metric scaling, monoMDS principal coordinates analysis (cmdscale stats, wcmdscale vegan).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"zero-dissimilarities-in-isomds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Zero dissimilarities in isoMDS","title":"","text":"Function metaMDS uses function monoMDS default method NMDS, function can handle zero dissimilarities. Alternative function isoMDS handle zero dissimilarities. want use isoMDS, can use argument zerodist = \"add\" metaMDS handle zero dissimilarities. argument, zero dissimilarities replaced small positive value, can handled isoMDS. kluge, people like . principal solution remove duplicate sites using R command unique. However, standardizations dissimilarity indices, originally non-unique sites can zero dissimilarity, resort kluge (work harder data). Usually better use monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-heard-that-you-cannot-fit-environmental-vectors-or-surfaces-to-nmds-results-which-only-have-rank-order-scores","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have heard that you cannot fit environmental vectors or surfaces to NMDS results which only have rank-order scores","title":"","text":"Claims like indeed large Internet, based grave misunderstanding plainly wrong. NMDS ordination results strictly metric, vegan metaMDS monoMDS even strictly Euclidean. method called “non-metric” Euclidean distances ordination space non-metric rank-order relationship community dissimilarities. can inspect non-linear step curve using function stressplot vegan. ordination scores strictly Euclidean, correct use vegan functions envfit ordisurf NMDS results.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"where-can-i-find-numerical-scores-of-ordination-axes","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Where can I find numerical scores of ordination axes?","title":"","text":"Normally can use function scores extract ordination scores ordination method. scores function can also find ordination scores many non-vegan functions prcomp princomp ade4 functions. cases ordination result object stores raw scores, axes also scaled appropriate access scores. instance, cca rda ordination object -called normalized scores, scaled ordination plots use accessed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-the-rda-results-are-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How the RDA results are scaled?","title":"","text":"scaling RDA results indeed differ software packages. scaling RDA complicated issue explained FAQ, explained separate pdf document “Design decision implementation details vegan” can read command browseVignettes(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-cannot-print-and-plot-rda-results-properly","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I cannot print and plot RDA results properly","title":"","text":"RDA ordination results weird format plot properly, probably name clash klaR package also function rda, klaR print, plot predict functions used vegan RDA results. can choose rda functions using vegan::rda() klaR::rda(): get obscure error messages use wrong function. general, vegan able work normally vegan loaded klaR, klaR loaded later, functions take precedence vegan. Sometimes vegan namespace loaded automatically restoring previously stored workspace start-, klaR methods always take precedence vegan. check loaded packages. klaR may also loaded indirectly via packages (reported cases often loaded via agricolae package). Vegan klaR function name (rda), may possible use packages simultaneously, safest choice unload one packages possible. See discussion vegan github issues.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"ordination-fails-with-error-in-la-svd","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Ordination fails with “Error in La.svd”","title":"","text":"Constrained ordination (cca, rda, dbrda, capscale) sometimes fail error message Error La.svd(x, nu, nv): error code 1 Lapack routine 'dgesdd'. seems basic problem svd function LAPACK used numerical analysis R. LAPACK external library beyond control package developers R core team problems may unsolvable. Reducing range constraints (environmental variables) helps sometimes. instance, multiplying constraints constant < 1. rescaling influence numerical results constrained ordination, can complicate analyses values constraints needed, scaling must applied . reports problems getting rare may problem fixed R LAPACK.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"variance-explained-by-ordination-axes-","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Variance explained by ordination axes.","title":"","text":"general, vegan directly give statistics “variance explained” ordination axes constrained axes. design decision: think information normally useless often misleading. community ordination, goal typically explain variance, find “gradients” main trends data. “total variation” often meaningless, proportions meaningless values also meaningless. Often better solution explains smaller part “total variation”. instance, unstandardized principal components analysis variance generated small number abundant species, easy “explain” data really multivariate. standardize data, species equally important. first axes explains much less “total variation”, now explain species equally, results typically much useful whole community. Correspondence analysis uses another measure variation (variance), typically explains “smaller proportion” principal components better result. Detrended correspondence analysis nonmetric multidimensional scaling even try “explain” variation, use criteria. methods incommensurable, impossible compare methods using “explanation variation”. still want get “explanation variation” (deranged editor requests ), possible get information methods: Eigenvector methods: Functions rda, cca, dbrda capscale give variation conditional (partialled), constrained (canonical) residual components. Function eigenvals extracts eigenvalues, summary(eigenvals(ord)) reports proportions explained result object ord, also works decorana wcmdscale. Function RsquareAdj gives R-squared adjusted R-squared (available) constrained components. Function goodness gives statistics individual species sites. addition, special function varpart unbiased partitioning variance four separate components redundancy analysis. Nonmetric multidimensional scaling. NMDS method nonlinear mapping, concept variation explained make sense. However, 1 - stress^2 transforms nonlinear stress quantity analogous squared correlation coefficient. Function stressplot displays nonlinear fit gives statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-have-random-effects-in-constrained-ordination-or-in-adonis","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I have random effects in constrained ordination or in adonis?","title":"","text":". Strictly speaking, impossible. However, can define models respond similar goals random effects models, although strictly speaking use fixed effects. Constrained ordination functions cca, rda dbrda can Condition() terms formula. Condition() define partial terms fitted constraints can used remove effects background variables, contribution decomposing inertia (variance) reported separately. partial terms often regarded similar random effects, still fitted way terms strictly speaking fixed terms. Function adonis2 can evaluate terms sequentially. model right-hand-side ~ + B effects evaluated first, effects B removing effects . Sequential tests also available anova function constrained ordination results setting argument = \"term\". way, first terms can serve similar role random effects, although fitted way terms, strictly speaking fixed terms. permutation tests vegan based permute package allows constructing various restricted permutation schemes. instance, can set levels plots blocks factor regarded random term. major reason real random effects models impossible vegan functions tests based permutation data. data given, fixed, therefore permutation tests basically tests fixed terms fixed data. Random effect terms require permutations data random component instead given, fixed data, tests available vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-possible-to-have-passive-points-in-ordination","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Is it possible to have passive points in ordination?","title":"","text":"Vegan concept passive points, point little influence ordination results. However, can add points eigenvector methods using predict functions newdata. can first perform ordination without species sites, can find scores points using complete data newdata. predict functions available basic eigenvector methods vegan (cca, rda, decorana, --date list, use command methods(\"predict\")).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"class-variables-and-dummies","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Class variables and dummies","title":"","text":"define class variable R factor, vegan automatically handle . R (vegan) knows unordered ordered factors. Unordered factors internally coded dummy variables, one redundant level removed aliased. default contrasts, removed level first one. Ordered factors expressed polynomial contrasts. contrasts explained standard R documentation.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-are-environmental-arrows-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How are environmental arrows scaled?","title":"","text":"printed output envfit gives direction cosines coordinates unit length arrows. plotting, scaled correlation (square roots column r2). can see scaled lengths envfit arrows using command scores. scaled environmental vectors envfit arrows continuous environmental variables constrained ordination (cca, rda, dbrda) adjusted fill current graph. lengths arrows fixed meaning respect points (species, sites), can compared , therefore relative lengths important. want change scaling arrows, can use text (plotting arrows text) points (plotting arrows) functions constrained ordination. functions argument arrow.mul sets multiplier. plot function envfit also arrow.mul argument set arrow multiplier. save invisible result constrained ordination plot command, can see value currently used arrow.mul saved attribute biplot scores. Function ordiArrowMul used find scaling current plot. can use function see arrows scaled:","code":"sol <- cca(varespec) ef <- envfit(sol ~ ., varechem) plot(sol) ordiArrowMul(scores(ef, display=\"vectors\"))"},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-want-to-use-helmert-or-sum-contrasts","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I want to use Helmert or sum contrasts","title":"","text":"vegan uses standard R utilities defining contrasts. default standard installations use treatment contrasts, can change behaviour globally setting options locally using keyword contrasts. Please check R help pages user manuals details.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-are-aliased-variables-and-how-to-see-them","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"What are aliased variables and how to see them?","title":"","text":"Aliased variable information can expressed help variables. variables automatically removed constrained ordination vegan. aliased variables can redundant levels factors whole variables. Vegan function alias gives defining equations aliased variables. want see names aliased variables levels solution sol, use alias(sol, names.=TRUE).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"plotting-aliased-variables","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Plotting aliased variables","title":"","text":"can fit vectors class centroids aliased variables using envfit function. envfit function uses weighted fitting, fitted vectors identical vectors correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"restricted-permutations-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Restricted permutations in vegan","title":"","text":"Vegan uses permute package permutation tests. permute package allow restricted permutation designs time series, line transects, spatial grids blocking factors. construction restricted permutation schemes explained manual page permutations vegan documentation permute package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-use-different-plotting-symbols-in-ordination-graphics","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to use different plotting symbols in ordination graphics?","title":"","text":"default ordination plot function intended fast plotting configurable. use different plotting symbols, first create empty ordination plot plot(..., type=\"n\"), add points text created empty frame (... means arguments want give plot command). points text commands fully configurable, allow different plotting symbols characters.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-avoid-cluttered-ordination-graphs","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to avoid cluttered ordination graphs?","title":"","text":"really high number species sites, graphs often congested many labels overwritten. may impossible complete readable graphics data sets. give brief overview tricks can use. Gavin Simpson’s blog bottom heap series articles “decluttering ordination plots” detailed discussion examples. Use points, possibly different types need see labels. may need first create empty plot using plot(..., type=\"n\"), satisfied default graph. (... means arguments want give plot command.) Use points add labels desired points using interactive identify command need see labels. Add labels using function ordilabel uses non-transparent background text. labels still shadow , uppermost labels readable. Argument priority help displaying interesting labels (see Decluttering blog, part 1). Use orditorp function uses labels can added graph without overwriting labels, points otherwise, need see labels. must first create empty plot using plot(..., type=\"n\"), add labels points orditorp (see Decluttering blog). Use ordipointlabel uses points text labels points, tries optimize location text minimize overlap (see Decluttering blog). Ordination text points functions argument select can used full control selecting items plotted text points. Use interactive orditkplot function (vegan3d package) lets drag labels points better positions need see labels. one set points can used (see Decluttering blog). plot functions allow zoom part graph using xlim ylim arguments reduce clutter congested areas.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-flip-an-axis-in-ordination-diagram","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I flip an axis in ordination diagram?","title":"","text":"Use xlim ylim flipped limits. model mod <- cca(dune) can flip first axis plot(mod, xlim = c(3, -2)).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-zoom-into-an-ordination-plot","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I zoom into an ordination plot?","title":"","text":"can use xlim ylim arguments plot ordiplot zoom ordination diagrams. Normally must set xlim ylim ordination plots keep equal aspect ratio axes, fill graph longer axis fit. Dynamic zooming can done function orditkplot CRAN package vegan3d. can directly save edited orditkplot graph various graphic formats, can export graph object back R session use plot display results.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-twinspan","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Is there TWINSPAN?","title":"","text":"TWINSPAN R available github.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-restricted-permutation-does-not-influence-adonis-results","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Why restricted permutation does not influence adonis results?","title":"","text":"permutation scheme influences permutation distribution statistics probably significance levels, influence calculation statistics.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-is-deviance-calculated","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"How is deviance calculated?","title":"","text":"vegan functions, radfit use base R facility family maximum likelihood estimation. allows use several alternative error distributions, among \"poisson\" \"gaussian\". R family also defines deviance. can see equations deviance commands like poisson()$dev gaussian()$dev. general, deviance 2 times log.likelihood shifted models exact fit zero deviance.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Jari Oksanen. Author, maintainer. Gavin L. Simpson. Author. F. Guillaume Blanchet. Author. Roeland Kindt. Author. Pierre Legendre. Author. Peter R. Minchin. Author. R.B. O'Hara. Author. Peter Solymos. Author. M. Henry H. Stevens. Author. Eduard Szoecs. Author. Helene Wagner. Author. Matt Barbour. Author. Michael Bedward. Author. Ben Bolker. Author. Daniel Borcard. Author. Gustavo Carvalho. Author. Michael Chirico. Author. Miquel De Caceres. Author. Sebastien Durand. Author. Heloisa Beatriz Antoniazi Evangelista. Author. Rich FitzJohn. Author. Michael Friendly. Author. Brendan Furneaux. Author. Geoffrey Hannigan. Author. Mark O. Hill. Author. Leo Lahti. Author. Dan McGlinn. Author. Marie-Helene Ouellette. Author. Eduardo Ribeiro Cunha. Author. Tyler Smith. Author. Adrian Stier. Author. Cajo J.F. Ter Braak. Author. James Weedon. Author.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O'Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier , Ter Braak C, Weedon J (2024). vegan: Community Ecology Package. R package version 2.6-7, https://github.com/vegandevs/vegan, https://vegandevs.github.io/vegan/.","code":"@Manual{, title = {vegan: Community Ecology Package}, author = {Jari Oksanen and Gavin L. Simpson and F. Guillaume Blanchet and Roeland Kindt and Pierre Legendre and Peter R. Minchin and R.B. O'Hara and Peter Solymos and M. Henry H. Stevens and Eduard Szoecs and Helene Wagner and Matt Barbour and Michael Bedward and Ben Bolker and Daniel Borcard and Gustavo Carvalho and Michael Chirico and Miquel {De Caceres} and Sebastien Durand and Heloisa Beatriz Antoniazi Evangelista and Rich FitzJohn and Michael Friendly and Brendan Furneaux and Geoffrey Hannigan and Mark O. Hill and Leo Lahti and Dan McGlinn and Marie-Helene Ouellette and Eduardo {Ribeiro Cunha} and Tyler Smith and Adrian Stier and Cajo J.F. {Ter Braak} and James Weedon}, year = {2024}, note = {R package version 2.6-7, https://github.com/vegandevs/vegan}, url = {https://vegandevs.github.io/vegan/}, }"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"vegan-an-r-package-for-community-ecologists","dir":"","previous_headings":"","what":"vegan: an R package for community ecologists","title":"vegan: an R package for community ecologists","text":"Ordination methods, diversity analysis functions community vegetation ecologists. Website development version vegan package. Vignettes available R-universe Introduction ordination vegan Partition Variation Diversity analysis vegan Design decisions implementation","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"vegan: an R package for community ecologists","text":"install development version vegan can use usual git R CMD build -> R CMD INSTALL dance cloned repo (downloaded sources). ’ll need able install packages source work; don’t relevant developer tools, won’t able install vegan way.","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"using-remotes","dir":"","previous_headings":"","what":"Using remotes","title":"vegan: an R package for community ecologists","text":"developer tools installed don’t want hassle keeping local source code tree --date, use remotes package:","code":"install.packages(\"remotes\") remotes::install_github(\"vegandevs/vegan\")"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installing-binaries-from-r-universe","dir":"","previous_headings":"","what":"Installing binaries from R Universe","title":"vegan: an R package for community ecologists","text":"just want install binary version packages, just CRAN, can install R Universe repository. Run following R session:","code":"install.packages('vegan', repos = c('https://vegandevs.r-universe.dev','https://cloud.r-project.org'))"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":null,"dir":"Reference","previous_headings":"","what":"Barro Colorado Island Tree Counts — BCI","title":"Barro Colorado Island Tree Counts — BCI","text":"Tree counts 1-hectare plots Barro Colorado Island associated site information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI) data(BCI.env)"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Barro Colorado Island Tree Counts — BCI","text":"data frame 50 plots (rows) 1 hectare counts trees plot total 225 species (columns). Full Latin names used tree species. names updated http://www.theplantlist.org Kress et al. (2009) allows matching 207 species doi:10.5061/dryad.63q27 (Zanne et al., 2014). original species names available attribute original.names BCI. See Examples changed names. BCI.env, data frame 50 plots (rows) nine site variables derived Pyke et al. (2001) Harms et al. (2001): UTM.EW: UTM coordinates (zone 17N) East-West. UTM.NS: UTM coordinates (zone 17N) North-South. Precipitation: Precipitation mm per year. Elevation: Elevation m sea level. Age.cat: Forest age category. Geology: Underlying geological formation. Habitat: Dominant habitat type based map habitat types 25 grid cells plot (Harms et al. 2001, excluding streamside habitat). habitat types Young forests (ca. 100 years), old forests > 7 degree slopes (OldSlope), old forests 152 m elevation (OldLow) higher elevation (OldHigh) Swamp forests. River: \"Yes\" streamside habitat plot. EnvHet: Environmental Heterogeneity assessed Simpson diversity frequencies Habitat types 25 grid cells plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Barro Colorado Island Tree Counts — BCI","text":"Data give numbers trees least 10 cm diameter breast height (DBH) one hectare quadrat 1982 BCI plot. Within plot, individuals tallied recorded table. full survey included smaller trees DBH 1 cm larger, BCI dataset subset larger trees compiled Condit et al. (2002). full data thinner trees densities 4000 stems per hectare, ten times stems data. dataset BCI provided (2003) illustrate analysis methods vegan. scientific research ecological issues strongly recommend access complete modern data (Condit et al. 2019) updated taxonomy (Condit et al. 2020). data frame contains Barro Colorado Island subset full data table Condit et al. (2002). quadrats located regular grid. See BCI.env coordinates. full description site information BCI.env given Pyke et al. (2001) Harms et al. (2001). N.B. Pyke et al. (2001) Harms et al. (2001) give conflicting information forest age categories elevation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Barro Colorado Island Tree Counts — BCI","text":"https://www.science.org/doi/10.1126/science.1066854 community data References environmental data. updated complete data (incl. thinner trees 1 cm), see Condit et al. (2019).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Barro Colorado Island Tree Counts — BCI","text":"Condit, R, Pitman, N, Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., Nuñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002). Beta-diversity tropical forest trees. Science 295, 666--669. Condit R., Pérez, R., Aguilar, S., Lao, S., Foster, R. & Hubbell, S. (2019). Complete data Barro Colorado 50-ha plot: 423617 trees, 35 years [Dataset]. Dryad. doi:10.15146/5xcp-0d46 Condit, R., Aguilar, S., Lao, S., Foster, R., Hubbell, S. (2020). BCI 50-ha Plot Taxonomy [Dataset]. Dryad. doi:10.15146/R3FH61 Harms K.E., Condit R., Hubbell S.P. & Foster R.B. (2001) Habitat associations trees shrubs 50-ha neotropical forest plot. J. Ecol. 89, 947--959. Kress W.J., Erickson D.L, Jones F.., Swenson N.G, Perez R., Sanjur O. & Bermingham E. (2009) Plant DNA barcodes community phylogeny tropical forest dynamics plot Panama. PNAS 106, 18621--18626. Pyke, C. R., Condit, R., Aguilar, S., & Lao, S. (2001). Floristic composition across climatic gradient neotropical lowland forest. Journal Vegetation Science 12, 553--566. doi:10.2307/3237007 Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506, 89--92. doi:10.1038/nature12872 (published online Dec 22, 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI, BCI.env) head(BCI.env) #> UTM.EW UTM.NS Precipitation Elevation Age.cat Geology Habitat Stream EnvHet #> 1 625754 1011569 2530 120 c3 Tb OldSlope Yes 0.6272 #> 2 625754 1011669 2530 120 c3 Tb OldLow Yes 0.3936 #> 3 625754 1011769 2530 120 c3 Tb OldLow No 0.0000 #> 4 625754 1011869 2530 120 c3 Tb OldLow No 0.0000 #> 5 625754 1011969 2530 120 c3 Tb OldSlope No 0.4608 #> 6 625854 1011569 2530 120 c3 Tb OldLow No 0.0768 ## see changed species names oldnames <- attr(BCI, \"original.names\") taxa <- cbind(\"Old Names\" = oldnames, \"Current Names\" = names(BCI)) noquote(taxa[taxa[,1] != taxa[,2], ]) #> Old Names Current Names #> [1,] Abarema.macradenium Abarema.macradenia #> [2,] Acacia.melanoceras Vachellia.melanoceras #> [3,] Apeiba.aspera Apeiba.glabra #> [4,] Aspidosperma.cruenta Aspidosperma.desmanthum #> [5,] Cassipourea.elliptica Cassipourea.guianensis #> [6,] Cespedezia.macrophylla Cespedesia.spathulata #> [7,] Chlorophora.tinctoria Maclura.tinctoria #> [8,] Coccoloba.manzanillensis Coccoloba.manzinellensis #> [9,] Coussarea.curvigemmia Coussarea.curvigemma #> [10,] Cupania.sylvatica Cupania.seemannii #> [11,] Dipteryx.panamensis Dipteryx.oleifera #> [12,] Eugenia.coloradensis Eugenia.florida #> [13,] Eugenia.oerstedeana Eugenia.oerstediana #> [14,] Guapira.standleyana Guapira.myrtiflora #> [15,] Hyeronima.alchorneoides Hieronyma.alchorneoides #> [16,] Inga.marginata Inga.semialata #> [17,] Lonchocarpus.latifolius Lonchocarpus.heptaphyllus #> [18,] Maquira.costaricana Maquira.guianensis.costaricana #> [19,] Phoebe.cinnamomifolia Cinnamomum.triplinerve #> [20,] Swartzia.simplex.var.ochnacea Swartzia.simplex.continentalis #> [21,] Tabebuia.guayacan Handroanthus.guayacan"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":null,"dir":"Reference","previous_headings":"","what":"Canonical Correlation Analysis — CCorA","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis, following Brian McArdle's unpublished graduate course notes, plus improvements allow calculations case sparse collinear matrices, permutation test Pillai's trace statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Canonical Correlation Analysis — CCorA","text":"","code":"CCorA(Y, X, stand.Y=FALSE, stand.X=FALSE, permutations = 0, ...) # S3 method for CCorA biplot(x, plot.type=\"ov\", xlabs, plot.axes = 1:2, int=0.5, col.Y=\"red\", col.X=\"blue\", cex=c(0.7,0.9), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Canonical Correlation Analysis — CCorA","text":"Y Left matrix (object class: matrix data.frame). X Right matrix (object class: matrix data.frame). stand.Y Logical; Y standardized? stand.X Logical; X standardized? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x CCoaR result object. plot.type character string indicating following plots produced: \"objects\", \"variables\", \"ov\" (separate graphs objects variables), \"biplots\". unambiguous subset containing first letters names can used instead full names. xlabs Row labels. default use row names, NULL uses row numbers instead, NA suppresses plotting row names completely. plot.axes vector 2 values containing order numbers canonical axes plotted. Default: first two axes. int Radius inner circles plotted visual references plots variables. Default: int=0.5. int=0, inner circle plotted. col.Y Color used objects variables first data table (Y) plots. biplots, objects black. col.X Color used objects variables second data table (X) plots. cex vector 2 values containing size reduction factors object variable names, respectively, plots. Default values: cex=c(0.7,0.9). ... arguments passed functions. function biplot.CCorA passes graphical arguments biplot biplot.default. CCorA currently ignores extra arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis (Hotelling 1936) seeks linear combinations variables Y maximally correlated linear combinations variables X. analysis estimates relationships displays graphs. Pillai's trace statistic computed tested parametrically (F-test); permutation test also available. Algorithmic note -- blunt approach read two matrices, compute covariance matrices, matrix S12 %*% inv(S22) %*% t(S12) %*% inv(S11). trace Pillai's trace statistic. approach may fail, however, heavy multicollinearity sparse data matrices. safe approach replace data matrices PCA object scores. function can produce different types plots depending option chosen: \"objects\" produces two plots objects, one space Y, second space X; \"variables\" produces two plots variables, one variables Y space Y, second variables X space X; \"ov\" produces four plots, two objects two variables; \"biplots\" produces two biplots, one first matrix (Y) one second matrix (X) solutions. biplots, function passes arguments biplot.default; consult help page configuring biplots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Canonical Correlation Analysis — CCorA","text":"Function CCorA returns list containing following elements: Pillai Pillai's trace statistic = sum canonical eigenvalues. Eigenvalues Canonical eigenvalues. squares canonical correlations. CanCorr Canonical correlations. Mat.ranks Ranks matrices Y X. RDA.Rsquares Bimultivariate redundancy coefficients (R-squares) RDAs Y|X X|Y. RDA.adj.Rsq RDA.Rsquares adjusted n number explanatory variables. nperm Number permutations. p.Pillai Parametric probability value associated Pillai's trace. p.perm Permutational probability associated Pillai's trace. Cy Object scores Y biplot. Cx Object scores X biplot. corr.Y.Cy Scores Y variables Y biplot, computed cor(Y,Cy). corr.X.Cx Scores X variables X biplot, computed cor(X,Cx). corr.Y.Cx cor(Y,Cy) available plotting variables Y space X manually. corr.X.Cy cor(X,Cx) available plotting variables X space Y manually. control list control values permutations returned function . call Call CCorA function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Canonical Correlation Analysis — CCorA","text":"Hotelling, H. 1936. Relations two sets variates. Biometrika 28: 321-377. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Canonical Correlation Analysis — CCorA","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal. Implemented vegan help Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Canonical Correlation Analysis — CCorA","text":"","code":"# Example using two mite groups. The mite data are available in vegan data(mite) # Two mite species associations (Legendre 2005, Fig. 4) group.1 <- c(1,2,4:8,10:15,17,19:22,24,26:30) group.2 <- c(3,9,16,18,23,25,31:35) # Separate Hellinger transformations of the two groups of species mite.hel.1 <- decostand(mite[,group.1], \"hel\") mite.hel.2 <- decostand(mite[,group.2], \"hel\") rownames(mite.hel.1) = paste(\"S\",1:nrow(mite),sep=\"\") rownames(mite.hel.2) = paste(\"S\",1:nrow(mite),sep=\"\") out <- CCorA(mite.hel.1, mite.hel.2) out #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mite.hel.1, X = mite.hel.2) #> #> Y X #> Matrix Ranks 24 11 #> #> Pillai's trace: 4.573009 #> #> Significance of Pillai's trace: #> from F-distribution: 0.0032737 #> CanAxis1 CanAxis2 CanAxis3 CanAxis4 CanAxis5 CanAxis6 #> Canonical Correlations 0.92810 0.82431 0.81209 0.74981 0.70795 0.65950 #> CanAxis7 CanAxis8 CanAxis9 CanAxis10 CanAxis11 #> Canonical Correlations 0.50189 0.48179 0.41089 0.37823 0.28 #> #> Y | X X | Y #> RDA R squares 0.33224 0.5376 #> adj. RDA R squares 0.20560 0.2910 #> biplot(out, \"ob\") # Two plots of objects biplot(out, \"v\", cex=c(0.7,0.6)) # Two plots of variables biplot(out, \"ov\", cex=c(0.7,0.6)) # Four plots (2 for objects, 2 for variables) biplot(out, \"b\", cex=c(0.7,0.6)) # Two biplots biplot(out, xlabs = NA, plot.axes = c(3,5)) # Plot axes 3, 5. No object names biplot(out, plot.type=\"biplots\", xlabs = NULL) # Replace object names by numbers # Example using random numbers. No significant relationship is expected mat1 <- matrix(rnorm(60),20,3) mat2 <- matrix(rnorm(100),20,5) out2 = CCorA(mat1, mat2, permutations=99) out2 #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mat1, X = mat2, permutations = 99) #> #> Y X #> Matrix Ranks 3 5 #> #> Pillai's trace: 0.480458 #> #> Significance of Pillai's trace: #> from F-distribution: 0.90606 #> based on permutations: 0.94 #> Permutation: free #> Number of permutations: 99 #> #> CanAxis1 CanAxis2 CanAxis3 #> Canonical Correlations 0.64421 0.23458 0.1021 #> #> Y | X X | Y #> RDA R squares 0.214302 0.0839 #> adj. RDA R squares -0.066305 -0.0879 #> biplot(out2, \"b\")"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Add New Points to NMDS ordination — MDSaddpoints","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Add new points existing metaMDS monoMDS ordination.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"MDSaddpoints(nmds, dis, neighbours = 5, maxit = 200) dist2xy(dist, pick, type = c(\"xy\", \"xx\"), invert = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"nmds Result object metaMDS monoMDS. configuration points fixed, new points added. dis Rectangular non-symmetric dissimilarity matrix among new points (rows) old fixed points (columns). matrix can extracted complete dissimilarities old new points dist2xy, calculated designdist2. neighbours Number nearest points used get starting locations new points. maxit Maximum number iterations. dist Input dissimilarities. pick Indices (integers) selected observations logical vector TRUE picked items. output original order reordered argument. type \"xy\" returns rectangular data picked picked observations, \"xx\" subset symmetric dissimilarities. invert Invert pick: drop elements listed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function return list class \"nmds\" (objects type vegan) following elements points Coordinates added new points seeds Starting coordinates new points. deltastress Change stress added points. iters Number iterations. cause Cause termination iterations. Integer convergence criteria monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function provides interface monoMDS Fortran code add new points existing ordination regarded fixed. function similar role predict functions newdata Euclidean ordination (e.g. predict.cca). Input data must rectangular matrix distances among new added points (rows) fixed old points (columns). matrices can extracted complete dissimilarities helper function dist2xy. Function designdist2 can directly calculate rectangular dissimilarity matrices sampling units (rows) two matries. addition, analogue distance function can calculate dissimilarities among two matrices, including functions specified designdist2. Great care needed preparing dissimilarities input. dissimilarity index must exactly fixed ordination, columns must match old fixed points, rows added new points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"## Cross-validation: remove a point when performing NMDS and add as ## a new points data(dune) d <- vegdist(dune) ## remove point 3 from ordination mod3 <- metaMDS(dist2xy(d, 3, \"xx\", invert = TRUE), trace=0) ## add point 3 to the result MDSaddpoints(mod3, dist2xy(d, 3)) #> $points #> MDS1 MDS2 #> 3 -0.09726881 -0.4560917 #> #> $seed #> [,1] [,2] #> [1,] 0.01222493 -0.4353716 #> #> $deltastress #> [1] 0.001811377 #> #> $iters #> [1] 18 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\" ## Use designdist2 d15 <- designdist(dune[1:15,]) m15 <- metaMDS(d15, trace=0) MDSaddpoints(m15, designdist2(dune[1:15,], dune[16:20,])) #> $points #> MDS1 MDS2 #> 16 0.4296366 -0.3068902 #> 17 -0.4122463 0.4118309 #> 18 -0.1135994 0.3120708 #> 19 0.1421922 0.4869128 #> 20 0.5892538 0.1751219 #> #> $seed #> [,1] [,2] #> [1,] 0.2832566 0.019571976 #> [2,] -0.1920994 0.137270818 #> [3,] -0.1529998 0.094374124 #> [4,] -0.0244597 0.137988191 #> [5,] 0.3212258 0.009941098 #> #> $deltastress #> [1] 0.03394768 #> #> $iters #> [1] 33 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":null,"dir":"Reference","previous_headings":"","what":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function rotates multidimensional scaling result first dimension parallel external (environmental variable). function can handle results metaMDS monoMDS functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"MDSrotate(object, vec, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"object result object metaMDS monoMDS. vec environmental variable matrix variables. number variables must lower number dimensions, solution rotated variables order appear matrix. Alternatively vec can factor, solution rotated optimal separation factor levels using lda. na.rm Remove missing values continuous variable vec. ... arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"orientation rotation undefined multidimensional scaling. Functions metaMDS metaMDS can rotate solutions principal components dispersion points highest first dimension. Sometimes different rotation intuitive, MDSrotate allows rotation result first axis parallel given external variable two first variables completely two-dimensional plane etc. several external variables supplied, applied order matrix. First axis rotated first supplied variable, second axis second variable. variables usually correlated, second variable usually aligned second axis, uncorrelated later dimensions. must least one free dimension: number external variables must lower number dimensions, used environmental variables uncorrelated free dimension. Alternatively method can rotate discriminate levels factor using linear discriminant analysis (lda). hardly meaningful two-dimensional solutions, since rotations two dimensions separation cluster levels. However, function can useful finding two-dimensional projection clusters two dimensions. last dimension always show residual variation, \\(k\\) dimensions, \\(k-1\\) discrimination vectors used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function returns original ordination result, rotated scores (site species available), pc attribute scores set FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Rotation factor variable experimental feature may removed. discriminant analysis weights dimensions discriminating power, MDSrotate performs rigid rotation. Therefore solution may optimal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"data(varespec) data(varechem) mod <- monoMDS(vegdist(varespec)) mod <- with(varechem, MDSrotate(mod, pH)) plot(mod) ef <- envfit(mod ~ pH, varechem, permutations = 0) plot(ef) ordisurf(mod ~ pH, varechem, knots = 1, add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ poly(x1, 1) + poly(x2, 1) #> Total model degrees of freedom 3 #> #> REML score: -2.736059"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":null,"dir":"Reference","previous_headings":"","what":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds & Shaw test concerns location highest (hump) lowest (pit) value quadratic curve given points. Typically, used study whether quadratic hump pit located within studied interval. current test generalized applies generalized linear models (glm) link function instead simple quadratic curve. test popularized ecology analysis humped species richness patterns (Mittelbach et al. 2001), general. logarithmic link function, quadratic response defines Gaussian response model ecological gradients (ter Braak & Looman 1986), test can used inspecting location Gaussian optimum within given range gradient. can also used replace Tokeshi's test “bimodal” species frequency distribution.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"MOStest(x, y, interval, ...) # S3 method for MOStest plot(x, which = c(1,2,3,6), ...) fieller.MOStest(object, level = 0.95) # S3 method for MOStest profile(fitted, alpha = 0.01, maxsteps = 10, del = zmax/5, ...) # S3 method for MOStest confint(object, parm = 1, level = 0.95, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"x independent variable plotting object plot. y dependent variable. interval two points test statistic evaluated. missing, extremes x used. Subset plots produced. Values =1 2 define plots specific MOStest (see Details), larger values select graphs plot.lm (minus 2). object, fitted result object MOStest. level confidence level required. alpha Maximum significance level allowed. maxsteps Maximum number steps profile. del step length parameter profile (see code). parm Ignored. ... variables passed functions. Function MOStest passes glm can include family. functions pass underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function fits quadratic curve \\(\\mu = b_0 + b_1 x + b_2 x^2\\) given family link function. \\(b_2 < 0\\), defines unimodal curve highest point \\(u = -b_1/(2 b_2)\\) (ter Braak & Looman 1986). \\(b_2 > 0\\), parabola minimum \\(u\\) response sometimes called “bimodal”. null hypothesis extreme point \\(u\\) located within interval given points \\(p_1\\) \\(p_2\\). extreme point \\(u\\) exactly \\(p_1\\), \\(b_1 = 0\\) shifted axis \\(x - p_1\\). test, origin x shifted values \\(p_1\\) \\(p_2\\), test statistic based differences deviances original model model origin forced given location using standard anova.glm function (Oksanen et al. 2001). Mitchell-Olds & Shaw (1987) used first degree coefficient significance estimated summary.glm function. give identical results Normal error, error distributions preferable use test based differences deviances fitted models. test often presented general test location hump, really dependent quadratic fitted curve. hump different form quadratic, test may insignificant. strong assumptions test, use support functions inspect fit. Function plot(..., =1) displays data points, fitted quadratic model, approximate 95% confidence intervals (2 times SE). Function plot = 2 displays approximate confidence interval polynomial coefficients, together two lines indicating combinations coefficients produce evaluated points x. Moreover, cross-hair shows approximate confidence intervals polynomial coefficients ignoring correlations. Higher values produce corresponding graphs plot.lm. , must add 2 value plot.lm. Function fieller.MOStest approximates confidence limits location extreme point (hump pit) using Fieller's theorem following ter Braak & Looman (1986). test based quasideviance except family poisson binomial. Function profile evaluates profile deviance fitted model, confint finds profile based confidence limits following Oksanen et al. (2001). test typically used assessing significance diversity hump productivity gradient (Mittelbach et al. 2001). also can used location pit (deepest points) instead Tokeshi test. , can used test location Gaussian optimum ecological gradient analysis (ter Braak & Looman 1986, Oksanen et al. 2001).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function based glm, returns result object glm amended result test. new items MOStest : isHump TRUE response hump. isBracketed TRUE hump pit bracketed evaluated points. hump Sorted vector location hump pit points test evaluated. coefficients Table test statistics significances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds, T. & Shaw, R.G. 1987. Regression analysis natural selection: statistical inference biological interpretation. Evolution 41, 1149--1161. Mittelbach, G.C. Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, R.M., Dodson, S.. & Gough, L. 2001. observed relationship species richness productivity? Ecology 82, 2381--2396. Oksanen, J., Läärä, E., Tolonen, K. & Warner, B.G. 2001. Confidence intervals optimum Gaussian response function. Ecology 82, 1191--1197. ter Braak, C.J.F & Looman, C.W.N 1986. Weighted averaging, logistic regression Gaussian response model. Vegetatio 65, 3--11.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Function fieller.MOStest based package optgrad Ecological Archives (https://figshare.com/articles/dataset/Full_Archive/3521975) accompanying Oksanen et al. (2001). Ecological Archive package optgrad also contains profile deviance method location hump pit, current implementation profile confint rather follow example profile.glm confint.glm MASS package.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"## The Al-Mufti data analysed in humpfit(): mass <- c(140,230,310,310,400,510,610,670,860,900,1050,1160,1900,2480) spno <- c(1, 4, 3, 9, 18, 30, 20, 14, 3, 2, 3, 2, 5, 2) mod <- MOStest(mass, spno) ## Insignificant mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: gaussian #> Link function: identity #> #> hump min max #> 46.89749 140.00000 2480.00000 #> ***** Caution: hump/pit not bracketed by the data ****** #> #> min/max F Pr(>F) #> hump at min 140 0.0006 0.9816 #> hump at max 2480 0.3161 0.5852 #> Combined 0.9924 ## ... but inadequate shape of the curve op <- par(mfrow=c(2,2), mar=c(4,4,1,1)+.1) plot(mod) ## Looks rather like log-link with Poisson error and logarithmic biomass mod <- MOStest(log(mass), spno, family=quasipoisson) mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: quasipoisson #> Link function: log #> #> min hump max #> 4.941642 6.243371 7.816014 #> #> min/max F Pr(>F) #> hump at min 4.9416 7.1367 0.02174 * #> hump at max 7.8160 9.0487 0.01191 * #> Combined 0.03338 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod) par(op) ## Confidence Limits fieller.MOStest(mod) #> 2.5 % 97.5 % #> 5.255827 6.782979 confint(mod) #> 2.5 % 97.5 % #> 5.816021 6.574378 plot(profile(mod))"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":null,"dir":"Reference","previous_headings":"","what":"Adjusted R-square — RsquareAdj","title":"Adjusted R-square — RsquareAdj","text":"functions finds adjusted R-square.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Adjusted R-square — RsquareAdj","text":"","code":"# S3 method for default RsquareAdj(x, n, m, ...) # S3 method for rda RsquareAdj(x, ...) # S3 method for cca RsquareAdj(x, permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Adjusted R-square — RsquareAdj","text":"x Unadjusted R-squared object terms evaluation adjusted R-squared can found. n, m Number observations number degrees freedom fitted model. permutations Number permutations use computing adjusted R-squared cca. permutations can calculated parallel specifying number cores passed permutest ... arguments (ignored) except case cca arguments passed permutest.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Adjusted R-square — RsquareAdj","text":"default method finds adjusted \\(R^2\\) unadjusted \\(R^2\\), number observations, number degrees freedom fitted model. specific methods find information fitted result object. specific methods rda (also used distance-based RDA), cca, lm glm. Adjusted, even unadjusted, \\(R^2\\) may available cases, functions return NA. \\(R^2\\) values available gaussian models glm. adjusted, \\(R^2\\) cca computed using permutation approach developed Peres-Neto et al. (2006). default 1000 permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Adjusted R-square — RsquareAdj","text":"functions return list items r.squared adj.r.squared.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Adjusted R-square — RsquareAdj","text":"Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277. Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87, 2614--2625.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Adjusted R-square — RsquareAdj","text":"","code":"data(mite) data(mite.env) ## rda m <- rda(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.5265047 #> #> $adj.r.squared #> [1] 0.4367038 #> ## cca m <- cca(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.4471676 #> #> $adj.r.squared #> [1] 0.3446879 #> ## default method RsquareAdj(0.8, 20, 5) #> [1] 0.7285714"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":null,"dir":"Reference","previous_headings":"","what":"Self-Starting nls Species-Area Models — SSarrhenius","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions provide self-starting species-area models non-linear regression (nls). can also used fitting species accumulation models fitspecaccum. models (many ) reviewed Dengler (2009).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"SSarrhenius(area, k, z) SSgleason(area, k, slope) SSgitay(area, k, slope) SSlomolino(area, Asym, xmid, slope)"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"area Area size sample: independent variable. k, z, slope, Asym, xmid Estimated model parameters: see Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions assumed used species richness (number species) independent variable, area sample size independent variable. Basically, define least squares models untransformed data, differ models transformed species richness models non-Gaussian error. Arrhenius model (SSarrhenius) expression k*area^z. classical model can found textbook ecology (also Dengler 2009). Parameter z steepness species-area curve, k expected number species unit area. Gleason model (SSgleason) linear expression k + slope*log(area) (Dengler 200). linear model, starting values give final estimates; provided ease comparison models. Gitay model (SSgitay) quadratic logarithmic expression (k + slope*log(area))^2 (Gitay et al. 1991, Dengler 2009). Parameter slope steepness species-area curve, k square root expected richness unit area. Lomolino model (SSlomolino) Asym/(1 + slope^log(xmid/area)) (Lomolino 2000, Dengler 2009). Parameter Asym asymptotic maximum number species, slope maximum slope increase richness, xmid area half maximum richness achieved. addition models, several models studied Dengler (2009) available standard R self-starting models: Michaelis-Menten (SSmicmen), Gompertz (SSgompertz), logistic (SSlogis), Weibull (SSweibull), others may useful.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Numeric vector length area. value expression model. arguments names objects gradient matrix respect names attached attribute named gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Dengler, J. (2009) function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gitay, H., Roxburgh, S.H. & Wilson, J.B. (1991) Species-area relationship New Zealand tussock grassland, implications nature reserve design community structure. Journal Vegetation Science 2, 113--118. Lomolino, M. V. (2000) Ecology's general, yet protean pattern: species-area relationship. Journal Biogeography 27, 17--26.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"## Get species area data: sipoo.map gives the areas of islands data(sipoo, sipoo.map) S <- specnumber(sipoo) plot(S ~ area, sipoo.map, xlab = \"Island Area (ha)\", ylab = \"Number of Species\", ylim = c(1, max(S))) ## The Arrhenius model marr <- nls(S ~ SSarrhenius(area, k, z), data=sipoo.map) marr #> Nonlinear regression model #> model: S ~ SSarrhenius(area, k, z) #> data: sipoo.map #> k z #> 3.4062 0.4364 #> residual sum-of-squares: 78.1 #> #> Number of iterations to convergence: 5 #> Achieved convergence tolerance: 1.056e-06 ## confidence limits from profile likelihood confint(marr) #> Waiting for profiling to be done... #> 2.5% 97.5% #> k 2.6220312 4.3033906 #> z 0.3813576 0.4944693 ## draw a line xtmp <- with(sipoo.map, seq(min(area), max(area), len=51)) lines(xtmp, predict(marr, newdata=data.frame(area = xtmp)), lwd=2) ## The normal way is to use linear regression on log-log data, ## but this will be different from the previous: mloglog <- lm(log(S) ~ log(area), data=sipoo.map) mloglog #> #> Call: #> lm(formula = log(S) ~ log(area), data = sipoo.map) #> #> Coefficients: #> (Intercept) log(area) #> 1.0111 0.4925 #> lines(xtmp, exp(predict(mloglog, newdata=data.frame(area=xtmp))), lty=2) ## Gleason: log-linear mgle <- nls(S ~ SSgleason(area, k, slope), sipoo.map) lines(xtmp, predict(mgle, newdata=data.frame(area=xtmp)), lwd=2, col=2) ## Gitay: quadratic of log-linear mgit <- nls(S ~ SSgitay(area, k, slope), sipoo.map) lines(xtmp, predict(mgit, newdata=data.frame(area=xtmp)), lwd=2, col = 3) ## Lomolino: using original names of the parameters (Lomolino 2000): mlom <- nls(S ~ SSlomolino(area, Smax, A50, Hill), sipoo.map) mlom #> Nonlinear regression model #> model: S ~ SSlomolino(area, Smax, A50, Hill) #> data: sipoo.map #> Smax A50 Hill #> 53.493 94.697 2.018 #> residual sum-of-squares: 55.37 #> #> Number of iterations to convergence: 6 #> Achieved convergence tolerance: 9.715e-07 lines(xtmp, predict(mlom, newdata=data.frame(area=xtmp)), lwd=2, col = 4) ## One canned model of standard R: mmic <- nls(S ~ SSmicmen(area, Asym, slope), sipoo.map) lines(xtmp, predict(mmic, newdata = data.frame(area=xtmp)), lwd =2, col = 5) legend(\"bottomright\", c(\"Arrhenius\", \"log-log linear\", \"Gleason\", \"Gitay\", \"Lomolino\", \"Michaelis-Menten\"), col=c(1,1,2,3,4,5), lwd=c(2,1,2,2,2,2), lty=c(1,2,1,1,1,1)) ## compare models (AIC) allmods <- list(Arrhenius = marr, Gleason = mgle, Gitay = mgit, Lomolino = mlom, MicMen= mmic) sapply(allmods, AIC) #> Arrhenius Gleason Gitay Lomolino MicMen #> 83.49847 96.94018 80.54984 79.30718 83.02003"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Compute single terms can added dropped constrained ordination model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"# S3 method for cca add1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...) # S3 method for cca drop1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"object constrained ordination object cca, rda, dbrda capscale. scope formula giving terms considered adding dropping; see add1 details. test permutation test added using anova.cca. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ... arguments passed add1.default, drop1.default, anova.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"argument test = \"none\" functions call add1.default drop1.default. argument test = \"permutation\" functions add test results anova.cca. Function drop1.cca call anova.cca argument = \"margin\". Function add1.cca implement test single term additions directly available anova.cca. Functions used implicitly step, ordiR2step ordistep. deviance.cca deviance.rda used step firm basis, setting argument test = \"permutation\" may help getting useful insight validity model building. Function ordistep calls alternately drop1.cca add1.cca argument test = \"permutation\" selects variables permutation \\(P\\)-values. Meticulous use add1.cca drop1.cca allow judicious model building. default number permutations set low value, permutation tests can take long time. sufficient give impression significances terms, higher values permutations used \\(P\\) values really important.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Returns similar object add1 drop1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"data(dune) data(dune.env) ## Automatic model building based on AIC but with permutation tests step(cca(dune ~ 1, dune.env), reformulate(names(dune.env)), test=\"perm\") #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 86.608 2.2536 0.005 ** #> + Management 3 86.935 2.1307 0.005 ** #> + A1 1 87.411 2.1400 0.045 * #> 87.657 #> + Manure 4 88.832 1.5251 0.030 * #> + Use 2 89.134 1.1431 0.200 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC F Pr(>F) #> 86.608 #> + Management 3 86.813 1.4565 0.045 * #> + A1 1 86.992 1.2624 0.175 #> + Use 2 87.259 1.2760 0.085 . #> + Manure 4 87.342 1.3143 0.050 * #> - Moisture 3 87.657 2.2536 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #> ## see ?ordistep to do the same, but based on permutation P-values if (FALSE) { ordistep(cca(dune ~ 1, dune.env), reformulate(names(dune.env))) } ## Manual model building ## -- define the maximal model for scope mbig <- rda(dune ~ ., dune.env) ## -- define an empty model to start with m0 <- rda(dune ~ 1, dune.env) ## -- manual selection and updating add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 89.620 #> A1 1 89.591 1.9217 0.015 * #> Moisture 3 87.707 2.5883 0.005 ** #> Management 3 87.082 2.8400 0.005 ** #> Use 2 91.032 1.1741 0.225 #> Manure 4 89.232 1.9539 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Management) add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 87.082 #> A1 1 87.424 1.2965 0.215 #> Moisture 3 85.567 1.9764 0.010 ** #> Use 2 88.284 1.0510 0.400 #> Manure 3 87.517 1.3902 0.105 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Moisture) ## -- included variables still significant? drop1(m0, test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> Management 3 87.707 2.1769 0.010 ** #> Moisture 3 87.082 1.9764 0.015 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> A1 1 86.220 0.8359 0.605 #> Use 2 86.842 0.8027 0.765 #> Manure 3 85.762 1.1225 0.390"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"additive diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set (gamma diversity). hierarchical null model testing, statistic returned function evaluated according nested hierarchical sampling design (hiersimu).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"adipart(...) # S3 method for default adipart(y, x, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula adipart(formula, data, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) hiersimu(...) # S3 method for default hiersimu(y, x, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula hiersimu(formula, data, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, function performs overall decomposition alpha, beta gamma diversities. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variables referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, diversity index calculated (see Details). weights Character, \"unif\" uniform weights, \"prop\" weighting proportional sample abundances use weighted averaging individual alpha values within strata given level sampling hierarchy. relative Logical, TRUE alpha beta diversity values given relative value gamma function adipart. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. FUN function used hiersimu. must fully specified, currently arguments passed function via .... location Character, identifies function (mean median) used calculate location samples. drop.highest Logical, drop highest level . FUN evaluates arrays least 2 dimensions, highest level dropped, selected . ... arguments passed functions, e.g. base logarithm Shannon diversity, method, thin burnin arguments oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Additive diversity partitioning means mean alpha beta diversities add gamma diversity, thus beta diversity measured dimensions alpha gamma (Lande 1996). additive procedure extended across multiple scales hierarchical sampling design \\(= 1, 2, 3, \\ldots, m\\) levels sampling (Crist et al. 2003). Samples lower hierarchical levels nested within higher level units, thus \\(=1\\) \\(=m\\) grain size increasing constant survey extent. level \\(\\), \\(\\alpha_i\\) denotes average diversity found within samples. highest sampling level, diversity components calculated $$\\beta_m = \\gamma - \\alpha_m$$ lower sampling level $$\\beta_i = \\alpha_{+1} - \\alpha_i$$ , additive partition diversity $$\\gamma = \\alpha_1 + \\sum_{=1}^m \\beta_i$$ Average alpha components can weighted uniformly (weight=\"unif\") calculate simple average, proportionally sample abundances (weight=\"prop\") calculate weighted average follows $$\\alpha_i = \\sum_{j=1}^{n_i} D_{ij} w_{ij}$$ \\(D_{ij}\\) diversity index \\(w_{ij}\\) weight calculated \\(j\\)th sample \\(\\)th sampling level. implementation additive diversity partitioning adipart follows Crist et al. 2003. based species richness (\\(S\\), \\(S-1\\)), Shannon's Simpson's diversity indices stated index argument. expected diversity components calculated nsimul times individual based randomisation community data matrix. done \"r2dtable\" method oecosimu default. hiersimu works almost way adipart, without comparing actual statistic values returned FUN highest possible value (cf. gamma diversity). , cases, difficult ensure additive properties mean statistic values along hierarchy.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"object class \"adipart\" \"hiersimu\" structure oecosimu objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Crist, T.O., Veech, J.., Gering, J.C. Summerville, K.S. (2003). Partitioning species diversity across landscapes regions: hierarchical analysis \\(\\alpha\\), \\(\\beta\\), \\(\\gamma\\)-diversity. . Nat., 162, 734--743. Lande, R. (1996). Statistics partitioning species diversity, similarity among multiple communities. Oikos, 76, 5--13.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l1=1:nrow(mite), l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)), l4=rep(1, nrow(mite)))) ## Let's see in a map par(mfrow=c(1,3)) plot(mite.xy, main=\"l1\", col=as.numeric(levsm$l1)+1, asp = 1) plot(mite.xy, main=\"l2\", col=as.numeric(levsm$l2)+1, asp = 1) plot(mite.xy, main=\"l3\", col=as.numeric(levsm$l3)+1, asp = 1) par(mfrow=c(1,1)) ## Additive diversity partitioning adipart(mite, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -28.979 22.393 21.978 22.429 22.765 0.05 * #> gamma 35.000 0.000 35.000 35.000 35.000 35.000 1.00 #> beta.1 19.886 28.979 12.607 12.235 12.571 13.022 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## the next two define identical models adipart(mite, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, x = levsm, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -54.1889 22.35038 22.10429 22.37143 22.584 0.05 * #> alpha.2 29.750 -25.1503 34.81579 34.50000 34.75000 35.000 0.05 * #> alpha.3 33.000 0.0000 35.00000 35.00000 35.00000 35.000 0.05 * #> gamma 35.000 0.0000 35.00000 35.00000 35.00000 35.000 1.00 #> beta.1 14.636 9.9124 12.46541 12.10000 12.45000 12.779 0.05 * #> beta.2 3.250 15.2208 0.18421 0.00000 0.25000 0.500 0.05 * #> beta.3 2.000 0.0000 0.00000 0.00000 0.00000 0.000 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adipart(mite ~ l2 + l3, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -51.771 22.378947 22.100714 22.400000 22.587 0.05 * #> alpha.2 29.750 -25.578 34.736842 34.500000 34.750000 35.000 0.05 * #> alpha.3 33.000 -17.206 34.973684 34.725000 35.000000 35.000 0.05 * #> gamma 35.000 0.000 35.000000 35.000000 35.000000 35.000 1.00 #> beta.1 14.636 11.165 12.357895 12.025714 12.428571 12.686 0.05 * #> beta.2 3.250 15.455 0.236842 0.000000 0.250000 0.500 0.05 * #> beta.3 2.000 17.206 0.026316 0.000000 0.000000 0.275 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical null model testing ## diversity analysis (similar to adipart) hiersimu(mite, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(y = mite, FUN = diversity, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> level_1 0.76064 -52.201 0.93892 0.93378 0.93866 0.9444 0.05 * #> leve_2 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 hiersimu(mite ~ l2 + l3, levsm, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ l2 + l3, data = levsm, FUN = diversity, #> relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> unit 0.76064 -60.783 0.93927 0.93425 0.93883 0.9435 0.05 * #> l2 0.89736 -116.072 0.99795 0.99617 0.99808 0.9991 0.05 * #> l3 0.92791 -421.205 0.99932 0.99902 0.99935 0.9995 0.05 * #> all 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical testing with the Morisita index morfun <- function(x) dispindmorisita(x)$imst hiersimu(mite ~., levsm, morfun, drop.highest=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ ., data = levsm, FUN = morfun, #> drop.highest = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> l1 0.52070 6.4968 0.35648 0.31140 0.35449 0.3978 0.05 * #> l2 0.60234 12.5142 0.15359 0.08900 0.14525 0.2039 0.05 * #> l3 0.67509 17.6125 -0.19189 -0.26300 -0.20619 -0.1149 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Analysis variance using distance matrices --- partitioning distance matrices among sources variation fitting linear models (e.g., factors, polynomial regression) distance matrices; uses permutation test pseudo-\\(F\\) ratios.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"adonis2(formula, data, permutations = 999, method = \"bray\", sqrt.dist = FALSE, add = FALSE, by = \"terms\", parallel = getOption(\"mc.cores\"), na.action = na.fail, strata = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"formula Model formula. left-hand side (LHS) formula must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist used find dissimilarities. right-hand side (RHS) formula defines independent variables. can continuous variables factors, can transformed within formula, can interactions typical formula. data data frame independent variables, rows order community data matrix dissimilarity matrix named LHS formula. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. method name method used vegdist calculate pairwise distances left hand side formula data frame matrix. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" analyse one-degree--freedom contrasts sequentially, = NULL assess overall significance terms together. arguments passed anova.cca. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. na.action Handling missing values right-hand-side formula (see na.fail explanation alternatives). Missing values allowed left-hand-side. NB, argument subset implemented. strata Groups within constrain permutations. traditional non-movable strata set Blocks permute package, flexible alternatives may appropriate. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"adonis2 function analysis partitioning sums squares using dissimilarities. function based principles McArdle & Anderson (2001) can perform sequential, marginal overall tests. function also allows using additive constants squareroot dissimilarities avoid negative eigenvalues, can also handle semimetric indices (Bray-Curtis) produce negative eigenvalues. adonis2 tests identical anova.cca dbrda. Euclidean distances, tests also identical anova.cca rda. function partitions sums squares multivariate data set, directly analogous MANOVA (multivariate analysis variance). McArdle Anderson (2001) Anderson (2001) refer method “permutational MANOVA” (formerly “nonparametric MANOVA”). , inputs linear predictors, response matrix arbitrary number columns, robust alternative parametric MANOVA ordination methods describing variation attributed different experimental treatments uncontrolled covariates. method also analogous distance-based redundancy analysis algorithmically similar dbrda (Legendre Anderson 1999), provides alternative AMOVA (nested analysis molecular variance, Excoffier, Smouse, Quattro, 1992; amova ade4 package) crossed nested factors.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"function returns anova.cca result object new column partial \\(R^2\\): proportion sum squares total, marginal models (= \"margin\") \\(R^2\\) terms add 1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson (2001, Fig. 4) warns method may confound location dispersion effects: significant differences may caused different within-group variation (dispersion) instead different mean values groups (see Warton et al. 2012 general analysis). However, seems adonis2 less sensitive dispersion effects alternatives (anosim, mrpp). Function betadisper sister function adonis2 study differences dispersion within geometric framework.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson, M.J. 2001. new method non-parametric multivariate analysis variance. Austral Ecology, 26: 32--46. Excoffier, L., P.E. Smouse, J.M. Quattro. 1992. Analysis molecular variance inferred metric distances among DNA haplotypes: Application human mitochondrial DNA restriction data. Genetics, 131:479--491. Legendre, P. M.J. Anderson. 1999. Distance-based redundancy analysis: Testing multispecies responses multifactorial ecological experiments. Ecological Monographs, 69:1--24. McArdle, B.H. M.J. Anderson. 2001. Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology, 82: 290--297. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Martin Henry H. Stevens Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"data(dune) data(dune.env) ## default test by terms adonis2(dune ~ Management*A1, data = dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.2629 0.002 ** #> A1 1 0.4409 0.10256 2.9387 0.012 * #> Management:A1 3 0.5892 0.13705 1.3090 0.217 #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## overall tests adonis2(dune ~ Management*A1, data = dune.env, by = NULL) #> Permutation test for adonis under reduced model #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env, by = NULL) #> Df SumOfSqs R2 F Pr(>F) #> Model 7 2.4987 0.58122 2.3792 0.004 ** #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ### Example of use with strata, for nested (e.g., block) designs. dat <- expand.grid(rep=gl(2,1), NO3=factor(c(0,10)),field=gl(3,1) ) dat #> rep NO3 field #> 1 1 0 1 #> 2 2 0 1 #> 3 1 10 1 #> 4 2 10 1 #> 5 1 0 2 #> 6 2 0 2 #> 7 1 10 2 #> 8 2 10 2 #> 9 1 0 3 #> 10 2 0 3 #> 11 1 10 3 #> 12 2 10 3 Agropyron <- with(dat, as.numeric(field) + as.numeric(NO3)+2) +rnorm(12)/2 Schizachyrium <- with(dat, as.numeric(field) - as.numeric(NO3)+2) +rnorm(12)/2 total <- Agropyron + Schizachyrium dotplot(total ~ NO3, dat, jitter.x=TRUE, groups=field, type=c('p','a'), xlab=\"NO3\", auto.key=list(columns=3, lines=TRUE) ) Y <- data.frame(Agropyron, Schizachyrium) mod <- metaMDS(Y, trace = FALSE) plot(mod) ### Ellipsoid hulls show treatment with(dat, ordiellipse(mod, NO3, kind = \"ehull\", label = TRUE)) ### Spider shows fields with(dat, ordispider(mod, field, lty=3, col=\"red\", label = TRUE)) ### Incorrect (no strata) adonis2(Y ~ NO3, data = dat, permutations = 199) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.06 . #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Correct with strata with(dat, adonis2(Y ~ NO3, data = dat, permutations = 199, strata = field)) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Blocks: strata #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199, strata = field) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.005 ** #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":null,"dir":"Reference","previous_headings":"","what":"Analysis of Similarities — anosim","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Analysis of Similarities — anosim","text":"","code":"anosim(x, grouping, permutations = 999, distance = \"bray\", strata = NULL, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Analysis of Similarities — anosim","text":"x Data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. distance Choice distance metric measures dissimilarity two observations. See vegdist options. used x dissimilarity structure symmetric square matrix. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units. Function anosim operates directly dissimilarity matrix. suitable dissimilarity matrix produced functions dist vegdist. method philosophically allied NMDS ordination (monoMDS), uses rank order dissimilarity values. two groups sampling units really different species composition, compositional dissimilarities groups greater within groups. anosim statistic \\(R\\) based difference mean ranks groups (\\(r_B\\)) within groups (\\(r_W\\)): $$R = (r_B - r_W)/(N (N-1) / 4)$$ divisor chosen \\(R\\) interval \\(-1 \\dots +1\\), value \\(0\\) indicating completely random grouping. statistical significance observed \\(R\\) assessed permuting grouping vector obtain empirical distribution \\(R\\) null-model. See permutations additional details permutation tests Vegan. distribution simulated values can inspected permustats function. function summary plot methods. show valuable information assess validity method: function assumes ranked dissimilarities within groups equal median range. plot method uses boxplot options notch=TRUE varwidth=TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Analysis of Similarities — anosim","text":"function returns list class \"anosim\" following items: call Function call. statistic value ANOSIM statistic \\(R\\) signif Significance permutation. perm Permutation values \\(R\\). distribution permutation values can inspected function permustats. class.vec Factor value dissimilarities classes class name corresponding dissimilarity within class. dis.rank Rank dissimilarity entry. dissimilarity name dissimilarity index: \"method\" entry dist object. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Analysis of Similarities — anosim","text":"Clarke, K. R. (1993). Non-parametric multivariate analysis changes community structure. Australian Journal Ecology 18, 117--143. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Analysis of Similarities — anosim","text":"Jari Oksanen, help Peter R. Minchin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Analysis of Similarities — anosim","text":"anosim function can confound differences groups dispersion within groups results can difficult interpret (cf. Warton et al. 2012). function returns lot information ease studying performance. anosim models analysed adonis2 seems robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Analysis of Similarities — anosim","text":"","code":"data(dune) data(dune.env) dune.dist <- vegdist(dune) dune.ano <- with(dune.env, anosim(dune.dist, Management)) summary(dune.ano) #> #> Call: #> anosim(x = dune.dist, grouping = Management) #> Dissimilarity: bray #> #> ANOSIM statistic R: 0.2579 #> Significance: 0.009 #> #> Permutation: free #> Number of permutations: 999 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.121 0.164 0.205 0.242 #> #> Dissimilarity ranks between and within classes: #> 0% 25% 50% 75% 100% N #> Between 4 58.50 104.00 145.500 188.0 147 #> BF 5 15.25 25.50 41.250 57.0 3 #> HF 1 7.25 46.25 68.125 89.5 10 #> NM 6 64.75 124.50 156.250 181.0 15 #> SF 3 32.75 53.50 99.250 184.0 15 #> plot(dune.ano) #> Warning: some notches went outside hinges ('box'): maybe set notch=FALSE"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function performs ANOVA like permutation test Constrained Correspondence Analysis (cca), Redundancy Analysis (rda) distance-based Redundancy Analysis (dbRDA, dbrda) assess significance constraints.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"# S3 method for cca anova(object, ..., permutations = how(nperm=999), by = NULL, model = c(\"reduced\", \"direct\", \"full\"), parallel = getOption(\"mc.cores\"), strata = NULL, cutoff = 1, scope = NULL) # S3 method for cca permutest(x, permutations = how(nperm = 99), model = c(\"reduced\", \"direct\", \"full\"), by = NULL, first = FALSE, strata = NULL, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"object One several result objects cca, rda, dbrda capscale. several result objects, compared order supplied. single object, test specified overall test given. x single ordination result object. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Setting = \"axis\" assess significance constrained axis, setting = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" assess sequentially one-degree--freedom contrasts split factors. model Permutation model: model=\"direct\" permutes community data, model=\"reduced\" permutes residuals community data Conditions (partial model), model = \"full\" permutes residuals Conditions Constraints. parallel Use parallel processing given number cores. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. error use permutations matrix, defines blocks. legacy argument deprecated future: use permutations = (..., blocks) instead. cutoff effective =\"axis\" stops permutations axis equals exceeds cutoff \\(p\\)-value. scope effective =\"margin\" can used select marginal terms testing. default test marginal terms drop.scope. first Analyse significance first axis. ... Parameters passed functions. anova.cca passes arguments permutest.cca. anova = \"axis\" can use argument cutoff (defaults 1) stops permutations exceeding given level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Functions anova.cca permutest.cca implement ANOVA like permutation tests joint effect constraints cca, rda, dbrda capscale. Function anova intended user-friendly alternative permutest (real workhorse). Function anova can analyse sequence constrained ordination models. analysis based differences residual deviance permutations nested models. default test sum constrained eigenvalues. Setting first = TRUE perform test first constrained eigenvalue. Argument first can set either anova.cca permutest.cca. also possible perform significance tests axis term (constraining variable) using argument anova.cca. Setting = \"axis\" perform separate significance tests constrained axis. previous constrained axes used conditions (“partialled ”) test first constrained eigenvalues performed (Legendre et al. 2011). can stop permutation tests exceeding given significance level argument cutoff speed calculations large models. Setting = \"terms\" perform separate significance test term (constraining variable). terms assessed sequentially first last, order terms influence significances. Setting = \"onedf\" perform similar sequential test one-degree--freedom effects, multi-level factors split contrasts. Setting = \"margin\" perform separate significance test marginal term model terms. marginal test also accepts scope argument drop.scope can character vector term labels analysed, fitted model lower scope. marginal effects also known “Type III” effects, current function evaluates marginal terms. , instance, ignore main effects included interaction terms. calculating pseudo-\\(F\\), terms compared residual full model. Community data permuted choice model=\"direct\", residuals partial CCA/ RDA/ dbRDA choice model=\"reduced\" (default). partial CCA/ RDA/ dbRDA stage, model=\"reduced\" simply permutes data equivalent model=\"direct\". test statistic “pseudo-\\(F\\)”, ratio constrained unconstrained total Inertia (Chi-squares, variances something similar), divided respective degrees freedom. conditions (“partial” terms), sum eigenvalues remains constant, pseudo-\\(F\\) eigenvalues give equal results. partial CCA/ RDA/ dbRDA, effect conditioning variables (“covariables”) removed permutation, total Chi-square fixed, test based pseudo-\\(F\\) differ test based plain eigenvalues.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function anova.cca calls permutest.cca fills anova table. Additional attributes Random.seed (random seeds used), control (permutation design, see ) F.perm (permuted test statistics).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Legendre, P. Legendre, L. (2012). Numerical Ecology. 3rd English ed. Elsevier. Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"data(dune, dune.env) mod <- cca(dune ~ Moisture + Management, dune.env) ## overall test anova(mod) #> Permutation test for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Model 6 1.0024 1.9515 0.003 ** #> Residual 13 1.1129 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## tests for individual terms anova(mod, by=\"term\") #> Permutation test for cca under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.62831 2.4465 0.001 *** #> Management 3 0.37407 1.4565 0.049 * #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 anova(mod, by=\"margin\") #> Permutation test for cca under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.39854 1.5518 0.036 * #> Management 3 0.37407 1.4565 0.053 . #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential test for contrasts anova(mod, by = \"onedf\") #> Permutation test for cca under reduced model #> Sequential test for contrasts #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture.L 1 0.41081 4.7988 0.001 *** #> Moisture.Q 1 0.11261 1.3154 0.167 #> Moisture.C 1 0.10489 1.2253 0.226 #> ManagementHF 1 0.08849 1.0337 0.359 #> ManagementNM 1 0.20326 2.3744 0.011 * #> ManagementSF 1 0.08231 0.9615 0.455 #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## test for adding all environmental variables anova(mod, cca(dune ~ ., dune.env)) #> Permutation tests for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model 1: dune ~ Moisture + Management #> Model 2: dune ~ A1 + Moisture + Management + Use + Manure #> ResDf ResChiSquare Df ChiSquare F Pr(>F) #> 1 13 1.1129 #> 2 7 0.6121 6 0.50079 0.9545 0.532"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Averaged Subsampled Dissimilarity Matrices — avgdist","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function computes dissimilarity matrix dataset multiple times using vegdist randomly subsampling dataset time. subsampled iterations averaged (mean) provide distance matrix represents average multiple subsampling iterations. emulates behavior distance matrix calculator within Mothur microbial ecology toolkit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"avgdist(x, sample, distfun = vegdist, meanfun = mean, transf = NULL, iterations = 100, dmethod = \"bray\", diag = TRUE, upper = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"x Community data matrix. sample subsampling depth used iteration. Samples meet threshold removed analysis, identity returned user stdout. distfun dissimilarity matrix function used. Default vegan vegdist meanfun calculation use average (mean median). transf Option transforming count data calculating distance matrix. base transformation option can used (e.g. sqrt) iterations number random iterations perform averaging. Default 100 iterations. dmethod Dissimilarity index used specified dissimilarity matrix function. Default Bray-Curtis diag, upper Return dissimilarities diagonal upper triangle. NB. default differs vegdist returns symmetric \"dist\" structure instead lower diagonal. However, object accessed matrix indices unless cast matrix .matrix. ... additional arguments add distance function mean/median function specified.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"Geoffrey Hannigan, minor tweaks Gavin L. Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function builds function rrarefy additional distance matrix function (e.g. vegdist) add meaningful representations distances among randomly subsampled datasets presenting average multiple random iterations. function runs using vegdist. functionality utilized Mothur standalone microbial ecology toolkit .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"# Import an example count dataset data(BCI) # Test the base functionality mean.avg.dist <- avgdist(BCI, sample = 50, iterations = 10) # Test the transformation function mean.avg.dist.t <- avgdist(BCI, sample = 50, iterations = 10, transf = sqrt) # Test the median functionality median.avg.dist <- avgdist(BCI, sample = 50, iterations = 10, meanfun = median) # Print the resulting tables head(as.matrix(mean.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 #> 1 0.000 0.556 0.606 0.648 0.620 0.624 0.610 0.606 0.654 0.624 0.638 0.648 0.708 #> 2 0.556 0.000 0.568 0.568 0.632 0.568 0.532 0.554 0.576 0.578 0.588 0.560 0.678 #> 3 0.606 0.568 0.000 0.582 0.578 0.618 0.600 0.568 0.610 0.574 0.610 0.622 0.738 #> 4 0.648 0.568 0.582 0.000 0.624 0.654 0.610 0.572 0.594 0.594 0.596 0.636 0.672 #> 5 0.620 0.632 0.578 0.624 0.000 0.602 0.662 0.626 0.670 0.606 0.678 0.716 0.754 #> 6 0.624 0.568 0.618 0.654 0.602 0.000 0.598 0.566 0.616 0.644 0.576 0.568 0.706 #> 14 15 16 17 18 19 20 21 22 23 24 25 26 #> 1 0.630 0.648 0.600 0.646 0.744 0.682 0.626 0.600 0.646 0.700 0.658 0.620 0.620 #> 2 0.600 0.592 0.554 0.564 0.660 0.584 0.598 0.606 0.544 0.642 0.574 0.624 0.592 #> 3 0.628 0.612 0.616 0.646 0.730 0.660 0.646 0.568 0.636 0.722 0.652 0.672 0.660 #> 4 0.602 0.602 0.618 0.634 0.666 0.622 0.636 0.656 0.568 0.642 0.614 0.654 0.638 #> 5 0.610 0.642 0.630 0.760 0.766 0.698 0.654 0.642 0.738 0.702 0.652 0.688 0.652 #> 6 0.618 0.696 0.626 0.618 0.668 0.634 0.664 0.648 0.630 0.694 0.608 0.682 0.674 #> 27 28 29 30 31 32 33 34 35 36 37 38 39 #> 1 0.702 0.684 0.674 0.634 0.680 0.724 0.692 0.684 0.744 0.666 0.694 0.716 0.700 #> 2 0.632 0.586 0.568 0.582 0.646 0.606 0.618 0.634 0.722 0.638 0.590 0.590 0.616 #> 3 0.668 0.660 0.654 0.676 0.660 0.674 0.694 0.700 0.764 0.684 0.628 0.684 0.708 #> 4 0.644 0.596 0.612 0.608 0.662 0.594 0.610 0.632 0.730 0.648 0.562 0.588 0.636 #> 5 0.712 0.752 0.728 0.734 0.676 0.710 0.698 0.768 0.830 0.678 0.690 0.752 0.760 #> 6 0.628 0.640 0.630 0.668 0.668 0.636 0.662 0.682 0.768 0.678 0.658 0.664 0.700 #> 40 41 42 43 44 45 46 47 48 49 50 #> 1 0.714 0.686 0.624 0.688 0.654 0.668 0.668 0.650 0.676 0.694 0.676 #> 2 0.638 0.670 0.644 0.654 0.626 0.664 0.648 0.638 0.648 0.678 0.672 #> 3 0.716 0.702 0.634 0.672 0.668 0.646 0.720 0.730 0.712 0.698 0.704 #> 4 0.680 0.652 0.618 0.680 0.698 0.696 0.658 0.666 0.658 0.696 0.708 #> 5 0.828 0.688 0.654 0.658 0.690 0.654 0.772 0.720 0.704 0.692 0.656 #> 6 0.724 0.686 0.678 0.670 0.666 0.698 0.734 0.692 0.714 0.734 0.704 head(as.matrix(mean.avg.dist.t)) #> 1 2 3 4 5 6 7 #> 1 0.0000000 0.5073247 0.5451162 0.5511308 0.5863431 0.5693314 0.5561033 #> 2 0.5073247 0.0000000 0.5076177 0.5454488 0.5590474 0.5361407 0.5075526 #> 3 0.5451162 0.5076177 0.0000000 0.5169507 0.5442375 0.5536933 0.5580432 #> 4 0.5511308 0.5454488 0.5169507 0.0000000 0.5556915 0.6029668 0.5894129 #> 5 0.5863431 0.5590474 0.5442375 0.5556915 0.0000000 0.5730661 0.5904097 #> 6 0.5693314 0.5361407 0.5536933 0.6029668 0.5730661 0.0000000 0.5161496 #> 8 9 10 11 12 13 14 #> 1 0.5673559 0.5627535 0.5792425 0.5729159 0.6020539 0.7016723 0.5490011 #> 2 0.5366592 0.5380985 0.5623811 0.5321790 0.5515688 0.6779465 0.5572373 #> 3 0.5073755 0.5536056 0.5389882 0.5801096 0.5868026 0.7260347 0.5816036 #> 4 0.5259962 0.5706750 0.5613498 0.5523447 0.5753112 0.6804780 0.5838035 #> 5 0.5484180 0.5855322 0.5881431 0.6281267 0.6276212 0.7377994 0.5875698 #> 6 0.5529963 0.5552535 0.6215621 0.5618841 0.5336650 0.6716372 0.5641578 #> 15 16 17 18 19 20 21 #> 1 0.6050878 0.5605091 0.6262520 0.6790944 0.6158975 0.5894641 0.6238942 #> 2 0.5492945 0.5571279 0.5683865 0.6719160 0.5864794 0.5817164 0.5785775 #> 3 0.5651808 0.5907658 0.5986803 0.7180656 0.6208023 0.5831755 0.6141896 #> 4 0.5846365 0.5995058 0.6019378 0.6723126 0.6185437 0.5788246 0.6204008 #> 5 0.5886468 0.6313116 0.6908350 0.7310610 0.6738026 0.6056946 0.6248085 #> 6 0.6245763 0.5734551 0.5897524 0.6553604 0.5945132 0.6253476 0.6236726 #> 22 23 24 25 26 27 28 #> 1 0.6087932 0.6864876 0.5674543 0.6124061 0.6000094 0.6287013 0.6246696 #> 2 0.5851407 0.6506247 0.5554757 0.5519726 0.6143138 0.5810440 0.5887911 #> 3 0.6301892 0.7101043 0.5608707 0.5964616 0.6536689 0.6015372 0.6254962 #> 4 0.6116276 0.6570546 0.5806733 0.5942875 0.6756654 0.6170196 0.6088559 #> 5 0.6692645 0.7281868 0.6019467 0.6240673 0.6717579 0.6824310 0.6781349 #> 6 0.5779917 0.6752783 0.5474248 0.6048853 0.6297599 0.6032439 0.6025756 #> 29 30 31 32 33 34 35 #> 1 0.5877126 0.5931557 0.6122546 0.6435203 0.6114740 0.6243358 0.7012874 #> 2 0.5493477 0.5852872 0.5988185 0.6153332 0.5921568 0.6244034 0.6905708 #> 3 0.5678376 0.6210073 0.6158757 0.6377790 0.6371270 0.6555867 0.7323038 #> 4 0.5759931 0.5912841 0.5983615 0.6066556 0.6016788 0.6087876 0.6950808 #> 5 0.6276522 0.6379576 0.6034350 0.6457655 0.6563073 0.6866576 0.7235042 #> 6 0.5816252 0.6242943 0.6152092 0.6137642 0.6116231 0.6246531 0.7312678 #> 36 37 38 39 40 41 42 #> 1 0.5974168 0.6467898 0.6716840 0.6643855 0.6683070 0.6353745 0.6004874 #> 2 0.5903695 0.5752574 0.6187077 0.6192344 0.6573943 0.6571387 0.6027516 #> 3 0.5687077 0.6126409 0.6617885 0.6959650 0.6909293 0.6525653 0.6014759 #> 4 0.6194344 0.5761296 0.6102164 0.6295049 0.6560977 0.5989171 0.6281164 #> 5 0.6056532 0.6441953 0.6593381 0.7110348 0.7252049 0.6568578 0.6156351 #> 6 0.6109414 0.6067715 0.6656711 0.6870593 0.6686455 0.6769139 0.6426746 #> 43 44 45 46 47 48 49 #> 1 0.6310309 0.5900500 0.5774603 0.6474425 0.6488217 0.6157111 0.6070289 #> 2 0.6028820 0.6126007 0.6208170 0.6652427 0.6134982 0.6243098 0.6319083 #> 3 0.6162330 0.6398741 0.6039060 0.7021972 0.6677691 0.6678864 0.6696487 #> 4 0.6197236 0.6379939 0.6156272 0.6582549 0.6062615 0.6139683 0.6466872 #> 5 0.6203780 0.6542059 0.6040162 0.7417980 0.6946591 0.6954255 0.6537119 #> 6 0.6386277 0.6418793 0.6306854 0.6948586 0.6326248 0.6678749 0.6692019 #> 50 #> 1 0.5867459 #> 2 0.6227903 #> 3 0.6365131 #> 4 0.6260672 #> 5 0.6250031 #> 6 0.6423431 head(as.matrix(median.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 #> 1 0.00 0.56 0.60 0.62 0.66 0.63 0.58 0.62 0.65 0.61 0.59 0.65 0.74 0.64 0.64 #> 2 0.56 0.00 0.53 0.59 0.64 0.63 0.60 0.58 0.58 0.60 0.59 0.62 0.68 0.58 0.61 #> 3 0.60 0.53 0.00 0.61 0.63 0.58 0.56 0.57 0.60 0.55 0.60 0.64 0.76 0.66 0.61 #> 4 0.62 0.59 0.61 0.00 0.64 0.66 0.62 0.58 0.57 0.58 0.63 0.62 0.66 0.55 0.57 #> 5 0.66 0.64 0.63 0.64 0.00 0.57 0.65 0.64 0.62 0.62 0.68 0.67 0.76 0.66 0.65 #> 6 0.63 0.63 0.58 0.66 0.57 0.00 0.58 0.62 0.61 0.66 0.60 0.59 0.74 0.69 0.70 #> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 #> 1 0.63 0.65 0.76 0.67 0.65 0.65 0.66 0.68 0.66 0.66 0.62 0.66 0.65 0.66 0.66 #> 2 0.62 0.59 0.69 0.61 0.56 0.64 0.62 0.67 0.60 0.58 0.62 0.64 0.63 0.56 0.63 #> 3 0.61 0.64 0.72 0.69 0.64 0.62 0.62 0.74 0.65 0.65 0.66 0.64 0.61 0.60 0.64 #> 4 0.63 0.62 0.67 0.61 0.60 0.68 0.59 0.67 0.62 0.60 0.66 0.60 0.62 0.59 0.62 #> 5 0.65 0.73 0.81 0.70 0.65 0.64 0.72 0.69 0.71 0.67 0.71 0.68 0.72 0.70 0.67 #> 6 0.63 0.60 0.68 0.63 0.66 0.68 0.63 0.70 0.61 0.64 0.66 0.63 0.65 0.62 0.69 #> 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 #> 1 0.65 0.69 0.66 0.66 0.74 0.65 0.68 0.69 0.70 0.68 0.67 0.63 0.63 0.64 0.67 #> 2 0.62 0.61 0.61 0.63 0.75 0.66 0.60 0.60 0.67 0.68 0.68 0.62 0.64 0.64 0.69 #> 3 0.63 0.61 0.69 0.68 0.78 0.67 0.61 0.65 0.68 0.72 0.71 0.65 0.66 0.64 0.65 #> 4 0.59 0.65 0.61 0.66 0.79 0.61 0.57 0.61 0.65 0.66 0.68 0.63 0.66 0.68 0.71 #> 5 0.65 0.71 0.70 0.75 0.82 0.70 0.65 0.73 0.79 0.76 0.72 0.64 0.73 0.70 0.68 #> 6 0.68 0.68 0.67 0.68 0.82 0.72 0.71 0.69 0.71 0.75 0.74 0.69 0.67 0.69 0.72 #> 46 47 48 49 50 #> 1 0.69 0.66 0.66 0.66 0.67 #> 2 0.69 0.62 0.67 0.63 0.65 #> 3 0.73 0.70 0.70 0.68 0.72 #> 4 0.68 0.65 0.65 0.69 0.64 #> 5 0.76 0.70 0.75 0.72 0.70 #> 6 0.72 0.71 0.73 0.72 0.71 # Run example to illustrate low variance of mean, median, and stdev results # Mean and median std dev are around 0.05 sdd <- avgdist(BCI, sample = 50, iterations = 100, meanfun = sd) summary(mean.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4640 0.6180 0.6540 0.6543 0.6900 0.8360 summary(median.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4600 0.6100 0.6500 0.6501 0.6900 0.8600 summary(sdd) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.03849 0.05617 0.05921 0.05919 0.06208 0.07661 # Test for when subsampling depth excludes some samples # Return samples that are removed for not meeting depth filter depth.avg.dist <- avgdist(BCI, sample = 450, iterations = 10) #> Warning: The following sampling units were removed because they were below sampling depth: 1, 2, 6, 7, 8, 9, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 31, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 # Print the result depth.avg.dist #> 3 4 5 10 15 30 32 #> 3 0.0000000 0.3282222 0.3635556 0.2942222 0.3562222 0.4555556 0.4726667 #> 4 0.3282222 0.0000000 0.3775556 0.3173333 0.3480000 0.3908889 0.4273333 #> 5 0.3635556 0.3775556 0.0000000 0.3900000 0.3995556 0.4968889 0.5231111 #> 10 0.2942222 0.3173333 0.3900000 0.0000000 0.3135556 0.4266667 0.4200000 #> 15 0.3562222 0.3480000 0.3995556 0.3135556 0.0000000 0.4535556 0.4657778 #> 30 0.4555556 0.3908889 0.4968889 0.4266667 0.4535556 0.0000000 0.3811111 #> 32 0.4726667 0.4273333 0.5231111 0.4200000 0.4657778 0.3811111 0.0000000 #> 35 0.6591111 0.6266667 0.6966667 0.6817778 0.6573333 0.5157778 0.5893333 #> 40 0.5624444 0.5286667 0.6413333 0.5240000 0.5628889 0.4540000 0.4057778 #> 35 40 #> 3 0.6591111 0.5624444 #> 4 0.6266667 0.5286667 #> 5 0.6966667 0.6413333 #> 10 0.6817778 0.5240000 #> 15 0.6573333 0.5628889 #> 30 0.5157778 0.4540000 #> 32 0.5893333 0.4057778 #> 35 0.0000000 0.4280000 #> 40 0.4280000 0.0000000"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":null,"dir":"Reference","previous_headings":"","what":"Beals Smoothing and Degree of Absence — beals","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing replaces entry community data probability target species occurring particular site, based joint occurrences target species species actually occur site. Swan's (1970) degree absence applies Beals smoothing zero items long zeros replaced smoothed values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"beals(x, species = NA, reference = x, type = 0, include = TRUE) swan(x, maxit = Inf, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Beals Smoothing and Degree of Absence — beals","text":"x Community data frame matrix. species Column index used compute Beals function single species. default (NA) indicates function computed species. reference Community data frame matrix used compute joint occurrences. default, x used reference compute joint occurrences. type Numeric. Specifies abundance values used function beals. See details explanation. include logical flag indicates whether target species included computing mean conditioned probabilities. original Beals (1984) definition equivalent include=TRUE, formulation Münzbergová Herben equal include=FALSE. maxit Maximum number iterations. default Inf means iterations continued zeros number zeros change. Probably maxit = 1 makes sense addition default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing estimated probability \\(p_{ij}\\) species \\(j\\) occurs site \\(\\). defined \\(p_{ij} = \\frac{1}{S_i} \\sum_k \\frac{N_{jk} I_{ik}}{N_k}\\), \\(S_i\\) number species site \\(\\), \\(N_{jk}\\) number joint occurrences species \\(j\\) \\(k\\), \\(N_k\\) number occurrences species \\(k\\), \\(\\) incidence (0 1) species (last term usually omitted equation, necessary). \\(N_{jk}\\) can interpreted mean conditional probability, beals function can interpreted mean conditioned probabilities (De Cáceres & Legendre 2008). present function generalized abundance values (De Cáceres & Legendre 2008). type argument specifies abundance values used. type = 0 presence/absence mode. type = 1 abundances reference (x) used compute conditioned probabilities. type = 2 abundances x used compute weighted averages conditioned probabilities. type = 3 abundances used compute conditioned probabilities weighted averages. Beals smoothing originally suggested method data transformation remove excessive zeros (Beals 1984, McCune 1994). However, suitable method purpose since maintain information species presences: species may higher probability occurrence site occur sites occurs. Moreover, regularizes data strongly. method may useful identifying species belong species pool (Ewald 2002) identify suitable unoccupied patches metapopulation analysis (Münzbergová & Herben 2004). case, function called include=FALSE cross-validation smoothing species; argument species can used one species studied. Swan (1970) suggested replacing zero values degrees absence species community data matrix. Swan expressed method terms similarity matrix, equivalent applying Beals smoothing zero values, step shifting smallest initially non-zero item value one, repeating many times zeros left data. actually similar extended dissimilarities (implemented function stepacross), rarely used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Beals Smoothing and Degree of Absence — beals","text":"function returns transformed data matrix vector Beals smoothing requested single species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals, E.W. 1984. Bray-Curtis ordination: effective strategy analysis multivariate ecological data. Pp. 1--55 : MacFadyen, . & E.D. Ford [eds.] Advances Ecological Research, 14. Academic Press, London. De Cáceres, M. & Legendre, P. 2008. Beals smoothing revisited. Oecologia 156: 657--669. Ewald, J. 2002. probabilistic approach estimating species pools large compositional matrices. J. Veg. Sci. 13: 191--198. McCune, B. 1994. Improving community ordination Beals smoothing function. Ecoscience 1: 82--86. Münzbergová, Z. & Herben, T. 2004. Identification suitable unoccupied habitats metapopulation studies using co-occurrence species. Oikos 105: 408--414. Swan, J.M.. 1970. examination ordination problems use simulated vegetational data. Ecology 51: 89--102.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Beals Smoothing and Degree of Absence — beals","text":"Miquel De Cáceres Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"data(dune) ## Default x <- beals(dune) ## Remove target species x <- beals(dune, include = FALSE) ## Smoothed values against presence or absence of species pa <- decostand(dune, \"pa\") boxplot(as.vector(x) ~ unlist(pa), xlab=\"Presence\", ylab=\"Beals\") ## Remove the bias of tarbet species: Yields lower values. beals(dune, type =3, include = FALSE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord #> 1 0.49590853 0.38333415 0.01157407 0.4923280 0.30827883 0.4935662 0.43263047 #> 2 0.47083676 0.39501120 0.03361524 0.4718807 0.34723984 0.4917791 0.42000984 #> 3 0.34063019 0.52738394 0.01520046 0.5309152 0.21609954 0.4033301 0.33010938 #> 4 0.30816435 0.51198853 0.02876960 0.5971801 0.21542662 0.4398775 0.35732610 #> 5 0.59949785 0.27622698 0.06632771 0.3349203 0.48876285 0.4322142 0.44309579 #> 6 0.58819821 0.26299306 0.05967771 0.2700508 0.53154426 0.3696613 0.39760652 #> 7 0.56496165 0.29412293 0.05329633 0.3403047 0.48010987 0.4051777 0.40471531 #> 8 0.21230502 0.66906674 0.02588333 0.5187956 0.16247716 0.2720122 0.21219877 #> 9 0.30323659 0.59744543 0.02213662 0.5792855 0.21896113 0.3292320 0.28613526 #> 10 0.54083871 0.26902092 0.07349127 0.3372958 0.42671693 0.4705094 0.42934344 #> 11 0.40509331 0.31656550 0.10259239 0.3185489 0.38766111 0.3713794 0.31413659 #> 12 0.21008725 0.66278454 0.03625297 0.5753377 0.20078932 0.2802946 0.22974415 #> 13 0.21850759 0.68239707 0.02191119 0.6404427 0.16737280 0.2939740 0.24942466 #> 14 0.13570397 0.76284476 0.02298398 0.4107645 0.12128973 0.1682755 0.13757552 #> 15 0.09168815 0.79412733 0.02538032 0.4505613 0.10117099 0.1420251 0.09794548 #> 16 0.06335463 0.87877202 0.00742115 0.5232448 0.05538377 0.1516354 0.09458531 #> 17 0.55254140 0.07330247 0.29233391 0.1013889 0.69331132 0.3129358 0.34982363 #> 18 0.37751017 0.34451209 0.08535723 0.2838834 0.36918166 0.3676424 0.30478244 #> 19 0.29826049 0.25952255 0.35137675 0.1934048 0.51929869 0.2237843 0.18074796 #> 20 0.05429986 0.76675441 0.06144615 0.4063662 0.10738280 0.1450721 0.06706410 #> Chenalbu Cirsarve Comapalu Eleopalu Elymrepe Empenigr #> 1 0.025132275 0.09504980 0.000000000 0.05592045 0.4667439 0.00000000 #> 2 0.043866562 0.08570299 0.026548839 0.08656209 0.4407282 0.01829337 #> 3 0.065338638 0.08967477 0.031898812 0.16099072 0.4137888 0.01074444 #> 4 0.057970906 0.12920228 0.039859621 0.16112450 0.4399661 0.02527165 #> 5 0.026434737 0.05520104 0.015892090 0.05419613 0.3575948 0.03029752 #> 6 0.021256367 0.03223112 0.030347896 0.08784329 0.3138879 0.03093489 #> 7 0.038467708 0.04706743 0.017083997 0.06694311 0.3586644 0.02304603 #> 8 0.063278453 0.06688407 0.100703044 0.29777644 0.3046956 0.02102222 #> 9 0.069879277 0.07647268 0.045830682 0.19018562 0.3523460 0.01838883 #> 10 0.025686639 0.06037513 0.029746617 0.07787078 0.3736128 0.03425596 #> 11 0.021234732 0.05778318 0.035740922 0.11146095 0.2884798 0.07310076 #> 12 0.103543341 0.07799259 0.045375827 0.19518888 0.3354080 0.03413656 #> 13 0.122547745 0.07905124 0.056084315 0.22437598 0.3511708 0.01840390 #> 14 0.042990591 0.03618335 0.241811837 0.55982776 0.1428372 0.01989756 #> 15 0.035609053 0.04022968 0.198176675 0.53973883 0.1462975 0.02215971 #> 16 0.056246994 0.05184498 0.201352298 0.51523810 0.1832397 0.00742115 #> 17 0.007716049 0.01049383 0.009876543 0.02777778 0.1929470 0.21968254 #> 18 0.014640428 0.04454602 0.042890320 0.17341352 0.2651538 0.06763669 #> 19 0.019591245 0.03668466 0.031845637 0.12592768 0.1422725 0.26011417 #> 20 0.037623741 0.03453783 0.185726965 0.58476297 0.1168700 0.05905666 #> Hyporadi Juncarti Juncbufo Lolipere Planlanc Poaprat Poatriv #> 1 0.07702746 0.14794933 0.1987270 0.9226190 0.40103107 0.9863946 0.8826329 #> 2 0.07454127 0.13017869 0.2070478 0.8272395 0.40700777 0.8972046 0.8288385 #> 3 0.05562332 0.22291082 0.2544828 0.7205525 0.27933493 0.8083020 0.8383185 #> 4 0.06985986 0.21320122 0.2318440 0.7197924 0.25797285 0.7940926 0.8197302 #> 5 0.10245961 0.10406655 0.2164230 0.8380779 0.52628928 0.9035899 0.8094632 #> 6 0.11463153 0.11631772 0.2166255 0.8000021 0.58765018 0.8666677 0.7782619 #> 7 0.10837376 0.11293676 0.2110045 0.8053380 0.51905808 0.8925059 0.8018775 #> 8 0.06550319 0.33219882 0.2323566 0.5403355 0.20596764 0.6160461 0.7101299 #> 9 0.05343787 0.23134366 0.2675624 0.6874068 0.25274756 0.7523318 0.8247374 #> 10 0.13692492 0.09080902 0.1678040 0.8102783 0.52588347 0.8915882 0.7543592 #> 11 0.18108995 0.13478872 0.1656396 0.7180948 0.47012501 0.8062720 0.6404351 #> 12 0.06777311 0.27306206 0.3231724 0.5875943 0.22110550 0.6932541 0.8199960 #> 13 0.04250245 0.28204736 0.3339728 0.5714581 0.18153869 0.7063028 0.7993754 #> 14 0.04665747 0.46685537 0.1206518 0.3356311 0.14342002 0.3817081 0.5090703 #> 15 0.05040404 0.51561767 0.1370235 0.3689922 0.13523214 0.4078219 0.5263520 #> 16 0.01731602 0.54304667 0.1776781 0.3561752 0.07269979 0.4124222 0.6071083 #> 17 0.36492870 0.03333333 0.1038156 0.5858415 0.59641331 0.7434618 0.5036834 #> 18 0.17491099 0.18956922 0.1376386 0.7124388 0.45087176 0.7368632 0.5859071 #> 19 0.39145281 0.13543701 0.1127832 0.4289185 0.40784415 0.5548077 0.3605827 #> 20 0.07795311 0.53056145 0.1192488 0.3262685 0.13059496 0.3662817 0.4523029 #> Ranuflam Rumeacet Sagiproc Salirepe Scorautu Trifprat Trifrepe #> 1 0.08105273 0.3160963 0.3371121 0.02729885 0.8898317 0.21701279 0.8782576 #> 2 0.13042865 0.3031318 0.3302063 0.05983781 0.9349640 0.20673650 0.9125666 #> 3 0.22632936 0.2909068 0.4204104 0.06065155 0.9036443 0.14654749 0.8817430 #> 4 0.21909541 0.2610006 0.4191908 0.07579199 0.9204237 0.12896524 0.8943213 #> 5 0.08063087 0.3979230 0.2612828 0.07589611 0.9576838 0.34808957 0.9142360 #> 6 0.10909966 0.4330705 0.2539380 0.08921540 0.9590466 0.35423465 0.9110822 #> 7 0.10541081 0.4113622 0.2954682 0.07094548 0.9550487 0.32489503 0.9171688 #> 8 0.40134447 0.2331043 0.4009544 0.11569906 0.8755515 0.09897600 0.8002526 #> 9 0.26006489 0.3464870 0.4531178 0.07351827 0.9145996 0.16269563 0.8714833 #> 10 0.10355742 0.3226025 0.2732735 0.09037489 0.9568824 0.26807372 0.9003730 #> 11 0.13269569 0.2753878 0.3673397 0.16465286 0.9442707 0.19982976 0.8979262 #> 12 0.29873222 0.3507140 0.5122033 0.08041977 0.9377963 0.13849854 0.9079979 #> 13 0.33309468 0.3107471 0.5131337 0.06572594 0.9255312 0.11199114 0.8841739 #> 14 0.64674225 0.1241545 0.2528665 0.15917563 0.8477706 0.06176123 0.6485949 #> 15 0.64449081 0.1459458 0.3151199 0.17750323 0.8430677 0.05831084 0.7170446 #> 16 0.66893881 0.1508409 0.3480368 0.15783292 0.8131968 0.03916718 0.6776273 #> 17 0.03549383 0.2913631 0.3292030 0.25651777 0.9839744 0.27593101 0.8141660 #> 18 0.18805395 0.2668100 0.3154533 0.17191937 0.9554011 0.20461193 0.8600701 #> 19 0.14551892 0.1831168 0.4798245 0.36429493 0.9902041 0.11966159 0.8147968 #> 20 0.62796060 0.1098600 0.3105433 0.21674174 0.8457313 0.03708580 0.6350394 #> Vicilath Bracruta Callcusp #> 1 0.17244420 0.7476589 0.003527337 #> 2 0.18494940 0.7415172 0.034597921 #> 3 0.12833142 0.7666969 0.075789630 #> 4 0.12550967 0.7919786 0.081110164 #> 5 0.16693075 0.8079786 0.023129027 #> 6 0.18035860 0.8387650 0.040981168 #> 7 0.19027523 0.8089116 0.024070054 #> 8 0.10213052 0.8109194 0.201958942 #> 9 0.08630413 0.7972178 0.092982775 #> 10 0.23383453 0.7660374 0.033527777 #> 11 0.24317802 0.8182692 0.043950322 #> 12 0.08049055 0.8061715 0.100954127 #> 13 0.06604026 0.7465509 0.122856392 #> 14 0.07857237 0.7238162 0.347514804 #> 15 0.07370069 0.7997141 0.381379395 #> 16 0.03353260 0.8029953 0.364128496 #> 17 0.20728700 0.7635487 0.009876543 #> 18 0.26222869 0.8397471 0.109959916 #> 19 0.18188455 0.8161275 0.082361157 #> 20 0.09111967 0.8397124 0.397924041 ## Uses abundance information. ## Vector with beals smoothing values corresponding to the first species ## in dune. beals(dune, species=1, include=TRUE) #> 1 2 3 4 5 6 7 8 #> 0.5923077 0.5032372 0.3499038 0.3306953 0.5944041 0.5928780 0.5824352 0.2082532 #> 9 10 11 12 13 14 15 16 #> 0.2960799 0.5462492 0.3659392 0.2610043 0.1982372 0.0922619 0.1140625 0.1066506 #> 17 18 19 20 #> 0.6020408 0.3844577 0.2865741 0.0750000"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Multivariate homogeneity of groups dispersions (variances) — betadisper","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Implements Marti Anderson's PERMDISP2 procedure analysis multivariate homogeneity group dispersions (variances). betadisper multivariate analogue Levene's test homogeneity variances. Non-euclidean distances objects group centres (centroids medians) handled reducing original distances principal coordinates. procedure latterly used means assessing beta diversity. anova, scores, plot boxplot methods. TukeyHSD.betadisper creates set confidence intervals differences mean distance--centroid levels grouping factor specified family-wise probability coverage. intervals based Studentized range statistic, Tukey's 'Honest Significant Difference' method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"betadisper(d, group, type = c(\"median\",\"centroid\"), bias.adjust = FALSE, sqrt.dist = FALSE, add = FALSE) # S3 method for betadisper anova(object, ...) # S3 method for betadisper scores(x, display = c(\"sites\", \"centroids\"), choices = c(1,2), ...) # S3 method for betadisper eigenvals(x, ...) # S3 method for betadisper plot(x, axes = c(1,2), cex = 0.7, pch = seq_len(ng), col = NULL, lty = \"solid\", lwd = 1, hull = TRUE, ellipse = FALSE, conf, segments = TRUE, seg.col = \"grey\", seg.lty = lty, seg.lwd = lwd, label = TRUE, label.cex = 1, ylab, xlab, main, sub, ...) # S3 method for betadisper boxplot(x, ylab = \"Distance to centroid\", ...) # S3 method for betadisper TukeyHSD(x, which = \"group\", ordered = FALSE, conf.level = 0.95, ...) # S3 method for betadisper print(x, digits = max(3, getOption(\"digits\") - 3), neigen = 8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"d distance structure returned dist, betadiver vegdist. group vector describing group structure, usually factor object can coerced factor using .factor. Can consist factor single level (.e., one group). type type analysis perform. Use spatial median group centroid? spatial median now default. bias.adjust logical: adjust small sample bias beta diversity estimates? sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. display character; partial match access scores \"sites\" \"species\". object, x object class \"betadisper\", result call betadisper. choices, axes principal coordinate axes wanted. hull logical; convex hull group plotted? ellipse logical; standard deviation data ellipse group plotted? conf Expected fractions data coverage data ellipses, e.g. 0.95. default draw 1 standard deviation data ellipse, supplied, conf multiplied corresponding value found Chi-squared distribution 2df provide requested coverage (probability contour). pch plot symbols groups, vector length equal number groups. col colors plot symbols centroid labels groups, vector length equal number groups. lty, lwd linetype, linewidth convex hulls confidence ellipses. segments logical; segments joining points centroid drawn? seg.col colour draw segments points centroid. Can vector, case one colour per group. seg.lty, seg.lwd linetype line width segments. label logical; centroids labelled respective factor label? label.cex numeric; character expansion centroid labels. cex, ylab, xlab, main, sub graphical parameters. details, see plot.default. character vector listing terms fitted model intervals calculated. Defaults grouping factor. ordered logical; see TukeyHSD. conf.level numeric value zero one giving family-wise confidence level use. digits, neigen numeric; print method, sets number digits use (per print.default) maximum number axes display eigenvalues , repsectively. ... arguments, including graphical parameters ( plot.betadisper boxplot.betadisper), passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"One measure multivariate dispersion (variance) group samples calculate average distance group members group centroid spatial median (referred 'centroid' now unless stated otherwise) multivariate space. test dispersions (variances) one groups different, distances group members group centroid subject ANOVA. multivariate analogue Levene's test homogeneity variances distances group members group centroids Euclidean distance. However, better measures distance Euclidean distance available ecological data. can accommodated reducing distances produced using dissimilarity coefficient principal coordinates, embeds within Euclidean space. analysis proceeds calculating Euclidean distances group members group centroid basis principal coordinate axes rather original distances. Non-metric dissimilarity coefficients can produce principal coordinate axes negative Eigenvalues. correspond imaginary, non-metric part distance objects. negative Eigenvalues produced, must correct imaginary distances. distance centroid point $$z_{ij}^c = \\sqrt{\\Delta^2(u_{ij}^+, c_i^+) - \\Delta^2(u_{ij}^-, c_i^-)},$$ \\(\\Delta^2\\) squared Euclidean distance \\(u_{ij}\\), principal coordinate \\(j\\)th point \\(\\)th group, \\(c_i\\), coordinate centroid \\(\\)th group. super-scripted ‘\\(+\\)’ ‘\\(-\\)’ indicate real imaginary parts respectively. equation (3) Anderson (2006). imaginary part greater magnitude real part, taking square root negative value, resulting NaN, cases changed zero distances (warning). line behaviour Marti Anderson's PERMDISP2 programme. test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance \\(F\\). alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution \\(F\\) Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can also performed using permutest.betadisper. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper. simple wrapper TukeyHSD. user directed read help file TukeyHSD using function. particular, note statement using function unbalanced designs. results analysis can visualised using plot boxplot methods. One additional use functions assessing beta diversity (Anderson et al 2006). Function betadiver provides popular dissimilarity measures purpose. noted passing Anderson (2006) related context O'Neill (2000), estimates dispersion around central location (median centroid) calculated data biased downward. bias matters comparing diversity among treatments small, unequal numbers samples. Setting bias.adjust=TRUE using betadisper imposes \\(\\sqrt{n/(n-1)}\\) correction (Stier et al. 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"anova method returns object class \"anova\" inheriting class \"data.frame\". scores method returns list one components \"sites\" \"centroids\". plot function invisibly returns object class \"ordiplot\", plotting structure can used identify.ordiplot (identify points) functions ordiplot family. boxplot function invisibly returns list whose components documented boxplot. eigenvals.betadisper returns named vector eigenvalues. TukeyHSD.betadisper returns list. See TukeyHSD details. betadisper returns list class \"betadisper\" following components: eig numeric; eigenvalues principal coordinates analysis. vectors matrix; eigenvectors principal coordinates analysis. distances numeric; Euclidean distances principal coordinate space samples respective group centroid median. group factor; vector describing group structure centroids matrix; locations group centroids medians principal coordinates. group.distances numeric; mean distance group centroid median. call matched function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"group consists single level group, anova permutest methods appropriate used data stop error. Missing values either d group removed prior performing analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Stewart Schultz noticed permutation test type=\"centroid\" wrong type error anti-conservative. , default type changed \"median\", uses spatial median group centroid. Tests suggests permutation test type analysis gives correct error rates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62, 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. O'Neill, M.E. (2000) Weighted Least Squares Approach Levene's Test Homogeneity Variance. Australian & New Zealand Journal Statistics 42, 81-–100. Stier, .C., Geange, S.W., Hanson, K.M., & Bolker, B.M. (2013) Predator density timing arrival affect reef fish community assembly. Ecology 94, 1057--1068.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Gavin L. Simpson; bias correction Adrian Stier Ben Bolker.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F permutest(mod, pairwise = TRUE, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 99 0.07 . #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Pairwise comparisons: #> (Observed p-value below diagonal, permuted p-value above diagonal) #> grazed ungrazed #> grazed 0.08 #> ungrazed 0.04295 ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Plot the groups and distances to centroids on the ## first two PCoA axes plot(mod) ## with data ellipses instead of hulls plot(mod, ellipse = TRUE, hull = FALSE) # 1 sd data ellipse plot(mod, ellipse = TRUE, hull = FALSE, conf = 0.90) # 90% data ellipse # plot with manual colour specification my_cols <- c(\"#1b9e77\", \"#7570b3\") plot(mod, col = my_cols, pch = c(16,17), cex = 1.1) ## can also specify which axes to plot, ordering respected plot(mod, axes = c(3,1), seg.col = \"forestgreen\", seg.lty = \"dashed\") ## Draw a boxplot of the distances to centroid for each group boxplot(mod) ## `scores` and `eigenvals` also work scrs <- scores(mod) str(scrs) #> List of 2 #> $ sites : num [1:24, 1:2] 0.0946 -0.3125 -0.3511 -0.3291 -0.1926 ... #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:24] \"18\" \"15\" \"24\" \"27\" ... #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" #> $ centroids: num [1:2, 1:2] -0.1455 0.2786 0.0758 -0.2111 #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:2] \"grazed\" \"ungrazed\" #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" head(scores(mod, 1:4, display = \"sites\")) #> PCoA1 PCoA2 PCoA3 PCoA4 #> 18 0.09459373 0.15914576 0.074400844 -0.202466025 #> 15 -0.31248809 0.10032751 -0.062243360 0.110844864 #> 24 -0.35106507 -0.05954096 -0.038079447 0.095060928 #> 27 -0.32914546 -0.17019348 0.231623720 0.019110623 #> 23 -0.19259443 -0.01459250 -0.005679372 -0.209718312 #> 19 -0.06794575 -0.14501690 -0.085645653 0.002431355 # group centroids/medians scores(mod, 1:4, display = \"centroids\") #> PCoA1 PCoA2 PCoA3 PCoA4 #> grazed -0.1455200 0.07584572 -0.01366220 -0.0178990 #> ungrazed 0.2786095 -0.21114993 -0.03475586 0.0220129 # eigenvalues from the underlying principal coordinates analysis eigenvals(mod) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 #> 1.7552165 1.1334455 0.4429018 0.3698054 0.2453532 0.1960921 0.1751131 #> PCoA8 PCoA9 PCoA10 PCoA11 PCoA12 PCoA13 PCoA14 #> 0.1284467 0.0971594 0.0759601 0.0637178 0.0583225 0.0394934 0.0172699 #> PCoA15 PCoA16 PCoA17 PCoA18 PCoA19 PCoA20 PCoA21 #> 0.0051011 -0.0004131 -0.0064654 -0.0133147 -0.0253944 -0.0375105 -0.0480069 #> PCoA22 PCoA23 #> -0.0537146 -0.0741390 ## try out bias correction; compare with mod3 (mod3B <- betadisper(dis, groups, type = \"median\", bias.adjust=TRUE)) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"median\", bias.adjust #> = TRUE) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.4055 0.2893 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 anova(mod3B) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 0.06468 . #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 permutest(mod3B, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 99 0.05 * #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## should always work for a single group group <- factor(rep(\"grazed\", NROW(varespec))) (tmp <- betadisper(dis, group, type = \"median\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"median\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed #> 0.4255 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 (tmp <- betadisper(dis, group, type = \"centroid\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"centroid\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to centroid: #> grazed #> 0.4261 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## simulate missing values in 'd' and 'group' ## using spatial medians groups[c(2,20)] <- NA dis[c(2, 20)] <- NA mod2 <- betadisper(dis, groups) ## messages #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod2 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to median: #> grazed ungrazed #> 0.3984 0.3008 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod2, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 99 0.16 #> Residuals 18 0.296910 0.016495 anova(mod2) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 0.1369 #> Residuals 18 0.296910 0.016495 plot(mod2) boxplot(mod2) plot(TukeyHSD(mod2)) ## Using group centroids mod3 <- betadisper(dis, groups, type = \"centroid\") #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod3 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"centroid\") #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to centroid: #> grazed ungrazed #> 0.4001 0.3108 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod3, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 99 0.12 #> Residuals 18 0.189749 0.010542 anova(mod3) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 0.09166 . #> Residuals 18 0.189749 0.010542 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod3) boxplot(mod3) plot(TukeyHSD(mod3))"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of beta Diversity — betadiver","title":"Indices of beta Diversity — betadiver","text":"function estimates 24 indices beta diversity reviewed Koleff et al. (2003). Alternatively, finds co-occurrence frequencies triangular plots (Koleff et al. 2003).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of beta Diversity — betadiver","text":"","code":"betadiver(x, method = NA, order = FALSE, help = FALSE, ...) # S3 method for betadiver plot(x, ...) # S3 method for betadiver scores(x, triangular = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of beta Diversity — betadiver","text":"x Community data matrix, betadiver result plot scores functions. method index beta diversity defined Koleff et al. (2003), Table 1. can use either subscript \\(\\beta\\) number index. See argument help . order Order sites increasing number species. influence configuration triangular plot non-symmetric indices. help Show numbers, subscript names defining equations indices exit. triangular Return scores suitable triangular plotting proportions. FALSE, returns 3-column matrix raw counts. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of beta Diversity — betadiver","text":"commonly used index beta diversity \\(\\beta_w = S/\\alpha - 1\\), \\(S\\) total number species, \\(\\alpha\\) average number species per site (Whittaker 1960). drawback model \\(S\\) increases sample size, expectation \\(\\alpha\\) remains constant, beta diversity increases sample size. solution problem study beta diversity pairs sites (Marion et al. 2017). denote number species shared two sites \\(\\) numbers unique species (shared) \\(b\\) \\(c\\), \\(S = + b + c\\) \\(\\alpha = (2 + b + c)/2\\) \\(\\beta_w = (b+c)/(2 + b + c)\\). Sørensen dissimilarity defined vegan function vegdist argument binary = TRUE. Many indices dissimilarity indices well. Function betadiver finds indices reviewed Koleff et al. (2003). indices found function designdist, current function provides conventional shortcut. function finds indices. proper analysis must done functions betadisper, adonis2 mantel. indices directly taken Table 1 Koleff et al. (2003), can selected either index number subscript name used Koleff et al. numbers, names defining equations can seen using betadiver(help = TRUE). cases two alternative forms, one term \\(-1\\) used. several duplicate indices, number distinct alternatives much lower 24 formally provided. formulations used functions differ occasionally Koleff et al. (2003), still mathematically equivalent. method = NA, index calculated, instead object class betadiver returned. list elements , b c. Function plot can used display proportions elements triangular plot suggested Koleff et al. (2003), scores extracts triangular coordinates raw scores. Function plot returns invisibly triangular coordinates \"ordiplot\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of beta Diversity — betadiver","text":"method = NA, function returns object class \"betadisper\" elements , b, c. method specified, function returns \"dist\" object can used function analysing dissimilarities. beta diversity, particularly useful functions betadisper study betadiversity groups, adonis2 model, mantel compare beta diversities dissimilarities distances (including geographical distances). Although betadiver returns \"dist\" object, indices similarities used place dissimilarities, user error. Functions 10 (\"j\"), 11 (\"sor\") 21 (\"rlb\") similarity indices. Function sets argument \"maxdist\" similarly vegdist, using NA fixed upper limit, 0 similarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of beta Diversity — betadiver","text":"Baselga, . (2010) Partitioning turnover nestedness components beta diversity. Global Ecology Biogeography 19, 134--143. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence-absence data. Journal Animal Ecology 72, 367--382. Marion, Z.H., Fordyce, J.. Fitzpatrick, B.M. (2017) Pairwise beta diversity resolves underappreciated source confusion calculating species turnover. Ecology 98, 933--939. Whittaker, R.H. (1960) Vegetation Siskiyou mountains, Oregon California. Ecological Monographs 30, 279--338.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of beta Diversity — betadiver","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Indices of beta Diversity — betadiver","text":"indices return similarities instead dissimilarities.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of beta Diversity — betadiver","text":"","code":"## Raw data and plotting data(sipoo) m <- betadiver(sipoo) plot(m) ## The indices betadiver(help=TRUE) #> 1 \"w\" = (b+c)/(2*a+b+c) #> 2 \"-1\" = (b+c)/(2*a+b+c) #> 3 \"c\" = (b+c)/2 #> 4 \"wb\" = b+c #> 5 \"r\" = 2*b*c/((a+b+c)^2-2*b*c) #> 6 \"I\" = log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c) #> 7 \"e\" = exp(log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c))-1 #> 8 \"t\" = (b+c)/(2*a+b+c) #> 9 \"me\" = (b+c)/(2*a+b+c) #> 10 \"j\" = a/(a+b+c) #> 11 \"sor\" = 2*a/(2*a+b+c) #> 12 \"m\" = (2*a+b+c)*(b+c)/(a+b+c) #> 13 \"-2\" = pmin.int(b,c)/(pmax.int(b,c)+a) #> 14 \"co\" = (a*c+a*b+2*b*c)/(2*(a+b)*(a+c)) #> 15 \"cc\" = (b+c)/(a+b+c) #> 16 \"g\" = (b+c)/(a+b+c) #> 17 \"-3\" = pmin.int(b,c)/(a+b+c) #> 18 \"l\" = (b+c)/2 #> 19 \"19\" = 2*(b*c+1)/(a+b+c)/(a+b+c-1) #> 20 \"hk\" = (b+c)/(2*a+b+c) #> 21 \"rlb\" = a/(a+c) #> 22 \"sim\" = pmin.int(b,c)/(pmin.int(b,c)+a) #> 23 \"gl\" = 2*abs(b-c)/(2*a+b+c) #> 24 \"z\" = (log(2)-log(2*a+b+c)+log(a+b+c))/log(2) ## The basic Whittaker index d <- betadiver(sipoo, \"w\") ## This should be equal to Sorensen index (binary Bray-Curtis in ## vegan) range(d - vegdist(sipoo, binary=TRUE)) #> [1] 0 0"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":null,"dir":"Reference","previous_headings":"","what":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function computes coefficients dispersal direction geographically connected areas, defined Legendre Legendre (1984), also described Legendre Legendre (2012, section 13.3.4).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"bgdispersal(mat, PAonly = FALSE, abc = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"mat Data frame matrix containing community composition data table (species presence-absence abundance data). PAonly FALSE four types coefficients, DD1 DD4, requested; TRUE DD1 DD2 sought (see Details). abc TRUE, return tables , b c used DD1 DD2.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"signs DD coefficients indicate direction dispersal, provided asymmetry significant. positive sign indicates dispersal first (row DD tables) second region (column); negative sign indicates opposite. McNemar test asymmetry computed presence-absence data test hypothesis significant asymmetry two areas comparison. input data table, rows sites areas, columns taxa. often, taxa species, coefficients can computed genera families well. DD1 DD2 computed presence-absence data. four types coefficients computed quantitative data, converted presence-absence computation DD1 DD2. PAonly = FALSE indicates four types coefficients requested. PAonly = TRUE DD1 DD2 sought.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Function bgdispersal returns list containing following matrices: DD1 \\(DD1_{j,k} = ((b - c))/((+ b + c)^2)\\) DD2 \\(DD2_{j,k} = (2 (b - c))/((2a + b + c) (+ b + c))\\) \\(\\), \\(b\\), \\(c\\) meaning computation binary similarity coefficients. DD3 \\(DD3_{j,k} = {W(-B) / (+B-W)^2} \\) DD4 \\(DD4_{j,k} = 2W(-B) / ((+B)(+B-W))\\) W = sum(pmin(vector1, vector2)), = sum(vector1), B = sum(vector2) McNemar McNemar chi-square statistic asymmetry (Sokal Rohlf 1995): \\(2(b \\log(b) + c \\log(c) - (b+c) \\log((b+c)/2)) / q\\), \\(q = 1 + 1/(2(b+c))\\) (Williams correction continuity) prob.McNemar probabilities associated McNemar statistics, chi-square test. H0: asymmetry \\((b-c)\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Legendre, P. V. Legendre. 1984. Postglacial dispersal freshwater fishes Québec peninsula. Can. J. Fish. Aquat. Sci. 41: 1781-1802. Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Sokal, R. R. F. J. Rohlf. 1995. Biometry. principles practice statistics biological research. 3rd edn. W. H. Freeman, New York.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function uses powerful alternative McNemar test classical formula. classical formula constructed spirit Pearson's Chi-square, formula function constructed spirit Wilks Chi-square \\(G\\) statistic. Function mcnemar.test uses classical formula. new formula introduced vegan version 1.10-11, older implementations bgdispersal used classical formula.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"mat <- matrix(c(32,15,14,10,70,30,100,4,10,30,25,0,18,0,40, 0,0,20,0,0,0,0,4,0,30,20,0,0,0,0,25,74,42,1,45,89,5,16,16,20), 4, 10, byrow=TRUE) bgdispersal(mat) #> $DD1 #> [,1] [,2] [,3] [,4] #> [1,] 0.00 0.24 0.21 0.00 #> [2,] -0.24 0.00 0.08 -0.24 #> [3,] -0.21 -0.08 0.00 -0.21 #> [4,] 0.00 0.24 0.21 0.00 #> #> $DD2 #> [,1] [,2] [,3] [,4] #> [1,] 0.0000000 0.3428571 0.3230769 0.0000000 #> [2,] -0.3428571 0.0000000 0.1142857 -0.3428571 #> [3,] -0.3230769 -0.1142857 0.0000000 -0.3230769 #> [4,] 0.0000000 0.3428571 0.3230769 0.0000000 #> #> $DD3 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.1567922 0.1420408 -0.01325831 #> [2,] -0.15679216 0.0000000 0.1101196 -0.20049485 #> [3,] -0.14204082 -0.1101196 0.0000000 -0.13586560 #> [4,] 0.01325831 0.2004949 0.1358656 0.00000000 #> #> $DD4 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.2513176 0.2425087 -0.01960102 #> [2,] -0.25131757 0.0000000 0.1725441 -0.30993929 #> [3,] -0.24250871 -0.1725441 0.0000000 -0.23381521 #> [4,] 0.01960102 0.3099393 0.2338152 0.00000000 #> #> $McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 7.677938 9.0571232 0.000000 #> [2,] NA NA 0.2912555 7.677938 #> [3,] NA NA NA 9.057123 #> [4,] NA NA NA NA #> #> $prob.McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 0.005590001 0.002616734 1.000000000 #> [2,] NA NA 0.589417103 0.005590001 #> [3,] NA NA NA 0.002616734 #> [4,] NA NA NA NA #>"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":null,"dir":"Reference","previous_headings":"","what":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Function finds best subset environmental variables, Euclidean distances scaled environmental variables maximum (rank) correlation community dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# S3 method for default bioenv(comm, env, method = \"spearman\", index = \"bray\", upto = ncol(env), trace = FALSE, partial = NULL, metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), parallel = getOption(\"mc.cores\"), ...) # S3 method for formula bioenv(formula, data, ...) bioenvdist(x, which = \"best\")"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"comm Community data frame dissimilarity object square matrix can interpreted dissimilarities. env Data frame continuous environmental variables. method correlation method used cor. index dissimilarity index used community data (comm) vegdist. ignored comm dissimilarities. upto Maximum number parameters studied subsets. formula, data Model formula data. trace Trace calculations partial Dissimilarities partialled inspecting variables env. metric Metric used distances environmental distances. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. x bioenv result object. number model environmental distances evaluated, \"best\" model. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function calculates community dissimilarity matrix using vegdist. selects possible subsets environmental variables, scales variables, calculates Euclidean distances subset using dist. function finds correlation community dissimilarities environmental distances, size subsets, saves best result. \\(2^p-1\\) subsets \\(p\\) variables, exhaustive search may take , , long time (parameter upto offers partial relief). argument metric defines distances given set environmental variables. metric = \"euclidean\", variables scaled unit variance Euclidean distances calculated. metric = \"mahalanobis\", Mahalanobis distances calculated: addition scaling unit variance, matrix current set environmental variables also made orthogonal (uncorrelated). metric = \"manhattan\", variables scaled unit range Manhattan distances calculated, distances sums differences environmental variables. metric = \"gower\", Gower distances calculated using function daisy. allows also using factor variables, continuous variables results equal metric = \"manhattan\". function can called model formula LHS data matrix RHS lists environmental variables. formula interface practical selecting transforming environmental variables. argument partial can perform “partial” analysis. partializing item must dissimilarity object class dist. partial item can used correlation method, strictly correct Pearson. Function bioenvdist recalculates environmental distances used within function. default calculate distances best model, number model can given. Clarke & Ainsworth (1993) suggested method used selecting best subset environmental variables interpreting results nonmetric multidimensional scaling (NMDS). recommended parallel display NMDS community dissimilarities NMDS Euclidean distances best subset scaled environmental variables. warned use Procrustes analysis, looks like good way comparing two ordinations. Clarke & Ainsworth wrote computer program BIO-ENV giving name current function. Presumably BIO-ENV later incorporated Clarke's PRIMER software (available Windows). addition, Clarke & Ainsworth suggested novel method rank correlation available current function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function returns object class bioenv summary method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Clarke, K. R & Ainsworth, M. 1993. method linking multivariate community structure environmental variables. Marine Ecology Progress Series, 92, 205--219.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"want study ‘significance’ bioenv results, can use function mantel mantel.partial use definition correlation. However, bioenv standardizes environmental variables depending used metric, must mantel comparable results (standardized data returned item x result object). safest use bioenvdist extract environmental distances really used within bioenv. NB., bioenv selects variables maximize Mantel correlation, significance tests based priori selection variables biased.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# The method is very slow for large number of possible subsets. # Therefore only 6 variables in this example. data(varespec) data(varechem) sol <- bioenv(wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, varechem) sol #> #> Call: #> bioenv(formula = wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, data = varechem) #> #> Subset of environmental variables with best correlation to community data. #> #> Correlations: spearman #> Dissimilarities: bray #> Metric: euclidean #> #> Best model has 3 parameters (max. 6 allowed): #> P Ca Al #> with correlation 0.4004806 #> ## IGNORE_RDIFF_BEGIN summary(sol) #> size correlation #> P 1 0.2513 #> P Al 2 0.4004 #> P Ca Al 3 0.4005 #> P Ca pH Al 4 0.3619 #> log(N) P Ca pH Al 5 0.3216 #> log(N) P K Ca pH Al 6 0.2822 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA biplot — biplot.rda","title":"PCA biplot — biplot.rda","text":"Draws PCA biplot species scores indicated biplot arrows","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA biplot — biplot.rda","text":"","code":"# S3 method for rda biplot(x, choices = c(1, 2), scaling = \"species\", display = c(\"sites\", \"species\"), type, xlim, ylim, col = c(1,2), const, correlation = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA biplot — biplot.rda","text":"x rda result object. choices Axes show. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. negative scaling values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. display Scores shown. must alternatives \"species\" species scores, /\"sites\" site scores. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. Can length 2 (e.g. type = c(\"text\", \"points\")), case first element describes species scores handled, second site scores drawn. xlim, ylim x y limits (min, max) plot. col Colours used sites species (order). one colour given, used . const General scaling constant scores.rda. ... parameters plotting functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA biplot — biplot.rda","text":"Produces plot biplot results call rda. common \"species\" scores PCA drawn biplot arrows point direction increasing values variable. biplot.rda function provides wrapper plot.cca allow easy production plot. biplot.rda suitable unconstrained models. used ordination object constraints, error issued. species scores drawn using \"text\", arrows drawn origin 0.85 * species score, whilst labels drawn species score. type used \"points\", labels drawn therefore arrows drawn origin actual species score.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"PCA biplot — biplot.rda","text":"plot function returns invisibly plotting structure can used identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"PCA biplot — biplot.rda","text":"Gavin Simpson, based plot.cca Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA biplot — biplot.rda","text":"","code":"data(dune) mod <- rda(dune, scale = TRUE) biplot(mod, scaling = \"symmetric\") ## different type for species and site scores biplot(mod, scaling = \"symmetric\", type = c(\"text\", \"points\")) ## We can use ordiplot pipes in R 4.1 to build similar plots with ## flexible control if (FALSE) { if (getRversion() >= \"4.1\") { plot(mod, scaling = \"symmetric\", type=\"n\") |> text(\"sites\", cex=0.8) |> text(\"species\", arrows=TRUE, length=0.02, col=\"red\", cex=0.6) } }"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":null,"dir":"Reference","previous_headings":"","what":"K-means partitioning using a range of values of K — cascadeKM","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function wrapper kmeans function. creates several partitions forming cascade small large number groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"cascadeKM(data, inf.gr, sup.gr, iter = 100, criterion = \"calinski\", parallel = getOption(\"mc.cores\")) cIndexKM(y, x, index = \"all\") # S3 method for cascadeKM plot(x, min.g, max.g, grpmts.plot = TRUE, sortg = FALSE, gridcol = NA, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"K-means partitioning using a range of values of K — cascadeKM","text":"data data matrix. objects (samples) rows. inf.gr number groups partition smallest number groups cascade (min). sup.gr number groups partition largest \t number groups cascade (max). iter number random starting configurations value \\(K\\). criterion criterion used select best partition. default value \"calinski\", refers Calinski-Harabasz (1974) criterion. simple structure index (\"ssi\") also available. indices available package cclust. experience, two indices work best likely return maximum value near optimal number clusters \"calinski\" \"ssi\". y Object class \"kmeans\" returned clustering algorithm kmeans x Data matrix columns correspond variables rows observations, plotting object plot index available indices : \"calinski\" \"ssi\". Type \"\" obtain indices. Abbreviations names also accepted. min.g, max.g minimum maximum numbers groups displayed. grpmts.plot Show plot (TRUE FALSE). sortg Sort objects function group membership produce easily interpretable graph. See Details. original object names kept; used labels output table x, although graph. row names, sequential row numbers used keep track original order objects. gridcol colour grid lines plots. NA, default value, removes grid lines. ... parameters functions (ignored). parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function creates several partitions forming cascade small large number groups formed kmeans. work performed function cIndex based clustIndex package cclust). criteria removed version computation errors generated one object found group. default value \"calinski\", refers well-known Calinski-Harabasz (1974) criterion. available index simple structure index \"ssi\" (Dolnicar et al. 1999). case groups equal sizes, \"calinski\" generally good criterion indicate correct number groups. Users take indications literally groups equal size. Type \"\" obtain indices. indices defined : calinski: \\((SSB/(K-1))/(SSW/(n-K))\\), \\(n\\) number data points \\(K\\) number clusters. \\(SSW\\) sum squares within clusters \\(SSB\\) sum squares among clusters. index simply \\(F\\) (ANOVA) statistic. ssi: “Simple Structure Index” multiplicatively combines several elements influence interpretability partitioning solution. best partition indicated highest SSI value. simulation study, Milligan Cooper (1985) found Calinski-Harabasz criterion recovered correct number groups often. recommend criterion , groups equal sizes, maximum value \"calinski\" usually indicates correct number groups. Another available index simple structure index \"ssi\". Users take indications indices literally groups equal size explore groups corresponding values \\(K\\). Function cascadeKM plot method. Two plots produced. graph left objects abscissa number groups ordinate. groups represented colours. graph right shows values criterion (\"calinski\" \"ssi\") determining best partition. highest value criterion marked red. Points marked orange, , indicate partitions producing increase criterion value number groups increases; may represent interesting partitions. sortg=TRUE, objects reordered following procedure: (1) simple matching distance matrix computed among objects, based table K-means assignments groups, \\(K\\) = min.g \\(K\\) = max.g. (2) principal coordinate analysis (PCoA, Gower 1966) computed centred distance matrix. (3) first principal coordinate used new order objects graph. simplified algorithm used compute first principal coordinate , using iterative algorithm described Legendre & Legendre (2012). full distance matrix among objects never computed; avoids problem storing number objects large. Distance values computed needed algorithm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Function cascadeKM returns object class cascadeKM items: partition Table partitions found different numbers groups \\(K\\), \\(K\\) = inf.gr \\(K\\) = sup.gr. results Values criterion select best partition. criterion name criterion used. size number objects found group, partitions (columns). Function cIndex returns vector index values. maximum value indices supposed indicate best partition. indices work best groups equal sizes. groups equal sizes, one put much faith maximum indices, also explore groups corresponding values \\(K\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Calinski, T. J. Harabasz. 1974. dendrite method cluster analysis. Commun. Stat. 3: 1--27. Dolnicar, S., K. Grabler J. . Mazanec. 1999. tale three cities: perceptual charting analyzing destination images. Pp. 39-62 : Woodside, . et al. [eds.] Consumer psychology tourism, hospitality leisure. CAB International, New York. Gower, J. C. 1966. distance properties latent root vector methods used multivariate analysis. Biometrika 53: 325--338. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Milligan, G. W. & M. C. Cooper. 1985. examination procedures determining number clusters data set. Psychometrika 50: 159--179. Weingessel, ., Dimitriadou, . Dolnicar, S. 2002. examination indexes determining number clusters binary data sets. Psychometrika 67: 137--160.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Marie-Helene Ouellette Marie-Helene.Ouellette@UMontreal.ca, Sebastien Durand Sebastien.Durand@UMontreal.ca Pierre Legendre Pierre.Legendre@UMontreal.ca. Parallel processing Virgilio Gómez-Rubio. Edited vegan Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"# Partitioning a (10 x 10) data matrix of random numbers mat <- matrix(runif(100),10,10) res <- cascadeKM(mat, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning an autocorrelated time series vec <- sort(matrix(runif(30),30,1)) res <- cascadeKM(vec, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning a large autocorrelated time series # Note that we remove the grid lines vec <- sort(matrix(runif(1000),1000,1)) res <- cascadeKM(vec, 2, 7, iter = 10, criterion = 'calinski') toto <- plot(res, gridcol=NA)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca performs correspondence analysis, optionally constrained correspondence analysis (.k.. canonical correspondence analysis), optionally partial constrained correspondence analysis. Function rda performs redundancy analysis, optionally principal components analysis. popular ordination techniques community ecology.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"# S3 method for formula cca(formula, data, na.action = na.fail, subset = NULL, ...) # S3 method for formula rda(formula, data, scale=FALSE, na.action = na.fail, subset = NULL, ...) # S3 method for default cca(X, Y, Z, ...) # S3 method for default rda(X, Y, Z, scale=FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"formula Model formula, left hand side gives community data matrix, right hand side gives constraining variables, conditioning variables can given within special function Condition. data Data frame containing variables right hand side model formula. X Community data matrix. Y Constraining matrix, typically environmental variables. Can missing. data.frame, expanded model.matrix factors expanded contrasts (“dummy variables”). better use formula instead argument, analyses work formula used. Z Conditioning matrix, effect removed (“partialled ”) next step. Can missing. data.frame, expanded similarly constraining matrix. scale Scale species unit variance (like correlations). na.action Handling missing values constraints conditions. default (na.fail) stop missing value. Choice na.omit removes rows missing values. Choice na.exclude keeps observations gives NA results calculated. WA scores rows may found also missing values constraints. Missing values never allowed dependent community data. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data. ... arguments print plot functions (ignored functions).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Since introduction (ter Braak 1986), constrained, canonical, correspondence analysis spin-, redundancy analysis, popular ordination methods community ecology. Functions cca rda similar popular proprietary software Canoco, although implementation completely different. functions based Legendre & Legendre's (2012) algorithm: cca Chi-square transformed data matrix subjected weighted linear regression constraining variables, fitted values submitted correspondence analysis performed via singular value decomposition (svd). Function rda similar, uses ordinary, unweighted linear regression unweighted SVD. Legendre & Legendre (2012), Table 11.5 (p. 650) give skeleton RDA algorithm vegan. algorithm CCA similar, involves standardization row column weights. functions can called either matrix-like entries community data constraints, formula interface. general, formula interface preferred, allows better control model allows factor constraints. analyses ordination results possible model fitted formula (e.g., cases anova.cca, automatic model building). following sections, X, Y Z, although referred matrices, commonly data frames. matrix interface, community data matrix X must given, data matrices may omitted, corresponding stage analysis skipped. matrix Z supplied, effects removed community matrix, residual matrix submitted next stage. called partial correspondence redundancy analysis. matrix Y supplied, used constrain ordination, resulting constrained canonical correspondence analysis, redundancy analysis. Finally, residual submitted ordinary correspondence analysis (principal components analysis). matrices Z Y missing, data matrix analysed ordinary correspondence analysis ( principal components analysis). Instead separate matrices, model can defined using model formula. left hand side must community data matrix (X). right hand side defines constraining model. constraints can contain ordered unordered factors, interactions among variables functions variables. defined contrasts honoured factor variables. constraints can also matrices (data frames). formula can include special term Condition conditioning variables (“covariables”) partialled analysis. following commands equivalent: cca(X, Y, Z), cca(X ~ Y + Condition(Z)), Y Z refer constraints conditions matrices respectively. Constrained correspondence analysis indeed constrained method: CCA try display variation data, part can explained used constraints. Consequently, results strongly dependent set constraints transformations interactions among constraints. shotgun method use environmental variables constraints. However, exploratory problems better analysed unconstrained methods correspondence analysis (decorana, corresp) non-metric multidimensional scaling (metaMDS) environmental interpretation analysis (envfit, ordisurf). CCA good choice user clear strong priori hypotheses constraints interested major structure data set. CCA able correct curve artefact commonly found correspondence analysis forcing configuration linear constraints. However, curve artefact can avoided low number constraints curvilinear relation . curve can reappear even two badly chosen constraints single factor. Although formula interface makes easy include polynomial interaction terms, terms often produce curved artefacts (difficult interpret), probably avoided. According folklore, rda used ``short gradients'' rather cca. However, based research finds methods based Euclidean metric uniformly weaker based Chi-squared metric. However, standardized Euclidean distance may appropriate measures (see Hellinger standardization decostand particular). Partial CCA (pCCA; alternatively partial RDA) can used remove effect conditioning background random variables covariables CCA proper. fact, pCCA compares models cca(X ~ Z) cca(X ~ Y + Z) attributes difference effect Y cleansed effect Z. people used method extracting “components variance” CCA. However, effect variables together stronger sum separately, can increase total Chi-square partialling variation, give negative “components variance”. general, components “variance” trusted due interactions two sets variables. functions summary plot methods documented separately (see plot.cca, summary.cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca returns huge object class cca, described separately cca.object. Function rda returns object class rda inherits class cca described cca.object. scaling used rda scores described separate vignette package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"original method ter Braak, current implementation follows Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. McCune, B. (1997) Influence noisy environmental data canonical correspondence analysis. Ecology 78, 2617-2623. Palmer, M. W. (1993) Putting things even better order: advantages canonical correspondence analysis. Ecology 74,2215-2230. Ter Braak, C. J. F. (1986) Canonical Correspondence Analysis: new eigenvector technique multivariate direct gradient analysis. Ecology 67, 1167-1179.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"responsible author Jari Oksanen, code borrows heavily Dave Roberts (Montana State University, USA).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"data(varespec) data(varechem) ## Common but bad way: use all variables you happen to have in your ## environmental data matrix vare.cca <- cca(varespec, varechem) vare.cca #> Call: cca(X = varespec, Y = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Constrained 1.4415 0.6920 14 #> Unconstrained 0.6417 0.3080 9 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4389 0.2918 0.1628 0.1421 0.1180 0.0890 0.0703 0.0584 0.0311 0.0133 0.0084 #> CCA12 CCA13 CCA14 #> 0.0065 0.0062 0.0047 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 #> 0.19776 0.14193 0.10117 0.07079 0.05330 0.03330 0.01887 0.01510 0.00949 #> plot(vare.cca) ## Formula interface and a better model vare.cca <- cca(varespec ~ Al + P*(K + Baresoil), data=varechem) vare.cca #> Call: cca(formula = varespec ~ Al + P * (K + Baresoil), data = #> varechem) #> #> Inertia Proportion Rank #> Total 2.083 1.000 #> Constrained 1.046 0.502 6 #> Unconstrained 1.038 0.498 17 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 0.3756 0.2342 0.1407 0.1323 0.1068 0.0561 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.27577 0.15411 0.13536 0.11803 0.08887 0.05511 0.04919 0.03781 #> (Showing 8 of 17 unconstrained eigenvalues) #> plot(vare.cca) ## Partialling out and negative components of variance cca(varespec ~ Ca, varechem) #> Call: cca(formula = varespec ~ Ca, data = varechem) #> #> Inertia Proportion Rank #> Total 2.08320 1.00000 #> Constrained 0.15722 0.07547 1 #> Unconstrained 1.92598 0.92453 22 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.15722 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.4745 0.2939 0.2140 0.1954 0.1748 0.1171 0.1121 0.0880 #> (Showing 8 of 22 unconstrained eigenvalues) #> cca(varespec ~ Ca + Condition(pH), varechem) #> Call: cca(formula = varespec ~ Ca + Condition(pH), data = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Conditional 0.1458 0.0700 1 #> Constrained 0.1827 0.0877 1 #> Unconstrained 1.7547 0.8423 21 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.18269 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3834 0.2749 0.2123 0.1760 0.1701 0.1161 0.1089 0.0880 #> (Showing 8 of 21 unconstrained eigenvalues) #> ## RDA data(dune) data(dune.env) dune.Manure <- rda(dune ~ Manure, dune.env) plot(dune.Manure)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":null,"dir":"Reference","previous_headings":"","what":"Result Object from Constrained Ordination — cca.object","title":"Result Object from Constrained Ordination — cca.object","text":"Ordination methods cca, rda, dbrda capscale return similar result objects. methods use internal function ordConstrained. differ (1) initial transformation data defining inertia, (2) weighting, (3) use rectangular rows \\(\\times\\) columns data symmetric rows \\(\\times\\) rows dissimilarities: rda initializes data give variance correlations inertia, cca based double-standardized data give Chi-square inertia uses row column weights, capscale maps real part dissimilarities rectangular data performs RDA, dbrda performs RDA-like analysis directly symmetric dissimilarities. Function ordConstrained returns result components methods, calling function may add components final result. However, access result components directly (using $): internal structure regarded stable application interface (API), can change release. access results components directly, take risk breakage vegan release. vegan provides wide set accessor functions components, functions updated result object changes. documentation gives overview accessor functions cca result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Result Object from Constrained Ordination — cca.object","text":"","code":"ordiYbar(x, model = c(\"CCA\", \"CA\", \"pCCA\", \"partial\", \"initial\")) # S3 method for cca model.frame(formula, ...) # S3 method for cca model.matrix(object, ...) # S3 method for cca weights(object, display = \"sites\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Result Object from Constrained Ordination — cca.object","text":"object, x, formula result object cca, rda, dbrda, capscale. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. ordiYbar value can also \"initial\" internal working input data, \"partial\" internal working input data removing partial effects. display Display either \"sites\" \"species\". ... arguments passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Result Object from Constrained Ordination — cca.object","text":"internal (“working”) form dependent (community) data can accessed function ordiYbar. form depends ordination method: instance, cca data weighted Chi-square transformed, dbrda Gower-centred dissimilarities. input data original (“response”) form can accessed fitted.cca residuals.cca. Function predict.cca can return either working response data, also lower-rank approximations. model matrix independent data (“Constraints” “Conditions”) can extracted model.matrix. partial analysis, function returns list design matrices called Conditions Constraints. either component missing, single matrix returned. redundant (aliased) terms appear model matrix. terms can found alias.cca. Function model.frame tries reconstruct data frame model matrices derived. possible original model fitted formula data arguments, still fails data unavailable. number observations can accessed nobs.cca, residual degrees freedom df.residual.cca. information observations missing values can accessed na.action. terms formula fitted model can accessed formula terms. weights used cca can accessed weights. unweighted methods (rda) weights equal. ordination results saved separate components partial terms, constraints residual unconstrained ordination. guarantee components internal names currently, cautious developing scripts functions directly access components. constrained ordination algorithm based QR decomposition constraints conditions (environmental data), QR component saved separately partial constrained components. QR decomposition constraints can accessed qr.cca. also include residual effects partial terms (Conditions), used together ordiYbar(x, \"partial\"). environmental data first centred rda weighted centred cca. QR decomposition used many functions access cca results, can used find many items directly stored object. examples, see coef.cca, coef.rda, vif.cca, permutest.cca, predict.cca, predict.rda, calibrate.cca. See qr possible uses component. instance, rank constraints can found QR decomposition. eigenvalues solution can accessed eigenvals.cca. Eigenvalues evaluated partial component, available constrained residual components. ordination scores internally stored (weighted) orthonormal scores matrices. results can accessed scores.cca scores.rda functions. ordination scores scaled accessed scores functions, internal (weighted) orthonormal scores can accessed setting scaling = FALSE. Unconstrained residual component species site scores, constrained component also fitted site scores linear combination scores sites biplot scores centroids constraint variables. biplot scores correspond model.matrix, centroids calculated factor variables used. scores can selected defining axes, direct way accessing scores certain component. number dimensions can assessed eigenvals. addition, types can derived results although saved results. instance, regression scores model coefficients can accessed scores coef functions. Partial component scores. Distance-based methods (dbrda, capscale) can negative eigenvalues associated imaginary axis scores. addition, species scores initially missing dbrda accessory found analysis capscale (may misleading). Function sppscores can used add species scores replace meaningful ones.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Result Object from Constrained Ordination — cca.object","text":"latest large change result object made release 2.5-1 2016. can modernize ancient stray results modernobject <- update(ancientobject).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Result Object from Constrained Ordination — cca.object","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Result Object from Constrained Ordination — cca.object","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":null,"dir":"Reference","previous_headings":"","what":"Multinomial Species Classification Method (CLAM) — clamtest","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"CLAM statistical approach classifying generalists specialists two distinct habitats described Chazdon et al. (2011).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"clamtest(comm, groups, coverage.limit = 10, specialization = 2/3, npoints = 20, alpha = 0.05/20) # S3 method for clamtest summary(object, ...) # S3 method for clamtest plot(x, xlab, ylab, main, pch = 21:24, col.points = 1:4, col.lines = 2:4, lty = 1:3, position = \"bottomright\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"comm Community matrix, consisting counts. groups vector identifying two habitats. Must exactly two unique values levels. Habitat IDs grouping vector must match corresponding rows community matrix comm. coverage.limit Integer, sample coverage based correction applied rare species counts limit. Sample coverage calculated separately two habitats. Sample relative abundances used species higher equal coverage.limit total counts per habitat. specialization Numeric, specialization threshold value 0 1. value \\(2/3\\) represents ‘supermajority’ rule, value \\(1/2\\) represents ‘simple majority’ rule assign shared species habitat specialists. npoints Integer, number points used determine boundary lines plots. alpha Numeric, nominal significance level individual tests. default value reduces conventional limit \\(0.05\\) account overdispersion multiple testing several species simultaneously. However, firm reason exactly limit. x, object Fitted model object class \"clamtest\". xlab, ylab Labels plot axes. main Main title plot. pch, col.points Symbols colors used plotting species groups. lty, col.lines Line types colors boundary lines plot separate species groups. position Position figure legend, see legend specification details. Legend shown position = NULL. ... Additional arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"method uses multinomial model based estimated species relative abundance two habitats (, B). minimizes bias due differences sampling intensities two habitat types well bias due insufficient sampling within habitat. method permits robust statistical classification habitat specialists generalists, without excluding rare species priori (Chazdon et al. 2011). Based user-defined specialization threshold, model classifies species one four groups: (1) generalists; (2) habitat specialists; (3) habitat B specialists; (4) rare classify confidence.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"data frame (class attribute \"clamtest\"), columns: Species: species name (column names comm), Total_**: total count habitat , Total_*B*: total count habitat B, Classes: species classification, factor levels Generalist, Specialist_**, Specialist_*B*, Too_rare. ** *B* placeholders habitat names/labels found data. summary method returns descriptive statistics results. plot method returns values invisibly produces bivariate scatterplot species total abundances two habitats. Symbols boundary lines shown species groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Chazdon, R. L., Chao, ., Colwell, R. K., Lin, S.-Y., Norden, N., Letcher, S. G., Clark, D. B., Finegan, B. Arroyo J. P.(2011). novel statistical method classifying habitat generalists specialists. Ecology 92, 1332--1343.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Peter Solymos solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"code tested standalone CLAM software provided website Anne Chao (http://chao.stat.nthu.edu.tw/wordpress); minor inconsistencies found, especially finding threshold 'rare' species. inconsistencies probably due numerical differences two implementation. current R implementation uses root finding iso-lines instead iterative search. original method (Chazdon et al. 2011) two major problems: assumes error distribution multinomial. justified choice individuals freely distributed, -dispersion clustering individuals. ecological data, variance much higher multinomial assumption, therefore test statistic optimistic. original authors suggest multiple testing adjustment multiple testing based number points (npoints) used draw critical lines plot, whereas adjustment based number tests (.e., tested species). function uses numerical values original paper, automatic connection npoints alpha arguments, must work adjustment .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"data(mite) data(mite.env) sol <- with(mite.env, clamtest(mite, Shrub==\"None\", alpha=0.005)) summary(sol) #> Two Groups Species Classification Method (CLAM) #> #> Specialization threshold = 0.6666667 #> Alpha level = 0.005 #> #> Estimated sample coverage: #> FALSE TRUE #> 1.0000 0.9996 #> #> Minimum abundance for classification: #> FALSE TRUE #> 27 9 #> #> Species Proportion #> Generalist 10 0.286 #> Specialist_FALSE 14 0.400 #> Specialist_TRUE 4 0.114 #> Too_rare 7 0.200 head(sol) #> Species Total_FALSE Total_TRUE Classes #> 1 Brachy 534 77 Generalist #> 2 PHTH 89 0 Specialist_FALSE #> 3 HPAV 389 207 Generalist #> 4 RARD 85 0 Specialist_FALSE #> 5 SSTR 22 0 Too_rare #> 6 Protopl 26 0 Too_rare plot(sol)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an Object for Null Model Algorithms — commsim","title":"Create an Object for Null Model Algorithms — commsim","text":"commsim function can used feed Null Model algorithms nullmodel analysis. make.commsim function returns various predefined algorithm types (see Details). functions represent low level interface community null model infrastructure vegan intent extensibility, less emphasis direct use users.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"commsim(method, fun, binary, isSeq, mode) make.commsim(method) # S3 method for commsim print(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create an Object for Null Model Algorithms — commsim","text":"method Character, name algorithm. fun function. possible formal arguments function see Details. binary Logical, algorithm applies presence-absence count matrices. isSeq Logical, algorithm sequential (needs burnin thinning) . mode Character, storage mode community matrix, either \"integer\" \"double\". x object class commsim. ... Additional arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create an Object for Null Model Algorithms — commsim","text":"function fun must return array dim(nr, nc, n), must take following arguments: x: input matrix, n: number permuted matrices output, nr: number rows, nc: number columns, rs: vector row sums, cs: vector column sums, rf: vector row frequencies (non-zero cells), cf: vector column frequencies (non-zero cells), s: total sum x, fill: matrix fill (non-zero cells), thin: thinning value sequential algorithms, ...: additional arguments. can define null model, several null model algorithm pre-defined can called name. predefined algorithms described detail following chapters. binary null models produce matrices zeros (absences) ones (presences) also input matrix quantitative. two types quantitative data: Counts integers natural unit individuals can shuffled, abundances can real (floating point) values natural subunit shuffling. quantitative models can handle counts, able handle real values. null models sequential next matrix derived current one. makes models dependent previous models, usually must thin matrices study sequences stability: see oecosimu details instructions. See Examples structural constraints imposed algorithm defining null model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"binary-null-models","dir":"Reference","previous_headings":"","what":"Binary null models","title":"Create an Object for Null Model Algorithms — commsim","text":"binary null models preserve fill: number presences conversely number absences. classic models may also preserve column (species) frequencies (c0) row frequencies species richness site (r0) take account commonness rarity species (r1, r2). Algorithms swap, tswap, curveball, quasiswap backtracking preserve row column frequencies. Three first ones sequential two latter non-sequential produce independent matrices. Basic algorithms reviewed Wright et al. (1998). \"r00\": non-sequential algorithm binary matrices preserves number presences (fill). \"r0\": non-sequential algorithm binary matrices preserves site (row) frequencies. \"r1\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses column marginal frequencies probabilities selecting species. \"r2\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses squared column marginal frequencies probabilities selecting species. \"c0\": non-sequential algorithm binary matrices preserves species frequencies (Jonsson 2001). \"swap\": sequential algorithm binary matrices changes matrix structure, influence marginal sums (Gotelli & Entsminger 2003). inspects \\(2 \\times 2\\) submatrices long swap can done. \"tswap\": sequential algorithm binary matrices. \"swap\" algorithm, tries fixed number times performs zero many swaps one step (according thin argument call). approach suggested Miklós & Podani (2004) found ordinary swap may lead biased sequences, since columns rows easily swapped. \"curveball\": sequential method binary matrices implements ‘Curveball’ algorithm Strona et al. (2014). algorithm selects two random rows finds set unique species occur one rows. algorithm distributes set unique species rows preserving original row frequencies. Zero several species swapped one step, usually matrix perturbed strongly sequential methods. \"quasiswap\": non-sequential algorithm binary matrices implements method matrix first filled honouring row column totals, integers may larger one. method inspects random \\(2 \\times 2\\) matrices performs quasiswap . addition ordinary swaps, quasiswap can reduce numbers one ones preserving marginal totals (Miklós & Podani 2004). method non-sequential, accepts thin argument: convergence checked every thin steps. allows performing several ordinary swaps addition fill changing swaps helps reducing removing bias. \"greedyqswap\": greedy variant quasiswap. greedy step, one element \\(2 \\times 2\\) matrix taken \\(> 1\\) elements. greedy steps biased, method can thinned, first thin steps greedy. Even modest thinning (say thin = 20) removes reduces bias, thin = 100 (1% greedy steps) looks completely safe still speeds simulation. code experimental provided scrutiny, tested bias use. \"backtracking\": non-sequential algorithm binary matrices implements filling method constraints row column frequencies (Gotelli & Entsminger 2001). matrix first filled randomly, typically row column sums reached incidences filled . begins \"backtracking\", incidences removed, filling started , backtracking done many times incidences filled matrix. results may biased inspected carefully use.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-models-for-counts-with-fixed-marginal-sums","dir":"Reference","previous_headings":"","what":"Quantitative Models for Counts with Fixed Marginal Sums","title":"Create an Object for Null Model Algorithms — commsim","text":"models shuffle individuals counts keep marginal sums fixed, marginal frequencies preserved. Algorithm r2dtable uses standard R function r2dtable also used simulated \\(P\\)-values chisq.test. Algorithm quasiswap_count uses , preserves original fill. Typically means increasing numbers zero cells result zero-inflated respect r2dtable. \"r2dtable\": non-sequential algorithm count matrices. algorithm keeps matrix sum row/column sums constant. Based r2dtable. \"quasiswap_count\": non-sequential algorithm count matrices. algorithm similar Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function preserving row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping \\(2 \\times 2\\) submatrices (see \"swap_count\" algorithm details) maintain row column totals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative swap models similar binary swap, swap largest permissible value. models section maintain fill perform quantitative swap can done without changing fill. Single step swap often changes matrix little. particular, cell counts variable, high values change slowly. Checking chain stability independence even crucial binary swap, strong thinning often needed. models never used without inspecting properties current data. null models can also defined using permatswap function. \"swap_count\": sequential algorithm count matrices. algorithm find \\(2 \\times 2\\) submatrices can swapped leaving column row totals fill unchanged. algorithm finds largest value submatrix can swapped (\\(d\\)). Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. \"abuswap_r\": sequential algorithm count nonnegative real valued matrices fixed row frequencies (see also permatswap). algorithm similar swap_count, uses different swap value row \\(2 \\times 2\\) submatrix. step changes corresponding column sums, honours matrix fill, row sums, row/column frequencies (Hardy 2008; randomization scheme 2x). \"abuswap_c\": sequential algorithm count nonnegative real valued matrices fixed column frequencies (see also permatswap). algorithm similar previous one, operates columns. step changes corresponding row sums, honours matrix fill, column sums, row/column frequencies (Hardy 2008; randomization scheme 3x).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-and-shuffle-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap and Shuffle Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative Swap Shuffle methods (swsh methods) preserve fill column row frequencies, also either row column sums. methods first perform binary quasiswap shuffle original quantitative data non-zero cells. samp methods shuffle original non-zero cell values can used also non-integer data. methods redistribute individuals randomly among non-zero cells can used integer data. shuffling either free whole matrix, within rows (r methods) within columns (c methods). Shuffling within row preserves row sums, shuffling within column preserves column sums. models can also defined permatswap. \"swsh_samp\": non-sequential algorithm quantitative data (either integer counts non-integer values). Original non-zero values values shuffled. \"swsh_both\": non-sequential algorithm count data. Individuals shuffled freely non-zero cells. \"swsh_samp_r\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately row. \"swsh_samp_c\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately column. \"swsh_both_r\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values within row. \"swsh_both_c\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values column.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-shuffle-methods","dir":"Reference","previous_headings":"","what":"Quantitative Shuffle Methods","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative shuffle methods generalizations binary models r00, r0 c0. _ind methods shuffle individuals grand sum, row sum column sums preserved. methods similar r2dtable still slacker constraints marginal sums. _samp _both methods first apply corresponding binary model similar restriction marginal frequencies distribute quantitative values non-zero cells. _samp models shuffle original cell values can therefore handle also non-count real values. _both models shuffle individuals among non-zero values. shuffling whole matrix r00_, within row r0_ within column c0_ cases. \"r00_ind\": non-sequential algorithm count matrices. algorithm preserves grand sum individuals shuffled among cells matrix. \"r0_ind\": non-sequential algorithm count matrices. algorithm preserves row sums individuals shuffled among cells row matrix. \"c0_ind\": non-sequential algorithm count matrices. algorithm preserves column sums individuals shuffled among cells column matrix. \"r00_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves grand sum cells matrix shuffled. \"r0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves row sums cells within row shuffled. \"c0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves column sums constant cells within column shuffled. \"r00_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells matrix shuffled. \"r0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells row shuffled. \"c0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells column shuffled.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create an Object for Null Model Algorithms — commsim","text":"object class commsim elements corresponding arguments (method, binary, isSeq, mode, fun). input make.comsimm commsim object, returned without evaluation. case, character method argument matched predefined algorithm names. error message issued none found. method argument missing, function returns names currently available null model algorithms character vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Create an Object for Null Model Algorithms — commsim","text":"Gotelli, N.J. & Entsminger, N.J. (2001). Swap fill algorithms null model analysis: rethinking knight's tour. Oecologia 129, 281--291. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97. Strona, G., Nappo, D., Boccacci, F., Fattorini, S. & San-Miguel-Ayanz, J. (2014). fast unbiased procedure randomize ecological binary matrices fixed row column totals. Nature Communications 5:4114 doi:10.1038/ncomms5114 . Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an Object for Null Model Algorithms — commsim","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"## write the r00 algorithm f <- function(x, n, ...) array(replicate(n, sample(x)), c(dim(x), n)) (cs <- commsim(\"r00\", fun=f, binary=TRUE, isSeq=FALSE, mode=\"integer\")) #> An object of class “commsim” #> ‘r00’ method (binary, non-sequential, integer mode) #> ## retrieving the sequential swap algorithm (cs <- make.commsim(\"swap\")) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## feeding a commsim object as argument make.commsim(cs) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## making the missing c1 model using r1 as a template ## non-sequential algorithm for binary matrices ## that preserves the species (column) frequencies, ## but uses row marginal frequencies ## as probabilities of selecting sites f <- function (x, n, nr, nc, rs, cs, ...) { out <- array(0L, c(nr, nc, n)) J <- seq_len(nc) storage.mode(rs) <- \"double\" for (k in seq_len(n)) for (j in J) out[sample.int(nr, cs[j], prob = rs), j, k] <- 1L out } cs <- make.commsim(\"r1\") cs$method <- \"c1\" cs$fun <- f ## structural constraints diagfun <- function(x, y) { c(sum = sum(y) == sum(x), fill = sum(y > 0) == sum(x > 0), rowSums = all(rowSums(y) == rowSums(x)), colSums = all(colSums(y) == colSums(x)), rowFreq = all(rowSums(y > 0) == rowSums(x > 0)), colFreq = all(colSums(y > 0) == colSums(x > 0))) } evalfun <- function(meth, x, n) { m <- nullmodel(x, meth) y <- simulate(m, nsim=n) out <- rowMeans(sapply(1:dim(y)[3], function(i) diagfun(attr(y, \"data\"), y[,,i]))) z <- as.numeric(c(attr(y, \"binary\"), attr(y, \"isSeq\"), attr(y, \"mode\") == \"double\")) names(z) <- c(\"binary\", \"isSeq\", \"double\") c(z, out) } x <- matrix(rbinom(10*12, 1, 0.5)*rpois(10*12, 3), 12, 10) algos <- make.commsim() a <- t(sapply(algos, evalfun, x=x, n=10)) print(as.table(ifelse(a==1,1,0)), zero.print = \".\") #> binary isSeq double sum fill rowSums colSums rowFreq colFreq #> r00 1 . . 1 1 . . . . #> c0 1 . . 1 1 . 1 . 1 #> r0 1 . . 1 1 1 . 1 . #> r1 1 . . 1 1 1 . 1 . #> r2 1 . . 1 1 1 . 1 . #> quasiswap 1 . . 1 1 1 1 1 1 #> greedyqswap 1 . . 1 1 1 1 1 1 #> swap 1 1 . 1 1 1 1 1 1 #> tswap 1 1 . 1 1 1 1 1 1 #> curveball 1 1 . 1 1 1 1 1 1 #> backtrack 1 . . 1 1 1 1 1 1 #> r2dtable . . . 1 . 1 1 . . #> swap_count . 1 . 1 1 1 1 . . #> quasiswap_count . . . 1 1 1 1 . . #> swsh_samp . . 1 1 1 . . 1 1 #> swsh_both . . . 1 1 . . 1 1 #> swsh_samp_r . . 1 1 1 1 . 1 1 #> swsh_samp_c . . 1 1 1 . 1 1 1 #> swsh_both_r . . . 1 1 1 . 1 1 #> swsh_both_c . . . 1 1 . 1 1 1 #> abuswap_r . 1 1 1 1 1 . 1 1 #> abuswap_c . 1 1 1 1 . 1 1 1 #> r00_samp . . 1 1 1 . . . . #> c0_samp . . 1 1 1 . 1 . 1 #> r0_samp . . 1 1 1 1 . 1 . #> r00_ind . . . 1 . . . . . #> c0_ind . . . 1 . . 1 . . #> r0_ind . . . 1 . 1 . . . #> r00_both . . . 1 1 . . . . #> c0_both . . . 1 1 . 1 . 1 #> r0_both . . . 1 1 1 . 1 ."},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":null,"dir":"Reference","previous_headings":"","what":"Contribution Diversity Approach — contribdiv","title":"Contribution Diversity Approach — contribdiv","text":"contribution diversity approach based differentiation within-unit among-unit diversity using additive diversity partitioning unit distinctiveness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Contribution Diversity Approach — contribdiv","text":"","code":"contribdiv(comm, index = c(\"richness\", \"simpson\"), relative = FALSE, scaled = TRUE, drop.zero = FALSE) # S3 method for contribdiv plot(x, sub, xlab, ylab, ylim, col, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Contribution Diversity Approach — contribdiv","text":"comm community data matrix samples rows species column. index Character, diversity index calculated. relative Logical, TRUE contribution diversity values expressed signed deviation mean. See details. scaled Logical, TRUE relative contribution diversity values scaled sum gamma values (index = \"richness\") sum gamma values times number rows comm (index = \"simpson\"). See details. drop.zero Logical, empty rows dropped result? empty rows dropped, corresponding results NAs. x object class \"contribdiv\". sub, xlab, ylab, ylim, col Graphical arguments passed plot. ... arguments passed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Contribution Diversity Approach — contribdiv","text":"approach proposed Lu et al. (2007). Additive diversity partitioning (see adipart references) deals relation mean alpha total (gamma) diversity. Although alpha diversity values often vary considerably. Thus, contributions sites total diversity uneven. site specific contribution measured contribution diversity components. unit e.g. many unique species contribute higher level (gamma) diversity another unit number species, common. Distinctiveness species \\(j\\) can defined number sites occurs (\\(n_j\\)), sum relative frequencies (\\(p_j\\)). Relative frequencies computed sitewise \\(sum_j{p_ij}\\)s site \\(\\) sum \\(1\\). contribution site \\(\\) total diversity given \\(alpha_i = sum_j(1 / n_ij)\\) dealing richness \\(alpha_i = sum(p_{ij} * (1 - p_{ij}))\\) Simpson index. unit distinctiveness site \\(\\) average species distinctiveness, averaging species occur site \\(\\). species richness: \\(alpha_i = mean(n_i)\\) (paper, second equation contains typo, \\(n\\) without index). Simpson index: \\(alpha_i = mean(n_i)\\). Lu et al. (2007) gives -depth description different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Contribution Diversity Approach — contribdiv","text":"object class \"contribdiv\" inheriting data frame. Returned values alpha, beta gamma components sites (rows) community matrix. \"diff.coef\" attribute gives differentiation coefficient (see Examples).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Contribution Diversity Approach — contribdiv","text":"Lu, H. P., Wagner, H. H. Chen, X. Y. 2007. contribution diversity approach evaluate species diversity. Basic Applied Ecology, 8, 1--12.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Contribution Diversity Approach — contribdiv","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Contribution Diversity Approach — contribdiv","text":"","code":"## Artificial example given in ## Table 2 in Lu et al. 2007 x <- matrix(c( 1/3,1/3,1/3,0,0,0, 0,0,1/3,1/3,1/3,0, 0,0,0,1/3,1/3,1/3), 3, 6, byrow = TRUE, dimnames = list(LETTERS[1:3],letters[1:6])) x #> a b c d e f #> A 0.3333333 0.3333333 0.3333333 0.0000000 0.0000000 0.0000000 #> B 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 0.0000000 #> C 0.0000000 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 ## Compare results with Table 2 contribdiv(x, \"richness\") #> alpha beta gamma #> A 1 1.5 2.5 #> B 1 0.5 1.5 #> C 1 1.0 2.0 contribdiv(x, \"simpson\") #> alpha beta gamma #> A 0.6666667 0.1851852 0.8518519 #> B 0.6666667 0.1111111 0.7777778 #> C 0.6666667 0.1481481 0.8148148 ## Relative contribution (C values), compare with Table 2 (cd1 <- contribdiv(x, \"richness\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.5 0.5 #> B 0 -0.5 -0.5 #> C 0 0.0 0.0 (cd2 <- contribdiv(x, \"simpson\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.03703704 0.03703704 #> B 0 -0.03703704 -0.03703704 #> C 0 0.00000000 0.00000000 ## Differentiation coefficients attr(cd1, \"diff.coef\") # D_ST #> [1] 0.5 attr(cd2, \"diff.coef\") # D_DT #> [1] 0.1818182 ## BCI data set data(BCI) opar <- par(mfrow=c(2,2)) plot(contribdiv(BCI, \"richness\"), main = \"Absolute\") plot(contribdiv(BCI, \"richness\", relative = TRUE), main = \"Relative\") plot(contribdiv(BCI, \"simpson\")) plot(contribdiv(BCI, \"simpson\", relative = TRUE)) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] Distance-based Redundancy Analysis — dbrda","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Distance-based redundancy analysis (dbRDA) ordination method similar Redundancy Analysis (rda), allows non-Euclidean dissimilarity indices, Manhattan Bray-Curtis distance. Despite non-Euclidean feature, analysis strictly linear metric. called Euclidean distance, results identical rda, dbRDA less efficient. Functions dbrda constrained versions metric scaling, .k.. principal coordinates analysis, based Euclidean distance can used, useful, dissimilarity measures. Function capscale simplified version based Euclidean approximation dissimilarities. functions can also perform unconstrained principal coordinates analysis, optionally using extended dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"dbrda(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...) capscale(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, comm = NULL, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"formula Model formula. function can called formula interface. usual features formula hold, especially defined cca rda. LHS must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist function given dfun used find dissimilarities. RHS defines constraints. constraints can continuous variables factors, can transformed within formula, can interactions typical formula. RHS can special term Condition defines variables “partialled ” constraints, just like rda cca. allows use partial dbRDA. data Data frame containing variables right hand side model formula. distance name dissimilarity (distance) index LHS formula data frame instead dissimilarity matrix. sqrt.dist Take square roots dissimilarities. See section Details . comm Community data frame used finding species scores LHS formula dissimilarity matrix. used LHS data frame. supplied, “species scores” unavailable dissimilarities supplied. N.B., available capscale: dbrda return species scores. Function sppscores can used add species scores missing. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). \"lingoes\" (TRUE) uses recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. latter one cmdscale. dfun Distance dissimilarity function used. function returning standard \"dist\" taking index name first argument can used. metaMDSdist Use metaMDSdist similarly metaMDS. means automatic data transformation using extended flexible shortest path dissimilarities (function stepacross) many dissimilarities based shared species. na.action Handling missing values constraints conditions. default (na.fail) stop missing values. Choices na.omit na.exclude delete rows missing values, differ representation results. na.omit non-missing site scores shown, na.exclude gives NA scores missing observations. Unlike rda, WA scores available missing constraints conditions. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data (given formula comm argument). ... parameters passed underlying functions (e.g., metaMDSdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Functions dbrda capscale provide two alternative implementations dbRDA. Function dbrda based McArdle & Anderson (2001) directly decomposes dissimilarities. Euclidean distances results identical rda. Non-Euclidean dissimilarities may give negative eigenvalues associated imaginary axes. Function capscale based Legendre & Anderson (1999): dissimilarity data first ordinated using metric scaling, ordination results analysed rda. capscale ignores imaginary component give negative eigenvalues (report magnitude imaginary component). user supplied community data frame instead dissimilarities, functions find dissimilarities using vegdist distance function given dfun specified distance. functions accept distance objects vegdist, dist, method producing compatible objects. constraining variables can continuous factors , can interaction terms, can transformed call. Moreover, can special term Condition just like rda cca “partial” analysis can performed. Function dbrda return species scores, can also missing capscale, can added analysis using function sppscores. Non-Euclidean dissimilarities can produce negative eigenvalues (Legendre & Anderson 1999, McArdle & Anderson 2001). negative eigenvalues, printed output capscale add column sums positive eigenvalues item sum negative eigenvalues, dbrda add column giving number real dimensions positive eigenvalues. negative eigenvalues disturbing, functions let distort dissimilarities non-negative eigenvalues produced argument add = TRUE. Alternatively, sqrt.dist = TRUE, square roots dissimilarities can used may help avoiding negative eigenvalues (Legendre & Anderson 1999). functions can also used perform ordinary metric scaling .k.. principal coordinates analysis using formula constant right hand side, comm ~ 1. metaMDSdist = TRUE, function can automatic data standardization use extended dissimilarities using function stepacross similarly non-metric multidimensional scaling metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"functions return object class dbrda capscale inherit rda. See cca.object description result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Anderson, M.J. & Willis, T.J. (2003). Canonical analysis principal coordinates: useful method constrained ordination ecology. Ecology 84, 511--525. Gower, J.C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. Legendre, P. & Legendre, L. (2012). Numerical Ecology. 3rd English Edition. Elsevier. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290--297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Function dbrda implements real distance-based RDA preferred capscale. capscale originally developed variant constrained analysis proximities (Anderson & Willis 2003), developments made similar dbRDA. However, discards imaginary dimensions negative eigenvalues ordination significance tests area based real dimensions positive eigenvalues. capscale may removed vegan future. vegan since 2003 (CRAN release 1.6-0) dbrda first released 2016 (version 2.4-0), removal capscale may disruptive historical examples scripts, modern times dbrda used. inertia named dissimilarity index defined dissimilarity data, unknown distance information missing. largest original dissimilarity larger 4, capscale handles input similarly rda bases analysis variance instead sum squares. Keyword mean added inertia cases, e.g. Euclidean Manhattan distances. Inertia based squared index, keyword squared added name distance, unless data square root transformed (argument sqrt.dist=TRUE). additive constant used argument add, Lingoes Cailliez adjusted added name inertia, value constant printed.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"data(varespec, varechem) ## dbrda dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\") #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\") #> #> Inertia Proportion Rank RealDims #> Total 4.5444 1.0000 #> Conditional 0.9726 0.2140 1 #> Constrained 0.9731 0.2141 3 3 #> Unconstrained 2.5987 0.5718 19 13 #> Inertia is squared Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5362 0.3198 0.1171 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9054 0.5070 0.3336 0.2581 0.2027 0.1605 0.1221 0.0825 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues with sqrt distances dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\", sqrt.dist = TRUE) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\", sqrt.dist = TRUE) #> #> Inertia Proportion Rank #> Total 6.9500 1.0000 #> Conditional 0.9535 0.1372 1 #> Constrained 1.2267 0.1765 3 #> Unconstrained 4.7698 0.6863 19 #> Inertia is Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5817 0.4086 0.2365 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9680 0.6100 0.4469 0.3837 0.3371 0.3012 0.2558 0.2010 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues also with Jaccard distances (m <- dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"jaccard\")) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"jaccard\") #> #> Inertia Proportion Rank #> Total 6.5044 1.0000 #> Conditional 1.0330 0.1588 1 #> Constrained 1.2068 0.1855 3 #> Unconstrained 4.2646 0.6557 19 #> Inertia is squared Jaccard distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5992 0.3994 0.2082 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.0388 0.6441 0.4518 0.3759 0.3239 0.2785 0.2279 0.1644 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## add species scores sppscores(m) <- wisconsin(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":null,"dir":"Reference","previous_headings":"","what":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Performs detrended correspondence analysis basic reciprocal averaging orthogonal correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"decorana(veg, iweigh=0, iresc=4, ira=0, mk=26, short=0, before=NULL, after=NULL) # S3 method for decorana plot(x, choices=c(1,2), origin=TRUE, display=c(\"both\",\"sites\",\"species\",\"none\"), cex = 0.8, cols = c(1,2), type, xlim, ylim, ...) # S3 method for decorana text(x, display = c(\"sites\", \"species\"), labels, choices = 1:2, origin = TRUE, select, ...) # S3 method for decorana points(x, display = c(\"sites\", \"species\"), choices=1:2, origin = TRUE, select, ...) # S3 method for decorana scores(x, display=\"sites\", choices=1:4, origin=TRUE, tidy=FALSE, ...) downweight(veg, fraction = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"veg Community data, matrix-like object. iweigh Downweighting rare species (0: ). iresc Number rescaling cycles (0: rescaling). ira Type analysis (0: detrended, 1: basic reciprocal averaging). mk Number segments rescaling. short Shortest gradient rescaled. Hill's piecewise transformation: values transformation. Hill's piecewise transformation: values transformation -- must correspond values . x decorana result object. choices Axes shown. origin Use true origin even detrended correspondence analysis. display Display sites, species, neither. cex Plot character size. cols Colours used sites species. type Type plots, partial match \"text\", \"points\" \"none\". labels Optional text used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. xlim, ylim x y limits (min,max) plot. fraction Abundance fraction downweighting begins. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score (\"sites\", \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. ... arguments plot function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"late 1970s, correspondence analysis became method choice ordination vegetation science, since seemed better able cope non-linear species responses principal components analysis. However, even correspondence analysis can produce arc-shaped configuration single gradient. Mark Hill developed detrended correspondence analysis correct two assumed ‘faults’ correspondence analysis: curvature straight gradients packing sites ends gradient. curvature removed replacing orthogonalization axes detrending. orthogonalization successive axes made non-correlated, detrending remove systematic dependence axes. Detrending performed using smoothing window mk segments. packing sites ends gradient undone rescaling axes extraction. rescaling, axis supposed scaled ‘SD’ units, average width Gaussian species responses supposed one whole axis. innovations piecewise linear transformation species abundances downweighting rare species regarded unduly high influence ordination axes. seems detrending actually works twisting ordination space, results look non-curved two-dimensional projections (‘lolly paper effect’). result, points usually easily recognized triangular diamond shaped pattern, obviously artefact detrending. Rescaling works differently commonly presented, . decorana use, even evaluate, widths species responses. Instead, tries equalize weighted standard deviation species scores axis segments (parameter mk effect, since decorana finds segments internally). Function tolerance returns internal criterion can used assess success rescaling. plot method plots species site scores. Classical decorana scaled axes smallest site score 0 ( smallest species score negative), summary, plot scores use true origin, unless origin = FALSE. addition proper eigenvalues, function reports ‘decorana values’ detrended analysis. ‘decorana values’ values legacy code decorana returns eigenvalues. estimated iteration, describe joint effects axes detrending. ‘decorana values’ estimated rescaling show effect eigenvalues. proper eigenvalues estimated extraction axes ratio weighted sum squares site species scores even detrended rescaled solutions. eigenvalues estimated axis separately, additive, higher decorana axes can show effects already explained prior axes. ‘Additive eigenvalues’ cleansed effects prior axes, can assumed add total inertia (scaled Chi-square). proportions cumulative proportions explained can use eigenvals.decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana returns object class \"decorana\", print, summary, scores, plot, points text methods, support functions eigenvals, bstick, screeplot, predict tolerance. downweight independent function can also used methods decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Hill, M.O. Gauch, H.G. (1980). Detrended correspondence analysis: improved ordination technique. Vegetatio 42, 47--58. Oksanen, J. Minchin, P.R. (1997). Instability ordination results changes input data order: explanations remedies. Journal Vegetation Science 8, 447--454.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Mark O. Hill wrote original Fortran code, R port Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana uses central numerical engine original Fortran code (public domain), 1/3 original program. tried implement original behaviour, although great part preparatory steps written R language, may differ somewhat original code. However, well-known bugs corrected strict criteria used (Oksanen & Minchin 1997). Please note really need piecewise transformation even downweighting within decorana, since powerful extensive alternatives R, options included compliance original software. different fraction abundance needed downweighting, function downweight must applied decorana. Function downweight indeed can applied prior correspondence analysis, can used together cca, . Github package natto R implementation decorana allows easier inspection algorithm also easier development function. vegan 2.6-6 earlier summary method, nothing useful now defunct. former information can extracted scores weights.decorana.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"data(varespec) vare.dca <- decorana(varespec) vare.dca #> #> Call: #> decorana(veg = varespec) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.0832 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.5235 0.3253 0.20010 0.19176 #> Additive Eigenvalues 0.5235 0.3217 0.17919 0.11922 #> Decorana values 0.5249 0.1572 0.09669 0.06075 #> Axis lengths 2.8161 2.2054 1.54650 1.64864 #> plot(vare.dca) ### the detrending rationale: gaussresp <- function(x,u) exp(-(x-u)^2/2) x <- seq(0,6,length=15) ## The gradient u <- seq(-2,8,len=23) ## The optima pack <- outer(x,u,gaussresp) matplot(x, pack, type=\"l\", main=\"Species packing\") opar <- par(mfrow=c(2,2)) plot(scores(prcomp(pack)), asp=1, type=\"b\", main=\"PCA\") plot(scores(decorana(pack, ira=1)), asp=1, type=\"b\", main=\"CA\") plot(scores(decorana(pack)), asp=1, type=\"b\", main=\"DCA\") plot(scores(cca(pack ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") ### Let's add some noise: noisy <- (0.5 + runif(length(pack)))*pack par(mfrow=c(2,1)) matplot(x, pack, type=\"l\", main=\"Ideal model\") matplot(x, noisy, type=\"l\", main=\"Noisy model\") par(mfrow=c(2,2)) plot(scores(prcomp(noisy)), type=\"b\", main=\"PCA\", asp=1) plot(scores(decorana(noisy, ira=1)), type=\"b\", main=\"CA\", asp=1) plot(scores(decorana(noisy)), type=\"b\", main=\"DCA\", asp=1) plot(scores(cca(noisy ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":null,"dir":"Reference","previous_headings":"","what":"Standardization Methods for Community Ecology — decostand","title":"Standardization Methods for Community Ecology — decostand","text":"function provides popular (effective) standardization methods community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"decostand(x, method, MARGIN, range.global, logbase = 2, na.rm=FALSE, ...) wisconsin(x) decobackstand(x, zap = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Standardization Methods for Community Ecology — decostand","text":"x Community data, matrix-like object. decobackstand standardized data. method Standardization method. See Details available options. MARGIN Margin, default acceptable. 1 = rows, 2 = columns x. range.global Matrix range found method = \"range\". allows using ranges across subsets data. dimensions MARGIN must match x. logbase logarithm base used method = \"log\". na.rm Ignore missing values row column standardizations. zap Make near-zero values exact zeros avoid negative values exaggerated estimates species richness. ... arguments function (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Standardization Methods for Community Ecology — decostand","text":"function offers following standardization methods community data: total: divide margin total (default MARGIN = 1). max: divide margin maximum (default MARGIN = 2). frequency: divide margin total multiply number non-zero items, average non-zero entries one (Oksanen 1983; default MARGIN = 2). normalize: make margin sum squares equal one (default MARGIN = 1). range: standardize values range 0 ... 1 (default MARGIN = 2). values constant, transformed 0. rank, rrank: rank replaces abundance values increasing ranks leaving zeros unchanged, rrank similar uses relative ranks maximum 1 (default MARGIN = 1). Average ranks used tied values. standardize: scale x zero mean unit variance (default MARGIN = 2). pa: scale x presence/absence scale (0/1). chi.square: divide row sums square root column sums, adjust square root matrix total (Legendre & Gallagher 2001). used Euclidean distance, distances similar Chi-square distance used correspondence analysis. However, results cmdscale still differ, since CA weighted ordination method (default MARGIN = 1). hellinger: square root method = \"total\" (Legendre & Gallagher 2001). log: logarithmic transformation suggested Anderson et al. (2006): \\(\\log_b (x) + 1\\) \\(x > 0\\), \\(b\\) base logarithm; zeros left zeros. Higher bases give less weight quantities presences, logbase = Inf gives presence/absence scaling. Please note \\(\\log(x+1)\\). Anderson et al. (2006) suggested (strongly) modified Gower distance (implemented method = \"altGower\" vegdist), standardization can used independently distance indices. alr: Additive log ratio (\"alr\") transformation (Aitchison 1986) reduces data skewness compositionality bias. transformation assumes positive values, pseudocounts can added argument pseudocount. One rows/columns reference can given reference (name index). first row/column used default (reference = 1). Note transformation drops one row column transformed output data. alr transformation defined formally follows: $$alr = [log\\frac{x_1}{x_D}, ..., log\\frac{x_{D-1}}{x_D}]$$ denominator sample \\(x_D\\) can chosen arbitrarily. transformation often used pH chemistry measurenments. also commonly used multinomial logistic regression. Default MARGIN = 1 uses row reference. clr: centered log ratio (\"clr\") transformation proposed Aitchison (1986) used reduce data skewness compositionality bias. transformation frequent applications microbial ecology (see e.g. Gloor et al., 2017). clr transformation defined : $$clr = log\\frac{x}{g(x)} = log x - log g(x)$$ \\(x\\) single value, g(x) geometric mean \\(x\\). method can operate positive data; common way deal zeroes add pseudocount (e.g. smallest positive value data), either adding manually input data, using argument pseudocount decostand(x, method = \"clr\", pseudocount = 1). Adding pseudocount inevitably introduce bias; see rclr method one available solution. rclr: robust clr (\"rclr\") similar regular clr (see ) allows data contains zeroes. method use pseudocounts, unlike standard clr. robust clr (rclr) divides values geometric mean observed features; zero values kept zeroes, taken account. high dimensional data, geometric mean rclr approximates true geometric mean; see e.g. Martino et al. (2019) rclr transformation defined formally follows: $$rclr = log\\frac{x}{g(x > 0)}$$ \\(x\\) single value, \\(g(x > 0)\\) geometric mean sample-wide values \\(x\\) positive (> 0). Standardization, contrasted transformation, means entries transformed relative entries. methods default margin. MARGIN=1 means rows (sites normal data set) MARGIN=2 means columns (species normal data set). Command wisconsin shortcut common Wisconsin double standardization species (MARGIN=2) first standardized maxima (max) sites (MARGIN=1) site totals (tot). standardization methods give nonsense results negative data entries normally occur community data. empty sites species (constant method = \"range\"), many standardization change NaN. Function decobackstand can used transform standardized data back original. possible standardization may implemented cases possible. round-errors back-transformation exact, wise overwrite original data. zap=TRUE original zeros exact.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Standardization Methods for Community Ecology — decostand","text":"Returns standardized data frame, adds attribute \"decostand\" giving name applied standardization \"method\" attribute \"parameters\" appropriate transformation parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Standardization Methods for Community Ecology — decostand","text":"Jari Oksanen, Etienne Laliberté (method = \"log\"), Leo Lahti (alr, \"clr\" \"rclr\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Standardization Methods for Community Ecology — decostand","text":"Common transformations can made standard R functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Standardization Methods for Community Ecology — decostand","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Egozcue, J.J., Pawlowsky-Glahn, V., Mateu-Figueras, G., Barcel'o-Vidal, C. (2003) Isometric logratio transformations compositional data analysis. Mathematical Geology 35, 279--300. Gloor, G.B., Macklaim, J.M., Pawlowsky-Glahn, V. & Egozcue, J.J. (2017) Microbiome Datasets Compositional: Optional. Frontiers Microbiology 8, 2224. Legendre, P. & Gallagher, E.D. (2001) Ecologically meaningful transformations ordination species data. Oecologia 129, 271--280. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Oksanen, J. (1983) Ordination boreal heath-like vegetation principal component analysis, correspondence analysis multidimensional scaling. Vegetatio 52, 181--189.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"data(varespec) sptrans <- decostand(varespec, \"max\") apply(sptrans, 2, max) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv Descflex Betupube #> 1 1 1 1 1 1 1 1 #> Vacculig Diphcomp Dicrsp Dicrfusc Dicrpoly Hylosple Pleuschr Polypili #> 1 1 1 1 1 1 1 1 #> Polyjuni Polycomm Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 1 1 1 1 1 1 1 1 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris Cladchlo Cladbotr #> 1 1 1 1 1 1 1 1 #> Cladamau Cladsp Cetreric Cetrisla Flavniva Nepharct Stersp Peltapht #> 1 1 1 1 1 1 1 1 #> Icmaeric Cladcerv Claddefo Cladphyl #> 1 1 1 1 sptrans <- wisconsin(varespec) # CLR transformation for rows, with pseudocount varespec.clr <- decostand(varespec, \"clr\", pseudocount=1) # ALR transformation for rows, with pseudocount and reference sample varespec.alr <- decostand(varespec, \"alr\", pseudocount=1, reference=1) ## Chi-square: PCA similar but not identical to CA. ## Use wcmdscale for weighted analysis and identical results. sptrans <- decostand(varespec, \"chi.square\") plot(procrustes(rda(sptrans), cca(varespec)))"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Design your own Dissimilarities — designdist","title":"Design your own Dissimilarities — designdist","text":"Function designdist lets define dissimilarities using terms shared total quantities, number rows number columns. shared total quantities can binary, quadratic minimum terms. binary terms, shared component number shared species, totals numbers species sites. quadratic terms cross-products sums squares, minimum terms sums parallel minima row totals. Function designdist2 similar, finds dissimilarities among two data sets. Function chaodist lets define dissimilarities using terms supposed take account “unseen species” (see Chao et al., 2005 Details vegdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Design your own Dissimilarities — designdist","text":"","code":"designdist(x, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) designdist2(x, y, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) chaodist(x, method = \"1 - 2*U*V/(U+V)\", name)"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Design your own Dissimilarities — designdist","text":"x Input data. y Another input data set: dissimilarities calculated among rows x rows y. method Equation dissimilarities. can use terms J shared quantity, B totals, N number rows (sites) P number columns (species) chaodist can use terms U V. equation can also contain R functions accepts vector arguments returns vectors length. terms shared total components found. vectors x y \"quadratic\" terms J = sum(x*y), = sum(x^2), B = sum(y^2), \"minimum\" terms J = sum(pmin(x,y)), = sum(x) B = sum(y), \"binary\" terms either transforming data binary form (shared number species, number species row). abcd Use 2x2 contingency table notation binary data: \\(\\) number shared species, \\(b\\) \\(c\\) numbers species occurring one sites , \\(d\\) number species occur neither sites. alphagamma Use beta diversity notation terms alpha average alpha diversity compared sites, gamma diversity pooled sites, delta absolute value difference average alpha alpha diversities compared sites. Terms B refer alpha diversities compared sites. name name want use index. default combine method equation terms argument. maxdist Theoretical maximum dissimilarity, NA index open absolute maximum. necessary argument, used vegan functions, certain maximum, better supply value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Design your own Dissimilarities — designdist","text":"popular dissimilarity measures ecology can expressed help terms J, B, also involve matrix dimensions N P. examples can define designdist : function designdist can implement dissimilarity indices vegdist elsewhere, can also used implement many indices, amongst , described Legendre & Legendre (2012). can also used implement indices beta diversity described Koleff et al. (2003), also specific function betadiver purpose. want implement binary dissimilarities based 2x2 contingency table notation, can set abcd = TRUE. notation = J, b = -J, c = B-J, d = P--B+J. notation often used instead tangible default notation reasons opaque . alphagamma = TRUE possible use beta diversity notation terms alpha average alpha diversity gamma gamma diversity two compared sites. terms calculated alpha = (+B)/2, gamma = +B-J delta = abs(-B)/2. Terms B also available give alpha diversities individual compared sites. beta diversity terms may make sense binary terms (diversities expressed numbers species), calculated quadratic minimum terms well ( warning). Function chaodist similar designgist, uses terms U V Chao et al. (2005). terms supposed take account effects unseen species. U V scaled range \\(0 \\dots 1\\). take place B product U*V used place J designdist. Function chaodist can implement commonly used Chao et al. (2005) style dissimilarity: Function vegdist implements Jaccard-type Chao distance, documentation contains complete discussion calculation terms.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Design your own Dissimilarities — designdist","text":"designdist returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Design your own Dissimilarities — designdist","text":"Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005) new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence--absence data. J. Animal Ecol. 72, 367--382. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Design your own Dissimilarities — designdist","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Design your own Dissimilarities — designdist","text":"designdist use compiled code, based vectorized R code. designdist function can much faster vegdist, although latter uses compiled code. However, designdist skip missing values uses much memory calculations. use sum terms can numerically unstable. particularly, terms large, precision may lost. risk large number columns high, particularly large quadratic terms. precise calculations better use functions like dist vegdist robust numerical problems.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Design your own Dissimilarities — designdist","text":"","code":"data(BCI) ## Four ways of calculating the same Sørensen dissimilarity d0 <- vegdist(BCI, \"bray\", binary = TRUE) d1 <- designdist(BCI, \"(A+B-2*J)/(A+B)\") d2 <- designdist(BCI, \"(b+c)/(2*a+b+c)\", abcd = TRUE) d3 <- designdist(BCI, \"gamma/alpha - 1\", alphagamma = TRUE) ## Arrhenius dissimilarity: the value of z in the species-area model ## S = c*A^z when combining two sites of equal areas, where S is the ## number of species, A is the area, and c and z are model parameters. ## The A below is not the area (which cancels out), but number of ## species in one of the sites, as defined in designdist(). dis <- designdist(BCI, \"(log(A+B-J)-log(A+B)+log(2))/log(2)\") ## This can be used in clustering or ordination... ordiplot(cmdscale(dis)) #> species scores not available ## ... or in analysing beta diversity (without gradients) summary(dis) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2733 0.3895 0.4192 0.4213 0.4537 0.5906"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions extract statistics resemble deviance AIC result constrained correspondence analysis cca redundancy analysis rda. functions rarely needed directly, called step automatic model building. Actually, cca rda AIC functions certainly wrong.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# S3 method for cca deviance(object, ...) # S3 method for cca extractAIC(fit, scale = 0, k = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"object result constrained ordination (cca rda). fit fitted model constrained ordination. scale optional numeric specifying scale parameter model, see scale step. k numeric specifying \"weight\" equivalent degrees freedom (=:edf) part AIC formula. ... arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions find statistics resemble deviance AIC constrained ordination. Actually, constrained ordination methods log-Likelihood, means AIC deviance. Therefore use functions, use , trust . use functions, remains responsibility check adequacy result. deviance cca equal Chi-square residual data matrix fitting constraints. deviance rda defined residual sum squares. deviance function rda also used distance-based RDA dbrda. Function extractAIC mimics extractAIC.lm translating deviance AIC. little need call functions directly. However, called implicitly step function used automatic selection constraining variables. check resulting model criteria, statistics used unfounded. particular, penalty k properly defined, default k = 2 justified theoretically. continuous covariates, step function base model building magnitude eigenvalues, value k influences stopping point ( variables highest eigenvalues necessarily significant permutation tests anova.cca). also multi-class factors, value k capricious effect model building. step function pass arguments add1.cca drop1.cca, setting test = \"permutation\" provide permutation tests deletion addition can help judging validity model building.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"deviance functions return “deviance”, extractAIC returns effective degrees freedom “AIC”.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Godínez-Domínguez, E. & Freire, J. (2003) Information-theoretic approach selection spatial temporal models community organization. Marine Ecology Progress Series 253, 17--24.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions unfounded untested used directly implicitly. Moreover, usual caveats using step valid.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# The deviance of correspondence analysis equals Chi-square data(dune) data(dune.env) chisq.test(dune) #> Warning: Chi-squared approximation may be incorrect #> #> \tPearson's Chi-squared test #> #> data: dune #> X-squared = 1449, df = 551, p-value < 2.2e-16 #> deviance(cca(dune)) #> [1] 1448.956 # Stepwise selection (forward from an empty model \"dune ~ 1\") ord <- cca(dune ~ ., dune.env) step(cca(dune ~ 1, dune.env), scope = formula(ord)) #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC #> + Moisture 3 86.608 #> + Management 3 86.935 #> + A1 1 87.411 #> 87.657 #> + Manure 4 88.832 #> + Use 2 89.134 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC #> 86.608 #> + Management 3 86.813 #> + A1 1 86.992 #> + Use 2 87.259 #> + Manure 4 87.342 #> - Moisture 3 87.657 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #>"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":null,"dir":"Reference","previous_headings":"","what":"Morisita index of intraspecific aggregation — dispindmorisita","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Calculates Morisita index dispersion, standardized index values, called clumpedness uniform indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"dispindmorisita(x, unique.rm = FALSE, crit = 0.05, na.rm = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"x community data matrix, sites (samples) rows species columns. unique.rm logical, TRUE, unique species (occurring one sample) removed result. crit two-sided p-value used calculate critical Chi-squared values. na.rm logical. missing values (including NaN) omitted calculations?","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita index dispersion defined (Morisita 1959, 1962): Imor = n * (sum(xi^2) - sum(xi)) / (sum(xi)^2 - sum(xi)) \\(xi\\) count individuals sample \\(\\), \\(n\\) number samples (\\(= 1, 2, \\ldots, n\\)). \\(Imor\\) values 0 \\(n\\). uniform (hyperdispersed) patterns value falls 0 1, clumped patterns falls 1 \\(n\\). increasing sample sizes (.e. joining neighbouring quadrats), \\(Imor\\) goes \\(n\\) quadrat size approaches clump size. random patterns, \\(Imor = 1\\) counts samples follow Poisson frequency distribution. deviation random expectation (null hypothesis) can tested using critical values Chi-squared distribution \\(n-1\\) degrees freedom. Confidence intervals around 1 can calculated clumped \\(Mclu\\) uniform \\(Muni\\) indices (Hairston et al. 1971, Krebs 1999) (Chi2Lower Chi2Upper refers e.g. 0.025 0.975 quantile values Chi-squared distribution \\(n-1\\) degrees freedom, respectively, crit = 0.05): Mclu = (Chi2Lower - n + sum(xi)) / (sum(xi) - 1) Muni = (Chi2Upper - n + sum(xi)) / (sum(xi) - 1) Smith-Gill (1975) proposed scaling Morisita index [0, n] interval [-1, 1], setting -0.5 0.5 values confidence limits around random distribution rescaled value 0. rescale Morisita index, one following four equations apply calculate standardized index \\(Imst\\): () Imor >= Mclu > 1: Imst = 0.5 + 0.5 (Imor - Mclu) / (n - Mclu), (b) Mclu > Imor >= 1: Imst = 0.5 (Imor - 1) / (Mclu - 1), (c) 1 > Imor > Muni: Imst = -0.5 (Imor - 1) / (Muni - 1), (d) 1 > Muni > Imor: Imst = -0.5 + 0.5 (Imor - Muni) / Muni.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Returns data frame many rows number columns input data, four columns. Columns : imor unstandardized Morisita index, mclu clumpedness index, muni uniform index, imst standardized Morisita index, pchisq Chi-squared based probability null hypothesis random expectation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita, M. 1959. Measuring dispersion individuals analysis distributional patterns. Mem. Fac. Sci. Kyushu Univ. Ser. E 2, 215--235. Morisita, M. 1962. Id-index, measure dispersion individuals. Res. Popul. Ecol. 4, 1--7. Smith-Gill, S. J. 1975. Cytophysiological basis disruptive pigmentary patterns leopard frog, Rana pipiens. II. Wild type mutant cell specific patterns. J. Morphol. 146, 35--54. Hairston, N. G., Hill, R. Ritte, U. 1971. interpretation aggregation patterns. : Patil, G. P., Pileou, E. C. Waters, W. E. eds. Statistical Ecology 1: Spatial Patterns Statistical Distributions. Penn. State Univ. Press, University Park. Krebs, C. J. 1999. Ecological Methodology. 2nd ed. Benjamin Cummings Publishers.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"common error found several papers standardizing case (b), denominator given Muni - 1. results hiatus [0, 0.5] interval standardized index. root typo book Krebs (1999), see Errata book (Page 217, https://www.zoology.ubc.ca/~krebs/downloads/errors_2nd_printing.pdf).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"data(dune) x <- dispindmorisita(dune) x #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Chenalbu NaN Inf -Inf NaN NaN #> Cirsarve 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Empenigr 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 y <- dispindmorisita(dune, unique.rm = TRUE) y #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 dim(x) ## with unique species #> [1] 30 5 dim(y) ## unique species removed #> [1] 27 5"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":null,"dir":"Reference","previous_headings":"","what":"Dispersion-based weighting of species counts — dispweight","title":"Dispersion-based weighting of species counts — dispweight","text":"Transform abundance data downweighting species overdispersed Poisson error.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"dispweight(comm, groups, nsimul = 999, nullmodel = \"c0_ind\", plimit = 0.05) gdispweight(formula, data, plimit = 0.05) # S3 method for dispweight summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dispersion-based weighting of species counts — dispweight","text":"comm Community data matrix. groups Factor describing group structure. missing, sites regarded belonging one group. NA values allowed. nsimul Number simulations. nullmodel nullmodel used commsim within groups. default follows Clarke et al. (2006). plimit Downweight species \\(p\\)-value limit. formula, data Formula left-hand side community data frame right-hand side gives explanatory variables. explanatory variables found data frame given data parent frame. object Result object dispweight gdispweight. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dispersion-based weighting of species counts — dispweight","text":"dispersion index (\\(D\\)) calculated ratio variance expected value species. species abundances follow Poisson distribution, expected dispersion \\(E(D) = 1\\), \\(D > 1\\), species overdispersed. inverse \\(1/D\\) can used downweight species abundances. Species downweighted overdispersion judged statistically significant (Clarke et al. 2006). Function dispweight implements original procedure Clarke et al. (2006). one factor can used group sites find species means. significance overdispersion assessed freely distributing individuals species within factor levels. achieved using nullmodel \"c0_ind\" (accords Clarke et al. 2006), nullmodels can used, though may meaningful (see commsim alternatives). species absent factor level, whole level ignored calculation overdispersion, number degrees freedom can vary among species. reduced number degrees freedom used divisor overdispersion \\(D\\), species higher dispersion hence lower weights transformation. Function gdispweight generalized parametric version dispweight. function based glm quasipoisson error family. glm model can used, including several factors continuous covariates. Function gdispweight uses test statistic dispweight (Pearson Chi-square), ignore factor levels species absent, number degrees freedom equal species. Therefore transformation weights can higher dispweight. gdispweight function evaluates significance overdispersion parametrically Chi-square distribution (pchisq). Functions dispweight gdispweight transform data, add information overdispersion weights attributes result. summary can used extract print information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dispersion-based weighting of species counts — dispweight","text":"Function returns transformed data following new attributes: D Dispersion statistic. df Degrees freedom species. p \\(p\\)-value Dispersion statistic \\(D\\). weights weights applied community data. nsimul Number simulations used assess \\(p\\)-value, NA simulations performed. nullmodel name commsim null model, NA simulations performed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dispersion-based weighting of species counts — dispweight","text":"Clarke, K. R., M. G. Chapman, P. J. Somerfield, H. R. Needham. 2006. Dispersion-based weighting species counts assemblage analyses. Marine Ecology Progress Series, 320, 11–27.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dispersion-based weighting of species counts — dispweight","text":"Eduard Szöcs eduardszoesc@gmail.com wrote original dispweight, Jari Oksanen significantly modified code, provided support functions developed gdispweight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"data(mite, mite.env) ## dispweight and its summary mite.dw <- with(mite.env, dispweight(mite, Shrub, nsimul = 99)) ## IGNORE_RDIFF_BEGIN summary(mite.dw) #> Dispersion Weight Df Pr(Disp.) #> Brachy 9.6908 0.1031909 67 0.01 ** #> PHTH 3.2809 0.3047900 49 0.01 ** #> HPAV 6.5263 0.1532264 67 0.01 ** #> RARD 6.0477 0.1653525 49 0.01 ** #> SSTR 2.2619 0.4421053 49 0.01 ** #> Protopl 5.4229 0.1844031 49 0.01 ** #> MEGR 4.5354 0.2204860 67 0.01 ** #> MPRO 1.2687 0.7882353 67 0.04 * #> TVIE 2.5956 0.3852706 67 0.01 ** #> HMIN 10.0714 0.0992906 67 0.01 ** #> HMIN2 7.5674 0.1321466 49 0.01 ** #> NPRA 2.6743 0.3739344 67 0.01 ** #> TVEL 9.6295 0.1038474 49 0.01 ** #> ONOV 11.3628 0.0880064 67 0.01 ** #> SUCT 8.7372 0.1144533 67 0.01 ** #> LCIL 129.4436 0.0077254 67 0.01 ** #> Oribatl1 4.1250 0.2424248 67 0.01 ** #> Ceratoz1 1.7150 0.5830768 67 0.01 ** #> PWIL 2.2943 0.4358538 67 0.01 ** #> Galumna1 2.8777 0.3474943 49 0.01 ** #> Stgncrs2 3.8242 0.2614953 49 0.01 ** #> HRUF 1.7575 0.5690021 67 0.01 ** #> Trhypch1 14.9225 0.0670128 67 0.01 ** #> PPEL 1.3628 1.0000000 49 0.07 . #> NCOR 2.5875 0.3864771 67 0.01 ** #> SLAT 2.7857 0.3589744 49 0.01 ** #> FSET 4.8901 0.2044944 49 0.01 ** #> Lepidzts 1.6577 0.6032360 49 0.02 * #> Eupelops 1.4611 0.6844033 67 0.04 * #> Miniglmn 1.6505 0.6058733 49 0.01 ** #> LRUG 12.0658 0.0828790 67 0.01 ** #> PLAG2 3.2403 0.3086090 67 0.01 ** #> Ceratoz3 3.5125 0.2846947 67 0.01 ** #> Oppiminu 3.1680 0.3156525 67 0.01 ** #> Trimalc2 10.5927 0.0944046 67 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Based on 99 simulations on 'c0_ind' nullmodel ## IGNORE_RDIFF_END ## generalized dispersion weighting mite.dw <- gdispweight(mite ~ Shrub + WatrCont, data = mite.env) rda(mite.dw ~ Shrub + WatrCont, data = mite.env) #> Call: rda(formula = mite.dw ~ Shrub + WatrCont, data = mite.env) #> #> Inertia Proportion Rank #> Total 38.1640 1.0000 #> Constrained 9.2129 0.2414 3 #> Unconstrained 28.9511 0.7586 35 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 7.986 0.748 0.480 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 5.886 3.634 2.791 2.592 1.932 1.573 1.210 1.078 #> (Showing 8 of 35 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":null,"dir":"Reference","previous_headings":"","what":"Connectedness of Dissimilarities — distconnected","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected finds groups connected disregarding dissimilarities threshold NA. function can used find groups can ordinated together transformed stepacross. Function .shared returns logical dissimilarity object, TRUE means sites species common. minimal structure distconnected can used set missing values dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"distconnected(dis, toolong = 1, trace = TRUE) no.shared(x)"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Connectedness of Dissimilarities — distconnected","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. trace Summarize results distconnected x Community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Connectedness of Dissimilarities — distconnected","text":"Data sets disconnected sample plots groups sample plots share species sites groups sites. data sets sensibly ordinated unconstrained method subsets related . instance, correspondence analysis polarize subsets eigenvalue 1. Neither can dissimilarities transformed stepacross, path points, result contain NAs. Function distconnected find subsets dissimilarity matrices. function return grouping vector can used sub-setting data. data connected, result vector \\(1\\)s. connectedness two points can defined either threshold toolong using input dissimilarities NAs. Function .shared returns dist structure value TRUE two sites nothing common, value FALSE least one shared species. minimal structure can analysed distconnected. function can used select dissimilarities shared species indices fixed upper limit. Function distconnected uses depth-first search (Sedgewick 1990).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected returns vector observations using integers identify connected groups. data connected, values 1. Function .shared returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Connectedness of Dissimilarities — distconnected","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Connectedness of Dissimilarities — distconnected","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"## There are no disconnected data in vegan, and the following uses an ## extremely low threshold limit for connectedness. This is for ## illustration only, and not a recommended practice. data(dune) dis <- vegdist(dune) gr <- distconnected(dis, toolong=0.4) #> Connectivity of distance matrix with threshold dissimilarity 0.4 #> Data are disconnected: 6 groups #> Groups sizes #> 1 2 3 4 5 6 #> 1 11 2 4 1 1 # Make sites with no shared species as NA in Manhattan dissimilarities dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":null,"dir":"Reference","previous_headings":"","what":"Ecological Diversity Indices — diversity","title":"Ecological Diversity Indices — diversity","text":"Shannon, Simpson, Fisher diversity indices species richness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ecological Diversity Indices — diversity","text":"","code":"diversity(x, index = \"shannon\", groups, equalize.groups = FALSE, MARGIN = 1, base = exp(1)) simpson.unb(x, inverse = FALSE) fisher.alpha(x, MARGIN = 1, ...) specnumber(x, groups, MARGIN = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ecological Diversity Indices — diversity","text":"x Community data, matrix-like object vector. index Diversity index, one \"shannon\", \"simpson\" \"invsimpson\". MARGIN Margin index computed. base logarithm base used shannon. inverse Use inverse Simpson similarly diversity(x, \"invsimpson\"). groups grouping factor: given, finds diversity communities pooled groups. equalize.groups Instead observed abundances, standardize communities unit total. ... Parameters passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ecological Diversity Indices — diversity","text":"Shannon Shannon--Weaver (Shannon--Wiener) index defined \\(H' = -\\sum_i p_i \\log_{b} p_i\\), \\(p_i\\) proportional abundance species \\(\\) \\(b\\) base logarithm. popular use natural logarithms, argue base \\(b = 2\\) (makes sense, real difference). variants Simpson's index based \\(D = \\sum p_i^2\\). Choice simpson returns \\(1-D\\) invsimpson returns \\(1/D\\). simpson.unb finds unbiased Simpson indices discrete samples (Hurlbert 1971, eq. 5). less sensitive sample size basic Simpson indices. unbiased indices can calculated data integer counts. diversity function can find total (gamma) diversity pooled communities argument groups. average alpha diversity can found mean diversities groups, difference ratio estimate beta diversity (see Examples). pooling can based either observed abundancies, communities can equalized unit total pooling; see Jost (2007) discussion. Functions adipart multipart provide canned alternatives estimating alpha, beta gamma diversities hierarchical settings. fisher.alpha estimates \\(\\alpha\\) parameter Fisher's logarithmic series (see fisherfit). estimation possible genuine counts individuals. None diversity indices usable empty sampling units without species, indices can give numeric value. Filtering cases left user. Function specnumber finds number species. MARGIN = 2, finds frequencies species. groups given, finds total number species group (see example finding one kind beta diversity option). Better stories can told Simpson's index Shannon's index, still grander narratives rarefaction (Hurlbert 1971). However, indices closely related (Hill 1973), reason despise one others (graduate student, drag , obey Professor's orders). particular, exponent Shannon index linearly related inverse Simpson (Hill 1973) although former may sensitive rare species. Moreover, inverse Simpson asymptotically equal rarefied species richness sample two individuals, Fisher's \\(\\alpha\\) similar inverse Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ecological Diversity Indices — diversity","text":"vector diversity indices numbers species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ecological Diversity Indices — diversity","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12, 42--58. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ecological Diversity Indices — diversity","text":"Jari Oksanen Bob O'Hara (fisher.alpha).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ecological Diversity Indices — diversity","text":"","code":"data(BCI, BCI.env) H <- diversity(BCI) simp <- diversity(BCI, \"simpson\") invsimp <- diversity(BCI, \"inv\") ## Unbiased Simpson unbias.simp <- simpson.unb(BCI) ## Fisher alpha alpha <- fisher.alpha(BCI) ## Plot all pairs(cbind(H, simp, invsimp, unbias.simp, alpha), pch=\"+\", col=\"blue\") ## Species richness (S) and Pielou's evenness (J): S <- specnumber(BCI) ## rowSums(BCI > 0) does the same... J <- H/log(S) ## beta diversity defined as gamma/alpha - 1: ## alpha is the average no. of species in a group, and gamma is the ## total number of species in the group (alpha <- with(BCI.env, tapply(specnumber(BCI), Habitat, mean))) #> OldHigh OldLow OldSlope Swamp Young #> 85.75000 91.76923 91.58333 94.00000 90.00000 (gamma <- with(BCI.env, specnumber(BCI, Habitat))) #> OldHigh OldLow OldSlope Swamp Young #> 158 210 183 128 117 gamma/alpha - 1 #> OldHigh OldLow OldSlope Swamp Young #> 0.8425656 1.2883487 0.9981802 0.3617021 0.3000000 ## similar calculations with Shannon diversity (alpha <- with(BCI.env, tapply(diversity(BCI), Habitat, mean))) # average #> OldHigh OldLow OldSlope Swamp Young #> 3.638598 3.876413 3.887122 4.003780 3.246729 (gamma <- with(BCI.env, diversity(BCI, groups=Habitat))) # pooled #> OldHigh OldLow OldSlope Swamp Young #> 3.873186 4.284972 4.212098 4.164335 3.387536 ## additive beta diversity based on Shannon index gamma-alpha #> OldHigh OldLow OldSlope Swamp Young #> 0.2345878 0.4085595 0.3249760 0.1605548 0.1408068"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and Environment in Dutch Dune Meadows. — dune","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune meadow vegetation data, dune, cover class values 30 species 20 sites. corresponding environmental data frame dune.env following entries:","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune data frame observations 30 species 20 sites. species names abbreviated 4+4 letters (see make.cepnames). following names changed original source (Jongman et al. 1987): Leontodon autumnalis Scorzoneroides, Potentilla palustris Comarum. dune.env data frame 20 observations following 5 variables: A1: numeric vector thickness soil A1 horizon. Moisture: ordered factor levels: 1 < 2 < 4 < 5. Management: factor levels: BF (Biological farming), HF (Hobby farming), NM (Nature Conservation Management), SF (Standard Farming). Use: ordered factor land-use levels: Hayfield < Haypastu < Pasture. Manure: ordered factor levels: 0 < 1 < 2 < 3 < 4.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"Jongman, R.H.G, ter Braak, C.J.F & van Tongeren, O.F.R. (1987). Data Analysis Community Landscape Ecology. Pudoc, Wageningen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":null,"dir":"Reference","previous_headings":"","what":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"Classification table species dune data set.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"dune.taxon data frame 30 species (rows) classified five taxonomic levels (columns). dune.phylodis dist object estimated coalescence ages extracted doi:10.5061/dryad.63q27 (Zanne et al. 2014) using tools packages ape phylobase.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"families orders based APG IV (2016) vascular plants Hill et al. (2006) mosses. higher levels (superorder subclass) based Chase & Reveal (2009). Chase & Reveal (2009) treat Angiosperms mosses subclasses class Equisetopsida (land plants), brylogists traditionally used much inflated levels adjusted match Angiosperm classification.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"APG IV [Angiosperm Phylogeny Group] (2016) update Angiosperm Phylogeny Group classification orders families flowering plants: APG IV. Bot. J. Linnean Soc. 181: 1--20. Chase, M.W. & Reveal, J. L. (2009) phylogenetic classification land plants accompany APG III. Bot. J. Linnean Soc. 161: 122--127. Hill, M.O et al. (2006) annotated checklist mosses Europe Macaronesia. J. Bryology 28: 198--267. Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506: 89--92.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Eigenvalues from an Ordination Object — eigenvals","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Function extracts eigenvalues object . Many multivariate methods return objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"eigenvals(x, ...) # S3 method for cca eigenvals(x, model = c(\"all\", \"unconstrained\", \"constrained\"), constrained = NULL, ...) # S3 method for decorana eigenvals(x, kind = c(\"additive\", \"axiswise\", \"decorana\"), ...) # S3 method for eigenvals summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"x object extract eigenvalues. object eigenvals result object. model eigenvalues return objects inherit class \"cca\" . constrained Return constrained eigenvalues. Deprecated vegan 2.5-0. Use model instead. kind Kind eigenvalues returned decorana. \"additive\" eigenvalues can used reporting importances components summary. \"axiswise\" gives non-additive eigenvalues, \"decorana\" decorana values (see decorana details). ... arguments functions (usually ignored)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"generic function methods cca, wcmdscale, pcnm, prcomp, princomp, dudi (ade4), pca pco ( labdsv) result objects. default method also extracts eigenvalues result looks like eigen svd. Functions prcomp princomp contain square roots eigenvalues called standard deviations, eigenvals function returns squares. Function svd contains singular values, function eigenvals returns squares. constrained ordination methods cca, rda capscale function returns constrained unconstrained eigenvalues concatenated one vector, partial component ignored. However, argument constrained = TRUE constrained eigenvalues returned. summary eigenvals result returns eigenvalues, proportion explained cumulative proportion explained. result object can negative eigenvalues (wcmdscale, dbrda, pcnm) correspond imaginary axes Euclidean mapping non-Euclidean distances (Gower 1985). case real axes (corresponding positive eigenvalues) \"explain\" proportion >1 total variation, negative eigenvalues bring cumulative proportion 1. capscale find positive eigenvalues used finding proportions. decorana importances cumulative proportions reported kind = \"additive\", alternatives add total inertia input data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"object class \"eigenvals\", vector eigenvalues. summary method returns object class \"summary.eigenvals\", matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Gower, J. C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ev <- eigenvals(mod) ev #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 CA5 #> 0.3615566 0.1699600 0.1126167 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 #> CA6 CA7 CA8 CA9 CA10 CA11 CA12 CA13 #> 0.1002670 0.0772991 0.0536938 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 #> CA14 CA15 CA16 CA17 CA18 CA19 CA20 #> 0.0101910 0.0094701 0.0055090 0.0030529 0.0025118 0.0019485 0.0005178 summary(ev) #> Importance of components: #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 #> Eigenvalue 0.3616 0.16996 0.11262 0.3500 0.2201 0.18507 0.15512 #> Proportion Explained 0.1736 0.08159 0.05406 0.1680 0.1056 0.08884 0.07446 #> Cumulative Proportion 0.1736 0.25514 0.30920 0.4772 0.5829 0.67172 0.74618 #> CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> Eigenvalue 0.13511 0.10027 0.07730 0.05369 0.03656 0.03509 0.02823 #> Proportion Explained 0.06485 0.04813 0.03711 0.02577 0.01755 0.01684 0.01355 #> Cumulative Proportion 0.81104 0.85917 0.89627 0.92205 0.93960 0.95644 0.96999 #> CA12 CA13 CA14 CA15 CA16 CA17 #> Eigenvalue 0.017065 0.012247 0.010191 0.009470 0.005509 0.003053 #> Proportion Explained 0.008192 0.005879 0.004892 0.004546 0.002644 0.001465 #> Cumulative Proportion 0.978183 0.984062 0.988954 0.993500 0.996145 0.997610 #> CA18 CA19 CA20 #> Eigenvalue 0.002512 0.0019485 0.0005178 #> Proportion Explained 0.001206 0.0009353 0.0002486 #> Cumulative Proportion 0.998816 0.9997514 1.0000000 ## choose which eignevalues to return eigenvals(mod, model = \"unconstrained\") #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 0.1002670 0.0772991 0.0536938 #> CA9 CA10 CA11 CA12 CA13 CA14 CA15 CA16 #> 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 0.0101910 0.0094701 0.0055090 #> CA17 CA18 CA19 CA20 #> 0.0030529 0.0025118 0.0019485 0.0005178"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fits an Environmental Vector or Factor onto an Ordination — envfit","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"function fits environmental vectors factors onto ordination. projections points onto vectors maximum correlation corresponding environmental variables, factors show averages factor levels. continuous varaibles equal fitting linear trend surface (plane 2D) variable (see ordisurf); trend surface can presented showing gradient (direction steepest increase) using arrow. environmental variables dependent variables explained ordination scores, dependent variable analysed separately.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"# S3 method for default envfit(ord, env, permutations = 999, strata = NULL, choices=c(1,2), display = \"sites\", w = weights(ord, display), na.rm = FALSE, ...) # S3 method for formula envfit(formula, data, ...) # S3 method for envfit plot(x, choices = c(1,2), labels, arrow.mul, at = c(0,0), axis = FALSE, p.max = NULL, col = \"blue\", bg, add = TRUE, ...) # S3 method for envfit scores(x, display, choices, arrow.mul=1, tidy = FALSE, ...) vectorfit(X, P, permutations = 0, strata = NULL, w, ...) factorfit(X, P, permutations = 0, strata = NULL, w, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"ord ordination object structure ordination scores can extracted (including data frame matrix scores). env Data frame, matrix vector environmental variables. variables can mixed type (factors, continuous variables) data frames. X Matrix data frame ordination scores. P Data frame, matrix vector environmental variable(s). must continuous vectorfit factors characters factorfit. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Set permutations = 0 skip permutations. formula, data Model formula data. na.rm Remove points missing values ordination scores environmental variables. operation casewise: whole row data removed missing value na.rm = TRUE. x result object envfit. ordiArrowMul ordiArrowTextXY must two-column matrix ( matrix-like object) containing coordinates arrow heads two plot axes, methods extract structure envfit results. choices Axes plotted. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable scores (\"vectors\" \"factors\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Change plotting labels. argument list elements vectors factors give new plotting labels. either elements omitted, default labels used. one type elements (vectors factors), labels can given vector. default labels can displayed labels command. arrow.mul Multiplier vector lengths. arrows automatically scaled similarly plot.cca given plot add = TRUE. However, scores can used adjust arrow lengths plot function used. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. axis Plot axis showing scaling fitted arrows. p.max Maximum estimated \\(P\\) value displayed variables. must calculate \\(P\\) values setting permutations use option. col Colour plotting. bg Background colour labels. bg set, labels displayed ordilabel instead text. See Examples using semitransparent background. add Results added existing ordination plot. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. display fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). scores function either \"vectors\" \"factors\" (synonyms \"bp\" \"cn\", resp.). w Weights used fitting (concerns mainly cca decorana results nonconstant weights). ... Parameters passed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Function envfit finds vectors factor averages environmental variables. Function plot.envfit adds ordination diagram. X data.frame, envfit uses factorfit factor variables vectorfit variables. X matrix vector, envfit uses vectorfit. Alternatively, model can defined simplified model formula, left hand side must ordination result object matrix ordination scores, right hand side lists environmental variables. formula interface can used easier selection /transformation environmental variables. main effects analysed even interaction terms defined formula. ordination results extracted scores extra arguments passed scores. fitted models apply results defined extracting scores using envfit. instance, scaling constrained ordination (see scores.rda, scores.cca) must set way envfit plot ordination results (see Examples). printed output continuous variables (vectors) gives direction cosines coordinates heads unit length vectors. plot scaled correlation (square root column r2) “weak” predictors shorter arrows “strong” predictors. can see scaled relative lengths using command scores. plotted (scaled) arrows adjusted current graph using constant multiplier: keep relative r2-scaled lengths arrows tries fill current plot. can see multiplier using ordiArrowMul(result_of_envfit), set argument arrow.mul. Functions vectorfit factorfit can called directly. Function vectorfit finds directions ordination space towards environmental vectors change rapidly maximal correlations ordination configuration. Function factorfit finds averages ordination scores factor levels. Function factorfit treats ordered unordered factors similarly. permutations \\(> 0\\), significance fitted vectors factors assessed using permutation environmental variables. goodness fit statistic squared correlation coefficient (\\(r^2\\)). factors defined \\(r^2 = 1 - ss_w/ss_t\\), \\(ss_w\\) \\(ss_t\\) within-group total sums squares. See permutations additional details permutation tests Vegan. User can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. fitted vectors similar biplot arrows constrained ordination fitted LC scores (display = \"lc\") set scaling = \"species\" (see scores.cca). weighted fitting gives similar results biplot arrows class centroids cca. lengths arrows fitted vectors automatically adjusted physical size plot, arrow lengths compared across plots. similar scaling arrows, must explicitly set arrow.mul argument plot command; see ordiArrowMul ordiArrowTextXY. results can accessed scores.envfit function returns either fitted vectors scaled correlation coefficient centroids fitted environmental variables, named list .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Functions vectorfit factorfit return lists classes vectorfit factorfit print method. result object following items: arrows Arrow endpoints vectorfit. arrows scaled unit length. centroids Class centroids factorfit. r Goodness fit statistic: Squared correlation coefficient permutations Number permutations. control list control values permutations returned function . pvals Empirical P-values variable. Function envfit returns list class envfit results vectorfit envfit items. Function plot.envfit scales vectors correlation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Jari Oksanen. permutation test derives code suggested Michael Scroggie.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Fitted vectors become method choice displaying environmental variables ordination. Indeed, optimal way presenting environmental variables Constrained Correspondence Analysis cca, since linear constraints. unconstrained ordination relation external variables ordination configuration may less linear, therefore methods arrows may useful. simplest adjust plotting symbol sizes (cex, symbols) environmental variables. Fancier methods involve smoothing regression methods abound R, ordisurf provides wrapper .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"data(varespec, varechem) library(MASS) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2276827 #> Run 2 stress 0.2204286 #> Run 3 stress 0.1967393 #> Run 4 stress 0.2066032 #> Run 5 stress 0.2234314 #> Run 6 stress 0.1869637 #> Run 7 stress 0.18584 #> Run 8 stress 0.18458 #> ... Procrustes: rmse 0.04935318 max resid 0.1575045 #> Run 9 stress 0.2178486 #> Run 10 stress 0.195049 #> Run 11 stress 0.2300388 #> Run 12 stress 0.1948413 #> Run 13 stress 0.1955836 #> Run 14 stress 0.2390981 #> Run 15 stress 0.2229031 #> Run 16 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04162605 max resid 0.1518104 #> Run 17 stress 0.2390411 #> Run 18 stress 0.1825658 #> ... Procrustes: rmse 1.358437e-05 max resid 2.641802e-05 #> ... Similar to previous best #> Run 19 stress 0.2092456 #> Run 20 stress 0.1982376 #> *** Best solution repeated 1 times (fit <- envfit(ord, varechem, perm = 999)) #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05732 -0.99836 0.2536 0.046 * #> P 0.61973 0.78481 0.1938 0.093 . #> K 0.76646 0.64229 0.1809 0.121 #> Ca 0.68520 0.72835 0.4119 0.008 ** #> Mg 0.63253 0.77454 0.4270 0.004 ** #> S 0.19140 0.98151 0.1752 0.123 #> Al -0.87159 0.49023 0.5269 0.001 *** #> Fe -0.93602 0.35195 0.4451 0.002 ** #> Mn 0.79871 -0.60172 0.5231 0.002 ** #> Zn 0.61756 0.78652 0.1879 0.107 #> Mo -0.90306 0.42951 0.0609 0.539 #> Baresoil 0.92488 -0.38025 0.2508 0.041 * #> Humdepth 0.93283 -0.36032 0.5201 0.002 ** #> pH -0.64798 0.76165 0.2308 0.070 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> scores(fit, \"vectors\") #> NMDS1 NMDS2 #> N -0.02886657 -0.5027884 #> P 0.27284738 0.3455254 #> K 0.32602389 0.2732093 #> Ca 0.43973450 0.4674260 #> Mg 0.41334179 0.5061403 #> S 0.08011287 0.4108193 #> Al -0.63269483 0.3558584 #> Fe -0.62444176 0.2347947 #> Mn 0.57765881 -0.4351900 #> Zn 0.26771371 0.3409563 #> Mo -0.22293623 0.1060315 #> Baresoil 0.46316234 -0.1904201 #> Humdepth 0.67271099 -0.2598473 #> pH -0.31130599 0.3659163 plot(ord) plot(fit) plot(fit, p.max = 0.05, col = \"red\") ## Adding fitted arrows to CCA. We use \"lc\" scores, and hope ## that arrows are scaled similarly in cca and envfit plots ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, type=\"p\") fit <- envfit(ord, varechem, perm = 999, display = \"lc\") plot(fit, p.max = 0.05, col = \"red\") ## 'scaling' must be set similarly in envfit and in ordination plot plot(ord, type = \"p\", scaling = \"sites\") fit <- envfit(ord, varechem, perm = 0, display = \"lc\", scaling = \"sites\") plot(fit, col = \"red\") ## Class variables, formula interface, and displaying the ## inter-class variability with ordispider, and semitransparent ## white background for labels (semitransparent colours are not ## supported by all graphics devices) data(dune) data(dune.env) ord <- cca(dune) fit <- envfit(ord ~ Moisture + A1, dune.env, perm = 0) plot(ord, type = \"n\") with(dune.env, ordispider(ord, Moisture, col=\"skyblue\")) with(dune.env, points(ord, display = \"sites\", col = as.numeric(Moisture), pch=16)) plot(fit, cex=1.2, axis=TRUE, bg = rgb(1, 1, 1, 0.5)) ## Use shorter labels for factor centroids labels(fit) #> $vectors #> [1] \"A1\" #> #> $factors #> [1] \"Moisture1\" \"Moisture2\" \"Moisture4\" \"Moisture5\" #> plot(ord) plot(fit, labels=list(factors = paste(\"M\", c(1,2,4,5), sep = \"\")), bg = rgb(1,1,0,0.5))"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds minimum (\\(q^*\\)) evenness profile based Tsallis entropy. scale factor entropy represents specific weighting species relative frequencies leads minimum evenness community (Mendes et al. 2008).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"eventstar(x, qmax = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"x community matrix numeric vector. qmax Maximum scale parameter Tsallis entropy used finding minimum Tsallis based evenness range c(0, qmax).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds characteristic value scale parameter \\(q\\) Tsallis entropy corresponding minimum evenness (equitability) profile based Tsallis entropy. value proposed Mendes et al. (2008) \\(q^*\\). \\(q^\\ast\\) index represents scale parameter one parameter Tsallis diversity family leads greatest deviation maximum equitability given relative abundance vector community. value \\(q^\\ast\\) found identifying minimum evenness profile scaling factor \\(q\\) one-dimensional minimization. evenness profile known convex function, guaranteed underlying optimize function find unique solution range c(0, qmax). scale parameter value \\(q^\\ast\\) used find corresponding values diversity (\\(H_{q^\\ast}\\)), evenness (\\(H_{q^\\ast}(\\max)\\)), numbers equivalent (\\(D_{q^\\ast}\\)). calculation details, see tsallis Examples . Mendes et al. (2008) advocated use \\(q^\\ast\\) corresponding diversity, evenness, Hill numbers, unique value representing diversity profile, positively associated rare species community, thus potentially useful indicator certain relative abundance distributions communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"data frame columns: qstar scale parameter value \\(q\\ast\\) corresponding minimum value Tsallis based evenness profile. Estar Value evenness based normalized Tsallis entropy \\(q^\\ast\\). Hstar Value Tsallis entropy \\(q^\\ast\\). Dstar Value Tsallis entropy \\(q^\\ast\\) converted numbers equivalents (also called Hill numbers, effective number species, ‘true’ diversity; cf. Jost 2007). See tsallis calculation details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Mendes, R.S., Evangelista, L.R., Thomaz, S.M., Agostinho, .. Gomes, L.C. (2008) unified index measure ecological diversity species rarity. Ecography 31, 450--456. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439. Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Values \\(q^\\ast\\) found Mendes et al. (2008) ranged 0.56 1.12 presenting low variability, interval 0 5 safely encompass possibly expected \\(q^\\ast\\) values practice, profiling evenness changing value qmax argument advised output values near range limits found.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Eduardo Ribeiro Cunha edurcunha@gmail.com Heloisa Beatriz Antoniazi Evangelista helobeatriz@gmail.com, technical input Péter Sólymos.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"data(BCI) (x <- eventstar(BCI[1:5,])) #> qstar Estar Hstar Dstar #> 1 0.6146389 0.4263687 10.524584 67.03551 #> 2 0.6249249 0.4080263 9.534034 57.66840 #> 3 0.6380858 0.4062032 9.225458 57.69174 #> 4 0.6245808 0.4062213 10.140189 65.50247 #> 5 0.6404825 0.4219957 9.828138 66.96440 ## profiling y <- as.numeric(BCI[10,]) (z <- eventstar(y)) #> qstar Estar Hstar Dstar #> 1 0.6372529 0.4117557 9.546332 61.77715 q <- seq(0, 2, 0.05) Eprof <- tsallis(y, scales=q, norm=TRUE) Hprof <- tsallis(y, scales=q) Dprof <- tsallis(y, scales=q, hill=TRUE) opar <- par(mfrow=c(3,1)) plot(q, Eprof, type=\"l\", main=\"Evenness\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, norm=TRUE), col=2) plot(q, Hprof, type=\"l\", main=\"Diversity\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar), col=2) plot(q, Dprof, type=\"l\", main=\"Effective number of species\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, hill=TRUE), col=2) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Function fisherfit fits Fisher's logseries abundance data. Function prestonfit groups species frequencies doubling octave classes fits Preston's lognormal model, function prestondistr fits truncated lognormal model without pooling data octaves.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"fisherfit(x, ...) prestonfit(x, tiesplit = TRUE, ...) prestondistr(x, truncate = -1, ...) # S3 method for prestonfit plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", line.col = \"red\", lwd = 2, ...) # S3 method for prestonfit lines(x, line.col = \"red\", lwd = 2, ...) veiledspec(x, ...) as.fisher(x, ...) # S3 method for fisher plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", kind = c(\"bar\", \"hiplot\", \"points\", \"lines\"), add = FALSE, ...) as.preston(x, tiesplit = TRUE, ...) # S3 method for preston plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", ...) # S3 method for preston lines(x, xadjust = 0.5, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"x Community data vector fitting functions result object plot functions. tiesplit Split frequencies \\(1, 2, 4, 8\\) etc adjacent octaves. truncate Truncation point log-Normal model, log2 units. Default value \\(-1\\) corresponds left border zero Octave. choice strongly influences fitting results. xlab, ylab Labels x y axes. bar.col Colour data bars. line.col Colour fitted line. lwd Width fitted line. kind Kind plot drawn: \"bar\" similar bar plot plot.fisherfit, \"hiplot\" draws vertical lines plot(..., type=\"h\"), \"points\" \"lines\" obvious. add Add existing plot. xadjust Adjustment horizontal positions octaves. ... parameters passed functions. Ignored prestonfit tiesplit passed .preston prestondistr.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher's logarithmic series expected number species \\(f\\) \\(n\\) observed individuals \\(f_n = \\alpha x^n / n\\) (Fisher et al. 1943). estimation possible genuine counts individuals. parameter \\(\\alpha\\) used diversity index can estimated separate function fisher.alpha. parameter \\(x\\) taken nuisance parameter estimated separately taken \\(n/(n+\\alpha)\\). Helper function .fisher transforms abundance data Fisher frequency table. Diversity given NA communities one (zero) species: reliable way estimating diversity, even equations return bogus numeric value cases. Preston (1948) satisfied Fisher's model seemed imply infinite species richness, postulated rare species diminishing class species middle frequency scale. achieved collapsing higher frequency classes wider wider “octaves” doubling class limits: 1, 2, 3--4, 5--8, 9--16 etc. occurrences. seems Preston regarded frequencies 1, 2, 4, etc.. “tied” octaves (Williamson & Gaston 2005). means half species frequency 1 shown lowest octave, rest transferred second octave. Half species second octave transferred higher one well, usually large number species. practise makes data look lognormal reducing usually high lowest octaves. can achieved setting argument tiesplit = TRUE. tiesplit = FALSE frequencies split, ones lowest octave, twos second, etc. Williamson & Gaston (2005) discuss alternative definitions detail, consulted critical review log-Normal model. logseries data look like lognormal plotted Preston's way. expected frequency \\(f\\) abundance octave \\(o\\) defined \\(f_o = S_0 \\exp(-(\\log_2(o) - \\mu)^2/2/\\sigma^2)\\), \\(\\mu\\) location mode \\(\\sigma\\) width, \\(\\log_2\\) scale, \\(S_0\\) expected number species mode. lognormal model usually truncated left rare species observed. Function prestonfit fits truncated lognormal model second degree log-polynomial octave pooled data using Poisson ( tiesplit = FALSE) quasi-Poisson (tiesplit = TRUE) error. Function prestondistr fits left-truncated Normal distribution \\(\\log_2\\) transformed non-pooled observations direct maximization log-likelihood. Function prestondistr modelled function fitdistr can used alternative distribution models. functions common print, plot lines methods. lines function adds fitted curve octave range line segments showing location mode width (sd) response. Function .preston transforms abundance data octaves. Argument tiesplit influence fit prestondistr, influence barplot octaves. total extrapolated richness fitted Preston model can found function veiledspec. function accepts results prestonfit prestondistr. veiledspec called species count vector, internally use prestonfit. Function specpool provides alternative ways estimating number unseen species. fact, Preston's lognormal model seems truncated ends, may main reason result differ lognormal models fitted Rank--Abundance diagrams functions rad.lognormal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"function prestonfit returns object fitted coefficients, observed (freq) fitted (fitted) frequencies, string describing fitting method. Function prestondistr omits entry fitted. function fisherfit returns result nlm, item estimate \\(\\alpha\\). result object amended nuisance parameter item fisher observed data .fisher","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12: 42--58. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Williamson, M. & Gaston, K.J. (2005). lognormal distribution appropriate null hypothesis species--abundance distribution. Journal Animal Ecology 74, 409--422.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Bob O'Hara Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"data(BCI) mod <- fisherfit(BCI[5,]) mod #> #> Fisher log series model #> No. of species: 101 #> Fisher alpha: 37.96423 #> # prestonfit seems to need large samples mod.oct <- prestonfit(colSums(BCI)) mod.ll <- prestondistr(colSums(BCI)) mod.oct #> #> Preston lognormal model #> Method: Quasi-Poisson fit to octaves #> No. of species: 225 #> #> mode width S0 #> 4.885798 2.932690 32.022923 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.500000 16.00000 18.00000 19.000 30.00000 35.00000 31.00000 26.50000 #> Fitted 7.994154 13.31175 19.73342 26.042 30.59502 31.99865 29.79321 24.69491 #> 8 9 10 11 #> Observed 18.00000 13.00000 7.000000 2.0000 #> Fitted 18.22226 11.97021 7.000122 3.6443 #> mod.ll #> #> Preston lognormal model #> Method: maximized likelihood to log2 abundances #> No. of species: 225 #> #> mode width S0 #> 4.365002 2.753531 33.458185 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.50000 16.00000 18.00000 19.00000 30.00000 35.00000 31.00000 26.50000 #> Fitted 9.52392 15.85637 23.13724 29.58961 33.16552 32.58022 28.05054 21.16645 #> 8 9 10 11 #> Observed 18.00000 13.000000 7.000000 2.00000 #> Fitted 13.99829 8.113746 4.121808 1.83516 #> plot(mod.oct) lines(mod.ll, line.col=\"blue3\") # Different ## Smoothed density den <- density(log2(colSums(BCI))) lines(den$x, ncol(BCI)*den$y, lwd=2) # Fairly similar to mod.oct ## Extrapolated richness veiledspec(mod.oct) #> Extrapolated Observed Veiled #> 235.40577 225.00000 10.40577 veiledspec(mod.ll) #> Extrapolated Observed Veiled #> 230.931018 225.000000 5.931018"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Functions goodness inertcomp can used assess goodness fit individual sites species. Function vif.cca alias.cca can used analyse linear dependencies among constraints conditions. addition, diagnostic tools (see 'Details').","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"# S3 method for cca goodness(object, choices, display = c(\"species\", \"sites\"), model = c(\"CCA\", \"CA\"), summarize = FALSE, addprevious = FALSE, ...) inertcomp(object, display = c(\"species\", \"sites\"), unity = FALSE, proportional = FALSE) spenvcor(object) intersetcor(object) vif.cca(object) # S3 method for cca alias(object, names.only = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"object result object cca, rda, dbrda capscale. display Display \"species\" \"sites\". Species available dbrda capscale. choices Axes shown. Default show axes \"model\". model Show constrained (\"CCA\") unconstrained (\"CA\") results. summarize Show accumulated total. addprevious Add variation explained previous components statistic=\"explained\". model = \"CCA\" add conditioned (partialled ) variation, model = \"CA\" add conditioned constrained variation. give cumulative explanation previous components. unity Scale inertia components unit sum (sum items 1). proportional Give inertia components proportional corresponding total item (sum row 1). option takes precedence unity. names.Return names aliased variable(s) instead defining equations. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Function goodness gives cumulative proportion inertia accounted species chosen axes. proportions can assessed either species sites depending argument display, species available distance-based dbrda. function implemented capscale. Function inertcomp decomposes inertia partial, constrained unconstrained components site species. Legendre & De Cáceres (2012) called inertia components local contributions beta-diversity (LCBD) species contributions beta-diversity (SCBD), give relative contributions summing unity (argument unity = TRUE). interpretation, appropriate dissimilarity measures used dbrda appropriate standardization rda (Legendre & De Cáceres 2012). function implemented capscale. Function spenvcor finds -called “species -- environment correlation” (weighted) correlation weighted average scores linear combination scores. bad measure goodness ordination, sensitive extreme scores (like correlations ), sensitive overfitting using many constraints. Better models often poorer correlations. Function ordispider can show graphically. Function intersetcor finds -called “interset correlation” (weighted) correlation weighted averages scores constraints. defined contrasts used factor variables. bad measure since correlation. , focuses correlations single contrasts single axes instead looking multivariate relationship. Fitted vectors (envfit) provide better alternative. Biplot scores (see scores.cca) multivariate alternative (weighted) correlation linear combination scores constraints. Function vif.cca gives variance inflation factors constraint contrast factor constraints. partial ordination, conditioning variables analysed together constraints. Variance inflation diagnostic tool identify useless constraints. common rule values 10 indicate redundant constraints. later constraints complete linear combinations conditions previous constraints, completely removed estimation, biplot scores centroids calculated aliased constraints. note printed default output aliased constraints. Function alias give linear coefficients defining aliased constraints, names argument names.= TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"functions return matrices vectors appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London. Gross, J. (2003). Variance inflation factors. R News 3(1), 13--15. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Jari Oksanen. vif.cca relies heavily code W. N. Venables. alias.cca simplified version alias.lm.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) goodness(mod, addprevious = TRUE) #> CCA1 CCA2 CCA3 CCA4 #> Achimill 0.36630013 0.3822685 0.3838616 0.4934158 #> Agrostol 0.67247051 0.6724758 0.6779597 0.7773267 #> Airaprae 0.36213737 0.3698100 0.3816619 0.3908018 #> Alopgeni 0.61547145 0.6966105 0.7042650 0.7212918 #> Anthodor 0.24619147 0.2795001 0.3509172 0.3609709 #> Bellpere 0.41185412 0.4179432 0.4847618 0.4849622 #> Bromhord 0.33487622 0.3397416 0.3870032 0.5505037 #> Chenalbu 0.23594716 0.2684323 0.2828928 0.2885321 #> Cirsarve 0.29041563 0.3013655 0.3080671 0.3591280 #> Comapalu 0.16338257 0.6836790 0.7390659 0.7963425 #> Eleopalu 0.55132024 0.6099415 0.6193301 0.6259818 #> Elymrepe 0.25239595 0.2710266 0.2761491 0.2882666 #> Empenigr 0.27089495 0.3132399 0.3153052 0.3154203 #> Hyporadi 0.31349648 0.3371809 0.3387669 0.3388716 #> Juncarti 0.43923609 0.4492937 0.4871043 0.5224072 #> Juncbufo 0.70439967 0.7226263 0.7228786 0.7257471 #> Lolipere 0.48141171 0.5720410 0.5727299 0.6034007 #> Planlanc 0.54969676 0.6084389 0.6802195 0.6826265 #> Poaprat 0.40267189 0.4944813 0.5014516 0.5326546 #> Poatriv 0.49694972 0.5409439 0.5468830 0.5594817 #> Ranuflam 0.68677962 0.6983001 0.7020461 0.7064850 #> Rumeacet 0.44788204 0.5211145 0.7673956 0.7691199 #> Sagiproc 0.27039747 0.3497634 0.3553109 0.3613746 #> Salirepe 0.64788354 0.7264891 0.7276110 0.7639711 #> Scorautu 0.54312496 0.5510319 0.6078931 0.6140593 #> Trifprat 0.37328840 0.4101104 0.6624199 0.6625703 #> Trifrepe 0.03048149 0.2115857 0.3300132 0.4207437 #> Vicilath 0.17824132 0.1784611 0.3762406 0.4279428 #> Bracruta 0.15585567 0.1641095 0.1672797 0.2449864 #> Callcusp 0.30771429 0.3143582 0.3308502 0.3518027 goodness(mod, addprevious = TRUE, summ = TRUE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 0.4934158 0.7773267 0.3908018 0.7212918 0.3609709 0.4849622 0.5505037 0.2885321 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 0.3591280 0.7963425 0.6259818 0.2882666 0.3154203 0.3388716 0.5224072 0.7257471 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 0.6034007 0.6826265 0.5326546 0.5594817 0.7064850 0.7691199 0.3613746 0.7639711 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 0.6140593 0.6625703 0.4207437 0.4279428 0.2449864 0.3518027 # Inertia components inertcomp(mod, prop = TRUE) #> pCCA CCA CA #> Achimill 0.34271900 0.15069678 0.5065842 #> Agrostol 0.55602406 0.22130269 0.2226733 #> Airaprae 0.06404726 0.32675457 0.6091982 #> Alopgeni 0.34238968 0.37890210 0.2787082 #> Anthodor 0.10259139 0.25837947 0.6390291 #> Bellpere 0.40972447 0.07523776 0.5150378 #> Bromhord 0.33046684 0.22003683 0.4494963 #> Chenalbu 0.11064346 0.17788865 0.7114679 #> Cirsarve 0.26649913 0.09262886 0.6408720 #> Comapalu 0.16096277 0.63537969 0.2036575 #> Eleopalu 0.53954819 0.08643366 0.3740182 #> Elymrepe 0.22234322 0.06592337 0.7117334 #> Empenigr 0.10361994 0.21180040 0.6845797 #> Hyporadi 0.03889627 0.29997533 0.6611284 #> Juncarti 0.43439190 0.08801527 0.4775928 #> Juncbufo 0.66622672 0.05952038 0.2742529 #> Lolipere 0.46273045 0.14067027 0.3965993 #> Planlanc 0.51993753 0.16268893 0.3173735 #> Poaprat 0.39408053 0.13857406 0.4673454 #> Poatriv 0.05598349 0.50349824 0.4405183 #> Ranuflam 0.68509904 0.02138594 0.2935150 #> Rumeacet 0.40125987 0.36786003 0.2308801 #> Sagiproc 0.26050435 0.10087025 0.6386254 #> Salirepe 0.12527838 0.63869277 0.2360289 #> Scorautu 0.10895437 0.50510492 0.3859407 #> Trifprat 0.34544815 0.31712212 0.3374297 #> Trifrepe 0.02132183 0.39942191 0.5792563 #> Vicilath 0.12125433 0.30668844 0.5720572 #> Bracruta 0.07222706 0.17275938 0.7550136 #> Callcusp 0.29447422 0.05732850 0.6481973 inertcomp(mod) #> pCCA CCA CA #> Achimill 0.0173766015 0.007640656 0.02568493 #> Agrostol 0.0456558521 0.018171449 0.01828399 #> Airaprae 0.0066672285 0.034014687 0.06341666 #> Alopgeni 0.0325977567 0.036073980 0.02653486 #> Anthodor 0.0096274015 0.024246897 0.05996790 #> Bellpere 0.0154640710 0.002839669 0.01943887 #> Bromhord 0.0180126793 0.011993496 0.02450059 #> Chenalbu 0.0031913088 0.005130874 0.02052099 #> Cirsarve 0.0110663060 0.003846389 0.02661204 #> Comapalu 0.0127652351 0.050389111 0.01615116 #> Eleopalu 0.0797827194 0.012780901 0.05530588 #> Elymrepe 0.0193932154 0.005749967 0.06207879 #> Empenigr 0.0063826176 0.013046147 0.04216766 #> Hyporadi 0.0046669914 0.035992710 0.07932587 #> Juncarti 0.0359126341 0.007276518 0.03948420 #> Juncbufo 0.0494087668 0.004414156 0.02033917 #> Lolipere 0.0368344271 0.011197683 0.03157023 #> Planlanc 0.0366139947 0.011456552 0.02234944 #> Poaprat 0.0142991623 0.005028142 0.01695757 #> Poatriv 0.0028845344 0.025942611 0.02269759 #> Ranuflam 0.0446783229 0.001394671 0.01914141 #> Rumeacet 0.0288221948 0.026423110 0.01658394 #> Sagiproc 0.0151161507 0.005853146 0.03705718 #> Salirepe 0.0142756439 0.072779924 0.02689581 #> Scorautu 0.0030643984 0.014206339 0.01085478 #> Trifprat 0.0228613139 0.020986733 0.02233067 #> Trifrepe 0.0008339368 0.015622139 0.02265580 #> Vicilath 0.0049088357 0.012415912 0.02315905 #> Bracruta 0.0032317812 0.007730074 0.03378289 #> Callcusp 0.0319130878 0.006212868 0.07024716 # vif.cca vif.cca(mod) #> Moisture.L Moisture.Q Moisture.C A1 ManagementHF ManagementNM #> 1.504327 1.284489 1.347660 1.367328 2.238653 2.570972 #> ManagementSF #> 2.424444 # Aliased constraints mod <- cca(dune ~ ., dune.env) mod #> Call: cca(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 1.5032 0.7106 12 #> Unconstrained 0.6121 0.2894 7 #> Inertia is scaled Chi-square #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4671 0.3410 0.1761 0.1532 0.0953 0.0703 0.0589 0.0499 0.0318 0.0260 0.0228 #> CCA12 #> 0.0108 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 #> 0.27237 0.10876 0.08975 0.06305 0.03489 0.02529 0.01798 #> vif.cca(mod) #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> 2.208249 2.858927 3.072715 3.587087 6.608315 142.359372 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> 12.862713 2.642718 3.007238 80.828330 49.294455 21.433337 #> Manure^4 #> NA alias(mod) #> Model : #> dune ~ A1 + Moisture + Management + Use + Manure #> #> Complete : #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> Manure^4 8.366600 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> Manure^4 5.291503 -4.472136 2.645751 #> with(dune.env, table(Management, Manure)) #> Manure #> Management 0 1 2 3 4 #> BF 0 2 1 0 0 #> HF 0 1 2 2 0 #> NM 6 0 0 0 0 #> SF 0 0 1 2 3 # The standard correlations (not recommended) ## IGNORE_RDIFF_BEGIN spenvcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 #> 0.9636709 0.9487249 0.9330741 0.8734876 0.9373716 0.8362687 0.9748793 0.8392720 #> CCA9 CCA10 CCA11 CCA12 #> 0.8748741 0.6087512 0.6633248 0.7581210 intersetcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 #> A1 -0.5332506 0.13691202 -0.47996401 -0.259859587 -0.09894964 #> Moisture.L -0.8785505 0.17867589 0.03714134 0.181952935 -0.09826534 #> Moisture.Q -0.1956664 -0.33044917 -0.27321286 -0.180333890 0.26609291 #> Moisture.C -0.2023782 -0.09698397 0.28596824 -0.261712720 -0.49103002 #> ManagementHF 0.3473460 0.01680324 -0.51205769 0.194144965 0.30752664 #> ManagementNM -0.5699549 -0.61111645 0.14751127 -0.013777789 0.04571982 #> ManagementSF -0.1197499 0.64084416 0.19780650 0.134892908 -0.09679992 #> Use.L -0.1871999 0.32990444 -0.30941161 -0.372747011 0.09586963 #> Use.Q -0.1820298 -0.48874152 -0.01997442 -0.009812946 0.04812588 #> Manure.L 0.3175126 0.65945634 0.03724864 -0.025383543 -0.04077470 #> Manure.Q -0.4075615 -0.21149073 0.49297244 -0.176686201 0.11973190 #> Manure.C 0.4676279 0.11376054 0.29132473 -0.173382982 0.14219924 #> Manure^4 0.2222349 -0.12789494 -0.12921227 0.108367170 -0.02559567 #> CCA6 CCA7 CCA8 CCA9 CCA10 #> A1 -0.15225816 0.25788462 0.19247720 -0.27694466 -0.1158449480 #> Moisture.L -0.02923342 0.07858647 -0.10772510 0.07101300 0.0952517164 #> Moisture.Q -0.11211675 0.05062810 -0.48302647 0.06138704 -0.2053304965 #> Moisture.C -0.23581275 -0.38693407 -0.10144580 -0.21907160 0.1875632770 #> ManagementHF -0.24278705 0.16364055 -0.14053438 0.31066725 0.1310215145 #> ManagementNM -0.06430101 0.23917584 0.14375754 -0.27103732 0.0002768613 #> ManagementSF -0.01611984 -0.49726250 0.08073472 -0.30235728 -0.1381281272 #> Use.L 0.19127262 -0.44624831 -0.18450714 0.12950951 0.0452826749 #> Use.Q 0.13485545 0.10367354 -0.11020112 0.41245485 -0.0766932005 #> Manure.L -0.22265819 -0.49627772 -0.16971786 -0.03943343 -0.0045229147 #> Manure.Q -0.19402211 -0.11937394 0.17611673 -0.44002593 0.0903998202 #> Manure.C 0.14760330 0.07842345 0.37774417 0.10181374 0.1055057288 #> Manure^4 -0.36683782 0.05953330 0.40927409 -0.06054381 -0.1500198368 #> CCA11 CCA12 #> A1 -0.03550223 -0.08881387 #> Moisture.L 0.06404776 -0.08587882 #> Moisture.Q -0.21810558 0.16917878 #> Moisture.C 0.13701079 -0.14260914 #> ManagementHF 0.17283125 0.13296499 #> ManagementNM -0.01358436 0.09533598 #> ManagementSF -0.01468592 -0.06614834 #> Use.L -0.08584883 0.32559307 #> Use.Q 0.41893616 0.04881247 #> Manure.L 0.02396993 0.13049087 #> Manure.Q 0.12987366 0.07137031 #> Manure.C 0.05176927 -0.41550238 #> Manure^4 -0.41603287 0.01661279 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS find goodness fit measure points nonmetric multidimensional scaling, function stressplot makes Shepard diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"# S3 method for metaMDS goodness(object, dis, ...) # S3 method for default stressplot(object, dis, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"object result object metaMDS, monoMDS isoMDS. dis Dissimilarities. used metaMDS monoMDS, must used isoMDS. pch Plotting character points. Default dependent number points. p.col, l.col Point line colours. lwd Line width. monoMDS default lwd = 1 two lines drawn, lwd = 2 otherwise. ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS finds goodness fit statistic observations (points). defined sum squared values equal squared stress. Large values indicate poor fit. absolute values goodness statistic depend definition stress: isoMDS expresses stress percents, therefore goodness values 100 times higher monoMDS expresses stress proportion. Function stressplot draws Shepard diagram plot ordination distances monotone linear fit line original dissimilarities. addition, displays two correlation-like statistics goodness fit graph. nonmetric fit based stress \\(S\\) defined \\(R^2 = 1-S^2\\). “linear fit” squared correlation fitted values ordination distances. monoMDS, “linear fit” \\(R^2\\) “stress type 2” equal. functions can used metaMDS, monoMDS isoMDS. original dissimilarities given monoMDS metaMDS results (latter tries reconstruct dissimilarities using metaMDSredist isoMDS used engine). isoMDS dissimilarities must given. either case, functions inspect dissimilarities consistent current ordination, refuse analyse inconsistent dissimilarities. Function goodness.metaMDS generic vegan, must spell name completely isoMDS class.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness returns vector values. Function stressplot returns invisibly object items original dissimilarities, ordination distances fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"data(varespec) mod <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2410643 #> Run 2 stress 0.1948413 #> Run 3 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.0416189 max resid 0.1517616 #> Run 4 stress 0.2383056 #> Run 5 stress 0.2096851 #> Run 6 stress 0.2329653 #> Run 7 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 2.732084e-05 max resid 8.096006e-05 #> ... Similar to previous best #> Run 8 stress 0.1825658 #> ... Procrustes: rmse 1.167373e-05 max resid 3.019892e-05 #> ... Similar to previous best #> Run 9 stress 0.2092456 #> Run 10 stress 0.1976154 #> Run 11 stress 0.2331577 #> Run 12 stress 0.2330486 #> Run 13 stress 0.1948413 #> Run 14 stress 0.2356726 #> Run 15 stress 0.1955838 #> Run 16 stress 0.2339637 #> Run 17 stress 0.2088293 #> Run 18 stress 0.2138946 #> Run 19 stress 0.2136761 #> Run 20 stress 0.2352729 #> *** Best solution repeated 2 times stressplot(mod) gof <- goodness(mod) gof #> [1] 0.02984516 0.03513715 0.04189194 0.04598243 0.04003134 0.03441442 #> [7] 0.03294898 0.03050145 0.03060754 0.02994058 0.03526332 0.02621426 #> [13] 0.03830984 0.02980913 0.03369611 0.02225897 0.03561569 0.03505254 #> [19] 0.06577496 0.03268352 0.03503121 0.02956628 0.05168013 0.04601957 plot(mod, display = \"sites\", type = \"n\") points(mod, display = \"sites\", cex = 2*gof/mean(gof))"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":null,"dir":"Reference","previous_headings":"","what":"Indicator Power of Species — indpower","title":"Indicator Power of Species — indpower","text":"Indicator power calculation Halme et al. (2009) congruence indicator target species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indicator Power of Species — indpower","text":"","code":"indpower(x, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indicator Power of Species — indpower","text":"x Community data frame matrix. type type statistic returned. See Details explanation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indicator Power of Species — indpower","text":"Halme et al. (2009) described index indicator power defined \\(IP_I = \\sqrt{\\times b}\\), \\(= S / O_I\\) \\(b = 1 - (O_T - S) / (N - O_I)\\). \\(N\\) number sites, \\(S\\) number shared occurrences indicator (\\(\\)) target (\\(T\\)) species. \\(O_I\\) \\(O_T\\) number occurrences indicator target species. type argument function call enables choose statistic return. type = 0 returns \\(IP_I\\), type = 1 returns \\(\\), type = 2 returns \\(b\\). Total indicator power (TIP) indicator species column mean (without value, see examples). Halme et al. (2009) explain calculate confidence intervals statistics, see Examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indicator Power of Species — indpower","text":"matrix indicator species rows target species columns (indicated first letters row/column names).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indicator Power of Species — indpower","text":"Halme, P., Mönkkönen, M., Kotiaho, J. S, Ylisirniö, -L. 2009. Quantifying indicator power indicator species. Conservation Biology 23: 1008--1016.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indicator Power of Species — indpower","text":"Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indicator Power of Species — indpower","text":"","code":"data(dune) ## IP values ip <- indpower(dune) ## and TIP values diag(ip) <- NA (TIP <- rowMeans(ip, na.rm=TRUE)) #> i.Achimill i.Agrostol i.Airaprae i.Alopgeni i.Anthodor i.Bellpere i.Bromhord #> 0.3186250 0.3342800 0.2168133 0.3416198 0.3567884 0.3432281 0.3665632 #> i.Chenalbu i.Cirsarve i.Comapalu i.Eleopalu i.Elymrepe i.Empenigr i.Hyporadi #> 0.2095044 0.2781640 0.1713273 0.2414787 0.3263516 0.2016196 0.2378197 #> i.Juncarti i.Juncbufo i.Lolipere i.Planlanc i.Poaprat i.Poatriv i.Ranuflam #> 0.2915850 0.3331330 0.3998442 0.3426064 0.4094319 0.3929520 0.2663080 #> i.Rumeacet i.Sagiproc i.Salirepe i.Scorautu i.Trifprat i.Trifrepe i.Vicilath #> 0.3484684 0.3788905 0.2898512 0.4362493 0.3145854 0.4503764 0.2605349 #> i.Bracruta i.Callcusp #> 0.4252676 0.2070766 ## p value calculation for a species ## from Halme et al. 2009 ## i is ID for the species i <- 1 fun <- function(x, i) indpower(x)[i,-i] ## 'c0' randomizes species occurrences os <- oecosimu(dune, fun, \"c0\", i=i, nsimul=99) #> Warning: nullmodel transformed 'comm' to binary data ## get z values from oecosimu output z <- os$oecosimu$z ## p-value (p <- sum(z) / sqrt(length(z))) #> [1] -1.471628 ## 'heterogeneity' measure (chi2 <- sum((z - mean(z))^2)) #> [1] 86.33867 pchisq(chi2, df=length(z)-1) #> [1] 0.9999999 ## Halme et al.'s suggested output out <- c(TIP=TIP[i], significance=p, heterogeneity=chi2, minIP=min(fun(dune, i=i)), varIP=sd(fun(dune, i=i)^2)) out #> TIP.i.Achimill significance heterogeneity minIP varIP #> 0.3186250 -1.4716280 86.3386705 0.0000000 0.2142097"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Linear Model Diagnostics for Constrained Ordination — influence.cca","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"set function extracts influence statistics linear model statistics directly constrained ordination result object cca, rda, capscale dbrda. constraints linear model functions support functions return identical results corresponding linear models (lm), can use documentation. main functions normal usage leverage values (hatvalues), standardized residuals (rstandard), studentized leave-one-residuals (rstudent), Cook's distance (cooks.distance). addition, vcov returns variance-covariance matrix coefficients, diagonal values variances coefficients. functions mainly support functions , can used directly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"# S3 method for cca hatvalues(model, ...) # S3 method for cca rstandard(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca rstudent(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca cooks.distance(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca sigma(object, type = c(\"response\", \"canoco\"), ...) # S3 method for cca vcov(object, type = \"canoco\", ...) # S3 method for cca SSD(object, type = \"canoco\", ...) # S3 method for cca qr(x, ...) # S3 method for cca df.residual(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"model, object, x constrained ordination result object. type Type statistics used extracting raw residuals residual standard deviation (sigma). Either \"response\" species data difference WA LC scores \"canoco\". ... arguments functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"vegan algorithm constrained ordination uses linear model (weighted linear model cca) find fitted values dependent community data, constrained ordination based fitted response (Legendre & Legendre 2012). hatvalues give leverage values constraints, leverage independent response data. influence statistics (rstandard, rstudent, cooks.distance) based leverage, raw residuals residual standard deviation (sigma). type = \"response\" raw residuals given unconstrained component constrained ordination, influence statistics matrix dimensions . observations times . species. cca statistics obtained lm model using Chi-square standardized species data (see decostand) dependent variable, row sums community data weights, rda lm model uses non-modified community data weights. algorithm CANOCO software constraints results iteration performing linear regression weighted averages (WA) scores constraints taking fitted values regression linear combination (LC) scores (ter Braak 1984). WA scores directly found species scores, LC scores linear combinations constraints regression. type = \"canoco\" raw residuals differences WA LC scores, residual standard deviation (sigma) taken axis sum squared WA scores minus one. quantities relationship residual component ordination, rather methodological artefacts algorithm used vegan. result matrix dimensions . observations times . constrained axes. Function vcov returns matrix variances covariances regression coefficients. diagonal values matrix variances, square roots give standard errors regression coefficients. function based SSD extracts sum squares crossproducts residuals. residuals defined similarly influence measures type similar properties limitations, define dimensions result matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Function .mlm casts ordination object multiple linear model class \"mlm\" (see lm), similar statistics can derived modified object set functions. However, problems R implementation analysis multiple linear model objects. results differ, current set functions probable correct. use .mlm objects avoided.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"data(varespec, varechem) mod <- cca(varespec ~ Al + P + K, varechem) ## leverage hatvalues(mod) #> 18 15 24 27 23 19 22 #> 0.06904416 0.06666628 0.15245083 0.18944882 0.09291510 0.05122338 0.15309307 #> 16 28 13 14 20 25 7 #> 0.09605909 0.27139695 0.75889765 0.04958141 0.06582891 0.10590183 0.20630888 #> 5 6 3 4 2 9 12 #> 0.19797654 0.16280522 0.22738889 0.30915530 0.15557066 0.14855598 0.09046701 #> 10 11 21 #> 0.12745850 0.10984996 0.14195559 plot(hatvalues(mod), type = \"h\") ## ordination plot with leverages: points with high leverage have ## similar LC and WA scores plot(mod, type = \"n\") ordispider(mod) # segment from LC to WA scores points(mod, dis=\"si\", cex=5*hatvalues(mod), pch=21, bg=2) # WA scores text(mod, dis=\"bp\", col=4) ## deviation and influence head(rstandard(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.4084518 0.9442480 -0.68178124 -0.798241724 0.9883838 -0.3086868 #> 15 -1.3902462 -1.5717947 -0.70784872 -0.645563228 0.2353736 -0.1679226 #> 24 0.9622453 -0.9520875 -0.08884556 -0.654099911 0.2420416 0.4832198 #> 27 -1.1080099 1.0938951 1.70146427 -0.196668562 -0.3937467 -0.7424140 #> 23 0.3979939 1.3218254 -0.63872221 -1.003315524 1.8996365 -0.4495408 #> 19 -1.5874575 0.7894087 -0.59609083 -0.006142973 0.1334143 -0.1060450 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 -0.5785258 -0.4585683 0.7640788 4.3748349 -0.39301720 -0.656213958 #> 15 -0.5416812 -0.4594716 -0.3410155 -0.2704388 -0.05769657 0.406022095 #> 24 -0.6409619 0.1908003 0.0198320 -0.2175720 3.75416938 -0.009140093 #> 27 4.2976822 -0.2704153 0.9211453 -0.2037075 -0.84410200 -0.712968237 #> 23 -0.8356637 -0.2779318 -0.0206861 -0.2899895 -0.67773316 -0.294674408 #> 19 -0.4453843 -0.3739569 -0.3174207 -0.2429740 -0.14281646 -0.819230368 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 -0.4933634 -0.6030042 -1.3812122 -0.07430854 -0.5332731 -0.7140513 #> 15 -0.3630434 -0.4055353 1.9930791 0.02126658 -0.2274864 -0.6730241 #> 24 2.2312025 -1.3907968 0.5792314 -0.45358405 -0.4628096 -0.2153846 #> 27 -0.5715008 1.6310289 0.8124329 -0.34628172 -0.8630615 0.9423113 #> 23 -0.4684368 -1.1995321 -0.8241477 -0.04953929 0.7890327 -0.6774675 #> 19 -0.3186606 -0.3948224 0.6180114 0.05035990 0.8260214 2.3626483 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 18 0.2918311 -0.42027512 -0.40191333 1.5303054 0.85056948 -0.2218833 #> 15 -0.6168627 -0.47096984 -0.42715787 0.1758993 -0.45409579 -0.3739920 #> 24 1.9274063 0.09984384 0.21066100 0.1668363 -0.08508535 -1.2835676 #> 27 -1.5729552 -0.31944598 -0.09167396 -0.1864165 0.51080623 -0.1083797 #> 23 0.4875575 0.61798970 -0.18848086 0.5574309 0.28079468 -0.4026464 #> 19 -0.1141521 -0.25344373 -0.31179839 -0.2391669 -0.65577452 0.6584127 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris #> 18 -0.39796095 0.93836573 -0.2564543 0.33864378 1.1572584 -0.2083305 #> 15 0.06761526 0.24340663 -0.1767166 0.27491203 1.0721632 1.9724410 #> 24 1.24902375 -0.98471253 -0.4801382 2.51184311 -1.4063518 -0.3084304 #> 27 -0.59021669 -1.25354423 -0.2460447 -1.09351514 -1.1681499 -1.0337232 #> 23 -0.34604539 -0.10730202 3.9477300 2.51924664 0.3536280 3.3882402 #> 19 -0.33866721 0.02698153 0.1776632 0.03968833 -0.7512944 -0.4763562 #> Cladchlo Cladbotr Cladamau Cladsp Cetreric Cetrisla #> 18 -0.5711604 -0.4914716 4.0852019 0.2489284 -0.4428064 -0.5834462 #> 15 -0.4347061 -0.6517740 -0.3155708 -0.2568784 0.2369559 -0.3468005 #> 24 0.6013607 0.4603779 -0.1576900 -0.5995616 2.7826114 0.3931826 #> 27 -0.5436659 -0.2788962 -0.1606997 0.1557879 -0.7298364 -0.5867612 #> 23 0.1029999 0.6494142 -0.3415411 -0.2330698 -0.6309632 -0.5301977 #> 19 0.2793712 -0.1315438 -0.2705486 -0.2865837 -0.4664078 -0.5056846 #> Flavniva Nepharct Stersp Peltapht Icmaeric Cladcerv #> 18 0.30428187 -0.3624631 -0.23665431 -0.1571633 -0.630333375 0.12058739 #> 15 0.17345018 -0.1919943 0.05469573 -0.3233311 -0.561177494 0.08518455 #> 24 -0.74154401 -0.4146848 -0.05500461 -0.7609417 0.255058737 -0.92761801 #> 27 0.11301489 -0.5064006 -0.08681568 -0.1247151 -0.001277338 -0.04038189 #> 23 0.09411988 -0.4627811 0.47668055 3.5826478 -0.274664798 -0.05517988 #> 19 0.07211309 -0.1693122 -0.17244475 -0.3155345 -0.461532920 -0.02709075 #> Claddefo Cladphyl #> 18 -0.43581630 -0.2098378 #> 15 0.94176661 -0.1028102 #> 24 -0.07508682 -1.0479632 #> 27 -1.06110299 -0.4908554 #> 23 2.66430575 -0.4244333 #> 19 -0.08427954 -0.1692474 head(cooks.distance(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.003093283 0.01653142 0.0086184263 1.181427e-02 0.0181129462 0.0017667454 #> 15 0.034513793 0.04411649 0.0089472619 7.441951e-03 0.0009892926 0.0005035324 #> 24 0.041636714 0.04076229 0.0003549575 1.923947e-02 0.0026344196 0.0105001237 #> 27 0.071736260 0.06992022 0.1691597848 2.260067e-03 0.0090591037 0.0322065174 #> 23 0.004056312 0.04474315 0.0104472601 2.577825e-02 0.0924100906 0.0051750754 #> 19 0.034013281 0.00841101 0.0047958896 5.093326e-07 0.0002402422 0.0001517834 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 0.006205594 0.003898934 1.082466e-02 0.3548634058 2.863921e-03 7.984152e-03 #> 15 0.005239584 0.003769873 2.076622e-03 0.0013060122 5.944416e-05 2.943802e-03 #> 24 0.018474359 0.001637053 1.768633e-05 0.0021286828 6.337714e-01 3.756697e-06 #> 27 1.079245024 0.004272814 4.958014e-02 0.0024247421 4.163335e-02 2.970242e-02 #> 23 0.017883042 0.001978130 1.095811e-05 0.0021534900 1.176240e-02 2.223634e-03 #> 19 0.002677405 0.001887502 1.359924e-03 0.0007968265 2.752966e-04 9.058502e-03 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 0.004513066 0.006741841 0.035371924 1.023801e-04 0.0052727512 0.009453588 #> 15 0.002353565 0.002936747 0.070934636 8.076156e-06 0.0009241023 0.008088546 #> 24 0.223863289 0.086982577 0.015087217 9.251676e-03 0.0096318471 0.002086096 #> 27 0.019084693 0.155444288 0.038567940 7.006653e-03 0.0435246182 0.051884812 #> 23 0.005619275 0.036846999 0.017393559 6.284604e-05 0.0159429407 0.011753178 #> 19 0.001370570 0.002104010 0.005155103 3.423056e-05 0.0092093061 0.075342921 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang #> 18 0.0015790689 0.0032749544 0.0029950405 0.0434204290 0.0134139742 #> 15 0.0067949512 0.0039609143 0.0032582625 0.0005525065 0.0036821731 #> 24 0.1670519112 0.0004482780 0.0019955986 0.0012516596 0.0003255476 #> 27 0.1445719869 0.0059627459 0.0004910701 0.0020305802 0.0152462776 #> 23 0.0060873754 0.0097800450 0.0009097308 0.0079572035 0.0020190921 #> 19 0.0001758783 0.0008669766 0.0013121758 0.0007720518 0.0058043470 #> Cladstel Cladunci Cladcocc Cladcorn Cladgrac Cladfimb #> 18 0.0009128242 2.936424e-03 1.632609e-02 0.0012194323 2.126298e-03 0.024831249 #> 15 0.0024976637 8.163919e-05 1.057972e-03 0.0005576530 1.349574e-03 0.020527262 #> 24 0.0740870642 7.015301e-02 4.360375e-02 0.0103666250 2.837200e-01 0.088939118 #> 27 0.0006863532 2.035516e-02 9.181862e-02 0.0035373629 6.987166e-02 0.079734929 #> 23 0.0041517010 3.066511e-03 2.948453e-04 0.3990922140 1.625248e-01 0.003202372 #> 19 0.0058511426 1.548070e-03 9.826017e-06 0.0004260291 2.126033e-05 0.007618415 #> Cladcris Cladchlo Cladbotr Cladamau Cladsp Cetreric #> 18 0.0008047178 0.0060485887 0.0044785217 0.3094317912 0.001148912 0.003635513 #> 15 0.0694732020 0.0033744318 0.0075858327 0.0017782910 0.001178323 0.001002639 #> 24 0.0042777886 0.0162620357 0.0095308986 0.0011181833 0.016164882 0.348184969 #> 27 0.0624395810 0.0172709295 0.0045450274 0.0015089732 0.001418140 0.031124511 #> 23 0.2939860714 0.0002716767 0.0107999564 0.0029872002 0.001391075 0.010194981 #> 19 0.0030627243 0.0010534342 0.0002335528 0.0009879496 0.001108529 0.002936134 #> Cetrisla Flavniva Nepharct Stersp Peltapht Icmaeric #> 18 0.006311601 1.716683e-03 0.0024359336 1.038405e-03 0.0004579733 7.366793e-03 #> 15 0.002147676 5.372282e-04 0.0006582422 5.342149e-05 0.0018668272 5.623539e-03 #> 24 0.006951741 2.472742e-02 0.0077328775 1.360514e-04 0.0260380002 2.925400e-03 #> 27 0.020117510 7.463161e-04 0.0149844191 4.404005e-04 0.0009088445 9.533741e-08 #> 23 0.007198701 2.268512e-04 0.0054844061 5.818798e-03 0.3286900860 1.931899e-03 #> 19 0.003451467 7.018956e-05 0.0003869197 4.013694e-04 0.0013438101 2.875078e-03 #> Cladcerv Claddefo Cladphyl #> 18 2.696135e-04 3.521639e-03 0.0008164040 #> 15 1.295779e-04 1.583784e-02 0.0001887477 #> 24 3.869397e-02 2.535317e-04 0.0493852158 #> 27 9.528504e-05 6.579101e-02 0.0140785723 #> 23 7.797221e-05 1.817802e-01 0.0046131462 #> 19 9.905729e-06 9.587131e-05 0.0003866237 ## Influence measures from lm y <- decostand(varespec, \"chi.square\") # needed in cca y1 <- with(y, Cladstel) # take one species for lm lmod1 <- lm(y1 ~ Al + P + K, varechem, weights = rowSums(varespec)) ## numerically identical within 2e-15 all(abs(cooks.distance(lmod1) - cooks.distance(mod)[, \"Cladstel\"]) < 1e-8) #> [1] TRUE ## t-values of regression coefficients based on type = \"canoco\" ## residuals coef(mod) #> CCA1 CCA2 CCA3 #> Al 0.007478556 -0.001883637 0.003380774 #> P -0.006491081 -0.102189737 -0.022306682 #> K -0.006755568 0.015343662 0.017067351 coef(mod)/sqrt(diag(vcov(mod, type = \"canoco\"))) #> CCA1 CCA2 CCA3 #> Al 6.5615451 -1.397643 3.313629 #> P -0.4576132 -6.092557 -1.756774 #> K -2.0862129 4.007159 5.887926"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":null,"dir":"Reference","previous_headings":"","what":"Isometric Feature Mapping Ordination — isomap","title":"Isometric Feature Mapping Ordination — isomap","text":"function performs isometric feature mapping consists three simple steps: (1) retain shortest dissimilarities among objects, (2) estimate dissimilarities shortest path distances, (3) perform metric scaling (Tenenbaum et al. 2000).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"isomap(dist, ndim=10, ...) isomapdist(dist, epsilon, k, path = \"shortest\", fragmentedOK =FALSE, ...) # S3 method for isomap summary(object, ...) # S3 method for isomap plot(x, net = TRUE, n.col = \"gray\", type = \"points\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Isometric Feature Mapping Ordination — isomap","text":"dist Dissimilarities. ndim Number axes metric scaling (argument k cmdscale). epsilon Shortest dissimilarity retained. k Number shortest dissimilarities retained point. epsilon k given, epsilon used. path Method used stepacross estimate shortest path, alternatives \"shortest\" \"extended\". fragmentedOK dissimilarity matrix fragmented. TRUE, analyse largest connected group, otherwise stop error. x, object isomap result object. net Draw net retained dissimilarities. n.col Colour drawn net segments. can also vector recycled points, colour net segment mixture joined points. type Plot observations either \"points\", \"text\" use \"none\" plot observations. \"text\" use ordilabel net = TRUE ordiplot net = FALSE, pass extra arguments functions. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Isometric Feature Mapping Ordination — isomap","text":"function isomap first calls function isomapdist dissimilarity transformation, performs metric scaling result. arguments isomap passed isomapdist. functions separate isompadist transformation easily used functions simple linear mapping cmdscale. Function isomapdist retains either dissimilarities equal shorter epsilon, epsilon given, least k shortest dissimilarities point. complete dissimilarity matrix reconstructed using stepacross using either flexible shortest paths extended dissimilarities (details, see stepacross). De'ath (1999) actually published essentially method Tenenbaum et al. (2000), De'ath's function available function xdiss non-CRAN package mvpart. differences isomap introduced k criterion, whereas De'ath used epsilon criterion. practice, De'ath also retains higher proportion dissimilarities typical isomap. plot function uses internally ordiplot, except adds text net using ordilabel. plot function passes extra arguments functions. addition, vegan3d package function rgl.isomap make dynamic 3D plots can rotated screen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Isometric Feature Mapping Ordination — isomap","text":"Function isomapdist returns dissimilarity object similar dist. Function isomap returns object class isomap plot summary methods. plot function returns invisibly object class ordiplot. Function scores can extract ordination scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Isometric Feature Mapping Ordination — isomap","text":"De'ath, G. (1999) Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecology 144, 191--199 Tenenbaum, J.B., de Silva, V. & Langford, J.C. (2000) global network framework nonlinear dimensionality reduction. Science 290, 2319--2323.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Isometric Feature Mapping Ordination — isomap","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Isometric Feature Mapping Ordination — isomap","text":"Tenenbaum et al. (2000) justify isomap tool unfolding manifold (e.g. 'Swiss Roll'). Even manifold structure, sampling must even dense dissimilarities along manifold shorter across folds. data manifold structure, results sensitive parameter values.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"## The following examples also overlay minimum spanning tree to ## the graphics in red. op <- par(mar=c(4,4,1,1)+0.2, mfrow=c(2,2)) data(BCI) dis <- vegdist(BCI) tr <- spantree(dis) pl <- ordiplot(cmdscale(dis), main=\"cmdscale\") #> species scores not available lines(tr, pl, col=\"red\") ord <- isomap(dis, k=3) ord #> #> Isometric Feature Mapping (isomap) #> #> Call: #> isomap(dist = dis, k = 3) #> #> Distance method: bray shortest isomap #> Criterion: k = 3 pl <- plot(ord, main=\"isomap k=3\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, k=5), main=\"isomap k=5\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, epsilon=0.45), main=\"isomap epsilon=0.45\") lines(tr, pl, col=\"red\") par(op) ## colour points and web by the dominant species dom <- apply(BCI, 1, which.max) ## need nine colours, but default palette has only eight op <- palette(c(palette(\"default\"), \"sienna\")) plot(ord, pch = 16, col = dom, n.col = dom) palette(op)"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":null,"dir":"Reference","previous_headings":"","what":"Kendall coefficient of concordance — kendall.global","title":"Kendall coefficient of concordance — kendall.global","text":"Function kendall.global computes tests coefficient concordance among several judges (variables, species) permutation test. Function kendall.post carries posteriori tests contributions individual judges (variables, species) overall concordance group permutation tests. several groups judges identified data table, coefficients concordance (kendall.global) posteriori tests (kendall.post) computed group separately. Use ecology: identify significant species associations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"kendall.global(Y, group, nperm = 999, mult = \"holm\") kendall.post(Y, group, nperm = 999, mult = \"holm\")"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kendall coefficient of concordance — kendall.global","text":"Y Data file (data frame matrix) containing quantitative semiquantitative data. Rows objects columns judges (variables). community ecology, table often site--species table. group vector defining judges divided groups. See example . groups explicitly defined, judges data file considered forming single group. nperm Number permutations performed. Default 999. mult Correct P-values multiple testing using alternatives described p.adjust addition \"sidak\" (see Details). Bonferroni correction overly conservative; recommended. included allow comparisons methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Kendall coefficient of concordance — kendall.global","text":"Y must contain quantitative data. transformed ranks within column computation coefficient concordance. search species associations described Legendre (2005) proceeds 3 steps: (1) Correlation analysis species. possible method compute Ward's agglomerative clustering matrix correlations among species. detail: (1.1) compute Pearson Spearman correlation matrix (correl.matrix) among species; (1.2) turn distance matrix: mat.D = .dist(1-correl.matrix); (1.3) carry Ward's hierarchical clustering matrix using hclust: clust.ward = hclust(mat.D, \"ward\"); (1.4) plot dendrogram: plot(clust.ward, hang=-1); (1.5) cut dendrogram two groups, retrieve vector species membership: group.2 = cutree(clust.ward, k=2). (1.6) steps 2 3 , may come back try divisions species k = \\(3, 4, 5, \\dots\\) groups. (2) Compute global tests significance 2 () groups using function kendall.global vector defining groups. Groups globally significant must refined abandoned. (3) Compute posteriori tests contribution individual species concordance group using function kendall.post vector defining groups. species negative values \"Spearman.mean\", means species clearly belong group, hence group inclusive. Go back (1.5) cut dendrogram finely. left right groups can cut separately, independently levels along dendrogram; write vector group membership cutree produce desired groups. corrections used multiple testing applied list P-values (P); take account number tests (k) carried simultaneously (number groups kendall.global, number species kendall.post). corrections performed using function p.adjust; see function description correction methods. addition, Šidák correction defined \\(P_{corr} = 1 -(1 - P)^k\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Kendall coefficient of concordance — kendall.global","text":"table containing following information rows. columns correspond groups \"judges\" defined vector \"group\". function Kendall.post used, many tables number predefined groups. W Kendall's coefficient concordance, W. F F statistic. F = W*(m-1)/(1-W) m number judges. Prob.F Probability associated F statistic, computed F distribution nu1 = n-1-(2/m) nu2 = nu1*(m-1); n number objects. Corrected prob.F Probabilities associated F, corrected using method selected parameter mult. Shown one group. Chi2 Friedman's chi-square statistic (Friedman 1937) used permutation test W. Prob.perm Permutational probabilities, uncorrected. Corrected prob.perm Permutational probabilities corrected using method selected parameter mult. Shown one group. Spearman.mean Mean Spearman correlations judge test judges group. W.per.species Contribution judge test overall concordance statistic group.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Kendall coefficient of concordance — kendall.global","text":"Friedman, M. 1937. use ranks avoid assumption normality implicit analysis variance. Journal American Statistical Association 32: 675-701. Kendall, M. G. B. Babington Smith. 1939. problem m rankings. Annals Mathematical Statistics 10: 275-287. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245. Legendre, P. 2009. Coefficient concordance. : Encyclopedia Research Design. SAGE Publications (press). Siegel, S. N. J. Castellan, Jr. 1988. Nonparametric statistics behavioral sciences. 2nd edition. McGraw-Hill, New York.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Kendall coefficient of concordance — kendall.global","text":"F. Guillaume Blanchet, University Alberta, Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"data(mite) mite.hel <- decostand(mite, \"hel\") # Reproduce the results shown in Table 2 of Legendre (2005), a single group mite.small <- mite.hel[c(4,9,14,22,31,34,45,53,61,69),c(13:15,23)] kendall.global(mite.small, nperm=49) #> $Concordance_analysis #> Group.1 #> W 0.44160305 #> F 2.37252221 #> Prob.F 0.04403791 #> Chi2 15.89770992 #> Prob.perm 0.12000000 #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.small, mult=\"holm\", nperm=49) #> $A_posteriori_tests #> TVEL ONOV SUCT Trhypch1 #> Spearman.mean 0.3265678 0.3965503 0.4570402 -0.1681251 #> W.per.species 0.4949258 0.5474127 0.5927802 0.1239061 #> Prob 0.0800000 0.0200000 0.0200000 0.7200000 #> Corrected prob 0.1600000 0.0800000 0.0800000 0.7200000 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # Reproduce the results shown in Tables 3 and 4 of Legendre (2005), 2 groups group <-c(1,1,2,1,1,1,1,1,2,1,1,1,1,1,1,2,1,2,1,1,1,1,2,1,2,1,1,1,1,1,2,2,2,2,2) kendall.global(mite.hel, group=group, nperm=49) #> $Concordance_analysis #> Group.1 Group.2 #> W 3.097870e-01 2.911888e-01 #> F 1.032305e+01 4.108130e+00 #> Prob.F 1.177138e-85 4.676566e-22 #> Corrected prob.F 2.354275e-85 4.676566e-22 #> Chi2 5.130073e+02 2.210123e+02 #> Prob.perm 2.000000e-02 2.000000e-02 #> Corrected prob.perm 4.000000e-02 4.000000e-02 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.hel, group=group, mult=\"holm\", nperm=49) #> $A_posteriori_tests_Group #> $A_posteriori_tests_Group[[1]] #> Brachy PHTH RARD SSTR Protopl MEGR #> Spearman.mean 0.1851177 0.4258111 0.359058 0.2505486 0.1802160 0.2833298 #> W.per.species 0.2190711 0.4497357 0.385764 0.2817757 0.2143736 0.3131911 #> Prob 0.0200000 0.0200000 0.020000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 #> MPRO HMIN HMIN2 NPRA TVEL ONOV #> Spearman.mean 0.09248024 0.2444656 0.4138494 0.1263751 0.4177343 0.3301159 #> W.per.species 0.13029357 0.2759462 0.4382723 0.1627761 0.4419954 0.3580278 #> Prob 0.14000000 0.0200000 0.0200000 0.0400000 0.0200000 0.0200000 #> Corrected prob 0.70000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> SUCT Oribatl1 PWIL Galumna1 Stgncrs2 HRUF #> Spearman.mean 0.2185421 0.421216 0.2574779 0.4180699 0.3623428 0.1250230 #> W.per.species 0.2511028 0.445332 0.2884163 0.4423170 0.3889118 0.1614804 #> Prob 0.0200000 0.020000 0.0200000 0.0200000 0.0200000 0.0600000 #> Corrected prob 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 0.7000000 #> PPEL SLAT FSET Lepidzts Eupelops Miniglmn #> Spearman.mean 0.2188216 0.3016159 0.4217606 0.2577037 0.1108022 0.2301430 #> W.per.species 0.2513707 0.3307153 0.4458539 0.2886327 0.1478521 0.2622203 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0600000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> $A_posteriori_tests_Group[[2]] #> HPAV TVIE LCIL Ceratoz1 Trhypch1 NCOR #> Spearman.mean 0.1222579 0.2712078 0.1906408 0.1375601 0.1342409 0.3342345 #> W.per.species 0.2020527 0.3374616 0.2642189 0.2159637 0.2129463 0.3947586 #> Prob 0.0400000 0.0200000 0.0200000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> LRUG PLAG2 Ceratoz3 Oppiminu Trimalc2 #> Spearman.mean 0.3446561 0.1833099 0.3188922 0.1764232 0.2498877 #> W.per.species 0.4042328 0.2575544 0.3808111 0.2512938 0.3180797 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Function linestack plots vertical one-dimensional plots numeric vectors. plots always labelled, labels moved vertically avoid overwriting.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"linestack(x, labels, cex = 0.8, side = \"right\", hoff = 2, air = 1.1, at = 0, add = FALSE, axis = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"x Numeric vector plotted. labels Labels used instead default (names x). May expressions drawn plotmath. cex Size labels. side Put labels \"right\" \"left\" axis. hoff Distance vertical axis label units width letter “m”. air Multiplier string height leave empty space labels. Position plot horizontal axis. add Add existing plot. axis Add axis plot. ... graphical parameters labels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function returns invisibly shifted positions labels user coordinates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Jari Oksanen modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function always draws labelled diagrams. want unlabelled diagrams, can use, e.g., plot, stripchart rug.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"## First DCA axis data(dune) ord <- decorana(dune) linestack(scores(ord, choices=1, display=\"sp\")) linestack(scores(ord, choices=1, display=\"si\"), side=\"left\", add=TRUE) title(main=\"DCA axis 1\") ## Expressions as labels N <- 10 # Number of sites df <- data.frame(Ca = rlnorm(N, 2), NO3 = rlnorm(N, 4), SO4 = rlnorm(N, 10), K = rlnorm(N, 3)) ord <- rda(df, scale = TRUE) ### vector of expressions for labels labs <- expression(Ca^{2+phantom()}, NO[3]^{-phantom()}, SO[4]^{2-phantom()}, K^{+phantom()}) scl <- \"sites\" linestack(scores(ord, choices = 1, display = \"species\", scaling = scl), labels = labs, air = 2) linestack(scores(ord, choices = 1, display = \"site\", scaling = scl), side = \"left\", add = TRUE) title(main = \"PCA axis 1\")"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":null,"dir":"Reference","previous_headings":"","what":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"standard CEP name four first letters generic name four first letters specific epithet Latin name. last epithet, may subspecific name, used current function. name one component, abbreviated eight characters (see abbreviate). returned names made unique function make.unique adds numbers end CEP names needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(names, seconditem = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"names names formatted CEP names. seconditem Take always second item original name abbreviated name instead last original item (default).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Cornell Ecology Programs (CEP) used eight-letter abbreviations species site names. species, names formed taking four first letters generic name four first letters specific subspecific epithet. current function first makes valid R names using make.names, splits elements. CEP name made taking four first letters first element, four first letters last (default) second element ( seconditem = TRUE). one name element, abbreviated eight letters. Finally, names made unique may add numbers duplicated names. CEP names originally used, old FORTRAN IV CHARACTER data type, text stored numerical variables, popular computers hold four characters. modern times, reason limitation, ecologists used names, may practical avoid congestion ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Function returns CEP names.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"function simpleminded rigid. must write better one need.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(c(\"Aa maderoi\", \"Poa sp.\", \"Cladina rangiferina\", \"Cladonia cornuta\", \"Cladonia cornuta var. groenlandica\", \"Cladonia rangiformis\", \"Bryoerythrophyllum\")) #> [1] \"Aamade\" \"Poasp\" \"Cladrang\" \"Cladcorn\" \"Cladgroe\" #> [6] \"Cladrang.1\" \"Bryrythr\" data(BCI) colnames(BCI) <- make.cepnames(colnames(BCI))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel Correlogram — mantel.correlog","title":"Mantel Correlogram — mantel.correlog","text":"Function mantel.correlog computes multivariate Mantel correlogram. Proposed Sokal (1986) Oden Sokal (1986), method also described Legendre Legendre (2012, pp. 819--821).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel Correlogram — mantel.correlog","text":"","code":"mantel.correlog(D.eco, D.geo=NULL, XY=NULL, n.class=0, break.pts=NULL, cutoff=TRUE, r.type=\"pearson\", nperm=999, mult=\"holm\", progressive=TRUE) # S3 method for mantel.correlog plot(x, alpha=0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel Correlogram — mantel.correlog","text":"D.eco ecological distance matrix, class either dist matrix. D.geo geographic distance matrix, class either dist matrix. Provide either D.geo XY. Default: D.geo=NULL. XY file Cartesian geographic coordinates points. Default: XY=NULL. n.class Number classes. n.class=0, Sturges equation used unless break points provided. break.pts Vector containing break points distance distribution. Provide (n.class+1) breakpoints, , list beginning ending point. Default: break.pts=NULL. cutoff second half distance classes, cutoff = TRUE limits correlogram distance classes include points. cutoff = FALSE, correlogram includes distance classes. r.type Type correlation calculation Mantel statistic. Default: r.type=\"pearson\". choices r.type=\"spearman\" r.type=\"kendall\", functions cor mantel. nperm Number permutations tests significance. Default: nperm=999. large data files, permutation tests rather slow. mult Correct P-values multiple testing. correction methods \"holm\" (default), \"hochberg\", \"sidak\", methods available p.adjust function: \"bonferroni\" (best known, recommended overly conservative), \"hommel\", \"BH\", \"\", \"fdr\", \"none\". progressive Default: progressive=TRUE progressive correction multiple-testing, described Legendre Legendre (1998, p. 721). Test first distance class: correction; second distance class: correct 2 simultaneous tests; distance class k: correct k simultaneous tests. progressive=FALSE: correct tests n.class simultaneous tests. x Output mantel.correlog. alpha Significance level points drawn black symbols correlogram. Default: alpha=0.05. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel Correlogram — mantel.correlog","text":"correlogram graph spatial correlation values plotted, ordinate, function geographic distance classes among study sites along abscissa. Mantel correlogram, Mantel correlation (Mantel 1967) computed multivariate (e.g. multi-species) distance matrix user's choice design matrix representing geographic distance classes turn. Mantel statistic tested permutational Mantel test performed vegan's mantel function. correction multiple testing applied, permutations necessary -correction case, obtain significant p-values higher correlogram classes. print.mantel.correlog function prints correlogram. See examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel Correlogram — mantel.correlog","text":"mantel.res table distance classes rows class indices, number distances per class, Mantel statistics (computed using Pearson's r, Spearman's r, Kendall's tau), p-values columns. positive Mantel statistic indicates positive spatial correlation. additional column p-values corrected multiple testing added unless mult=\"none\". n.class n umber distance classes. break.pts break points provided user computed program. mult name correction multiple testing. correction: mult=\"none\". progressive logical (TRUE, FALSE) value indicating whether progressive correction multiple testing requested. n.tests number distance classes Mantel tests computed tested significance. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel Correlogram — mantel.correlog","text":"Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel Correlogram — mantel.correlog","text":"Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Mantel, N. 1967. detection disease clustering generalized regression approach. Cancer Res. 27: 209-220. Oden, N. L. R. R. Sokal. 1986. Directional autocorrelation: extension spatial correlograms two dimensions. Syst. Zool. 35: 608-617. Sokal, R. R. 1986. Spatial data analysis historical processes. 29-43 : E. Diday et al. [eds.] Data analysis informatics, IV. North-Holland, Amsterdam. Sturges, H. . 1926. choice class interval. Journal American Statistical Association 21: 65–66.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel Correlogram — mantel.correlog","text":"","code":"# Mite data available in \"vegan\" data(mite) data(mite.xy) mite.hel <- decostand(mite, \"hellinger\") # Detrend the species data by regression on the site coordinates mite.hel.resid <- resid(lm(as.matrix(mite.hel) ~ ., data=mite.xy)) # Compute the detrended species distance matrix mite.hel.D <- dist(mite.hel.resid) # Compute Mantel correlogram with cutoff, Pearson statistic mite.correlog <- mantel.correlog(mite.hel.D, XY=mite.xy, nperm=49) summary(mite.correlog) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 4 -none- call mite.correlog #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.135713 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.118174 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.037820 0.04 0.06 . #> D.cl.4 2.699274 696.000000 -0.098605 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.112682 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.107603 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.022264 0.12 0.14 #> D.cl.8 5.612730 326.000000 NA NA NA #> D.cl.9 6.341094 260.000000 NA NA NA #> D.cl.10 7.069458 184.000000 NA NA NA #> D.cl.11 7.797822 130.000000 NA NA NA #> D.cl.12 8.526186 66.000000 NA NA NA #> D.cl.13 9.254550 32.000000 NA NA NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # or: print(mite.correlog) # or: print.mantel.correlog(mite.correlog) plot(mite.correlog) # Compute Mantel correlogram without cutoff, Spearman statistic mite.correlog2 <- mantel.correlog(mite.hel.D, XY=mite.xy, cutoff=FALSE, r.type=\"spearman\", nperm=49) summary(mite.correlog2) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 6 -none- call mite.correlog2 #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, cutoff = FALSE, r.type = \"spearman\", nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.134229 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.121270 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.035413 0.02 0.06 . #> D.cl.4 2.699274 696.000000 -0.095899 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.118692 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.117148 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.031123 0.08 0.14 #> D.cl.8 5.612730 326.000000 0.026064 0.14 0.16 #> D.cl.9 6.341094 260.000000 0.050573 0.08 0.24 #> D.cl.10 7.069458 184.000000 0.057017 0.04 0.20 #> D.cl.11 7.797822 130.000000 0.036195 0.10 0.32 #> D.cl.12 8.526186 66.000000 -0.054242 0.02 0.24 #> D.cl.13 9.254550 32.000000 -0.066677 0.04 0.26 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mite.correlog2) # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Function mantel finds Mantel statistic matrix correlation two dissimilarity matrices, function mantel.partial finds partial Mantel statistic partial matrix correlation three dissimilarity matrices. significance statistic evaluated permuting rows columns first dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"mantel(xdis, ydis, method=\"pearson\", permutations=999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\")) mantel.partial(xdis, ydis, zdis, method = \"pearson\", permutations = 999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"xdis, ydis, zdis Dissimilarity matrices ordist objects. first object xdis permuted permutation tests. method Correlation method, accepted cor: \"pearson\", \"spearman\" \"kendall\". permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. na.rm Remove missing values calculation Mantel correlation. Use option care: Permutation tests can biased, particular two matrices missing values matching positions. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Mantel statistic simply correlation entries two dissimilarity matrices (use cross products, linearly related). However, significance directly assessed, \\(N(N-1)/2\\) entries just \\(N\\) observations. Mantel developed asymptotic test, use permutations \\(N\\) rows columns dissimilarity matrix. first matrix (xdist) permuted, second kept constant. See permutations additional details permutation tests Vegan. Partial Mantel statistic uses partial correlation conditioned third matrix. first matrix permuted correlation structure second first matrices kept constant. Although mantel.partial silently accepts methods \"pearson\", partial correlations probably wrong methods. function uses cor, accept alternatives pearson product moment correlations spearman kendall rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"function returns list class mantel following components: Call Function call. method Correlation method used, returned cor.test. statistic Mantel statistic. signif Empirical significance level permutations. perm vector permuted values. distribution permuted values can inspected permustats function. permutations Number permutations. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"test due Mantel, course, current implementation based Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English Edition. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Legendre & Legendre (2012, Box 10.4) warn using partial Mantel correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"## Is vegetation related to environment? data(varespec) data(varechem) veg.dist <- vegdist(varespec) # Bray-Curtis env.dist <- vegdist(scale(varechem), \"euclid\") mantel(veg.dist, env.dist) #> #> Mantel statistic based on Pearson's product-moment correlation #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist) #> #> Mantel statistic r: 0.3047 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.116 0.147 0.176 0.210 #> Permutation: free #> Number of permutations: 999 #> mantel(veg.dist, env.dist, method=\"spear\") #> #> Mantel statistic based on Spearman's rank correlation rho #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist, method = \"spear\") #> #> Mantel statistic r: 0.2838 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.123 0.162 0.182 0.208 #> Permutation: free #> Number of permutations: 999 #>"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS performs Nonmetric Multidimensional Scaling (NMDS), tries find stable solution using several random starts. addition, standardizes scaling result, configurations easier interpret, adds species scores site ordination. metaMDS function provide actual NMDS, calls another function purpose. Currently monoMDS default choice, also possible call isoMDS (MASS package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"metaMDS(comm, distance = \"bray\", k = 2, try = 20, trymax = 20, engine = c(\"monoMDS\", \"isoMDS\"), autotransform =TRUE, noshare = (engine == \"isoMDS\"), wascores = TRUE, expand = TRUE, trace = 1, plot = FALSE, previous.best, ...) # S3 method for metaMDS plot(x, display = c(\"sites\", \"species\"), choices = c(1, 2), type = \"p\", shrink = FALSE, ...) # S3 method for metaMDS points(x, display = c(\"sites\", \"species\"), choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS text(x, display = c(\"sites\", \"species\"), labels, choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS scores(x, display = c(\"sites\", \"species\"), shrink = FALSE, choices, tidy = FALSE, ...) metaMDSdist(comm, distance = \"bray\", autotransform = TRUE, noshare = TRUE, trace = 1, commname, zerodist = \"ignore\", distfun = vegdist, ...) metaMDSiter(dist, k = 2, try = 20, trymax = 20, trace = 1, plot = FALSE, previous.best, engine = \"monoMDS\", maxit = 200, parallel = getOption(\"mc.cores\"), ...) initMDS(x, k=2) postMDS(X, dist, pc=TRUE, center=TRUE, halfchange, threshold=0.8, nthreshold=10, plot=FALSE, ...) metaMDSredist(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"comm Community data. Alternatively, dissimilarities either dist structure symmetric square matrix. latter case stages skipped except random starts centring pc rotation axes. distance Dissimilarity index used vegdist. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher global non-metric MDS, still higher local NMDS. try, trymax Minimum maximum numbers random starts search stable solution. try reached, iteration stop similar solutions repeated trymax reached. engine function used MDS. default use monoMDS function vegan, backward compatibility also possible use isoMDS MASS. autotransform Use simple heuristics possible data transformation typical community data (see ). community data, probably set autotransform = FALSE. noshare Triggering calculation step-across extended dissimilarities function stepacross. argument can logical numerical value greater zero less one. TRUE, extended dissimilarities used always shared species sites, FALSE, never used. noshare numerical value, stepacross used proportion site pairs shared species exceeds noshare. number pairs shared species found .shared function, noshare effect input data dissimilarities instead community data. wascores Calculate species scores using function wascores. expand Expand weighted averages species wascores. trace Trace function; trace = 2 higher voluminous. plot Graphical tracing: plot interim results. may want set par(ask = TRUE) option. previous.best Start searches previous solution. x metaMDS result (dissimilarity structure initMDS). choices Axes shown. type Plot type: \"p\" points, \"t\" text, \"n\" axes . display Display \"sites\" \"species\". shrink Shrink back species scores expanded originally. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable code (\"sites\" \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Optional test used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. X Configuration multidimensional scaling. commname name comm: given function called directly. zerodist Handling zero dissimilarities: either \"fail\" \"add\" small positive value, \"ignore\". monoMDS accepts zero dissimilarities default zerodist = \"ignore\", isoMDS may need set zerodist = \"add\". distfun Dissimilarity function. function returning dist object accepting argument method can used (extra arguments may cause name conflicts). maxit Maximum number iterations single NMDS run; passed engine function monoMDS isoMDS. parallel Number parallel processes predefined socket cluster. use pre-defined socket clusters (say, clus), must issue clusterEvalQ(clus, library(vegan)) make available internal vegan functions. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist Dissimilarity matrix used multidimensional scaling. pc Rotate principal components. center Centre configuration. halfchange Scale axes half-change units. defaults TRUE dissimilarities known theoretical maximum value (ceiling). Function vegdist information attribute maxdist, distfun interpreted simple test (can fail), information may available input data distances. FALSE, ordination dissimilarities scaled range input dissimilarities. threshold Largest dissimilarity used half-change scaling. dissimilarities known (inferred) ceiling, threshold relative ceiling (see halfchange). nthreshold Minimum number points half-change scaling. object result object metaMDS. ... parameters passed functions. Function metaMDS passes arguments component functions metaMDSdist, metaMDSiter, postMDS, distfun engine.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-metric Multidimensional Scaling (NMDS) commonly regarded robust unconstrained ordination method community ecology (Minchin 1987). Function metaMDS wrapper function calls several functions combine Minchin's (1987) recommendations one command. complete steps metaMDS : Transformation: data values larger common abundance class scales, function performs Wisconsin double standardization (wisconsin). values look large, function also performs sqrt transformation. standardizations generally found improve results. However, limits completely arbitrary (present, data maximum 50 triggers sqrt \\(>9\\) triggers wisconsin). want full control analysis, set autotransform = FALSE standardize transform data independently. autotransform intended community data, data types, set autotransform = FALSE. step perfomed using metaMDSdist, step skipped input dissimilarities. Choice dissimilarity: good result, use dissimilarity indices good rank order relation ordering sites along gradients (Faith et al. 1987). default Bray-Curtis dissimilarity, often test winner. However, dissimilarity index vegdist can used. Function rankindex can used finding test winner data gradients. default choice may bad analyse community data, probably select appropriate index using argument distance. step performed using metaMDSdist, step skipped input dissimilarities. Step-across dissimilarities: Ordination may difficult large proportion sites shared species. case, results may improved stepacross dissimilarities, flexible shortest paths among sites. default NMDS engine monoMDS able break tied values maximum dissimilarity, often sufficient handle cases shared species, therefore default use stepacross monoMDS. Function isoMDS handle tied values adequately, therefore default use stepacross always sites shared species engine = \"isoMDS\". stepacross triggered option noshare. like manipulation original distances, set noshare = FALSE. step skipped input data dissimilarities instead community data. step performed using metaMDSdist, step skipped always input dissimilarities. NMDS random starts: NMDS easily gets trapped local optima, must start NMDS several times random starts confident found global solution. strategy metaMDS first run NMDS starting metric scaling (cmdscale usually finds good solution often close local optimum), use previous.best solution supplied, take solution standard (Run 0). metaMDS starts NMDS several random starts (minimum number given try maximum number trymax). random starts generated initMDS. solution better (lower stress) previous standard, taken new standard. solution better close standard, metaMDS compares two solutions using Procrustes analysis (function procrustes option symmetric = TRUE). solutions similar Procrustes rmse largest residual small, solutions regarded repeated better one taken new standard. conditions stringent, may found good relatively similar solutions although function yet satisfied. Setting trace = TRUE monitor final stresses, plot = TRUE display Procrustes overlay plots comparison. step performed using metaMDSiter. first step performed input data (comm) dissimilarities. Random starts can run parallel processing (argument parallel). Scaling results: metaMDS run postMDS final result. Function postMDS provides following ways “fixing” indeterminacy scaling orientation axes NMDS: Centring moves origin average axes; Principal components rotate configuration variance points maximized first dimension (function MDSrotate can alternatively rotate configuration first axis parallel environmental variable); Half-change scaling scales configuration one unit means halving community similarity replicate similarity. Half-change scaling based closer dissimilarities relation ordination distance community dissimilarity rather linear (limit set argument threshold). enough points threshold (controlled parameter nthreshold), dissimilarities regressed distances. intercept regression taken replicate dissimilarity, half-change distance similarity halves according linear regression. Obviously method applicable dissimilarity indices scaled \\(0 \\ldots 1\\), Kulczynski, Bray-Curtis Canberra indices. half-change scaling used, ordination scaled range original dissimilarities. Half-change scaling skipped default input dissimilarities, can turned argument halfchange = TRUE. NB., PC rotation changes directions reference axes, influence configuration solution general. Species scores: Function adds species scores final solution weighted averages using function wascores given value parameter expand. expansion weighted averages can undone shrink = TRUE plot scores functions, calculation species scores can suppressed wascores = FALSE. step skipped input dissimilarities community data unavailable. However, species scores can added replaced sppscores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"results-could-not-be-repeated","dir":"Reference","previous_headings":"","what":"Results Could Not Be Repeated","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-linear optimization hard task, best possible solution (“global optimum”) may found random starting configuration. software solve starting result metric scaling (cmdscale). probably give good result, necessarily “global optimum”. Vegan , metaMDS tries verify improve first solution (“try 0”) using several random starts seeing result can repeated improved improved solution repeated. succeed, get message result repeated. However, result least good usual standard strategy starting metric scaling may improved. may need anything message, can satisfied result. want sure probably “global optimum” may try following instructions. default engine = \"monoMDS\" function tabulate stopping criteria used, can see criterion made stringent. criteria can given arguments metaMDS current values described monoMDS. particular, reach maximum number iterations, increase value maxit. may ask larger number random starts without losing old ones giving previous solution argument previous.best. addition slack convergence criteria low number random starts, wrong number dimensions (argument k) common reason able repeat similar solutions. NMDS usually run low number dimensions (k=2 k=3), complex data increasing k one may help. run NMDS much higher number dimensions (say, k=10 ), reconsider drastically reduce k. heterogeneous data sets partial disjunctions, may help set stepacross, data sets default weakties = TRUE sufficient. Please note can give arguments metaMDS* functions NMDS engine (default monoMDS) metaMDS command, check documentation functions details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"common-wrong-claims","dir":"Reference","previous_headings":"","what":"Common Wrong Claims","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"NMDS often misunderstood wrong claims properties common Web even publications. often claimed NMDS configuration non-metric means fit environmental variables species onto space. false statement. fact, result configuration NMDS metric, can used like ordination result. NMDS rank orders Euclidean distances among points ordination non-metric monotone relationship observed dissimilarities. transfer function observed dissimilarities ordination distances non-metric (Kruskal 1964a, 1964b), ordination result configuration metric observed dissimilarities can kind (metric non-metric). ordination configuration usually rotated principal components metaMDS. rotation performed finding result, changes direction reference axes. important feature NMDS solution ordination distances, change rotation. Similarly, rank order distances change uniform scaling centring configuration points. can also rotate NMDS solution external environmental variables MDSrotate. rotation also change orientation axes, change configuration points distances points ordination space. Function stressplot displays method graphically: plots observed dissimilarities distances ordination space, also shows non-metric monotone regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS returns object class metaMDS. final site ordination stored item points, species ordination item species, stress item stress (NB, scaling stress depends engine: isoMDS uses percents, monoMDS proportions range \\(0 \\ldots 1\\)). items store information steps taken items returned engine function. object print, plot, points text methods. Functions metaMDSdist metaMDSredist return vegdist objects. Function initMDS returns random configuration intended used within isoMDS . Functions metaMDSiter postMDS returns result NMDS updated configuration.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Kruskal, J.B. (1964a). Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. (1964b). Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. (1987). evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS simple wrapper NMDS engine (either monoMDS isoMDS) support functions (metaMDSdist, stepacross, metaMDSiter, initMDS, postMDS, wascores). can call support functions separately better control results. Data transformation, dissimilarities possible stepacross made function metaMDSdist returns dissimilarity result. Iterative search ( starting values initMDS monoMDS) made metaMDSiter. Processing result configuration done postMDS, species scores added wascores. want certain reaching global solution, can compare results several independent runs. can also continue analysis previous results configuration. Function may save used dissimilarity matrix (monoMDS ), metaMDSredist tries reconstruct used dissimilarities original data transformation possible stepacross. metaMDS function designed used community data. type data, probably set arguments non-default values: probably least wascores, autotransform noshare FALSE. negative data entries, metaMDS set previous FALSE warning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"metaMDS uses monoMDS NMDS engine vegan version 2.0-0, replaced isoMDS function. can set argument engine select old engine.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"## The recommended way of running NMDS (Minchin 1987) ## data(dune) ## IGNORE_RDIFF_BEGIN ## Global NMDS using monoMDS sol <- metaMDS(dune) #> Run 0 stress 0.1192678 #> Run 1 stress 0.3016052 #> Run 2 stress 0.1192678 #> ... New best solution #> ... Procrustes: rmse 9.804396e-06 max resid 2.993113e-05 #> ... Similar to previous best #> Run 3 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 0.02027462 max resid 0.06499128 #> Run 4 stress 0.1192679 #> Run 5 stress 0.1183186 #> ... Procrustes: rmse 3.964007e-06 max resid 1.2573e-05 #> ... Similar to previous best #> Run 6 stress 0.1192679 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 6.675204e-06 max resid 1.892118e-05 #> ... Similar to previous best #> Run 8 stress 0.1192678 #> Run 9 stress 0.1183186 #> ... Procrustes: rmse 2.31839e-06 max resid 6.973206e-06 #> ... Similar to previous best #> Run 10 stress 0.1889645 #> Run 11 stress 0.1183186 #> ... Procrustes: rmse 4.550156e-06 max resid 1.4618e-05 #> ... Similar to previous best #> Run 12 stress 0.1192678 #> Run 13 stress 0.1192678 #> Run 14 stress 0.1192678 #> Run 15 stress 0.1192678 #> Run 16 stress 0.1192679 #> Run 17 stress 0.1808912 #> Run 18 stress 0.1183186 #> ... Procrustes: rmse 1.954831e-05 max resid 6.115607e-05 #> ... Similar to previous best #> Run 19 stress 0.1808911 #> Run 20 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 1.3197e-06 max resid 3.321941e-06 #> ... Similar to previous best #> *** Best solution repeated 1 times sol #> #> Call: #> metaMDS(comm = dune) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1183186 #> Stress type 1, weak ties #> Best solution was repeated 1 time in 20 tries #> The best solution was from try 20 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> plot(sol, type=\"t\") ## Start from previous best solution sol <- metaMDS(dune, previous.best = sol) #> Starting from 2-dimensional configuration #> Run 0 stress 0.1183186 #> Run 1 stress 0.1183186 #> ... Procrustes: rmse 5.279158e-06 max resid 1.797511e-05 #> ... Similar to previous best #> Run 2 stress 0.1183186 #> ... Procrustes: rmse 2.460953e-06 max resid 5.877957e-06 #> ... Similar to previous best #> Run 3 stress 0.1192678 #> Run 4 stress 0.1886532 #> Run 5 stress 0.1192678 #> Run 6 stress 0.1808911 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 7.828309e-06 max resid 2.405585e-05 #> ... Similar to previous best #> Run 8 stress 0.1183186 #> ... Procrustes: rmse 6.32253e-06 max resid 2.037397e-05 #> ... Similar to previous best #> Run 9 stress 0.1808911 #> Run 10 stress 0.1183186 #> ... Procrustes: rmse 1.809694e-06 max resid 5.978548e-06 #> ... Similar to previous best #> Run 11 stress 0.1901491 #> Run 12 stress 0.1183186 #> ... Procrustes: rmse 1.546268e-05 max resid 4.957148e-05 #> ... Similar to previous best #> Run 13 stress 0.1812932 #> Run 14 stress 0.1183186 #> ... Procrustes: rmse 5.112711e-06 max resid 1.098817e-05 #> ... Similar to previous best #> Run 15 stress 0.1183186 #> ... Procrustes: rmse 6.246131e-06 max resid 2.000824e-05 #> ... Similar to previous best #> Run 16 stress 0.1192679 #> Run 17 stress 0.1192678 #> Run 18 stress 0.1886532 #> Run 19 stress 0.1192679 #> Run 20 stress 0.1192678 #> *** Best solution repeated 9 times ## Local NMDS and stress 2 of monoMDS sol2 <- metaMDS(dune, model = \"local\", stress=2) #> Run 0 stress 0.1928478 #> Run 1 stress 0.1928482 #> ... Procrustes: rmse 0.0006537695 max resid 0.001899655 #> ... Similar to previous best #> Run 2 stress 0.1928476 #> ... New best solution #> ... Procrustes: rmse 0.0004471422 max resid 0.001294366 #> ... Similar to previous best #> Run 3 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 0.0002885977 max resid 0.0008320155 #> ... Similar to previous best #> Run 4 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 5.59458e-05 max resid 0.0001616763 #> ... Similar to previous best #> Run 5 stress 0.1928475 #> ... Procrustes: rmse 4.864069e-05 max resid 0.0001571056 #> ... Similar to previous best #> Run 6 stress 0.1928477 #> ... Procrustes: rmse 0.0002420844 max resid 0.0006909829 #> ... Similar to previous best #> Run 7 stress 0.1928475 #> ... Procrustes: rmse 0.0001346077 max resid 0.0003887498 #> ... Similar to previous best #> Run 8 stress 0.1928475 #> ... Procrustes: rmse 0.0001625818 max resid 0.0004729866 #> ... Similar to previous best #> Run 9 stress 0.1928475 #> ... Procrustes: rmse 8.000053e-05 max resid 0.0002277773 #> ... Similar to previous best #> Run 10 stress 0.192849 #> ... Procrustes: rmse 0.0006301494 max resid 0.001813666 #> ... Similar to previous best #> Run 11 stress 0.1928475 #> ... Procrustes: rmse 0.0001014964 max resid 0.0002751379 #> ... Similar to previous best #> Run 12 stress 0.192848 #> ... Procrustes: rmse 0.0002964116 max resid 0.0008527077 #> ... Similar to previous best #> Run 13 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 6.461832e-05 max resid 0.0001901839 #> ... Similar to previous best #> Run 14 stress 0.1928482 #> ... Procrustes: rmse 0.0003439501 max resid 0.001017732 #> ... Similar to previous best #> Run 15 stress 0.1928479 #> ... Procrustes: rmse 0.0002821687 max resid 0.0008172892 #> ... Similar to previous best #> Run 16 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 3.13617e-05 max resid 8.909983e-05 #> ... Similar to previous best #> Run 17 stress 0.1928475 #> ... Procrustes: rmse 4.709197e-05 max resid 0.0001417594 #> ... Similar to previous best #> Run 18 stress 0.1928477 #> ... Procrustes: rmse 0.0002414341 max resid 0.0007100825 #> ... Similar to previous best #> Run 19 stress 0.1928479 #> ... Procrustes: rmse 0.0003023992 max resid 0.0008897382 #> ... Similar to previous best #> Run 20 stress 0.1928476 #> ... Procrustes: rmse 0.0002063278 max resid 0.0006048887 #> ... Similar to previous best #> *** Best solution repeated 5 times sol2 #> #> Call: #> metaMDS(comm = dune, model = \"local\", stress = 2) #> #> local Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1928475 #> Stress type 2, weak ties #> Best solution was repeated 5 times in 20 tries #> The best solution was from try 16 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## Use Arrhenius exponent 'z' as a binary dissimilarity measure sol <- metaMDS(dune, distfun = betadiver, distance = \"z\") #> Run 0 stress 0.1067169 #> Run 1 stress 0.1073148 #> Run 2 stress 0.1073148 #> Run 3 stress 0.1073148 #> Run 4 stress 0.1073148 #> Run 5 stress 0.1073148 #> Run 6 stress 0.1067169 #> ... New best solution #> ... Procrustes: rmse 2.925047e-06 max resid 7.049929e-06 #> ... Similar to previous best #> Run 7 stress 0.1069781 #> ... Procrustes: rmse 0.006690411 max resid 0.02349817 #> Run 8 stress 0.1067169 #> ... Procrustes: rmse 6.432458e-06 max resid 1.886945e-05 #> ... Similar to previous best #> Run 9 stress 0.1069786 #> ... Procrustes: rmse 0.006774522 max resid 0.02386629 #> Run 10 stress 0.1067169 #> ... Procrustes: rmse 2.346391e-06 max resid 5.306599e-06 #> ... Similar to previous best #> Run 11 stress 0.1067169 #> ... Procrustes: rmse 2.816635e-06 max resid 7.563776e-06 #> ... Similar to previous best #> Run 12 stress 0.1069784 #> ... Procrustes: rmse 0.006735205 max resid 0.02369698 #> Run 13 stress 0.1067169 #> ... Procrustes: rmse 3.36442e-06 max resid 1.127286e-05 #> ... Similar to previous best #> Run 14 stress 0.2355185 #> Run 15 stress 0.1644741 #> Run 16 stress 0.1067169 #> ... Procrustes: rmse 8.389288e-06 max resid 2.174836e-05 #> ... Similar to previous best #> Run 17 stress 0.1073148 #> Run 18 stress 0.1067169 #> ... Procrustes: rmse 5.169725e-06 max resid 1.633663e-05 #> ... Similar to previous best #> Run 19 stress 0.1067169 #> ... Procrustes: rmse 3.667758e-06 max resid 9.091241e-06 #> ... Similar to previous best #> Run 20 stress 0.1073148 #> *** Best solution repeated 8 times sol #> #> Call: #> metaMDS(comm = dune, distance = \"z\", distfun = betadiver) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: beta.z #> #> Dimensions: 2 #> Stress: 0.1067169 #> Stress type 1, weak ties #> Best solution was repeated 8 times in 20 tries #> The best solution was from try 6 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":null,"dir":"Reference","previous_headings":"","what":"Oribatid Mite Data with Explanatory Variables — mite","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Oribatid mite data. 70 soil cores collected Daniel Borcard 1989. See Borcard et al. (1992, 1994) details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) data(mite.xy)"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"three linked data sets: mite contains data 35 species Oribatid mites, mite.env contains environmental data sampling sites, mite.xy contains geographic coordinates, mite.pcnm contains 22 PCNM base functions (columns) computed geographic coordinates 70 sampling sites (Borcard & Legendre 2002). whole sampling area 2.5 m x 10 m size. fields environmental data : SubsDens Substrate density (g/L) WatrCont Water content substrate (g/L) Substrate Substrate type, factor levels Sphagn1, \tSphagn2 Sphagn3 Sphagn Litter Barepeat Interface Shrub Shrub density, ordered factor levels 1 < 2 < 3 Topo Microtopography, factor levels Blanket Hummock","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Pierre Legendre","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Borcard, D., P. Legendre P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045-1055. Borcard, D. P. Legendre. 1994. Environmental control spatial structure ecological communities: example using Oribatid mites (Acari, Oribatei). Environmental Ecological Statistics 1: 37-61. Borcard, D. P. Legendre. 2002. -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153: 51-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Function implements Kruskal's (1964a,b) non-metric multidimensional scaling (NMDS) using monotone regression primary (“weak”) treatment ties. addition traditional global NMDS, function implements local NMDS, linear hybrid multidimensional scaling.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"monoMDS(dist, y, k = 2, model = c(\"global\", \"local\", \"linear\", \"hybrid\"), threshold = 0.8, maxit = 200, weakties = TRUE, stress = 1, scaling = TRUE, pc = TRUE, smin = 1e-4, sfgrmin = 1e-7, sratmax=0.999999, ...) # S3 method for monoMDS scores(x, choices = NA, ...) # S3 method for monoMDS plot(x, choices = c(1,2), type = \"t\", ...) # S3 method for monoMDS points(x, choices = c(1,2), select, ...) # S3 method for monoMDS text(x, labels, choices = c(1,2), select, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"dist Input dissimilarities. y Starting configuration. random configuration generated missing. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher non-metric MDS. model MDS model: \"global\" normal non-metric MDS monotone regression, \"local\" non-metric MDS separate regressions point, \"linear\" uses linear regression, \"hybrid\" uses linear regression dissimilarities threshold addition monotone regression. See Details. threshold Dissimilarity linear regression used alternately monotone regression. maxit Maximum number iterations. weakties Use primary weak tie treatment, equal observed dissimilarities allowed different fitted values. FALSE, secondary (strong) tie treatment used, tied values broken. stress Use stress type 1 2 (see Details). scaling Scale final scores unit root mean squares. pc Rotate final scores principal components. smin, sfgrmin, sratmax Convergence criteria: iterations stop stress drops smin, scale factor gradient drops sfgrmin, stress ratio two iterations goes sratmax (still \\(< 1\\)). x monoMDS result. choices Dimensions returned plotted. default NA returns dimensions. type type plot: \"t\" text, \"p\" points, \"n\" none. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. labels Labels use used instead row names. ... parameters functions (ignored monoMDS, passed graphical functions plot.).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"several versions non-metric multidimensional scaling R, monoMDS offers following unique combination features: “Weak” treatment ties (Kruskal 1964a,b), tied dissimilarities can broken monotone regression. especially important cases compared sites share species dissimilarities tied maximum value one. Breaking ties allows points different distances can help recovering long coenoclines (gradients). Functions smacof package also hav adequate tie treatment. Handles missing values meaningful way. Offers “local” “hybrid” scaling addition usual “global” NMDS (see ). Uses fast compiled code (isoMDS MASS package also uses compiled code). Function monoMDS uses Kruskal's (1964b) original monotone regression minimize stress. two alternatives stress: Kruskal's (1964a,b) original “stress 1” alternative version “stress 2” (Sibson 1972). stresses can expressed general formula $$s^2 = \\frac{\\sum (d - \\hat d)^2}{\\sum(d - d_0)^2}$$ \\(d\\) distances among points ordination configuration, \\(\\hat d\\) fitted ordination distances, \\(d_0\\) ordination distances null model. “stress 1” \\(d_0 = 0\\), “stress 2” \\(d_0 = \\bar{d}\\) mean distances. “Stress 2” can expressed \\(s^2 = 1 - R^2\\), \\(R^2\\) squared correlation fitted values ordination distances, related “linear fit” stressplot. Function monoMDS can fit several alternative NMDS variants can selected argument model. default model = \"global\" fits global NMDS, Kruskal's (1964a,b) original NMDS similar isoMDS (MASS). Alternative model = \"local\" fits local NMDS independent monotone regression used point (Sibson 1972). Alternative model = \"linear\" fits linear MDS. fits linear regression instead monotone, identical metric scaling principal coordinates analysis (cmdscale) performs eigenvector decomposition dissimilarities (Gower 1966). Alternative model = \"hybrid\" implements hybrid MDS uses monotone regression points linear regression dissimilarities threshold dissimilarity alternating steps (Faith et al. 1987). Function stressplot can used display kind regression model. Scaling, orientation direction axes arbitrary. However, function always centres axes, default scaling scale configuration unit root mean square rotate axes (argument pc) principal components first dimension shows major variation. possible rotate solution first axis parallel given environmental variable using function MDSrotate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"convergence-criteria","dir":"Reference","previous_headings":"","what":"Convergence Criteria","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"NMDS iterative, function stops convergence criteria met. actually criterion assured convergence, solution can local optimum. compare several random starts (use monoMDS via metaMDS) assess solutions likely global optimum. stopping criteria : maxit: Maximum number iterations. Reaching criterion means solutions almost certainly found, maxit increased. smin: Minimum stress. stress nearly zero, fit almost perfect. Usually means data set small requested analysis, may several different solutions almost perfect. reduce number dimensions (k), get data ( observations) use method, metric scaling (cmdscale, wcmdscale). sratmax: Change stress. Values close one mean almost unchanged stress. may mean solution, can also signal stranding suboptimal solution flat stress surface. sfgrmin: Minimum scale factor. Values close zero mean almost unchanged configuration. may mean solution, also happen local optima.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Returns object class \"monoMDS\". final scores returned item points (function scores extracts results), stress item stress. addition, large number items (may change without notice future releases).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Faith, D.P., Minchin, P.R Belbin, L. 1987. Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J.C. (1966). distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Kruskal, J.B. 1964a. Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. 1964b. Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. 1987. evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107. Sibson, R. 1972. Order invariant methods data analysis. Journal Royal Statistical Society B 34, 311--349.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Peter R. Michin (Fortran core) Jari Oksanen (R interface).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"default NMDS function used metaMDS. Function metaMDS adds support functions NMDS can run like recommended Minchin (1987).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"data(dune) dis <- vegdist(dune) m <- monoMDS(dis, model = \"loc\") m #> #> Call: #> monoMDS(dist = dis, model = \"loc\") #> #> Local non-metric Multidimensional Scaling #> #> 20 points, dissimilarity ‘bray’, call ‘vegdist(x = dune)’ #> #> Dimensions: 2 #> Stress: 0.07596745 #> Stress type 1, weak ties #> Scores scaled to unit root mean square, rotated to principal components #> Stopped after 62 iterations: Stress nearly unchanged (ratio > sratmax) plot(m)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":null,"dir":"Reference","previous_headings":"","what":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. Function meandist finds mean within block dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"mrpp(dat, grouping, permutations = 999, distance = \"euclidean\", weight.type = 1, strata = NULL, parallel = getOption(\"mc.cores\")) meandist(dist, grouping, ...) # S3 method for meandist summary(object, ...) # S3 method for meandist plot(x, kind = c(\"dendrogram\", \"histogram\"), cluster = \"average\", ylim, axes = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"dat data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor numeric index grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. used assess significance MRPP statistic, \\(delta\\). distance Choice distance metric measures dissimilarity two observations . See vegdist options. used dat dissimilarity structure symmetric square matrix. weight.type choice group weights. See Details options. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist dist object dissimilarities, produced functions dist, vegdist designdist. . object, x meandist result object. kind Draw dendrogram histogram; see Details. cluster clustering method hclust function kind = \"dendrogram\". hclust method can used, perhaps \"average\" \"single\" make sense. ylim Limits vertical axes (optional). axes Draw scale vertical axis. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. difference may one location (differences mean) one spread (differences within-group distance; cf. Warton et al. 2012). Function mrpp operates data.frame matrix rows observations responses data matrix. response(s) may uni- multivariate. method philosophically mathematically allied analysis variance, compares dissimilarities within among groups. two groups sampling units really different (e.g. species composition), average within-group compositional dissimilarities less average dissimilarities two random collection sampling units drawn entire population. mrpp statistic \\(\\delta\\) overall weighted mean within-group means pairwise dissimilarities among sampling units. choice group weights currently clear. mrpp function offers three choices: (1) group size (\\(n\\)), (2) degrees--freedom analogue (\\(n-1\\)), (3) weight number unique distances calculated among \\(n\\) sampling units (\\(n(n-1)/2\\)). mrpp algorithm first calculates pairwise distances entire dataset, calculates \\(\\delta\\). permutes sampling units associated pairwise distances, recalculates \\(\\delta\\) based permuted data. repeats permutation step permutations times. significance test fraction permuted deltas less observed delta, small sample correction. function also calculates change-corrected within-group agreement \\(= 1 -\\delta/E(\\delta)\\), \\(E(\\delta)\\) expected \\(\\delta\\) assessed average dissimilarities. first argument dat can interpreted dissimilarities, used directly. cases function treats dat observations, uses vegdist find dissimilarities. default distance Euclidean traditional use method, dissimilarities vegdist also available. Function meandist calculates matrix mean within-cluster dissimilarities (diagonal) -cluster dissimilarities (-diagonal elements), attribute n grouping counts. Function summary finds within-class, -class overall means dissimilarities, MRPP statistics weight.type options Classification Strength, CS (Van Sickle Hughes, 2000). CS defined dissimilarities \\(\\bar{B} - \\bar{W}\\), \\(\\bar{B}\\) mean cluster dissimilarity \\(\\bar{W}\\) mean within cluster dissimilarity weight.type = 1. function perform significance tests statistics, must use mrpp appropriate weight.type. currently significance test CS, mrpp weight.type = 1 gives correct test \\(\\bar{W}\\) good approximation CS. Function plot draws dendrogram histogram result matrix based within-group group dissimilarities. dendrogram found method given cluster argument using function hclust. terminal segments hang within-cluster dissimilarity. clusters heterogeneous combined class, leaf segment reversed. histograms based dissimilarities, ore otherwise similar Van Sickle Hughes (2000): horizontal line drawn level mean -cluster dissimilarity vertical lines connect within-cluster dissimilarities line.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"function returns list class mrpp following items: call Function call. delta overall weighted mean group mean distances. E.delta expected delta, null hypothesis group structure. mean original dissimilarities. CS Classification strength (Van Sickle Hughes, 2000). Currently implemented always NA. n Number observations class. classdelta Mean dissimilarities within classes. overall \\(\\delta\\) weighted average values given weight.type . Pvalue Significance test. chance-corrected estimate proportion distances explained group identity; value analogous coefficient determination linear model. distance Choice distance metric used; \"method\" entry dist object. weight.type choice group weights used. boot.deltas vector \"permuted deltas,\" deltas calculated permuted datasets. distribution item can inspected permustats function. permutations number permutations used. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"B. McCune J. B. Grace. 2002. Analysis Ecological Communities. MjM Software Design, Gleneden Beach, Oregon, USA. P. W. Mielke K. J. Berry. 2001. Permutation Methods: Distance Function Approach. Springer Series Statistics. Springer. J. Van Sickle R. M. Hughes 2000. Classification strengths ecoregions, catchments, geographic clusters aquatic vertebrates Oregon. J. N. . Benthol. Soc. 19:370--384. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"M. Henry H. Stevens HStevens@muohio.edu Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"difference may one location (differences mean) one spread (differences within-group distance). , may find significant difference two groups simply one groups greater dissimilarities among sampling units. mrpp models can analysed adonis2 seems suffer problems mrpp robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"data(dune) data(dune.env) dune.mrpp <- with(dune.env, mrpp(dune, Management)) dune.mrpp #> #> Call: #> mrpp(dat = dune, grouping = Management) #> #> Dissimilarity index: euclidean #> Weights for groups: n #> #> Class means and counts: #> #> BF HF NM SF #> delta 10.03 11.08 10.66 12.27 #> n 3 5 6 6 #> #> Chance corrected within-group agreement A: 0.1246 #> Based on observed delta 11.15 and expected delta 12.74 #> #> Significance of delta: 0.001 #> Permutation: free #> Number of permutations: 999 #> # Save and change plotting parameters def.par <- par(no.readonly = TRUE) layout(matrix(1:2,nr=1)) plot(dune.ord <- metaMDS(dune, trace=0), type=\"text\", display=\"sites\" ) with(dune.env, ordihull(dune.ord, Management)) with(dune.mrpp, { fig.dist <- hist(boot.deltas, xlim=range(c(delta,boot.deltas)), main=\"Test of Differences Among Groups\") abline(v=delta); text(delta, 2*mean(fig.dist$counts), adj = -0.5, expression(bold(delta)), cex=1.5 ) } ) par(def.par) ## meandist dune.md <- with(dune.env, meandist(vegdist(dune), Management)) dune.md #> BF HF NM SF #> BF 0.4159972 0.4736637 0.7296979 0.6247169 #> HF 0.4736637 0.4418115 0.7217933 0.5673664 #> NM 0.7296979 0.7217933 0.6882438 0.7723367 #> SF 0.6247169 0.5673664 0.7723367 0.5813015 #> attr(,\"class\") #> [1] \"meandist\" \"matrix\" #> attr(,\"n\") #> grouping #> BF HF NM SF #> 3 5 6 6 summary(dune.md) #> #> Mean distances: #> Average #> within groups 0.5746346 #> between groups 0.6664172 #> overall 0.6456454 #> #> Summary statistics: #> Statistic #> MRPP A weights n 0.1423836 #> MRPP A weights n-1 0.1339124 #> MRPP A weights n(n-1) 0.1099842 #> Classification strength 0.1127012 plot(dune.md) plot(dune.md, kind=\"histogram\")"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":null,"dir":"Reference","previous_headings":"","what":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso adds attribute vario object class \"cca\" describes spatial partitioning cca object performs optional permutation test spatial independence residuals. function plot.mso creates diagnostic plot spatial partitioning \"cca\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"mso(object.cca, object.xy, grain = 1, round.up = FALSE, permutations = 0) msoplot(x, alpha = 0.05, explained = FALSE, ylim = NULL, legend = \"topleft\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"object.cca object class cca, created cca rda function. object.xy vector, matrix data frame spatial coordinates data represented object.cca. number rows must match number observations (given nobs) cca.object. Alternatively, interpoint distances can supplied dist object. grain Interval size distance classes. round.Determines choice breaks. false, distances rounded nearest multiple grain. true, distances rounded upper multiple grain. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x result object mso. alpha Significance level two-sided permutation test Mantel statistic spatial independence residual inertia point-wise envelope variogram total variance. Bonferroni-type correction can achieved dividing overall significance value (e.g. 0.05) number distance classes. explained false, suppresses plotting variogram explained variance. ylim Limits y-axis. legend x y co-ordinates used position legend. can specified keyword way accepted legend. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Mantel test adaptation function mantel vegan package parallel testing several distance classes. compares mean inertia distance class pooled mean inertia distance classes. explanatory variables (RDA, CCA, pRDA, pCCA) significance test residual autocorrelation performed running function mso, function plot.mso print estimate much autocorrelation (based significant distance classes) causes global error variance regression analysis underestimated","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso returns amended cca rda object additional attributes grain, H, H.test vario. grain grain attribute defines interval size distance classes . H H object class 'dist' contains geographic distances observations. H.test H.test contains set dummy variables describe pairs observations (rows = elements object$H) fall distance class (columns). vario vario attribute data frame contains following components rda case (cca case brackets): H Distance class multiples grain. Dist Average distance pairs observations distance class H. n Number unique pairs observations distance class \tH. Empirical (chi-square) variogram total variance \t(inertia). Sum Sum empirical (chi-square) variograms explained \tresidual variance (inertia). CA Empirical (chi-square) variogram residual variance \t(inertia). CCA Empirical (chi-square) variogram explained variance \t(inertia). pCCA Empirical (chi-square) variogram conditioned \tvariance (inertia). se Standard error empirical (chi-square) variogram \ttotal variance (inertia). CA.signif P-value permutation test spatial \tindependence residual variance (inertia).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Wagner, H.H. 2004. Direct multi-scale ordination canonical correspondence analysis. Ecology 85: 342--351.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"responsible author Helene Wagner.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function based code published Ecological Archives E085-006 (doi:10.1890/02-0738 ).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"## Reconstruct worked example of Wagner (submitted): X <- matrix(c(1, 2, 3, 2, 1, 0), 3, 2) Y <- c(3, -1, -2) tmat <- c(1:3) ## Canonical correspondence analysis (cca): Example.cca <- cca(X, Y) Example.cca <- mso(Example.cca, tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.cca) Example.cca$vario #> H Dist n All Sum CA CCA se #> 1 1 1 2 0.25 0.3456633 0.07461735 0.2710459 0 #> 2 2 2 1 1.00 0.8086735 0.01147959 0.7971939 NA ## Correspondence analysis (ca): Example.ca <- mso(cca(X), tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.ca) ## Unconstrained ordination with test for autocorrelation ## using oribatid mite data set as in Wagner (2004) data(mite) data(mite.env) data(mite.xy) mite.cca <- cca(log(mite + 1)) mite.cca <- mso(mite.cca, mite.xy, grain = 1, permutations = 99) msoplot(mite.cca) mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, grain = 1, #> permutations = 99) #> #> Inertia Rank #> Total 1.164 #> Unconstrained 1.164 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3662 0.1328 0.0723 0.0658 0.0559 0.0481 0.0418 0.0391 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All CA CA.signif #> 0 0 0.3555 63 0.6250 0.6250 0.01 #> 1 1 1.0659 393 0.7556 0.7556 0.01 #> 2 2 2.0089 534 0.8931 0.8931 0.01 #> 3 3 2.9786 417 1.0988 1.0988 0.01 #> 4 4 3.9817 322 1.3321 1.3321 0.01 #> 5 5 5.0204 245 1.5109 1.5109 0.01 #> 10 10 6.8069 441 1.7466 1.7466 0.01 #> #> Permutation: free #> Number of permutations: 99 #> ## Constrained ordination with test for residual autocorrelation ## and scale-invariance of species-environment relationships mite.cca <- cca(log(mite + 1) ~ SubsDens + WatrCont + Substrate + Shrub + Topo, mite.env) mite.cca <- mso(mite.cca, mite.xy, permutations = 99) msoplot(mite.cca) #> Error variance of regression model underestimated by 0.4 percent mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, permutations = #> 99) #> #> Inertia Proportion Rank #> Total 1.1638 1.0000 #> Constrained 0.5211 0.4478 11 #> Unconstrained 0.6427 0.5522 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 #> 0.31207 0.06601 0.04117 0.02938 0.02438 0.01591 0.01201 0.00752 0.00612 0.00373 #> CCA11 #> 0.00284 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.07888 0.06752 0.05457 0.04023 0.03855 0.03491 0.03233 0.02692 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All Sum CA CCA se CA.signif #> 0 0 0.3555 63 0.6250 0.7479 0.5512 0.1967 0.03506 0.01 #> 1 1 1.0659 393 0.7556 0.8820 0.6339 0.2482 0.01573 0.18 #> 2 2 2.0089 534 0.8931 0.9573 0.6473 0.3100 0.01487 0.68 #> 3 3 2.9786 417 1.0988 1.1010 0.6403 0.4607 0.01858 0.46 #> 4 4 3.9817 322 1.3321 1.2548 0.6521 0.6027 0.02439 0.98 #> 5 5 5.0204 245 1.5109 1.4564 0.6636 0.7928 0.02801 0.41 #> 10 10 6.8069 441 1.7466 1.6266 0.6914 0.9351 0.02052 0.19 #> #> Permutation: free #> Number of permutations: 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Multiplicative Diversity Partitioning — multipart","title":"Multiplicative Diversity Partitioning — multipart","text":"multiplicative diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set pooled samples (gamma diversity).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"multipart(...) # S3 method for default multipart(y, x, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula multipart(formula, data, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multiplicative Diversity Partitioning — multipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, two levels assumed: row group first level, rows group second level. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variable(s) referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, entropy index calculated (see Details). relative Logical, TRUE beta diversity standardized maximum (see Details). scales Numeric, length 1, order generalized diversity index used. global Logical, indicates calculation beta diversity values, see Details. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. ... arguments passed oecosimu, .e. method, thin burnin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multiplicative Diversity Partitioning — multipart","text":"Multiplicative diversity partitioning based Whittaker's (1972) ideas, recently generalised one parametric diversity families (.e. Rényi Tsallis) Jost (2006, 2007). Jost recommends use numbers equivalents (Hill numbers), instead pure diversities, proofs, satisfies multiplicative partitioning requirements. current implementation multipart calculates Hill numbers based functions renyi tsallis (provided index argument). values one scales desired, done separate runs, adds extra dimensionality implementation, resolved efficiently. Alpha diversities averages Hill numbers hierarchy levels, global gamma diversity alpha value calculated highest hierarchy level. global = TRUE, beta calculated relative global gamma value: $$\\beta_i = \\gamma / \\alpha_{}$$ global = FALSE, beta calculated relative local gamma values (local gamma means diversity calculated particular cluster based pooled abundance vector): $$\\beta_ij = \\alpha_{(+1)j} / mean(\\alpha_{ij})$$ \\(j\\) particular cluster hierarchy level \\(\\). beta diversity value level \\(\\) mean beta values clusters level, \\(\\beta_{} = mean(\\beta_{ij})\\). relative = TRUE, respective beta diversity values standardized maximum possible values (\\(mean(\\beta_{ij}) / \\beta_{max,ij}\\)) given \\(\\beta_{max,ij} = n_{j}\\) (number lower level units given cluster \\(j\\)). expected diversity components calculated nsimul times individual based randomization community data matrix. done \"r2dtable\" method oecosimu default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multiplicative Diversity Partitioning — multipart","text":"object class \"multipart\" structure \"oecosimu\" objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multiplicative Diversity Partitioning — multipart","text":"Jost, L. (2006). Entropy diversity. Oikos, 113, 363--375. Jost, L. (2007). Partitioning diversity independent alpha beta components. Ecology, 88, 2427--2439. Whittaker, R. (1972). Evolution measurement species diversity. Taxon, 21, 213--251.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multiplicative Diversity Partitioning — multipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)))) ## Multiplicative diversity partitioning multipart(mite, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(y = mite, x = levsm, index = \"renyi\", scales = 1, #> nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 11.235 -81.754 14.07974 14.02539 14.07261 14.148 0.05 * #> gamma 12.006 -256.950 14.13219 14.11654 14.13205 14.146 0.05 * #> beta.1 1.071 29.363 1.00373 0.99935 1.00399 1.008 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ l2 + l3, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"renyi\", scales = 1, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -46.004 12.17819 12.02119 12.15824 12.3415 0.05 * #> alpha.2 11.2353 -87.290 14.08463 14.03637 14.07817 14.1383 0.05 * #> alpha.3 12.0064 -490.481 14.13774 14.13229 14.13634 14.1450 0.05 * #> gamma 14.1603 0.000 14.16027 14.16027 14.16027 14.1603 1.00 #> beta.1 1.3568 20.034 1.16000 1.14399 1.16150 1.1779 0.05 * #> beta.2 1.0710 29.394 1.00379 0.99989 1.00410 1.0073 0.05 * #> beta.3 1.1794 577.616 1.00159 1.00108 1.00169 1.0020 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, relative=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.055481 -46.949 12.221716 12.073934 12.244586 12.3654 0.05 * #> alpha.2 11.235261 -123.496 14.087872 14.043345 14.090654 14.1218 0.05 * #> alpha.3 12.006443 -323.402 14.135613 14.124838 14.135306 14.1443 0.05 * #> gamma 14.160271 0.000 14.160271 14.160271 14.160271 14.1603 1.00 #> beta.1 0.078594 17.624 0.068099 0.067172 0.068222 0.0691 0.05 * #> beta.2 0.535514 42.418 0.501697 0.500646 0.501814 0.5031 0.05 * #> beta.3 0.589695 380.700 0.500872 0.500566 0.500883 0.5013 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, global=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, global = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global TRUE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -74.528 12.1813 12.0860 12.1748 12.2676 0.05 * #> alpha.2 11.2353 -92.796 14.0859 14.0459 14.0812 14.1484 0.05 * #> alpha.3 12.0064 -331.084 14.1370 14.1256 14.1378 14.1455 0.05 * #> gamma 14.1603 0.000 14.1603 14.1603 14.1603 14.1603 1.00 #> beta.1 1.7578 112.658 1.1625 1.1543 1.1631 1.1716 0.05 * #> beta.2 1.2603 116.533 1.0053 1.0008 1.0056 1.0081 0.05 * #> beta.3 1.1794 389.764 1.0016 1.0010 1.0016 1.0025 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":null,"dir":"Reference","previous_headings":"","what":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Patches local communities regarded nested subsets community. general, species poor communities subsets species rich communities, rare species occur species rich communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"nestedchecker(comm) nestedn0(comm) nesteddisc(comm, niter = 200) nestedtemp(comm, ...) nestednodf(comm, order = TRUE, weighted = FALSE, wbinary = FALSE) nestedbetasor(comm) nestedbetajac(comm) # S3 method for nestedtemp plot(x, kind = c(\"temperature\", \"incidence\"), col=rev(heat.colors(100)), names = FALSE, ...) # S3 method for nestednodf plot(x, col = \"red\", names = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"comm Community data. niter Number iterations reorder tied columns. x Result object plot. col Colour scheme matrix temperatures. kind kind plot produced. names Label columns rows plot using names comm. logical vector length 2, row column labels returned accordingly. order Order rows columns frequencies. weighted Use species abundances weights interactions. wbinary Modify original method binary data give result weighted unweighted analysis. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"nestedness functions evaluate alternative indices nestedness. functions intended used together Null model communities used argument oecosimu analyse non-randomness results. Function nestedchecker gives number checkerboard units, 2x2 submatrices species occur different sites (Stone & Roberts 1990). Function nestedn0 implements nestedness measure N0 number absences sites richer pauperate site species occurs (Patterson & Atmar 1986). Function nesteddisc implements discrepancy index number ones shifted fill row ones table arranged species frequencies (Brualdi & Sanderson 1999). original definition arranges species (columns) frequencies, method handling tied frequencies. nesteddisc function tries order tied columns minimize discrepancy statistic rather slow, large number tied columns guarantee best ordering found (argument niter gives maximum number tried orders). case warning tied columns issued. Function nestedtemp finds matrix temperature defined sum “surprises” arranged matrix. arranged unsurprising matrix species within proportion given matrix fill upper left corner matrix, surprise absence presences diagonal distance fill line (Atmar & Patterson 1993). Function tries pack species sites low temperature (Rodríguez-Gironés & Santamaria 2006), iterative procedure, temperatures usually vary among runs. Function nestedtemp also plot method can display either incidences temperatures surprises. Matrix temperature rather vaguely described (Atmar & Patterson 1993), Rodríguez-Gironés & Santamaria (2006) explicit description used . However, results probably differ implementations, users cautious interpreting results. details calculations explained vignette Design decisions implementation can read using functions browseVignettes. Function nestedness bipartite package direct port BINMATNEST programme Rodríguez-Gironés & Santamaria (2006). Function nestednodf implements nestedness metric based overlap decreasing fill (Almeida-Neto et al., 2008). Two basic properties required matrix maximum degree nestedness according metric: (1) complete overlap 1's right left columns rows, (2) decreasing marginal totals pairs columns pairs rows. nestedness statistic evaluated separately columns (N columns) rows (N rows) combined whole matrix (NODF). set order = FALSE, statistic evaluated current matrix ordering allowing tests meaningful hypothesis matrix structure default ordering row column totals (breaking ties total abundances weighted = TRUE) (see Almeida-Neto et al. 2008). weighted = TRUE, function finds weighted version index (Almeida-Neto & Ulrich, 2011). However, requires quantitative null models adequate testing. Almeida-Neto & Ulrich (2011) say positive nestedness values first row/column higher second. condition, weighted analysis binary data always give zero nestedness. argument wbinary = TRUE, equality rows/columns also indicates nestedness, binary data give identical results weighted unweighted analysis. However, can also influence results weighted analysis results may differ Almeida-Neto & Ulrich (2011). Functions nestedbetasor nestedbetajac find multiple-site dissimilarities decompose components turnover nestedness following Baselga (2012); pairwise dissimilarities can found designdist. can seen decomposition beta diversity (see betadiver). Function nestedbetasor uses Sørensen dissimilarity turnover component Simpson dissimilarity (Baselga 2012), nestedbetajac uses analogous methods Jaccard index. functions return vector three items: turnover, nestedness sum multiple Sørensen Jaccard dissimilarity. last one total beta diversity (Baselga 2012). functions treat data presence/absence (binary) can used binary nullmodel. overall dissimilarity constant nullmodels fix species (column) frequencies (\"c0\"), components constant row columns also fixed (e.g., model \"quasiswap\"), functions meaningful null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"result returned nestedness function contains item called statistic, components differ among functions. functions constructed can handled oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Almeida-Neto, M., Guimarães, P., Guimarães, P.R., Loyola, R.D. & Ulrich, W. (2008). consistent metric nestedness analysis ecological systems: reconciling concept measurement. Oikos 117, 1227--1239. Almeida-Neto, M. & Ulrich, W. (2011). straightforward computational approach measuring nestedness using quantitative matrices. Env. Mod. Software 26, 173--178. Atmar, W. & Patterson, B.D. (1993). measurement order disorder distribution species fragmented habitat. Oecologia 96, 373--382. Baselga, . (2012). relationship species replacement, dissimilarity derived nestedness, nestedness. Global Ecol. Biogeogr. 21, 1223--1232. Brualdi, R.. & Sanderson, J.G. (1999). Nested species subsets, gaps, discrepancy. Oecologia 119, 256--264. Patterson, B.D. & Atmar, W. (1986). Nested subsets structure insular mammalian faunas archipelagos. Biol. J. Linnean Soc. 28, 65--82. Rodríguez-Gironés, M.. & Santamaria, L. (2006). new algorithm calculate nestedness temperature presence-absence matrices. J. Biogeogr. 33, 924--935. Stone, L. & Roberts, . (1990). checkerboard score species distributions. Oecologia 85, 74--79. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Jari Oksanen Gustavo Carvalho (nestednodf).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"data(sipoo) ## Matrix temperature out <- nestedtemp(sipoo) out #> nestedness temperature: 10.23506 #> with matrix fill 0.2233333 plot(out) plot(out, kind=\"incid\") ## Use oecosimu to assess the non-randomness of checker board units nestedchecker(sipoo) #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 oecosimu(sipoo, nestedchecker, \"quasiswap\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = #> \"quasiswap\") #> #> nullmodel method ‘quasiswap’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.55254 2715.3 2564.9 2713.0 2911.2 0.57 ## Another Null model and standardized checkerboard score oecosimu(sipoo, nestedchecker, \"r00\", statistic = \"C.score\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r00\", #> statistic = \"C.score\") #> #> nullmodel method ‘r00’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> C.score 2.2588 -28.215 9.2285 8.7036 9.2294 9.6735 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract the Number of Observations from a vegan Fit. — nobs.cca","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Extract number ‘observations’ vegan model fit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"","code":"# S3 method for cca nobs(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"object fitted model object. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Function nobs generic R, vegan provides methods objects betadisper, cca related methods, CCorA, decorana, isomap, metaMDS, pcnm, procrustes, radfit, varpart wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"single number, normally integer, giving number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Null Model and Simulation — nullmodel","title":"Null Model and Simulation — nullmodel","text":"nullmodel function creates object, can serve basis Null Model simulation via simulate method. update method updates nullmodel object without sampling (effective sequential algorithms). smbind binds together multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Null Model and Simulation — nullmodel","text":"","code":"nullmodel(x, method) # S3 method for nullmodel print(x, ...) # S3 method for nullmodel simulate(object, nsim = 1, seed = NULL, burnin = 0, thin = 1, ...) # S3 method for nullmodel update(object, nsim = 1, seed = NULL, ...) # S3 method for simmat print(x, ...) smbind(object, ..., MARGIN, strict = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Null Model and Simulation — nullmodel","text":"x community matrix. print method, object printed. method Character, specifying one null model algorithms listed help page commsim. can user supplied object class commsim. object object class nullmodel returned function nullmodel. case smbind simmat object returned update simulate methods. nsim Positive integer, number simulated matrices return. update method, number burnin steps made sequential algorithms update status input model object. seed object specifying random number generator initialized (\"seeded\"). Either NULL integer used call set.seed simulating matrices. set, value saved \"seed\" attribute returned value. default, NULL change random generator state, return .Random.seed \"seed\" attribute, see Value. burnin Nonnegative integer, specifying number steps discarded starting simulation. Active sequential null model algorithms. Ignored non-sequential null model algorithms. thin Positive integer, number simulation steps made returned matrix. Active sequential null model algorithms. Ignored non-sequential null model algorithms. MARGIN Integer, indicating dimension multiple simmat objects bound together smbind. 1: matrices stacked (row bound), 2: matrices column bound, 3: iterations combined. Needs length 1. dimensions expected match across objects. strict Logical, consistency time series attributes (\"start\", \"end\", \"thin\", number simulated matrices) simmat objects strictly enforced binding multiple objects together using smbind. Applies input objects based sequential null model algorithms. ... Additional arguments supplied algorithms. case smbind can contain multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Null Model and Simulation — nullmodel","text":"purpose nullmodel function create object, necessary statistics input matrix calculated . information reused, recalculated step simulation process done simulate method. simulate method carries simulation, simulated matrices stored array. sequential algorithms, method updates state input nullmodel object. Therefore, possible diagnostic tests returned simmat object, make simulations, use increased thinning value desired. update method makes burnin steps case sequential algorithms update status input model without attempt return matrices. non-sequential algorithms method nothing. update preferred way making burnin iterations without sampling. Alternatively, burnin can done via simulate method. convergence diagnostics, recommended use simulate method without burnin. input nullmodel object updated, samples can simulated desired without start process . See Examples. smbind function can used combine multiple simmat objects. comes handy null model simulations stratified sites (MARGIN = 1) species (MARGIN = 2), case multiple objects returned identical/consistent settings e.g. parallel computations (MARGIN = 3). Sanity checks made ensure combining multiple objects sensible, user's responsibility check independence simulated matrices null distribution converged case sequential null model algorithms. strict = FALSE setting can relax checks regarding start, end, thinning values sequential null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Null Model and Simulation — nullmodel","text":"function nullmodel returns object class nullmodel. set objects sharing environment: data: original matrix integer mode. nrow: number rows. ncol: number columns. rowSums: row sums. colSums: column sums. rowFreq: row frequencies (number nonzero cells). colFreq: column frequencies (number nonzero cells). totalSum: total sum. fill: number nonzero cells matrix. commsim: commsim object result method argument. state: current state permutations, matrix similar original. NULL non-sequential algorithms. iter: current number iterations sequential algorithms. NULL non-sequential algorithms. simulate method returns object class simmat. array simulated matrices (third dimension corresponding nsim argument). update method returns current state (last updated matrix) invisibly, update input object sequential algorithms. non sequential algorithms, returns NULL. smbind function returns object class simmat.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Null Model and Simulation — nullmodel","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Null Model and Simulation — nullmodel","text":"Care must taken input matrix contains single row column. input invalid swapping hypergeometric distribution (calling r2dtable) based algorithms. also applies cases input stratified subsets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Null Model and Simulation — nullmodel","text":"","code":"data(mite) x <- as.matrix(mite)[1:12, 21:30] ## non-sequential nullmodel (nm <- nullmodel(x, \"r00\")) #> An object of class “nullmodel” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> (sm <- simulate(nm, nsim=10)) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## sequential nullmodel (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 55, End = 100, Thin = 5 #> ## sequential nullmodel with burnin and extra updating (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, burnin=10, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> ## sequential nullmodel with separate initial burnin (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> nm <- update(nm, nsim=10) (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> ## combining multiple simmat objects ## stratification nm1 <- nullmodel(x[1:6,], \"r00\") sm1 <- simulate(nm1, nsim=10) nm2 <- nullmodel(x[7:12,], \"r00\") sm2 <- simulate(nm2, nsim=10) smbind(sm1, sm2, MARGIN=1) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## binding subsequent samples from sequential algorithms ## start, end, thin retained nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=10) sm1 <- simulate(nm, nsim=10, thin=5) sm2 <- simulate(nm, nsim=20, thin=5) sm3 <- simulate(nm, nsim=10, thin=5) smbind(sm3, sm2, sm1, MARGIN=3) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 40 #> Start = 15, End = 210, Thin = 5 #> ## 'replicate' based usage which is similar to the output ## of 'parLapply' or 'mclapply' in the 'parallel' package ## start, end, thin are set, also noting number of chains smfun <- function(x, burnin, nsim, thin) { nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=burnin) simulate(nm, nsim=nsim, thin=thin) } smlist <- replicate(3, smfun(x, burnin=50, nsim=10, thin=5), simplify=FALSE) smbind(smlist, MARGIN=3) # Number of permuted matrices = 30 #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 30 #> Start = 55, End = 100, Thin = 5 (3 chains) #> if (FALSE) { ## parallel null model calculations library(parallel) if (.Platform$OS.type == \"unix\") { ## forking on Unix systems smlist <- mclapply(1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) smbind(smlist, MARGIN=3) } ## socket type cluster, works on all platforms cl <- makeCluster(3) clusterEvalQ(cl, library(vegan)) clusterExport(cl, c(\"smfun\", \"x\")) smlist <- parLapply(cl, 1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) stopCluster(cl) smbind(smlist, MARGIN=3) }"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function evaluates statistic vector statistics community evaluates significance series simulated random communities. approach used traditionally analysis nestedness, function general can used statistics evaluated simulated communities. Function oecosimu collects evaluates statistics. Null model communities described make.commsim permatfull/ permatswap, definition Null models nullmodel, nestedness statistics nestednodf (describes several alternative statistics, including nestedness temperature, \\(N0\\), checker board units, nestedness discrepancy NODF).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"oecosimu(comm, nestfun, method, nsimul = 99, burnin = 0, thin = 1, statistic = \"statistic\", alternative = c(\"two.sided\", \"less\", \"greater\"), batchsize = NA, parallel = getOption(\"mc.cores\"), ...) # S3 method for oecosimu as.ts(x, ...) # S3 method for oecosimu toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"comm Community data, Null model object generated nullmodel object class simmat (array permuted matrices simulate.nullmodel). comm community data, null model simulation method must specified. comm nullmodel, simulation method ignored, comm simmat object, arguments ignored except nestfun, statistic alternative. nestfun Function analysed. nestedness functions provided vegan (see nestedtemp), function can used accepts community first argument, returns either plain number vector result list item name defined argument statistic. See Examples defining functions. method Null model method: either name (character string) method defined make.commsim commsim function. argument ignored comm nullmodel simmat object. See Details Examples. nsimul Number simulated null communities (ignored comm simmat object). burnin Number null communities discarded proper analysis sequential methods (\"tswap\") (ignored non-sequential methods comm simmat object). thin Number discarded null communities two evaluations nestedness statistic sequential methods (ignored non-sequential methods comm simmat object). statistic name statistic returned nestfun. alternative character string specifying alternative hypothesis, must one \"two.sided\" (default), \"greater\" \"less\". Please note \\(p\\)-value two-sided test approximately two times higher corresponding one-sided test (\"greater\" \"less\" depending sign difference). batchsize Size Megabytes largest simulation object. larger structure produced, analysis broken internally batches. default NA analysis broken batches. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. define nestfun Windows needs R packages vegan permute, must set socket cluster call. x oecosimu result object. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu wrapper evaluates statistic using function given nestfun, simulates series null models based nullmodel, evaluates statistic null models. vegan packages contains nestedness functions described separately (nestedchecker, nesteddisc, nestedn0, nestedtemp, nestednodf), many functions can used long meaningful simulated communities. applicable function must return either statistic plain number vector, list element \"statistic\" (like chisq.test), item whose name given argument statistic. statistic can single number (like typical nestedness index), can vector. vector indices can used analyse site (row) species (column) properties, see treedive example. Raup-Crick index (raupcrick) gives example using dissimilarities. Null model type can given name (quoted character string) used define Null model make.commsim. include binary models described Wright et al. (1998), Jonsson (2001), Gotelli & Entsminger (2003), Miklós & Podani (2004), others. several quantitative Null models, discussed Hardy (2008), several unpublished (see make.commsim, permatfull, permatswap discussion). user can also define commsim function (see Examples). Function works first defining nullmodel given commsim, generating series simulated communities simulate.nullmodel. shortcut can used stages input can Community data (comm), Null model function (nestfun) number simulations (nsimul). nullmodel object number simulations, argument method ignored. three-dimensional array simulated communities generated simulate.nullmodel, arguments method nsimul ignored. last case allows analysing several statistics simulations. function first generates simulations given nullmodel analyses using nestfun. large data sets /large number simulations, generated objects can large, memory exhausted, analysis can become slow system can become unresponsive. simulation broken several smaller batches simulated nullmodel objective set batchsize avoid memory problems (see object.size estimating size current data set). parallel processing still increases memory needs. parallel processing used evaluating nestfun. main load may simulation nullmodel, parallel argument help . Function .ts transforms simulated results sequential methods time series ts object. allows using analytic tools time series studying sequences (see examples). Function toCoda transforms simulated results sequential methods \"mcmc\" object coda package. coda package provides functions analysis stationarity, adequacy sample size, autocorrelation, need burn-much sequential methods, summary results. Please consult documentation coda package. Function permustats provides support standard density, densityplot, qqnorm qqmath functions simulated values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu returns object class \"oecosimu\". result object items statistic oecosimu. statistic contains complete object returned nestfun original data. oecosimu component contains following items: statistic Observed values statistic. simulated Simulated values statistic. means Mean values statistic simulations. z Standardized effect sizes (SES, .k.. \\(z\\)-values) observed statistic based simulations. pval \\(P\\)-values statistic based simulations. alternative type testing given argument alternative. method method used nullmodel. isSeq TRUE method sequential.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Jari Oksanen Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"wonder name oecosimu, look journal names References (nestedtemp). internal structure function radically changed vegan 2.2-0 introduction commsim nullmodel deprecation commsimulator.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"## Use the first eigenvalue of correspondence analysis as an index ## of structure: a model for making your own functions. data(sipoo) ## Traditional nestedness statistics (number of checkerboard units) oecosimu(sipoo, nestedchecker, \"r0\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r0\") #> #> nullmodel method ‘r0’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 -18.742 8017.0 7469.0 8021.0 8450.5 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential model, one-sided test, a vector statistic out <- oecosimu(sipoo, decorana, \"swap\", burnin=100, thin=10, statistic=\"evals\", alt = \"greater\") out #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = decorana, method = \"swap\", #> burnin = 100, thin = 10, statistic = \"evals\", alternative = \"greater\") #> #> nullmodel method ‘swap’ with 99 simulations #> options: thin 10, burnin 100 #> alternative hypothesis: statistic is greater than simulated values #> #> #> Call: #> nestfun(veg = comm) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.4436 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.3822 0.2612 0.1668 0.08723 #> Additive Eigenvalues 0.3822 0.2609 0.1631 0.07650 #> Decorana values 0.4154 0.2465 0.1391 0.04992 #> Axis lengths 2.9197 2.5442 2.7546 1.78074 #> #> #> statistic SES mean 50% 95% Pr(sim.) #> DCA1 0.382249 2.04544 0.32960 0.33249 0.3677 0.01 ** #> DCA2 0.261208 1.77368 0.21549 0.21404 0.2587 0.05 * #> DCA3 0.166788 0.63257 0.15363 0.15405 0.1907 0.22 #> DCA4 0.087226 -1.69622 0.12533 0.12788 0.1636 0.96 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Inspect the swap sequence as a time series object plot(as.ts(out)) lag.plot(as.ts(out)) acf(as.ts(out)) ## Density plot densityplot(permustats(out), as.table = TRUE, layout = c(1,4)) ## Use quantitative null models to compare ## mean Bray-Curtis dissimilarities data(dune) meandist <- function(x) mean(vegdist(x, \"bray\")) mbc1 <- oecosimu(dune, meandist, \"r2dtable\") mbc1 #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = \"r2dtable\") #> #> nullmodel method ‘r2dtable’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 13.84 0.46601 0.44155 0.46756 0.4928 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Define your own null model as a 'commsim' function: shuffle cells ## in each row foo <- function(x, n, nr, nc, ...) { out <- array(0, c(nr, nc, n)) for (k in seq_len(n)) out[,,k] <- apply(x, 2, function(z) sample(z, length(z))) out } cf <- commsim(\"myshuffle\", foo, isSeq = FALSE, binary = FALSE, mode = \"double\") oecosimu(dune, meandist, cf) #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = cf) #> #> nullmodel method ‘myshuffle’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 3.8862 0.63505 0.62995 0.63508 0.64 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Use pre-built null model nm <- simulate(nullmodel(sipoo, \"curveball\"), 99) oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘curveball’ with 99 simulations #> options: thin 1, burnin 0 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 1.4459 2710.6 2635.0 2723.0 2762.7 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Several chains of a sequential model -- this can be generalized ## for parallel processing (see ?smbind) nm <- replicate(5, simulate(nullmodel(sipoo, \"swap\"), 99, thin=10, burnin=100), simplify = FALSE) ## nm is now a list of nullmodels: use smbind to combine these into one ## nullmodel with several chains ## IGNORE_RDIFF_BEGIN nm <- smbind(nm, MARGIN = 3) nm #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 18 x 50 matrix #> Number of permuted matrices = 495 #> Start = 110, End = 1090, Thin = 10 (5 chains) #> oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘swap’ with 495 simulations #> options: thin 10, burnin 100, chains 5 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.71842 2698.1 2572.0 2676.0 2904.7 0.4577 ## IGNORE_RDIFF_END ## After this you can use toCoda() and tools in the coda package to ## analyse the chains (these will show that thin, burnin and nsimul are ## all too low for real analysis)."},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":null,"dir":"Reference","previous_headings":"","what":"Support Functions for Drawing Vectors — ordiArrowTextXY","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Support functions assist drawing vectors (arrows) ordination plots. ordiArrowMul finds multiplier coordinates head vector occupy fill proportion plot region. ordiArrowTextXY finds coordinates locations labels drawn just beyond head vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"ordiArrowTextXY(x, labels, display, choices = c(1,2), rescale = TRUE, fill = 0.75, at = c(0,0), cex = NULL, ...) ordiArrowMul(x, at = c(0,0), fill = 0.75, display, choices = c(1,2), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"x R object, scores can determine suitable ordination scores object created envfit, two-column matrix coordinates arrow heads two plot axes. labels Change plotting labels. character vector labels label coordinates sought. supplied, determined row names x, scores(x, ...) required. either defined, suitable labels generated. display character string known scores one methods indicates type scores extract. fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). x created envfit display can set user takes value \"vectors\". Ignored x matrix. choices Axes plotted. rescale logical; coordinates extracted x rescaled fill fill proportion plot region? default always rescale coordinates usually desired objects x coordinates retrieved. supplying x 2-column matrix already rescaled, set FALSE. fill numeric; proportion plot fill span arrows. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. cex Character expansion text. ... Parameters passed scores, strwidth strheight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowMul finds multiplier scale bunch arrows fill ordination plot, ordiArrowTextXY finds coordinates labels arrows. NB., ordiArrowTextXY draw labels; simply returns coordinates labels drawn use another function, text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowTextXY, 2-column matrix coordinates label centres coordinate system currently active plotting device. ordiArrowMul, length-1 vector containing scaling factor.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Jari Oksanen, modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"## Scale arrows by hand to fill 80% of the plot ## Biplot arrows by hand data(varespec, varechem) ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, display = c(\"species\",\"sites\")) ## biplot scores bip <- scores(ord, choices = 1:2, display = \"bp\") ## scaling factor for arrows to fill 80% of plot (mul <- ordiArrowMul(bip, fill = 0.8)) #> [1] 2.092173 bip.scl <- bip * mul # Scale the biplot scores labs <- rownames(bip) # Arrow labels ## calculate coordinate of labels for arrows (bip.lab <- ordiArrowTextXY(bip.scl, rescale = FALSE, labels = labs)) #> [,1] [,2] #> Al 1.9098765 -0.3562415 #> P -0.9298005 -1.6652122 #> K -1.0069931 -0.3764923 ## draw arrows and text labels arrows(0, 0, bip.scl[,1], bip.scl[,2], length = 0.1) text(bip.lab, labels = labs) ## Handling of ordination objects directly mul2 <- ordiArrowMul(ord, display = \"bp\", fill = 0.8) stopifnot(all.equal(mul, mul2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Functions add arrows, line segments, regular grids points. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"ordiarrows(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, startmark, label = FALSE, length = 0.1, ...) ordisegments(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, label = FALSE, ...) ordigrid(ord, levels, replicates, display = \"sites\", lty = c(1,1), col = c(1,1), lwd = c(1,1), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. levels, replicates Alternatively, regular groups can defined arguments levels replicates, levels gives number groups, replicates number successive items group. order.Order points increasing order variable within groups. Reverse sign variable decreasing ordering. display Item displayed. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names. ordiellipse, ordihull ordispider group name centroid object, ordiarrows start arrow, ordisegments ends. ordiellipse ordihull use standard text, others use ordilabel. startmark plotting character used mark first item. default use mark, instance, startmark = 1 draw circle. plotting characters, see pch points. col Colour lines, label borders startmark ordiarrows ordisegments. can vector recycled groups. ordigrid can vector length 2 used levels replicates. length Length edges arrow head (inches). lty, lwd Line type, line width used levels replicates ordigrid. ... Parameters passed graphical functions lines, segments, arrows, scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Function ordiarrows draws arrows ordisegments draws line segments successive items groups. Function ordigrid draws line segments within groups corresponding items among groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"functions add graphical items ordination graph: must draw graph first.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"example(pyrifos) #> #> pyrifs> data(pyrifos) #> #> pyrifs> ditch <- gl(12, 1, length=132) #> #> pyrifs> week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) #> #> pyrifs> dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) mod <- rda(pyrifos) plot(mod, type = \"n\") ## Annual succession by ditches, colour by dose ordiarrows(mod, ditch, label = TRUE, col = as.numeric(dose)) legend(\"topright\", levels(dose), lty=1, col=1:5, title=\"Dose\") ## Show only control and highest Pyrifos treatment plot(mod, type = \"n\") ordiarrows(mod, ditch, label = TRUE, show.groups = c(\"2\", \"3\", \"5\", \"11\")) ordiarrows(mod, ditch, label = TRUE, show = c(\"6\", \"9\"), col = 2) legend(\"topright\", c(\"Control\", \"Pyrifos 44\"), lty = 1, col = c(1,2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions add convex hulls, “spider” graphs, ellipses cluster dendrogram ordination diagrams. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"ordihull(ord, groups, display = \"sites\", draw = c(\"lines\",\"polygon\", \"none\"), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd = NULL, ...) ordiellipse(ord, groups, display=\"sites\", kind = c(\"sd\",\"se\", \"ehull\"), conf, draw = c(\"lines\",\"polygon\", \"none\"), w = weights(ord, display), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd=NULL, ...) ordibar(ord, groups, display = \"sites\", kind = c(\"sd\", \"se\"), conf, w = weights(ord, display), col = 1, show.groups, label = FALSE, lwd = NULL, length = 0, ...) ordispider(ord, groups, display=\"sites\", w = weights(ord, display), spiders = c(\"centroid\", \"median\"), show.groups, label = FALSE, col = NULL, lty = NULL, lwd = NULL, ...) ordicluster(ord, cluster, prune = 0, display = \"sites\", w = weights(ord, display), col = 1, draw = c(\"segments\", \"none\"), ...) # S3 method for ordihull summary(object, ...) # S3 method for ordiellipse summary(object, ...) ordiareatest(ord, groups, area = c(\"hull\", \"ellipse\"), kind = \"sd\", permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. display Item displayed. draw character; objects represented plot? ordihull ordiellipse use either lines polygon draw lines. ordicluster, line segments drawn using segments. suppress plotting, use \"none\". Graphical parameters passed . main difference polygons may filled non-transparent. none nothing drawn, function returns invisible plotting. col Colour hull ellipse lines (draw = \"lines\") fills (draw = \"polygon\") ordihull ordiellipse. draw = \"polygon\", colour bordering lines can set argument border polygon function. functions effect depends underlining functions argument passed . multiple values col specified used element names(table(groups)) (order), shorter vectors recycled. Function ordicluster groups, argument recycled points, colour connecting lines mixture point s cluster. alpha Transparency fill colour draw = \"polygon\" ordihull ordiellipse. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names centroid object. ordiellipse ordihull use standard text, others use ordilabel. w Weights used find average within group. Weights used automatically cca decorana results, unless undone user. w=NULL sets equal weights points. kind Draw standard deviations points (sd), standard errors (se) ellipsoid hulls enclose points group (ehull). conf Confidence limit ellipses, e.g. 0.95. given, corresponding sd se multiplied corresponding value found Chi-squared distribution 2df. spiders centres spider bodies calculated either centroids (averages) spatial medians. cluster Result hierarchic cluster analysis, hclust agnes. prune Number upper level hierarchies removed dendrogram. prune \\(>0\\), dendrogram disconnected. object result object ordihull ordiellipse. result invisible, can saved, used summaries (areas etc. hulls ellipses). area Evaluate area convex hulls ordihull, ellipses ordiellipse. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. lty, lwd, border Vectors parameters can supplied applied (appropriate) element names(table(groups)) (order). Shorter vectors recycled. length Width (inches) small (“caps”) ends bar segment (passed arrows). ... Parameters passed graphical functions scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Function ordihull draws lines polygons convex hulls found function chull encircling items groups. Function ordiellipse draws lines polygons ellipses groups. function can either draw standard deviation points (kind=\"sd\") standard error (weighted) centroids (kind=\"se\"), (weighted) correlation defines direction principal axis ellipse. kind = \"se\" used together argument conf, ellipses show confidence regions locations group centroids. kind=\"ehull\" function draws ellipse encloses points group using ellipsoidhull (cluster package). Function ordibar draws crossed “error bars” using either either standard deviation point scores standard error (weighted) average scores. principal axes corresponding ordiellipse, found principal component analysis (weighted) covariance matrix. Functions ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups. Argument kind can used select kind ellipse, effect convex hulls. Function ordispider draws ‘spider’ diagram point connected group centroid segments. Weighted centroids used correspondence analysis methods cca decorana user gives weights call. ordispider called cca rda result without groups argument, function connects ‘WA’ scores corresponding ‘LC’ score. argument (invisible) ordihull object, function connect points hull centroid. Function ordicluster overlays cluster dendrogram onto ordination. needs result hierarchic clustering hclust agnes, similar structure. Function ordicluster connects cluster centroids line segments. Function uses centroids points clusters, therefore similar average linkage methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"functions add graphical items ordination graph: must draw graph first. draw line segments, grids arrows, see ordisegments, ordigrid andordiarrows.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions ordihull, ordiellipse ordispider return invisible plotting structure. Function ordispider return coordinates point connected (centroids ‘LC’ scores). Function ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ Management, dune.env) plot(mod, type=\"n\", scaling = \"symmetric\") ## Catch the invisible result of ordihull... pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", label = TRUE)) ## ... and find centres and areas of the hulls summary(pl) #> BF HF NM SF #> CCA1 0.2917476 0.36826105 -1.3505642 0.2762936 #> CCA2 0.8632208 0.09419919 0.2681515 -0.8139398 #> Area 0.1951715 0.59943363 1.7398193 1.0144372 ## use more colours and add ellipsoid hulls plot(mod, type = \"n\") pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", col = 1:4, draw=\"polygon\", label =TRUE)) with(dune.env, ordiellipse(mod, Management, scaling = \"symmetric\", kind = \"ehull\", col = 1:4, lwd=3)) ## ordispider to connect WA and LC scores plot(mod, dis=c(\"wa\",\"lc\"), type=\"p\") ordispider(mod) ## Other types of plots plot(mod, type = \"p\", display=\"sites\") cl <- hclust(vegdist(dune)) ordicluster(mod, cl, prune=3, col = cutree(cl, 4)) ## confidence ellipse: location of the class centroids plot(mod, type=\"n\", display = \"sites\") with(dune.env, text(mod, display=\"sites\", labels = as.character(Management), col=as.numeric(Management))) pl <- with(dune.env, ordiellipse(mod, Management, kind=\"se\", conf=0.95, lwd=2, draw = \"polygon\", col=1:4, border=1:4, alpha=63)) summary(pl) #> BF HF NM SF #> CCA1 0.4312652 0.5583211 -1.87848340 0.5601499 #> CCA2 1.3273917 0.6373120 -0.05503211 -1.3859924 #> Area 1.4559842 1.3806668 2.73667419 1.5559135 ## add confidence bars with(dune.env, ordibar(mod, Management, kind=\"se\", conf=0.95, lwd=2, col=1:4, label=TRUE))"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Function ordilabel similar text, text opaque label. can help crowded ordination plots: still see text labels, least uppermost readable. Argument priority helps make important labels visible.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"ordilabel(x, display, labels, choices = c(1, 2), priority, select, cex = 0.8, fill = \"white\", border = NULL, col = NULL, xpd = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"x ordination object object known scores. display Kind scores displayed (passed scores). labels Optional text used plots. given, text found ordination object. choices Axes shown (passed scores). priority Vector length number labels. items high priority plotted uppermost. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. cex Character expansion text (passed text). fill Background colour labels (col argument polygon). border colour visibility border label defined polygon. col Text colour. Default NULL give value border par(\"fg\") border NULL. xpd Draw labels also outside plot region (see par). ... arguments (passed text).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"function may useful crowded ordination plots, particular together argument priority. see text labels, least readable. alternatives crowded plots identify.ordiplot, orditorp orditkplot ( vegan3d package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"data(dune) ord <- cca(dune) plot(ord, type = \"n\") ordilabel(ord, dis=\"sites\", cex=1.2, font=3, fill=\"hotpink\", col=\"blue\") ## You may prefer separate plots, but here species as well ordilabel(ord, dis=\"sp\", font=2, priority=colSums(dune))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Alternative plot and identify Functions for Ordination — ordiplot","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot alternative plotting function can worked vegan ordination result many non-vegan results. addition, plot functions vegan ordinations return invisibly \"ordiplot\" result object, allows using ordiplot support functions result: identify can used add labels selected site, species constraint points, points text can add elements plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"ordiplot(ord, choices = c(1, 2), type=\"points\", display, xlim, ylim, cex = 0.7, ...) # S3 method for ordiplot identify(x, what, labels, ...) # S3 method for ordiplot points(x, what, select, arrows = FALSE, ...) # S3 method for ordiplot text(x, what, labels, select, arrows = FALSE, length = 0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"ord result ordination. choices Axes shown. type type graph may \"points\", \"text\" \"none\" ordination method. display Display \"sites\" \"species\". default methods display , cca, rda, dbrda capscale plot.cca. xlim, ylim x y limits (min,max) plot. cex Character expansion factor points text. ... graphical parameters. x result object ordiplot. Items identified ordination plot. types depend kind plot used. methods know sites species, functions cca rda know addition constraints (LC scores), centroids biplot, plot.procrustes ordination plot heads points. labels Optional text used labels. Row names used missing. arrows Draw arrows origin. always TRUE biplot scores value ignored. Setting TRUE draw arrows type scores. allows, e.g, using biplot arrows species. arrow head value scores, possible text moved outwards. length Length arrow heads (see arrows). select Items displayed. can either logical vector TRUE displayed items vector indices displayed items.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot draws ordination diagram using black circles sites red crosses species. returns invisibly object class ordiplot can used identify.ordiplot label selected sites species, constraints cca rda. function can handle output several alternative ordination methods. cca, rda decorana uses plot method option type = \"points\". addition, plot functions methods return invisibly ordiplot object can used identify.ordiplot label points. ordinations relies scores extract scores. full user control plots, best call ordiplot type = \"none\" save result, add sites species using points.ordiplot text.ordiplot pass arguments corresponding default graphical functions. functions can chained pipes allows alternative intuitive way building plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot returns invisibly object class ordiplot used scores. general, vegan plot functions ordination results also return invisible ordiplot object. plot(..., type = \"n\") used originally, plot empty, items can added invisible object. Functions points text return input object without modification, allows chaining commands pipes. Function identify.ordiplot uses object label point.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"purpose functions provide similar functionality plot, plotid specid methods library labdsv. functions somewhat limited parametrization, can call directly standard identify plot functions better user control.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"## Draw a plot for a non-vegan ordination (cmdscale). data(dune) dune.dis <- vegdist(wisconsin(dune)) dune.mds <- cmdscale(dune.dis, eig = TRUE) dune.mds$species <- wascores(dune.mds$points, dune, expand = TRUE) pl <- ordiplot(dune.mds, type = \"none\") points(pl, \"sites\", pch=21, col=\"red\", bg=\"yellow\") text(pl, \"species\", col=\"blue\", cex=0.9) if (FALSE) { ## same plot using pipes (|>) ordiplot(dune.mds, type=\"n\") |> points(\"sites\", pch=21, col=\"red\", bg=\"yellow\") |> text(\"species\", col=\"blue\", cex=0.9) ## Some people think that species should be shown with arrows in PCA. ## Other ordination methods also return an invisible ordiplot object and ## we can use pipes to draw those arrows. mod <- rda(dune) plot(mod, type=\"n\") |> points(\"sites\", pch=16, col=\"red\") |> text(\"species\", arrows = TRUE, length=0.05, col=\"blue\") } ## Default plot of the previous using identify to label selected points if (FALSE) { pl <- ordiplot(dune.mds) identify(pl, \"spec\")}"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function ordipointlabel produces ordination plots points text label points. points exact location given ordination, function tries optimize location text labels minimize overplotting text. function may useful moderately crowded ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"ordipointlabel(x, display = c(\"sites\", \"species\"), choices = c(1, 2), col = c(1, 2), pch = c(\"o\", \"+\"), font = c(1, 1), cex = c(0.8, 0.8), add = FALSE, select, ...) # S3 method for ordipointlabel plot(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"x ordipointlabel() result object ordination function. plot.ordipointlabel object resulting call ordipointlabel(). display Scores displayed plot. choices Axes shown. col, pch, font, cex Colours, point types, font style character expansion kind scores displayed plot. vectors length number items display. add Add existing plot. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. select used single set scores plotted (.e. length(display) == 1), otherwise ignored warning issued. logical vector used, must length scores plotted. ... arguments passed points text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function uses simulated annealing (optim, method = \"SANN\") optimize location text labels points. eight possible locations: , , sides corners. weak preference text right point, weak avoidance corner positions. exact locations goodness solution varies runs, guarantee finding global optimum. optimization can take long time difficult cases high number potential overlaps. Several sets scores can displayed one plot. function modelled pointLabel maptools package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function returns invisibly object class ordipointlabel items xy coordinates points, labels coordinates labels, items pch, cex font graphical parameters point label. addition, returns result optim attribute \"optim\". unit overlap area character \"m\", variable cex smallest alternative. plot method based orditkplot alter reset graphical parameters via par. result object ordipointlabel inherits orditkplot vegan3d package, may possible edit result object orditkplot, good results necessary points span whole horizontal axis without empty margins.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function designed ordination graphics, optimization works properly plots isometric aspect ratio.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"data(dune) ord <- cca(dune) plt <- ordipointlabel(ord) ## set scaling - should be no warnings! ordipointlabel(ord, scaling = \"sites\") ## plot then add plot(ord, scaling = \"symmetric\", type = \"n\") ordipointlabel(ord, display = \"species\", scaling = \"symm\", add = TRUE) ordipointlabel(ord, display = \"sites\", scaling = \"symm\", add = TRUE) ## redraw plot without rerunning SANN optimisation plot(plt)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function provides plot.lm style diagnostic plots results constrained ordination cca, rda, dbrda capscale. Normally need plots, ordination descriptive make assumptions distribution residuals. However, permute residuals significance tests (anova.cca), may interested inspecting residuals really exchangeable independent fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"ordiresids(x, kind = c(\"residuals\", \"scale\", \"qqmath\"), residuals = \"working\", type = c(\"p\", \"smooth\", \"g\"), formula, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"x Ordination result cca, rda, dbrda, capscale. kind type plot: \"residuals\" plot residuals fitted values, \"scale\" square root absolute residuals fitted values, \"qqmath\" residuals expected distribution (defaults qnorm), unless defined differently formula argument. residuals kind residuals fitted values, alternatives \"working\", \"response\", \"standardized\" \"studentized\" (see Details). type type plot. argument passed lattice functions. formula Formula override default plot. formula can contain items Fitted, Residuals, Species Sites (provided names species sites available ordination result). ... arguments passed lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"default plots similar plot.lm, use Lattice functions xyplot qqmath. alternatives default formulae can replaced user. elements available formula groups argument Fitted, Residuals, Species Sites. residuals = \"response\" residuals = \"working\" fitted values residuals found functions fitted.cca residuals.cca. residuals = \"standardized\" residuals found rstandard.cca, residuals = \"studentized\" found rstudent.cca, cases fitted values standardized sigma.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function returns Lattice object can displayed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ordiresids(mod) ordiresids(mod, formula = Residuals ~ Fitted | Species, residuals=\"standard\", cex = 0.5)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":null,"dir":"Reference","previous_headings":"","what":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Automatic stepwise model building constrained ordination methods (cca, rda, dbrda, capscale). function ordistep modelled step can forward, backward stepwise model selection using permutation tests. Function ordiR2step performs forward model choice solely adjusted \\(R^2\\) \\(P\\)-value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"ordistep(object, scope, direction = c(\"both\", \"backward\", \"forward\"), Pin = 0.05, Pout = 0.1, permutations = how(nperm = 199), steps = 50, trace = TRUE, ...) ordiR2step(object, scope, Pin = 0.05, R2scope = TRUE, permutations = how(nperm = 499), trace = TRUE, R2permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"object ordistep, ordination object inheriting cca rda. scope Defines range models examined stepwise search. can list containing components upper lower, formulae. single item, interpreted target scope, depending direction. direction \"forward\", single item interpreted upper scope formula input object lower scope. See step details. ordiR2step, defines upper scope; can also ordination object model extracted. direction mode stepwise search, can one \"\", \"backward\", \"forward\", default \"\". scope argument missing, default direction \"backward\" ordistep ( ordiR2step argument, works forward). Pin, Pout Limits permutation \\(P\\)-values adding (Pin) term model, dropping (Pout) model. Term added \\(P \\le\\) Pin, removed \\(P >\\) Pout. R2scope Use adjusted \\(R^2\\) stopping criterion: models lower adjusted \\(R^2\\) scope accepted. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. passed anova.cca: see details. steps Maximum number iteration steps dropping adding terms. trace positive, information printed model building. Larger values may give information. R2permutations Number permutations used estimation adjusted \\(R^2\\) cca using RsquareAdj. ... additional arguments add1.cca drop1.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"basic functions model choice constrained ordination add1.cca drop1.cca. functions, ordination models can chosen standard R function step bases term choice AIC. AIC-like statistics ordination provided functions deviance.cca extractAIC.cca ( similar functions rda). Actually, constrained ordination methods AIC, therefore step may trusted. function provides alternative using permutation \\(P\\)-values. Function ordistep defines model, scope models considered, direction procedure similarly step. function alternates drop add steps stops model changed one step. - + signs summary table indicate stage performed. often sensible Pout \\(>\\) Pin stepwise models avoid cyclic adds drops single terms. Function ordiR2step builds model forward maximizes adjusted \\(R^2\\) (function RsquareAdj) every step, stopping adjusted \\(R^2\\) starts decrease, adjusted \\(R^2\\) scope exceeded, selected permutation \\(P\\)-value exceeded (Blanchet et al. 2008). second criterion ignored option R2scope = FALSE, third criterion can ignored setting Pin = 1 (higher). function used adjusted \\(R^2\\) calculated. number predictors higher number observations, adjusted \\(R^2\\) also unavailable. models can analysed R2scope = FALSE, variable selection stop models become overfitted adjusted \\(R^2\\) calculated, adjusted \\(R^2\\) reported zero. \\(R^2\\) cca based simulations (see RsquareAdj) different runs ordiR2step can give different results. Functions ordistep (based \\(P\\) values) ordiR2step (based adjusted \\(R^2\\) hence eigenvalues) can select variables different order.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Functions return selected model one additional component, anova, contains brief information steps taken. can suppress voluminous output model building setting trace = FALSE, find summary model history anova item.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Blanchet, F. G., Legendre, P. & Borcard, D. (2008) Forward selection explanatory variables. Ecology 89, 2623--2632.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"## See add1.cca for another example ### Dune data data(dune) data(dune.env) mod0 <- rda(dune ~ 1, dune.env) # Model with intercept only mod1 <- rda(dune ~ ., dune.env) # Model with all explanatory variables ## With scope present, the default direction is \"both\" mod <- ordistep(mod0, scope = formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.035 * #> + Use 2 91.032 1.1741 0.300 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.085 . #> + A1 1 87.424 1.2965 0.170 #> + Use 2 88.284 1.0510 0.355 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.010 ** #> - Management 3 87.707 2.1769 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.535 #> + Use 2 86.842 0.8027 0.785 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## summary table of steps mod$anova #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 85.567 1.9764 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Example of ordistep, forward ordistep(mod0, scope = formula(mod1), direction=\"forward\") #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.020 * #> + Use 2 91.032 1.1741 0.290 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.055 . #> + A1 1 87.424 1.2965 0.225 #> + Use 2 88.284 1.0510 0.410 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.595 #> + Use 2 86.842 0.8027 0.730 #> #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## Example of ordiR2step (always forward) ## stops because R2 of 'mod1' exceeded ordiR2step(mod0, mod1) #> Step: R2.adj= 0 #> Call: dune ~ 1 #> #> R2.adjusted #> 0.32508817 #> + Management 0.22512409 #> + Moisture 0.20050225 #> + Manure 0.16723149 #> + A1 0.04626579 #> + Use 0.01799755 #> 0.00000000 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.84 0.002 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: R2.adj= 0.2251241 #> Call: dune ~ Management #> #> R2.adjusted #> + Moisture 0.3450334 #> 0.3250882 #> + Manure 0.2779515 #> + A1 0.2392216 #> + Use 0.2300349 #> 0.2251241 #> #> Call: rda(formula = dune ~ Management, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 29.2307 0.3475 3 #> Unconstrained 54.8930 0.6525 16 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 14.865 10.690 3.675 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 #> 15.270 8.428 6.899 5.675 3.988 3.121 2.588 2.380 1.818 1.376 0.995 #> PC12 PC13 PC14 PC15 PC16 #> 0.785 0.661 0.467 0.283 0.159 #>"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface given variable plots result ordination diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"# S3 method for default ordisurf(x, y, choices = c(1, 2), knots = 10, family = \"gaussian\", col = \"red\", isotropic = TRUE, thinplate = TRUE, bs = \"tp\", fx = FALSE, add = FALSE, display = \"sites\", w = weights(x, display), main, nlevels = 10, levels, npoints = 31, labcex = 0.6, bubble = FALSE, cex = 1, select = TRUE, method = \"REML\", gamma = 1, plot = TRUE, lwd.cl = par(\"lwd\"), ...) # S3 method for formula ordisurf(formula, data, ...) # S3 method for ordisurf calibrate(object, newdata, ...) # S3 method for ordisurf plot(x, what = c(\"contour\",\"persp\",\"gam\"), add = FALSE, bubble = FALSE, col = \"red\", cex = 1, nlevels = 10, levels, labcex = 0.6, lwd.cl = par(\"lwd\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"x ordisurf ordination configuration, either matrix result known scores. plot.ordisurf object class \"ordisurf\" returned ordisurf. y Variable plotted / modelled function ordination scores. choices Ordination axes. knots Number initial knots gam (one degrees freedom). knots = 0 knots = 1 function fit linear trend surface, knots = 2 function fit quadratic trend surface instead smooth surface. vector length 2 allowed isotropic = FALSE, first second elements knots referring first second ordination dimensions (indicated choices) respectively. family Error distribution gam. col Colour contours. isotropic, thinplate Fit isotropic smooth surface (.e. smoothness ordination dimensions) via gam. Use thinplate deprecated removed future version package. bs two letter character string indicating smoothing basis use. (e.g. \"tp\" thin plate regression spline, \"cr\" cubic regression spline). One c(\"tp\", \"ts\", \"cr\", \"cs\", \"ds\", \"ps\", \"ad\"). See smooth.terms view refer . default use thin plate splines: bs = \"tp\". fx indicates whether smoothers fixed degree freedom regression splines (fx = FALSE) penalised regression splines (fx = TRUE). Can vector length 2 anisotropic surfaces (isotropic = FALSE). make sense use fx = TRUE select = TRUE error . warning issued specify fx = TRUE forget use select = FALSE though fitting continues using select = FALSE. add Add contours existing diagram draw new plot? display Type scores known scores: typically \"sites\" ordinary site scores \"lc\" linear combination scores. w Prior weights data. Concerns mainly cca decorana results nonconstant weights. main main title plot, default name plotted variable new plot. nlevels, levels Either vector levels contours drawn, suggested number contours nlevels levels supplied. npoints numeric; number locations evaluate fitted surface. represents number locations dimension. labcex Label size contours. Setting zero suppress labels. bubble Use “bubble plot” points, vary point diameter value plotted variable. bubble numeric, value used maximum symbol size ( cex), bubble = TRUE, value cex gives maximum. minimum size always cex = 0.4. option effect add = FALSE. cex Character expansion plotting symbols. select Logical; specify gam argument \"select\". TRUE gam can add extra penalty term can penalized zero. means smoothing parameter estimation part fitting can completely remove terms model. corresponding smoothing parameter estimated zero extra penalty effect. method character; smoothing parameter estimation method. Options allowed : \"GCV.Cp\" uses GCV models unknown scale parameter Mallows' Cp/UBRE/AIC models known scale; \"GACV.Cp\" \"GCV.Cp\" uses GACV (Generalised Approximate CV) instead GCV; \"REML\" \"ML\" use restricted maximum likelihood maximum likelihood estimation known unknown scale; \"P-REML\" \"P-ML\" use REML ML estimation use Pearson estimate scale. gamma Multiplier inflate model degrees freedom GCV UBRE/AIC score . effectively places extra penalty complex models. oft-used value gamma = 1.4. plot logical; plotting done ordisurf? Useful want fitted response surface model. lwd.cl numeric; lwd (line width) parameter use drawing contour lines. formula, data Alternative definition fitted model x ~ y, left-hand side ordination x right-hand side single fitted continuous variable y. variable y must working environment data frame environment given data. arguments passed default method. object ordisurf result object. newdata Coordinates two-dimensional ordination new points. character; type plot produce. \"contour\" produces contour plot response surface, see contour details. \"persp\" produces perspective plot , see persp details. \"gam\" plots fitted GAM model, object inherits class \"gam\" returned ordisurf, see plot.gam. ... parameters passed scores, graphical functions. See Note exceptions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface using penalised splines (Wood 2003) gam, uses predict.gam find fitted values regular grid. smooth surface can fitted extra penalty allows entire smoother penalized back 0 degrees freedom, effectively removing term model (see Marra & Wood, 2011). addition extra penalty invoked setting argument select TRUE. alternative use spline basis includes shrinkage (bs = \"ts\" bs = \"cs\"). ordisurf() exposes large number options gam specifying basis functions used surface. stray defaults, read Notes section relevant documentation s smooth.terms. function plots fitted contours convex hull data points either existing ordination diagram draws new plot. select = TRUE smooth effectively penalised model, contours plotted. gam determines degree smoothness fitted response surface model fitting, unless fx = TRUE. Argument method controls gam performs smoothness selection. See gam details available options. Using \"REML\" \"ML\" yields p-values smooths best coverage properties things matter . function uses scores extract ordination scores, x can result object known function. user can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. behaviour consistent envfit. complete accordance constrained cca, set display = \"lc\". Function calibrate returns fitted values response variable. newdata must coordinates points fitted values desired. function based predict.gam pass extra arguments function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"ordisurf usually called side effect drawing contour plot. function returns result object class \"ordisurf\" inherits gam used internally fit surface, adds item grid contains data grid surface. item grid elements x y vectors axis coordinates, element z matrix fitted values contour. values outside convex hull observed points indicated NA z. gam component result can used analysis like predicting new values (see predict.gam).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Dave Roberts, Jari Oksanen Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"default use isotropic smoother via s employing thin plate regression splines (bs = \"tp\"). make sense ordination equal smoothing directions rotation invariant. However, different degrees smoothness along dimensions required, anisotropic smooth surface may applicable. can achieved use isotropic = FALSE, wherein surface fitted via tensor product smoother via te (unless bs = \"ad\", case separate splines dimension fitted using s). Cubic regression splines P splines can used isotropic = FALSE. Adaptive smooths (bs = \"ad\"), especially two dimensions, require large number observations; without many hundreds observations, default complexities smoother exceed number observations fitting fail. get old behaviour ordisurf use select = FALSE, method = \"GCV.Cp\", fx = FALSE, bs = \"tp\". latter two options current defaults. Graphical arguments supplied plot.ordisurf passed underlying plotting functions, contour, persp, plot.gam. exception arguments col cex can currently passed plot.gam bug way function evaluates arguments arranging plot. work-around call plot.gam directly result call ordisurf. See Examples illustration .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"fitted GAM regression model usual assumptions models. particular note assumption independence residuals. observations independent (e.g. repeat measures set objects, experimental design, inter alia) trust p-values GAM output. need control (.e. add additional fixed effects model, use complex smoothers), extract ordination scores using scores function generate gam call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Marra, G.P & Wood, S.N. (2011) Practical variable selection generalized additive models. Comput. Stat. Data Analysis 55, 2372--2387. Wood, S.N. (2003) Thin plate regression splines. J. R. Statist. Soc. B 65, 95--114.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(varespec) vare.mds <- monoMDS(vare.dist) ## IGNORE_RDIFF_BEGIN ordisurf(vare.mds ~ Baresoil, varechem, bubble = 5) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## as above but without the extra penalties on smooth terms, ## and using GCV smoothness selection (old behaviour of `ordisurf()`): ordisurf(vare.mds ~ Baresoil, varechem, col = \"blue\", add = TRUE, select = FALSE, method = \"GCV.Cp\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.34 total = 8.34 #> #> GCV score: 154.2897 ## Cover of Cladina arbuscula fit <- ordisurf(vare.mds ~ Cladarbu, varespec, family=quasipoisson) ## Get fitted values calibrate(fit) #> 18 15 24 27 23 19 22 #> 21.6245333 8.0344376 3.8614798 2.4595115 6.3958275 5.4260377 6.7158154 #> 16 28 13 14 20 25 7 #> 11.6759768 0.8041457 31.1726268 16.1942820 9.5897933 5.4219853 29.8214589 #> 5 6 3 4 2 9 12 #> 22.8500283 29.9936019 7.1653707 15.5309226 2.8467735 0.9051372 3.5541639 #> 10 11 21 #> 1.3099294 10.7783687 0.9177922 ## Variable selection via additional shrinkage penalties ## This allows non-significant smooths to be selected out ## of the model not just to a linear surface. There are 2 ## options available: ## - option 1: `select = TRUE` --- the *default* ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", select = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## - option 2: use a basis with shrinkage ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", bs = \"ts\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ts\", fx = FALSE) #> #> Estimated degrees of freedom: #> 6.26 total = 7.26 #> #> REML score: 98.83943 ## or bs = \"cs\" with `isotropic = FALSE` ## IGNORE_RDIFF_END ## Plot method plot(fit, what = \"contour\") ## Plotting the \"gam\" object plot(fit, what = \"gam\") ## 'col' and 'cex' not passed on ## or via plot.gam directly library(mgcv) #> Loading required package: nlme #> This is mgcv 1.9-1. For overview type 'help(\"mgcv-package\")'. plot.gam(fit, cex = 2, pch = 1, col = \"blue\") ## 'col' effects all objects drawn... ### controlling the basis functions used ## Use Duchon splines ordisurf(vare.mds ~ Baresoil, varechem, bs = \"ds\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ds\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94117 ## A fixed degrees of freedom smooth, must use 'select = FALSE' ordisurf(vare.mds ~ Baresoil, varechem, knots = 4, fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 4, bs = \"tp\", fx = TRUE) #> #> Estimated degrees of freedom: #> 3 total = 4 #> #> REML score: 85.55126 ## An anisotropic smoother with cubic regression spline bases ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cr\", knots = 4) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(4, 4), bs = c(\"cr\", \"cr\"), fx = c(FALSE, #> FALSE)) #> #> Estimated degrees of freedom: #> 3.08 total = 4.08 #> #> REML score: 93.02782 ## An anisotropic smoother with cubic regression spline with ## shrinkage bases & different degrees of freedom in each dimension ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cs\", knots = c(3,4), fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(3, 4), bs = c(\"cs\", \"cs\"), fx = c(TRUE, #> TRUE)) #> #> Estimated degrees of freedom: #> 11 total = 12 #> #> REML score: 42.9834"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text or Points to Ordination Plots — orditorp","title":"Add Text or Points to Ordination Plots — orditorp","text":"function adds text points ordination plots. Text used can done without overwriting text labels, points used otherwise. function can help reducing clutter ordination graphics, manual editing may still necessary.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"orditorp(x, display, labels, choices = c(1, 2), priority, select, cex = 0.7, pcex, col = par(\"col\"), pcol, pch = par(\"pch\"), air = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text or Points to Ordination Plots — orditorp","text":"x result object ordination ordiplot result. display Items displayed plot. one alternative allowed. Typically \"sites\" \"species\". labels Optional text used labels. Row names used missing. choices Axes shown. priority Text used items higher priority labels overlap. vector length number items plotted. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. logical vector used, must length scores plotted. cex, pcex Text point sizes, see plot.default.. col, pcol Text point colours, see plot.default. pch Plotting character, see points. air Amount empty space text labels. Values <1 allow overlapping text. ... arguments scores (various methods), text points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text or Points to Ordination Plots — orditorp","text":"Function orditorp add either text points existing plot. items high priority added first text used can done without overwriting previous labels,points used otherwise. priority missing, labels added outskirts centre. Function orditorp can used ordination results, plotting results ordiplot ordination plot functions (plot.cca, plot.decorana, plot.metaMDS). Arguments can passed relevant scores method ordination object (x) drawn. See relevant scores help page arguments can used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add Text or Points to Ordination Plots — orditorp","text":"function returns invisibly logical vector TRUE means item labelled text FALSE means marked point. returned vector can used select argument ordination text points functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text or Points to Ordination Plots — orditorp","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"## A cluttered ordination plot : data(BCI) mod <- cca(BCI) plot(mod, dis=\"sp\", type=\"t\") # Now with orditorp and abbreviated species names cnam <- make.cepnames(names(BCI)) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, pch=\"+\", pcol=\"grey\") ## show select in action set.seed(1) take <- sample(ncol(BCI), 50) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, select = take, pch=\"+\", pcol=\"grey\") # \\dontshow{ ## example(orditorp) should not set random seed in the user session rm(.Random.seed) #> Warning: object '.Random.seed' not found # }"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Trellis (Lattice) Plots for Ordination — ordixyplot","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Functions ordicloud, ordisplom ordixyplot provide interface plot ordination results using Trellis functions cloud, splom xyplot package lattice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"ordixyplot(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", aspect = \"iso\", envfit, type = c(\"p\", \"biplot\"), ...) ordisplom(x, data=NULL, formula = NULL, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", type = \"p\", ...) ordicloud(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi3d\", prepanel = \"prepanel.ordi3d\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"x ordination result scores knows: ordination result vegan many others. data Optional data amend ordination results. ordination results found x, may give data variables needed plots. Typically environmental data. formula Formula define plots. default formula used omitted. ordination axes must called names ordination results (names vary among methods). ordisplom, special character . refers ordination result. display kind scores: argument passed scores. choices axes selected: argument passed scores. panel, prepanel names panel prepanel functions. aspect aspect plot (passed lattice function). envfit Result envfit function displayed ordixyplot. Please note needs choices ordixyplot. type type plot. knows alternatives panel.xyplot. addition ordixyplot alternatives \"biplot\", \"arrows\" \"polygon\". first displays fitted vectors factor centroids envfit, constrained ordination, biplot arrows factor centroids envfit given. second (type = \"arrows\") trellis variant ordiarrows draws arrows groups. line parameters controlled trellis.par.set superpose.line, user can set length, angle ends parameters panel.arrows. last one (type = \"polygon\") draws polygon enclosing points panel polygon enclosing points data. overall polygon controlled trellis.par.set plot.polygon, panel polygon controlled superpose.polygon. ... Arguments passed scores methods lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"functions provide interface corresponding lattice functions. graphical parameters passed lattice function graphs extremely configurable. See Lattice xyplot, splom cloud details, usage possibilities. argument x must always ordination result. scores extracted vegan function scores functions work vegan ordinations many others. formula used define models. functions simple default formulae used formula missing. formula omitted ordisplom produces pairs plot ordination axes variables data. formula given, ordination results must referred . variables names. functions, formula must use names ordination scores names data. ordination scores found x, data optional. data contain variables ordination scores used plots. Typically, environmental variables (typically factors) define panels plot symbols. proper work done panel function. layout can changed defining panel functions. See panel.xyplot, panel.splom panel.cloud details survey possibilities. Ordination graphics always isometric: scale used axes. controlled (can changed) argument aspect ordixyplot. ordicloud isometric scaling defined panel prepanel functions. must replace functions want non-isometric scaling graphs. select isometric scaling ordisplom.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"function return Lattice objects class \"trellis\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"data(dune, dune.env) ord <- cca(dune) ## Pairs plots ordisplom(ord) ordisplom(ord, data=dune.env, choices=1:2) ordisplom(ord, data=dune.env, form = ~ . | Management, groups=Manure) ## Scatter plot with polygons ordixyplot(ord, data=dune.env, form = CA1 ~ CA2 | Management, groups=Manure, type = c(\"p\",\"polygon\")) ## Choose a different scaling ordixyplot(ord, scaling = \"symmetric\") ## ... Slices of third axis ordixyplot(ord, form = CA1 ~ CA2 | equal.count(CA3, 4), type = c(\"g\",\"p\", \"polygon\")) ## Display environmental variables ordixyplot(ord, envfit = envfit(ord ~ Management + A1, dune.env, choices=1:3)) ## 3D Scatter plots ordicloud(ord, form = CA2 ~ CA3*CA1, groups = Manure, data = dune.env) ordicloud(ord, form = CA2 ~ CA3*CA1 | Management, groups = Manure, data = dune.env, auto.key = TRUE, type = c(\"p\",\"h\"))"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Coordinates of Neighbourhood Matrix — pcnm","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"function computed classical PCNM principal coordinate analysis truncated distance matrix. commonly used transform (spatial) distances rectangular data suitable constrained ordination regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"pcnm(dis, threshold, w, dist.ret = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"dis distance matrix. threshold threshold value truncation distance. missing, minimum distance giving connected network used. found longest distance minimum spanning tree dis. w Prior weights rows. dist.ret Return distances used calculate PCNMs.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Principal Coordinates Neighbourhood Matrix (PCNM) map distances rows onto rectangular matrix rows using truncation threshold long distances (Borcard & Legendre 2002). original distances Euclidean distances two dimensions (like normal spatial distances), mapped onto two dimensions truncation distances. truncation, higher number principal coordinates. selection truncation distance huge influence PCNM vectors. default use longest distance keep data connected. distances truncation threshold given arbitrary value 4 times threshold. regular data, first PCNM vectors show wide scale variation later PCNM vectors show smaller scale variation (Borcard & Legendre 2002), irregular data interpretation clear. PCNM functions used express distances rectangular form similar normal explanatory variables used , e.g., constrained ordination (rda, cca dbrda) univariate regression (lm) together environmental variables (row weights supplied cca; see Examples). regarded powerful method forcing rectangular environmental data distances using partial mantel analysis (mantel.partial) together geographic distances (Legendre et al. 2008, see Tuomisto & Ruokolainen 2008). function based pcnm function Dray's unreleased spacemakeR package. differences current function uses spantree internal support function. current function also can use prior weights rows using weighted metric scaling wcmdscale. use row weights allows finding orthonormal PCNMs also correspondence analysis (e.g., cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"list following elements: values Eigenvalues obtained principal coordinates analysis. vectors Eigenvectors obtained principal coordinates analysis. scaled unit norm. vectors can extracted scores function. default return PCNM vectors, argument choices selects given vectors. threshold Truncation distance. dist distance matrix values threshold replaced arbitrary value four times threshold. String \"pcnm\" added method attribute, new attribute threshold added distances. returned dist.ret = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Borcard D. Legendre P. (2002) -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153, 51--68. Legendre, P., Borcard, D Peres-Neto, P. (2008) Analyzing explaining beta diversity? Comment. Ecology 89, 3238--3244. Tuomisto, H. & Ruokolainen, K. (2008) Analyzing explaining beta diversity? reply. Ecology 89, 3244--3256.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Jari Oksanen, based code Stephane Dray.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"## Example from Borcard & Legendre (2002) data(mite.xy) pcnm1 <- pcnm(dist(mite.xy)) op <- par(mfrow=c(1,3)) ## Map of PCNMs in the sample plot ordisurf(mite.xy, scores(pcnm1, choi=1), bubble = 4, main = \"PCNM 1\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.71 total = 9.71 #> #> REML score: -120.7705 ordisurf(mite.xy, scores(pcnm1, choi=2), bubble = 4, main = \"PCNM 2\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.18 total = 8.18 #> #> REML score: -103.4662 ordisurf(mite.xy, scores(pcnm1, choi=3), bubble = 4, main = \"PCNM 3\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.32 total = 9.32 #> #> REML score: -94.19053 par(op) ## Plot first PCNMs against each other ordisplom(pcnm1, choices=1:4) ## Weighted PCNM for CCA data(mite) rs <- rowSums(mite)/sum(mite) pcnmw <- pcnm(dist(mite.xy), w = rs) ord <- cca(mite ~ scores(pcnmw)) ## Multiscale ordination: residual variance should have no distance ## trend msoplot(mso(ord, mite.xy))"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":null,"dir":"Reference","previous_headings":"","what":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Individual (count data) incidence (presence-absence data) based null models can generated community level simulations. Options preserving characteristics original matrix (rows/columns sums, matrix fill) restricted permutations (based strata) discussed Details section.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"permatfull(m, fixedmar = \"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, ...) permatswap(m, method = \"quasiswap\", fixedmar=\"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, burnin = 0, thin = 1, ...) # S3 method for permat print(x, digits = 3, ...) # S3 method for permat summary(object, ...) # S3 method for summary.permat print(x, digits = 2, ...) # S3 method for permat plot(x, type = \"bray\", ylab, xlab, col, lty, lowess = TRUE, plot = TRUE, text = TRUE, ...) # S3 method for permat lines(x, type = \"bray\", ...) # S3 method for permat as.ts(x, type = \"bray\", ...) # S3 method for permat toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"m community data matrix plots (samples) rows species (taxa) columns. fixedmar character, stating row/column sums preserved (\"none\", \"rows\", \"columns\", \"\"). strata Numeric vector factor length nrow(m) grouping rows within strata restricted permutations. Unique values levels used. mtype Matrix data type, either \"count\" count data, \"prab\" presence-absence type incidence data. times Number permuted matrices. method Character method used swap algorithm (\"swap\", \"tswap\", \"quasiswap\", \"backtrack\") described function make.commsim. mtype=\"count\" \"quasiswap\", \"swap\", \"swsh\" \"abuswap\" methods available (see details). shuffle Character, indicating whether individuals (\"ind\"), samples (\"samp\") (\"\") shuffled, see details. burnin Number null communities discarded proper analysis sequential (\"swap\", \"tswap\") methods. thin Number discarded permuted matrices two evaluations sequential (\"swap\", \"tswap\") methods. x, object Object class \"permat\" digits Number digits used rounding. ylab, xlab, col, lty graphical parameters plot method. type Character, type plot displayed: \"bray\" Bray-Curtis dissimilarities, \"chisq\" Chi-squared values. lowess, plot, text Logical arguments plot method, whether locally weighted regression curve drawn, plot drawn, statistic values printed plot. ... arguments passed simulate.nullmodel methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"function permatfull useful matrix fill allowed vary, matrix type count. fixedmar argument used set constraints permutation. none margins fixed, cells randomised within matrix. rows columns fixed, cells within rows columns randomised, respectively. margins fixed, r2dtable function used based Patefield's (1981) algorithm. presence absence data, matrix fill necessarily fixed, permatfull wrapper function make.commsim. r00, r0, c0, quasiswap algorithms make.commsim used \"none\", \"rows\", \"columns\", \"\" values fixedmar argument, respectively shuffle argument effect mtype = \"count\" permatfull function used \"none\", \"rows\", \"columns\" values fixedmar. cases count data individual based randomisations. \"samp\" \"\" options result fixed matrix fill. \"\" option means individuals shuffled among non zero cells ensuring cell zeros result, cell (zero new valued cells) shuffled. function permatswap useful matrix fill (.e. proportion empty cells) row/columns sums kept constant. permatswap uses different kinds swap algorithms, row columns sums fixed cases. presence-absence data, swap tswap methods make.commsim can used. count data, special swap algorithm ('swapcount') implemented results permuted matrices fixed marginals matrix fill time. 'quasiswapcount' algorithm (method=\"quasiswap\" mtype=\"count\") uses trick Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function retaining row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping 2x2 submatrices (see 'swapcount' algorithm details) maintain row column totals. algorithm generates independent matrices step, burnin thin arguments considered. default method, sequential (swapcount ) independence subsequent matrices checked. swapcount algorithm (method=\"swap\" mtype=\"count\") tries find 2x2 submatrices (identified 2 random row 2 random column indices), can swapped order leave column row totals fill unchanged. First, algorithm finds largest value submatrix can swapped (\\(d\\)) whether diagonal antidiagonal way. Submatrices contain values larger zero either diagonal antidiagonal position can swapped. Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. algorithm sequential, subsequent matrices independent, swaps modify little matrix large. cases many burnin steps thinning needed get independent random matrices. Although algorithm implemented C, large burnin thin values can slow considerably. WARNING: according simulations, algorithm seems biased non random, thus use avoided! algorithm \"swsh\" function permatswap hybrid algorithm. First, makes binary quasiswaps keep row column incidences constant, non-zero values modified according shuffle argument (\"samp\" \"\" available case, applied non-zero values). also recognizes fixedmar argument \"\" (vegan versions <= 2.0 algorithm fixedmar = \"none\"). algorithm \"abuswap\" produces two kinds null models (based fixedmar=\"columns\" fixedmar=\"rows\") described Hardy (2008; randomization scheme 2x 3x, respectively). preserve column row occurrences, column row sums time. (Note similar constraints can achieved non sequential \"swsh\" algorithm fixedmar argument set \"columns\" \"rows\", respectively.) Constraints row/column sums, matrix fill, total sum sums within strata can checked summary method. plot method visually testing randomness permuted matrices, especially sequential swap algorithms. tendency graph, higher burnin thin values can help sequential methods. New lines can added existing plot lines method. Unrestricted restricted permutations: strata NULL, functions perform unrestricted permutations. Otherwise, used restricted permutations. strata contain least 2 rows order perform randomization (case low row numbers, swap algorithms can rather slow). design well balanced (.e. number observations within stratum), permuted matrices may biased constraints forced submatrices different dimensions. often means, number potential permutations decrease dimensions. constraints put, less randomness can expected. plot method useful graphically testing trend independence permuted matrices. especially important using sequential algorithms (\"swap\", \"tswap\", \"abuswap\"). .ts method can used extract Bray-Curtis dissimilarities Chi-squared values time series. can used testing independence (see Examples). method toCoda useful accessing diagnostic tools available coda package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Functions permatfull permatswap return object class \"permat\" containing function call (call), original data matrix used permutations (orig) list permuted matrices length times (perm). summary method returns various statistics list (including mean Bray-Curtis dissimilarities calculated pairwise among original permuted matrices, Chi-square statistics, check results constraints; see Examples). Note strata used original call, summary calculation may take longer. plot creates plot side effect. .ts method returns object class \"ts\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Original references presence-absence algorithms given help page make.commsim. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Péter Sólymos, solymos@ualberta.ca Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"## A simple artificial community data matrix. m <- matrix(c( 1,3,2,0,3,1, 0,2,1,0,2,1, 0,0,1,2,0,3, 0,0,0,1,4,3 ), 4, 6, byrow=TRUE) ## Using the quasiswap algorithm to create a ## list of permuted matrices, where ## row/columns sums and matrix fill are preserved: x1 <- permatswap(m, \"quasiswap\") summary(x1) #> Summary of object of class 'permat' #> #> Call: permatswap(m = m, method = \"quasiswap\") #> #> Matrix type: count #> Permutation type: swap #> Method: quasiswap_count, burnin: 0, thin: 1 #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 13.13 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3667 0.4333 0.4145 0.4667 0.6000 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 16.27 19.59 21.10 21.51 23.36 31.69 ## Unrestricted permutation retaining ## row/columns sums but not matrix fill: x2 <- permatfull(m) summary(x2) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 16.16 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 0 % #> Column incidences retained: 1.01 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.3667 0.3865 0.4333 0.6333 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 7.824 12.046 15.660 16.015 19.413 28.132 ## Unrestricted permutation of presence-absence type ## not retaining row/columns sums: x3 <- permatfull(m, \"none\", mtype=\"prab\") x3$orig ## note: original matrix is binarized! #> [,1] [,2] [,3] [,4] [,5] [,6] #> [1,] 1 1 1 0 1 1 #> [2,] 0 1 1 0 1 1 #> [3,] 0 0 1 1 0 1 #> [4,] 0 0 0 1 1 1 summary(x3) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, fixedmar = \"none\", mtype = \"prab\") #> #> Matrix type: prab #> Permutation type: full #> Method: r00 #> Restricted: FALSE #> Fixed margins: none #> Individuals and samples are shuffled #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 15 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 4.04 % #> Column sums retained: 0 % #> Row incidences retained: 4.04 % #> Column incidences retained: 0 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.4000 0.3852 0.4000 0.5333 #> #> Chi-squared for original matrix: 8.4 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 8.812 13.583 15.208 15.295 17.083 21.896 ## Restricted permutation, ## check sums within strata: x4 <- permatfull(m, strata=c(1,1,2,2)) summary(x4) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, strata = c(1, 1, 2, 2)) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: TRUE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 38.38 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 2.02 % #> Sums within strata retained: 100 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.06667 0.20000 0.26667 0.23502 0.26667 0.46667 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 14.21 18.68 21.05 22.09 25.26 36.50 ## NOTE: 'times' argument usually needs to be >= 99 ## here much lower value is used for demonstration ## Not sequential algorithm data(BCI) a <- permatswap(BCI, \"quasiswap\", times=19) ## Sequential algorithm b <- permatswap(BCI, \"abuswap\", fixedmar=\"col\", burnin=0, thin=100, times=19) opar <- par(mfrow=c(2,2)) plot(a, main=\"Not sequential\") plot(b, main=\"Sequential\") plot(a, \"chisq\") plot(b, \"chisq\") par(opar) ## Extract Bray-Curtis dissimilarities ## as time series bc <- as.ts(b) ## Lag plot lag.plot(bc) ## First order autoregressive model mar <- arima(bc, c(1,0,0)) mar #> #> Call: #> arima(x = bc, order = c(1, 0, 0)) #> #> Coefficients: #> ar1 intercept #> 0.9915 0.1850 #> s.e. 0.0120 0.1374 #> #> sigma^2 estimated as 0.000346: log likelihood = 46.71, aic = -87.42 ## Ljung-Box test of residuals Box.test(residuals(mar)) #> #> \tBox-Pierce test #> #> data: residuals(mar) #> X-squared = 0.35725, df = 1, p-value = 0.55 #> ## Graphical diagnostics tsdiag(mar)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract, Analyse and Display Permutation Results — permustats","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results vegan functions. support functions can find quantiles standardized effect sizes, plot densities Q-Q plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"permustats(x, ...) # S3 method for permustats summary(object, interval = 0.95, alternative, ...) # S3 method for permustats densityplot(x, data, xlab = \"Permutations\", ...) # S3 method for permustats density(x, observed = TRUE, ...) # S3 method for permustats qqnorm(y, observed = TRUE, ...) # S3 method for permustats qqmath(x, data, observed = TRUE, sd.scale = FALSE, ylab = \"Permutations\", ...) # S3 method for permustats boxplot(x, scale = FALSE, names, ...) # S3 method for permustats pairs(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract, Analyse and Display Permutation Results — permustats","text":"object, x, y object handled. interval numeric; coverage interval reported. alternative character string specifying limits used interval direction test evaluating \\(p\\)-values. Must one \"two.sided\" (upper lower limit), \"greater\" (upper limit), \"less\" (lower limit). Usually alternative given result object, can specified argument. xlab, ylab Arguments densityplot qqmath functions. observed Add observed statistic among permutations. sd.scale Scale permutations unit standard deviation observed statistic standardized effect size. data Ignored. scale Use standardized effect size (SES). names Names boxes (default: names statistics). ... arguments passed function. density passed density.default, boxplot boxplot.default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results observed statistics several vegan functions perform permutations simulations. summary method permustats estimates standardized effect sizes (SES) difference observed statistic mean permutations divided standard deviation permutations (also known \\(z\\)-values). also prints mean, median, limits contain interval percent permuted values. default (interval = 0.95), two-sided test (2.5%, 97.5%) one-sided tests either 5% 95% quantile \\(p\\)-value depending test direction. mean, quantiles \\(z\\) values evaluated permuted values without observed statistic, \\(p\\)-value evaluated observed statistic. intervals \\(p\\)-value evaluated test direction original test, can changed argument alternative. Several permustats objects can combined c function. c function checks statistics equal, performs sanity tests. density densityplot methods display kernel density estimates permuted values. observed value statistic included permuted values, densityplot method marks observed statistic vertical line. However density method uses standard plot method mark observed value. qqnorm qqmath display Q-Q plots permutations, optionally together observed value (default) shown horizontal line plots. qqnorm plots permutation values standard Normal variate. qqmath defaults standard Normal well, can accept alternatives (see standard qqmath). qqmath function can also plot observed statistic standardized effect size (SES) standandized permutations (argument sd.scale). permutations standardized without observed statistic, similarly summary. Functions density qqnorm based standard R methods accept arguments. handle one statistic, used several test statistic evaluated. densityplot qqmath lattice graphics, can used either one several statistics. functions pass arguments underlying functions; see documentation. Functions qqmath densityplot default use axis scaling subplots lattice. can use argument scales set independent scaling subplots appropriate (see xyplot exhaustive list arguments). Function boxplot draws box--whiskers plots effect size, difference permutations observed statistic. scale = TRUE, permutations standardized unit standard deviation, plot show standardized effect sizes. Function pairs plots permutation values statistics . function passes extra arguments pairs. permustats can extract permutation statistics results adonis2, anosim, anova.cca, mantel, mantel.partial, mrpp, oecosimu, ordiareatest, permutest.cca, protest, permutest.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function returns object class \"permustats\". list items \"statistic\" observed statistics, permutations contains permuted values, alternative contains text defining character test (\"two.sided\", \"less\" \"greater\"). qqnorm density methods return standard result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract, Analyse and Display Permutation Results — permustats","text":"Jari Oksanen contributions Gavin L. Simpson (permustats.permutest.betadisper method related modifications summary.permustats print method) Eduard Szöcs (permustats.anova.cca).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"data(dune, dune.env) mod <- adonis2(dune ~ Management + A1, data = dune.env) ## use permustats perm <- permustats(mod) summary(perm) #> #> statistic SES mean lower median upper Pr(perm) #> Management 3.0730 4.6870 1.0387 0.9565 1.8217 0.004 ** #> A1 2.7676 2.7175 1.0022 0.8484 2.2436 0.028 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.95) densityplot(perm) qqmath(perm) boxplot(perm, scale=TRUE, lty=1, pch=16, cex=0.6, col=\"hotpink\", ylab=\"SES\") abline(h=0, col=\"skyblue\") ## example of multiple types of statistic mod <- with(dune.env, betadisper(vegdist(dune), Management)) pmod <- permutest(mod, nperm = 99, pairwise = TRUE) perm <- permustats(pmod) summary(perm, interval = 0.90) #> #> statistic SES mean lower median upper Pr(perm) #> Overall (F) 1.9506 0.7173 1.1427 0.8211 2.4909 0.154 #> BF-HF (t) -0.5634 -0.4124 -0.0443 -2.0202 -0.0293 1.8851 0.591 #> BF-NM (t) -2.2387 -1.8672 -0.0045 -1.8423 0.0074 2.0628 0.067 . #> BF-SF (t) -1.1675 -0.9341 -0.0086 -1.9337 -0.0450 1.9486 0.283 #> HF-NM (t) -2.1017 -1.9328 0.0277 -1.6716 0.0346 1.7582 0.067 . #> HF-SF (t) -0.8789 -0.7872 0.0321 -1.8598 0.0284 1.8394 0.379 #> NM-SF (t) 0.9485 0.8265 0.0121 -1.9118 0.0690 1.7827 0.379 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.9)"},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation tests in Vegan — permutations","title":"Permutation tests in Vegan — permutations","text":"version 2.2-0, vegan significantly improved access restricted permutations brings line offered Canoco. permutation designs modelled permutation schemes Canoco 3.1 (ter Braak, 1990). vegan currently provides following features within permutation tests: Free permutation DATA, also known randomisation, Free permutation DATA within levels grouping variable, Restricted permutations line transects time series, Permutation groups samples whilst retaining within-group ordering, Restricted permutations spatial grids, Blocking, samples never permuted blocks, Split-plot designs, permutation whole plots, split plots, . , use DATA mean either observed data function data, example residuals ordination model presence covariables. capabilities provided functions permute package. user can request particular type permutation supplying permutations argument function object returned , defines samples permuted. Alternatively, user can simply specify required number permutations simple randomisation procedure performed. Finally, user can supply matrix permutations ( number rows equal number permutations number columns equal number observations data) vegan use permutations instead generating new permutations. majority functions vegan allow full range possibilities outlined . Exceptions include kendall.post kendall.global. Null hypothesis first two types permutation test listed assumes free exchangeability DATA (within levels grouping variable, specified). Dependence observations, arises due spatial temporal autocorrelation, -complicated experimental designs, split-plot designs, violates fundamental assumption test requires complex restricted permutation test designs. designs available via permute package vegan provides access version 2.2-0 onwards. Unless otherwise stated help pages specific functions, permutation tests vegan follow format/structure: appropriate test statistic chosen. statistic chosen described help pages individual functions. value test statistic evaluate observed data analysis/model recorded. Denote value \\(x_0\\). DATA randomly permuted according one schemes, value test statistic permutation evaluated recorded. Step 3 repeated total \\(n\\) times, \\(n\\) number permutations requested. Denote values \\(x_i\\), \\(= 1, ..., n\\) Count number values test statistic, \\(x_i\\), Null distribution extreme test statistic observed data \\(x_0\\). Denote count \\(N\\). use phrase extreme include cases two-sided test performed large negative values test statistic considered. permutation p-value computed $$p = \\frac{N + 1}{n + 1}$$ description illustrates default number permutations specified vegan functions takes values 199 999 example. Pretty p values achieved \\(+ 1\\) denominator results division 200 1000, 199 999 random permutations used test. simple intuition behind presence \\(+ 1\\) numerator denominator represent inclusion observed value statistic Null distribution (e.g. Manly 2006). Phipson & Smyth (2010) present compelling explanation inclusion \\(+ 1\\) numerator denominator p value calculation. Fisher (1935) mind permutation test involve enumeration possible permutations data yielding exact test. However, complete enumeration may feasible practice owing potentially vast number arrangements data, even modestly-sized data sets free permutation samples. result evaluate p value tail probability Null distribution test statistic directly random sample possible permutations. Phipson & Smyth (2010) show naive calculation permutation p value $$p = \\frac{N}{n}$$ leads invalid test incorrect type error rate. go show replacing unknown tail probability ( p value) Null distribution biased estimator $$p = \\frac{N + 1}{n + 1}$$ positive bias induced just right size account uncertainty estimation tail probability set randomly sampled permutations yield test correct type error rate. estimator described correct situation permutations data samples randomly without replacement. strictly happens vegan permutations drawn pseudo-randomly independent one another. Note actual chance happening practice small functions permute guarantee generate unique set permutations unless complete enumeration permutations requested. feasible smallest data sets restrictive permutation designs, cases chance drawing set permutations repeats lessened sample size, thence size set possible permutations, increases. situation sampling permutations replacement , tail probability \\(p\\) calculated biased estimator described somewhat conservative, large amount depends number possible values test statistic can take permutation data (Phipson & Smyth, 2010). represents slight loss statistical power conservative p value calculation used . However, unless sample sizes small permutation design set values test statistic can take also small, loss power unlikely critical. minimum achievable p-value $$p_{\\mathrm{min}} = \\frac{1}{n + 1}$$ hence depends number permutations evaluated. However, one simply increase number permutations (\\(n\\)) achieve potentially lower p-value unless number observations available permits number permutations. unlikely problem smallest data sets free permutation (randomisation) valid, restricted permutation designs low number observations, may many unique permutations data might desire reach required level significance. currently responsibility user determine total number possible permutations DATA. number possible permutations allowed specified design can calculated using numPerms permute package. Heuristics employed within shuffleSet function used vegan can triggered generate entire set permutations instead random set. settings controlling triggering complete enumeration step contained within permutation design created using link[permute]{} can set user. See details. Limits total number permutations DATA severe temporally spatially ordered data experimental designs low replication. example, time series \\(n = 100\\) observations just 100 possible permutations including observed ordering. situations low number permutations possible due nature DATA experimental design, enumeration permutations becomes important achievable computationally. , provided brief overview capabilities vegan permute. get best new functionality details set permutation designs using , consult vignette Restricted permutations; using permute package supplied permute accessible via vignette(\"permutations\", package = \"permute\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"random-number-generation","dir":"Reference","previous_headings":"","what":"Random Number Generation","title":"Permutation tests in Vegan — permutations","text":"permutations based random number generator provided R. may change R releases change permutations vegan test results. One change R release 3.6.0. new version clearly better permutation tests use . However, need reproduce old results, can set R random number generator previous version RNGversion.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation tests in Vegan — permutations","text":"Manly, B. F. J. (2006). Randomization, Bootstrap Monte Carlo Methods Biology, Third Edition. Chapman Hall/CRC. Phipson, B., & Smyth, G. K. (2010). Permutation P-values never zero: calculating exact P-values permutations randomly drawn. Statistical Applications Genetics Molecular Biology, 9, Article 39. DOI: 10.2202/1544-6115.1585 ter Braak, C. J. F. (1990). Update notes: CANOCO version 3.1. Wageningen: Agricultural Mathematics Group. (UR). See also: Davison, . C., & Hinkley, D. V. (1997). Bootstrap Methods Application. Cambridge University Press.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation tests in Vegan — permutations","text":"Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Implements permutation-based test multivariate homogeneity group dispersions (variances) results call betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"# S3 method for betadisper permutest(x, pairwise = FALSE, permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"x object class \"betadisper\", result call betadisper. pairwise logical; perform pairwise comparisons group means? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. ... Arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance F. alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution F Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can performed setting argument pairwise TRUE. classical t test performed pairwise group dispersions. combined permutation test based t statistic calculated pairwise group dispersions. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"permutest.betadisper returns list class \"permutest.betadisper\" following components: tab ANOVA table object inheriting class \"data.frame\". pairwise list components observed permuted containing observed permuted p-values pairwise comparisons group mean distances (dispersions variances). groups character; levels grouping factor. control list, result call .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62(1), 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9(6), 683--693.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F pmod <- permutest(mod, permutations = 99, pairwise = TRUE) ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Has permustats() method pstat <- permustats(pmod) densityplot(pstat, scales = list(x = list(relation = \"free\"))) qqmath(pstat, scales = list(relation = \"free\"))"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Functions plot extract results constrained correspondence analysis (cca), redundancy analysis (rda), distance-based redundancy analysis (dbrda) constrained analysis principal coordinates (capscale).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"# S3 method for cca plot(x, choices = c(1, 2), display = c(\"sp\", \"wa\", \"cn\"), scaling = \"species\", type, xlim, ylim, const, correlation = FALSE, hill = FALSE, cex = 0.7, ...) # S3 method for cca text(x, display = \"sites\", labels, choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca points(x, display = \"sites\", choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", hill = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for rda scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", const, correlation = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for cca summary(object, digits = max(3, getOption(\"digits\") - 3), ...) # S3 method for cca labels(object, display, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"x, object cca result object. choices Axes shown. display Scores shown. must include alternatives \"species\" \"sp\" species scores, sites \"wa\" site scores, \"lc\" linear constraints LC scores, \"bp\" biplot arrows \"cn\" centroids factor constraints instead arrow, \"reg\" regression coefficients (.k.. canonical coefficients). alternative \"\" selects available scores. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. Corresponding negative values can used cca additionally multiply results \\(\\sqrt(1/(1-\\lambda))\\). scaling know Hill scaling (although nothing Hill's rescaling decorana). corresponding negative values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Arguments correlation hill scores.rda scores.cca respectively can used combination character descriptions get corresponding negative value. correlation, hill logical; scaling character description scaling type, correlation hill used select corresponding negative scaling type; either correlation-like scores Hill's scaling PCA/RDA CA/CCA respectively. See argument scaling details. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score, names variable label, weights (CCA) variable weight. possible values score species, sites (WA scores), constraints (LC scores sites calculated directly constraining variables), biplot (biplot arrows), centroids (levels factor variables), factorbiplot (biplot arrows model centroids), regression (regression coefficients find LC scores constraints). scores used conventional plot, directly suitable used ggplot2 package. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. xlim, ylim x y limits (min,max) plot. labels Optional text used instead row names. use , good check default labels order using labels command. arrow.mul Factor expand arrows graph. Arrows scaled automatically fit graph missing. head.arrow Default length arrow heads. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. const General scaling constant rda scores. default use constant gives biplot scores, , scores approximate original data (see vignette ‘Design Decisions’ browseVignettes(\"vegan\") details discussion). const vector two items, first used species, second item site scores. droplist Return matrix instead named list one kind scores requested. axis.bp Draw axis biplot arrows. digits Number digits output. cex Character expansion. ... Parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function used cca rda. produces quick, standard plot current scaling. plot function sets colours (col), plotting characters (pch) character sizes (cex) certain standard values. fuller control produced plot, best call plot type=\"none\" first, add plotting item separately using text.cca points.cca functions. use default settings standard text points functions accept parameters, allowing full user control produced plots. Environmental variables receive special treatment. display=\"bp\", arrows drawn. labelled text unlabelled points. arrows basically unit scaling, sites scaled (scaling \"sites\" \"symmetric\"), scores requested axes adjusted relative axis highest eigenvalue. scaling = \"species\" scaling = \"none\", arrows consistent vectors fitted linear combination scores (display = \"lc\" function envfit), scaling alternatives differ. basic plot function uses simple heuristics adjusting unit-length arrows current plot area, user can give expansion factor mul.arrow. display=\"cn\" centroids levels factor variables displayed (available factors formula interface used cca rda). option continuous variables still presented arrows ordered factors arrows centroids. display = \"reg\" arrows drawn regression coefficients (.k.. canonical coefficients) constraints conditions. Biplot arrows can interpreted individually, regression coefficients must interpreted together: LC score site sum regressions displayed arrows. partialled conditions zero shown biplot arrows, shown regressions, show effect must partialled get LC scores. biplot arrows standard easily interpreted, regression arrows used know need . want better control plots, best construct plot text points commands accept graphical parameters. important remember use scaling, correlation hill arguments calls. plot.cca command returns invisibly ordiplot result object, consistent scaling elements. also possible use R pipes (|>) maintain consistent scaling setting. easiest way full control graphics first set plot frame using plot type = \"n\" needed scores display save result use first command pipe. points text commands ordiplot allow full graphical control (see section Examples). Utility function labels returns default labels order applied text. Palmer (1993) suggested using linear constraints (“LC scores”) ordination diagrams, gave better results simulations site scores (“WA scores”) step constrained unconstrained analysis. However, McCune (1997) showed noisy environmental variables (environmental measurements noisy) destroy “LC scores” whereas “WA scores” little affected. Therefore plot function uses site scores (“WA scores”) default. consistent usage statistics functions R (lda, cancor).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function returns invisibly plotting structure can used function identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Moisture + Management, dune.env) ## better control -- remember to set scaling etc identically plot(mod, type=\"n\", scaling=\"sites\") text(mod, dis=\"cn\", scaling=\"sites\") points(mod, pch=21, col=\"red\", bg=\"yellow\", cex=1.2, scaling=\"sites\") text(mod, \"species\", col=\"blue\", cex=0.8, scaling=\"sites\") ## catch the invisible result and use ordiplot support - the example ## will make a biplot with arrows for species and correlation scaling pca <- rda(dune) pl <- plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) with(dune.env, points(pl, \"site\", pch=21, col=1, bg=Management)) text(pl, \"sp\", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE) with(dune.env, legend(\"bottomleft\", levels(Management), pch=21, pt.bg=1:4, bty=\"n\")) ## Configuration with pipes: make an arrow biplot plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) |> points(\"sites\", pch=21, col = 1, cex=1.5, bg = dune.env$Management) |> text(\"species\", col = \"blue\", arrow = TRUE, xpd = TRUE) ## Scaling can be numeric or more user-friendly names ## e.g. Hill's scaling for (C)CA scrs <- scores(mod, scaling = \"sites\", hill = TRUE) ## or correlation-based scores in PCA/RDA scrs <- scores(rda(dune ~ A1 + Moisture + Management, dune.env), scaling = \"sites\", correlation = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Response Curves for Treatments with Repeated Observations — prc","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Principal Response Curves (PRC) special case Redundancy Analysis (rda) multivariate responses repeated observation design. originally suggested ecological communities. easier interpret traditional constrained ordination. can also used study effects factor depend levels factor B, + :B, multivariate response experiment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"prc(response, treatment, time, ...) # S3 method for prc summary(object, axis = 1, scaling = \"sites\", const, digits = 4, correlation = FALSE, ...) # S3 method for prc plot(x, species = TRUE, select, scaling = \"symmetric\", axis = 1, correlation = FALSE, const, type = \"l\", xlab, ylab, ylim, lty = 1:5, col = 1:6, pch, legpos, cex = 0.8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"response Multivariate response data. Typically community (species) data. data counts, probably log transformed prior analysis. treatment factor treatments. time unordered factor defining observations times repeated design. object, x prc result object. axis Axis shown (one axis can selected). scaling Scaling species scores, identical scaling scores.rda. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. const General scaling constant species scores (see scores.rda details). Lower values reduce range species scores, influence regression coefficients. digits Number significant digits displayed. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. species Display species scores. select Vector select displayed species. can vector indices logical vector TRUE selected species type Type plot: \"l\" lines, \"p\" points \"b\" . xlab, ylab Text replace default axis labels. ylim Limits vertical axis. lty, col, pch Line type, colour plotting characters (defaults supplied). legpos position legend. guess made supplied, NA suppress legend. cex Character expansion symbols species labels. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"PRC special case rda single factor treatment single factor time points repeated observations. vegan, corresponding rda model defined rda(response ~ treatment * time + Condition(time)). Since time appears twice model formula, main effects aliased, main effect treatment interaction terms available, used PRC. Instead usual multivariate ordination diagrams, PRC uses canonical (regression) coefficients species scores single axis. current functions provide special summary plot methods display rda results PRC fashion. current version works default contrasts (contr.treatment) coefficients contrasts first level, levels must arranged first level control ( baseline). necessary, must change baseline level function relevel. Function summary prints species scores coefficients. Function plot plots coefficients time using matplot, similar defaults. graph (PRC) meaningful first treatment level control, results contrasts first level unordered factors used. plot also displays species scores right vertical axis using function linestack. Typically number species high can displayed default settings, users can reduce character size padding (air) linestack, select subset species. legend displayed unless suppressed legpos = NA, functions tries guess put legend legpos supplied.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"function special case rda returns result object (see cca.object). However, special summary plot methods display returns differently rda.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Jari Oksanen Cajo ter Braak","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"first level treatment must control: use function relevel guarantee correct reference level. current version ignore user setting contrasts always use treatment contrasts (contr.treatment). time must unordered factor.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"## Chlorpyrifos experiment and experimental design: Pesticide ## treatment in ditches (replicated) and followed over from 4 weeks ## before to 24 weeks after exposure data(pyrifos) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) ditch <- gl(12, 1, length=132) ## IGNORE_RDIFF_BEGIN ## PRC mod <- prc(pyrifos, dose, week) mod # RDA #> Call: prc(response = pyrifos, treatment = dose, time = week) #> #> Inertia Proportion Rank #> Total 288.9920 1.0000 #> Conditional 63.3493 0.2192 10 #> Constrained 96.6837 0.3346 44 #> Unconstrained 128.9589 0.4462 77 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 25.282 8.297 6.044 4.766 4.148 3.857 3.587 3.334 3.087 2.551 2.466 #> RDA12 RDA13 RDA14 RDA15 RDA16 RDA17 RDA18 RDA19 RDA20 RDA21 RDA22 #> 2.209 2.129 1.941 1.799 1.622 1.579 1.440 1.398 1.284 1.211 1.133 #> RDA23 RDA24 RDA25 RDA26 RDA27 RDA28 RDA29 RDA30 RDA31 RDA32 RDA33 #> 1.001 0.923 0.862 0.788 0.750 0.712 0.685 0.611 0.584 0.537 0.516 #> RDA34 RDA35 RDA36 RDA37 RDA38 RDA39 RDA40 RDA41 RDA42 RDA43 RDA44 #> 0.442 0.417 0.404 0.368 0.340 0.339 0.306 0.279 0.271 0.205 0.179 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 17.156 9.189 7.585 6.064 5.730 4.843 4.518 4.105 #> (Showing 8 of 77 unconstrained eigenvalues) #> summary(mod) # PRC #> #> Call: #> prc(response = pyrifos, treatment = dose, time = week) #> Species scores: #> Simve Daplo Cerpu Alogu Aloco Alore Aloaf Copsp #> 2.688099 1.464566 0.542739 0.280040 0.177019 0.315038 0.426524 1.169368 #> Ostsp Slyla Acrha Aloex Chysp Alona Plead Oxyte #> 2.312186 -0.556899 0.105535 0.228092 0.095042 0.063689 0.138397 0.025401 #> Grate Copdi NauLa CilHa Strvi amosp Ascmo Synsp #> 0.096840 1.428854 4.847070 0.895241 3.069709 -1.357663 0.069736 -0.026494 #> Squro Squmu Polar Kerqu Anufi Mytve Mytvi Mytmu #> 0.264390 -0.452667 0.461989 0.495348 0.432767 0.074372 0.090928 0.105891 #> Lepsp Leppa Colob Colbi Colun Lecsp Lecqu Lecco #> 0.998286 0.084809 -0.723051 -0.139569 -0.828338 0.472866 -0.088860 -0.290531 #> Leclu Lecfl Tripo Cepsp Monlo Monae Scalo Trilo #> 0.049788 -0.408035 0.215234 -0.809597 -0.527913 -0.089948 -0.077192 -0.039086 #> Tripo.1 Tricy Trisp Tepat Rotne Notla Filsp Lopox #> 0.246435 0.335400 0.078423 0.007719 0.143730 -0.114301 -0.168356 -0.030990 #> hydrspec bothrosp olchaeta erpoocto glsicomp alglhete hebdstag sphidae #> -0.048698 0.398665 -1.165154 -0.901030 -0.144389 -0.073049 0.902223 1.463655 #> ansuvote armicris bathcont binitent gyraalbu hippcomp lymnstag lymnaes7 #> 0.140685 1.680010 0.073282 -1.950500 0.033051 0.404473 -0.263679 0.135150 #> physfont plbacorn popyanti radiovat radipere valvcris valvpisc hycarina #> -0.026383 0.084761 1.272223 -0.019815 -0.625468 0.010579 -0.267577 1.044034 #> gammpule aselaqua proameri collembo caenhora caenluct caenrobu cloedipt #> 1.526450 1.578743 0.116577 0.029906 5.767844 2.376188 0.126181 4.734035 #> cloesimi aeshniae libellae conagrae corident coripanz coripunc cymabons #> 1.242207 0.212699 -0.081867 1.630574 0.013761 0.120439 -0.176746 0.046360 #> hesplinn hespsahl notoglau notomacu notoobli notoviri pacoconc pleaminu #> -0.069465 0.033240 0.555201 0.050060 0.082356 0.215863 0.016709 0.071168 #> sigadist sigafall sigastri sigarasp colyfusc donacis6 gyrimari haliconf #> 0.076479 -0.018364 0.060206 -0.277312 0.036493 -0.078608 0.010579 0.446911 #> haliflav haligruf haliobli herubrev hya_herm hyglpusi hyhyovat hypoplan #> 0.044538 0.450146 0.131472 -0.128486 -0.322379 -0.011775 0.024196 0.014976 #> hyporusp hytuinae hytuvers laphminu noteclav rhantusp sialluta ablalong #> 0.232042 2.316179 1.772351 0.632897 0.007912 0.067618 1.109341 0.014976 #> ablaphmo cltanerv malopisp mopetenu prdiussp pstavari chironsp crchirsp #> 2.992695 0.075631 0.047658 0.008716 0.554911 0.082842 1.889916 0.016709 #> crclglat ditendsp mitegchl pachgarc pachgvit popegnub popedisp acriluce #> 0.028953 0.083481 0.230629 -0.012187 -0.029907 0.224272 -0.069649 -0.007950 #> chclpige conescut cricotsp liesspec psclbarb psclgsli psclobvi psclplat #> -0.008744 0.821036 0.121530 0.107387 -0.028639 0.601568 -0.362378 -0.052054 #> psclpsil pscladsp cladotsp laa_spec patanysp tatarssp zaa_spec anopmacu #> 0.007339 0.005674 0.539894 0.034105 0.146807 0.669430 0.049949 -0.163731 #> cepogoae chaoobsc cucidae4 tabanusp agdasphr athrater cyrncren holodubi #> 2.555403 2.442310 0.033240 -0.011601 0.271815 0.067618 0.071168 0.094754 #> holopici leceriae lilurhom monaangu mystazur mystloni oecefurv oecelacu #> 0.611618 0.298633 0.009063 0.644295 0.033240 2.998460 0.536628 0.259064 #> triabico paponysp #> 0.088915 0.097788 #> #> Coefficients for dose + week:dose interaction #> which are contrasts to dose 0 #> rows are dose, columns are week #> -4 -1 0.1 1 2 4 8 12 15 #> 0.1 -0.07218 -0.1375 -0.1020 -0.04068 -0.2101 -0.1364 -0.08077 -0.1536 -0.1123 #> 0.9 -0.08106 -0.1935 -0.1936 -0.47699 -0.4977 -0.4306 -0.13532 -0.3548 -0.2408 #> 6 -0.16616 -0.1232 -0.4539 -1.15638 -1.0835 -1.1511 -0.56112 -0.4698 -0.3078 #> 44 -0.13979 -0.1958 -0.7308 -1.26088 -1.2978 -1.4627 -1.29139 -1.0081 -0.7819 #> 19 24 #> 0.1 -0.2163 -0.07835 #> 0.9 -0.1756 -0.15442 #> 6 -0.3293 -0.18227 #> 44 -0.5768 -0.31022 logabu <- colSums(pyrifos) plot(mod, select = logabu > 100) ## IGNORE_RDIFF_END ## Ditches are randomized, we have a time series, and are only ## interested in the first axis ctrl <- how(plots = Plots(strata = ditch,type = \"free\"), within = Within(type = \"series\"), nperm = 99) anova(mod, permutations = ctrl, first=TRUE) #> Permutation test for rda under reduced model #> Plots: ditch, plot permutation: free #> Permutation: series #> Number of permutations: 99 #> #> Model: prc(response = pyrifos, treatment = dose, time = week) #> Df Variance F Pr(>F) #> RDA1 1 25.282 15.096 0.01 ** #> Residual 77 128.959 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function predict can used find site species scores estimates response data new data sets, Function calibrate estimates values constraints new data set. Functions fitted residuals return estimates response data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"# S3 method for cca fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\"), type = c(\"response\", \"working\"), ...) # S3 method for capscale fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\", \"Imaginary\"), type = c(\"response\", \"working\"), ...) # S3 method for cca residuals(object, ...) # S3 method for cca predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", hill = FALSE, ...) # S3 method for rda predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", correlation = FALSE, const, ...) # S3 method for dbrda predict(object, newdata, type = c(\"response\", \"lc\", \"wa\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", const, ...) # S3 method for cca calibrate(object, newdata, rank = \"full\", ...) # S3 method for cca coef(object, norm = FALSE, ...) # S3 method for decorana predict(object, newdata, type = c(\"response\", \"sites\", \"species\"), rank = 4, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"object result object cca, rda, dbrda, capscale decorana. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. fitted method capscale can also \"Imaginary\" imaginary components negative eigenvalues newdata New data frame used prediction calibration. Usually new community data frame, type = \"lc\" constrained component type = \"response\" type = \"working\" must data frame constraints. newdata must number rows original community data cca result type = \"response\" type = \"working\". original model row column names, new data must contain rows columns names (row names species scores, column names \"wa\" scores constraint names \"lc\" scores). cases rows columns must match directly. argument implemented \"wa\" scores dbrda. type type prediction, fitted values residuals: \"response\" scales results ordination gives results, \"working\" gives values used internally, Chi-square standardization cca scaling centring rda. capscale dbrda \"response\" gives dissimilarities, \"working\" internal data structure analysed ordination. Alternative \"wa\" gives site scores weighted averages community data, \"lc\" site scores linear combinations environmental data, \"sp\" species scores. predict.decorana alternatives scores \"sites\" \"species\". rank rank number axes used approximation. default use axes (full rank) \"model\" available four axes predict.decorana. scaling logical, character, numeric; Scaling predicted scores meaning cca, rda, dbrda, capscale. See scores.cca details acceptable values. correlation, hill logical; correlation-like scores Hill's scaling appropriate RDA CCA respectively. See scores.cca additional details. const Constant multiplier RDA scores. used scaling FALSE, default value give similar scaling scores.rda. norm Coefficients variables centred scaled unit norm. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function fitted gives approximation original data matrix dissimilarities ordination result either scale response scaled internally function. Function residuals gives approximation original data unconstrained ordination. argument type = \"response\" fitted.cca residuals.cca function give marginal totals original data matrix, fitted residuals add original data. Functions fitted residuals dbrda capscale give dissimilarities type = \"response\", additive. However, \"working\" scores additive capscale ( dbrda). fitted residuals capscale dbrda include additive constant requested function call. variants fitted residuals defined model mod <- cca(y ~ x), cca(fitted(mod)) equal constrained ordination, cca(residuals(mod)) equal unconstrained part ordination. Function predict can find estimate original data matrix dissimilarities (type = \"response\") rank. rank = \"full\" identical fitted. addition, function can find species scores site scores community data matrix cca rda. function can used new data, can used add new species site scores existing ordinations. function returns (weighted) orthonormal scores default, must specify explicit scaling add scores ordination diagrams. type = \"wa\" function finds site scores species scores. case, new data can contain new sites, species must match original new data. type=\"sp\" function finds species scores site constraints (linear combination scores). case new data can contain new species, sites must match original new data. type = \"lc\" function finds linear combination scores sites environmental data. case new data frame must contain constraining conditioning environmental variables model formula. type = \"response\" type = \"working\" new data must contain environmental variables constrained component desired, community data matrix residual unconstrained component desired. types, function uses newdata find new \"lc\" (constrained) \"wa\" scores (unconstrained) finds response working data new row scores species scores. original site (row) species (column) weights used type = \"response\" type = \"working\" correspondence analysis (cca) therefore number rows must match original data newdata. completely new data frame created, extreme care needed defining variables similarly original model, particular (ordered) factors. ordination performed formula interface, newdata can data frame matrix, extreme care needed columns match original newdata. Function calibrate.cca finds estimates constraints community ordination \"wa\" scores cca, rda capscale. often known calibration, bioindication environmental reconstruction. Basically, method similar projecting site scores onto biplot arrows, uses regression coefficients. function can called newdata cross-validation possible. newdata may contain new sites, species must match original new data. function work ‘partial’ models Condition term, used newdata capscale dbrda results. results may interpretable continuous variables. Function coef give regression coefficients centred environmental variables (constraints conditions) linear combination scores. coefficients unstandardized environmental variables. coefficients NA aliased effects. Function predict.decorana similar predict.cca. However, type = \"species\" available detrended correspondence analysis (DCA), detrending destroys mutual reciprocal averaging (except first axis rescaling used). Detrended CA attempt approximate original data matrix, type = \"response\" meaning detrended analysis (except rank = 1).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"functions return matrices, vectors dissimilarities appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) # Definition of the concepts 'fitted' and 'residuals' mod #> Call: cca(formula = dune ~ A1 + Management + Condition(Moisture), data #> = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Conditional 0.6283 0.2970 3 #> Constrained 0.5109 0.2415 4 #> Unconstrained 0.9761 0.4615 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 #> 0.24932 0.12090 0.08160 0.05904 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> cca(fitted(mod)) #> Call: cca(X = fitted(mod)) #> #> Inertia Rank #> Total 0.5109 #> Unconstrained 0.5109 4 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 #> 0.24932 0.12090 0.08160 0.05904 #> cca(residuals(mod)) #> Call: cca(X = residuals(mod)) #> #> Inertia Rank #> Total 0.9761 #> Unconstrained 0.9761 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> # Remove rare species (freq==1) from 'cca' and find their scores # 'passively'. freq <- specnumber(dune, MARGIN=2) freq #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 7 10 2 8 6 6 5 1 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 1 2 5 6 1 3 5 4 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 12 7 14 13 6 5 7 3 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 18 3 16 3 15 3 mod <- cca(dune[, freq>1] ~ A1 + Management + Condition(Moisture), dune.env) ## IGNORE_RDIFF_BEGIN predict(mod, type=\"sp\", newdata=dune[, freq==1], scaling=\"species\") #> CCA1 CCA2 CCA3 CCA4 #> Chenalbu 1.5737337 0.7842538 0.5503660 -0.35108333 #> Cirsarve 0.5945146 0.3714228 -0.2862647 -0.88373727 #> Empenigr -1.8771953 0.9904299 -0.2446222 -0.04858656 # New sites predict(mod, type=\"lc\", new=data.frame(A1 = 3, Management=\"NM\", Moisture=\"2\"), scal=2) #> CCA1 CCA2 CCA3 CCA4 #> 1 -2.38829 1.230652 -0.2363485 0.3338258 # Calibration and residual plot mod <- cca(dune ~ A1 + Moisture, dune.env) pred <- calibrate(mod) pred #> A1 Moisture.L Moisture.Q Moisture.C #> 1 2.2630533 -0.62633470 -0.20456759 0.220761764 #> 2 4.0510042 -0.47341146 -0.36986691 0.474939409 #> 3 4.2752294 -0.07214500 -0.60797514 0.303213289 #> 4 4.5398659 0.03192745 -1.12417368 0.932223234 #> 5 5.0409406 -0.84235946 0.43000738 -0.291599200 #> 6 5.1962100 -0.91316862 1.11354235 -0.804453944 #> 7 4.2452549 -0.76452556 0.60464291 -0.484842066 #> 8 5.0208369 0.43886340 0.08169514 0.132995916 #> 9 4.2663219 0.10720486 -0.34067849 -0.675151598 #> 10 4.0411356 -0.65472729 0.02832164 0.558402684 #> 11 2.8280051 -0.45762457 0.63079135 -0.089977975 #> 12 5.1204137 0.36328912 -0.69118581 -0.665622948 #> 13 4.9034218 0.47069541 -0.54378271 -0.118643453 #> 14 11.6455841 0.60920550 0.78341426 0.532852308 #> 15 10.7829689 0.69208513 0.82190786 0.237311062 #> 16 7.9892176 0.96421599 0.46793089 0.373647014 #> 17 0.9218684 -0.15822891 0.14593271 1.189161582 #> 18 3.1680733 -0.41737900 1.03352732 -0.236938282 #> 19 -1.2003506 0.57033354 0.72777285 0.509955590 #> 20 4.7876770 1.00324330 1.49898460 0.009202396 ## IGNORE_RDIFF_END with(dune.env, plot(A1, pred[,\"A1\"] - A1, ylab=\"Prediction Error\")) abline(h=0)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":null,"dir":"Reference","previous_headings":"","what":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes rotates configuration maximum similarity another configuration. Function protest tests non-randomness (significance) two configurations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"procrustes(X, Y, scale = TRUE, symmetric = FALSE, scores = \"sites\", ...) # S3 method for procrustes summary(object, digits = getOption(\"digits\"), ...) # S3 method for procrustes plot(x, kind=1, choices=c(1,2), to.target = TRUE, type = \"p\", xlab, ylab, main, ar.col = \"blue\", length=0.05, cex = 0.7, ...) # S3 method for procrustes points(x, display = c(\"target\", \"rotated\"), choices = c(1,2), truemean = FALSE, ...) # S3 method for procrustes text(x, display = c(\"target\", \"rotated\"), choices = c(1,2), labels, truemean = FALSE, ...) # S3 method for procrustes lines(x, type = c(\"segments\", \"arrows\"), choices = c(1, 2), truemean = FALSE, ...) # S3 method for procrustes residuals(object, ...) # S3 method for procrustes fitted(object, truemean = TRUE, ...) # S3 method for procrustes predict(object, newdata, truemean = TRUE, ...) protest(X, Y, scores = \"sites\", permutations = how(nperm = 999), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"X Target matrix Y Matrix rotated. scale Allow scaling axes Y. symmetric Use symmetric Procrustes statistic (rotation still non-symmetric). scores Kind scores used. display argument used corresponding scores function: see scores, scores.cca scores.cca alternatives. x, object object class procrustes. digits Number digits output. kind plot function, kind plot produced: kind = 1 plots shifts two configurations, kind = 0 draws corresponding empty plot, kind = 2 plots impulse diagram residuals. choices Axes (dimensions) plotted. xlab, ylab Axis labels, defaults unacceptable. main Plot title, default unacceptable. display Show \"target\" \"rotated\" matrix points. .target Draw arrows point target. type type plot drawn. plot, type can \"points\" \"text\" select marker tail arrow, \"none\" drawing empty plot. lines type selects either arrows line segments connect target rotated configuration. truemean Use original range target matrix instead centring fitted values. Function plot.procrustes needs truemean = FALSE, adding graphical items plots original results may need truemean = TRUE. newdata Matrix coordinates rotated translated target. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ar.col Arrow colour. length Width arrow head. labels Character vector text labels. Rownames result object used default. cex Character expansion points text. ... parameters passed functions. procrustes protest parameters passed scores, graphical functions underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Procrustes rotation rotates matrix maximum similarity target matrix minimizing sum squared differences. Procrustes rotation typically used comparison ordination results. particularly useful comparing alternative solutions multidimensional scaling. scale=FALSE, function rotates matrix Y. scale=TRUE, scales linearly configuration Y maximum similarity. Since Y scaled fit X, scaling non-symmetric. However, symmetric=TRUE, configurations scaled equal dispersions symmetric version Procrustes statistic computed. Instead matrix, X Y can results ordination scores can extract results. Function procrustes passes extra arguments scores, scores.cca etc. can specify arguments scaling. Function plot plots procrustes object returns invisibly ordiplot object function identify.ordiplot can used identifying points. items ordiplot object called heads points kind=1 (ordination diagram) sites kind=2 (residuals). ordination diagrams, arrow heads point target configuration .target = TRUE, rotated configuration .target = FALSE. Target original rotated axes shown cross hairs two-dimensional Procrustes analysis, higher number dimensions, rotated axes projected onto plot scaled centred range. Function plot passes parameters underlying plotting functions. full control plots, can draw axes using plot kind = 0, add items points lines. functions pass parameters underlying functions can select plotting characters, size, colours etc., can select width, colour type line segments arrows, can select orientation head width arrows. Function residuals returns pointwise residuals, fitted fitted values, either centred zero mean (truemean=FALSE) original scale ( hardly make sense symmetric = TRUE). addition, summary print methods. matrix X lower number columns matrix Y, matrix X filled zero columns match dimensions. means function can used rotate ordination configuration environmental variable ( practically extracting result fitted function). Function predict can used add new rotated coordinates target. predict function always translate coordinates original non-centred matrix. function used newdata symmetric analysis. Function protest performs symmetric Procrustes analysis repeatedly estimate significance Procrustes statistic. Function protest uses correlation-like statistic derived symmetric Procrustes sum squares \\(ss\\) \\(r =\\sqrt{1-ss}\\), also prints sum squares symmetric analysis, sometimes called \\(m_{12}^2\\). Function protest print method, otherwise uses procrustes methods. Thus plot protest object yields Procrustean superimposition plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes returns object class procrustes items. Function protest inherits procrustes, amends new items: Yrot Rotated matrix Y. X Target matrix. ss Sum squared differences X Yrot. rotation Orthogonal rotation matrix. translation Translation origin. scale Scaling factor. xmean centroid target. symmetric Type ss statistic. call Function call. t0 following items class protest: Procrustes correlation non-permuted solution. t Procrustes correlations permutations. distribution correlations can inspected permustats function. signif Significance t permutations Number permutations. control list control values permutations returned function . control list passed argument control describing permutation design.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate Analysis. Academic Press. Peres-Neto, P.R. Jackson, D.. (2001). well multivariate data sets match? advantages Procrustean superimposition approach Mantel test. Oecologia 129: 169-178.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"function protest follows Peres-Neto & Jackson (2001), implementation still Mardia et al. (1979).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"## IGNORE_RDIFF_BEGIN data(varespec) vare.dist <- vegdist(wisconsin(varespec)) mds.null <- monoMDS(vare.dist, y = cmdscale(vare.dist)) mds.alt <- monoMDS(vare.dist) vare.proc <- procrustes(mds.alt, mds.null) vare.proc #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Procrustes sum of squares: #> 11.17 #> summary(vare.proc) #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Number of objects: 24 Number of dimensions: 2 #> #> Procrustes sum of squares: #> 11.17448 #> Procrustes root mean squared error: #> 0.6823512 #> Quantiles of Procrustes errors: #> Min 1Q Median 3Q Max #> 0.1642438 0.2425785 0.2783603 0.4983976 2.4447632 #> #> Rotation matrix: #> [,1] [,2] #> [1,] 0.99937107 -0.03546065 #> [2,] 0.03546065 0.99937107 #> #> Translation of averages: #> [,1] [,2] #> [1,] -1.713781e-17 1.56769e-17 #> #> Scaling of target: #> [1] 0.7310245 #> plot(vare.proc) plot(vare.proc, kind=2) residuals(vare.proc) #> 18 15 24 27 23 19 22 16 #> 0.2734040 0.2032392 0.4708118 0.4420710 0.3547337 0.1642438 0.2515286 0.2611623 #> 28 13 14 20 25 7 5 6 #> 0.7773604 0.3075051 0.2833167 0.1749943 0.2684784 0.5167965 0.9747233 0.2437827 #> 3 4 2 9 12 10 11 21 #> 0.2252071 0.7586954 2.4447632 0.2389659 0.2093104 0.2597721 0.4922646 1.0884414 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":null,"dir":"Reference","previous_headings":"","what":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data log transformed abundances aquatic invertebrate twelve ditches studied eleven times insecticide treatment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos)"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data frame 132 observations log-transformed (log(10*x + 1)) abundances 178 species. twelve sites (ditches, mesocosms), studied repeatedly eleven occasions. treatment levels, treatment times, ditch ID's data frame, data regular, example shows obtain external variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data set obtained experiment outdoor experimental ditches. Twelve mesocosms allocated random treatments; four served controls, remaining eight treated insecticide chlorpyrifos, nominal dose levels 0.1, 0.9, 6, 44 \\(\\mu\\)g/ L two mesocosms . example data set invertebrates. Sampling done 11 times, week -4 pre-treatment week 24 post-treatment, giving total 132 samples (12 mesocosms times 11 sampling dates), see van den Brink & ter Braak (1999) details. data set contains species data, example shows obtain treatment, time ditch ID variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"CANOCO 4 example data, permission Cajo J. F. ter Braak.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos) ditch <- gl(12, 1, length=132) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank -- Abundance or Dominance / Diversity Models — radfit","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions construct rank -- abundance dominance / diversity Whittaker plots fit brokenstick, preemption, log-Normal, Zipf Zipf-Mandelbrot models species abundance.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"# S3 method for default radfit(x, ...) rad.null(x, family=poisson, ...) rad.preempt(x, family = poisson, ...) rad.lognormal(x, family = poisson, ...) rad.zipf(x, family = poisson, ...) rad.zipfbrot(x, family = poisson, ...) # S3 method for radline predict(object, newdata, total, ...) # S3 method for radfit plot(x, BIC = FALSE, legend = TRUE, ...) # S3 method for radfit.frame plot(x, order.by, BIC = FALSE, model, legend = TRUE, as.table = TRUE, ...) # S3 method for radline plot(x, xlab = \"Rank\", ylab = \"Abundance\", type = \"b\", ...) radlattice(x, BIC = FALSE, ...) # S3 method for radfit lines(x, ...) # S3 method for radfit points(x, ...) as.rad(x) # S3 method for rad plot(x, xlab = \"Rank\", ylab = \"Abundance\", log = \"y\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"x Data frame, matrix vector giving species abundances, object plotted. family Error distribution (passed glm). alternatives accepting link = \"log\" family can used, although make sense. object fitted result object. newdata Ranks used ordinations. models can interpolate non-integer “ranks” (although may approximate), extrapolation may fail total new total used predicting abundance. Observed total count used omitted. order.vector used ordering sites plots. BIC Use Bayesian Information Criterion, BIC, instead Akaike's AIC. penalty BIC \\(k = \\log(S)\\) \\(S\\) number species, whereas AIC uses \\(k = 2\\). model Show specified model. missing, AIC used select model. model names (can abbreviated) Null, Preemption, Lognormal, Zipf, Mandelbrot. legend Add legend line colours. .table Arrange panels starting upper left corner (passed xyplot). xlab,ylab Labels x y axes. type Type plot, \"b\" plotting observed points fitted lines, \"p\" points, \"l\" fitted lines, \"n\" setting frame. log Use logarithmic scale given axis. default log = \"y\" gives traditional plot community ecology preemption model straight line, log = \"xy\" Zipf model straight line. log = \"\" axes original arithmetic scale. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Rank--Abundance Dominance (RAD) Dominance/Diversity plots (Whittaker 1965) display logarithmic species abundances species rank order. plots supposed effective analysing types abundance distributions communities. functions fit popular models mainly following Wilson (1991). Functions rad.null, rad.preempt, rad.lognormal, rad.zipf zipfbrot fit individual models (described ) single vector (row data frame), function radfit fits models. argument function radfit can either vector single community data frame row represents distinct community. Function rad.null fits brokenstick model expected abundance species rank \\(r\\) \\(a_r = (J/S) \\sum_{x=r}^S (1/x)\\) (Pielou 1975), \\(J\\) total number individuals (site total) \\(S\\) total number species community. gives Null model individuals randomly distributed among observed species, fitted parameters. Function rad.preempt fits niche preemption model, .k.. geometric series Motomura model, expected abundance \\(\\) species rank \\(r\\) \\(a_r = J \\alpha (1 - \\alpha)^{r-1}\\). estimated parameter preemption coefficient \\(\\alpha\\) gives decay rate abundance per rank. niche preemption model straight line RAD plot. Function rad.lognormal fits log-Normal model assumes logarithmic abundances distributed Normally, \\(a_r = \\exp( \\log \\mu + \\log \\sigma N)\\), \\(N\\) Normal deviate. Function rad.zipf fits Zipf model \\(a_r = J p_1 r^\\gamma\\) \\(p_1\\) fitted proportion abundant species, \\(\\gamma\\) decay coefficient. Zipf--Mandelbrot model (rad.zipfbrot) adds one parameter: \\(a_r = J c (r + \\beta)^\\gamma\\) \\(p_1\\) Zipf model changes meaningless scaling constant \\(c\\). Log-Normal Zipf models generalized linear models (glm) logarithmic link function. Zipf--Mandelbrot adds one nonlinear parameter Zipf model, fitted using nlm nonlinear parameter estimating parameters log-Likelihood glm. Preemption model fitted purely nonlinear model. estimated parameters Null model. default family poisson appropriate genuine counts (integers), families accept link = \"log\" can used. Families Gamma gaussian may appropriate abundance data, cover. best model selected AIC. Therefore ‘quasi’ families quasipoisson used: AIC log-Likelihood needed non-linear models. functions plot functions. radfit applied data frame, plot uses Lattice graphics, plot functions use ordinary graphics. ordinary graphics functions return invisibly ordiplot object observed points, function identify.ordiplot can used label selected species. Alternatively, radlattice uses Lattice graphics display radfit model single site separate panel together AIC BIC values. Function .rad base function construct ordered RAD data. plot used RAD plot functions pass extra arguments (xlab log) function. function returns ordered vector taxa occurring site, corresponding attribute \"index\" included taxa.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions rad.null, rad.preempt, rad.lognormal, zipf zipfbrot fit single RAD model single site. result object class \"radline\" inherits glm, can handled ( ) glm methods. Function radfit fits models either single site rows data frame matrix. fitted single site, function returns object class \"radfit\" items y (observed values), family, models list fitted \"radline\" models. applied data frame matrix, radfit function returns object class \"radfit.frame\" list \"radfit\" objects, item names corresponding row name. result objects (\"radline\", \"radfit\", \"radfit.frame\") can accessed method functions. following methods available: AIC, coef, deviance, logLik. addition fit results can accessed fitted, predict residuals (inheriting residuals.glm). graphical functions discussed Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Pielou, E.C. (1975) Ecological Diversity. Wiley & Sons. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Whittaker, R. H. (1965) Dominance diversity plant communities. Science 147, 250--260. Wilson, J. B. (1991) Methods fitting dominance/diversity curves. Journal Vegetation Science 2, 35--46.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"RAD models usually fitted proportions instead original abundances. However, nothing models seems require division abundances site totals, original observations used functions. wish use proportions, must standardize data site totals, e.g. decostand use appropriate family Gamma. lognormal model fitted standard way, think quite correct -- least equivalent fitting Normal density log abundances like originally suggested (Preston 1948). models may fail. particular, estimation Zipf-Mandelbrot model difficult. fitting fails, NA returned. Wilson (1991) defined preemption model \\(a_r = J p_1 (1 - \\alpha)^{r-1}\\), \\(p_1\\) fitted proportion first species. However, parameter \\(p_1\\) completely defined \\(\\alpha\\) since fitted proportions must add one, therefore handle preemption one-parameter model. Veiled log-Normal model included earlier releases function, removed flawed: implicit veil line also appears ordinary log-Normal. latest release version rad.veil 1.6-10.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"data(BCI) mod <- rad.lognormal(BCI[5,]) mod #> #> RAD model: Log-Normal #> Family: poisson #> No. of species: 101 #> Total abundance: 505 #> #> log.mu log.sigma Deviance AIC BIC #> 0.951926 1.165929 17.077549 317.656487 322.886728 plot(mod) mod <- radfit(BCI[1,]) ## Standard plot overlaid for all models ## Preemption model is a line plot(mod) ## log for both axes: Zipf model is a line plot(mod, log = \"xy\") ## Lattice graphics separately for each model radlattice(mod) # Take a subset of BCI to save time and nerves mod <- radfit(BCI[3:5,]) mod #> #> Deviance for RAD models: #> #> 3 4 5 #> Null 86.1127 49.8111 80.855 #> Preemption 58.9295 39.7817 76.311 #> Lognormal 29.2719 16.6588 17.078 #> Zipf 50.1262 47.9108 30.936 #> Mandelbrot 5.7342 5.5665 10.573 plot(mod, pch=\".\")"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":null,"dir":"Reference","previous_headings":"","what":"Compares Dissimilarity Indices for Gradient Detection — rankindex","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Rank correlations dissimilarity indices gradient separation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"rankindex(grad, veg, indices = c(\"euc\", \"man\", \"gow\", \"bra\", \"kul\"), stepacross = FALSE, method = \"spearman\", metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"grad gradient variable matrix. veg community data matrix. indices Dissimilarity indices compared, partial matches alternatives vegdist. Alternatively, can (named) list functions returning objects class 'dist'. stepacross Use stepacross find shorter path dissimilarity. dissimilarities site pairs shared species set NA using .shared indices fixed upper limit can also analysed. method Correlation method used. metric Metric evaluate gradient separation. See Details. ... parameters stepacross.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"good dissimilarity index multidimensional scaling high rank-order similarity gradient separation. function compares indices vegdist gradient separation using rank correlation coefficients cor. gradient separation point assessed using given metric. default use Euclidean distance continuous variables scaled unit variance, use Gower metric mixed data using function daisy grad factors. alternatives Mahalanabis distances based grad matrix scaled columns orthogonal (uncorrelated) unit variance, Manhattan distances grad variables scaled unit range. indices argument can accept dissimilarity indices besides ones calculated vegdist function. , argument value (possibly named) list functions. function must return valid 'dist' object dissimilarities, similarities accepted converted dissimilarities beforehand.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Returns named vector rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Faith, F.P., Minchin, P.R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Jari Oksanen, additions Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"several problems using rank correlation coefficients. Typically many ties \\(n(n-1)/2\\) gradient separation values derived just \\(n\\) observations. Due floating point arithmetics, many tied values differ machine epsilon arbitrarily ranked differently rank used cor.test. Two indices identical certain transformation standardization may differ slightly (magnitude \\(10^{-15}\\)) may lead third fourth decimal instability rank correlations. Small differences rank correlations taken seriously. Probably method replaced sounder method, yet know ... may experiment mantel, anosim even protest. Earlier version function used method = \"kendall\", far slow large data sets. functions returning dissimilarity objects self contained, ... argument passes additional parameters stepacross functions supplied via indices argument.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"data(varespec) data(varechem) ## The variables are automatically scaled rankindex(varechem, varespec) #> euc man gow bra kul #> 0.2396330 0.2735087 0.2288358 0.2837910 0.2839834 rankindex(varechem, wisconsin(varespec)) #> euc man gow bra kul #> 0.4200990 0.4215642 0.3708606 0.4215642 0.4215642 ## Using non vegdist indices as functions funs <- list(Manhattan=function(x) dist(x, \"manhattan\"), Gower=function(x) cluster:::daisy(x, \"gower\"), Ochiai=function(x) designdist(x, \"1-J/sqrt(A*B)\")) rankindex(scale(varechem), varespec, funs) #> Manhattan Gower Ochiai #> 0.2735087 0.2288358 0.1696862"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":null,"dir":"Reference","previous_headings":"","what":"Rarefaction Species Richness — rarefy","title":"Rarefaction Species Richness — rarefy","text":"Rarefied species richness community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rarefaction Species Richness — rarefy","text":"","code":"rarefy(x, sample, se = FALSE, MARGIN = 1) rrarefy(x, sample) drarefy(x, sample) rarecurve(x, step = 1, sample, xlab = \"Sample Size\", ylab = \"Species\", label = TRUE, col, lty, tidy = FALSE, ...) rareslope(x, sample)"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rarefaction Species Richness — rarefy","text":"x Community data, matrix-like object vector. MARGIN Margin index computed. sample Subsample size rarefying community, either single value vector. se Estimate standard errors. step Step size sample sizes rarefaction curves. xlab, ylab Axis labels plots rarefaction curves. label Label rarefaction curves rownames x (logical). col, lty plotting colour line type, see par. Can vector length nrow(x), one per sample, extended length internally. tidy Instead drawing plot, return “tidy” data frame can used ggplot2 graphics. data frame variables Site (factor), Sample Species. ... Parameters passed nlm, plot, lines ordilabel rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rarefaction Species Richness — rarefy","text":"Function rarefy gives expected species richness random subsamples size sample community. size sample smaller total community size, function work larger sample well (warning) return non-rarefied species richness (standard error = 0). sample vector, rarefaction observations performed sample size separately. Rarefaction can performed genuine counts individuals. function rarefy based Hurlbert's (1971) formulation, standard errors Heck et al. (1975). Function rrarefy generates one randomly rarefied community data frame vector given sample size. sample can vector giving sample sizes row. sample size equal larger observed number individuals, non-rarefied community returned. random rarefaction made without replacement variance rarefied communities rather related rarefaction proportion size sample. Random rarefaction sometimes used remove effects different sample sizes. usually bad idea: random rarefaction discards valid data, introduces random error reduces quality data (McMurdie & Holmes 2014). better use normalizing transformations (decostand vegan) possible variance stabilization (decostand dispweight vegan) methods sensitive sample sizes. Function drarefy returns probabilities species occur rarefied community size sample. sample can vector giving sample sizes row. sample equal larger observed number individuals, observed species sampling probability 1. Function rarecurve draws rarefaction curve row input data. rarefaction curves evaluated using interval step sample sizes, always including 1 total sample size. sample specified, vertical line drawn sample horizontal lines rarefied species richnesses. Function rareslope calculates slope rarecurve (derivative rarefy) given sample size; sample need integer. Rarefaction functions used observed counts. think necessary use multiplier data, rarefy first multiply. Removing rare species rarefaction can also give biased results. Observed count data normally include singletons (species count 1), missing, functions issue warnings. may false positives, recommended check observed counts multiplied rare taxa removed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rarefaction Species Richness — rarefy","text":"vector rarefied species richness values. single sample se = TRUE, function rarefy returns 2-row matrix rarefied richness (S) standard error (se). sample vector rarefy, function returns matrix column sample size, se = TRUE, rarefied richness standard error consecutive lines. Function rarecurve returns invisible list rarefy results corresponding drawn curve. Alternatively, tidy = TRUE returns data frame can used ggplot2 graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rarefaction Species Richness — rarefy","text":"Heck, K.L., van Belle, G. & Simberloff, D. (1975). Explicit calculation rarefaction diversity measurement determination sufficient sample size. Ecology 56, 1459--1461. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. McMurdie, P.J. & Holmes, S. (2014). Waste , want : rarefying microbiome data inadmissible. PLoS Comput Biol 10(4): e1003531. doi:10.1371/journal.pcbi.1003531","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rarefaction Species Richness — rarefy","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rarefaction Species Richness — rarefy","text":"","code":"data(BCI) S <- specnumber(BCI) # observed number of species (raremax <- min(rowSums(BCI))) #> [1] 340 Srare <- rarefy(BCI, raremax) plot(S, Srare, xlab = \"Observed No. of Species\", ylab = \"Rarefied No. of Species\") abline(0, 1) rarecurve(BCI, step = 20, sample = raremax, col = \"blue\", cex = 0.6)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":null,"dir":"Reference","previous_headings":"","what":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Function finds Raup-Crick dissimilarity probability number co-occurring species species occurrence probabilities proportional species frequencies.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"raupcrick(comm, null = \"r1\", nsimul = 999, chase = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"comm Community data treated presence/absence data. null Null model used method oecosimu. nsimul Number null communities assessing dissimilarity index. chase Use Chase et al. (2011) method tie handling ( recommended except comparing results Chase script). ... parameters passed oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Raup-Crick index probability compared sampling units non-identical species composition. probability can regarded dissimilarity, although metric: identical sampling units can dissimilarity slightly \\(0\\), dissimilarity can nearly zero range shared species, sampling units shared species can dissimilarity slightly \\(1\\). Moreover, communities sharing rare species appear similar (lower probability finding rare species together), communities sharing number common species. function always treat data binary (presence/ absence). probability assessed using simulation oecosimu test statistic observed number shared species sampling units evaluated community null model (see Examples). default null model \"r1\" probability selecting species proportional species frequencies. vegdist function implements variant Raup-Crick index equal sampling probabilities species using exact analytic equations without simulation. corresponds null model \"r0\" also can used current function. null model methods oecosimu can used current function, new unpublished methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function returns object inheriting dist can interpreted dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function developed Brian Inouye contacted us informed us method Chase et al. (2011), function takes idea code published paper. current function written Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"test statistic number shared species, typically tied large number simulation results. tied values handled differently current function function published Chase et al. (2011). vegan, index number simulated values smaller equal observed value, smaller observed value used Chase et al. (2011) option split = FALSE script; can achieved chase = TRUE vegan. Chase et al. (2011) script split = TRUE uses half tied simulation values calculate distance measure, choice directly reproduced vegan ( average vegan raupcrick results chase = TRUE chase = FALSE).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"## data set with variable species richness data(sipoo) ## default raupcrick dr1 <- raupcrick(sipoo) ## use null model \"r0\" of oecosimu dr0 <- raupcrick(sipoo, null = \"r0\") ## vegdist(..., method = \"raup\") corresponds to 'null = \"r0\"' d <- vegdist(sipoo, \"raup\") op <- par(mfrow=c(2,1), mar=c(4,4,1,1)+.1) plot(dr1 ~ d, xlab = \"Raup-Crick with Null R1\", ylab=\"vegdist\") plot(dr0 ~ d, xlab = \"Raup-Crick with Null R0\", ylab=\"vegdist\") par(op) ## The calculation is essentially as in the following oecosimu() call, ## except that designdist() is replaced with faster code if (FALSE) oecosimu(sipoo, function(x) designdist(x, \"J\", \"binary\"), method = \"r1\")"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":null,"dir":"Reference","previous_headings":"","what":"Reads a CEP (Canoco) data file — read.cep","title":"Reads a CEP (Canoco) data file — read.cep","text":"read.cep reads file formatted relaxed strict CEP format used Canoco software, among others.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"read.cep(file, positive=TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reads a CEP (Canoco) data file — read.cep","text":"file File name (character variable). positive positive entries, like community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reads a CEP (Canoco) data file — read.cep","text":"Cornell Ecology Programs (CEP) introduced several data formats designed punched cards. One ‘condensed strict’ format adopted popular software DECORANA TWINSPAN. relaxed variant format later adopted Canoco software (ter Braak 1984). Function read.cep reads legacy files written format. condensed CEP CANOCO formats : Two three title cards, importantly specifying format number items per record. Data condensed format: First number line site identifier (integer), followed pairs (‘couplets’) numbers identifying species abundance (integer floating point number). Species site names, given Fortran format (10A8): Ten names per line, eight columns . option positive = TRUE function removes rows columns zero negative marginal sums. community data positive entries, removes empty sites species. data entries can negative, ruins data, data sets read option positive = FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reads a CEP (Canoco) data file — read.cep","text":"Returns data frame, columns species rows sites. Column row names taken CEP file, changed unique R names make.names stripping blanks.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Reads a CEP (Canoco) data file — read.cep","text":"ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reads a CEP (Canoco) data file — read.cep","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reads a CEP (Canoco) data file — read.cep","text":"Function read.cep used Fortran read data vegan 2.4-5 earlier, Fortran /O longer allowed CRAN packages, function re-written R. original Fortran code robust, several legacy data sets may fail current version, read previous Fortran version. CRAN package cepreader makes available original Fortran-based code run separate subprocess. cepreader package can also read ‘free’ ‘open’ Canoco formats handled function. function based read.fortran. REAL format defines decimal part species abundances (F5.1), read.fortran divides input corresponding power 10 even input data explicit decimal separator. F5.1, 100 become 10, 0.1 become 0.01. Function read.cep tries undo division, check scaling results reading data, necessary, multiply results original scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"## Provided that you have the file \"dune.spe\" if (FALSE) { theclassic <- read.cep(\"dune.spe\")}"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":null,"dir":"Reference","previous_headings":"","what":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi find Rényi diversities scale corresponding Hill number (Hill 1973). Function renyiaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"renyi(x, scales = c(0, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64, Inf), hill = FALSE) # S3 method for renyi plot(x, ...) renyiaccum(x, scales = c(0, 0.5, 1, 2, 4, Inf), permutations = 100, raw = FALSE, collector = FALSE, subset, ...) # S3 method for renyiaccum plot(x, what = c(\"Collector\", \"mean\", \"Qnt 0.025\", \"Qnt 0.975\"), type = \"l\", ...) # S3 method for renyiaccum persp(x, theta = 220, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"x Community data matrix plotting object. scales Scales Rényi diversity. hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. collector Accumulate diversities order sites data set, collector curve can plotted summary permutations. argument ignored raw = TRUE. subset logical expression indicating sites (rows) keep: missing values taken FALSE. Items plotted. type Type plot, type = \"l\" means lines. theta Angle defining viewing direction (azimuthal) persp. col Colours used surface. Single colour passed , vector colours selected midpoint rectangle persp. zlim Limits vertical axis. ... arguments passed renyi graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Common diversity indices special cases Rényi diversity $$H_a = \\frac{1}{1-} \\log \\sum p_i^$$ \\(\\) scale parameter, Hill (1975) suggested use -called ‘Hill numbers’ defined \\(N_a = \\exp(H_a)\\). Hill numbers number species \\(= 0\\), \\(\\exp(H')\\) exponent Shannon diversity \\(= 1\\), inverse Simpson \\(= 2\\) \\(1/ \\max(p_i)\\) \\(= \\infty\\). According theory diversity ordering, one community can regarded diverse another Rényi diversities higher (Tóthmérész 1995). plot method renyi uses lattice graphics, displays diversity values scale separate panel site together minimum, maximum median values complete data. Function renyiaccum similar specaccum finds Rényi Hill diversities given scales random permutations accumulated sites. plot function uses lattice function xyplot display accumulation curves value scales separate panel. addition, persp method plot diversity surface scale number sites. Similar dynamic graphics can made rgl.renyiaccum vegan3d package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi returns data frame selected indices. Function renyiaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Hill, M.O. (1973). Diversity evenness: unifying notation consequences. Ecology 54, 427--473. Kindt, R., Van Damme, P., Simons, .J. (2006). Tree diversity western Kenya: using profiles characterise richness evenness. Biodiversity Conservation 15, 1253--1270. Tóthmérész, B. (1995). Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Roeland Kindt Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) mod <- renyi(BCI[i,]) plot(mod) mod <- renyiaccum(BCI[i,]) plot(mod, as.table=TRUE, col = c(1, 2, 2)) persp(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":null,"dir":"Reference","previous_headings":"","what":"Reorder a Hierarchical Clustering Tree — reorder.hclust","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Function takes hierarchical clustering tree hclust vector values reorders clustering tree order supplied vector, maintaining constraints tree. method generic function reorder alternative reordering \"dendrogram\" object reorder.dendrogram","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"# S3 method for hclust reorder(x, wts, agglo.FUN = c(\"mean\", \"min\", \"max\", \"sum\", \"uwmean\"), ...) # S3 method for hclust rev(x) # S3 method for hclust scores(x, display = \"internal\", ...) cutreeord(tree, k = NULL, h = NULL)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"x, tree hierarchical clustering hclust. wts numeric vector reordering. agglo.FUN function weights agglomeration, see . display return \"internal\" nodes \"terminal\" nodes (also called \"leaves\"). k, h scalars vectors giving numbers desired groups heights tree cut (passed function cutree). ... additional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Dendrograms can ordered many ways. reorder function reorders hclust tree provides alternative reorder.dendrogram can reorder dendrogram. current function also work differently agglo.FUN \"mean\": reorder.dendrogram always take direct mean member groups ignoring sizes, function used weighted.mean weighted group sizes, group mean always mean member leaves (terminal nodes). want ignore group sizes, can use unweighted mean \"uwmean\". function accepts limited list agglo.FUN functions assessing value wts groups. ordering always ascending, order leaves can reversed rev. Function scores finds coordinates nodes two-column matrix. terminal nodes (leaves) value item merged tree, labels can still hang level (see plot.hclust). Function cutreeord cuts tree groups numbered left right tree. based standard function cutree numbers groups order appear input data instead order tree.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Reordered hclust result object added item value gives value statistic merge level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"functions really base R.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"## reorder by water content of soil data(mite, mite.env) hc <- hclust(vegdist(wisconsin(sqrt(mite)))) ohc <- with(mite.env, reorder(hc, WatrCont)) plot(hc) plot(ohc) ## label leaves by the observed value, and each branching point ## (internal node) by the cluster mean with(mite.env, plot(ohc, labels=round(WatrCont), cex=0.7)) ordilabel(scores(ohc), label=round(ohc$value), cex=0.7) ## Slightly different from reordered 'dendrogram' which ignores group ## sizes in assessing means. den <- as.dendrogram(hc) den <- with(mite.env, reorder(den, WatrCont, agglo.FUN = mean)) plot(den)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Species or Site Scores from an Ordination — scores","title":"Get Species or Site Scores from an Ordination — scores","text":"Function access either species site scores specified axes ordination methods. scores function generic vegan, vegan ordination functions scores functions documented separately method (see e.g. scores.cca, scores.metaMDS, scores.decorana). help file documents default scores method used non-vegan ordination objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"# S3 method for default scores(x, choices, display=c(\"sites\", \"species\", \"both\"), tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Species or Site Scores from an Ordination — scores","text":"x ordination result. choices Ordination axes. missing, default method returns axes. display Partial match access scores \"sites\" \"species\" \"\". tidy Return \"\" scores data frame compatible ggplot2, variable score labelling scores \"sites\" \"species\". ... parameters (unused).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Species or Site Scores from an Ordination — scores","text":"Function scores generic method vegan. Several vegan functions scores methods defaults new arguments. help page describes default method. methods, see, e.g., scores.cca, scores.rda, scores.decorana. vegan ordination functions scores method used extract scores instead directly accessing . Scaling transformation scores also happen scores function. scores function available, results can plotted using ordiplot, ordixyplot etc., ordination results can compared procrustes analysis. scores.default function used extract scores non-vegan ordination results. Many standard ordination methods libraries specific class, specific method can written . However, scores.default guesses commonly used functions keep site scores possible species scores. x matrix, scores.default returns chosen columns matrix, ignoring whether species sites requested (regard bug feature, please). Currently function seems work least isoMDS, prcomp, princomp ade4 objects. may work cases fail mysteriously.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Species or Site Scores from an Ordination — scores","text":"function returns matrix scores one type requested, named list matrices display = \"\", ggplot2 compatible data frame tidy = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get Species or Site Scores from an Ordination — scores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"data(varespec) vare.pca <- prcomp(varespec) scores(vare.pca, choices=c(1,2)) #> PC1 PC2 #> 18 -10.7847878 18.7094315 #> 15 -27.8036826 -11.7414745 #> 24 -25.6919559 -14.5399684 #> 27 -31.7820166 -31.2216800 #> 23 -19.6315869 -2.5541193 #> 19 -0.2413294 -11.4974077 #> 22 -26.6771373 -12.3140897 #> 16 -21.9230366 0.4449159 #> 28 -39.6083051 -41.8877392 #> 13 -4.0664328 20.4191153 #> 14 -18.4416245 5.4406988 #> 20 -17.3999191 2.3653380 #> 25 -25.1673547 -13.2508067 #> 7 -11.4065430 41.7356300 #> 5 -8.4243752 45.3805255 #> 6 -2.0759474 36.9311222 #> 3 39.8617580 8.0590041 #> 4 13.1065901 12.8377217 #> 2 57.6827011 -4.8983565 #> 9 63.3138332 -22.4481549 #> 12 44.1073111 -10.1653935 #> 10 64.9418975 -16.7633564 #> 11 11.5313633 3.9720890 #> 21 -3.4194194 -3.0130455"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Screeplot methods plotting variances ordination axes/components overlaying broken stick distributions. Also, provides alternative screeplot methods princomp prcomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"# S3 method for cca screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, if (is.null(x$CCA) || x$CCA$rank == 0) x$CA$rank else x$CCA$rank), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for decorana screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = 4, ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for prcomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for princomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) bstick(n, ...) # S3 method for default bstick(n, tot.var = 1, ...) # S3 method for cca bstick(n, ...) # S3 method for prcomp bstick(n, ...) # S3 method for princomp bstick(n, ...) # S3 method for decorana bstick(n, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"x object component variances can determined. bstick logical; broken stick distribution drawn? npcs number components plotted. type type plot. ptype type == \"lines\" bstick = TRUE, character indicating type plotting used lines; actually types plot.default. bst.col, bst.lty colour line type used draw broken stick distribution. xlab, ylab, main graphics parameters. legend logical; draw legend? n object variances can extracted number variances (components) case bstick.default. tot.var total variance split. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"functions provide screeplots ordination methods vegan enhanced versions broken stick prcomp princomp. Function bstick gives brokenstick values ordered random proportions, defined \\(p_i = (tot/n) \\sum_{x=}^n (1/x)\\) (Legendre & Legendre 2012), \\(tot\\) total \\(n\\) number brokenstick components (cf. radfit). Broken stick recommended stopping rule principal component analysis (Jackson 1993): principal components retained long observed eigenvalues higher corresponding random broken stick components. bstick function generic. default needs number components total, specific methods extract information ordination results. also bstick method cca. However, broken stick model strictly valid correspondence analysis (CA), eigenvalues CA defined \\(\\leq 1\\), whereas brokenstick components restrictions. brokenstick components detrended correspondence analysis (DCA) assume input data full rank, additive eigenvalues used screeplot (see decorana).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Function screeplot draws plot currently active device, returns invisibly xy.coords points bars eigenvalues. Function bstick returns numeric vector broken stick components.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Jackson, D. . (1993). Stopping rules principal components analysis: comparison heuristical statistical approaches. Ecology 74, 2204--2214. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"data(varespec) vare.pca <- rda(varespec, scale = TRUE) bstick(vare.pca) #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 7.1438620 5.2308185 4.2742968 3.6366156 3.1583548 2.7757461 2.4569055 2.1836136 #> PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 #> 1.9444831 1.7319228 1.5406184 1.3667054 1.2072851 1.0601279 0.9234819 0.7959457 #> PC17 PC18 PC19 PC20 PC21 PC22 PC23 #> 0.6763805 0.5638485 0.4575683 0.3568818 0.2612296 0.1701323 0.0831758 screeplot(vare.pca, bstick = TRUE, type = \"lines\")"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":null,"dir":"Reference","previous_headings":"","what":"Similarity Percentages — simper","title":"Similarity Percentages — simper","text":"Discriminating species two groups using Bray-Curtis dissimilarities","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Similarity Percentages — simper","text":"","code":"simper(comm, group, permutations = 999, parallel = 1, ...) # S3 method for simper summary(object, ordered = TRUE, digits = max(3,getOption(\"digits\") - 3), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Similarity Percentages — simper","text":"comm Community data. group Factor describing group structure. missing one level, contributions estimated non-grouped data dissimilarities show overall heterogeneity species abundances. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. object object returned simper. ordered Logical; species ordered average contribution? digits Number digits output. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. (yet implemented). ... Parameters passed functions. simper extra parameters passed shuffleSet permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Similarity Percentages — simper","text":"Similarity percentage, simper (Clarke 1993) based decomposition Bray-Curtis dissimilarity index (see vegdist, designdist). contribution individual species \\(\\) overall Bray-Curtis dissimilarity \\(d_{jk}\\) given $$d_{ijk} = \\frac{|x_{ij}-x_{ik}|}{\\sum_{=1}^S (x_{ij}+x_{ik})}$$ \\(x\\) abundance species \\(\\) sampling units \\(j\\) \\(k\\). overall index sum individual contributions \\(S\\) species \\(d_{jk}=\\sum_{=1}^S d_{ijk}\\). simper functions performs pairwise comparisons groups sampling units finds contribution species average -group Bray-Curtis dissimilarity. Although method called “Similarity Percentages”, really studied dissimilarities instead similarities (Clarke 1993). function displays important species pair groups. species contribute least 70 % differences groups. function returns much extensive results (including species) can accessed directly result object (see section Value). Function summary transforms result list data frames. argument ordered = TRUE data frames also include cumulative contributions ordered species contribution. results simper can difficult interpret often misunderstood even publications. method gives contribution species overall dissimilarities, caused variation species abundances, partly differences among groups. Even make groups copies , method single species high contribution, contributions non-existing -group differences random noise variation species abundances. abundant species usually highest variances, high contributions even differ among groups. Permutation tests study differences among groups, can used find species differences among groups important component contribution dissimilarities. Analysis without group argument find species contributions average overall dissimilarity among sampling units. non-grouped contributions can compared grouped contributions see much added value grouping species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Similarity Percentages — simper","text":"list class \"simper\" following items: species species names. average Species contribution average -group dissimilarity. overall average -group dissimilarity. sum item average. sd Standard deviation contribution. ratio Average sd ratio. ava, avb Average abundances per group. ord index vector order vectors contribution order cusum back original data order. cusum Ordered cumulative contribution. based item average, sum total 1. p Permutation \\(p\\)-value. Probability getting larger equal average contribution random permutation group factor. area available permutations used (default: calculated).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Similarity Percentages — simper","text":"Eduard Szöcs Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Similarity Percentages — simper","text":"Clarke, K.R. 1993. Non-parametric multivariate analyses changes community structure. Australian Journal Ecology, 18, 117–143.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Similarity Percentages — simper","text":"","code":"data(dune) data(dune.env) (sim <- with(dune.env, simper(dune, Management, permutations = 99))) #> cumulative contributions of most influential species: #> #> $SF_BF #> Agrostol Alopgeni Lolipere Trifrepe Poatriv Scorautu Bromhord #> 0.09824271 0.18254830 0.25956958 0.33367870 0.40734444 0.47729205 0.53120026 #> Achimill Planlanc Elymrepe Bracruta #> 0.57946526 0.62522255 0.67016196 0.71098133 #> #> $SF_HF #> Agrostol Alopgeni Lolipere Planlanc Rumeacet Elymrepe Poatriv #> 0.08350879 0.16534834 0.23934930 0.30843624 0.37716139 0.43334492 0.48351753 #> Bracruta Eleopalu Poaprat Anthodor Sagiproc Trifprat #> 0.52804045 0.57205850 0.61423981 0.65549838 0.69628951 0.73696831 #> #> $SF_NM #> Poatriv Alopgeni Agrostol Lolipere Eleopalu Poaprat Bracruta Elymrepe #> 0.1013601 0.1935731 0.2667383 0.3377578 0.3999419 0.4526707 0.5044725 0.5505643 #> Scorautu Trifrepe Sagiproc Salirepe #> 0.5926117 0.6320111 0.6712478 0.7091528 #> #> $BF_HF #> Rumeacet Poatriv Planlanc Bromhord Lolipere Elymrepe Trifrepe #> 0.08163219 0.15193797 0.21918333 0.27967181 0.33969561 0.39843338 0.45298204 #> Anthodor Achimill Bracruta Alopgeni Trifprat Juncarti #> 0.50276849 0.55222648 0.60021994 0.64584333 0.69126471 0.73366621 #> #> $BF_NM #> Lolipere Poatriv Poaprat Trifrepe Bromhord Bracruta Eleopalu Agrostol #> 0.1242718 0.1992126 0.2711756 0.3414609 0.3958520 0.4448077 0.4910724 0.5369083 #> Achimill Scorautu Anthodor Planlanc #> 0.5823926 0.6253645 0.6638182 0.7012577 #> #> $HF_NM #> Poatriv Lolipere Rumeacet Poaprat Planlanc Bracruta Eleopalu #> 0.09913221 0.17468460 0.23917190 0.29701331 0.35469313 0.40365488 0.44804851 #> Agrostol Trifrepe Elymrepe Anthodor Juncarti Trifprat Salirepe #> 0.49226546 0.53434466 0.57564661 0.61543243 0.65341300 0.68921695 0.72432408 #> ## IGNORE_RDIFF_BEGIN summary(sim) #> #> Contrast: SF_BF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.06137 0.03419 1.79490 4.66700 0.00000 0.098 0.05 * #> Alopgeni 0.05267 0.03648 1.44390 4.33300 0.66700 0.182 0.14 #> Lolipere 0.04812 0.03945 1.21980 3.00000 6.00000 0.260 0.39 #> Trifrepe 0.04630 0.02553 1.81380 1.33300 4.66700 0.334 0.09 . #> Poatriv 0.04602 0.03380 1.36150 4.66700 3.66700 0.407 0.46 #> Scorautu 0.04370 0.02492 1.75340 1.33300 4.33300 0.477 0.04 * #> Bromhord 0.03368 0.02586 1.30230 0.50000 2.66700 0.531 0.02 * #> Achimill 0.03015 0.02082 1.44820 0.16700 2.33300 0.580 0.04 * #> Planlanc 0.02859 0.02155 1.32650 0.00000 2.00000 0.625 0.49 #> Elymrepe 0.02807 0.02978 0.94280 2.00000 1.33300 0.670 0.52 #> Bracruta 0.02550 0.02390 1.06690 2.00000 2.00000 0.711 0.83 #> Poaprat 0.02513 0.02397 1.04850 2.50000 4.00000 0.751 0.82 #> Sagiproc 0.02433 0.02215 1.09830 1.83300 0.66700 0.790 0.39 #> Bellpere 0.01986 0.01709 1.16220 0.66700 1.66700 0.822 0.10 . #> Eleopalu 0.01861 0.04296 0.43330 1.33300 0.00000 0.852 0.82 #> Anthodor 0.01754 0.02580 0.67980 0.00000 1.33300 0.880 0.75 #> Juncbufo 0.01603 0.02371 0.67620 1.16700 0.00000 0.905 0.57 #> Vicilath 0.01467 0.01331 1.10260 0.00000 1.00000 0.929 0.04 * #> Hyporadi 0.01029 0.01520 0.67680 0.00000 0.66700 0.945 0.62 #> Ranuflam 0.00931 0.01360 0.68450 0.66700 0.00000 0.960 0.93 #> Juncarti 0.00698 0.01611 0.43330 0.50000 0.00000 0.972 0.95 #> Callcusp 0.00698 0.01611 0.43330 0.50000 0.00000 0.983 0.79 #> Rumeacet 0.00453 0.01044 0.43330 0.33300 0.00000 0.990 0.95 #> Cirsarve 0.00398 0.00918 0.43360 0.33300 0.00000 0.996 0.37 #> Chenalbu 0.00233 0.00537 0.43330 0.16700 0.00000 1.000 0.41 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_HF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.04738 0.03127 1.51510 4.66700 1.40000 0.084 0.35 #> Alopgeni 0.04643 0.03290 1.41150 4.33300 1.60000 0.165 0.20 #> Lolipere 0.04199 0.02701 1.55460 3.00000 4.00000 0.239 0.75 #> Planlanc 0.03920 0.03321 1.18040 0.00000 3.00000 0.308 0.02 * #> Rumeacet 0.03899 0.02737 1.42470 0.33300 3.20000 0.377 0.01 ** #> Elymrepe 0.03188 0.02955 1.07870 2.00000 2.00000 0.433 0.30 #> Poatriv 0.02847 0.02152 1.32270 4.66700 4.80000 0.484 1.00 #> Bracruta 0.02526 0.02104 1.20040 2.00000 2.80000 0.528 0.92 #> Eleopalu 0.02497 0.03888 0.64240 1.33300 0.80000 0.572 0.77 #> Poaprat 0.02393 0.01918 1.24780 2.50000 3.40000 0.614 0.98 #> Anthodor 0.02341 0.02143 1.09230 0.00000 1.80000 0.655 0.67 #> Sagiproc 0.02314 0.02048 1.13010 1.83300 0.80000 0.696 0.40 #> Trifprat 0.02308 0.02343 0.98500 0.00000 1.80000 0.737 0.01 ** #> Juncarti 0.02285 0.02568 0.88990 0.50000 1.60000 0.777 0.51 #> Trifrepe 0.02238 0.01949 1.14860 1.33300 2.80000 0.817 0.94 #> Juncbufo 0.02164 0.02224 0.97330 1.16700 1.20000 0.855 0.24 #> Scorautu 0.02051 0.01642 1.24890 1.33300 2.80000 0.891 0.79 #> Achimill 0.01518 0.01139 1.33260 0.16700 1.20000 0.918 0.75 #> Bromhord 0.01338 0.01450 0.92220 0.50000 0.80000 0.941 0.79 #> Ranuflam 0.01066 0.01339 0.79640 0.66700 0.40000 0.960 0.86 #> Bellpere 0.00999 0.01257 0.79480 0.66700 0.40000 0.978 0.84 #> Callcusp 0.00662 0.01508 0.43930 0.50000 0.00000 0.989 0.92 #> Cirsarve 0.00381 0.00867 0.43940 0.33300 0.00000 0.996 0.54 #> Chenalbu 0.00221 0.00503 0.43930 0.16700 0.00000 1.000 0.51 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Hyporadi 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Vicilath 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07828 0.04095 1.91180 4.66700 0.00000 0.101 0.01 ** #> Alopgeni 0.07122 0.04696 1.51670 4.33300 0.00000 0.194 0.01 ** #> Agrostol 0.05651 0.04418 1.27920 4.66700 2.16700 0.267 0.05 * #> Lolipere 0.05485 0.05991 0.91550 3.00000 0.33300 0.338 0.15 #> Eleopalu 0.04803 0.04717 1.01820 1.33300 2.16700 0.400 0.04 * #> Poaprat 0.04072 0.03179 1.28100 2.50000 0.66700 0.453 0.06 . #> Bracruta 0.04001 0.03440 1.16310 2.00000 2.83300 0.504 0.08 . #> Elymrepe 0.03560 0.03852 0.92430 2.00000 0.00000 0.551 0.12 #> Scorautu 0.03247 0.03481 0.93280 1.33300 3.16700 0.593 0.13 #> Trifrepe 0.03043 0.03163 0.96190 1.33300 1.83300 0.632 0.59 #> Sagiproc 0.03030 0.03048 0.99430 1.83300 0.50000 0.671 0.02 * #> Salirepe 0.02928 0.03201 0.91440 0.00000 1.83300 0.709 0.02 * #> Anthodor 0.02454 0.03669 0.66880 0.00000 1.33300 0.741 0.56 #> Callcusp 0.02276 0.02944 0.77310 0.50000 1.16700 0.770 0.08 . #> Ranuflam 0.02257 0.02282 0.98890 0.66700 1.33300 0.800 0.08 . #> Juncarti 0.02254 0.02860 0.78830 0.50000 1.16700 0.829 0.53 #> Hyporadi 0.02011 0.03129 0.64260 0.00000 1.16700 0.855 0.21 #> Juncbufo 0.01986 0.02903 0.68400 1.16700 0.00000 0.881 0.21 #> Planlanc 0.01542 0.02277 0.67720 0.00000 0.83300 0.900 0.98 #> Airaprae 0.01488 0.02188 0.68020 0.00000 0.83300 0.920 0.06 . #> Bellpere 0.01232 0.01592 0.77370 0.66700 0.33300 0.936 0.72 #> Comapalu 0.01188 0.01741 0.68260 0.00000 0.66700 0.951 0.05 * #> Achimill 0.00929 0.01493 0.62240 0.16700 0.33300 0.963 0.98 #> Bromhord 0.00717 0.01633 0.43910 0.50000 0.00000 0.972 0.98 #> Rumeacet 0.00559 0.01275 0.43840 0.33300 0.00000 0.980 0.98 #> Empenigr 0.00523 0.01200 0.43540 0.00000 0.33300 0.986 0.29 #> Cirsarve 0.00478 0.01089 0.43910 0.33300 0.00000 0.993 0.02 * #> Chenalbu 0.00289 0.00660 0.43820 0.16700 0.00000 0.996 0.02 * #> Vicilath 0.00279 0.00642 0.43450 0.00000 0.16700 1.000 0.81 #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_HF #> #> average sd ratio ava avb cumsum p #> Rumeacet 0.03867 0.02606 1.48380 0.00000 3.20000 0.082 0.03 * #> Poatriv 0.03330 0.02579 1.29110 3.66700 4.80000 0.152 0.96 #> Planlanc 0.03185 0.01830 1.74010 2.00000 3.00000 0.219 0.34 #> Bromhord 0.02865 0.01799 1.59260 2.66700 0.80000 0.280 0.07 . #> Lolipere 0.02843 0.02215 1.28340 6.00000 4.00000 0.340 1.00 #> Elymrepe 0.02782 0.02959 0.94040 1.33300 2.00000 0.398 0.56 #> Trifrepe 0.02584 0.01656 1.56030 4.66700 2.80000 0.453 0.78 #> Anthodor 0.02358 0.02042 1.15470 1.33300 1.80000 0.503 0.55 #> Achimill 0.02343 0.01474 1.58930 2.33300 1.20000 0.552 0.24 #> Bracruta 0.02273 0.01802 1.26170 2.00000 2.80000 0.600 0.89 #> Alopgeni 0.02161 0.02308 0.93630 0.66700 1.60000 0.646 0.91 #> Trifprat 0.02151 0.02207 0.97470 0.00000 1.80000 0.691 0.12 #> Juncarti 0.02008 0.02555 0.78600 0.00000 1.60000 0.734 0.59 #> Scorautu 0.01932 0.01357 1.42410 4.33300 2.80000 0.774 0.78 #> Bellpere 0.01829 0.01486 1.23050 1.66700 0.40000 0.813 0.22 #> Agrostol 0.01761 0.02284 0.77080 0.00000 1.40000 0.850 1.00 #> Juncbufo 0.01500 0.02066 0.72600 0.00000 1.20000 0.882 0.67 #> Vicilath 0.01285 0.01140 1.12740 1.00000 0.00000 0.909 0.04 * #> Sagiproc 0.01168 0.01297 0.90080 0.66700 0.80000 0.934 0.92 #> Eleopalu 0.01017 0.02111 0.48170 0.00000 0.80000 0.955 0.93 #> Hyporadi 0.00895 0.01312 0.68240 0.66700 0.00000 0.974 0.64 #> Poaprat 0.00720 0.01010 0.71330 4.00000 3.40000 0.989 1.00 #> Ranuflam 0.00508 0.01055 0.48170 0.00000 0.40000 1.000 0.97 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Callcusp 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_NM #> #> average sd ratio ava avb cumsum p #> Lolipere 0.09068 0.02644 3.42900 6.00000 0.33300 0.124 0.04 * #> Poatriv 0.05468 0.04465 1.22500 3.66700 0.00000 0.199 0.28 #> Poaprat 0.05251 0.01813 2.89700 4.00000 0.66700 0.271 0.02 * #> Trifrepe 0.05129 0.02756 1.86100 4.66700 1.83300 0.342 0.04 * #> Bromhord 0.03969 0.02920 1.35900 2.66700 0.00000 0.396 0.01 ** #> Bracruta 0.03572 0.02869 1.24500 2.00000 2.83300 0.445 0.40 #> Eleopalu 0.03376 0.03573 0.94500 0.00000 2.16700 0.491 0.43 #> Agrostol 0.03345 0.03473 0.96300 0.00000 2.16700 0.537 0.87 #> Achimill 0.03319 0.02338 1.42000 2.33300 0.33300 0.582 0.02 * #> Scorautu 0.03136 0.02026 1.54800 4.33300 3.16700 0.625 0.29 #> Anthodor 0.02806 0.03295 0.85200 1.33300 1.33300 0.664 0.37 #> Planlanc 0.02732 0.02193 1.24600 2.00000 0.83300 0.701 0.62 #> Salirepe 0.02677 0.02927 0.91400 0.00000 1.83300 0.738 0.11 #> Bellpere 0.02353 0.01909 1.23200 1.66700 0.33300 0.770 0.04 * #> Hyporadi 0.02172 0.02450 0.88600 0.66700 1.16700 0.800 0.26 #> Ranuflam 0.02031 0.02275 0.89300 0.00000 1.33300 0.828 0.27 #> Elymrepe 0.01999 0.02926 0.68300 1.33300 0.00000 0.855 0.79 #> Callcusp 0.01783 0.02681 0.66500 0.00000 1.16700 0.880 0.44 #> Juncarti 0.01769 0.02600 0.68100 0.00000 1.16700 0.904 0.73 #> Vicilath 0.01577 0.01447 1.09000 1.00000 0.16700 0.925 0.01 ** #> Sagiproc 0.01543 0.01857 0.83100 0.66700 0.50000 0.947 0.79 #> Airaprae 0.01341 0.01969 0.68100 0.00000 0.83300 0.965 0.30 #> Comapalu 0.01074 0.01571 0.68400 0.00000 0.66700 0.980 0.42 #> Alopgeni 0.01000 0.01463 0.68300 0.66700 0.00000 0.993 0.99 #> Empenigr 0.00479 0.01105 0.43300 0.00000 0.33300 1.000 0.41 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Juncbufo 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Rumeacet 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: HF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07155 0.01368 5.23000 4.80000 0.00000 0.099 0.01 ** #> Lolipere 0.05453 0.02962 1.84100 4.00000 0.33300 0.175 0.22 #> Rumeacet 0.04655 0.03081 1.51100 3.20000 0.00000 0.239 0.01 ** #> Poaprat 0.04175 0.01885 2.21500 3.40000 0.66700 0.297 0.06 . #> Planlanc 0.04163 0.02956 1.40800 3.00000 0.83300 0.355 0.03 * #> Bracruta 0.03534 0.02010 1.75800 2.80000 2.83300 0.404 0.40 #> Eleopalu 0.03204 0.03231 0.99200 0.80000 2.16700 0.448 0.51 #> Agrostol 0.03192 0.02889 1.10500 1.40000 2.16700 0.492 0.99 #> Trifrepe 0.03037 0.02287 1.32800 2.80000 1.83300 0.534 0.71 #> Elymrepe 0.02981 0.03868 0.77100 2.00000 0.00000 0.576 0.53 #> Anthodor 0.02872 0.02480 1.15800 1.80000 1.33300 0.615 0.24 #> Juncarti 0.02741 0.02854 0.96100 1.60000 1.16700 0.653 0.19 #> Trifprat 0.02584 0.02597 0.99500 1.80000 0.00000 0.689 0.01 ** #> Salirepe 0.02534 0.02729 0.92900 0.00000 1.83300 0.724 0.18 #> Alopgeni 0.02446 0.03240 0.75500 1.60000 0.00000 0.758 0.92 #> Scorautu 0.02070 0.01412 1.46600 2.80000 3.16700 0.787 0.85 #> Ranuflam 0.01928 0.01994 0.96700 0.40000 1.33300 0.814 0.36 #> Juncbufo 0.01818 0.02465 0.73800 1.20000 0.00000 0.839 0.43 #> Hyporadi 0.01714 0.02655 0.64600 0.00000 1.16700 0.863 0.42 #> Callcusp 0.01683 0.02490 0.67600 0.00000 1.16700 0.886 0.42 #> Achimill 0.01656 0.01490 1.11100 1.20000 0.33300 0.909 0.59 #> Sagiproc 0.01528 0.01653 0.92400 0.80000 0.50000 0.930 0.89 #> Airaprae 0.01261 0.01824 0.69100 0.00000 0.83300 0.947 0.30 #> Bromhord 0.01209 0.01517 0.79700 0.80000 0.00000 0.964 0.84 #> Comapalu 0.01011 0.01456 0.69400 0.00000 0.66700 0.978 0.25 #> Bellpere 0.00880 0.01373 0.64100 0.40000 0.33300 0.990 0.94 #> Empenigr 0.00454 0.01033 0.43900 0.00000 0.33300 0.997 0.60 #> Vicilath 0.00240 0.00546 0.43900 0.00000 0.16700 1.000 0.86 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 99 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Function simulates response data frame adds Gaussian error fitted responses Redundancy Analysis (rda), Constrained Correspondence Analysis (cca) distance-based RDA (capscale). function special case generic simulate, works similarly simulate.lm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"# S3 method for rda simulate(object, nsim = 1, seed = NULL, indx = NULL, rank = \"full\", correlated = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"object object representing fitted rda, cca capscale model. nsim number response matrices simulated. one dissimilarity matrix returned capscale, larger nsim error. seed object specifying random number generator initialized (‘seeded’). See simulate details. indx Index residuals added fitted values, produced shuffleSet sample. index can duplicate entries bootstrapping allowed. nsim \\(>1\\), output compliant shuffleSet one line simulation. nsim missing, number rows indx used define number simulations, nsim given, match number rows indx. null, parametric simulation used Gaussian error added fitted values. rank rank constrained component: passed predict.rda predict.cca. correlated species regarded correlated parametric simulation indx given? correlated = TRUE, multivariate Gaussian random error generated, FALSE, Gaussian random error generated separately species. argument effect capscale information species. ... additional optional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"implementation follows \"lm\" method simulate, adds Gaussian (Normal) error fitted values (fitted.rda) using function rnorm correlated = FALSE mvrnorm correlated = TRUE. standard deviations (rnorm) covariance matrices species (mvrnorm) estimated residuals fitting constraints. Alternatively, function can take permutation index used add permuted residuals (unconstrained component) fitted values. Raw data used rda. Internal Chi-square transformed data used cca within function, returned matrix similar original input data. simulation performed internal metric scaling data capscale, function returns Euclidean distances calculated simulated data. simulation uses real components, imaginary dimensions ignored.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"nsim = 1, returns matrix dissimilarities ( capscale) similar additional arguments random number seed simulate. nsim > 1, returns similar array returned simulate.nullmodel similar attributes.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"data(dune) data(dune.env) mod <- rda(dune ~ Moisture + Management, dune.env) ## One simulation update(mod, simulate(mod) ~ .) #> Call: rda(formula = simulate(mod) ~ Moisture + Management, data = #> dune.env) #> #> Inertia Proportion Rank #> Total 79.3906 1.0000 #> Constrained 52.2955 0.6587 6 #> Unconstrained 27.0951 0.3413 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 24.007 14.238 5.712 3.314 2.879 2.145 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 6.128 4.391 3.331 3.143 2.121 1.757 1.488 1.152 1.115 1.044 0.682 0.539 0.205 #> ## An impression of confidence regions of site scores plot(mod, display=\"sites\") for (i in 1:5) lines(procrustes(mod, update(mod, simulate(mod) ~ .)), col=\"blue\") ## Simulate a set of null communities with permutation of residuals simulate(mod, indx = shuffleSet(nrow(dune), 99)) #> An object of class “simulate.rda” #> ‘simulate index’ method (abundance, non-sequential) #> 20 x 30 matrix #> Number of permuted matrices = 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":null,"dir":"Reference","previous_headings":"","what":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Land birds islands covered coniferous forest Sipoo Archipelago, southern Finland.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map)"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"sipoo data frame contains data occurrences 50 land bird species 18 islands Sipoo Archipelago (Simberloff & Martin, 1991, Appendix 3). species referred 4+4 letter abbreviation Latin names (using five letters two species names make unique). sipoo.map data contains geographic coordinates islands ETRS89-TM35FIN coordinate system (EPSG:3067) areas islands hectares.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Simberloff, D. & Martin, J.-L. (1991). Nestedness insular avifaunas: simple summary statistics masking complex species patterns. Ornis Fennica 68:178--192.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map) plot(N ~ E, data=sipoo.map, asp = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":null,"dir":"Reference","previous_headings":"","what":"Minimum Spanning Tree — spantree","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree connecting points, disregarding dissimilarities threshold NA.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Minimum Spanning Tree — spantree","text":"","code":"spantree(d, toolong = 0) # S3 method for spantree as.hclust(x, ...) # S3 method for spantree cophenetic(x) spandepth(x) # S3 method for spantree plot(x, ord, cex = 0.7, type = \"p\", labels, dlim, FUN = sammon, ...) # S3 method for spantree lines(x, ord, display=\"sites\", col = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Minimum Spanning Tree — spantree","text":"d Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. x spantree result object. ord ordination configuration, ordination result known scores. cex Character expansion factor. type Observations plotted points type=\"p\" type=\"b\", text label type=\"t\". tree (lines) always plotted. labels Text used type=\"t\" node names missing. dlim ceiling value used highest cophenetic dissimilarity. FUN Ordination function find configuration cophenetic dissimilarities. supplied FUN work, supply ordination result argument ord. display Type scores used ord. col Colour line segments. can vector recycled points, line colour mixture two joined points. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree dissimilarities (may several minimum spanning trees, function finds one). Dissimilarities threshold toolong NAs disregarded, spanning tree found dissimilarities. data disconnected, function return disconnected tree (forest), corresponding link NA. Connected subtrees can identified using distconnected. Minimum spanning tree closely related single linkage clustering, .k.. nearest neighbour clustering, genetics neighbour joining tree available hclust agnes functions. important practical difference minimum spanning tree concept cluster membership, always joins individual points . Function .hclust can change spantree result corresponding hclust object. Function cophenetic finds distances points along tree segments. Function spandepth returns depth node. nodes tree either leaves (one link) internal nodes (one link). leaves recursively removed tree, depth layer leaf removed. disconnected spantree object (forest) tree analysed separately disconnected nodes tree depth zero. Function plot displays tree supplied ordination configuration, lines adds spanning tree ordination graph. configuration supplied plot, function ordinates cophenetic dissimilarities spanning tree overlays tree result. default ordination function sammon (package MASS), Sammon scaling emphasizes structure neighbourhood nodes may able beautifully represent tree (may need set dlim, sometimes results remain twisted). ordination methods work disconnected trees, must supply ordination configuration. Function lines overlay tree existing plot. Function spantree uses Prim's method implemented priority-first search dense graphs (Sedgewick 1990). Function cophenetic uses function stepacross option path = \"extended\". spantree fast, cophenetic slow large data sets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Minimum Spanning Tree — spantree","text":"Function spantree returns object class spantree list two vectors, length \\(n-1\\). number links tree one less number observations, first item omitted. items kid child node parent, starting parent number two. link parent, value NA tree disconnected node. dist Corresponding distance. kid = NA, dist = 0. labels Names nodes found input dissimilarities. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Minimum Spanning Tree — spantree","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Minimum Spanning Tree — spantree","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Minimum Spanning Tree — spantree","text":"principle, minimum spanning tree equivalent single linkage clustering can performed using hclust agnes. However, functions combine clusters information actually connected points (“single link”) recovered result. graphical output single linkage clustering plotted ordicluster look different equivalent spanning tree plotted lines.spantree.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Minimum Spanning Tree — spantree","text":"","code":"data(dune) dis <- vegdist(dune) tr <- spantree(dis) ## Add tree to a metric scaling plot(tr, cmdscale(dis), type = \"t\") ## Find a configuration to display the tree neatly plot(tr, type = \"t\") #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Depths of nodes depths <- spandepth(tr) plot(tr, type = \"t\", label = depths) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Plot as a dendrogram cl <- as.hclust(tr) plot(cl) ## cut hclust tree to classes and show in colours in spantree plot(tr, col = cutree(cl, 5), pch=16) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":null,"dir":"Reference","previous_headings":"","what":"Species Accumulation Curves — specaccum","title":"Species Accumulation Curves — specaccum","text":"Function specaccum finds species accumulation curves number species certain number sampled sites individuals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species Accumulation Curves — specaccum","text":"","code":"specaccum(comm, method = \"exact\", permutations = 100, conditioned =TRUE, gamma = \"jack1\", w = NULL, subset, ...) # S3 method for specaccum plot(x, add = FALSE, random = FALSE, ci = 2, ci.type = c(\"bar\", \"line\", \"polygon\"), col = par(\"fg\"), lty = 1, ci.col = col, ci.lty = 1, ci.length = 0, xlab, ylab = x$method, ylim, xvar = c(\"sites\", \"individuals\", \"effort\"), ...) # S3 method for specaccum boxplot(x, add = FALSE, ...) fitspecaccum(object, model, method = \"random\", ...) # S3 method for fitspecaccum plot(x, col = par(\"fg\"), lty = 1, xlab = \"Sites\", ylab = x$method, ...) # S3 method for specaccum predict(object, newdata, interpolation = c(\"linear\", \"spline\"), ...) # S3 method for fitspecaccum predict(object, newdata, ...) specslope(object, at)"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species Accumulation Curves — specaccum","text":"comm Community data set. method Species accumulation method (partial match). Method \"collector\" adds sites order happen data, \"random\" adds sites random order, \"exact\" finds expected (mean) species richness, \"coleman\" finds expected richness following Coleman et al. 1982, \"rarefaction\" finds mean accumulating individuals instead sites. permutations Number permutations method = \"random\". Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. conditioned Estimation standard deviation conditional empirical dataset exact SAC gamma Method estimating total extrapolated number species survey area function specpool w Weights giving sampling effort. subset logical expression indicating sites (rows) keep: missing values taken FALSE. x specaccum result object add Add existing graph. random Draw random simulation separately instead drawing average confidence intervals. ci Multiplier used get confidence intervals standard deviation (standard error estimate). Value ci = 0 suppresses drawing confidence intervals. ci.type Type confidence intervals graph: \"bar\" draws vertical bars, \"line\" draws lines, \"polygon\" draws shaded area. col Colour drawing lines. lty line type (see par). ci.col Colour drawing lines filling \"polygon\". ci.lty Line type confidence intervals border \"polygon\". ci.length Length horizontal bars (inches) end vertical bars ci.type = \"bar\". xlab,ylab Labels x (defaults xvar) y axis. ylim y limits plot. xvar Variable used horizontal axis: \"individuals\" can used method = \"rarefaction\". object Either community data set fitted specaccum model. model Nonlinear regression model (nls). See Details. newdata Optional data used prediction interpreted number sampling units (sites). missing, fitted values returned. interpolation Interpolation method used newdata. Number plots slope evaluated. Can real number. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species Accumulation Curves — specaccum","text":"Species accumulation curves (SAC) used compare diversity properties community data sets using different accumulator functions. classic method \"random\" finds mean SAC standard deviation random permutations data, subsampling without replacement (Gotelli & Colwell 2001). \"exact\" method finds expected SAC using sample-based rarefaction method independently developed numerous times (Chiarucci et al. 2008) often known Mao Tau estimate (Colwell et al. 2012). unconditional standard deviation exact SAC represents moment-based estimation conditioned empirical data set (sd samples > 0). unconditional standard deviation based estimation extrapolated number species survey area (.k.. gamma diversity), estimated function specpool. conditional standard deviation developed Jari Oksanen ( published, sd=0 samples). Method \"coleman\" finds expected SAC standard deviation following Coleman et al. (1982). methods based sampling sites without replacement. contrast, method = \"rarefaction\" finds expected species richness standard deviation sampling individuals instead sites. achieves applying function rarefy number individuals corresponding average number individuals per site. Methods \"random\" \"collector\" can take weights (w) give sampling effort site. weights w influence order sites accumulated, value sampling effort sites equal. summary results expressed sites even accumulation uses weights (methods \"random\", \"collector\"), based individuals (\"rarefaction\"). actual sampling effort given item Effort Individuals printed result. weighted \"random\" method effort refers average effort per site, sum weights per number sites. weighted method = \"random\", averaged species richness found linear interpolation single random permutations. Therefore least first value (often several first) NA richness, values interpolated cases extrapolated. plot function defaults display results scaled sites, can changed selecting xvar = \"effort\" (weighted methods) xvar = \"individuals\" (method = \"rarefaction\"). summary boxplot methods available method = \"random\". Function predict specaccum can return values corresponding newdata. method \"exact\", \"rarefaction\" \"coleman\" function uses analytic equations interpolated non-integer values, methods linear (approx) spline (spline) interpolation. newdata given, function returns values corresponding data. NB., fitted values method=\"rarefaction\" based rounded integer counts, predict can use fractional non-integer counts newdata give slightly different results. Function fitspecaccum fits nonlinear (nls) self-starting species accumulation model. input object can result specaccum community data frame. latter case function first fits specaccum model proceeds fitting nonlinear model. function can apply limited set nonlinear regression models suggested species-area relationship (Dengler 2009). selfStart models. permissible alternatives \"arrhenius\" (SSarrhenius), \"gleason\" (SSgleason), \"gitay\" (SSgitay), \"lomolino\" (SSlomolino) vegan package. addition following standard R models available: \"asymp\" (SSasymp), \"gompertz\" (SSgompertz), \"michaelis-menten\" (SSmicmen), \"logis\" (SSlogis), \"weibull\" (SSweibull). See functions model specification details. weights w used fit based accumulated effort model = \"rarefaction\" accumulated number individuals. plot still based sites, unless alternative selected xvar. Function predict fitspecaccum uses predict.nls, can pass arguments function. addition, fitted, residuals, nobs, coef, AIC, logLik deviance work result object. Function specslope evaluates derivative species accumulation curve given number sample plots, gives rate increase number species. function works specaccum result object based analytic models \"exact\", \"rarefaction\" \"coleman\", non-linear regression results fitspecaccum. Nonlinear regression may fail reason, fitspecaccum models fragile may succeed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species Accumulation Curves — specaccum","text":"Function specaccum returns object class \"specaccum\", fitspecaccum model class \"fitspecaccum\" adds items \"specaccum\" (see end list ): call Function call. method Accumulator method. sites Number sites. method = \"rarefaction\" number sites corresponding certain number individuals generally integer, average number individuals also returned item individuals. effort Average sum weights corresponding number sites model fitted argument w richness number species corresponding number sites. method = \"collector\" observed richness, methods average expected richness. sd standard deviation SAC (standard error). NULL method = \"collector\", estimated permutations method = \"random\", analytic equations methods. perm Permutation results method = \"random\" NULL cases. column perm holds one permutation. weights Matrix accumulated weights corresponding columns perm matrix model fitted argument w. fitted, residuals, coefficients fitspecacum: fitted values, residuals nonlinear model coefficients. method = \"random\" matrices column random accumulation. models fitspecaccum: list fitted nls models (see Examples accessing models).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Species Accumulation Curves — specaccum","text":"Chiarucci, ., Bacaro, G., Rocchini, D. & Fattorini, L. (2008). Discovering rediscovering sample-based rarefaction formula ecological literature. Commun. Ecol. 9: 121--123. Coleman, B.D, Mares, M.., Willis, M.R. & Hsieh, Y. (1982). Randomness, area species richness. Ecology 63: 1121--1133. Colwell, R.K., Chao, ., Gotelli, N.J., Lin, S.Y., Mao, C.X., Chazdon, R.L. & Longino, J.T. (2012). Models estimators linking individual-based sample-based rarefaction, extrapolation comparison assemblages. J. Plant Ecol. 5: 3--21. Dengler, J. (2009). function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity: procedures pitfalls measurement comparison species richness. Ecol. Lett. 4, 379--391.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species Accumulation Curves — specaccum","text":"Roeland Kindt r.kindt@cgiar.org Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Species Accumulation Curves — specaccum","text":"SAC method = \"exact\" developed Roeland Kindt, standard deviation Jari Oksanen (unpublished). method = \"coleman\" underestimates SAC handle properly sampling without replacement. , standard deviation take account species correlations, generally low.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species Accumulation Curves — specaccum","text":"","code":"data(BCI) sp1 <- specaccum(BCI) #> Warning: the standard deviation is zero sp2 <- specaccum(BCI, \"random\") sp2 #> Species Accumulation Curve #> Accumulation method: random, with 100 permutations #> Call: specaccum(comm = BCI, method = \"random\") #> #> #> Sites 1.00000 2.00000 3.00000 4.00000 5.00000 6.00000 7.00000 #> Richness 90.42000 121.29000 138.34000 150.46000 158.46000 165.38000 170.34000 #> sd 7.23931 7.43361 7.38305 6.80377 5.92959 5.07475 4.58196 #> #> Sites 8.00000 9.00000 10.00000 11.0000 12.00000 13.00000 14.0000 #> Richness 175.02000 178.93000 182.15000 184.8500 187.47000 189.88000 191.9000 #> sd 4.43808 4.74662 4.58671 4.4093 4.23419 4.08565 4.1451 #> #> Sites 15.00000 16.00000 17.00000 18.00000 19.00000 20.00000 21.00000 #> Richness 193.75000 195.50000 197.32000 198.94000 200.32000 201.68000 202.95000 #> sd 4.13503 4.01889 3.54447 3.53859 3.62617 3.62895 3.41528 #> #> Sites 22.00000 23.00000 24.00000 25.00000 26.00000 27.000 28.00000 #> Richness 204.17000 205.39000 206.65000 207.73000 208.91000 209.880 210.90000 #> sd 3.47009 3.28417 3.22357 3.06448 3.06856 3.217 3.14787 #> #> Sites 29.00000 30.00000 31.00000 32.00000 33.00000 34.00000 35.00000 #> Richness 211.64000 212.74000 213.57000 214.43000 215.15000 215.93000 216.75000 #> sd 3.12862 3.01719 3.01931 2.92069 2.76111 2.77163 2.69446 #> #> Sites 36.00000 37.00000 38.00000 39.00000 40.00000 41.00000 42.00000 #> Richness 217.54000 218.20000 219.00000 219.61000 220.14000 220.77000 221.28000 #> sd 2.62629 2.64384 2.45361 2.22427 2.11307 1.93769 1.82618 #> #> Sites 43.00000 44.00000 45.00000 46.00000 47.00000 48.0000 49.00000 #> Richness 221.86000 222.33000 222.85000 223.28000 223.69000 224.1400 224.60000 #> sd 1.79235 1.49784 1.45904 1.36389 1.07961 0.8764 0.58603 #> #> Sites 50 #> Richness 225 #> sd 0 summary(sp2) #> 1 sites 2 sites 3 sites 4 sites #> Min. : 77.00 Min. :105.0 Min. :119.0 Min. :132.0 #> 1st Qu.: 85.00 1st Qu.:115.8 1st Qu.:134.0 1st Qu.:146.0 #> Median : 88.00 Median :121.5 Median :139.0 Median :151.0 #> Mean : 90.42 Mean :121.3 Mean :138.3 Mean :150.5 #> 3rd Qu.: 94.25 3rd Qu.:126.0 3rd Qu.:143.2 3rd Qu.:155.0 #> Max. :109.00 Max. :144.0 Max. :155.0 Max. :166.0 #> 5 sites 6 sites 7 sites 8 sites 9 sites #> Min. :140.0 Min. :149.0 Min. :157.0 Min. :161 Min. :166.0 #> 1st Qu.:155.0 1st Qu.:162.0 1st Qu.:168.0 1st Qu.:173 1st Qu.:176.0 #> Median :159.0 Median :165.0 Median :171.0 Median :175 Median :179.0 #> Mean :158.5 Mean :165.4 Mean :170.3 Mean :175 Mean :178.9 #> 3rd Qu.:163.0 3rd Qu.:170.0 3rd Qu.:174.0 3rd Qu.:178 3rd Qu.:182.0 #> Max. :172.0 Max. :176.0 Max. :181.0 Max. :186 Max. :190.0 #> 10 sites 11 sites 12 sites 13 sites #> Min. :170.0 Min. :174.0 Min. :174.0 Min. :179.0 #> 1st Qu.:179.0 1st Qu.:182.0 1st Qu.:184.0 1st Qu.:187.0 #> Median :183.0 Median :185.0 Median :188.0 Median :190.5 #> Mean :182.2 Mean :184.8 Mean :187.5 Mean :189.9 #> 3rd Qu.:185.0 3rd Qu.:188.0 3rd Qu.:191.0 3rd Qu.:193.0 #> Max. :191.0 Max. :193.0 Max. :200.0 Max. :202.0 #> 14 sites 15 sites 16 sites 17 sites #> Min. :181.0 Min. :183.0 Min. :185.0 Min. :188.0 #> 1st Qu.:189.0 1st Qu.:191.0 1st Qu.:193.0 1st Qu.:195.0 #> Median :193.0 Median :194.0 Median :196.0 Median :197.5 #> Mean :191.9 Mean :193.8 Mean :195.5 Mean :197.3 #> 3rd Qu.:194.0 3rd Qu.:196.0 3rd Qu.:198.0 3rd Qu.:200.0 #> Max. :203.0 Max. :204.0 Max. :205.0 Max. :205.0 #> 18 sites 19 sites 20 sites 21 sites #> Min. :188.0 Min. :188.0 Min. :190.0 Min. :195.0 #> 1st Qu.:196.8 1st Qu.:198.0 1st Qu.:199.0 1st Qu.:201.0 #> Median :199.0 Median :200.0 Median :202.0 Median :203.0 #> Mean :198.9 Mean :200.3 Mean :201.7 Mean :202.9 #> 3rd Qu.:201.0 3rd Qu.:203.0 3rd Qu.:204.0 3rd Qu.:205.0 #> Max. :208.0 Max. :209.0 Max. :210.0 Max. :211.0 #> 22 sites 23 sites 24 sites 25 sites #> Min. :195.0 Min. :199.0 Min. :199.0 Min. :200.0 #> 1st Qu.:202.0 1st Qu.:203.0 1st Qu.:204.0 1st Qu.:206.0 #> Median :204.0 Median :205.0 Median :207.0 Median :207.5 #> Mean :204.2 Mean :205.4 Mean :206.7 Mean :207.7 #> 3rd Qu.:206.2 3rd Qu.:207.0 3rd Qu.:209.0 3rd Qu.:210.0 #> Max. :214.0 Max. :214.0 Max. :214.0 Max. :215.0 #> 26 sites 27 sites 28 sites 29 sites #> Min. :203.0 Min. :203.0 Min. :204.0 Min. :204.0 #> 1st Qu.:207.0 1st Qu.:208.0 1st Qu.:209.0 1st Qu.:210.0 #> Median :209.0 Median :210.0 Median :211.0 Median :212.0 #> Mean :208.9 Mean :209.9 Mean :210.9 Mean :211.6 #> 3rd Qu.:211.0 3rd Qu.:212.0 3rd Qu.:213.0 3rd Qu.:214.0 #> Max. :217.0 Max. :218.0 Max. :220.0 Max. :220.0 #> 30 sites 31 sites 32 sites 33 sites #> Min. :206.0 Min. :206.0 Min. :206.0 Min. :207.0 #> 1st Qu.:211.0 1st Qu.:212.0 1st Qu.:213.0 1st Qu.:214.0 #> Median :212.5 Median :213.0 Median :214.0 Median :215.0 #> Mean :212.7 Mean :213.6 Mean :214.4 Mean :215.2 #> 3rd Qu.:215.0 3rd Qu.:215.0 3rd Qu.:216.0 3rd Qu.:216.0 #> Max. :221.0 Max. :221.0 Max. :222.0 Max. :222.0 #> 34 sites 35 sites 36 sites 37 sites #> Min. :208.0 Min. :208.0 Min. :208.0 Min. :208.0 #> 1st Qu.:214.0 1st Qu.:215.0 1st Qu.:216.0 1st Qu.:217.0 #> Median :216.0 Median :217.0 Median :218.0 Median :218.0 #> Mean :215.9 Mean :216.8 Mean :217.5 Mean :218.2 #> 3rd Qu.:218.0 3rd Qu.:219.0 3rd Qu.:219.0 3rd Qu.:220.0 #> Max. :222.0 Max. :222.0 Max. :223.0 Max. :224.0 #> 38 sites 39 sites 40 sites 41 sites #> Min. :212.0 Min. :212.0 Min. :212.0 Min. :212.0 #> 1st Qu.:217.0 1st Qu.:218.0 1st Qu.:219.0 1st Qu.:220.0 #> Median :219.0 Median :220.0 Median :220.0 Median :221.0 #> Mean :219.0 Mean :219.6 Mean :220.1 Mean :220.8 #> 3rd Qu.:220.2 3rd Qu.:221.0 3rd Qu.:221.0 3rd Qu.:222.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 42 sites 43 sites 44 sites 45 sites #> Min. :215.0 Min. :215.0 Min. :218.0 Min. :219.0 #> 1st Qu.:220.0 1st Qu.:221.0 1st Qu.:221.0 1st Qu.:222.0 #> Median :221.0 Median :222.0 Median :222.0 Median :223.0 #> Mean :221.3 Mean :221.9 Mean :222.3 Mean :222.8 #> 3rd Qu.:222.2 3rd Qu.:223.0 3rd Qu.:223.2 3rd Qu.:224.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 46 sites 47 sites 48 sites 49 sites 50 sites #> Min. :220.0 Min. :221.0 Min. :222.0 Min. :223.0 Min. :225 #> 1st Qu.:222.8 1st Qu.:223.0 1st Qu.:224.0 1st Qu.:224.0 1st Qu.:225 #> Median :223.0 Median :224.0 Median :224.0 Median :225.0 Median :225 #> Mean :223.3 Mean :223.7 Mean :224.1 Mean :224.6 Mean :225 #> 3rd Qu.:224.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 Max. :225 plot(sp1, ci.type=\"poly\", col=\"blue\", lwd=2, ci.lty=0, ci.col=\"lightblue\") boxplot(sp2, col=\"yellow\", add=TRUE, pch=\"+\") ## Fit Lomolino model to the exact accumulation mod1 <- fitspecaccum(sp1, \"lomolino\") coef(mod1) #> Asym xmid slope #> 258.440682 2.442061 1.858694 fitted(mod1) #> [1] 94.34749 121.23271 137.45031 148.83053 157.45735 164.31866 169.95946 #> [8] 174.71115 178.78954 182.34254 185.47566 188.26658 190.77402 193.04337 #> [15] 195.11033 197.00350 198.74606 200.35705 201.85227 203.24499 204.54643 #> [22] 205.76612 206.91229 207.99203 209.01150 209.97609 210.89054 211.75903 #> [29] 212.58527 213.37256 214.12386 214.84180 215.52877 216.18692 216.81820 #> [36] 217.42437 218.00703 218.56767 219.10762 219.62811 220.13027 220.61514 #> [43] 221.08369 221.53679 221.97528 222.39991 222.81138 223.21037 223.59747 #> [50] 223.97327 plot(sp1) ## Add Lomolino model using argument 'add' plot(mod1, add = TRUE, col=2, lwd=2) ## Fit Arrhenius models to all random accumulations mods <- fitspecaccum(sp2, \"arrh\") plot(mods, col=\"hotpink\") boxplot(sp2, col = \"yellow\", border = \"blue\", lty=1, cex=0.3, add= TRUE) ## Use nls() methods to the list of models sapply(mods$models, AIC) #> [1] 329.2456 321.8614 331.1568 352.1309 354.8606 323.3767 322.0897 347.7218 #> [9] 327.6283 327.3510 335.4843 308.5300 373.4959 334.3691 320.6975 338.1084 #> [17] 330.2018 343.5560 348.5827 300.2513 315.7909 310.7837 305.5217 295.1005 #> [25] 359.2639 342.7111 339.7456 328.4175 338.4335 368.4988 354.7097 328.8704 #> [33] 342.1528 292.9875 328.3752 299.7201 336.5796 364.3153 336.4822 339.6666 #> [41] 354.7445 320.7841 304.1628 314.9640 349.6753 350.7169 307.6840 296.7971 #> [49] 359.6018 302.9863 319.6832 339.5021 328.8450 329.0553 298.0970 334.0182 #> [57] 291.9027 349.7161 325.1376 347.7797 340.6111 338.6148 352.0292 308.8538 #> [65] 315.1890 289.7947 351.6732 339.3587 320.3515 300.9377 323.3054 290.9419 #> [73] 350.8587 322.6970 340.7948 334.1965 358.2890 300.7034 338.5079 338.8941 #> [81] 341.7687 331.4553 337.5975 349.7867 331.8581 270.3068 354.5083 326.0026 #> [89] 337.8555 322.7217 302.5019 324.7469 322.1415 324.6922 324.5427 347.2050 #> [97] 309.1766 316.1194 343.9958 350.5935"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":null,"dir":"Reference","previous_headings":"","what":"Extrapolated Species Richness in a Species Pool — specpool","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions estimate extrapolated species richness species pool, number unobserved species. Function specpool based incidences sample sites, gives single estimate collection sample sites (matrix). Function estimateR based abundances (counts) single sample site.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"specpool(x, pool, smallsample = TRUE) estimateR(x, ...) specpool2vect(X, index = c(\"jack1\",\"jack2\", \"chao\", \"boot\",\"Species\")) poolaccum(x, permutations = 100, minsize = 3) estaccumR(x, permutations = 100, parallel = getOption(\"mc.cores\")) # S3 method for poolaccum summary(object, display, alpha = 0.05, ...) # S3 method for poolaccum plot(x, alpha = 0.05, type = c(\"l\",\"g\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"x Data frame matrix species data analysis result plot function. pool vector giving classification pooling sites species data. missing, sites pooled together. smallsample Use small sample correction \\((N-1)/N\\), \\(N\\) number sites within pool. X, object specpool result object. index selected index extrapolated richness. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. minsize Smallest number sampling units reported. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. display Indices displayed. alpha Level quantiles shown. proportion left outside symmetric limits. type Type graph produced xyplot. ... parameters (used).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Many species always remain unseen undetected collection sample plots. function uses popular ways estimating number unseen species adding observed species richness (Palmer 1990, Colwell & Coddington 1994). incidence-based estimates specpool use frequencies species collection sites. following, \\(S_P\\) extrapolated richness pool, \\(S_0\\) observed number species collection, \\(a_1\\) \\(a_2\\) number species occurring one two sites collection, \\(p_i\\) frequency species \\(\\), \\(N\\) number sites collection. variants extrapolated richness specpool : specpool normally uses basic Chao equation, doubletons (\\(a2=0\\)) switches bias-corrected version. case Chao equation simplifies \\(S_0 + \\frac{1}{2} a_1 (a_1-1) \\frac{N-1}{N}\\). abundance-based estimates estimateR use counts (numbers individuals) species single site. called matrix data frame, function give separate estimates site. two variants extrapolated richness estimateR bias-corrected Chao ACE (O'Hara 2005, Chiu et al. 2014). Chao estimate similar bias corrected one , \\(a_i\\) refers number species abundance \\(\\) instead number sites, small-sample correction used. ACE estimate defined : \\(a_i\\) refers number species abundance \\(\\) \\(S_{rare}\\) number rare species, \\(S_{abund}\\) number abundant species, arbitrary threshold abundance 10 rare species, \\(N_{rare}\\) number individuals rare species. Functions estimate standard errors estimates. concern number added species, assume variance observed richness. equations standard errors complicated reproduced help page, can studied R source code function discussed vignette can read browseVignettes(\"vegan\"). standard error based following sources: Chiu et al. (2014) Chao estimates Smith van Belle (1984) first-order Jackknife bootstrap (second-order jackknife still missing). variance estimator \\(S_{ace}\\) see O'Hara (2005). Functions poolaccum estaccumR similar specaccum, estimate extrapolated richness indices specpool estimateR addition number species random ordering sampling units. Function specpool uses presence data estaccumR count data. functions share summary plot methods. summary returns quantile envelopes permutations corresponding given level alpha standard deviation permutations sample size. NB., based standard deviations estimated within specpool estimateR, based permutations. plot function shows mean envelope permutations given alpha models. selection models can restricted order changes using display argument summary plot. configuration plot command, see xyplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Function specpool returns data frame entries observed richness indices class pool vector. utility function specpool2vect maps pooled values vector giving value selected index original site. Function estimateR returns estimates standard errors site. Functions poolaccum estimateR return matrices permutation results richness estimator, vector sample sizes table means permutations estimator.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Chao, . (1987). Estimating population size capture-recapture data unequal catchability. Biometrics 43, 783--791. Chiu, C.H., Wang, Y.T., Walther, B.. & Chao, . (2014). Improved nonparametric lower bound species richness via modified Good-Turing frequency formula. Biometrics 70, 671--682. Colwell, R.K. & Coddington, J.. (1994). Estimating terrestrial biodiversity extrapolation. Phil. Trans. Roy. Soc. London B 345, 101--118. O'Hara, R.B. (2005). Species richness estimators: many species can dance head pin? J. Anim. Ecol. 74, 375--386. Palmer, M.W. (1990). estimation species richness extrapolation. Ecology 71, 1195--1198. Smith, E.P & van Belle, G. (1984). Nonparametric estimation species richness. Biometrics 40, 119--129.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Bob O'Hara (estimateR) Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions based assumption species pool: community closed fixed pool size \\(S_P\\). general, functions give lower limit species richness: real richness \\(S >= S_P\\), consistent bias estimates. Even bias-correction Chao reduces bias, remove completely (Chiu et al. 2014). Optional small sample correction added specpool vegan 2.2-0. used older literature (Chao 1987), recommended recently (Chiu et al. 2014).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"data(dune) data(dune.env) pool <- with(dune.env, specpool(dune, Management)) pool #> Species chao chao.se jack1 jack1.se jack2 boot boot.se n #> BF 16 17.19048 1.5895675 19.33333 2.211083 19.83333 17.74074 1.646379 3 #> HF 21 21.51429 0.9511693 23.40000 1.876166 22.05000 22.56864 1.821518 5 #> NM 21 22.87500 2.1582871 26.00000 3.291403 25.73333 23.77696 2.300982 6 #> SF 21 29.88889 8.6447967 27.66667 3.496029 31.40000 23.99496 1.850288 6 op <- par(mfrow=c(1,2)) boxplot(specnumber(dune) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") boxplot(specnumber(dune)/specpool2vect(pool) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") par(op) data(BCI) ## Accumulation model pool <- poolaccum(BCI) summary(pool, display = \"chao\") #> $chao #> N Chao 2.5% 97.5% Std.Dev #> [1,] 3 162.3374 142.5272 186.2227 11.217082 #> [2,] 4 176.1243 156.4704 206.3243 12.158304 #> [3,] 5 183.8507 162.9868 209.2821 12.488675 #> [4,] 6 188.9018 165.6732 214.2779 12.888148 #> [5,] 7 193.8679 175.8712 216.1384 11.602711 #> [6,] 8 199.0315 180.2949 227.8918 12.604562 #> [7,] 9 202.0946 183.3317 228.8381 11.958432 #> [8,] 10 204.9843 185.3556 227.5898 12.637762 #> [9,] 11 206.6160 186.7205 234.5167 11.679759 #> [10,] 12 209.1286 189.2824 232.6359 12.351073 #> [11,] 13 211.3655 191.7918 232.0423 11.991967 #> [12,] 14 212.8128 195.8592 239.3318 11.537817 #> [13,] 15 215.5275 198.2739 239.7978 11.422201 #> [14,] 16 218.7094 198.3891 243.3826 12.165145 #> [15,] 17 221.6814 197.6029 256.5328 14.464302 #> [16,] 18 222.7418 201.4964 257.9494 13.227892 #> [17,] 19 224.8619 206.6236 251.9979 13.165360 #> [18,] 20 226.3324 210.2832 249.5616 12.858116 #> [19,] 21 228.4284 210.9920 257.9636 13.784663 #> [20,] 22 228.8425 211.1903 254.3500 11.629250 #> [21,] 23 230.6429 214.1737 251.1681 10.510649 #> [22,] 24 231.9109 213.0874 252.7553 10.925851 #> [23,] 25 232.9972 214.3610 258.4917 11.207306 #> [24,] 26 233.8067 216.9697 263.2733 11.546339 #> [25,] 27 235.6347 217.3831 266.9230 12.800245 #> [26,] 28 235.8942 218.6913 261.4115 11.761873 #> [27,] 29 235.9612 218.6591 258.5682 11.133700 #> [28,] 30 236.6706 219.4315 259.9315 10.291499 #> [29,] 31 236.8530 220.2338 259.2903 9.810390 #> [30,] 32 237.6018 221.4868 262.0864 10.195825 #> [31,] 33 237.4291 222.2990 256.2778 9.227891 #> [32,] 34 237.6712 222.7844 258.2268 8.979536 #> [33,] 35 237.7856 222.2750 257.2540 8.537867 #> [34,] 36 238.4325 223.6473 255.5610 8.566487 #> [35,] 37 238.1834 224.9892 253.5811 7.776992 #> [36,] 38 238.4685 225.2802 254.0155 7.626532 #> [37,] 39 238.2082 226.0390 252.7975 7.642443 #> [38,] 40 238.7991 227.1971 254.1619 7.155802 #> [39,] 41 239.2246 227.3368 258.5178 7.926574 #> [40,] 42 238.4935 228.5195 256.2137 7.287824 #> [41,] 43 238.3856 227.8327 255.8470 7.300153 #> [42,] 44 238.0092 228.5279 252.9488 6.520961 #> [43,] 45 238.1537 228.7980 253.3757 6.443894 #> [44,] 46 238.0532 230.5990 250.9539 5.466587 #> [45,] 47 237.7542 231.3089 251.0192 4.821564 #> [46,] 48 237.2842 231.9015 248.6399 4.038862 #> [47,] 49 236.8407 233.3115 245.4082 2.803888 #> [48,] 50 236.3732 236.3732 236.3732 0.000000 #> #> attr(,\"class\") #> [1] \"summary.poolaccum\" plot(pool) ## Quantitative model estimateR(BCI[1:5,]) #> 1 2 3 4 5 #> S.obs 93.000000 84.000000 90.000000 94.000000 101.000000 #> S.chao1 117.473684 117.214286 141.230769 111.550000 136.000000 #> se.chao1 11.583785 15.918953 23.001405 8.919663 15.467344 #> S.ACE 122.848959 117.317307 134.669844 118.729941 137.114088 #> se.ACE 5.736054 5.571998 6.191618 5.367571 5.848474"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distance-based ordination (dbrda, capscale, metaMDS) information species, methods may add species scores community data available. However, species scores may missing ( always dbrda), may close relation used dissimilarity index. function add species scores replace existing species scores distance-based methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"sppscores(object) <- value"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"object Ordination result. value Community data find species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distances information species (columns, variables), hence distance-based ordination information species scores. However, species scores can added supplementary information analysis help interpretation results. ordination methods (capscale, metaMDS) can supplement species scores analysis community data available analysis. capscale species scores found projecting community data site ordination (linear combination scores), scores accurate analysis used Euclidean distances. dissimilarity index can expressed Euclidean distances transformed data (instance, Chord Hellinger Distances), species scores based transformed data accurate, function still finds dissimilarities untransformed data. Usually community dissimilarities differ two significant ways Euclidean distances: bound maximum 1, use absolute differences instead squared differences. cases, may better use species scores transformed Euclidean distances good linear relation used dissimilarities. often useful standardize data row unit total, perform squareroot transformation damp effect squared differences (see Examples). Function dbrda never finds species scores, mathematically similar capscale, similar rules followed supplementing species scores. Function metaMDS uses weighted averages (wascores) find species scores. better relationship dissimilarities projection scores used metric ordination, similar transformation community data used dissimilarities species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Replacement function adds species scores replaces old scores ordination object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"data(BCI, BCI.env) mod <- dbrda(vegdist(BCI) ~ Habitat, BCI.env) ## add species scores sppscores(mod) <- BCI ## Euclidean distances of BCI differ from used dissimilarity plot(vegdist(BCI), dist(BCI)) ## more linear relationship plot(vegdist(BCI), dist(sqrt(decostand(BCI, \"total\")))) ## better species scores sppscores(mod) <- sqrt(decostand(BCI, \"total\"))"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":null,"dir":"Reference","previous_headings":"","what":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function stepacross tries replace dissimilarities shortest paths stepping across intermediate sites regarding dissimilarities threshold missing data (NA). path = \"shortest\" flexible shortest path (Williamson 1978, Bradfield & Kenkel 1987), path = \"extended\" approximation known extended dissimilarities (De'ath 1999). use stepacross improve ordination high beta diversity, many sites species common.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"stepacross(dis, path = \"shortest\", toolong = 1, trace = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. path method stepping across (partial match) Alternative \"shortest\" finds shortest paths, \"extended\" approximation known extended dissimilarities. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . trace Trace calculations. ... parameters (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Williamson (1978) suggested using flexible shortest paths estimate dissimilarities sites nothing common, shared species. path = \"shortest\" function stepacross replaces dissimilarities toolong longer NA, tries find shortest paths sites using remaining dissimilarities. Several dissimilarity indices semi-metric means obey triangle inequality \\(d_{ij} \\leq d_{ik} + d_{kj}\\), shortest path algorithm can replace dissimilarities well, even shorter toolong. De'ath (1999) suggested simplified method known extended dissimilarities, calculated path = \"extended\". method, dissimilarities toolong longer first made NA, function tries replace NA dissimilarities path single stepping stone points. NA replaced one pass, function make new passes updated dissimilarities long NA replaced extended dissimilarities. mean second passes, remaining NA dissimilarities allowed one stepping stone site, previously replaced dissimilarities updated. , function consider dissimilarities shorter toolong, although replaced shorter path semi-metric indices, used part paths. optimal cases, extended dissimilarities equal shortest paths, may longer. alternative defining long dissimilarities parameter toolong, input dissimilarities can contain NAs. toolong zero negative, function make dissimilarities NA. NAs input toolong = 0, path = \"shortest\" find shorter paths semi-metric indices, path = \"extended\" nothing. Function .shared can used set dissimilarities NA. data disconnected path points, result contain NAs warning issued. Several methods handle NA dissimilarities, warning taken seriously. Function distconnected can used find connected groups remove rare outlier observations groups observations. Alternative path = \"shortest\" uses Dijkstra's method finding flexible shortest paths, implemented priority-first search dense graphs (Sedgewick 1990). Alternative path = \"extended\" follows De'ath (1999), implementation simpler code.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function returns object class dist extended dissimilarities (see functions vegdist dist). value path appended method attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Bradfield, G.E. & Kenkel, N.C. (1987). Nonlinear ordination using flexible shortest path adjustment ecological distances. Ecology 68, 750--753. De'ath, G. (1999). Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecol. 144, 191--199. Sedgewick, R. (1990). Algorithms C. Addison Wesley. Williamson, M.H. (1978). ordination incidence data. J. Ecol. 66, 911-920.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"function changes original dissimilarities, like . may best use function really must: extremely high beta diversity large proportion dissimilarities upper limit (species common). Semi-metric indices vary degree violating triangle inequality. Morisita Horn--Morisita indices vegdist may strongly semi-metric, shortest paths can change indices much. Mountford index violates basic rules dissimilarities: non-identical sites zero dissimilarity species composition poorer site subset richer. Mountford index, can find three sites \\(, j, k\\) \\(d_{ik} = 0\\) \\(d_{jk} = 0\\), \\(d_{ij} > 0\\). results stepacross Mountford index can weird. stepacross needed, best try use metric indices .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"# There are no data sets with high beta diversity in vegan, but this # should give an idea. data(dune) dis <- vegdist(dune) edis <- stepacross(dis) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities... plot(edis, dis, xlab = \"Shortest path\", ylab = \"Original\") ## Manhattan distance have no fixed upper limit. dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune) dis <- stepacross(dis, toolong=0) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities..."},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions plot ordination distances given number dimensions observed distances distances full space eigenvector methods. display similar Shepard diagram (stressplot non-metric multidimensional scaling metaMDS monoMDS), shows linear relationship eigenvector ordinations. stressplot methods available wcmdscale, rda, cca, capscale, dbrda, prcomp princomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"# S3 method for wcmdscale stressplot(object, k = 2, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"object Result object eigenvector ordination (wcmdscale, rda, cca, dbrda, capscale) k Number dimensions ordination distances displayed. pch, p.col, l.col, lwd Plotting character, point colour line colour like default stressplot ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"functions offer similar display eigenvector ordinations standard Shepard diagram (stressplot) non-metric multidimensional scaling. ordination distances given number dimensions plotted observed distances. metric distances, ordination distances full space (ordination axes) equal observed distances, fit line shows equality. general, fit line go points, points observed distances approach fit line . However, non-Euclidean distances ( wcmdscale, dbrda capscale) negative eigenvalues ordination distances can exceed observed distances real dimensions; imaginary dimensions negative eigenvalues correct excess distances. used dbrda, capscale wcmdscale argument add avoid negative eigenvalues, ordination distances exceed observed dissimilarities. partial ordination (cca, rda, capscale Condition formula), distances partial component included observed distances ordination distances. k=0, ordination distances refer partial ordination. exception dbrda distances partial, constrained residual components additive, first components can shown, partial models shown .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions draw graph return invisibly ordination distances ordination distances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"data(dune, dune.env) mod <- rda(dune) stressplot(mod) mod <- rda(dune ~ Management, dune.env) stressplot(mod, k=3)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of Taxonomic Diversity and Distinctness — taxondive","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function finds indices taxonomic diversity distinctness, averaged taxonomic distances among species individuals community (Clarke & Warwick 1998, 2001)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"taxondive(comm, dis, match.force = FALSE) taxa2dist(x, varstep = FALSE, check = TRUE, labels)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"comm Community data. dis Taxonomic distances among taxa comm. dist object symmetric square matrix. match.force Force matching column names comm labels dis. FALSE, matching happens dimensions differ, case species must identical order . x Classification table row species basic taxon, columns identifiers classification higher levels. varstep Vary step lengths successive levels relative proportional loss number distinct classes. check TRUE, remove redundant levels different rows constant rows regard row different basal taxon (species). FALSE levels retained basal taxa (species) also must coded variables (columns). get warning species coded, can ignore intention. labels labels attribute taxonomic distances. Row names used given. Species matched labels comm dis taxondive different dimensions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke & Warwick (1998, 2001) suggested several alternative indices taxonomic diversity distinctness. Two basic indices called taxonomic diversity (\\(\\Delta\\)) distinctness (\\(\\Delta^*\\)): equations give index value single site, summation goes species \\(\\) \\(j\\). \\(\\omega\\) taxonomic distances among taxa, \\(x\\) species abundances, \\(n\\) total abundance site. presence/absence data indices reduce index \\(\\Delta^+\\), index Clarke & Warwick (1998) also estimate standard deviation. Clarke & Warwick (2001) presented two new indices: \\(s\\Delta^+\\) product species richness \\(\\Delta^+\\), index variation taxonomic distinctness (\\(\\Lambda^+\\)) defined dis argument must species dissimilarities. must similar dissimilarities produced dist. customary integer steps taxonomic hierarchies, kind dissimilarities can used, phylogenetic trees genetic differences. , dis need taxonomic, species classifications can used. Function taxa2dist can produce suitable dist object classification table. species (basic taxon) corresponds row classification table, columns give classification different levels. varstep = FALSE successive levels separated equal steps, varstep = TRUE step length relative proportional decrease number classes (Clarke & Warwick 1999). check = TRUE, function removes classes distinct species combine species one class, assumes row presents distinct basic taxon. function scales distances longest path length taxa 100 (necessarily check = FALSE). Function plot.taxondive plots \\(\\Delta^+\\) Number species, together expectation approximate 2*sd limits. Function summary.taxondive finds \\(z\\) values significances Normal distribution \\(\\Delta^+\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function returns object class taxondive following items: Species Number species site. D, Dstar, Dplus, SDplus, Lambda \\(\\Delta\\), \\(\\Delta^*\\), \\(\\Delta^+\\), \\(s\\Delta^+\\) \\(\\Lambda^+\\) site. sd.Dplus Standard deviation \\(\\Delta^+\\). ED, EDstar, EDplus Expected values corresponding statistics. Function taxa2dist returns object class \"dist\", attribute \"steps\" step lengths successive levels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke, K.R & Warwick, R.M. (1998) taxonomic distinctness index statistical properties. Journal Applied Ecology 35, 523--531. Clarke, K.R. & Warwick, R.M. (1999) taxonomic distinctness measure biodiversity: weighting step lengths hierarchical levels. Marine Ecology Progress Series 184: 21--29. Clarke, K.R. & Warwick, R.M. (2001) biodiversity index applicable species lists: variation taxonomic distinctness. Marine Ecology Progress Series 216, 265--278.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"function still preliminary may change. scaling taxonomic dissimilarities influences results. multiply taxonomic distances (step lengths) constant, values Deltas multiplied constant, value \\(\\Lambda^+\\) square constant.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"## Preliminary: needs better data and some support functions data(dune) data(dune.taxon) # Taxonomic distances from a classification table with variable step lengths. taxdis <- taxa2dist(dune.taxon, varstep=TRUE) plot(hclust(taxdis), hang = -1) # Indices mod <- taxondive(dune, taxdis) mod #> Species Delta Delta* Lambda+ Delta+ S Delta+ #> 1 5.000 22.736 29.232 900.298 43.364 216.82 #> 2 10.000 51.046 55.988 822.191 56.232 562.32 #> 3 10.000 41.633 46.194 1025.471 62.869 628.69 #> 4 13.000 50.795 55.140 888.244 64.837 842.88 #> 5 14.000 63.498 67.856 715.393 69.211 968.95 #> 6 11.000 70.201 76.361 628.743 73.281 806.09 #> 7 13.000 61.605 66.187 679.337 69.918 908.94 #> 8 12.000 52.544 56.374 756.375 66.729 800.74 #> 9 13.000 50.526 54.108 849.448 63.205 821.67 #> 10 12.000 60.068 64.960 730.736 69.291 831.49 #> 11 9.000 69.589 77.740 404.609 77.803 700.23 #> 12 9.000 62.405 69.795 552.129 74.470 670.23 #> 13 10.000 47.316 53.842 536.429 66.657 666.57 #> 14 7.000 71.383 82.091 239.543 82.013 574.09 #> 15 8.000 68.564 77.097 334.889 79.010 632.08 #> 16 8.000 55.984 64.400 978.014 69.708 557.66 #> 17 7.000 53.913 60.222 632.990 59.286 415.00 #> 18 9.000 73.235 81.865 438.355 76.288 686.59 #> 19 9.000 68.727 76.091 336.364 78.636 707.73 #> 20 8.000 72.343 80.670 444.915 82.078 656.62 #> Expected 65.330 62.560 71.031 summary(mod) #> Delta Delta* Delta+ sd(Delta+) z(Delta+) Pr(>|z|) #> 1 22.7362 29.2322 43.3636 10.0499 -2.7530 0.005905 ** #> 2 51.0458 55.9878 56.2323 5.7727 -2.5636 0.010359 * #> 3 41.6334 46.1936 62.8687 5.7727 -1.4140 0.157360 #> 4 50.7952 55.1396 64.8368 4.5677 -1.3562 0.175047 #> 5 63.4979 67.8564 69.2108 4.2482 -0.4285 0.668251 #> 6 70.2011 76.3614 73.2810 5.3189 0.4229 0.672332 #> 7 61.6049 66.1871 69.9184 4.5677 -0.2437 0.807499 #> 8 52.5437 56.3743 66.7287 4.9215 -0.8743 0.381975 #> 9 50.5258 54.1079 63.2051 4.5677 -1.7134 0.086640 . #> 10 60.0680 64.9597 69.2906 4.9215 -0.3537 0.723567 #> 11 69.5894 77.7396 77.8030 6.3011 1.0747 0.282517 #> 12 62.4049 69.7949 74.4697 6.3011 0.5457 0.585290 #> 13 47.3158 53.8421 66.6566 5.7727 -0.7578 0.448548 #> 14 71.3834 82.0909 82.0130 7.7069 1.4249 0.154186 #> 15 68.5645 77.0970 79.0097 6.9314 1.1510 0.249714 #> 16 55.9840 64.3999 69.7078 6.9314 -0.1909 0.848565 #> 17 53.9134 60.2224 59.2857 7.7069 -1.5240 0.127501 #> 18 73.2349 81.8645 76.2879 6.3011 0.8342 0.404155 #> 19 68.7273 76.0909 78.6364 6.3011 1.2069 0.227457 #> 20 72.3431 80.6704 82.0779 6.9314 1.5937 0.111005 #> Expected 65.3302 62.5603 71.0313 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":null,"dir":"Reference","previous_headings":"","what":"Species tolerances and sample heterogeneities — tolerance","title":"Species tolerances and sample heterogeneities — tolerance","text":"Species tolerances sample heterogeneities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"tolerance(x, ...) # S3 method for cca tolerance(x, choices = 1:2, which = c(\"species\",\"sites\"), scaling = \"species\", useN2 = TRUE, hill = FALSE, ...) # S3 method for decorana tolerance(x, data, choices = 1:4, which = c(\"sites\", \"species\"), useN2 = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species tolerances and sample heterogeneities — tolerance","text":"Function compute species tolerances site heterogeneity measures unimodal ordinations (CCA & CA). Implements Eq 6.47 6.48 Canoco 4.5 Reference Manual (pages 178--179).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species tolerances and sample heterogeneities — tolerance","text":"Matrix tolerances/heterogeneities additional attributes: , scaling, N2, latter NA useN2 = FALSE N2 estimated.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species tolerances and sample heterogeneities — tolerance","text":"Gavin L. Simpson Jari Oksanen (decorana method).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species tolerances and sample heterogeneities — tolerance","text":"x object class \"cca\". choices numeric; ordination axes compute tolerances heterogeneities . Defaults axes 1 2. character; one \"species\" \"sites\", indicating whether species tolerances sample heterogeneities respectively computed. scaling character numeric; ordination scaling use. See scores.cca details. hill logical; scaling character, control whether Hill's scaling used (C)CA respectively. See scores.cca details. useN2 logical; bias tolerances / heterogeneities reduced via scaling Hill's N2? data Original input data used decorana. missing, function tries get data used decorana call. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ ., data = dune.env) ## defaults to species tolerances tolerance(mod) #> #> Species Tolerance #> #> Scaling: 2 #> #> CCA1 CCA2 #> Achimill 0.32968099 0.9241988 #> Agrostol 0.93670069 0.9238455 #> Airaprae 1.04694096 0.5889849 #> Alopgeni 0.72227472 0.3760138 #> Anthodor 1.00596787 0.8338212 #> Bellpere 0.32891011 0.9962790 #> Bromhord 0.27740999 0.6236199 #> Chenalbu 0.00000000 0.0000000 #> Cirsarve 0.00000000 0.0000000 #> Comapalu 0.47185632 0.8029414 #> Eleopalu 0.50344134 0.9384960 #> Elymrepe 0.35119963 0.5642491 #> Empenigr 0.00000000 0.0000000 #> Hyporadi 1.05840696 0.7523003 #> Juncarti 0.78397702 1.0686743 #> Juncbufo 0.69275956 0.6180830 #> Lolipere 0.51006235 0.8278177 #> Planlanc 0.36040676 0.6962294 #> Poaprat 0.58184277 0.9547104 #> Poatriv 0.78695928 0.7433503 #> Ranuflam 0.56576326 1.1725628 #> Rumeacet 0.58715663 0.8751491 #> Sagiproc 0.70922180 1.1153129 #> Salirepe 0.98530179 0.1077917 #> Scorautu 1.04355761 1.0724439 #> Trifprat 0.03045846 0.3651949 #> Trifrepe 1.21543364 0.9115613 #> Vicilath 0.24853962 0.6194084 #> Bracruta 1.03787313 1.0958331 #> Callcusp 0.57882025 1.0418623 #> ## sample heterogeneities for CCA axes 1:6 tolerance(mod, which = \"sites\", choices = 1:6) #> #> Sample Heterogeneity #> #> Scaling: 2 #> #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 1 0.2350112 0.8611530 1.7964571 0.4445499 2.4235732 0.5496289 #> 2 0.7100754 0.4136311 0.8151643 0.6311751 1.0467901 0.2514646 #> 3 0.5076492 0.7279717 0.8306874 0.5590739 0.3904998 0.9162012 #> 4 0.5955037 0.6901907 0.7931255 0.4873638 0.3966068 0.8700581 #> 5 0.6001048 0.5614830 1.1481560 0.3569604 0.4423909 1.9420043 #> 6 0.7272637 0.6867342 1.6068628 0.7778498 0.9187843 0.4938865 #> 7 0.6478967 0.4993262 0.7207318 0.3817131 0.4130713 0.7228173 #> 8 0.8563491 0.5498552 0.4217718 0.3370226 0.3013276 0.9535190 #> 9 0.5599722 0.7399384 0.4170304 1.0535541 1.4612437 0.7626183 #> 10 0.5210280 0.5806978 0.5856634 0.4174860 1.8559344 0.8890262 #> 11 0.4489323 0.6016877 0.3317371 1.8780211 1.2965939 2.1953737 #> 12 0.4948094 1.1084494 0.5226746 1.5064446 0.5703077 1.1561020 #> 13 0.6998985 0.8859365 0.4215474 0.8582272 0.5673698 0.5186678 #> 14 1.5925779 0.6747926 0.8927360 1.6798300 0.3480218 0.1575892 #> 15 1.0107648 0.5294221 1.0975629 1.7632888 0.2240900 0.3727240 #> 16 0.8031479 0.6058313 0.4871527 0.4227451 0.5341256 0.6990815 #> 17 0.5936276 1.5142792 0.5137979 1.0224938 1.7931775 0.6261853 #> 18 0.5689409 1.4067575 0.6398557 0.4983399 0.4364791 0.6590394 #> 19 1.1330387 0.9816332 1.1242398 0.7238920 0.5577662 0.7036044 #> 20 0.6737757 1.4458326 1.4380928 1.0959027 0.4142423 0.5332460 #> ## average should be 1 with scaling = \"sites\", hill = TRUE tol <- tolerance(mod, which = \"sites\", scaling = \"sites\", hill = TRUE, choices = 1:4) colMeans(tol) #> CCA1 CCA2 CCA3 CCA4 #> 1.059199 1.048823 1.000551 1.077612 apply(tol, 2, sd) #> CCA1 CCA2 CCA3 CCA4 #> 0.3174462 0.2793521 0.3714540 0.2681931 ## Rescaling tries to set all tolerances to 1 tol <- tolerance(decorana(dune)) colMeans(tol) #> DCA1 DCA2 DCA3 DCA4 #> 0.9817657 0.9249544 0.9444811 0.9821666 apply(tol, 2, sd) #> DCA1 DCA2 DCA3 DCA4 #> 0.1977777 0.3204058 0.2646872 0.1210543"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":null,"dir":"Reference","previous_headings":"","what":"Functional Diversity and Community Distances from Species Trees — treedive","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Functional diversity defined total branch length trait dendrogram connecting species, excluding unnecessary root segments tree (Petchey Gaston 2006). Tree distance increase total branch length combining two sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"treedive(comm, tree, match.force = TRUE, verbose = TRUE) treeheight(tree) treedist(x, tree, relative = TRUE, match.force = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"comm, x Community data frame matrix. tree dendrogram treedive must species (columns). match.force Force matching column names data (comm, x) labels tree. FALSE, matching happens dimensions differ (warning message). order data must match order tree matching names done. verbose Print diagnostic messages warnings. relative Use distances relative height combined tree. ... arguments passed functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Function treeheight finds sum lengths connecting segments dendrogram produced hclust, dendrogram can coerced correct type using .hclust. applied clustering species traits, measure functional diversity (Petchey Gaston 2002, 2006), applied phylogenetic trees phylogenetic diversity. Function treedive finds treeheight site (row) community matrix. function uses subset dendrogram species occur site, excludes tree root needed connect species (Petchey Gaston 2006). subset dendrogram found first calculating cophenetic distances input dendrogram, reconstructing dendrogram subset cophenetic distance matrix species occurring site. Diversity 0 one species, NA empty communities. Function treedist finds dissimilarities among trees. Pairwise dissimilarity two trees found combining species common tree seeing much tree height shared much unique. relative = FALSE dissimilarity defined \\(2 (\\cup B) - - B\\), \\(\\) \\(B\\) heights component trees \\(\\cup B\\) height combined tree. relative = TRUE dissimilarity \\((2(\\cup B)--B)/(\\cup B)\\). Although latter formula similar Jaccard dissimilarity (see vegdist, designdist), range \\(0 \\ldots 1\\), since combined tree can add new root. two zero-height trees combined tree zero height, relative index attains maximum value \\(2\\). dissimilarity zero combined zero-height tree. functions need dendrogram species traits phylogenies input. species traits contain factor ordered factor variables, recommended use Gower distances mixed data (function daisy package cluster), usually recommended clustering method UPGMA (method = \"average\" function hclust) (Podani Schmera 2006). Phylogenetic trees can changed dendrograms using function .hclust.phylo ape package. possible analyse non-randomness tree diversity using oecosimu. needs specifying adequate Null model, results change choice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"vector diversity values single tree height, dissimilarity structure inherits dist can used similarly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Lozupone, C. Knight, R. 2005. UniFrac: new phylogenetic method comparing microbial communities. Applied Environmental Microbiology 71, 8228--8235. Petchey, O.L. Gaston, K.J. 2002. Functional diversity (FD), species richness community composition. Ecology Letters 5, 402--411. Petchey, O.L. Gaston, K.J. 2006. Functional diversity: back basics looking forward. Ecology Letters 9, 741--758. Podani J. Schmera, D. 2006. dendrogram-based methods functional diversity. Oikos 115, 179--185.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"## There is no data set on species properties yet, and we demonstrate ## the methods using phylogenetic trees data(dune) data(dune.phylodis) cl <- hclust(dune.phylodis) treedive(dune, cl) #> forced matching of 'tree' labels and 'comm' names #> 1 2 3 4 5 6 7 8 #> 384.0913 568.8791 1172.9455 1327.9317 1426.9067 1391.1628 1479.5062 1523.0792 #> 9 10 11 12 13 14 15 16 #> 1460.0423 1316.4832 1366.9960 1423.5582 895.1120 1457.2705 1505.9501 1187.5165 #> 17 18 19 20 #> 517.6920 1394.5162 1470.4671 1439.5571 ## Significance test using Null model communities. ## The current choice fixes numbers of species and picks species ## proportionally to their overall frequency oecosimu(dune, treedive, \"r1\", tree = cl, verbose = FALSE) #> Warning: nullmodel transformed 'comm' to binary data #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = treedive, method = \"r1\", tree = #> cl, verbose = FALSE) #> #> nullmodel method ‘r1’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> 1 384.09 -1.238698 773.72 383.47 628.63 1237.7 0.11 #> 2 568.88 -2.353877 1222.61 663.93 1337.50 1561.4 0.01 ** #> 3 1172.95 -0.136353 1210.76 679.53 1327.94 1581.5 0.63 #> 4 1327.93 -0.418624 1427.25 886.63 1505.15 1733.1 0.45 #> 5 1426.91 -0.322462 1496.97 926.98 1559.25 1734.4 0.43 #> 6 1391.16 0.254066 1326.70 753.19 1408.30 1634.7 0.87 #> 7 1479.51 0.280433 1407.90 880.79 1476.23 1735.1 0.99 #> 8 1523.08 0.567886 1389.63 792.74 1428.01 1655.3 0.65 #> 9 1460.04 0.065245 1448.21 899.67 1472.92 1654.4 0.87 #> 10 1316.48 -0.192836 1362.43 708.60 1441.63 1608.0 0.47 #> 11 1367.00 0.768160 1150.05 621.65 1266.73 1486.4 0.43 #> 12 1423.56 1.101039 1088.13 622.24 1208.95 1474.2 0.25 #> 13 895.11 -0.972053 1188.10 656.34 1325.53 1558.5 0.61 #> 14 1457.27 1.477151 980.91 491.11 1101.65 1459.3 0.07 . #> 15 1505.95 1.532058 1052.35 575.91 1171.88 1460.7 0.03 * #> 16 1187.52 0.306608 1089.58 543.75 1238.26 1490.7 0.89 #> 17 517.69 -1.446211 959.50 504.26 1101.16 1367.8 0.09 . #> 18 1394.52 0.961778 1099.50 618.22 1249.49 1493.8 0.33 #> 19 1470.47 1.134795 1121.18 564.70 1240.26 1562.8 0.15 #> 20 1439.56 1.160566 1104.59 629.08 1244.20 1467.4 0.17 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Phylogenetically ordered community table dtree <- treedist(dune, cl) tabasco(dune, hclust(dtree), cl) ## Use tree distances in distance-based RDA dbrda(dtree ~ 1) #> Call: dbrda(formula = dtree ~ 1) #> #> Inertia Rank RealDims #> Total 2.183 #> Unconstrained 2.183 19 10 #> Inertia is squared Treedist distance #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.1971 0.4546 0.2967 0.1346 0.1067 0.0912 0.0391 0.0190 #> (Showing 8 of 19 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":null,"dir":"Reference","previous_headings":"","what":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis find Tsallis diversities scale corresponding evenness measures. Function tsallisaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"tsallis(x, scales = seq(0, 2, 0.2), norm = FALSE, hill = FALSE) tsallisaccum(x, scales = seq(0, 2, 0.2), permutations = 100, raw = FALSE, subset, ...) # S3 method for tsallisaccum persp(x, theta = 220, phi = 15, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"x Community data matrix plotting object. scales Scales Tsallis diversity. norm Logical, TRUE diversity values normalized maximum (diversity value equiprobability conditions). hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. subset logical expression indicating sites (rows) keep: missing values taken FALSE. theta, phi angles defining viewing direction. theta gives azimuthal direction phi colatitude. col Colours used surface. zlim Limits vertical axis. ... arguments passed tsallis graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis diversity (also equivalent Patil Taillie diversity) one-parametric generalised entropy function, defined : $$H_q = \\frac{1}{q-1} (1-\\sum_{=1}^S p_i^q)$$ \\(q\\) scale parameter, \\(S\\) number species sample (Tsallis 1988, Tothmeresz 1995). diversity concave \\(q>0\\), non-additive (Keylock 2005). \\(q=0\\) gives number species minus one, \\(q\\) tends 1 gives Shannon diversity, \\(q=2\\) gives Simpson index (see function diversity). norm = TRUE, tsallis gives values normalized maximum: $$H_q(max) = \\frac{S^{1-q}-1}{1-q}$$ \\(S\\) number species. \\(q\\) tends 1, maximum defined \\(ln(S)\\). hill = TRUE, tsallis gives Hill numbers (numbers equivalents, see Jost 2007): $$D_q = (1-(q-1) H)^{1/(1-q)}$$ Details plotting methods accumulating values can found help pages functions renyi renyiaccum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis returns data frame selected indices. Function tsallisaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487. Tothmeresz, B. (1995) Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290. Patil, G. P. Taillie, C. (1982) Diversity concept measurement. J. . Stat. Ass. 77, 548--567. Keylock, C. J. (2005) Simpson diversity Shannon-Wiener index special cases generalized entropy. Oikos 109, 203--207. Jost, L (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Péter Sólymos, solymos@ualberta.ca, based code Roeland Kindt Jari Oksanen written renyi","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) x1 <- tsallis(BCI[i,]) x1 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 #> 8 87 39.98200 19.77567 10.60391 6.181934 3.908381 2.659907 1.928610 1.473408 #> 10 93 41.50369 20.10059 10.63988 6.164157 3.889803 2.648176 1.922207 1.470121 #> 5 100 44.20598 21.17686 11.08126 6.350040 3.969940 2.683412 1.937940 1.477217 #> 17 92 40.88788 19.62643 10.29183 5.926929 3.736897 2.553271 1.864739 1.435868 #> 19 108 47.27018 22.32602 11.49839 6.492802 4.013094 2.692466 1.936696 1.473828 #> 13 92 41.98171 20.60707 10.96297 6.342501 3.982373 2.694847 1.945425 1.481616 #> 46 85 38.71044 19.06049 10.22455 5.987209 3.810489 2.611431 1.904913 1.461976 #> 16 92 41.45601 20.19256 10.71573 6.210870 3.916821 2.663870 1.931567 1.475861 #> 36 91 40.76595 19.77916 10.48193 6.081803 3.846109 2.625076 1.910150 1.463936 #> 7 81 37.47840 18.73414 10.17269 6.004908 3.836811 2.631757 1.918091 1.469869 #> 24 94 42.55654 20.75558 10.99284 6.343926 3.979427 2.692702 1.944478 1.481404 #> 22 90 40.16368 19.41416 10.25630 5.939677 3.755413 2.566788 1.872586 1.439722 #> 1.8 2 #> 8 1.173986 0.9671998 #> 10 1.172341 0.9663808 #> 5 1.175553 0.9678267 #> 17 1.152122 0.9545126 #> 19 1.172481 0.9655820 #> 13 1.178033 0.9692075 #> 46 1.168556 0.9646728 #> 16 1.175935 0.9686598 #> 36 1.169232 0.9648567 #> 7 1.173083 0.9672014 #> 24 1.178152 0.9694268 #> 22 1.153640 0.9548316 diversity(BCI[i,],\"simpson\") == x1[[\"2\"]] #> 8 10 5 17 19 13 46 16 36 7 24 22 #> TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE plot(x1) x2 <- tsallis(BCI[i,],norm=TRUE) x2 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 #> 8 1 0.9154260 0.8674322 0.8491531 0.8535754 0.8729254 0.8992501 0.9258854 #> 10 1 0.9001004 0.8450759 0.8255446 0.8324355 0.8561634 0.8872568 0.9180261 #> 5 1 0.9038026 0.8502581 0.8308757 0.8372445 0.8602030 0.8904867 0.9204832 #> 17 1 0.8945733 0.8308281 0.8026305 0.8032747 0.8244489 0.8566981 0.8913214 #> 19 1 0.9079031 0.8537683 0.8315799 0.8347890 0.8554245 0.8846675 0.9147428 #> 13 1 0.9185048 0.8723405 0.8549707 0.8595970 0.8786069 0.9042012 0.9298883 #> 46 1 0.9032439 0.8485357 0.8278693 0.8331492 0.8554539 0.8856795 0.9162035 #> 16 1 0.9070033 0.8547937 0.8356888 0.8417570 0.8641446 0.8938075 0.9232641 #> 36 1 0.8998669 0.8431295 0.8216927 0.8272622 0.8505725 0.8820846 0.9137994 #> 7 1 0.9094899 0.8600339 0.8427936 0.8492653 0.8706729 0.8985570 0.9261475 #> 24 1 0.9149445 0.8667059 0.8486326 0.8536824 0.8738548 0.9008891 0.9279024 #> 22 1 0.8945746 0.8334045 0.8082215 0.8109045 0.8325271 0.8637842 0.8965990 #> 1.6 1.8 2 #> 8 0.9486740 0.9660687 0.9783170 #> 10 0.9438798 0.9632975 0.9767719 #> 5 0.9456409 0.9644767 0.9775050 #> 17 0.9223038 0.9469046 0.9648877 #> 19 0.9406572 0.9605041 0.9745226 #> 13 0.9516896 0.9681999 0.9797424 #> 46 0.9422698 0.9621112 0.9760219 #> 16 0.9479928 0.9664763 0.9791887 #> 36 0.9407648 0.9611954 0.9754595 #> 7 0.9493990 0.9669335 0.9791421 #> 24 0.9507028 0.9678509 0.9797399 #> 22 0.9256369 0.9486075 0.9654409 plot(x2) mod1 <- tsallisaccum(BCI[i,]) plot(mod1, as.table=TRUE, col = c(1, 2, 2)) persp(mod1) mod2 <- tsallisaccum(BCI[i,], norm=TRUE) persp(mod2,theta=100,phi=30)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and environment in lichen pastures — varespec","title":"Vegetation and environment in lichen pastures — varespec","text":"varespec data frame 24 rows 44 columns. Columns estimated cover values 44 species. variable names formed scientific names, self explanatory anybody familiar vegetation type. varechem data frame 24 rows 14 columns, giving soil characteristics sites varespec data frame. chemical measurements obvious names. Baresoil gives estimated cover bare soil, Humdepth thickness humus layer.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varechem) data(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Vegetation and environment in lichen pastures — varespec","text":"Väre, H., Ohtonen, R. Oksanen, J. (1995) Effects reindeer grazing understorey vegetation dry Pinus sylvestris forests. Journal Vegetation Science 6, 523--530.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varespec) data(varechem)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":null,"dir":"Reference","previous_headings":"","what":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"function partitions variation community data community dissimilarities respect two, three, four explanatory tables, using adjusted \\(R^2\\) redundancy analysis ordination (RDA) distance-based redundancy analysis. response single vector, partitioning partial regression. Collinear variables explanatory tables removed prior partitioning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"varpart(Y, X, ..., data, chisquare = FALSE, transfo, scale = FALSE, add = FALSE, sqrt.dist = FALSE, permutations) # S3 method for varpart summary(object, ...) showvarparts(parts, labels, bg = NULL, alpha = 63, Xnames, id.size = 1.2, ...) # S3 method for varpart234 plot(x, cutoff = 0, digits = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Y Data frame matrix containing response data table dissimilarity structure inheriting dist. community ecology, table often site--species table dissimilarity object. X Two four explanatory models, variables tables. can defined three alternative ways: (1) one-sided model formulae beginning ~ defining model, (2) name single numeric factor variable, (3) name matrix numeric data frame numeric factor variables. model formulae can factors, interaction terms transformations variables. names variables model formula found data frame given data argument, found , user environment. Single variables, data frames matrices found user environment. entries till next argument (data transfo) interpreted explanatory models, names extra arguments abbreviated omitted. ... parameters passed functions. NB, arguments dots abbreviated must spelt completely. data data frame variables used formulae X. chisquare Partition Chi-square inertia Correspondence Analysis (cca). transfo Transformation Y (community data) using decostand. alternatives decostand can used, preserving Euclidean metric include \"hellinger\", \"chi.square\", \"total\", \"norm\". Ignored Y dissimilarities. scale columns Y standardized unit variance. Ignored Y dissimilarities. add Add constant non-diagonal values euclidify dissimilarities (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. argument effect Y dissimilarities. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. NB., argument name abbreviated. argument effect Y dissimilarities. permutations chisquare = TRUE, adjusted \\(R^2\\) estimated permutations, paramater can list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parts Number explanatory tables (circles) displayed. labels Labels used displayed fractions. Default use letters printed output. bg Fill colours circles ellipses. alpha Transparency fill colour. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. Xnames Names sources variation. Default names X1, X2, X3 X4. Xnames=NA, Xnames=NULL Xnames=\"\" produce names. names can changed names. often best use short names. id.size numerical value giving character expansion factor names circles ellipses. x, object varpart result. cutoff values cutoff displayed. digits number significant digits; number decimal places least one higher.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"functions partition variation Y components accounted two four explanatory tables combined effects. Y multicolumn data frame matrix, partitioning based redundancy analysis (RDA, see rda) constrained correspondence analysis chisquare = TRUE (CCA, see cca). Y single variable, partitioning based linear regression. Y dissimilarities, decomposition based distance-based redundancy analysis (db-RDA, see dbrda) following McArdle & Anderson (2001). input dissimilarities must compatible results dist. Vegan functions vegdist, designdist, raupcrick betadiver produce objects, many dissimilarity functions R packages. Partitioning made squared dissimilarities analogously using variance rectangular data -- unless sqrt.dist = TRUE specified. function primarily uses adjusted \\(R^2\\) assess partitions explained explanatory tables combinations (see RsquareAdj), unbiased method (Peres-Neto et al., 2006). raw \\(R^2\\) basic fractions also displayed, biased estimates variation explained explanatory table. correspondence analysis (chisquare = TRUE), adjusted \\(R^2\\) found permutation vary repeated analyses. identifiable fractions designated lower case alphabets. meaning symbols can found separate document (use browseVignettes(\"vegan\")), can displayed graphically using function showvarparts. fraction testable can directly expressed RDA db-RDA model. cases printed output also displays corresponding RDA model using notation explanatory tables | conditions (partialled ; see rda details). Although single fractions can testable, mean fractions simultaneously can tested, since number testable fractions higher number estimated models. non-testable components found differences testable components. testable components permutation variance correspondence analysis (chisquare = TRUE), non-testable components even higher variance. abridged explanation alphabetic symbols individual fractions follows, computational details checked vignette (readable browseVignettes(\"vegan\")) source code. two explanatory tables, fractions explained uniquely two tables [] [b], joint effect [c]. three explanatory tables, fractions explained uniquely three tables [] [c], joint fractions two tables [d] [f], joint fraction three tables [g]. four explanatory tables, fractions explained uniquely four tables [] [d], joint fractions two tables [e] [j], joint fractions three variables [k] [n], joint fraction four tables [o]. summary give overview unique overall contribution group variables. overall contribution (labelled “Contributed”) consists unique contribution variable equal shares fraction variable contributes. summary tabulates fraction divided variables, contributed component sum divided fractions. summary based idea Lai et al. (2022), similar output rdacca.hp package. plot function displays Venn diagram labels intersection (individual fraction) adjusted R squared higher cutoff. helper function showvarpart displays fraction labels. circles ellipses labelled short default names names defined user argument Xnames. Longer explanatory file names can written varpart output plot follows: use option Xnames=NA, add new names using text function. bit fiddling coordinates (see locator) character size allow users place names reasonably short lengths varpart plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Function varpart returns object class \"varpart\" items scale transfo (can missing) hold information standardizations, tables contains names explanatory tables, call function call. function varpart calls function varpart2, varpart3 varpart4 return object class \"varpart234\" saves result item part. items object : SS.Y Sum squares matrix Y. n Number observations (rows). nsets Number explanatory tables bigwarning Warnings collinearity. fract Basic fractions estimated constrained models. indfract Individual fractions possible subsections Venn diagram (see showvarparts). contr1 Fractions can found conditioning single explanatory table models three four explanatory tables. contr2 Fractions can found conditioning two explanatory tables models four explanatory tables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"fraction-data-frames","dir":"Reference","previous_headings":"","what":"Fraction Data Frames","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Items fract, indfract, contr1 contr2 data frames items: Df: Degrees freedom numerator \\(F\\)-statistic fraction. R.square: Raw \\(R^2\\). calculated fract NA items. Adj.R.square: Adjusted \\(R^2\\). Testable: fraction can expressed (partial) RDA model, directly Testable, field TRUE. case fraction label also gives specification testable RDA model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"() References variation partitioning Borcard, D., P. Legendre & P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045--1055. Lai J., Y. Zou, J. Zhang & P. Peres-Neto. 2022. Generalizing hierarchical variation partitioning multiple regression canonical analysis using rdacca.hp R package. Methods Ecology Evolution, 13: 782--788. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. (b) Reference transformations species data Legendre, P. E. D. Gallagher. 2001. Ecologically meaningful transformations ordination species data. Oecologia 129: 271--280. (c) Reference adjustment bimultivariate redundancy statistic Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87: 2614--2625. (d) References partitioning dissimilarities Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290-297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal, Canada. developed Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"can use command browseVignettes(\"vegan\") display document presents Venn diagrams showing fraction names partitioning variation Y respect 2, 3, 4 tables explanatory variables, well equations used variation partitioning. functions frequently give negative estimates variation. Adjusted \\(R^2\\) can negative fraction; unadjusted \\(R^2\\) testable fractions variances non-negative. Non-testable fractions found directly, subtracting different models, subtraction results can negative. fractions orthogonal, linearly independent, complicated nonlinear dependencies can cause negative non-testable fractions. fraction can negative non-Euclidean dissimilarities underlying db-RDA model can yield negative eigenvalues (see dbrda). negative eigenvalues underlying analysis can avoided arguments sqrt.dist add similar effect dbrda: square roots several dissimilarities negative eigenvalues, negative eigenvalues produced Lingoes Cailliez adjustment, effect add random variation dissimilarities. simplified, fast version RDA, CCA adn dbRDA used (functions simpleRDA2, simpleCCA simpleDBRDA). actual calculations done functions varpart2 varpart4, intended called directly user.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) # Two explanatory data frames -- Hellinger-transform Y mod <- varpart(mite, mite.env, mite.pcnm, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = mite.env, mite.pcnm, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.52650 0.43670 TRUE #> [b+c] = X2 22 0.62300 0.44653 TRUE #> [a+b+c] = X1+X2 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09141 TRUE #> [b] = X2|X1 22 0.10124 TRUE #> [c] 0 0.34530 FALSE #> [d] = Residuals 0.46206 FALSE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0914 0.264 mite.env #> X2 0.1012 0.274 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 #> [a] 0.0914 #> [b] 0.1012 #> [c] 0.1726 0.1726 ## Use fill colours showvarparts(2, bg = c(\"hotpink\",\"skyblue\")) plot(mod, bg = c(\"hotpink\",\"skyblue\")) ## Test fraction [a] using partial RDA, '~ .' in formula tells to use ## all variables of data mite.env. aFrac <- rda(decostand(mite, \"hel\"), mite.env, mite.pcnm) anova(aFrac) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(X = decostand(mite, \"hel\"), Y = mite.env, Z = mite.pcnm) #> Df Variance F Pr(>F) #> Model 11 0.053592 1.8453 0.001 *** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## RsquareAdj gives the same result as component [a] of varpart RsquareAdj(aFrac) #> $r.squared #> [1] 0.1359251 #> #> $adj.r.squared #> [1] 0.09140797 #> ## Partition Bray-Curtis dissimilarities varpart(vegdist(mite), mite.env, mite.pcnm) #> #> Partition of squared Bray distance in dbRDA #> #> Call: varpart(Y = vegdist(mite), X = mite.env, mite.pcnm) #> #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 14.696 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.50512 0.41127 TRUE #> [b+c] = X2 22 0.60144 0.41489 TRUE #> [a+b+c] = X1+X2 33 0.74631 0.51375 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09887 TRUE #> [b] = X2|X1 22 0.10249 TRUE #> [c] 0 0.31240 FALSE #> [d] = Residuals 0.48625 FALSE #> --- #> Use function ‘dbrda’ to test significance of fractions of interest ## Three explanatory tables with formula interface mod <- varpart(mite, ~ SubsDens + WatrCont, ~ Substrate + Shrub + Topo, mite.pcnm, data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm, data = mite.env, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm #> #> No. of explanatory tables: 3 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [a+d+f+g] = X1 2 0.32677 0.30667 TRUE #> [b+d+e+g] = X2 9 0.40395 0.31454 TRUE #> [c+e+f+g] = X3 22 0.62300 0.44653 TRUE #> [a+b+d+e+f+g] = X1+X2 11 0.52650 0.43670 TRUE #> [a+c+d+e+f+g] = X1+X3 24 0.67372 0.49970 TRUE #> [b+c+d+e+f+g] = X2+X3 31 0.72400 0.49884 TRUE #> [a+b+c+d+e+f+g] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3 2 0.03910 TRUE #> [b] = X2 | X1+X3 9 0.03824 TRUE #> [c] = X3 | X1+X2 22 0.10124 TRUE #> [d] 0 0.01407 FALSE #> [e] 0 0.09179 FALSE #> [f] 0 0.08306 FALSE #> [g] 0 0.17045 FALSE #> [h] = Residuals 0.46206 FALSE #> Controlling 1 table X #> [a+d] = X1 | X3 2 0.05317 TRUE #> [a+f] = X1 | X2 2 0.12216 TRUE #> [b+d] = X2 | X3 9 0.05231 TRUE #> [b+e] = X2 | X1 9 0.13003 TRUE #> [c+e] = X3 | X1 22 0.19303 TRUE #> [c+f] = X3 | X2 22 0.18429 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.144 ~SubsDens + WatrCont #> X2 0.0382 0.148 ~Substrate + Shrub + Topo #> X3 0.1012 0.245 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 X3 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10124 #> [d] 0.00703 0.00703 #> [e] 0.04590 0.04590 #> [f] 0.04153 0.04153 #> [g] 0.05682 0.05682 0.05682 showvarparts(3, bg=2:4) plot(mod, bg=2:4) ## Use RDA to test fraction [a] ## Matrix can be an argument in formula rda.result <- rda(decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) anova(rda.result) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) #> Df Variance F Pr(>F) #> Model 2 0.013771 2.6079 0.005 ** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Four explanatory tables mod <- varpart(mite, ~ SubsDens + WatrCont, ~Substrate + Shrub + Topo, mite.pcnm[,1:11], mite.pcnm[,12:22], data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm[, 1:11], mite.pcnm[, 12:22], data = mite.env, transfo = #> \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm[, 1:11] #> X4: mite.pcnm[, 12:22] #> #> No. of explanatory tables: 4 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [aeghklno] = X1 2 0.32677 0.30667 TRUE #> [befiklmo] = X2 9 0.40395 0.31454 TRUE #> [cfgjlmno] = X3 11 0.53231 0.44361 TRUE #> [dhijkmno] = X4 11 0.09069 -0.08176 TRUE #> [abefghiklmno] = X1+X2 11 0.52650 0.43670 TRUE #> [acefghjklmno] = X1+X3 13 0.59150 0.49667 TRUE #> [adeghijklmno] = X1+X4 13 0.40374 0.26533 TRUE #> [bcefgijklmno] = X2+X3 20 0.63650 0.48813 TRUE #> [bdefhijklmno] = X2+X4 20 0.53338 0.34292 TRUE #> [cdfghijklmno] = X3+X4 22 0.62300 0.44653 TRUE #> [abcefghijklmno] = X1+X2+X3 22 0.67947 0.52944 TRUE #> [abdefghijklmno] = X1+X2+X4 22 0.61553 0.43557 TRUE #> [acdefghijklmno] = X1+X3+X4 24 0.67372 0.49970 TRUE #> [bcdefghijklmno] = X2+X3+X4 31 0.72400 0.49884 TRUE #> [abcdefghijklmno] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3+X4 2 0.03910 TRUE #> [b] = X2 | X1+X3+X4 9 0.03824 TRUE #> [c] = X3 | X1+X2+X4 11 0.10237 TRUE #> [d] = X4 | X1+X2+X3 11 0.00850 TRUE #> [e] 0 0.01407 FALSE #> [f] 0 0.13200 FALSE #> [g] 0 0.05355 FALSE #> [h] 0 0.00220 FALSE #> [i] 0 -0.00547 FALSE #> [j] 0 -0.00963 FALSE #> [k] 0 -0.00231 FALSE #> [l] 0 0.24037 FALSE #> [m] 0 -0.03474 FALSE #> [n] 0 0.02730 FALSE #> [o] 0 -0.06761 FALSE #> [p] = Residuals 0 0.46206 FALSE #> Controlling 2 tables X #> [ae] = X1 | X3+X4 2 0.05317 TRUE #> [ag] = X1 | X2+X4 2 0.09265 TRUE #> [ah] = X1 | X2+X3 2 0.04131 TRUE #> [be] = X2 | X3+X4 9 0.05231 TRUE #> [bf] = X2 | X1+X4 9 0.17024 TRUE #> [bi] = X2 | X1+X3 9 0.03277 TRUE #> [cf] = X3 | X1+X4 11 0.23437 TRUE #> [cg] = X3 | X2+X4 11 0.15592 TRUE #> [cj] = X3 | X1+X2 11 0.09274 TRUE #> [dh] = X4 | X2+X3 11 0.01071 TRUE #> [di] = X4 | X1+X3 11 0.00303 TRUE #> [dj] = X4 | X1+X2 11 -0.00113 TRUE #> Controlling 1 table X #> [aghn] = X1 | X2 2 0.12216 TRUE #> [aehk] = X1 | X3 2 0.05306 TRUE #> [aegl] = X1 | X4 2 0.34709 TRUE #> [bfim] = X2 | X1 9 0.13003 TRUE #> [beik] = X2 | X3 9 0.04452 TRUE #> [befl] = X2 | X4 9 0.42468 TRUE #> [cfjm] = X3 | X1 11 0.19000 TRUE #> [cgjn] = X3 | X2 11 0.17359 TRUE #> [cfgl] = X3 | X4 11 0.52830 TRUE #> [dijm] = X4 | X1 11 -0.04134 TRUE #> [dhjn] = X4 | X2 11 0.02837 TRUE #> [dhik] = X4 | X3 11 0.00292 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.1456 ~SubsDens + WatrCont #> X2 0.0382 0.1594 ~Substrate + Shrub + Topo #> X3 0.1024 0.2511 mite.pcnm[, 1:11] #> X4 0.0085 -0.0181 mite.pcnm[, 12:22] #> #> Contributions of fractions to sets: #> #> X1 X2 X3 X4 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10237 #> [d] 0.00850 #> [e] 0.00703 0.00703 #> [f] 0.06600 0.06600 #> [g] 0.02678 0.02678 #> [h] 0.00110 0.00110 #> [i] -0.00274 -0.00274 #> [j] -0.00482 -0.00482 #> [k] -0.00077 -0.00077 -0.00077 #> [l] 0.08012 0.08012 0.08012 #> [m] -0.01158 -0.01158 -0.01158 #> [n] 0.00910 0.00910 0.00910 #> [o] -0.01690 -0.01690 -0.01690 -0.01690 plot(mod, bg=2:5) ## Show values for all partitions by putting 'cutoff' low enough: plot(mod, cutoff = -Inf, cex = 0.7, bg=2:5)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":null,"dir":"Reference","previous_headings":"","what":"Defunct Functions in Package vegan — vegan-defunct","title":"Defunct Functions in Package vegan — vegan-defunct","text":"functions variables listed longer part vegan longer needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Defunct Functions in Package vegan — vegan-defunct","text":"","code":"## defunct in vegan 2.7-0 (development 2.6-7) adonis(formula, data, permutations = 999, method = \"bray\", strata = NULL, contr.unordered = \"contr.sum\", contr.ordered = \"contr.poly\", parallel = getOption(\"mc.cores\"), ...) ## defunct in vegan 2.6-0 as.mlm(x) humpfit(mass, spno, family = poisson, start) vegandocs(doc = c(\"NEWS\", \"ONEWS\", \"FAQ-vegan\", \"intro-vegan\", \"diversity-vegan\", \"decision-vegan\", \"partitioning\", \"permutations\")) ## defunct in vegan 2.5-0 commsimulator(x, method, thin=1) ## defunct in vegan 2.4-0 # S3 method for adonis density(x, ...) # S3 method for vegandensity plot(x, main = NULL, xlab = NULL, ylab = \"Density\", type = \"l\", zero.line = TRUE, obs.line = TRUE, ...) # S3 method for adonis densityplot(x, data, xlab = \"Null\", ...) ## defunct in vegan 2.2-0 metaMDSrotate(object, vec, na.rm = FALSE, ...) ## defunct in vegan 2.0-0 getNumObs(object, ...) permuted.index2(n, control = permControl())"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Defunct Functions in Package vegan — vegan-defunct","text":"adonis2 replaces adonis extended functionality completely new internal design. shared arguments adonis similar adonis2, arguments contr.unordered contr.ordered can set contrasts within adonis. .mlm function replaced set functions can find statistics directly ordination result object: see hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca, vcov.cca. Function humpfit transferred natto package still available https://github.com/jarioksa/natto/. R functions news used read vegan NEWS (news(package = \"vegan\")), browseVignettes better tool reading vignettes vegandocs. Function commsimulator replaced make.commsim defines Null models, functions nullmodel simulate.nullmodel check input data generate Null model communities. deprecated density densityplot methods replaced similar methods permustats. permustats offers powerful analysis tools permutations, including summary.permustats giving \\(z\\) values (.k.. standardized effect sizes, SES), Q-Q plots (qqnorm.permustats, qqmath.permustats). Function metaMDSrotate replaced MDSrotate can handle monoMDS results addition metaMDS. permutation functions moved permute package, documented . permute package replaces permuted.index permuted.index2 shuffle getNumObs specific nobs-methods.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions in vegan package — vegan-deprecated","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"functions provided compatibility older versions vegan , may defunct soon next release.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"","code":"## moved to vegan3d package: install from CRAN orditkplot(...) ## use toCoda instead as.mcmc.oecosimu(x) as.mcmc.permat(x)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"x object transformed. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"orditkplot moved vegan3d (version 1.3-0). Install package CRAN use old way. .mcmc functions replaced toCoda.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":null,"dir":"Reference","previous_headings":"","what":"Internal vegan functions — vegan-internal","title":"Internal vegan functions — vegan-internal","text":"Internal vegan functions intended called directly, within functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Internal vegan functions — vegan-internal","text":"","code":"ordiParseFormula(formula, data, xlev = NULL, na.action = na.fail, subset = NULL, X) ordiTerminfo(d, data) ordiNAexclude(x, excluded) ordiNApredict(omit, x) ordiArgAbsorber(..., shrink, origin, scaling, triangular, display, choices, const, truemean, FUN) centroids.cca(x, mf, wt) getPermuteMatrix(perm, N, strata = NULL) howHead(x, ...) pasteCall(call, prefix = \"Call:\") veganCovEllipse(cov, center = c(0, 0), scale = 1, npoints = 100) veganMahatrans(x, s2, tol = sqrt(.Machine$double.eps)) hierParseFormula(formula, data) GowerDblcen(x, na.rm = TRUE) addLingoes(d) addCailliez(d)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Internal vegan functions — vegan-internal","text":"description intended vegan developers: functions intended users, used within functions. general, functions exported namespace, must use get ::: directly call functions. ordiParseFormula returns list three matrices (dependent variables, model.matrix constraints conditions, possibly NULL) needed constrained ordination. Argument xlev passed model.frame. left-hand-side already evaluated calling code, can given argument X re-evaluated. ordiTermInfo finds term information constrained ordination described cca.object. ordiNAexclude implements na.action = na.exclude constrained ordination finding WA scores CCA components site scores unconstrained component excluded rows observations. Function ordiNApredict pads result object WA scores similarly napredict. ordiArgAbsorber absorbs arguments scores function vegan cause superfluous warnings graphical function FUN. implement scores functions new arguments, update ordiArgAbsorber. centroids.cca finds weighted centroids variables. getPermuteMatrix interprets user input returns permutation matrix row gives indices observations permutation. input perm can single number number simple permutations, result defining permutation scheme permutation matrix. howHead formats permutation scheme display. formatting compact one used print permute package, shows non-default choices. output normally used printing results vegan permutations. pasteCall prints function call nicely wrapped Sweave output. veganCovEllipse finds coordinates drawing covariance ellipse. veganMahatrans transforms data matrix Euclidean distances Mahalanobis distances. input data x must matrix centred columns, s2 covariance matrix. s2 given, covariance matrix found x within function. hierParseFormula returns list one matrix (left hand side) model frame factors representing hierarchy levels (right hand side) used adipart, multipart hiersimu. GowerDblcen performs Gower double centring matrix dissimilarities. Similar function earlier available compiled code stats, part official API, therefore poorer replacement. addLingoes addCailliez find constant added non-diagonal (squared) dissimilarities make eigenvalues non-negative Principal Co-ordinates Analysis (wcmdscale, dbrda, capscale). Function cmdscale implements Cailliez method. argument matrix dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":null,"dir":"Reference","previous_headings":"","what":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan package provides tools descriptive community ecology. basic functions diversity analysis, community ordination dissimilarity analysis. multivariate tools can used data types well.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"functions vegan package contain tools diversity analysis, ordination methods tools analysis dissimilarities. Together labdsv package, vegan package provides standard tools descriptive community analysis. Package ade4 provides alternative comprehensive package, several packages complement vegan provide tools deeper analysis specific fields. Package BiodiversityR provides GUI large subset vegan functionality. vegan package developed GitHub (https://github.com/vegandevs/vegan/). GitHub provides --date information forums bug reports. important changes vegan documents can read news(package=\"vegan\") vignettes can browsed browseVignettes(\"vegan\"). vignettes include vegan FAQ, discussion design decisions, short introduction ordination discussion diversity methods. see preferable citation package, type citation(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan development team Jari Oksanen, F. Guillaume Blanchet, Roeland Kindt, Pierre Legendre, Peter R. Minchin, R. B. O'Hara, Gavin L. Simpson, Peter Solymos, M. Henry H. Stevens, Helene Wagner. Many people contributed individual functions: see credits function help pages.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"","code":"### Example 1: Unconstrained ordination ## NMDS data(varespec) data(varechem) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.184583 #> ... Procrustes: rmse 0.04902295 max resid 0.1551746 #> Run 2 stress 0.1948413 #> Run 3 stress 0.2136761 #> Run 4 stress 0.2079056 #> Run 5 stress 0.2087937 #> Run 6 stress 0.196245 #> Run 7 stress 0.2069725 #> Run 8 stress 0.1948413 #> Run 9 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04165046 max resid 0.1519318 #> Run 10 stress 0.1843196 #> Run 11 stress 0.2109853 #> Run 12 stress 0.2177543 #> Run 13 stress 0.2224267 #> Run 14 stress 0.2044974 #> Run 15 stress 0.195049 #> Run 16 stress 0.1969805 #> Run 17 stress 0.22911 #> Run 18 stress 0.2397062 #> Run 19 stress 0.18584 #> Run 20 stress 0.2295494 #> *** Best solution was not repeated -- monoMDS stopping criteria: #> 1: no. of iterations >= maxit #> 19: stress ratio > sratmax plot(ord, type = \"t\") ## Fit environmental variables ef <- envfit(ord, varechem) ef #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05721 -0.99836 0.2537 0.038 * #> P 0.61964 0.78489 0.1938 0.102 #> K 0.76634 0.64244 0.1809 0.127 #> Ca 0.68509 0.72846 0.4119 0.005 ** #> Mg 0.63243 0.77462 0.4271 0.003 ** #> S 0.19127 0.98154 0.1752 0.110 #> Al -0.87166 0.49011 0.5269 0.001 *** #> Fe -0.93612 0.35169 0.4451 0.001 *** #> Mn 0.79870 -0.60173 0.5231 0.001 *** #> Zn 0.61752 0.78655 0.1879 0.102 #> Mo -0.90302 0.42960 0.0609 0.521 #> Baresoil 0.92500 -0.37996 0.2508 0.050 * #> Humdepth 0.93288 -0.36018 0.5200 0.002 ** #> pH -0.64804 0.76161 0.2308 0.074 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> plot(ef, p.max = 0.05) ### Example 2: Constrained ordination (RDA) ## The example uses formula interface to define the model data(dune) data(dune.env) ## No constraints: PCA mod0 <- rda(dune ~ 1, dune.env) mod0 #> Call: rda(formula = dune ~ 1, data = dune.env) #> #> Inertia Rank #> Total 84.12 #> Unconstrained 84.12 19 #> Inertia is variance #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 24.795 18.147 7.629 7.153 5.695 4.333 3.199 2.782 #> (Showing 8 of 19 unconstrained eigenvalues) #> plot(mod0) ## All environmental variables: Full model mod1 <- rda(dune ~ ., dune.env) mod1 #> Call: rda(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 63.2062 0.7513 12 #> Unconstrained 20.9175 0.2487 7 #> Inertia is variance #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 22.396 16.208 7.039 4.038 3.760 2.609 2.167 1.803 1.404 0.917 0.582 #> RDA12 #> 0.284 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 #> 6.627 4.309 3.549 2.546 2.340 0.934 0.612 #> plot(mod1) ## Automatic selection of variables by permutation P-values mod <- ordistep(mod0, scope=formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.015 * #> + A1 1 89.591 1.9217 0.065 . #> + Use 2 91.032 1.1741 0.240 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.095 . #> + A1 1 87.424 1.2965 0.200 #> + Use 2 88.284 1.0510 0.360 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.015 * #> - Management 3 87.707 2.1769 0.010 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.325 #> + A1 1 86.220 0.8359 0.585 #> + Use 2 86.842 0.8027 0.770 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> plot(mod) ## Permutation test for all variables anova(mod) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Model 6 46.425 2.6682 0.001 *** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test of \"type III\" effects, or significance when a term ## is added to the model after all other terms anova(mod, by = \"margin\") #> Permutation test for rda under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Management 3 18.938 2.1769 0.005 ** #> Moisture 3 17.194 1.9764 0.008 ** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Plot only sample plots, use different symbols and draw SD ellipses ## for Managemenet classes plot(mod, display = \"sites\", type = \"n\") with(dune.env, points(mod, disp = \"si\", pch = as.numeric(Management))) with(dune.env, legend(\"topleft\", levels(Management), pch = 1:4, title = \"Management\")) with(dune.env, ordiellipse(mod, Management, label = TRUE)) ## add fitted surface of diversity to the model ordisurf(mod, diversity(dune), add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 1.28 total = 2.28 #> #> REML score: 3.00623 ### Example 3: analysis of dissimilarites a.k.a. non-parametric ### permutational anova adonis2(dune ~ ., dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ ., data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> A1 1 0.7230 0.16817 5.2038 0.002 ** #> Moisture 3 1.1871 0.27613 2.8482 0.006 ** #> Management 3 0.9036 0.21019 2.1681 0.028 * #> Use 2 0.0921 0.02143 0.3315 0.983 #> Manure 3 0.4208 0.09787 1.0096 0.439 #> Residual 7 0.9725 0.22621 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adonis2(dune ~ Management + Moisture, dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management + Moisture, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.7907 0.001 *** #> Moisture 3 1.1516 0.26788 2.9726 0.002 ** #> Residual 13 1.6788 0.39051 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Dissimilarity Indices for Community Ecologists — vegdist","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function computes dissimilarity indices useful popular community ecologists. indices use quantitative data, although named corresponding binary index, can calculate binary index using appropriate argument. find favourite index , can see can implemented using designdist. Gower, Bray--Curtis, Jaccard Kulczynski indices good detecting underlying ecological gradients (Faith et al. 1987). Morisita, Horn--Morisita, Binomial, Cao Chao indices able handle different sample sizes (Wolda 1981, Krebs 1999, Anderson & Millar 2004), Mountford (1962) Raup-Crick indices presence--absence data able handle unknown (variable) sample sizes. indices discussed Krebs (1999) Legendre & Legendre (2012), properties compared Wolda (1981) Legendre & De Cáceres (2012). Aitchison (1986) distance equivalent Euclidean distance CLR-transformed samples (\"clr\") deals positive compositional data. Robust Aitchison distance Martino et al. (2019) uses robust CLR (\"rlcr\"), making applicable non-negative data including zeroes (unlike standard Aitchison).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"vegdist(x, method=\"bray\", binary=FALSE, diag=FALSE, upper=FALSE, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"x Community data matrix. method Dissimilarity index, partial match \"manhattan\", \"euclidean\", \"canberra\", \"clark\", \"bray\", \"kulczynski\", \"jaccard\", \"gower\", \"altGower\", \"morisita\", \"horn\", \"mountford\", \"raup\", \"binomial\", \"chao\", \"cao\", \"mahalanobis\", \"chisq\", \"chord\", \"hellinger\", \"aitchison\", \"robust.aitchison\". binary Perform presence/absence standardization analysis using decostand. diag Compute diagonals. upper Return upper diagonal. na.rm Pairwise deletion missing observations computing dissimilarities. ... parameters. ignored, except method =\"gower\" accepts range.global parameter decostand, method=\"aitchison\", accepts pseudocount parameter decostand used clr transformation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jaccard (\"jaccard\"), Mountford (\"mountford\"), Raup--Crick (\"raup\"), Binomial Chao indices discussed later section. function also finds indices presence/ absence data setting binary = TRUE. following overview gives first quantitative version, \\(x_{ij}\\) \\(x_{ik}\\) refer quantity species (column) \\(\\) sites (rows) \\(j\\) \\(k\\). binary versions \\(\\) \\(B\\) numbers species compared sites, \\(J\\) number species occur compared sites similarly designdist (many indices produce identical binary versions): Jaccard index computed \\(2B/(1+B)\\), \\(B\\) Bray--Curtis dissimilarity. Binomial index derived Binomial deviance null hypothesis two compared communities equal. able handle variable sample sizes. index fixed upper limit, can vary among sites shared species. discussion, see Anderson & Millar (2004). Cao index CYd index (Cao et al. 1997) suggested minimally biased index high beta diversity variable sampling intensity. Cao index fixed upper limit, can vary among sites shared species. index intended count (integer) data, undefined zero abundances; replaced arbitrary value \\(0.1\\) following Cao et al. (1997). Cao et al. (1997) used \\(\\log_{10}\\), current function uses natural logarithms values approximately \\(2.30\\) times higher 10-based logarithms. Anderson & Thompson (2004) give alternative formulation Cao index highlight relationship Binomial index (). Mountford index defined \\(M = 1/\\alpha\\) \\(\\alpha\\) parameter Fisher's logseries assuming compared communities samples community (cf. fisherfit, fisher.alpha). index \\(M\\) found positive root equation \\(\\exp() + \\exp(bM) = 1 + \\exp[(+b-j)M]\\), \\(j\\) number species occurring communities, \\(\\) \\(b\\) number species separate community (index uses presence--absence information). Mountford index usually misrepresented literature: indeed Mountford (1962) suggested approximation used starting value iterations, proper index defined root equation . function vegdist solves \\(M\\) Newton method. Please note either \\(\\) \\(b\\) equal \\(j\\), one communities subset , dissimilarity \\(0\\) meaning non-identical objects may regarded similar index non-metric. Mountford index range \\(0 \\dots \\log(2)\\). Raup--Crick dissimilarity (method = \"raup\") probabilistic index based presence/absence data. defined \\(1 - prob(j)\\), based probability observing least \\(j\\) species shared compared communities. current function uses analytic result hypergeometric distribution (phyper) find probabilities. probability (index) dependent number species missing sites, adding -zero species data removing missing species data influence index. probability ( index) may almost zero almost one wide range parameter values. index nonmetric: two communities shared species may dissimilarity slightly one, two identical communities may dissimilarity slightly zero. index uses equal occurrence probabilities species, Raup Crick originally suggested sampling probabilities proportional species frequencies (Chase et al. 2011). simulation approach unequal species sampling probabilities implemented raupcrick function following Chase et al. (2011). index can also used transposed data give probabilistic dissimilarity index species co-occurrence (identical Veech 2013). Chao index tries take account number unseen species pairs, similarly method = \"chao\" specpool. Function vegdist implements Jaccard, index defined \\(1-\\frac{U \\times V}{U + V - U \\times V}\\); types can defined function chaodist. Chao equation, \\(U = C_j/N_j + (N_k - 1)/N_k \\times a_1/(2 a_2) \\times S_j/N_j\\), \\(V\\) similar except site index \\(k\\). \\(C_j\\) total number individuals species site \\(j\\) shared site \\(k\\), \\(N_j\\) total number individuals site \\(j\\), \\(a_1\\) (\\(a_2\\)) number species occurring site \\(j\\) one (two) individuals site \\(k\\), \\(S_j\\) total number individuals species present site \\(j\\) occur one individual site \\(k\\) (Chao et al. 2005). Morisita index can used genuine count data (integers) . Horn--Morisita variant able handle abundance data. Mahalanobis distances Euclidean distances matrix columns centred, unit variance, uncorrelated. index commonly used community data, sometimes used environmental variables. calculation based transforming data matrix using Euclidean distances following Mardia et al. (1979). Mahalanobis transformation usually fails number columns larger number rows (sampling units). transformation fails, distances nearly constant except small numeric noise. Users must check returned Mahalanobis distances meaningful. Euclidean Manhattan dissimilarities good gradient separation without proper standardization still included comparison special needs. Chi-square distances (\"chisq\") Euclidean distances Chi-square transformed data (see decostand). internal standardization used correspondence analysis (cca, decorana). Weighted principal coordinates analysis distances row sums weights equal correspondence analysis (see Example wcmdscale). Chi-square distance intended non-negative data, typical community data. However, can calculated long margin sums positive, warning issued negative data entries. Chord distances (\"chord\") Euclidean distance matrix rows standardized unit norm (sums squares 1) using decostand. Geometrically standardization moves row points surface multidimensional unit sphere, distances chords across hypersphere. Hellinger distances (\"hellinger\") related Chord distances, data standardized unit total (row sums 1) using decostand, square root transformed. distances upper limit \\(\\sqrt{2}\\). Bray--Curtis Jaccard indices rank-order similar, indices become identical rank-order similar standardizations, especially presence/absence transformation equalizing site totals decostand. Jaccard index metric, probably preferred instead default Bray-Curtis semimetric. Aitchison distance (1986) robust Aitchison distance (Martino et al. 2019) metrics deal compositional data. Aitchison distance said outperform Jensen-Shannon divergence Bray-Curtis dissimilarity, due better stability subsetting aggregation, proper distance (Aitchison et al., 2000). naming conventions vary. one adopted traditional rather truthful priority. function finds either quantitative binary variants indices name, correctly may refer one alternatives instance, Bray index known also Steinhaus, Czekanowski Sørensen index. quantitative version Jaccard probably called Ružička index. abbreviation \"horn\" Horn--Morisita index misleading, since separate Horn index. abbreviation changed index implemented vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Function drop-replacement dist function returns distance object type. result object adds attribute maxdist gives theoretical maximum index sampling units share species, NA maximum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Aitchison, J., Barceló-Vidal, C., Martín-Fernández, J.., Pawlowsky-Glahn, V. (2000). Logratio analysis compositional distance. Math. Geol. 32, 271–275. Anderson, M.J. Millar, R.B. (2004). Spatial variation effects habitat temperate reef fish assemblages northeastern New Zealand. Journal Experimental Marine Biology Ecology 305, 191--221. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006). Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Anderson, M.J & Thompson, .. (2004). Multivariate control charts ecological environmental monitoring. Ecological Applications 14, 1921--1935. Cao, Y., Williams, W.P. & Bark, .W. (1997). Similarity measure bias river benthic Auswuchs community analysis. Water Environment Research 69, 95--106. Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005). new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1 Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J. C. (1971). general coefficient similarity properties. Biometrics 27, 623--637. Krebs, C. J. (1999). Ecological Methodology. Addison Wesley Longman. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141 Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate analysis. Academic Press. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Mountford, M. D. (1962). index similarity application classification problems. : P.W.Murphy (ed.), Progress Soil Zoology, 43--50. Butterworths. Veech, J. . (2013). probabilistic model analysing species co-occurrence. Global Ecology Biogeography 22, 252--260. Wolda, H. (1981). Similarity indices, sample size diversity. Oecologia 50, 296--302.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jari Oksanen, contributions Tyler Smith (Gower index), Michael Bedward (Raup--Crick index), Leo Lahti (Aitchison robust Aitchison distance).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function alternative dist adding ecologically meaningful indices. methods produce similar types objects can interchanged method accepting either. Manhattan Euclidean dissimilarities identical methods. Canberra index divided number variables vegdist, dist. differ constant multiplier, alternative vegdist range (0,1). Function daisy (package cluster) provides alternative implementation Gower index also can handle mixed data numeric class variables. two versions Gower distance (\"gower\", \"altGower\") differ scaling: \"gower\" divides distances number observations (rows) scales column unit range, \"altGower\" omits double-zeros divides number pairs least one -zero value, scale columns (Anderson et al. 2006). can use decostand add range standardization \"altGower\" (see Examples). Gower (1971) suggested omitting double zeros presences, often taken general feature Gower distances. See Examples implementing Anderson et al. (2006) variant Gower index. dissimilarity indices vegdist designed community data, give misleading values negative data entries. results may also misleading NA NaN empty sites. principle, study species composition without species remove empty sites community data.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"data(varespec) vare.dist <- vegdist(varespec) # Orlóci's Chord distance: range 0 .. sqrt(2) vare.dist <- vegdist(decostand(varespec, \"norm\"), \"euclidean\") # Anderson et al. (2006) version of Gower vare.dist <- vegdist(decostand(varespec, \"log\"), \"altGower\") #> Warning: non-integer data: divided by smallest positive value # Range standardization with \"altGower\" (that excludes double-zeros) vare.dist <- vegdist(decostand(varespec, \"range\"), \"altGower\")"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Compact Ordered Community Tables — vegemite","title":"Display Compact Ordered Community Tables — vegemite","text":"Functions vegemite tabasco display compact community tables. Function vegemite prints text tables species rows, site takes one column without spaces. Function tabasco provides interface heatmap colour image data. community table can ordered explicit indexing, environmental variables results ordination cluster analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"vegemite(x, use, scale, sp.ind, site.ind, zero=\".\", select ,...) tabasco(x, use, sp.ind = NULL, site.ind = NULL, select, Rowv = TRUE, Colv = TRUE, labRow = NULL, labCol = NULL, scale, col = heat.colors(12), ...) coverscale(x, scale=c(\"Braun.Blanquet\", \"Domin\", \"Hult\", \"Hill\", \"fix\",\"log\"), maxabund, character = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Compact Ordered Community Tables — vegemite","text":"x Community data. use Either vector, object cca, decorana etc. hclust dendrogram ordering sites species. sp.ind, site.ind Species site indices. tabasco, can also hclust tree, agnes clusterings dendrograms. zero Character used zeros. select Select subset sites. can logical vector (TRUE selected sites), vector indices selected sites. order indices influence results, must specify use site.ind reorder sites. Rowv, Colv Re-order dendrograms rows (sites) columns (species) x. Rowv = TRUE, row dendrograms ordered first axis correspondence analysis, Colv = TRUE column dendrograms weighted average (wascores) row order. Alternatively, arguments can vectors used reorder dendrogram. labRow, labCol character vectors row column labels used heatmap instead default. NB., input matrix transposed row labels used data columns. scale vegemite coverscale: cover scale used (can abbreviated). tabasco: scaling colours heatmap. alternatives coverscale can used tabasco, addition \"column\" \"row\" scale columns rows equal maxima (NB., refer transposed data heatmap), \"none\" uses original values. col vector colours used -zero abundance values. maxabund Maximum abundance used scale = \"log\". Data maximum selected subset used missing. character Return character codes suitable vegemite. FALSE, returns corresponding integers. ... Arguments passed coverscale (.e., maxabund) vegemite heatmap tabasco.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Compact Ordered Community Tables — vegemite","text":"function vegemite prints traditional community table. display transposed, species rows sites columns. table printed compact form: one character can used abundance, spaces columns. Species occurrences dropped table. Function tabasco produces similar table vegemite using heatmap, abundances coded colours. function scales abundances equal intervals colour palette, either rows columns can scaled equal maxima, coverscale class systems can used. function can also display dendrograms sites (columns) species given argument (use sites, sp.ind species). parameter use used re-order output. use can vector object hclust agnes, dendrogram ordination result recognized scores (ordination methods vegan vegan). hclust, agnes dendrogram must sites. dendrogram displayed sites tabasco, shown vegemite. dendrogram species displayed, except given sp.ind. use vector, used ordering sites. use object ordination, sites species arranged first axis (provided results available also species). use object hclust, agnes dendrogram, sites ordered similarly cluster dendrogram. Function tabasco re-orders dendrogram Rowv = TRUE Rowv vector. re-ordering available vegemite, can done hand using reorder.dendrogram reorder.hclust. Please note dendrogram hclust reordering can differ: unweighted means merged branches used dendrogram, weighted means (= means leaves cluster) used reorder.hclust. cases species scores missing, species ordered weighted averages (wascores) site order. Species sites can ordered explicitly giving indices names parameters sp.ind site.ind. given, take precedence use. subset sites can displayed using argument select, used order sites, still must give use site.ind. However, tabasco makes two exceptions: site.ind select used use dendrogram (clustering result). addition, sp.ind can hclust tree, agnes clustering dendrogram, case dendrogram plotted left side heatmap. Phylogenetic trees directly used, package ape tools transform hclust trees. scale given, vegemite calls coverscale transform percent cover scale scales traditional class scales used vegetation science (coverscale can called directly, ). Function tabasco can also use traditional class scales, treats transformed values corresponding integers. Braun-Blanquet Domin scales actually strict cover scales, limits used codes r + arbitrary. Scale Hill may inappropriately named, since Mark O. Hill probably never intended cover scale. However, used default “cut levels” TWINSPAN, surprisingly many users stick default, de facto standard publications. traditional scales assume values cover percentages maximum 100. However, non-traditional alternative log can used scale range. class limits integer powers 1/2 maximum (argument maxabund), + used non-zero entries less 1/512 maximum (log stands alternatively logarithmic logical). Scale fix intended “fixing” 10-point scales: truncates scale values integers, replaces 10 X positive values 1 +.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Compact Ordered Community Tables — vegemite","text":"functions used mainly display table, return (invisibly) list items species ordered species index, sites ordered site index, table final ordered community table. items can used arguments sp.ind site.ind reproduce table, table can edited. addition table, vegemite prints numbers species sites name used cover scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Display Compact Ordered Community Tables — vegemite","text":"cover scales presented many textbooks vegetation science; used: Shimwell, D.W. (1971) Description Classification Vegetation. Sidgwick & Jackson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Compact Ordered Community Tables — vegemite","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Compact Ordered Community Tables — vegemite","text":"name vegemite chosen output compact, tabasco just compact, uses heat colours.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"data(varespec) ## Print only more common species freq <- apply(varespec > 0, 2, sum) vegemite(varespec, scale=\"Hult\", sp.ind = freq > 10) #> #> 1122212121122 1112 #> 854739268340575634292011 #> Callvulg 111...11.311...1111.111. #> Empenigr 211332121112211111212213 #> Vaccviti 323332212113221211233234 #> Pinusylv 111.111111111.11.1111111 #> Dicrfusc 12.111441121211111111111 #> Dicrpoly .11.11.11..1.11....1.111 #> Pleuschr 144533435123411111111131 #> Polyjuni 111.11.111112111.111.111 #> Pohlnuta 111111111111.1..11.11111 #> Ptilcili 1111111111..1..11.11..12 #> Cladarbu 321122121332143423111121 #> Cladrang 321221131312145443313241 #> Cladstel 11111311.211.11254555542 #> Cladunci 112111111131111111111111 #> Cladcocc 11..11111111111111.1.11. #> Cladcorn 111111111111111111111111 #> Cladgrac 111111111111111111111.11 #> Cladfimb 11.1111111111111111111.1 #> Cladcris 111111111111111111111111 #> Cladchlo ..1.111..1...1.11.11.1.1 #> Cetreric 111...11.111111111.1.1.. #> Cetrisla .11....1111.1....1.11111 #> Stersp 111.11.1111.112111...11. #> Claddefo 1111111111111111111111.1 #> 24 sites, 24 species #> scale: Hult ## Order by correspondence analysis, use Hill scaling and layout: dca <- decorana(varespec) vegemite(varespec, dca, \"Hill\", zero=\"-\") #> #> 1 1 1 11122211122222 #> 203942561738913046572458 #> Flavniva -1114-11-1-11-1--------- #> Cladstel 5555551451425411111211-- #> Cladphyl -1-1----1-------1------- #> Cladcerv 1---1-----------------1- #> Cladsp ---11--1--11---1-1-11-1- #> Cladamau --1---1----1------------ #> Cladchlo 1111---1-11-111-----11-- #> Cladrang 535254555555223414332321 #> Diphcomp --11-----112----------1- #> Stersp -11-1-4111111-1-111--111 #> Pinusylv 11-111111-111111111-1111 #> Polypili ------111--11-11--1----- #> Cetrisla -1-111--1-1--1--111--111 #> Cladcocc -1111-1111111-11111-1-11 #> Cladarbu 113142453555313343413231 #> Vacculig --1-1----3-1---1---12-11 #> Pohlnuta -11111--11111111111111-1 #> Cladfimb 11111111-111111111111-11 #> Callvulg 111-21-12-51---1221-21-- #> Icmaeric ------1--1------11------ #> Empenigr 342214131314344333143131 #> Vaccviti 342514244334455432444443 #> Cladgrac 1-1111111111111111111111 #> Cetreric -1111-11-111---1111-111- #> Cladcorn 111111111111111111111111 #> Cladcris 111111111111111111111111 #> Peltapht --------1-11--1----1--1- #> Ptilcili 1-11---11-11141--1111111 #> Barbhatc ----------1-121----1---- #> Claddefo 11111111-111111111111111 #> Cladbotr ----------1-1111---1-1-1 #> Betupube -------------1------1-1- #> Dicrpoly -1-1--1-11--1211-11--1-1 #> Cladunci 111111122111111251212311 #> Polycomm ------------11--1--1--1- #> Polyjuni 11-11-1111111121111--131 #> Rhodtome ----------1--2-----21--1 #> Dicrfusc 111111111121112145425-41 #> Pleuschr 111213114132524434555555 #> Vaccmyrt ---1------1-24--12133--4 #> Descflex ----1----1--11-----21-11 #> Nepharct --1------1-1---1------2- #> Dicrsp -----1----1--1-1-11-1541 #> Hylosple ---------------1---3--13 #> 24 sites, 44 species #> scale: Hill ## Show one class from cluster analysis, but retain the ordering above clus <- hclust(vegdist(varespec)) cl <- cutree(clus, 3) sel <- vegemite(varespec, use=dca, select = cl == 3, scale=\"Br\") #> #> 1 12 #> 20921 #> Flavniva .++.. #> Cladstel 55542 #> Cladphyl .++.. #> Cladcerv r.... #> Cladsp ..+.. #> Cladchlo r++.+ #> Cladrang 22121 #> Diphcomp ..+.. #> Stersp .+... #> Pinusylv r++1+ #> Cetrisla .++++ #> Cladcocc .++.. #> Cladarbu +1+1+ #> Pohlnuta .++++ #> Cladfimb r++++ #> Callvulg r+.+. #> Empenigr 22122 #> Vaccviti 22223 #> Cladgrac r.+++ #> Cetreric .++.. #> Cladcorn r+++r #> Cladcris r++++ #> Ptilcili r.+.2 #> Barbhatc ....1 #> Claddefo r++++ #> Cladbotr ....+ #> Betupube ....+ #> Dicrpoly .++.1 #> Cladunci +++1+ #> Polycomm ....+ #> Polyjuni r++.+ #> Rhodtome ....1 #> Dicrfusc r++++ #> Pleuschr ++121 #> Vaccmyrt ..+.2 #> Descflex ....+ #> Dicrsp ...++ #> 5 sites, 37 species #> scale: Braun.Blanquet ## Re-create previous vegemite(varespec, sp=sel$sp, site=sel$site, scale=\"Hult\") #> #> 1 12 #> 20921 #> Flavniva .11.. #> Cladstel 55552 #> Cladphyl .11.. #> Cladcerv 1.... #> Cladsp ..1.. #> Cladchlo 111.1 #> Cladrang 32131 #> Diphcomp ..1.. #> Stersp .1... #> Pinusylv 11111 #> Cetrisla .1111 #> Cladcocc .11.. #> Cladarbu 11111 #> Pohlnuta .1111 #> Cladfimb 11111 #> Callvulg 11.1. #> Empenigr 22123 #> Vaccviti 22334 #> Cladgrac 1.111 #> Cetreric .11.. #> Cladcorn 11111 #> Cladcris 11111 #> Ptilcili 1.1.2 #> Barbhatc ....1 #> Claddefo 11111 #> Cladbotr ....1 #> Betupube ....1 #> Dicrpoly .11.1 #> Cladunci 11111 #> Polycomm ....1 #> Polyjuni 111.1 #> Rhodtome ....1 #> Dicrfusc 11111 #> Pleuschr 11111 #> Vaccmyrt ..1.3 #> Descflex ....1 #> Dicrsp ...11 #> 5 sites, 37 species #> scale: Hult ## Re-order clusters by ordination clus <- as.dendrogram(clus) clus <- reorder(clus, scores(dca, choices=1, display=\"sites\"), agglo.FUN = mean) vegemite(varespec, clus, scale = \"Hult\") #> #> 1 111 1211221212222 #> 431567380922149306254578 #> Flavniva 21.111.111....11........ #> Cladamau .1.1...1................ #> Stersp 111211111....111.1.111.1 #> Polypili ..111..1......111..1.... #> Diphcomp .1...111.1...........1.. #> Cladphyl ..1.....11...1.......... #> Cladrang 344544332133111223121121 #> Cladcerv 1.........1..........1.. #> Cladstel 454121215555213111111.1. #> Cladarbu 322344331111132222121111 #> Vacculig 11...2.1........1.1..111 #> Callvulg 111.1.311.11.1..11111... #> Icmaeric ...1.1.......1...1...... #> Cladsp 1...1.11.1......111..11. #> Cladcocc 1111111111...1111111.1.1 #> Pinusylv 1.111.1111111111111111.1 #> Cladchlo .1..111.111.1.11..1.1... #> Cetrisla 1.1...1.11.111...1.111.1 #> Cladfimb 11.11111111111111111.111 #> Peltapht ..1...11.......1.....11. #> Cetreric 11.1111111...1..111111.. #> Cladgrac 11111111.111111111111111 #> Pohlnuta 111..11111.1111111111.11 #> Ptilcili .11.1.11.11.2.11.1111111 #> Barbhatc ......1.....1.11......1. #> Cladcorn 111111111111111111111111 #> Vaccviti 113122132323412331223232 #> Cladcris 111111111111111111111111 #> Empenigr 111111122122312322111231 #> Cladbotr ......1.....1.111...1.11 #> Betupube ............1.....1..1.. #> Cladunci 111111111111131111112111 #> Claddefo 11.111111111111111111111 #> Dicrpoly ..11.1..11..1.1111.11..1 #> Polycomm ............111......11. #> Rhodtome ......1.....1.....1...11 #> Polyjuni 1.111111111.111111.112.1 #> Dicrfusc 11111111111112111442.211 #> Pleuschr 113111111111123333444455 #> Vaccmyrt ......1..1..311..111..13 #> Nepharct .1...1.1........1....1.. #> Dicrsp ......1....11...111133.1 #> Descflex 1....1......1.1...1..111 #> Hylosple ................1....122 #> 24 sites, 44 species #> scale: Hult ## Abundance values have such a wide range that they must be rescaled tabasco(varespec, dca, scale=\"Braun\") ## Classification trees for species data(dune, dune.taxon) taxontree <- hclust(taxa2dist(dune.taxon)) plotree <- hclust(vegdist(dune), \"average\") ## Automatic reordering of clusters tabasco(dune, plotree, sp.ind = taxontree) ## No reordering of taxonomy tabasco(dune, plotree, sp.ind = taxontree, Colv = FALSE) ## Species cluster: most dissimilarity indices do a bad job when ## comparing rare and common species, but Raup-Crick makes sense sptree <- hclust(vegdist(t(dune), \"raup\"), \"average\") tabasco(dune, plotree, sptree)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Averages Scores for Species — wascores","title":"Weighted Averages Scores for Species — wascores","text":"Computes Weighted Averages scores species ordination configuration environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Averages Scores for Species — wascores","text":"","code":"wascores(x, w, expand=FALSE) eigengrad(x, w)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Averages Scores for Species — wascores","text":"x Environmental variables ordination scores. w Weights: species abundances. expand Expand weighted averages weighted variance corresponding environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores computes weighted averages. Weighted averages “shrink”: extreme values used calculating averages. expand = TRUE, function “deshrinks” weighted averages making biased weighted variance equal biased weighted variance corresponding environmental variable. Function eigengrad returns inverses squared expansion factors attribute shrinkage wascores result environmental gradient. equal constrained eigenvalue cca one gradient used constraint, describes strength gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores returns matrix species define rows ordination axes environmental variables define columns. expand = TRUE, attribute shrinkage inverses squared expansion factors cca eigenvalues variable. Function eigengrad returns shrinkage attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Weighted Averages Scores for Species — wascores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Averages Scores for Species — wascores","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(wisconsin(varespec)) vare.mds <- monoMDS(vare.dist) vare.points <- postMDS(vare.mds$points, vare.dist) vare.wa <- wascores(vare.points, varespec) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa, rownames(vare.wa), cex=0.8, col=\"blue\") ## Omit rare species (frequency <= 4) freq <- apply(varespec>0, 2, sum) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa[freq > 4,], rownames(vare.wa)[freq > 4],cex=0.8,col=\"blue\") ## Works for environmental variables, too. wascores(varechem, varespec) #> N P K Ca Mg S Al #> Callvulg 25.12401 41.66188 246.92383 572.3431 99.40828 49.10552 245.15751 #> Empenigr 21.61371 44.17350 158.92517 580.8403 89.20628 35.75327 107.05670 #> Rhodtome 22.34553 40.35530 162.31108 643.8819 95.07712 30.84303 28.83159 #> Vaccmyrt 24.96352 49.80649 189.70177 656.4179 96.75529 34.55190 32.75705 #> Vaccviti 21.14028 45.86984 162.12372 613.2126 94.30084 37.26975 116.14656 #> Pinusylv 18.37299 44.24818 163.60195 670.9387 93.52214 37.02238 150.45523 #> Descflex 22.24089 54.74804 212.20357 771.0159 114.15179 38.00000 24.64143 #> Betupube 21.51034 28.86552 112.28276 513.5586 75.15172 23.55172 33.12414 #> Vacculig 28.00729 33.48758 114.90230 372.9653 70.89954 29.32378 202.89008 #> Diphcomp 22.34228 39.71049 127.44259 446.1565 80.16173 32.32963 122.31574 #> Dicrsp 21.33007 60.03758 185.04563 828.0544 148.67509 46.75427 90.42294 #> Dicrfusc 23.45681 39.14575 162.91954 578.2309 77.81648 33.48086 60.66890 #> Dicrpoly 20.65446 43.87409 150.51485 665.5845 115.22112 36.17079 90.16733 #> Hylosple 26.10599 67.88980 245.78681 779.6520 111.96685 42.27433 24.92738 #> Pleuschr 22.60476 54.22534 199.96241 712.6278 109.23425 40.01132 70.43900 #> Polypili 23.17377 43.75902 144.82623 724.9738 85.42623 30.58525 145.73115 #> Polyjuni 22.89480 47.98022 154.81906 643.3864 87.27819 33.63863 53.24888 #> Polycomm 21.73521 41.17042 154.91549 631.8704 101.84789 32.17324 46.80986 #> Pohlnuta 19.99885 48.59198 169.68855 678.3813 104.31641 39.94427 132.05458 #> Ptilcili 21.27880 33.44211 127.08522 564.5652 85.96417 27.11720 56.60692 #> Barbhatc 21.17461 27.93323 113.13542 497.9138 77.50564 23.72288 42.09749 #> Cladarbu 23.56127 38.04952 142.03073 454.9019 74.00779 33.81002 173.12698 #> Cladrang 24.28421 38.60534 135.31177 463.2750 70.54209 32.53349 183.79979 #> Cladstel 19.28049 46.71060 158.00576 540.4904 80.19153 40.29106 225.89526 #> Cladunci 21.41240 45.49844 163.40402 621.9100 98.35538 40.00734 119.59481 #> Cladcocc 21.72473 42.80681 156.32330 557.9007 80.95448 36.25161 149.82616 #> Cladcorn 22.11640 47.06656 160.36881 623.5185 95.17781 36.75273 104.71463 #> Cladgrac 22.51887 44.06576 156.50214 583.1558 94.10292 36.93930 134.13424 #> Cladfimb 21.77980 41.82652 153.29444 512.4646 78.28232 35.62323 128.96061 #> Cladcris 20.88795 44.12262 171.04016 574.5672 92.52169 38.24003 116.03507 #> Cladchlo 19.51207 45.39655 150.93190 571.0233 95.77586 39.50862 156.81983 #> Cladbotr 22.97660 38.89574 167.20000 590.8021 99.57234 34.79362 87.75957 #> Cladamau 25.07143 35.84286 105.07857 395.2214 68.18571 27.11429 95.91429 #> Cladsp 19.21923 47.37308 168.49231 526.7654 79.54423 45.15385 215.33846 #> Cetreric 21.00944 47.76972 165.07972 579.6322 99.14944 42.25472 163.46000 #> Cetrisla 18.36552 42.73695 151.78374 626.2813 89.77833 35.33498 132.68227 #> Flavniva 18.56110 61.18194 207.67705 502.9203 60.91755 50.22532 396.82405 #> Nepharct 23.33099 49.10019 146.84715 618.1601 64.27319 29.95760 31.72300 #> Stersp 28.19743 32.84800 94.33459 389.5143 53.25377 24.22175 95.39326 #> Peltapht 21.08553 54.45395 193.38816 886.5487 119.35132 37.92500 106.16447 #> Icmaeric 28.88636 27.00000 87.86818 307.0500 40.48182 22.17273 89.94091 #> Cladcerv 20.25000 56.79000 192.36000 519.2300 62.10000 45.18000 314.92000 #> Claddefo 22.19198 45.22981 167.73069 583.7983 92.01320 38.51369 100.46139 #> Cladphyl 15.73750 54.56875 180.39375 775.4500 99.65625 43.35000 208.55000 #> Fe Mn Zn Mo Baresoil Humdepth pH #> Callvulg 75.457843 52.38247 8.281074 0.4734635 27.241036 2.187819 2.845108 #> Empenigr 38.146102 53.49357 7.159938 0.3289657 27.324317 2.367439 2.888078 #> Rhodtome 5.560906 70.48260 7.444100 0.2251490 37.325030 2.689154 2.895352 #> Vaccmyrt 5.589213 75.17221 7.838533 0.2666732 31.404171 2.798935 2.855216 #> Vaccviti 37.586067 51.81515 7.617213 0.3680289 26.307701 2.307879 2.923128 #> Pinusylv 39.121898 35.22311 7.733333 0.3485401 17.762968 1.996350 3.049148 #> Descflex 6.066429 110.87232 9.526607 0.2316071 22.740179 2.834821 2.822857 #> Betupube 5.417241 37.53448 5.637931 0.2068966 51.496552 2.527586 2.979310 #> Vacculig 93.963929 37.73062 4.593824 0.3780552 21.410710 2.041196 3.006965 #> Diphcomp 73.281173 46.88025 4.593827 0.3725309 31.836574 2.103704 2.856790 #> Dicrsp 22.504617 65.35020 13.060765 0.5610370 23.182889 2.232247 2.954272 #> Dicrfusc 13.922252 61.40958 6.922859 0.3218816 26.918674 2.484399 2.806431 #> Dicrpoly 20.973927 33.28779 9.110561 0.3892739 37.304043 2.228713 3.015842 #> Hylosple 4.729157 115.14606 9.885976 0.2851996 20.956264 2.925000 2.807594 #> Pleuschr 19.113811 77.12277 9.007860 0.3405945 24.584979 2.596881 2.858446 #> Polypili 51.993443 36.74754 8.045902 0.2204918 17.368852 1.493443 3.227869 #> Polyjuni 12.885704 82.56274 7.945126 0.2760289 28.116303 2.615523 2.874729 #> Polycomm 7.895775 68.22535 7.843662 0.2591549 38.687324 2.926761 2.860563 #> Pohlnuta 33.089313 42.02290 8.452290 0.3935115 24.709351 2.147328 2.985496 #> Ptilcili 14.036188 33.93547 5.906924 0.2303712 48.941884 2.502498 2.973376 #> Barbhatc 8.199687 31.11379 5.550784 0.2084639 54.331975 2.514734 2.986834 #> Cladarbu 65.470394 38.28429 6.387500 0.4464046 22.592997 2.048540 2.937879 #> Cladrang 76.612752 34.86010 6.616452 0.3903501 17.270158 1.799128 3.022346 #> Cladstel 84.639467 33.27903 7.287216 0.4054879 9.854042 1.851973 3.052167 #> Cladunci 27.463504 40.10322 9.108102 0.5120114 28.312376 2.362687 2.858564 #> Cladcocc 46.653763 43.53584 7.269176 0.3698925 19.972222 2.025090 2.974194 #> Cladcorn 32.916238 53.38441 7.518489 0.3639871 26.620868 2.399518 2.890997 #> Cladgrac 45.758366 44.63911 7.702724 0.4244163 25.546654 2.230350 2.933658 #> Cladfimb 41.710354 47.83687 6.815152 0.3887626 24.763889 2.239394 2.902778 #> Cladcris 34.729585 44.75984 6.933735 0.3746988 29.711352 2.400000 2.841633 #> Cladchlo 42.089655 35.13276 7.908621 0.3814655 22.319655 2.075000 3.022414 #> Cladbotr 23.374468 46.54255 7.263830 0.3106383 45.725532 2.580851 2.904255 #> Cladamau 68.971429 41.98571 4.928571 0.3214286 27.592857 1.857143 2.914286 #> Cladsp 47.913462 49.61923 8.421154 0.6173077 16.864423 2.213462 2.921154 #> Cetreric 42.614167 36.96694 9.516389 0.5687500 21.452639 2.058056 2.923889 #> Cetrisla 29.617734 33.24138 7.518227 0.3192118 26.417980 2.027586 3.065025 #> Flavniva 94.339916 37.03232 9.116371 0.9987764 19.692312 1.799241 2.923629 #> Nepharct 12.910837 115.15684 7.743536 0.2180608 23.135932 2.541065 2.918251 #> Stersp 30.226998 31.97061 7.635502 0.2828767 15.844007 1.477740 3.038756 #> Peltapht 37.598684 56.64079 7.652632 0.2046053 28.321053 2.286842 3.026316 #> Icmaeric 24.236364 23.95909 6.618182 0.2863636 18.727273 1.568182 2.968182 #> Cladcerv 111.090000 52.04000 8.530000 0.6800000 15.393000 1.870000 2.900000 #> Claddefo 25.116325 48.81105 7.599609 0.4096285 33.814545 2.468133 2.823656 #> Cladphyl 50.475000 35.28125 8.568750 0.2812500 7.728125 1.575000 3.231250 ## And the strengths of these variables are: eigengrad(varechem, varespec) #> N P K Ca Mg S Al #> 0.13000842 0.18880078 0.16246365 0.15722067 0.16359171 0.13391967 0.29817406 #> Fe Mn Zn Mo Baresoil Humdepth pH #> 0.20766831 0.27254480 0.16783834 0.09542514 0.20931501 0.25051326 0.14583161"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Weighted classical multidimensional scaling, also known weighted principal coordinates analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"wcmdscale(d, k, eig = FALSE, add = FALSE, x.ret = FALSE, w) # S3 method for wcmdscale plot(x, choices = c(1, 2), type = \"t\", ...) # S3 method for wcmdscale scores(x, choices = NA, tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"d distance structure returned dist full symmetric matrix containing dissimilarities. k dimension space data represented ; must \\(\\{1,2,\\ldots,n-1\\}\\). missing, dimensions zero eigenvalue. eig indicates whether eigenvalues returned. add additive constant \\(c\\) added non-diagonal dissimilarities \\(n-1\\) eigenvalues non-negative. Alternatives \"lingoes\" (default, also used TRUE) \"cailliez\" ( alternative cmdscale). See Legendre & Anderson (1999). x.ret indicates whether doubly centred symmetric distance matrix returned. w Weights points. x wcmdscale result object function called options eig = TRUE x.ret = TRUE (See Details). choices Axes returned; NA returns real axes. type Type graph may \"t\"ext, \"p\"oints \"n\"one. tidy Return scores compatible ggplot2: scores data.frame, score type variable score labelled \"sites\", weights variable weigth, names variable label. ... arguments passed graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Function wcmdscale based function cmdscale (package stats base R), uses point weights. Points high weights stronger influence result low weights. Setting equal weights w = 1 give ordinary multidimensional scaling. default options, function returns matrix scores scaled eigenvalues real axes. function called eig = TRUE x.ret = TRUE, function returns object class \"wcmdscale\" print, plot, scores, eigenvals stressplot methods, described section Value. method Euclidean, non-Euclidean dissimilarities eigenvalues can negative. disturbs , can avoided adding constant non-diagonal dissimilarities making eigenvalues non-negative. function implements methods discussed Legendre & Anderson (1999): method Lingoes (add=\"lingoes\") adds constant \\(c\\) squared dissimilarities \\(d\\) using \\(\\sqrt{d^2 + 2 c}\\) method Cailliez (add=\"cailliez\") dissimilarities using \\(d + c\\). Legendre & Anderson (1999) recommend method Lingoes, base R function cmdscale implements method Cailliez.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"eig = FALSE x.ret = FALSE (default), matrix k columns whose rows give coordinates points corresponding positive eigenvalues. Otherwise, object class wcmdscale containing components mostly similar cmdscale: points matrix k columns whose rows give coordinates points chosen represent dissimilarities. eig \\(n-1\\) eigenvalues computed scaling process eig true. x doubly centred weighted distance matrix x.ret true. ac, add additive constant adjustment method used avoid negative eigenvalues. NA FALSE adjustment done. GOF Goodness fit statistics k axes. first value based sum absolute values eigenvalues, second value based sum positive eigenvalues weights Weights. negaxes matrix scores axes negative eigenvalues scaled absolute eigenvalues similarly points. NULL negative eigenvalues k specified, include negative eigenvalues. call Function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Gower, J. C. (1966) distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecology 69, 1--24. Mardia, K. V., Kent, J. T. Bibby, J. M. (1979). Chapter 14 Multivariate Analysis, London: Academic Press.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"## Correspondence analysis as a weighted principal coordinates ## analysis of Euclidean distances of Chi-square transformed data data(dune) rs <- rowSums(dune)/sum(dune) d <- dist(decostand(dune, \"chi\")) ord <- wcmdscale(d, w = rs, eig = TRUE) ## Ordinary CA ca <- cca(dune) ## IGNORE_RDIFF_BEGIN ## Eigevalues are numerically similar ca$CA$eig - ord$eig #> CA1 CA2 CA3 CA4 CA5 #> 6.661338e-16 4.440892e-16 0.000000e+00 1.942890e-16 4.163336e-16 #> CA6 CA7 CA8 CA9 CA10 #> -2.359224e-16 -2.081668e-16 1.110223e-16 -1.387779e-17 -2.775558e-17 #> CA11 CA12 CA13 CA14 CA15 #> 7.632783e-17 1.387779e-17 1.387779e-17 -3.122502e-17 6.938894e-18 #> CA16 CA17 CA18 CA19 #> -3.295975e-17 -1.214306e-17 -1.647987e-17 6.028164e-17 ## Configurations are similar when site scores are scaled by ## eigenvalues in CA procrustes(ord, ca, choices=1:19, scaling = \"sites\") #> #> Call: #> procrustes(X = ord, Y = ca, choices = 1:19, scaling = \"sites\") #> #> Procrustes sum of squares: #> -4.263e-14 #> ## IGNORE_RDIFF_END plot(procrustes(ord, ca, choices=1:2, scaling=\"sites\")) ## Reconstruction of non-Euclidean distances with negative eigenvalues d <- vegdist(dune) ord <- wcmdscale(d, eig = TRUE) ## Only positive eigenvalues: cor(d, dist(ord$points)) #> [1] 0.9975185 ## Correction with negative eigenvalues: cor(d, sqrt(dist(ord$points)^2 - dist(ord$negaxes)^2)) #> [1] 1"},{"path":[]},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-7","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-7 (in development)","text":"vegan longer suggests tcltk, suggests vegan3d (version 1.3-0). See orditkplot section Deprecated Defunct.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-6-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.6-7 (in development)","text":"set functions add new points existing NMDS ordination metaMDS monoMDS. serves purpose adding new points existing eigenvector ordination (instance, predict.rda). main function MDSaddpoints. needs input rectangular matrix dissimilarities new points (rows) old points (columns). Support function dist2xy can extract needed matrices dissimilarities (old new) points, function designdist2 can directly find needed dissimilarities two data matrices. addition, analogue package can calculate rectangular dissimilarities, including many indices defined designdist2. function still experimental. particular user interface may need development. Comments welcome.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-7","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-7 (in development)","text":"adonis defunct: use adonis2 improved functionality. Disabled use summary get ordination scores: use scores! summary.cca see #644. summary.decorana defunct. nothing useful, can extract information scores weights. orditkplot code removed, try launch vegan3d::orditkplot loudly (deprecation message go vegan3d). experimental, may go direct error, depending user experience: please comment!","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-61","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6.1","title":"vegan 2.6-6.1","text":"CRAN release: 2024-05-21","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6.1","text":"C function do_wcentre (weighted centring) can segfault due protection error. problem found automatic CRAN checks. do_wcentre internal function called envfit (vectorfit), wcmdscale varpart (simpleCCA) Fixes bug #653.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-6","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6","title":"vegan 2.6-6","text":"CRAN release: 2024-05-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-6","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-6","text":"vegan depends R version 4.1.0. possible build vegan webR/wasm Fortran compiler. Issue #623.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-6","text":"Permutation tests CCA completely redesigned follow C.J.F ter Braak & D.E. te Beest: Environ Ecol Stat 29, 849–868 (2022) (https://doi.org/10.1007/s10651-022-00545-4). constraints now re-weighted permuted response data, partial model also residualized conditions (partial terms). vegan (release 2.4-6) tests identical Canoco, ter Braak & te Beest demonstrated results biased. old vegan (release 2.4-2 earlier) predictors re-weighted residualized. Re-weighting sufficient remove bias moderate variation weights, residualizing predictors necessary strongly varying weights. See discussion issue #542. new scheme concerns CCA weighted method, RDA dbRDA permutation unchanged. summary ordination results longer prints ordination scores often voluminous hide real summary; see issue #203. Ordination scores extracted scores function. breaks CRAN packages use summary.cca extract scores. switch use scores. maintainers contacted patch files suggested adapt change. See instructions fix packages. scores function constrained ordination (CCA, RDA,dbRDA) default return types scores (display = \"\"). Function can optionally return single type scores list one matrix instead returning matrix (new argument droplist). Constrained ordination objects (cca, rda, dbrda) fitted without formula interface can permutation tests (anova) \"axis\" \"onedf\". Models \"terms\" \"margin\" possible formula interface. Permutation tests constrained ordination objects (cca, rda, dbrda) = \"axis\" stop permutations later axis cutoff limit reached. Earlier cutoff exceeded. default stop permutations P-value 1 reached. analysis takes care P-values axes non-decreasing similarly Canoco. Coefficients effects prc models scaled similarly scaled vegan pre 2.5-1. change suggested Cajo ter Braak. Handling negative eigenvalues changed summary eigenvals. Negative eigenvalues given negative “explanation”, accumulated proportions add 1 last non-negative eigenvalue, 1 last negative eigenvalue. printed output capscale shows proportions real components ignores imaginary dimensions. consistent summary support methods. Issue #636. RsquareAdj capscale based positive eigenvalues, imaginary components ignored. stressplot.dbrda refuses handle partial models. first component variation can displayed dbrda internal (“working”) data structures additive. unconstrained model \"CA\", constrained \"CCA\" partial none. predict dbrda return actual type = \"working\". Earlier returned \"lc\" scores weighted eigenvalues. generated distances eigenvalues, though.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6","text":"Parallel processing inefficiently implemented slower non-parallel permutation tests constrained ordination adonis2. plot scores cca rda family methods gave error non-existing axes requested. Now ignores requests axes numbers higher result object. summary prc ignored extra parameters (const). -fitted models high number aliased variables caused rare failure adonis2 permutation tests constrained ordination methods (cca, rda, dbrda, capscale) arguments = \"margin\" = \"axis\". also concerned vif.cca intersetcor. Typically occurred high-order interactions factor variables. See issues #452 #622 methods accept rectangular raw data input alternative distances, pass arguments distance functions. arguments vegdist binary = TRUE pseudocount Aitchison distance. concerns dbrda, capscale bioenv. See issue #631 simper gave arbitrary p-values species occur subset. Now given NA. See https://stackoverflow.com/questions/77881877/ Rsquare.adj gave arbitrary p-values -fitted models residual variation. Now returns NA R2 adjusted. Automatic model building proceed cases, fixed ordiR2step returns R2 = 0 overfitted cases. constrained ordination methods issue warning model residual component. See issue #610 inertcomp(..., display = \"sites\", proportional = TRUE) gave wrong values.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"data-sets-2-6-6","dir":"Changelog","previous_headings":"","what":"Data Sets","title":"vegan 2.6-6","text":"Extended description BCI data sets avoid confusion. complete BCI survey includes stems 1 cm DBH, BCI data set vegan subset stems DBH 10 cm published Science 295, 666—669, 2002. data set intended demonstrate methods vegan ecological research suggest contacting BCI team using complete surveys made available Dryad.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-6","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-6","text":"adonis deprecated: use adonis2. several CRAN packages still use adonis although contacted authors June 2022 April 2024, printed message forthcoming deprecation since vegan 2.6-2. See issue #523. See instructions adapt packages functions use adonis2. orditkplot moved CRAN package vegan3d deprecated vegan. See issue #585 announcement #632 use summary extract ordination scores deprecated: use scores extract scores. version still allows extracting scores summary, fail next versions. summary.cca see instructions change package. Support removed ancient cca objects (results cca, rda, dbrda capscale) generated CRAN release 2.5 (2016). still stray relics, use newobject <- update(ancientobject) modernize result. .mcmc.oecosimu .mcmc.permat defunct: use toCoda. Code defunct functions completely removed.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-4","dir":"Changelog","previous_headings":"","what":"vegan 2.6-4","title":"vegan 2.6-4","text":"CRAN release: 2022-10-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-4","text":"Support scores ggplot2 graphics improved extended ordination functions. Suitable scores can requested argument tidy = TRUE, general available types scores returned data frame variable score labelling type. option implemented default method scores structured wcmdscale objects, glitches fixed rda family decorana. Previously tidy scores implemented cca, rda, dbrda family methods, metaMDS, envfit rarecurve. adonis2 anova constrained ordination results can perform sequential test one-degree--freedom effects multi-level factors split contrasts. Previously test available permutest. New summary function varpart brief overview. summary shows unique overall contributed variation set variables. fractions shared several sets variables divided equally contributing sets following Lai J, Zou Y, Zhang J, Peres-Neto P (2022) Methods Ecology Evolution, 13: 782–788. decorana estimates orthogonalized eigenvalues total inertia (scaled Chi-square). Orthogonalized eigenvalues can add total inertia. Together enabled implementing eigenvals, bstick screeplot methods decorana. Axis lengths reported decorana methods. Implemented tolerance method decorana. returns criterion used rescaling DCA, can used inspect success rescaling: constant 1 whole axis. New toCoda function transform sequential null model results oecosimu object can analysed coda convergence independence MCMC model. Function replaces .mcmc.oecosimu .mcmc.permat. metaMDS informative finding similar repeated results random starts uses less confusing language reporting results. Hellinger distance directly available vegdist. vegdist, betadiver raupcrick set attribute maxdist giving numeric value theoretical maximum dissimilarity index. many dissimilarities 1, √2 Chord Hellinger distances, instance. attribute NA open indices ceiling. betadiver three similarity indices set maxdist 0. metaMDS defaults halfchange scaling dissimilarities numeric maxdist attribute, adapt threshold ceiling value. open indices without ceiling, threshold scale dissimilarities. metaMDS used simple test detect index ceiling 1, test now robust can also find maximum values. inference made, function broadcast message assumed value ceiling. Mountford index vegdist now scaled maximum value log(2). Earlier Mountford distances scaled maximum 1. hatvalues constrained ordination objects can sometimes practically 1 1, now cases exactly 1. cases rstandard, rstudent cooks.distance NaN. behaviour similar stats::lm.influence functions. .rad can handle multi-row data frames matrices return list Rank-Abundance data row. Earlier one site handled. decostand returns attribute parameters settings variables used standardization. New function decobackstand can use parameters reconstruct original non-standardized data. Back-transformation exact round-errors, although attempt keep original zeros exact. Back-transformation possible methods pa, rank rrank implemented alr. Back-transformation queried https://stackoverflow.com/questions/73263526/ Rarefaction rarefaction-based methods make sense original observed counts give misleading results data multiplied rare species removed. Observed counts usually singletons (species count one), method issue warning minimum count higher one (may false positive, inspect data). Concerns functions rarefy, drarefy, rrarefy, rarecurve, specaccum(..., method=\"rarefy\"), rareslope avgdist. See github discussion #537. avgdist exposes .dist arguments can return \"dist\"ance objects appear lower triangles instead appearing symmetric matrices. betadisper plots accept col argument (PR #300).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-4","text":"decorana returned wrong results Hill’s piecewise transformation (arguments /) used, unless downweighting also used. scores failed metaMDS result species scores. Bug introduced release 2.6-2. Issue raised https://stackoverflow.com/questions/72483924/ tolerance.cca failed one axis (choice) requested. decostand(..., method=\"alr\") accept name reference, fail cases. CRAN package proxy interfered simper caused obscure error (github issue #528).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-4","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-4","text":"adonis way deprecation. Use adonis2 instead. .mcmc.oecosimu .mcmc.permat deprecated: used S3 methods without depending coda package. Use toCoda instead.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-2","dir":"Changelog","previous_headings":"","what":"vegan 2.6-2","title":"vegan 2.6-2","text":"CRAN release: 2022-04-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-2","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-2","text":"Compiled code adapted changes R 4.2.0. See issues #447, #507. Cross-references function packages adapted stringent tests CRAN","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-2","text":"Aitchison robust Aitchison distances added vegdist. Similar data transformations added decostand. Several functions can return “tidy” data structures can used ggplot2 graphics: rarecurve, scores functions constrained ordination (cca etc.), decorana, envfit, metaMDS. scores.envfit gained argument arrow.mul. vegan plot functions used automatically, now easier use envfit non-vegan plotting. Added function simpson.unb unbiased Simpson diversity robust variation sample sizes. diversity gained argument group calculate indices pooled data. Discussed issue #393. simper much faster even though parallel processing implemented new code. pairs function added plot permustats variables . varpart accepts dissimilarities given symmetric square matrix instead \"dist\" object per wish issue #497. metaMDS adopted user-friendly policy, trymax always maximum number tries. See dicussion https://stackoverflow.com/questions/66748605/. adonis2 accepts strata. adonis2 new main function replaces old adonis. See issue #427. Fisher alpha (fisherfit) badly suited extreme communities follow Fisher’s model. Now fisherfit returns NA communities 0 1 species, issues warning communities consisting singletons extreme Fisher alpha. adipart multipart formulae automatically add unique id constant. always sandwich requested grouping alpha gamma diversities, change results requested groupings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-2","text":"anova function failed marginal tests constrained partial ordination model (cca, rda etc.) interaction terms. Issue #463. Constrained ordination (cca etc.) gave misleading results external variables (constraints, condition) constant explained nothing. decorana fail axes zero eigenvalues. Issue #401. Species accumulation (specaccum) failed one species, several “communities”. Issue #501. Parallel processing failed Windows socket clusters permutest betadisper. Issue #369. orditorp failed numeric labels supplied. Reported https://stackoverflow.com/questions/69272366/. Argument summarize accidentally dropped goodness.cca 2017. taxa2dist failed one usable taxonomic level. See https://stackoverflow.com/questions/67231431/.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-2","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-2","text":"Function adonis2 replace adonis. humpfit functions defunct removed. available non-CRAN package natto https://github.com/jarioksa/natto. commisimulator defunct. Use simulate nullmodel objects. permuted.index finally defunct (deprecated vegan 2.2-0). .mlm defunct. Use functions documented influence.cca, hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca others.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-7","dir":"Changelog","previous_headings":"","what":"vegan 2.5-7","title":"vegan 2.5-7","text":"CRAN release: 2020-11-28","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-7","text":"Several distance-based functions failed distances zero (betadisper, capscale, isomap, monoMDS, pcnm, wcmdscale). Reported github issue #372. Non-linear self-starting regression models SSarrhenius, SSgitay, SSgleason SSlomolino failed future R. failure caused internal changes R-devel. Github issue #382. Arrow labels wrong position plot.envfit(..., add = FALSE). rarecurve added unnecessary names results. Github issue #352. permutest betadisper failed parallel processing Windows systems socket clusters used. Github issue #369.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-7","text":"Chi-square Chord distances added vegdist. distances can calculated Euclidean distances transformed data, actually available earlier, many users notice . monoMDS (hence metaMDS) uses stricter convergence criteria. improves possibilities find stable solutions. However, users may still need tweak convergence criteria data. See discussion Github issue #354. text functions constrained ordination plots (cca, rda, dbrda, capscale) accept now expression labels. allows using subscripts, superscripts mathematical expressions. New support function labels.cca returns current text labels authors can change desired ones. See github issue #374. vegemite returns invisibly final formatted table allowing processing. ordiplot passes cex argument linestack decorana plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-6","dir":"Changelog","previous_headings":"","what":"vegan 2.5-6","title":"vegan 2.5-6","text":"CRAN release: 2019-09-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-6","text":"vegdist silently accepted missing values (NA) removed analysis also option na.rm = FALSE. behaviour introduced vegan version 2.5-1. See GitHub issue #319. labels displaced bunch arrows drawn origin ordination graph envfit. See GitHub issue #315. Hill scale coverscale open-ended limited percent data, unlike traditional cover class scales undefined 100% cover.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-6","text":"results .rad longer print index attribute: attribute still object, printing made output messy.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-5","dir":"Changelog","previous_headings":"","what":"vegan 2.5-5","title":"vegan 2.5-5","text":"CRAN release: 2019-05-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-5","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-5","text":"vegan depends R 3.4.0 higher. next vegan release may increase dependence R 3.6.0. R 3.6.0 improved method find random indices permuting sampling data. Vegan relies now R functions ecological null models (functions nullmodel, oecosimu, commsim, permatfull, permatswap others). Technically change compatible R 3.4.0 later, can gain benefits improved code current release R. null models may change due change, certainly change R 3.6.0. See NEWS R 3.6.0 release discussion github issue #312. vegan permutation routines rely permute, gain similar benefits improved randomness upgrade R. Thanks new R dependence, sigma constrained ordination results works without workarounds vegan 2.5-2. fixes completely issue discussed #274. Vegan test results reproduced older versions R 3.6.0. worried , upgrade R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-5","text":"metaMDS failed scaling results engine monoMDS used. However, recommend use monoMDS. See github issue #310.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-5","text":"betadisper changed interpretation negative squared distances give complex-valued distances. Now regarded zero-distances whereas earlier used modulus. change results cases negative squared distances. discussion, see github issue #306.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-4","dir":"Changelog","previous_headings":"","what":"vegan 2.5-4","title":"vegan 2.5-4","text":"CRAN release: 2019-02-04","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-testing-2-5-4","dir":"Changelog","previous_headings":"","what":"Installation and Testing","title":"vegan 2.5-4","text":"code interpreting formula change R 3.6.0, makes constrained ordination methods (cca, rda, dbrda, capscale) fail. See github issue #299. R 3.6.0 introduces new environment variable _R_CHECK_LENGTH_1_LOGIC2_, several functions fail variable set. Changes concern ordiplot, plot summary constrained ordination objects, ordixyplot. See github issue #305.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-4","text":"decorana gave incorrect results downweighting used (argument iweigh = 1). bug introduced vegan 2.5-1 reported github issue #303. goodness constrained ordination methods failed constraints rank = 1 (one constraining variable). Reported Pierre Legendre.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-4","text":"Adjusted R2 enabled partial RDA models (functions rda dbrda) partial CCA models (function cca) function RsquareAdj. feature disabled vegan 2.5-1 . RDA, calculation similar vegan 2.4-6 earlier. Partial CCA now consistent RDA differs earlier implementation. methods, partial models consistent varpart. See github issue #295.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-3","dir":"Changelog","previous_headings":"","what":"vegan 2.5-3","title":"vegan 2.5-3","text":"CRAN release: 2018-10-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-3","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-3","text":"Tests numerical analysis written robustly give similar results alternative platforms versions R BLAS/Lapack libraries. See github issue #282.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-3","text":"Constrained ordination gave misleading results constraints conditions data NULL variables. rarely happens normal usage, happen marginal anova reported github issue #291. Several functions numerical analysis wrongly accepted non-numeric data (instance, factors) gave either meaningless results confusing error messages. Fixed functions include beals, designdist, diversity, gdispweight, indpower, spantree, specpool, tsallis, tsallisaccum vegdist. See github issue #292. envfit vectors fail missing data. original data scaled centred similarly simulations simulate.rda several simulations returned simmat object (compatible nullmodel simulations can used oecosimu).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-3","text":"anosim checks input avoid confusing error messages like reported StackOverflow question 52082743. Broken-stick distribution (function bstick) longer calculated distance-based Redundancy Analysis (dbrda) negative eigenvalues, clear done. Now dbrda capscale similar respect. print function betadisper results gained new argument neigen select number eigenvalues shown. print robust number eigenvalues lower requested neigen.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-5-3","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.5-3","text":"Function humpfit moved natto package still available https://github.com/jarioksa/natto. scheduled complete removal vegan 2.6-0.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-2","dir":"Changelog","previous_headings":"","what":"vegan 2.5-2","title":"vegan 2.5-2","text":"CRAN release: 2018-05-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-compatibility-2-5-2","dir":"Changelog","previous_headings":"","what":"Installation and Compatibility","title":"vegan 2.5-2","text":"Vegan declares dependence R version 3.2.0. dependence yet noticed previous vegan release. However, generic sigma function defined R-3.3.0, therefore sigma.cca vegan must spelt completely using R-3.2.x. See discussion issue #274. CRAN package klaR function rda, loaded together vegan clashes vegan rda Redundancy Analysis. Vegan tries mitigate problem. cases vegan functions used vegan loaded klaR, error message issued klaR objects handled vegan functions. klaR also tricked print informative message handles vegan objects. However, vegan namespace can attached automatically start-klaR functions take precedence. reported issue #277. Bioconductor package phyloseq problem vegdist function dissimilarities. problem can fixed re-installing phyloseq source package. , must either downgrade vegan version 2.4-6 wait till Bioconductor binary packages upgraded. reported Stackoverflow, vegan issue #272, phyloseq issues #918 #921.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-2","text":"Plotting betadisper failed groups one member. Reported Stackoverflow “Error: Incorrect .dimensions” plotting multivariate data Vegan. Permutation tests constrained ordination (anova.cca, permutest.cca) fail parallel processing socket clusters. Socket clusters always used Windows can also used operating systems created makeCluster. See issue #276.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-1","dir":"Changelog","previous_headings":"","what":"vegan 2.5-1","title":"vegan 2.5-1","text":"CRAN release: 2018-04-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-5-1","dir":"Changelog","previous_headings":"","what":"GENERAL","title":"vegan 2.5-1","text":"major new release changes package: Nearly 40% program files changed previous release. Please report regressions issues https://github.com/vegandevs/vegan/issues/. Compiled code used much extensively, compiled functions use .Call interface. gives smaller memory footprint also faster. wall clock time, greatest gains permutation tests constrained ordination methods (anova.cca) binary null models (nullmodel). Constrained ordination functions (cca, rda, dbrda, capscale) completely rewritten share code. makes consistent robust. internal structure changed constrained ordination objects, scripts may fail try access result object directly. never guarantee unchanged internal structure, scripts changed use provided support functions access result object (see documentation cca.object github issue #262). support analysis functions may longer work result objects created previous vegan versions. use update(old.result.object) fix old result objects. See github issues #218, #227. vegan includes tests run checking package installation. See github issues #181, #271. informative messages (warnings, notes error messages) cleaned unified also makes possible provide translations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.5-1","text":"avgdist: new function find averaged dissimilarities several random rarefactions communities. Code Geoffrey Hannigan. See github issues #242, #243, #246. chaodist: new function similar designdist, uses Chao terms supposed take account effects unseen species (Chao et al., Ecology Letters 8, 148-159; 2005). Earlier Jaccard-type Chao dissimilarity vegdist, new code allows defining kind Chao dissimilarity. New functions find influence statistics constrained ordination objects: hatvalues, sigma, rstandard, rstudent, cooks.distance, SSD, vcov, df.residual. earlier found via .mlm function deprecated. See github issue #234. boxplot added permustats results display (standardized) effect sizes. sppscores: new function add replace species scores distance-based ordination dbrda, capscale metaMDS. Earlier dbrda species scores, species scores capscale metaMDS based raw input data may consistent used dissimilarity measure. See github issue #254. cutreeord: new function similar stats::cutree, numbers cluster order appear dendrogram (left right) instead labelling order appeared data. sipoo.map: new data set locations sizes islands Sipoo archipelago bird data set sipoo.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-constrained-ordination-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Constrained Ordination","title":"vegan 2.5-1","text":"inertia Correspondence Analysis (cca) called “scaled Chi-square” instead using name little known statistic. elements Constraints Conditions data frames non-formula call rda cca, automatically expanded model matrices can contain factor variables. Earlier numerical model matrices factors used formula interface. Regression scores constraints can extracted plotted constrained ordination methods. See github issue #226. Full model (model = \"full\") enabled permutations tests constrained ordination results anova.cca permutest.cca. permutest.cca gained new option = \"onedf\" perform tests sequential one degree--freedom contrasts factors. option (yet) enabled anova.cca. permutation tests robust, scoping issues fixed. Permutation tests use compiled C code much faster. See github issue #211. permutest printed layout similar anova.cca. eigenvals gained new argument (model) select either constrained unconstrained scores. old argument (constrained) deprecated. See github issue #207. summary.eigenvals returns matrix instead list containing matrix. Adjusted R2 calculated partial ordination, unclear done (function RsquareAdj). ordiresids can display standardized studentized residuals. Function construct model.frame model.matrix constrained ordination robust fail fewer cases. goodness inertcomp constrained ordination result object longer option find distances: explained variation available. inertcomp gained argument unity. give “local contributions beta-diversity” (LCBD) “species contribution beta-diversity” (SCBD) Legendre & De Cáceres (Ecology Letters 16, 951-963; 2012). goodness disabled capscale. prc gained argument const general scaling results similarly rda. prc uses regression scores Canoco-compatibility.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-null-model-communities-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Null Model Communities","title":"vegan 2.5-1","text":"C code swap-based binary null models made efficients, models faster. Many models selected 2 times 2 submatrix, generated four random numbers (two rows, two columns). Now skip selecting third fourth random number obvious matrix swapped. Since time used generating random numbers functions, candidates rejected, speeds functions. However, also means random number sequences change previous vegan versions, old binary model results replicated exactly. See github issues #197, #255 details timing. Ecological null models (nullmodel, simulate, make.commsim, oecosimu) gained new null model \"greedyqswap\" can radically speed quasi-swap models minimal risk introducing bias. Backtracking written C much faster. However, backtracking models biased, provided classic legacy models.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-other-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Other Functions","title":"vegan 2.5-1","text":"adonis2 gained column R2 similarly old adonis. Great part R code decorana written C makes faster reduces memory footprint. metaMDS results gained new points text methods. ordiplot ordination plot functions can chained points text functions allowing use magrittr pipes. points text functions gained argument draw arrows allowing use drawing biplots adding vectors environmental variables ordiplot. Since many ordination plot methods return invisible \"ordiplot\" object, points text methods also work . See github issue #257. Lattice graphics (ordixyplot) ordination can add polygons enclose points panel complete data. ordicluster gained option suppress drawing plots can easily embedded functions calculations. .rad returns index included taxa attribute. Random rarefaction (function rrarefy) uses compiled C code much faster. plot specaccum can draw short horizontal bars vertical error bars. See StackOverflow question 45378751. decostand gained new standardization methods rank rrank replace abundance values ranks relative ranks. See github issue #225. Clark dissimilarity added vegdist (calculated designdist). designdist evaluates minimum terms compiled code, function faster vegdist also dissimilarities using minimum terms. Although designdist usually faster vegdist, numerically less stable, particular large data sets. swan passes type argument beals. tabasco can use traditional cover scale values function coverscale. Function coverscale can return scaled values integers numerical analysis instead returning characters printing. varpart can partition Chi-squared inertia correspondence analysis new argument chisquare. adjusted R2 based permutation tests, replicate analysis random variation. explanatory tables can data frames factors single factors varpart automatically expanded model matrices. Earlier factors used one-sided model formulae. Based code suggested Daniel Borcard, Univ. Montréal.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-1","text":"long Condition() statements (> 500 characters) failed partial constrained ordination models (cca, rda, dbrda, capscale). problem detected StackOverflow question 49249816. Labels adjusted arrows rescaled envfit plots. See StackOverflow question 49259747. ordiArrowMul failed one arrow plotted envfit.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-5-1","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.5-1","text":".mlm function constrained correspondence analysis deprecated favour new functions directly give influence statistics. See github issue #234. commsimulator now defunct: use simulate nullmodel objects. ade4 cca objects longer handled vegan: ade4 cca since version 1.7-8 (August 9, 2017).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-6","dir":"Changelog","previous_headings":"","what":"vegan 2.4-6","title":"vegan 2.4-6","text":"CRAN release: 2018-01-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-6","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-6","text":"CRAN packages longer allowed use FORTRAN input, read.cep function used FORTRAN format read legacy CEP Canoco files. avoid NOTEs WARNINGs, function re-written R. new read.cep less powerful fragile, can read data “condensed” format, can fail several cases successful old code. old FORTRAN-based function still available cepreader. See github issue #263. cepreader package developed https://github.com/vegandevs/cepreader.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-6","text":"functions rarefaction (rrarefy), species abundance distribution (preston) species pool (estimateR) need exact integer data, test allowed small fuzz. functions worked correctly original data, data transformed back-transformed, pass integer test fuzz give wrong results. instance, sqrt(3)^2 pass test 3, interpreted strictly integer 2. See github issue #259.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-6","text":"ordiresids uses now weighted residuals cca results.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-5","dir":"Changelog","previous_headings":"","what":"vegan 2.4-5","title":"vegan 2.4-5","text":"CRAN release: 2017-12-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-5","text":"Several “Swap & Shuffle” null models generated wrong number initial matrices. Usually generated many, dangerous, slow. However, random sequences change fix. Lattice graphics ordination (ordixyplot friends) colour arrows groups instead randomly mixed colours. Information constant mirrored permutations omitted reporting permutation tests (e.g., anova constrained ordination).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-5","text":"ordistep improved interpretation scope: lower scope missing, formula starting solution taken lower scope instead using empty model. See Stackoverflow question 46985029. fitspecaccum gained new support functions nobs logLik allow better co-operation packages functions. See GitHub issue #250. “backtracking” null model community simulation faster. However, “backtracking” biased legacy model used except comparative studies.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-4","dir":"Changelog","previous_headings":"","what":"vegan 2.4-4","title":"vegan 2.4-4","text":"CRAN release: 2017-08-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-4","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-4","text":"orditkplot longer give warnings CRAN tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-4","text":"anova(..., = \"axis\") constrained ordination (cca, rda, dbrda) ignored partial terms Condition(). inertcomp summary.cca failed constrained component defined, explained nothing zero rank. See StackOverflow: R - Error message RDA analysis - vegan package. Labels longer cropped meandist plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-4","text":"significance tests axes constrained ordination use now forward testing strategy. extensive analysis indicated previous marginal tests biased. conflict Legendre, Oksanen & ter Braak, Methods Ecol Evol 2, 269–277 (2011) regarded marginal tests unbiased. Canberra distance vegdist can now handle negative input entries similarly latest versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-3","dir":"Changelog","previous_headings":"","what":"vegan 2.4-3","title":"vegan 2.4-3","text":"CRAN release: 2017-04-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-3","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-3","text":"vegan registers native C Fortran routines. avoids warnings model checking, may also give small gain speed. Future versions vegan deprecate remove elements pCCA$Fit, CCA$Xbar, CA$Xbar cca result objects. release provides new function ordiYbar able construct elements current future releases. Scripts functions directly accessing elements switch ordiYbar smooth transition.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-3","text":".mlm methods constrained ordination include zero intercept give correct residual degrees freedom derived statistics. biplot method rda passes correlation argument scaling algorithm. Biplot scores wrongly centred cca caused small error values. Weighting centring corrected intersetcor spenvcor. fix can make small difference analysing cca results. Partial models correctly handled intersetcor. envfit ordisurf functions failed applied species scores. Non-standard variable names can used within Condition() partial ordination. Partial models used internally within several functions, problem reported Albin Meyer (Univ Lorraine, Metz, France) ordiR2step using variable name contained hyphen (wrongly interpreted minus sign partial ordination). ordispider pass graphical arguments used show difference LC WA scores constrained ordination. ordiR2step uses forward selection avoid several problems model evaluation. tolerance function return NaN cases returned 0. Partial models correctly analysed. Misleading (non-zero) tolerances sometimes given species occurred sampling units one species.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-2","dir":"Changelog","previous_headings":"","what":"vegan 2.4-2","title":"vegan 2.4-2","text":"CRAN release: 2017-01-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-2","text":"Permutation tests (permutests, anova) first axis failed constrained distance-based ordination (dbrda, capscale). Now capscale also throw away negative eigenvalues first eigenvalues tested. permutation tests first axis now faster. problem reported Cleo Tebby fixes discussed GitHub issue #198 pull request #199. support functions dbrda capscale gave results components wrong scale. Fixes stressplot, simulate, predict fitted functions. intersetcor use correct weighting cca results slightly . anova permutest failed betadisper fitted argument bias.adjust = TRUE. Fixes Github issue #219 reported Ross Cunning, O’ahu, Hawaii. ordicluster return invisibly coordinates internal points (clusters points joined), last rows contained coordinates external points (ordination scores points). cca method tolerance returning incorrect values second axis sample heterogeneities species tolerances. See issue #216 details.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-2","text":"Biplot scores scaled similarly site scores constrained ordination methods cca, rda, capscale dbrda. Earlier unscaled (technically, equal scaling axes). tabasco adds argument scale colours rows columns addition old equal scale whole plot. New arguments labRow labCex can used change column row labels. Function also takes care -zero observations coloured: earlier tiny observed values merged zeros distinct plots. Sequential null models somewhat faster (10%). Non-sequential null models may marginally faster. null models generated function nullmodel also used oecosimu. vegdist much faster. used clearly slower stats::dist, now nearly equally fast dissimilarity measure. Handling data= formula interface robust, messages user errors improved. fixes points raised Github issue #200. families orders dune.taxon updated APG IV (Bot J Linnean Soc 181, 1–20; 2016) corresponding classification higher levels (Chase & Reveal, Bot J Linnean Soc 161, 122-127; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-1","dir":"Changelog","previous_headings":"","what":"vegan 2.4-1","title":"vegan 2.4-1","text":"CRAN release: 2016-09-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-4-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.4-1","text":"Fortran code modernized avoid warnings latest R. modernization visible effect functions. Please report suspect cases vegan issues.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-1","text":"Several support functions ordination methods failed solution one ordination axis, instance, one constraining variable CCA, RDA friends. concerned goodness constrained ordination, inertcomp, fitted capscale, stressplot RDA, CCA (GitHub issue #189). goodness CCA & friends ignored choices argument (GitHub issue #190). goodness function consider negative eigenvalues db-RDA (function dbrda). Function meandist failed cases one groups one observation. linestack handle expressions labels. regression discussed GitHub issue #195. Nestedness measures nestedbetajac nestedbetasor expecting binary data cope quantitative input evaluating Baselga’s matrix-wide Jaccard Sørensen dissimilarity indices. Function .mcmc cast oecosimu result MCMC object (coda package) failed one chain.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-1","text":"diversity function returns now NA observation NA values instead returning 0. function also checks input refuses handle data negative values. GitHub issue #187. rarefy function work robustly marginal case user asks one individual can one species zero variance. Several functions robust factor arguments contain missing values (NA): betadisper, adipart, multipart, hiersimu, envfit constrained ordination methods cca, rda, capscale dbrda. GitHub issues #192 #193.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-0","dir":"Changelog","previous_headings":"","what":"vegan 2.4-0","title":"vegan 2.4-0","text":"CRAN release: 2016-06-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"distance-based-analysis-2-4-0","dir":"Changelog","previous_headings":"","what":"Distance-based Analysis","title":"vegan 2.4-0","text":"Distance-based methods redesigned made consistent ordination (capscale, new dbrda), permutational ANOVA (adonis, new adonis2), multivariate dispersion (betadisper) variation partitioning (varpart). methods can produce negative eigenvalues several popular semimetric dissimilarity indices, handled similarly functions. Now functions designed McArdle & Anderson (Ecology 82, 290–297; 2001). dbrda new function distance-based Redundancy Analysis following McArdle & Anderson (Ecology 82, 290–297; 2001). metric dissimilarities, function equivalent old capscale, negative eigenvalues semimetric indices handled differently. dbrda dissimilarities decomposed directly conditions, constraints residuals negative eigenvalues, components can imaginary dimensions. Function mostly compatible capscale constrained ordination methods, full compatibility achieved (see issue #140 Github). function based code Pierre Legendre. old capscale function constrained ordination still based real components, total inertia components assessed similarly dbrda. significance tests differ previous version, function oldCapscale cast capscale result similar form previously. adonis2 new function permutational ANOVA dissimilarities. based algorithm dbrda. function can perform overall tests independent variables well sequential marginal tests term. old adonis still available, can perform sequential tests. settings, adonis adonis2 give identical results (see Github issue #156 differences). Function varpart can partition dissimilarities using algorithm dbrda. Argument sqrt.dist takes square roots dissimilarities can change many popular semimetric indices metric distances capscale, dbrda, wcmdscale, adonis2, varpart betadisper (issue #179 Github). Lingoes Cailliez adjustments change dissimilarity metric distance capscale, dbrda, adonis2, varpart, betadisper wcmdscale. Earlier Cailliez adjustment capscale (issue #179 Github). RsquareAdj works capscale dbrda allows using ordiR2step model building.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-0","text":"specaccum: plot failed line type (lty) given. Reported Lila Nath Sharma (Univ Bergen, Norway)","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-4-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.4-0","text":"ordibar new function draw crosses standard deviations standard errors ordination diagrams instead corresponding ellipses. Several permustats results can combined new c() function. New function smbind binds together null models row, column replication. sequential models bound together, can treated parallel chains subsequent analysis (e.g., .mcmc). See issue #164 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-0","text":"Null model analysis upgraded: New \"curveball\" algorithm provides fast null model fixed row column sums binary matrices Strona et al. (Nature Commun. 5: 4114; 2014). \"quasiswap\" algorithm gained argument thin can reduce bias null models. \"backtracking\" now much faster, still slow, provided mainly allow comparison better faster methods. Compiled code can now interrupted null model simulations. designdist can now use beta diversity notation (gamma, alpha) easier definition beta diversity indices. metaMDS new iteration strategy: Argument try gives minimum number random starts, trymax maximum number. Earlier hand try gave maximum number, now run least try times. reduces risk trapped local optimum (issue #154 Github). convergent solutions, metaMDS now tabulate stopping criteria (trace = TRUE). can help deciding criteria made stringent number iterations increased. documentation monoMDS metaMDS give detailed information convergence criteria. summary permustats prints now P-values, test direction (alternative) can changed. qqmath function permustats can now plot standardized statistics. partial solution issue #172 Github. MDSrotate can rotate ordination show maximum separation factor levels (classes) using linear discriminant analysis (lda MASS package). adipart, hiersimu multipart expose argument method specify null model. RsquareAdj works cca allows using ordiR2step model building. code developed Dan McGlinn (issue #161 Github). However, cca still used varpart. ordiellipse ordihull allow setting colours, line types graphical parameters. alpha channel can now given also real number 0 … 1 addition integer 0 … 255. ordiellipse can now draw ellipsoid hulls enclose points group. ordicluster, ordisegments, ordispider lines plot functions isomap spantree can use mixture colours connected points. behaviour similar analogous functions vegan3d package. plot betadisper configurable. See issues #128 #166 Github details. text points methods orditkplot respect stored graphical parameters. Environmental data Barro Colorado Island forest plots gained new variables Harms et al. (J. Ecol. 89, 947–959; 2001). Issue #178 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-4-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.4-0","text":"Function metaMDSrotate removed replaced MDSrotate. density densityplot methods various vegan objects deprecated replaced density densityplot permustats. Function permustats can extract permutation simulation results vegan result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-5","dir":"Changelog","previous_headings":"","what":"vegan 2.3-5","title":"vegan 2.3-5","text":"CRAN release: 2016-04-09","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-5","text":"eigenvals fails prcomp results R-devel. next version prcomp argument limit number eigenvalues shown (rank.), breaks eigenvals vegan. calibrate failed cca friends rank given.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-4","dir":"Changelog","previous_headings":"","what":"vegan 2.3-4","title":"vegan 2.3-4","text":"CRAN release: 2016-02-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-4","text":"betadiver index 19 wrong sign one terms. linestack failed labels given, input scores names. Reported Jeff Wood (ANU, Canberra, ACT).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-3-4","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.3-4","text":"vegandocs deprecated. Current R provides better tools seeing extra documentation (news() browseVignettes()).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vignettes-2-3-4","dir":"Changelog","previous_headings":"","what":"Vignettes","title":"vegan 2.3-4","text":"vignettes built standard R tools can browsed browseVignettes. FAQ-vegan partitioning accessible vegandocs function.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"building-2-3-4","dir":"Changelog","previous_headings":"","what":"Building","title":"vegan 2.3-4","text":"Dependence external software texi2dvi removed. Version 6.1 texi2dvi incompatible R prevented building vegan. FAQ-vegan earlier built texi2dvi uses now knitr. , vegan now dependent R-3.0.0. Fixes issue #158 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-3","dir":"Changelog","previous_headings":"","what":"vegan 2.3-3","title":"vegan 2.3-3","text":"CRAN release: 2016-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-3","text":"metaMDS monoMDS fail input dissimilarities huge: reported case magnitude 1E85. Fixes issue #152 Github. Permutations failed defined permute control structures estaccum, ordiareatest, renyiaccum tsallisaccum. Reported Dan Gafta (Cluj-Napoca) renyiaccum. rarefy gave false warnings input vector single sampling unit. extrapolated richness indices specpool needed number doubletons (= number species occurring two sampling units), failed one sampling unit supplied. extrapolated richness estimated single sampling unit, now cases handled smoothly instead failing: observed non-extrapolated richness zero standard error reported. issue reported StackOverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-3","text":"treedist treedive refuse handle trees reversals, .e, higher levels homogeneous lower levels. Function treeheight estimate total height absolute values branch lengths. Function treedive refuses handle trees negative branch heights indicating negative dissimilarities. Function treedive faster. gdispweight works input data matrix instead data frame. Input dissimilarities supplied symmetric matrices data frames robustly recognized anosim, bioenv mrpp.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-2","dir":"Changelog","previous_headings":"","what":"vegan 2.3-2","title":"vegan 2.3-2","text":"CRAN release: 2015-11-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-2","text":"Printing details gridded permutation design fail grid within-plot level. ordicluster joined branches wrong coordinates cases. ordiellipse ignored weights calculating standard errors (kind = \"se\"). influenced plots cca, also influenced ordiareatest.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-2","text":"adonis capscale functions recognize symmetric square matrices dissimilarities. Formerly dissimilarities given \"dist\" objects produced dist vegdist functions, data frames matrices regarded observations x variables data confuse users (e.g., issue #147). mso accepts \"dist\" objects distances among locations alternative coordinates locations. text, points lines functions procrustes analysis gained new argument truemean allows adding procrustes items plots original analysis. rrarefy returns observed non-rarefied communities (warning) users request subsamples larger observed community instead failing. Function drarefy similar returned sampling probabilities 1, now also issues warning. Fixes issue #144 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-1","dir":"Changelog","previous_headings":"","what":"vegan 2.3-1","title":"vegan 2.3-1","text":"CRAN release: 2015-09-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-1","text":"Permutation tests always correctly recognize ties observed statistic result low P-values. happen particular predictor variables factors (classes). changes concern functions adonis, anosim, anova permutest functions cca, rda capscale, permutest betadisper, envfit, mantel mantel.partial, mrpp, mso, oecosimu, ordiareatest, protest simper. also fixes issues #120 #132 GitHub. Automated model building constrained ordination (cca, rda, capscale) step, ordistep ordiR2step fail aliased candidate variables, constraints completely explained variables already model. regression introduced vegan 2.2-0. Constrained ordination methods cca, rda capscale treat character variables factors analysis, return centroids plotting. Recovery original data metaMDS computing WA scores species fail expression supplied argument comm long & got deparsed multiple strings. metaMDSdist now returns (possibly modified) data frame community data comm attribute \"comm\" returned dist object. metaMDS now uses compute WA species scores NMDS. addition, deparsed expression comm now robust long expressions. Reported Richard Telford. metaMDS monoMDS rejected dissimilarities missing values. Function rarecurve check input cause confusing error messages. Now function checks input data integers can interpreted counts individuals sampling units species. Unchecked bad inputs reason problems reported Stackoverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-1","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-1","text":"Scaling ordination axes cca, rda capscale can now expressed descriptive strings \"none\", \"sites\", \"species\" \"symmetric\" tell kind scores scaled eigenvalues. can modified arguments hill cca correlation rda. old numeric scaling can still used. permutation data can extracted anova results constrained ordination (cca, rda, capscale) analysed permustats function. New data set BCI.env site information Barro Colorado Island tree community data. useful variables UTM coordinates sample plots. variables constant nearly constant little use normal analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-0","dir":"Changelog","previous_headings":"","what":"vegan 2.3-0","title":"vegan 2.3-0","text":"CRAN release: 2015-05-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-0","text":"Constrained ordination functions cca, rda capscale now robust. Scoping data set names variable names much improved. fix numerous long-standing problems, instance reported Benedicte Bachelot (email) Richard Telford (Twitter), well issues #16 #100 GitHub. Ordination functions cca rda silently accepted dissimilarities input although analysis makes sense methods. Dissimilarities analysed distance-based redundancy analysis (capscale). variance conditional component -estimated goodness rda results, results wrong partial RDA. problems reported R-sig-ecology message Christoph von Redwitz.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"windows-2-3-0","dir":"Changelog","previous_headings":"","what":"Windows","title":"vegan 2.3-0","text":"orditkplot add file type identifier saved graphics Windows although required. problem concerned Windows OS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-0","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-0","text":"goodness function constrained ordination (cca, rda, capscale) redesigned. Function gained argument addprevious add variation explained previous ordination components axes statistic = \"explained\". option, model = \"CCA\" include variation explained partialled-conditions, model = \"CA\" include accumulated variation explained conditions constraints. former behaviour addprevious = TRUE model = \"CCA\", addprevious = FALSE model = \"CA\". argument effect statistic = \"distance\", always show residual distance previous components. Formerly displayed residual distance currently analysed model. Functions ordiArrowMul ordiArrowTextXY exported can used normal interactive sessions. functions used scale bunch arrows fit ordination graphics, formerly internal functions used within vegan functions. orditkplot can export graphics SVG format. SVG vector graphics format can edited several external programs, Illustrator Inkscape. Rarefaction curve (rarecurve) species accumulation models (specaccum, fitspecaccum) gained new functions estimate slope curve given location. Originally based response R-SIG-ecology query. rarefaction curves, function rareslope, species accumulation models specslope. functions based analytic equations, can also evaluated interpolated non-integer values. specaccum models functions can evaluated analytic models \"exact\", \"rarefaction\" \"coleman\". \"random\" \"collector\" methods can use finite differences (diff(fitted())). Analytic functions slope used non-linear regression models known fitspecaccum. Species accumulation models (specaccum) non-liner regression models species accumulation (fitspecaccum) work consistently weights. cases, models defined using number sites independent variable, weights means observations can non-integer numbers virtual sites. predict models also use number sites newdata, analytic models can estimate expected values non-integer number sites, non-analytic randomized collector models can interpolate non-integer values. fitspecaccum gained support functions AIC deviance. varpart plots four-component models redesigned following Legendre, Borcard & Roberts Ecology 93, 1234–1240 (2012), use now four ellipses instead three circles two rectangles. components now labelled plots, circles ellipses can easily filled transparent background colour.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-1","dir":"Changelog","previous_headings":"","what":"vegan 2.2-1","title":"vegan 2.2-1","text":"CRAN release: 2015-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-1","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-1","text":"maintenance release avoid warning messages caused changes CRAN repository. namespace usage also stringent avoid warnings notes development versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-1","text":"vegan can installed loaded without tcltk package. tcltk package needed orditkplot function interactive editing ordination graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-1","text":"ordisurf failed gam package loaded due namespace issues: support functions gam used instead mgcv functions. tolerance function failed unconstrained correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-1","text":"estimateR uses exact variance formula bias-corrected Chao estimate extrapolated number species. new formula may unpublished, derived following guidelines Chiu, Wang, Walther & Chao, Biometrics 70, 671–682 (2014), doi:10.1111/biom.12200, online supplementary material. Diversity accumulation functions specaccum, renyiaccum, tsallisaccum, poolaccum estaccumR use now permute package permutations order sampling sites. Normally functions need simple random permutation sites, restricted permutation permute package user-supplied permutation matrices can used. estaccumR function can use parallel processing. linestack accepts now expressions labels. allows using mathematical symbols formula given mathematical expressions.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-0","dir":"Changelog","previous_headings":"","what":"vegan 2.2-0","title":"vegan 2.2-0","text":"CRAN release: 2014-11-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-0","text":"Several vegan functions can now use parallel processing slow repeating calculations. functions argument parallel. argument can integer giving number parallel processes. unix-alikes (Mac OS, Linux) launch \"multicore\" processing Windows set \"snow\" clusters desribed documentation parallel package. option \"mc.cores\" set integer > 1, used automatically start parallel processing. Finally, argument can also previously set \"snow\" cluster used Windows unix-alikes. Vegan vignette Design decision explains implementation (use vegandocs(\"decission\"), parallel package extensive documentation parallel processing R. following function use parallel processing analysing permutation statistics: adonis, anosim, anova.cca (permutest.cca), mantel (mantel.partial), mrpp, ordiareatest, permutest.betadisper simper. addition, bioenv can compare several candidate sets models paralle, metaMDS can launch several random starts parallel, oecosimu can evaluate test statistics several null models parallel. permutation tests based permute package offers strong tools restricted permutation. functions argument permutations. default usage simple non-restricted permutations achieved giving single integer number. Restricted permutations can defined using function permute package. Finally, argument can permutation matrix rows define permutations. possible use external user constructed permutations. See help(permutations) brief introduction permutations vegan, permute package full documention. vignette permute package can read vegan command vegandocs(\"permutations\"). following functions use permute package: CCorA, adonis, anosim, anova.cca (plus associated permutest.cca, add1.cca, drop1.cca, ordistep, ordiR2step), envfit (plus associated factorfit vectorfit), mantel (mantel.partial), mrpp, mso, ordiareatest, permutest.betadisper, protest simper. Community null model generation completely redesigned rewritten. communities constructed new nullmodel function defined low level commsim function. actual null models generated simulate function builds array null models. new null models include wide array quantitative models addition old binary models, users can plug generating functions. basic tool invoking analysing null models oecosimu. null models often used analysis nestedness, implementation oecosimu allows analysing statistic, null models better seen alternative permutation tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-0","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-0","text":"vegan package dependencies namespace imports adapted changes R, trigger warnings notes package tests. Three-dimensional ordination graphics using scatterplot3d static plots rgl dynamic plots removed vegan moved companion package vegan3d. package available CRAN.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-2-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.2-0","text":"Function dispweight implements dispersion weighting Clarke et al. (Marine Ecology Progress Series, 320, 11–27). addition, implemented new method generalized dispersion weighting gdispweight. methods downweight species significantly -dispersed. New hclust support functions reorder, rev scores. Functions reorder rev similar functions dendrogram objects base R. However, reorder can use (defaults ) weighted mean. weighted mean node average always mean member leaves, whereas dendrogram uses always unweighted means joined branches. Function ordiareatest supplements ordihull ordiellipse provides randomization test one-sided alternative hypothesis convex hulls ellipses two-dimensional ordination space smaller areas randomized groups. Function permustats extracts inspects permutation results support functions summary, density, densityplot, qqnorm qqmath. density qqnorm standard R tools work one statistic, densityplot qqmath lattice graphics work univariate multivariate statistics. results following functions can extracted: anosim, adonis, mantel (mantel.partial), mrpp, oecosimu, permustest.cca (corresponding anova methods), permutest.betadisper, protest. stressplot functions display ordination distances given number dimensions original distances. method functins similar stressplot metaMDS, always use inherent distances ordination method. functions available results capscale, cca, princomp, prcomp, rda, wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-0","text":"cascadeKM one group NA instead random value. ordiellipse can handle points exactly line, including two points (warning). plotting radfit results several species failed communities species one species. RsquareAdj capscale negative eigenvalues now report NA instead using biased method rda results. simper failed group single member.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-0","text":"anova.cca functions re-written use permute package. Old results may exactly reproduced, models missing data may fail several cases. new option analysing sequence models . simulate functions cca rda can return several simulations nullmodel compatible object. functions can produce simulations correlated errors (also capscale) parametric simulation Gaussian error. bioenv can use Manhattan, Gower Mahalanobis distances addition default Euclidean. New helper function bioenvdist can extract dissimilarities applied best model model. metaMDS(..., trace = 2) show convergence information default monoMDS engine. Function MDSrotate can rotate k-dimensional ordination k-1 variables. variables correlated (like usually case), vectors can also correlated previously rotated dimensions, uncorrelated later ones. vegan 2.0-10 changed weighted nestednodf weighted analysis binary data equivalent binary analysis. However, broke equivalence original method. Now function argument wbinary select method analysis. problem reported fix submitted Vanderlei Debastiani (Universidade Federal Rio Grande Sul, Brasil). ordiellipse, ordihull ordiellipse can handle missing values groups. ordispider can now use spatial medians instead means. rankindex can use Manhattan, Gower Mahalanobis distance addition default Euclidean. User can set colours line types function rarecurve plotting rarefaction curves. spantree gained support function .hclust change minimum spanning tree hclust tree. fitspecaccum can weighted analysis. Gained lines method. Functions extrapolated number species size species pool using Chao method modified following Chiu et al., Biometrics 70, 671–682 (2014). Incidence based specpool can now use (defaults ) small sample correction number sites sample size. Function uses basic Chao extrapolation based ratio singletons doubletons, switches now bias corrected Chao extrapolation doubletons (species found twice). variance formula bias corrected Chao derived following supporting line material doi:10.1111/biom.12200 differs slightly Chiu et al. (2014). poolaccum function changed similarly, small sample correction used always. abundance based estimateR uses bias corrected Chao extrapolation, earlier estimated variance classic Chao model. Now use widespread approximate estimate EstimateS variance. changes functions similar EstimateS tabasco uses now reorder.hclust hclust object better ordering previously cast trees dendrogram objects. treedive treedist default now match.force = TRUE can silenced verbose = FALSE. vegdist gained Mahalanobis distance. Nomenclature updated plant community data help Taxonstand taxize packages. taxonomy dune data adapted sources APG III. varespec dune use 8-character names (4 genus + 4 species epithet). New data set phylogenetic distances dune extracted Zanne et al. (Nature 506, 89–92; 2014). User configurable plots rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-2-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.2-0","text":"strata deprecated permutations. still accepted phased next releases. Use permute package. cca, rda capscale return scores scaled eigenvalues: use scores function extract scaled results. commsimulator deprecated. Replace commsimulator(x, method) simulate(nullmodel(x, method)). density densityplot permutation results deprecated: use permustats density densityplot method.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-10","dir":"Changelog","previous_headings":"","what":"vegan 2.0-10","title":"vegan 2.0-10","text":"CRAN release: 2013-12-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-10","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-10","text":"version adapted changes permute package version 0.8-0 triggers NOTEs package checks. release may last 2.0 series, next vegan release scheduled major release newly designed oecosimu community pattern simulation, support parallel processing, full support permute package. interested developments, may try development versions vegan GitHub report problems user experience us.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-10","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-10","text":"envfit function assumed external variables either numeric factors, failed , say, character strings. Now numeric variables taken continuous vectors, variables (character strings, logical) coerced factors possible. function also work degenerate data, like one level factor constant value continuous environmental variable. ties wrongly assessing permutation P-values vectorfit. nestednodf quantitative data consistent binary models, fill wrongly calculated quantitative data. oecosimu now correctly adapts displayed quantiles simulated values alternative test direction. renyiaccum plotting failed one level diversity scale used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-10","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-10","text":"Kempton Taylor algorithm found unreliable fisherfit fisher.alpha, now estimation Fisher α based number species number individuals. estimation standard errors profile confidence intervals also scrapped. renyiaccum, specaccum tsallisaccum functions gained subset argument. renyiaccum can now add collector curve analysis. collector curve diversity accumulation order sampling units. interesting ordering sampling units allows comparing actual species accumulations expected randomized accumulation. specaccum can now perform weighted accumulation using sampling effort weights.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-9","dir":"Changelog","previous_headings":"","what":"vegan 2.0-9","title":"vegan 2.0-9","text":"CRAN release: 2013-09-25 version released due changes programming interface testing procedures R 3.0.2. using older version R, need upgrade vegan. new features bug fixes. user-visible changes documentation output messages formatting. R changes, version dependent R version 2.14.0 newer lattice package.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-8","dir":"Changelog","previous_headings":"","what":"vegan 2.0-8","title":"vegan 2.0-8","text":"CRAN release: 2013-07-10","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-8","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-8","text":"maintenance release fixes issues raised changed R toolset processing vignettes. also fix typographic issues vignettes.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-8","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-8","text":"ordisurf gained new arguments flexible definition fitted models better utilize mgcv::gam function. linewidth contours can now set argument lwd. Labels arrows positioned better way plot functions results envfit, cca, rda capscale. labels longer overlap arrow tips. setting test direction clearer oecosimu. ordipointlabel gained plot method can used replot saved result.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-7","dir":"Changelog","previous_headings":"","what":"vegan 2.0-7","title":"vegan 2.0-7","text":"CRAN release: 2013-03-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-7","text":"tabasco() new function graphical display community data matrix. Technically interface R heatmap, use closer vegan function vegemite. function can reorder community data matrix similarly vegemite, instance, ordination results. Unlike heatmap, displays dendrograms supplied user, defaults re-order dendrograms correspondence analysis. Species ordered match site ordering like determined user.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-7","text":"Function fitspecaccum(..., model = \"asymp\") fitted logistic model instead asymptotic model (model = \"logis\"). nestedtemp() failed sparse data (fill < 0.38%).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-7","text":"plot function constrained ordination results (cca, rda, capscale) gained argument axis.bp (defaults TRUE) can used suppress axis scale biplot arrays. Number iterations nonmetric multidimensional scaling (NMDS) can set keyword maxit (defaults 200) metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-0-7","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.0-7","text":"result objects cca, rda capscale longer scores u.eig, v.eig wa.eig future versions vegan. change influence normal usage, vegan functions need items. However, external scripts packages may need changes future versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-6","dir":"Changelog","previous_headings":"","what":"vegan 2.0-6","title":"vegan 2.0-6","text":"CRAN release: 2013-02-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-6","text":"species scores scaled wrongly capscale(). scaled correctly Euclidean distances used, usually capscale() used non-Euclidean distances. graphics change redone. change scaling mainly influences spread species scores respect site scores. Function clamtest() failed set minimum abundance threshold cases. addition, output wrong possible species groups missing. problems reported Richard Telford (Bergen, Norway). Plotting object fitted envfit() fail p.max used unused levels one factors. unused levels result deletion observations missing values simply result supplying subset larger data set envfit(). multipart() printed wrong information analysis type (analysis correctly). Reported Valerie Coudrain. oecosimu() failed nestedfun returned data frame. fundamental fix vegan 2.2-0, structure oecosimu() result change. plot two-dimensional procrustes() solutions often draw original axes wrong angle. problem reported Elizabeth Ottesen (MIT). Function treedive() functional phylogenetic diversity correctly match species names community data species tree tree contained species occur data. Related function treedist() phylogenetic distances try match names .","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-6","text":"output capscale() displays value additive constant argument add = TRUE used. fitted() functions cca(), rda() capscale() can now return conditioned (partial) component response: Argument model gained new alternative model = \"pCCA\". dispindmorisita() output gained new column Chi-squared based probabilities null hypothesis (random distribution) true. metaMDS() monoMDS() new default convergence criteria. importantly, scale factor gradient (sfgrmin) stricter. former limit slack large data sets iterations stopped early without getting close solution. addition, scores() ignore now requests dimensions beyond calculated instead failing, scores() metaMDS() results drop dimensions. msoplot() gained legend argument positioning legend. Nestedness function nestednodf() gained plot method. ordiR2step() gained new argument R2scope (defaults TRUE) can used turn criterion stopping adjusted R2 current model exceeds scope. option allows model building scope overdetermined (number predictors higher number observations). ordiR2step() now handles partial redundancy analysis (pRDA). orditorp() gained argument select select rows columns results display. protest() prints standardized residual statistic squared m12 addition squared Procrustes correlation R2. calculated, latter displayed. Permutation tests much faster protest(). Instead calling repeatedly procrustes(), goodness fit statistic evaluated within function. wcmdscale() gained methods print, plot etc. results. methods used full wcmdscale result returned , e.g., argument eig = TRUE. default still return matrix scores similarly standard R function cmdscale(), case new methods used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-5","dir":"Changelog","previous_headings":"","what":"vegan 2.0-5","title":"vegan 2.0-5","text":"CRAN release: 2012-10-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-5","text":"anova(, ...) failed = \"axis\" = \"term\". bug reported Dr Sven Neulinger (Christian Albrecht University, Kiel, Germany). radlattice honour argument BIC = TRUE, always displayed AIC.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-5","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-5","text":"vegan functions permutation tests now density method can used find empirical probability distributions permutations. new plot method functions displays density observed statistic. density function available adonis, anosim, mantel, mantel.partial, mrpp, permutest.cca procrustes. Function adonis can return several statistics, now densityplot method (based lattice). Function oecosimu already density densityplot, now similar vegan methods, also work adipart, hiersimu multipart. radfit functions got predict method also accepts arguments newdata total new ranks site totals prediction. functions can also interpolate non-integer “ranks”, models also extrapolate.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-5","text":"Labels can now set plot envfit results. labels must given order function uses internally, new support function labels can used display default labels correct order. Mantel tests (functions mantel mantel.partial) gained argument na.rm can used remove missing values. options used care: Permutation tests can biased missing values originally matching fixed positions. radfit results can consistently accessed methods whether single model single site, models single site models sites data. functions now methods AIC, coef, deviance, logLik, fitted, predict residuals.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-0-5","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.0-5","text":"Building vegan vignettes failed latest version LaTeX (TeXLive 2012). R versions later 2.15-1 (including development version) report warnings errors installing checking vegan, must upgrade vegan version. warnings concern functions cIndexKM betadisper, error occurs betadisper. errors warnings triggered internal changes R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-4","dir":"Changelog","previous_headings":"","what":"vegan 2.0-4","title":"vegan 2.0-4","text":"CRAN release: 2012-06-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-4","text":"adipart assumed constant gamma diversity simulations assessing P-value. give biased results null model produces variable gamma diversities option weights = \"prop\" used. default null model (\"r2dtable\") default option (weights = \"unif\") analysed correctly. anova(, = \"axis\") cases failed due ‘NAMESPACE’ issues. clamtest wrongly used frequencies instead counts calculating sample coverage. detectable differences produced rerunning examples Chazdon et al. 2011 vegan help page. envfit failed unused factor levels. predict cca results type = \"response\" type = \"working\" failed newdata number rows match original data. Now newdata ignored wrong number rows. number rows must match results cca must weighted original row totals. problem concern rda capscale results need row weights. Reported Glenn De’ath.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-4","text":"Functions diversity partitioning (adipart, hiersimu multipart) now formula default methods. formula method identical previous functions, default method can take two matrices input. Functions adipart multipart can used fast easy overall partitioning alpha, beta gamma diversities omitting argument describing hierarchy. method betadisper biased small sample sizes. effects bias strongest unequal sample sizes. bias adjusted version developed Adrian Stier Ben Bolker, can invoked argument bias.adjust (defaults FALSE). bioenv accepts dissimilarities (square matrices can interpreted dissimilarities) alternative community data. allows using dissimilarities available vegdist. plot function envfit results gained new argument bg can used set background colour plotted labels. msoplot configurable, allows, instance, setting y-axis limits. Hulls ellipses now filled using semitransparent colours ordihull ordiellipse, user can set degree transparency new argument alpha. filled shapes used functions called argument draw = \"polygon\". Function ordihull puts labels (argument label = TRUE) now real polygon centre. ordiplot3d returns function envfit.convert projected location origin. Together can used add envfit results existing ordiplot3d plots. Equal aspect ratio set exactly ordiplot3d underlying core routines allow . Now ordiplot3d sets equal axis ranges, documents urge users verify aspect ratio reasonably equal graph looks like cube. problems solved future, ordiplot3d may removed next releases vegan. Function ordipointlabel gained argument select items plotting. argument can used one set points.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-3","dir":"Changelog","previous_headings":"","what":"vegan 2.0-3","title":"vegan 2.0-3","text":"CRAN release: 2012-03-03","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-3","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-3","text":"Added new nestedness functions nestedbetasor nestedbetajac implement multiple-site dissimilarity indices decomposition turnover nestedness components following Baselga (Global Ecology Biogeography 19, 134–143; 2010). Added function rarecurve draw rarefaction curves row (sampling unit) input data, optionally lines showing rarefied species richness given sample size curve. Added function simper implements “similarity percentages” Clarke (Australian Journal Ecology 18, 117–143; 1993). method compares two groups decomposes average -group Bray-Curtis dissimilarity index contributions individual species. code developed GitHub Eduard Szöcs (Uni Landau, Germany).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-3","text":"betadisper() failed groups factor empty levels. constrained ordination methods support functions robust border cases (completely aliased effects, saturated models, user requests non-existng scores etc). Concerns capscale, ordistep, varpart, plot function constrained ordination, anova(, = \"margin\"). scores function monoMDS honour choices argument hence dimensions chosen plot. default scores method failed number requested axes higher ordination object . reported error ordiplot R-sig-ecology mailing list.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-3","text":"metaMDS argument noshare = 0 now regarded numeric threshold always triggers extended dissimilarities (stepacross), instead treated synonymous noshare = FALSE always suppresses extended dissimilarities. Nestedness discrepancy index nesteddisc gained new argument allows user set number iterations optimizing index. oecosimu displays mean simulations describes alternative hypothesis clearly printed output. Implemented adjusted R2 partial RDA. partial model rda(Y ~ X1 + Condition(X2)) component [] = X1|X2 variance partition varpart describes marginal (unique) effect constraining term adjusted R2. Added Cao dissimilarity (CYd) new dissimilarity method vegdist following Cao et al., Water Envir Res 69, 95–106 (1997). index good data high beta diversity variable sampling intensity. Thanks consultation Yong Cao (Univ Illinois, USA).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-2","dir":"Changelog","previous_headings":"","what":"vegan 2.0-2","title":"vegan 2.0-2","text":"CRAN release: 2011-11-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-2","text":"Function capscale failed constrained component zero rank. happened likely partial models conditions aliased constraints. problem observed anova(..., =\"margin\") uses partial models analyses marginal effects, reported email message R-News mailing list. stressplot goodness sometimes failed metaMDS based isoMDS (MASS package) metaMDSdist use defaults step-across (extended) dissimilarities metaMDS(..., engine = \"isoMDS\"). change defaults can also influence triggering step-across capscale(..., metaMDSdist = TRUE). adonis contained minor bug resulting incomplete implementation speed-affect results. fixing bug, bug identified transposing hat matrices. second bug active following fixing first bug. fixing bugs, speed-internal f.test() function fully realised. Reported Nicholas Lewin-Koh.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-2","text":"ordiarrows ordisegments gained argument order.gives variable sort points within groups. Earlier points assumed order. Function ordispider invisibly returns coordinates points connected. Typically class centroids point, constrained ordination groups LC scores.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-1","dir":"Changelog","previous_headings":"","what":"vegan 2.0-1","title":"vegan 2.0-1","text":"CRAN release: 2011-10-20","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-1","text":"clamtest: new function classify species generalists specialists two distinct habitats (CLAM test Chazdon et al., Ecology 92, 1332–1343; 2011). test based multinomial distribution individuals two habitat types sampling units, applicable count data -dispersion. .preston gained plot lines methods, .fisher gained plot method (also can add items existing plots). similar plot lines prestonfit fisherfit, display data without fitted lines. raupcrick: new function implement Raup-Crick dissimilarity probability number co-occurring species occurrence probabilities proportional species frequencies. Vegan Raup-Crick index choice vegdist, uses equal sampling probabilities species analytic equations. new raupcrick function uses simulation oecosimu. function follows Chase et al. (2011) Ecosphere 2:art24 [doi:10.1890/ES10-00117.1], developed consultation Brian Inouye.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-1","text":"Function meandist scramble items give wrong results, especially grouping numerical. problem reported Dr Miguel Alvarez (Univ. Bonn). metaMDS reset tries new model started previous.best solution different model. Function permatswap community null models using quantitative swap never swapped items 2x2 submatrix cells filled. result permutest.cca updated ‘NAMESPACE’ issue. R 2.14.0 changed accept using sd() function matrices (behaviour least since R 1.0-0), several vegan functions changed adapt change (rda, capscale, simulate methods rda, cca capscale). change R 2.14.0 influence results probably wish upgrade vegan avoid annoying warnings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-1","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-1","text":"nesteddisc slacker hence faster trying optimize statistic tied column frequencies. Tracing showed cases improved ordering found rather early tries, results equally good cases.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-0","dir":"Changelog","previous_headings":"","what":"vegan 2.0-0","title":"vegan 2.0-0","text":"CRAN release: 2011-09-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-0","text":"Peter Minchin joins vegan team. vegan implements standard R ‘NAMESPACE’. general, S3 methods exported means directly use see contents functions like cca.default, plot.cca anova.ccabyterm. use functions rely R delegation simply use cca result objects use plot anova without suffix .cca. see contents function can use :::, vegan:::cca.default. change may break packages, documents scripts rely non-exported names. vegan depends permute package. package provides powerful tools restricted permutation schemes. vegan permutation gradually move use permute, currently betadisper uses new feature.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-0","text":"monoMDS: new function non-metric multidimensional scaling (NMDS). function replaces MASS::isoMDS default method metaMDS. Major advantages monoMDS ‘weak’ (‘primary’) tie treatment means can split tied observed dissimilarities. ‘Weak’ tie treatment improves ordination heterogeneous data sets, maximum dissimilarities 1 can split. addition global NMDS, monoMDS can perform local hybrid NMDS metric MDS. can also handle missing zero dissimilarities. Moreover, monoMDS faster previous alternatives. function uses Fortran code written Peter Minchin. MDSrotate new function replace metaMDSrotate. function can rotate metaMDS monoMDS results first axis parallel environmental vector. eventstar finds minimum evenness profile Tsallis entropy, uses find corresponding values diversity, evenness numbers equivalent following Mendes et al. (Ecography 31, 450-456; 2008). code contributed Eduardo Ribeira Cunha Heloisa Beatriz Antoniazi Evangelista adapted vegan Peter Solymos. fitspecaccum fits non-linear regression models species accumulation results specaccum. function can use new self-starting species accumulation models vegan self-starting non-linear regression models R. function can fit Arrhenius, Gleason, Gitay, Lomolino (vegan), asymptotic, Gompertz, Michaelis-Menten, logistic Weibull (base R) models. function plot predict methods. Self-starting non-linear species accumulation models SSarrhenius, SSgleason, SSgitay SSlomolino. can used fitspecaccum directly non-linear regression nls. functions implemented found good species-area models Dengler (J. Biogeogr. 36, 728-744; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-0","text":"adonis, anosim, meandist mrpp warn negative dissimilarities, betadisper refuses analyse . functions expect dissimilarities, giving something else (like correlations) probably user error. betadisper uses restricted permutation permute package. metaMDS uses monoMDS default ordination engine. Function gains new argument engine can used alternatively select MASS::isoMDS. default use stepacross monoMDS ‘weak’ tie treatment can cope tied maximum dissimilarities one. However, stepacross default isoMDS handle adequately tied maximum dissimilarities. specaccum gained predict method uses either linear spline interpolation data observed points. Extrapolation possible spline interpolation, may make little sense. specpool can handle missing values empty factor levels grouping factor pool. Now also checks length pool matches number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-0-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.0-0","text":"metaMDSrotate replaced MDSrotate can also handle results monoMDS. permuted.index2 “new” permutation code removed favour permute package. code intended normal use, packages depending code vegan instead depend permute.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-0","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-0","text":"treeheight uses much snappier code. results unchanged.","code":""}]
# S3 method for cca
plot(x, choices = c(1, 2), display = c("sp", "wa", "cn"),
scaling = "species", type, xlim, ylim, const,
- correlation = FALSE, hill = FALSE, ...)
+ correlation = FALSE, hill = FALSE, cex = 0.7, ...)
# S3 method for cca
text(x, display = "sites", labels, choices = c(1, 2),
scaling = "species", arrow.mul, head.arrow = 0.05, select, const,
@@ -199,6 +199,9 @@ ArgumentsDetails
-If you want to have a better control of plots, it is best to
- construct the plot text and points commands which
- accept graphical parameters. It is important to remember to use the
- same scaling, correlation and hill arguments
- in all calls. The plot.cca command returns invisibly an
+
If you want to have a better control of plots, it is best to construct
+ the plot text and points commands which accept graphical
+ parameters. It is important to remember to use the same
+ scaling, correlation and hill arguments in all
+ calls. The plot.cca command returns invisibly an
ordiplot result object, and this will have consistent
- scaling for all its elements. The easiest way for full control of
- graphics is to first set up the plot frame using plot with
- type = "n" and all needed scores in display and save
- this result. The points and text commands for
- ordiplot will allow full graphical control (see
- section Examples). Utility function labels returns the default
- labels in the order they are applied in text.
+ scaling for all its elements. It is also possible to use R pipes
+ (|>) which maintain consistent scaling and setting. The easiest
+ way for full control of graphics is to first set up the plot frame
+ using plot with type = "n" and all needed scores in
+ display and save this result or use it as a first command in a
+ pipe. The points and text commands for
+ ordiplot will allow full graphical control (see section
+ Examples). Utility function labels returns the default labels
+ in the order they are applied in text.
Palmer (1993) suggested using linear constraints (“LC scores”)
in ordination diagrams, because these gave better results in
simulations and site scores (“WA scores”) are a step from
@@ -305,8 +310,14 @@
Examplespl <- plot(pca, type="n", scaling="sites", correlation=TRUE)
with(dune.env, points(pl, "site", pch=21, col=1, bg=Management))
text(pl, "sp", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE)
-with(dune.env, legend("bottomleft", levels(Management), pch=21, pt.bg=1:4, bty="n"))
+with(dune.env, legend("bottomleft", levels(Management), pch=21,
+ pt.bg=1:4, bty="n"))
+## Configuration with pipes: make an arrow biplot
+plot(pca, type="n", scaling="sites", correlation=TRUE) |>
+ points("sites", pch=21, col = 1, cex=1.5, bg = dune.env$Management) |>
+ text("species", col = "blue", arrow = TRUE, xpd = TRUE)
+
## Scaling can be numeric or more user-friendly names
## e.g. Hill's scaling for (C)CA
scrs <- scores(mod, scaling = "sites", hill = TRUE)
diff --git a/docs/reference/vegan-package-3.png b/docs/reference/vegan-package-3.png
index 5b6055386..266e97ddc 100644
Binary files a/docs/reference/vegan-package-3.png and b/docs/reference/vegan-package-3.png differ
diff --git a/docs/reference/vegan-package-4.png b/docs/reference/vegan-package-4.png
index 6ed16b20b..c9cb6bd9a 100644
Binary files a/docs/reference/vegan-package-4.png and b/docs/reference/vegan-package-4.png differ
diff --git a/docs/search.json b/docs/search.json
index 8afe3f072..3d0395916 100644
--- a/docs/search.json
+++ b/docs/search.json
@@ -1 +1 @@
-[{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"vegan-faq","dir":"Articles","previous_headings":"","what":"vegan FAQ","title":"","text":"document contains answers frequently asked questions R package vegan. work licensed Creative Commons Attribution 3.0 License. view copy license, visit https://creativecommons.org/licenses//3.0/ send letter Creative Commons, 543 Howard Street, 5th Floor, San Francisco, California, 94105, USA. Copyright © 2008-2016 vegan development team","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is vegan?","title":"","text":"Vegan R package community ecologists. contains popular methods multivariate analysis needed analysing ecological communities, tools diversity analysis, potentially useful functions. Vegan self-contained must run R statistical environment, also depends many R packages. Vegan free software distributed GPL2 license.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is R?","title":"","text":"R system statistical computation graphics. consists language plus run-time environment graphics, debugger, access certain system functions, ability run programs stored script files. R home page https://www.R-project.org/. free software distributed GNU-style copyleft, official part GNU project (“GNU S”).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-obtain-vegan-and-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to obtain vegan and R?","title":"","text":"R latest release version vegan can obtained CRAN. Unstable development version vegan can obtained GitHub. github page gives instructions obtaining installing development versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-r-packages-vegan-depends-on","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What R packages vegan depends on?","title":"","text":"Vegan depends permute package provide advanced flexible permutation routines vegan. permute package developed together vegan GitHub. individual vegan functions depend packages MASS, mgcv, parallel, cluster lattice. Vegan dependence tcltk deprecated removed future releases. base recommended R packages available every R installation. Vegan declares suggested imported packages, can install vegan use functions without packages. Vegan accompanied supporting package vegan3d three-dimensional dynamic plotting. vegan3d package needs tcltk non-standard packages rgl scatterplot3d.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-packages-are-available-for-ecologists","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other packages are available for ecologists?","title":"","text":"CRAN Task Views include entries like Environmetrics, Multivariate Spatial describe several useful packages functions. install R package ctv, can inspect Task Views R session, automatically install sets important packages.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-documentation-is-available-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other documentation is available for vegan?","title":"","text":"Vegan fully documented R package standard help pages. authoritative sources documentation (last resource can use force read source, vegan open source). Vegan package ships documents can read browseVignettes(\"vegan\") command. documents included vegan package Vegan NEWS can accessed via news() command. document (FAQ-vegan). Short introduction basic ordination methods vegan (intro-vegan). Introduction diversity methods vegan (diversity-vegan). Discussion design decisions vegan (decision-vegan). Description variance partition procedures function varpart (partitioning). Web documents outside package include: https://github.com/vegandevs/vegan: development page. https://vegandevs.github.io/vegan/: vegan homepage.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-a-graphical-user-interface-gui-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is there a Graphical User Interface (GUI) for vegan?","title":"","text":"Roeland Kindt made package BiodiversityR provides GUI vegan. package available CRAN. mere GUI vegan, adds new functions complements vegan functions order provide workbench biodiversity analysis. can install BiodiversityR using install.packages(\"BiodiversityR\") graphical package management menu R. GUI works Windows, MacOS X Linux.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-cite-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to cite vegan?","title":"","text":"Use command citation(\"vegan\") R see recommended citation used publications.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-build-vegan-from-sources","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to build vegan from sources?","title":"","text":"general, need build vegan sources, binary builds release versions available CRAN Windows MacOS X. use operating systems, may use source packages. Vegan standard R package, can built like instructed R documentation. Vegan contains source files C FORTRAN, need appropriate compilers (may need work Windows MacOS X).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"are-there-binaries-for-devel-versions","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Are there binaries for devel versions?","title":"","text":"Binaries can available R Universe: see https://github.com/vegandevs/vegan instructions.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-report-a-bug-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to report a bug in vegan?","title":"","text":"think found bug vegan, report vegan maintainers developers. preferred forum report bugs GitHub. bug report detailed bug can replicated corrected. Preferably, send example causes bug. needs data set available R, send minimal data set well. also paste output error message message. also specify version vegan used. Bug reports welcome: way make vegan non-buggy. Please note shall send bug reports R mailing lists, since vegan standard R package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-a-bug-or-a-feature","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is it a bug or a feature?","title":"","text":"necessarily bug function gives different results expect: may deliberate design decision. may useful check documentation function see intended behaviour. may also happen function argument switch behaviour match expectation. instance, function vegdist always calculates quantitative indices (possible). expect calculate binary index, use argument binary = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-contribute-to-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Can I contribute to vegan?","title":"","text":"Vegan dependent user contribution. feedback welcome. problems vegan, may simple incomplete documentation, shall best improve documents. Feature requests also welcome, necessarily fulfilled. new feature added easy looks useful, submit code. can write code , best forum contribute vegan GitHub.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-only-numeric-and-positive-data-but-vegan-still-complains","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have only numeric and positive data but vegan still complains","title":"","text":"wrong! Computers painfully pedantic, find non-numeric negative data entries, really . Check data! common reasons non-numeric data row names read non-numeric variable instead used row names (check argument row.names reading data), column names interpreted data (check argument header = TRUE reading data). Another common reason empty cells input data, interpreted missing values.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-analyse-binary-or-cover-class-data","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I analyse binary or cover class data?","title":"","text":"Yes. vegan methods can handle binary data cover abundance data. statistical tests based permutation, make distributional assumptions. methods (mainly diversity analysis) need count data. methods check input data integers, may fooled cover class data.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-dissimilarities-in-vegan-differ-from-other-sources","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why dissimilarities in vegan differ from other sources?","title":"","text":"commonly reason software use presence–absence data whereas vegan used quantitative data. Usually vegan indices quantitative, can use argument binary = TRUE make presence–absence. However, index name cases, although different names usually occur literature. instance, Jaccard index actually refers binary index, vegan uses name \"jaccard\" quantitative index, . Another reason may indices indeed defined differently, people use names different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-nmds-stress-is-sometimes-0-1-and-sometimes-10","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why NMDS stress is sometimes 0.1 and sometimes 10?","title":"","text":"Stress proportional measure badness fit. proportions can expressed either parts one percents. Function isoMDS (MASS package) uses percents, function monoMDS (vegan package) uses proportions, therefore stress 100 times higher isoMDS. results goodness function also depend definition stress, goodness 100 times higher isoMDS monoMDS. conventions equally correct.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-get-zero-stress-but-no-repeated-solutions-in-metamds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I get zero stress but no repeated solutions in metaMDS","title":"","text":"first (try 0) run metaMDS starts metric scaling solution usually good, sofware return solution. However, metaMDS tries see standard solution can repeated, improved improved solution still repeated. cases, return best solution found, burning need anything get message tha solution repeated. keen know solution really global optimum, may follow instructions metaMDS help section “Results Repeated” try . common reason observations NMDS. n observations (points) k dimensions need estimate n*k parameters (ordination scores) using n*(n-1)/2 dissimilarities. k dimensions must n > 2*k + 1, two dimensions least six points. degenerate situations may need even larger number points. lower number points, can find undefined number perfect (stress zero) different solutions. Conventional wisdom due Kruskal n > 4*k + 1 points k dimensions. typical symptom insufficient data (nearly) zero stress two convergent solutions. cases reduce number dimensions (k) small data sets use NMDS, rely metric methods. seems local hybrid scaling monoMDS similar lower limits practice (although theoretically differ). However, higher number dimensions can used metric scaling, monoMDS principal coordinates analysis (cmdscale stats, wcmdscale vegan).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"zero-dissimilarities-in-isomds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Zero dissimilarities in isoMDS","title":"","text":"Function metaMDS uses function monoMDS default method NMDS, function can handle zero dissimilarities. Alternative function isoMDS handle zero dissimilarities. want use isoMDS, can use argument zerodist = \"add\" metaMDS handle zero dissimilarities. argument, zero dissimilarities replaced small positive value, can handled isoMDS. kluge, people like . principal solution remove duplicate sites using R command unique. However, standardizations dissimilarity indices, originally non-unique sites can zero dissimilarity, resort kluge (work harder data). Usually better use monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-heard-that-you-cannot-fit-environmental-vectors-or-surfaces-to-nmds-results-which-only-have-rank-order-scores","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have heard that you cannot fit environmental vectors or surfaces to NMDS results which only have rank-order scores","title":"","text":"Claims like indeed large Internet, based grave misunderstanding plainly wrong. NMDS ordination results strictly metric, vegan metaMDS monoMDS even strictly Euclidean. method called “non-metric” Euclidean distances ordination space non-metric rank-order relationship community dissimilarities. can inspect non-linear step curve using function stressplot vegan. ordination scores strictly Euclidean, correct use vegan functions envfit ordisurf NMDS results.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"where-can-i-find-numerical-scores-of-ordination-axes","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Where can I find numerical scores of ordination axes?","title":"","text":"Normally can use function scores extract ordination scores ordination method. scores function can also find ordination scores many non-vegan functions prcomp princomp ade4 functions. cases ordination result object stores raw scores, axes also scaled appropriate access scores. instance, cca rda ordination object -called normalized scores, scaled ordination plots use accessed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-the-rda-results-are-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How the RDA results are scaled?","title":"","text":"scaling RDA results indeed differ software packages. scaling RDA complicated issue explained FAQ, explained separate pdf document “Design decision implementation details vegan” can read command browseVignettes(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-cannot-print-and-plot-rda-results-properly","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I cannot print and plot RDA results properly","title":"","text":"RDA ordination results weird format plot properly, probably name clash klaR package also function rda, klaR print, plot predict functions used vegan RDA results. can choose rda functions using vegan::rda() klaR::rda(): get obscure error messages use wrong function. general, vegan able work normally vegan loaded klaR, klaR loaded later, functions take precedence vegan. Sometimes vegan namespace loaded automatically restoring previously stored workspace start-, klaR methods always take precedence vegan. check loaded packages. klaR may also loaded indirectly via packages (reported cases often loaded via agricolae package). Vegan klaR function name (rda), may possible use packages simultaneously, safest choice unload one packages possible. See discussion vegan github issues.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"ordination-fails-with-error-in-la-svd","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Ordination fails with “Error in La.svd”","title":"","text":"Constrained ordination (cca, rda, dbrda, capscale) sometimes fail error message Error La.svd(x, nu, nv): error code 1 Lapack routine 'dgesdd'. seems basic problem svd function LAPACK used numerical analysis R. LAPACK external library beyond control package developers R core team problems may unsolvable. Reducing range constraints (environmental variables) helps sometimes. instance, multiplying constraints constant < 1. rescaling influence numerical results constrained ordination, can complicate analyses values constraints needed, scaling must applied . reports problems getting rare may problem fixed R LAPACK.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"variance-explained-by-ordination-axes-","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Variance explained by ordination axes.","title":"","text":"general, vegan directly give statistics “variance explained” ordination axes constrained axes. design decision: think information normally useless often misleading. community ordination, goal typically explain variance, find “gradients” main trends data. “total variation” often meaningless, proportions meaningless values also meaningless. Often better solution explains smaller part “total variation”. instance, unstandardized principal components analysis variance generated small number abundant species, easy “explain” data really multivariate. standardize data, species equally important. first axes explains much less “total variation”, now explain species equally, results typically much useful whole community. Correspondence analysis uses another measure variation (variance), typically explains “smaller proportion” principal components better result. Detrended correspondence analysis nonmetric multidimensional scaling even try “explain” variation, use criteria. methods incommensurable, impossible compare methods using “explanation variation”. still want get “explanation variation” (deranged editor requests ), possible get information methods: Eigenvector methods: Functions rda, cca, dbrda capscale give variation conditional (partialled), constrained (canonical) residual components. Function eigenvals extracts eigenvalues, summary(eigenvals(ord)) reports proportions explained result object ord, also works decorana wcmdscale. Function RsquareAdj gives R-squared adjusted R-squared (available) constrained components. Function goodness gives statistics individual species sites. addition, special function varpart unbiased partitioning variance four separate components redundancy analysis. Nonmetric multidimensional scaling. NMDS method nonlinear mapping, concept variation explained make sense. However, 1 - stress^2 transforms nonlinear stress quantity analogous squared correlation coefficient. Function stressplot displays nonlinear fit gives statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-have-random-effects-in-constrained-ordination-or-in-adonis","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I have random effects in constrained ordination or in adonis?","title":"","text":". Strictly speaking, impossible. However, can define models respond similar goals random effects models, although strictly speaking use fixed effects. Constrained ordination functions cca, rda dbrda can Condition() terms formula. Condition() define partial terms fitted constraints can used remove effects background variables, contribution decomposing inertia (variance) reported separately. partial terms often regarded similar random effects, still fitted way terms strictly speaking fixed terms. Function adonis2 can evaluate terms sequentially. model right-hand-side ~ + B effects evaluated first, effects B removing effects . Sequential tests also available anova function constrained ordination results setting argument = \"term\". way, first terms can serve similar role random effects, although fitted way terms, strictly speaking fixed terms. permutation tests vegan based permute package allows constructing various restricted permutation schemes. instance, can set levels plots blocks factor regarded random term. major reason real random effects models impossible vegan functions tests based permutation data. data given, fixed, therefore permutation tests basically tests fixed terms fixed data. Random effect terms require permutations data random component instead given, fixed data, tests available vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-possible-to-have-passive-points-in-ordination","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Is it possible to have passive points in ordination?","title":"","text":"Vegan concept passive points, point little influence ordination results. However, can add points eigenvector methods using predict functions newdata. can first perform ordination without species sites, can find scores points using complete data newdata. predict functions available basic eigenvector methods vegan (cca, rda, decorana, --date list, use command methods(\"predict\")).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"class-variables-and-dummies","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Class variables and dummies","title":"","text":"define class variable R factor, vegan automatically handle . R (vegan) knows unordered ordered factors. Unordered factors internally coded dummy variables, one redundant level removed aliased. default contrasts, removed level first one. Ordered factors expressed polynomial contrasts. contrasts explained standard R documentation.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-are-environmental-arrows-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How are environmental arrows scaled?","title":"","text":"printed output envfit gives direction cosines coordinates unit length arrows. plotting, scaled correlation (square roots column r2). can see scaled lengths envfit arrows using command scores. scaled environmental vectors envfit arrows continuous environmental variables constrained ordination (cca, rda, dbrda) adjusted fill current graph. lengths arrows fixed meaning respect points (species, sites), can compared , therefore relative lengths important. want change scaling arrows, can use text (plotting arrows text) points (plotting arrows) functions constrained ordination. functions argument arrow.mul sets multiplier. plot function envfit also arrow.mul argument set arrow multiplier. save invisible result constrained ordination plot command, can see value currently used arrow.mul saved attribute biplot scores. Function ordiArrowMul used find scaling current plot. can use function see arrows scaled:","code":"sol <- cca(varespec) ef <- envfit(sol ~ ., varechem) plot(sol) ordiArrowMul(scores(ef, display=\"vectors\"))"},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-want-to-use-helmert-or-sum-contrasts","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I want to use Helmert or sum contrasts","title":"","text":"vegan uses standard R utilities defining contrasts. default standard installations use treatment contrasts, can change behaviour globally setting options locally using keyword contrasts. Please check R help pages user manuals details.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-are-aliased-variables-and-how-to-see-them","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"What are aliased variables and how to see them?","title":"","text":"Aliased variable information can expressed help variables. variables automatically removed constrained ordination vegan. aliased variables can redundant levels factors whole variables. Vegan function alias gives defining equations aliased variables. want see names aliased variables levels solution sol, use alias(sol, names.=TRUE).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"plotting-aliased-variables","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Plotting aliased variables","title":"","text":"can fit vectors class centroids aliased variables using envfit function. envfit function uses weighted fitting, fitted vectors identical vectors correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"restricted-permutations-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Restricted permutations in vegan","title":"","text":"Vegan uses permute package permutation tests. permute package allow restricted permutation designs time series, line transects, spatial grids blocking factors. construction restricted permutation schemes explained manual page permutations vegan documentation permute package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-use-different-plotting-symbols-in-ordination-graphics","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to use different plotting symbols in ordination graphics?","title":"","text":"default ordination plot function intended fast plotting configurable. use different plotting symbols, first create empty ordination plot plot(..., type=\"n\"), add points text created empty frame (... means arguments want give plot command). points text commands fully configurable, allow different plotting symbols characters.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-avoid-cluttered-ordination-graphs","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to avoid cluttered ordination graphs?","title":"","text":"really high number species sites, graphs often congested many labels overwritten. may impossible complete readable graphics data sets. give brief overview tricks can use. Gavin Simpson’s blog bottom heap series articles “decluttering ordination plots” detailed discussion examples. Use points, possibly different types need see labels. may need first create empty plot using plot(..., type=\"n\"), satisfied default graph. (... means arguments want give plot command.) Use points add labels desired points using interactive identify command need see labels. Add labels using function ordilabel uses non-transparent background text. labels still shadow , uppermost labels readable. Argument priority help displaying interesting labels (see Decluttering blog, part 1). Use orditorp function uses labels can added graph without overwriting labels, points otherwise, need see labels. must first create empty plot using plot(..., type=\"n\"), add labels points orditorp (see Decluttering blog). Use ordipointlabel uses points text labels points, tries optimize location text minimize overlap (see Decluttering blog). Ordination text points functions argument select can used full control selecting items plotted text points. Use interactive orditkplot function (vegan3d package) lets drag labels points better positions need see labels. one set points can used (see Decluttering blog). plot functions allow zoom part graph using xlim ylim arguments reduce clutter congested areas.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-flip-an-axis-in-ordination-diagram","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I flip an axis in ordination diagram?","title":"","text":"Use xlim ylim flipped limits. model mod <- cca(dune) can flip first axis plot(mod, xlim = c(3, -2)).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-zoom-into-an-ordination-plot","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I zoom into an ordination plot?","title":"","text":"can use xlim ylim arguments plot ordiplot zoom ordination diagrams. Normally must set xlim ylim ordination plots keep equal aspect ratio axes, fill graph longer axis fit. Dynamic zooming can done function orditkplot CRAN package vegan3d. can directly save edited orditkplot graph various graphic formats, can export graph object back R session use plot display results.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-twinspan","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Is there TWINSPAN?","title":"","text":"TWINSPAN R available github.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-restricted-permutation-does-not-influence-adonis-results","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Why restricted permutation does not influence adonis results?","title":"","text":"permutation scheme influences permutation distribution statistics probably significance levels, influence calculation statistics.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-is-deviance-calculated","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"How is deviance calculated?","title":"","text":"vegan functions, radfit use base R facility family maximum likelihood estimation. allows use several alternative error distributions, among \"poisson\" \"gaussian\". R family also defines deviance. can see equations deviance commands like poisson()$dev gaussian()$dev. general, deviance 2 times log.likelihood shifted models exact fit zero deviance.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Jari Oksanen. Author, maintainer. Gavin L. Simpson. Author. F. Guillaume Blanchet. Author. Roeland Kindt. Author. Pierre Legendre. Author. Peter R. Minchin. Author. R.B. O'Hara. Author. Peter Solymos. Author. M. Henry H. Stevens. Author. Eduard Szoecs. Author. Helene Wagner. Author. Matt Barbour. Author. Michael Bedward. Author. Ben Bolker. Author. Daniel Borcard. Author. Gustavo Carvalho. Author. Michael Chirico. Author. Miquel De Caceres. Author. Sebastien Durand. Author. Heloisa Beatriz Antoniazi Evangelista. Author. Rich FitzJohn. Author. Michael Friendly. Author. Brendan Furneaux. Author. Geoffrey Hannigan. Author. Mark O. Hill. Author. Leo Lahti. Author. Dan McGlinn. Author. Marie-Helene Ouellette. Author. Eduardo Ribeiro Cunha. Author. Tyler Smith. Author. Adrian Stier. Author. Cajo J.F. Ter Braak. Author. James Weedon. Author.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O'Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier , Ter Braak C, Weedon J (2024). vegan: Community Ecology Package. R package version 2.6-7, https://github.com/vegandevs/vegan, https://vegandevs.github.io/vegan/.","code":"@Manual{, title = {vegan: Community Ecology Package}, author = {Jari Oksanen and Gavin L. Simpson and F. Guillaume Blanchet and Roeland Kindt and Pierre Legendre and Peter R. Minchin and R.B. O'Hara and Peter Solymos and M. Henry H. Stevens and Eduard Szoecs and Helene Wagner and Matt Barbour and Michael Bedward and Ben Bolker and Daniel Borcard and Gustavo Carvalho and Michael Chirico and Miquel {De Caceres} and Sebastien Durand and Heloisa Beatriz Antoniazi Evangelista and Rich FitzJohn and Michael Friendly and Brendan Furneaux and Geoffrey Hannigan and Mark O. Hill and Leo Lahti and Dan McGlinn and Marie-Helene Ouellette and Eduardo {Ribeiro Cunha} and Tyler Smith and Adrian Stier and Cajo J.F. {Ter Braak} and James Weedon}, year = {2024}, note = {R package version 2.6-7, https://github.com/vegandevs/vegan}, url = {https://vegandevs.github.io/vegan/}, }"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"vegan-an-r-package-for-community-ecologists","dir":"","previous_headings":"","what":"vegan: an R package for community ecologists","title":"vegan: an R package for community ecologists","text":"Ordination methods, diversity analysis functions community vegetation ecologists. Website development version vegan package. Vignettes available R-universe Introduction ordination vegan Partition Variation Diversity analysis vegan Design decisions implementation","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"vegan: an R package for community ecologists","text":"install development version vegan can use usual git R CMD build -> R CMD INSTALL dance cloned repo (downloaded sources). ’ll need able install packages source work; don’t relevant developer tools, won’t able install vegan way.","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"using-remotes","dir":"","previous_headings":"","what":"Using remotes","title":"vegan: an R package for community ecologists","text":"developer tools installed don’t want hassle keeping local source code tree --date, use remotes package:","code":"install.packages(\"remotes\") remotes::install_github(\"vegandevs/vegan\")"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installing-binaries-from-r-universe","dir":"","previous_headings":"","what":"Installing binaries from R Universe","title":"vegan: an R package for community ecologists","text":"just want install binary version packages, just CRAN, can install R Universe repository. Run following R session:","code":"install.packages('vegan', repos = c('https://vegandevs.r-universe.dev','https://cloud.r-project.org'))"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":null,"dir":"Reference","previous_headings":"","what":"Barro Colorado Island Tree Counts — BCI","title":"Barro Colorado Island Tree Counts — BCI","text":"Tree counts 1-hectare plots Barro Colorado Island associated site information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI) data(BCI.env)"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Barro Colorado Island Tree Counts — BCI","text":"data frame 50 plots (rows) 1 hectare counts trees plot total 225 species (columns). Full Latin names used tree species. names updated http://www.theplantlist.org Kress et al. (2009) allows matching 207 species doi:10.5061/dryad.63q27 (Zanne et al., 2014). original species names available attribute original.names BCI. See Examples changed names. BCI.env, data frame 50 plots (rows) nine site variables derived Pyke et al. (2001) Harms et al. (2001): UTM.EW: UTM coordinates (zone 17N) East-West. UTM.NS: UTM coordinates (zone 17N) North-South. Precipitation: Precipitation mm per year. Elevation: Elevation m sea level. Age.cat: Forest age category. Geology: Underlying geological formation. Habitat: Dominant habitat type based map habitat types 25 grid cells plot (Harms et al. 2001, excluding streamside habitat). habitat types Young forests (ca. 100 years), old forests > 7 degree slopes (OldSlope), old forests 152 m elevation (OldLow) higher elevation (OldHigh) Swamp forests. River: \"Yes\" streamside habitat plot. EnvHet: Environmental Heterogeneity assessed Simpson diversity frequencies Habitat types 25 grid cells plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Barro Colorado Island Tree Counts — BCI","text":"Data give numbers trees least 10 cm diameter breast height (DBH) one hectare quadrat 1982 BCI plot. Within plot, individuals tallied recorded table. full survey included smaller trees DBH 1 cm larger, BCI dataset subset larger trees compiled Condit et al. (2002). full data thinner trees densities 4000 stems per hectare, ten times stems data. dataset BCI provided (2003) illustrate analysis methods vegan. scientific research ecological issues strongly recommend access complete modern data (Condit et al. 2019) updated taxonomy (Condit et al. 2020). data frame contains Barro Colorado Island subset full data table Condit et al. (2002). quadrats located regular grid. See BCI.env coordinates. full description site information BCI.env given Pyke et al. (2001) Harms et al. (2001). N.B. Pyke et al. (2001) Harms et al. (2001) give conflicting information forest age categories elevation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Barro Colorado Island Tree Counts — BCI","text":"https://www.science.org/doi/10.1126/science.1066854 community data References environmental data. updated complete data (incl. thinner trees 1 cm), see Condit et al. (2019).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Barro Colorado Island Tree Counts — BCI","text":"Condit, R, Pitman, N, Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., Nuñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002). Beta-diversity tropical forest trees. Science 295, 666--669. Condit R., Pérez, R., Aguilar, S., Lao, S., Foster, R. & Hubbell, S. (2019). Complete data Barro Colorado 50-ha plot: 423617 trees, 35 years [Dataset]. Dryad. doi:10.15146/5xcp-0d46 Condit, R., Aguilar, S., Lao, S., Foster, R., Hubbell, S. (2020). BCI 50-ha Plot Taxonomy [Dataset]. Dryad. doi:10.15146/R3FH61 Harms K.E., Condit R., Hubbell S.P. & Foster R.B. (2001) Habitat associations trees shrubs 50-ha neotropical forest plot. J. Ecol. 89, 947--959. Kress W.J., Erickson D.L, Jones F.., Swenson N.G, Perez R., Sanjur O. & Bermingham E. (2009) Plant DNA barcodes community phylogeny tropical forest dynamics plot Panama. PNAS 106, 18621--18626. Pyke, C. R., Condit, R., Aguilar, S., & Lao, S. (2001). Floristic composition across climatic gradient neotropical lowland forest. Journal Vegetation Science 12, 553--566. doi:10.2307/3237007 Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506, 89--92. doi:10.1038/nature12872 (published online Dec 22, 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI, BCI.env) head(BCI.env) #> UTM.EW UTM.NS Precipitation Elevation Age.cat Geology Habitat Stream EnvHet #> 1 625754 1011569 2530 120 c3 Tb OldSlope Yes 0.6272 #> 2 625754 1011669 2530 120 c3 Tb OldLow Yes 0.3936 #> 3 625754 1011769 2530 120 c3 Tb OldLow No 0.0000 #> 4 625754 1011869 2530 120 c3 Tb OldLow No 0.0000 #> 5 625754 1011969 2530 120 c3 Tb OldSlope No 0.4608 #> 6 625854 1011569 2530 120 c3 Tb OldLow No 0.0768 ## see changed species names oldnames <- attr(BCI, \"original.names\") taxa <- cbind(\"Old Names\" = oldnames, \"Current Names\" = names(BCI)) noquote(taxa[taxa[,1] != taxa[,2], ]) #> Old Names Current Names #> [1,] Abarema.macradenium Abarema.macradenia #> [2,] Acacia.melanoceras Vachellia.melanoceras #> [3,] Apeiba.aspera Apeiba.glabra #> [4,] Aspidosperma.cruenta Aspidosperma.desmanthum #> [5,] Cassipourea.elliptica Cassipourea.guianensis #> [6,] Cespedezia.macrophylla Cespedesia.spathulata #> [7,] Chlorophora.tinctoria Maclura.tinctoria #> [8,] Coccoloba.manzanillensis Coccoloba.manzinellensis #> [9,] Coussarea.curvigemmia Coussarea.curvigemma #> [10,] Cupania.sylvatica Cupania.seemannii #> [11,] Dipteryx.panamensis Dipteryx.oleifera #> [12,] Eugenia.coloradensis Eugenia.florida #> [13,] Eugenia.oerstedeana Eugenia.oerstediana #> [14,] Guapira.standleyana Guapira.myrtiflora #> [15,] Hyeronima.alchorneoides Hieronyma.alchorneoides #> [16,] Inga.marginata Inga.semialata #> [17,] Lonchocarpus.latifolius Lonchocarpus.heptaphyllus #> [18,] Maquira.costaricana Maquira.guianensis.costaricana #> [19,] Phoebe.cinnamomifolia Cinnamomum.triplinerve #> [20,] Swartzia.simplex.var.ochnacea Swartzia.simplex.continentalis #> [21,] Tabebuia.guayacan Handroanthus.guayacan"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":null,"dir":"Reference","previous_headings":"","what":"Canonical Correlation Analysis — CCorA","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis, following Brian McArdle's unpublished graduate course notes, plus improvements allow calculations case sparse collinear matrices, permutation test Pillai's trace statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Canonical Correlation Analysis — CCorA","text":"","code":"CCorA(Y, X, stand.Y=FALSE, stand.X=FALSE, permutations = 0, ...) # S3 method for CCorA biplot(x, plot.type=\"ov\", xlabs, plot.axes = 1:2, int=0.5, col.Y=\"red\", col.X=\"blue\", cex=c(0.7,0.9), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Canonical Correlation Analysis — CCorA","text":"Y Left matrix (object class: matrix data.frame). X Right matrix (object class: matrix data.frame). stand.Y Logical; Y standardized? stand.X Logical; X standardized? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x CCoaR result object. plot.type character string indicating following plots produced: \"objects\", \"variables\", \"ov\" (separate graphs objects variables), \"biplots\". unambiguous subset containing first letters names can used instead full names. xlabs Row labels. default use row names, NULL uses row numbers instead, NA suppresses plotting row names completely. plot.axes vector 2 values containing order numbers canonical axes plotted. Default: first two axes. int Radius inner circles plotted visual references plots variables. Default: int=0.5. int=0, inner circle plotted. col.Y Color used objects variables first data table (Y) plots. biplots, objects black. col.X Color used objects variables second data table (X) plots. cex vector 2 values containing size reduction factors object variable names, respectively, plots. Default values: cex=c(0.7,0.9). ... arguments passed functions. function biplot.CCorA passes graphical arguments biplot biplot.default. CCorA currently ignores extra arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis (Hotelling 1936) seeks linear combinations variables Y maximally correlated linear combinations variables X. analysis estimates relationships displays graphs. Pillai's trace statistic computed tested parametrically (F-test); permutation test also available. Algorithmic note -- blunt approach read two matrices, compute covariance matrices, matrix S12 %*% inv(S22) %*% t(S12) %*% inv(S11). trace Pillai's trace statistic. approach may fail, however, heavy multicollinearity sparse data matrices. safe approach replace data matrices PCA object scores. function can produce different types plots depending option chosen: \"objects\" produces two plots objects, one space Y, second space X; \"variables\" produces two plots variables, one variables Y space Y, second variables X space X; \"ov\" produces four plots, two objects two variables; \"biplots\" produces two biplots, one first matrix (Y) one second matrix (X) solutions. biplots, function passes arguments biplot.default; consult help page configuring biplots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Canonical Correlation Analysis — CCorA","text":"Function CCorA returns list containing following elements: Pillai Pillai's trace statistic = sum canonical eigenvalues. Eigenvalues Canonical eigenvalues. squares canonical correlations. CanCorr Canonical correlations. Mat.ranks Ranks matrices Y X. RDA.Rsquares Bimultivariate redundancy coefficients (R-squares) RDAs Y|X X|Y. RDA.adj.Rsq RDA.Rsquares adjusted n number explanatory variables. nperm Number permutations. p.Pillai Parametric probability value associated Pillai's trace. p.perm Permutational probability associated Pillai's trace. Cy Object scores Y biplot. Cx Object scores X biplot. corr.Y.Cy Scores Y variables Y biplot, computed cor(Y,Cy). corr.X.Cx Scores X variables X biplot, computed cor(X,Cx). corr.Y.Cx cor(Y,Cy) available plotting variables Y space X manually. corr.X.Cy cor(X,Cx) available plotting variables X space Y manually. control list control values permutations returned function . call Call CCorA function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Canonical Correlation Analysis — CCorA","text":"Hotelling, H. 1936. Relations two sets variates. Biometrika 28: 321-377. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Canonical Correlation Analysis — CCorA","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal. Implemented vegan help Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Canonical Correlation Analysis — CCorA","text":"","code":"# Example using two mite groups. The mite data are available in vegan data(mite) # Two mite species associations (Legendre 2005, Fig. 4) group.1 <- c(1,2,4:8,10:15,17,19:22,24,26:30) group.2 <- c(3,9,16,18,23,25,31:35) # Separate Hellinger transformations of the two groups of species mite.hel.1 <- decostand(mite[,group.1], \"hel\") mite.hel.2 <- decostand(mite[,group.2], \"hel\") rownames(mite.hel.1) = paste(\"S\",1:nrow(mite),sep=\"\") rownames(mite.hel.2) = paste(\"S\",1:nrow(mite),sep=\"\") out <- CCorA(mite.hel.1, mite.hel.2) out #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mite.hel.1, X = mite.hel.2) #> #> Y X #> Matrix Ranks 24 11 #> #> Pillai's trace: 4.573009 #> #> Significance of Pillai's trace: #> from F-distribution: 0.0032737 #> CanAxis1 CanAxis2 CanAxis3 CanAxis4 CanAxis5 CanAxis6 #> Canonical Correlations 0.92810 0.82431 0.81209 0.74981 0.70795 0.65950 #> CanAxis7 CanAxis8 CanAxis9 CanAxis10 CanAxis11 #> Canonical Correlations 0.50189 0.48179 0.41089 0.37823 0.28 #> #> Y | X X | Y #> RDA R squares 0.33224 0.5376 #> adj. RDA R squares 0.20560 0.2910 #> biplot(out, \"ob\") # Two plots of objects biplot(out, \"v\", cex=c(0.7,0.6)) # Two plots of variables biplot(out, \"ov\", cex=c(0.7,0.6)) # Four plots (2 for objects, 2 for variables) biplot(out, \"b\", cex=c(0.7,0.6)) # Two biplots biplot(out, xlabs = NA, plot.axes = c(3,5)) # Plot axes 3, 5. No object names biplot(out, plot.type=\"biplots\", xlabs = NULL) # Replace object names by numbers # Example using random numbers. No significant relationship is expected mat1 <- matrix(rnorm(60),20,3) mat2 <- matrix(rnorm(100),20,5) out2 = CCorA(mat1, mat2, permutations=99) out2 #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mat1, X = mat2, permutations = 99) #> #> Y X #> Matrix Ranks 3 5 #> #> Pillai's trace: 0.480458 #> #> Significance of Pillai's trace: #> from F-distribution: 0.90606 #> based on permutations: 0.94 #> Permutation: free #> Number of permutations: 99 #> #> CanAxis1 CanAxis2 CanAxis3 #> Canonical Correlations 0.64421 0.23458 0.1021 #> #> Y | X X | Y #> RDA R squares 0.214302 0.0839 #> adj. RDA R squares -0.066305 -0.0879 #> biplot(out2, \"b\")"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Add New Points to NMDS ordination — MDSaddpoints","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Add new points existing metaMDS monoMDS ordination.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"MDSaddpoints(nmds, dis, neighbours = 5, maxit = 200) dist2xy(dist, pick, type = c(\"xy\", \"xx\"), invert = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"nmds Result object metaMDS monoMDS. configuration points fixed, new points added. dis Rectangular non-symmetric dissimilarity matrix among new points (rows) old fixed points (columns). matrix can extracted complete dissimilarities old new points dist2xy, calculated designdist2. neighbours Number nearest points used get starting locations new points. maxit Maximum number iterations. dist Input dissimilarities. pick Indices (integers) selected observations logical vector TRUE picked items. output original order reordered argument. type \"xy\" returns rectangular data picked picked observations, \"xx\" subset symmetric dissimilarities. invert Invert pick: drop elements listed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function return list class \"nmds\" (objects type vegan) following elements points Coordinates added new points seeds Starting coordinates new points. deltastress Change stress added points. iters Number iterations. cause Cause termination iterations. Integer convergence criteria monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function provides interface monoMDS Fortran code add new points existing ordination regarded fixed. function similar role predict functions newdata Euclidean ordination (e.g. predict.cca). Input data must rectangular matrix distances among new added points (rows) fixed old points (columns). matrices can extracted complete dissimilarities helper function dist2xy. Function designdist2 can directly calculate rectangular dissimilarity matrices sampling units (rows) two matries. addition, analogue distance function can calculate dissimilarities among two matrices, including functions specified designdist2. Great care needed preparing dissimilarities input. dissimilarity index must exactly fixed ordination, columns must match old fixed points, rows added new points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"## Cross-validation: remove a point when performing NMDS and add as ## a new points data(dune) d <- vegdist(dune) ## remove point 3 from ordination mod3 <- metaMDS(dist2xy(d, 3, \"xx\", invert = TRUE), trace=0) ## add point 3 to the result MDSaddpoints(mod3, dist2xy(d, 3)) #> $points #> MDS1 MDS2 #> 3 -0.09726881 -0.4560917 #> #> $seed #> [,1] [,2] #> [1,] 0.01222493 -0.4353716 #> #> $deltastress #> [1] 0.001811377 #> #> $iters #> [1] 18 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\" ## Use designdist2 d15 <- designdist(dune[1:15,]) m15 <- metaMDS(d15, trace=0) MDSaddpoints(m15, designdist2(dune[1:15,], dune[16:20,])) #> $points #> MDS1 MDS2 #> 16 0.4296366 -0.3068902 #> 17 -0.4122463 0.4118309 #> 18 -0.1135994 0.3120708 #> 19 0.1421922 0.4869128 #> 20 0.5892538 0.1751219 #> #> $seed #> [,1] [,2] #> [1,] 0.2832566 0.019571976 #> [2,] -0.1920994 0.137270818 #> [3,] -0.1529998 0.094374124 #> [4,] -0.0244597 0.137988191 #> [5,] 0.3212258 0.009941098 #> #> $deltastress #> [1] 0.03394768 #> #> $iters #> [1] 33 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":null,"dir":"Reference","previous_headings":"","what":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function rotates multidimensional scaling result first dimension parallel external (environmental variable). function can handle results metaMDS monoMDS functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"MDSrotate(object, vec, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"object result object metaMDS monoMDS. vec environmental variable matrix variables. number variables must lower number dimensions, solution rotated variables order appear matrix. Alternatively vec can factor, solution rotated optimal separation factor levels using lda. na.rm Remove missing values continuous variable vec. ... arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"orientation rotation undefined multidimensional scaling. Functions metaMDS metaMDS can rotate solutions principal components dispersion points highest first dimension. Sometimes different rotation intuitive, MDSrotate allows rotation result first axis parallel given external variable two first variables completely two-dimensional plane etc. several external variables supplied, applied order matrix. First axis rotated first supplied variable, second axis second variable. variables usually correlated, second variable usually aligned second axis, uncorrelated later dimensions. must least one free dimension: number external variables must lower number dimensions, used environmental variables uncorrelated free dimension. Alternatively method can rotate discriminate levels factor using linear discriminant analysis (lda). hardly meaningful two-dimensional solutions, since rotations two dimensions separation cluster levels. However, function can useful finding two-dimensional projection clusters two dimensions. last dimension always show residual variation, \\(k\\) dimensions, \\(k-1\\) discrimination vectors used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function returns original ordination result, rotated scores (site species available), pc attribute scores set FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Rotation factor variable experimental feature may removed. discriminant analysis weights dimensions discriminating power, MDSrotate performs rigid rotation. Therefore solution may optimal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"data(varespec) data(varechem) mod <- monoMDS(vegdist(varespec)) mod <- with(varechem, MDSrotate(mod, pH)) plot(mod) ef <- envfit(mod ~ pH, varechem, permutations = 0) plot(ef) ordisurf(mod ~ pH, varechem, knots = 1, add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ poly(x1, 1) + poly(x2, 1) #> Total model degrees of freedom 3 #> #> REML score: -2.736059"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":null,"dir":"Reference","previous_headings":"","what":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds & Shaw test concerns location highest (hump) lowest (pit) value quadratic curve given points. Typically, used study whether quadratic hump pit located within studied interval. current test generalized applies generalized linear models (glm) link function instead simple quadratic curve. test popularized ecology analysis humped species richness patterns (Mittelbach et al. 2001), general. logarithmic link function, quadratic response defines Gaussian response model ecological gradients (ter Braak & Looman 1986), test can used inspecting location Gaussian optimum within given range gradient. can also used replace Tokeshi's test “bimodal” species frequency distribution.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"MOStest(x, y, interval, ...) # S3 method for MOStest plot(x, which = c(1,2,3,6), ...) fieller.MOStest(object, level = 0.95) # S3 method for MOStest profile(fitted, alpha = 0.01, maxsteps = 10, del = zmax/5, ...) # S3 method for MOStest confint(object, parm = 1, level = 0.95, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"x independent variable plotting object plot. y dependent variable. interval two points test statistic evaluated. missing, extremes x used. Subset plots produced. Values =1 2 define plots specific MOStest (see Details), larger values select graphs plot.lm (minus 2). object, fitted result object MOStest. level confidence level required. alpha Maximum significance level allowed. maxsteps Maximum number steps profile. del step length parameter profile (see code). parm Ignored. ... variables passed functions. Function MOStest passes glm can include family. functions pass underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function fits quadratic curve \\(\\mu = b_0 + b_1 x + b_2 x^2\\) given family link function. \\(b_2 < 0\\), defines unimodal curve highest point \\(u = -b_1/(2 b_2)\\) (ter Braak & Looman 1986). \\(b_2 > 0\\), parabola minimum \\(u\\) response sometimes called “bimodal”. null hypothesis extreme point \\(u\\) located within interval given points \\(p_1\\) \\(p_2\\). extreme point \\(u\\) exactly \\(p_1\\), \\(b_1 = 0\\) shifted axis \\(x - p_1\\). test, origin x shifted values \\(p_1\\) \\(p_2\\), test statistic based differences deviances original model model origin forced given location using standard anova.glm function (Oksanen et al. 2001). Mitchell-Olds & Shaw (1987) used first degree coefficient significance estimated summary.glm function. give identical results Normal error, error distributions preferable use test based differences deviances fitted models. test often presented general test location hump, really dependent quadratic fitted curve. hump different form quadratic, test may insignificant. strong assumptions test, use support functions inspect fit. Function plot(..., =1) displays data points, fitted quadratic model, approximate 95% confidence intervals (2 times SE). Function plot = 2 displays approximate confidence interval polynomial coefficients, together two lines indicating combinations coefficients produce evaluated points x. Moreover, cross-hair shows approximate confidence intervals polynomial coefficients ignoring correlations. Higher values produce corresponding graphs plot.lm. , must add 2 value plot.lm. Function fieller.MOStest approximates confidence limits location extreme point (hump pit) using Fieller's theorem following ter Braak & Looman (1986). test based quasideviance except family poisson binomial. Function profile evaluates profile deviance fitted model, confint finds profile based confidence limits following Oksanen et al. (2001). test typically used assessing significance diversity hump productivity gradient (Mittelbach et al. 2001). also can used location pit (deepest points) instead Tokeshi test. , can used test location Gaussian optimum ecological gradient analysis (ter Braak & Looman 1986, Oksanen et al. 2001).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function based glm, returns result object glm amended result test. new items MOStest : isHump TRUE response hump. isBracketed TRUE hump pit bracketed evaluated points. hump Sorted vector location hump pit points test evaluated. coefficients Table test statistics significances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds, T. & Shaw, R.G. 1987. Regression analysis natural selection: statistical inference biological interpretation. Evolution 41, 1149--1161. Mittelbach, G.C. Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, R.M., Dodson, S.. & Gough, L. 2001. observed relationship species richness productivity? Ecology 82, 2381--2396. Oksanen, J., Läärä, E., Tolonen, K. & Warner, B.G. 2001. Confidence intervals optimum Gaussian response function. Ecology 82, 1191--1197. ter Braak, C.J.F & Looman, C.W.N 1986. Weighted averaging, logistic regression Gaussian response model. Vegetatio 65, 3--11.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Function fieller.MOStest based package optgrad Ecological Archives (https://figshare.com/articles/dataset/Full_Archive/3521975) accompanying Oksanen et al. (2001). Ecological Archive package optgrad also contains profile deviance method location hump pit, current implementation profile confint rather follow example profile.glm confint.glm MASS package.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"## The Al-Mufti data analysed in humpfit(): mass <- c(140,230,310,310,400,510,610,670,860,900,1050,1160,1900,2480) spno <- c(1, 4, 3, 9, 18, 30, 20, 14, 3, 2, 3, 2, 5, 2) mod <- MOStest(mass, spno) ## Insignificant mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: gaussian #> Link function: identity #> #> hump min max #> 46.89749 140.00000 2480.00000 #> ***** Caution: hump/pit not bracketed by the data ****** #> #> min/max F Pr(>F) #> hump at min 140 0.0006 0.9816 #> hump at max 2480 0.3161 0.5852 #> Combined 0.9924 ## ... but inadequate shape of the curve op <- par(mfrow=c(2,2), mar=c(4,4,1,1)+.1) plot(mod) ## Looks rather like log-link with Poisson error and logarithmic biomass mod <- MOStest(log(mass), spno, family=quasipoisson) mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: quasipoisson #> Link function: log #> #> min hump max #> 4.941642 6.243371 7.816014 #> #> min/max F Pr(>F) #> hump at min 4.9416 7.1367 0.02174 * #> hump at max 7.8160 9.0487 0.01191 * #> Combined 0.03338 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod) par(op) ## Confidence Limits fieller.MOStest(mod) #> 2.5 % 97.5 % #> 5.255827 6.782979 confint(mod) #> 2.5 % 97.5 % #> 5.816021 6.574378 plot(profile(mod))"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":null,"dir":"Reference","previous_headings":"","what":"Adjusted R-square — RsquareAdj","title":"Adjusted R-square — RsquareAdj","text":"functions finds adjusted R-square.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Adjusted R-square — RsquareAdj","text":"","code":"# S3 method for default RsquareAdj(x, n, m, ...) # S3 method for rda RsquareAdj(x, ...) # S3 method for cca RsquareAdj(x, permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Adjusted R-square — RsquareAdj","text":"x Unadjusted R-squared object terms evaluation adjusted R-squared can found. n, m Number observations number degrees freedom fitted model. permutations Number permutations use computing adjusted R-squared cca. permutations can calculated parallel specifying number cores passed permutest ... arguments (ignored) except case cca arguments passed permutest.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Adjusted R-square — RsquareAdj","text":"default method finds adjusted \\(R^2\\) unadjusted \\(R^2\\), number observations, number degrees freedom fitted model. specific methods find information fitted result object. specific methods rda (also used distance-based RDA), cca, lm glm. Adjusted, even unadjusted, \\(R^2\\) may available cases, functions return NA. \\(R^2\\) values available gaussian models glm. adjusted, \\(R^2\\) cca computed using permutation approach developed Peres-Neto et al. (2006). default 1000 permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Adjusted R-square — RsquareAdj","text":"functions return list items r.squared adj.r.squared.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Adjusted R-square — RsquareAdj","text":"Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277. Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87, 2614--2625.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Adjusted R-square — RsquareAdj","text":"","code":"data(mite) data(mite.env) ## rda m <- rda(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.5265047 #> #> $adj.r.squared #> [1] 0.4367038 #> ## cca m <- cca(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.4471676 #> #> $adj.r.squared #> [1] 0.3446879 #> ## default method RsquareAdj(0.8, 20, 5) #> [1] 0.7285714"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":null,"dir":"Reference","previous_headings":"","what":"Self-Starting nls Species-Area Models — SSarrhenius","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions provide self-starting species-area models non-linear regression (nls). can also used fitting species accumulation models fitspecaccum. models (many ) reviewed Dengler (2009).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"SSarrhenius(area, k, z) SSgleason(area, k, slope) SSgitay(area, k, slope) SSlomolino(area, Asym, xmid, slope)"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"area Area size sample: independent variable. k, z, slope, Asym, xmid Estimated model parameters: see Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions assumed used species richness (number species) independent variable, area sample size independent variable. Basically, define least squares models untransformed data, differ models transformed species richness models non-Gaussian error. Arrhenius model (SSarrhenius) expression k*area^z. classical model can found textbook ecology (also Dengler 2009). Parameter z steepness species-area curve, k expected number species unit area. Gleason model (SSgleason) linear expression k + slope*log(area) (Dengler 200). linear model, starting values give final estimates; provided ease comparison models. Gitay model (SSgitay) quadratic logarithmic expression (k + slope*log(area))^2 (Gitay et al. 1991, Dengler 2009). Parameter slope steepness species-area curve, k square root expected richness unit area. Lomolino model (SSlomolino) Asym/(1 + slope^log(xmid/area)) (Lomolino 2000, Dengler 2009). Parameter Asym asymptotic maximum number species, slope maximum slope increase richness, xmid area half maximum richness achieved. addition models, several models studied Dengler (2009) available standard R self-starting models: Michaelis-Menten (SSmicmen), Gompertz (SSgompertz), logistic (SSlogis), Weibull (SSweibull), others may useful.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Numeric vector length area. value expression model. arguments names objects gradient matrix respect names attached attribute named gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Dengler, J. (2009) function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gitay, H., Roxburgh, S.H. & Wilson, J.B. (1991) Species-area relationship New Zealand tussock grassland, implications nature reserve design community structure. Journal Vegetation Science 2, 113--118. Lomolino, M. V. (2000) Ecology's general, yet protean pattern: species-area relationship. Journal Biogeography 27, 17--26.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"## Get species area data: sipoo.map gives the areas of islands data(sipoo, sipoo.map) S <- specnumber(sipoo) plot(S ~ area, sipoo.map, xlab = \"Island Area (ha)\", ylab = \"Number of Species\", ylim = c(1, max(S))) ## The Arrhenius model marr <- nls(S ~ SSarrhenius(area, k, z), data=sipoo.map) marr #> Nonlinear regression model #> model: S ~ SSarrhenius(area, k, z) #> data: sipoo.map #> k z #> 3.4062 0.4364 #> residual sum-of-squares: 78.1 #> #> Number of iterations to convergence: 5 #> Achieved convergence tolerance: 1.056e-06 ## confidence limits from profile likelihood confint(marr) #> Waiting for profiling to be done... #> 2.5% 97.5% #> k 2.6220312 4.3033906 #> z 0.3813576 0.4944693 ## draw a line xtmp <- with(sipoo.map, seq(min(area), max(area), len=51)) lines(xtmp, predict(marr, newdata=data.frame(area = xtmp)), lwd=2) ## The normal way is to use linear regression on log-log data, ## but this will be different from the previous: mloglog <- lm(log(S) ~ log(area), data=sipoo.map) mloglog #> #> Call: #> lm(formula = log(S) ~ log(area), data = sipoo.map) #> #> Coefficients: #> (Intercept) log(area) #> 1.0111 0.4925 #> lines(xtmp, exp(predict(mloglog, newdata=data.frame(area=xtmp))), lty=2) ## Gleason: log-linear mgle <- nls(S ~ SSgleason(area, k, slope), sipoo.map) lines(xtmp, predict(mgle, newdata=data.frame(area=xtmp)), lwd=2, col=2) ## Gitay: quadratic of log-linear mgit <- nls(S ~ SSgitay(area, k, slope), sipoo.map) lines(xtmp, predict(mgit, newdata=data.frame(area=xtmp)), lwd=2, col = 3) ## Lomolino: using original names of the parameters (Lomolino 2000): mlom <- nls(S ~ SSlomolino(area, Smax, A50, Hill), sipoo.map) mlom #> Nonlinear regression model #> model: S ~ SSlomolino(area, Smax, A50, Hill) #> data: sipoo.map #> Smax A50 Hill #> 53.493 94.697 2.018 #> residual sum-of-squares: 55.37 #> #> Number of iterations to convergence: 6 #> Achieved convergence tolerance: 9.715e-07 lines(xtmp, predict(mlom, newdata=data.frame(area=xtmp)), lwd=2, col = 4) ## One canned model of standard R: mmic <- nls(S ~ SSmicmen(area, Asym, slope), sipoo.map) lines(xtmp, predict(mmic, newdata = data.frame(area=xtmp)), lwd =2, col = 5) legend(\"bottomright\", c(\"Arrhenius\", \"log-log linear\", \"Gleason\", \"Gitay\", \"Lomolino\", \"Michaelis-Menten\"), col=c(1,1,2,3,4,5), lwd=c(2,1,2,2,2,2), lty=c(1,2,1,1,1,1)) ## compare models (AIC) allmods <- list(Arrhenius = marr, Gleason = mgle, Gitay = mgit, Lomolino = mlom, MicMen= mmic) sapply(allmods, AIC) #> Arrhenius Gleason Gitay Lomolino MicMen #> 83.49847 96.94018 80.54984 79.30718 83.02003"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Compute single terms can added dropped constrained ordination model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"# S3 method for cca add1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...) # S3 method for cca drop1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"object constrained ordination object cca, rda, dbrda capscale. scope formula giving terms considered adding dropping; see add1 details. test permutation test added using anova.cca. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ... arguments passed add1.default, drop1.default, anova.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"argument test = \"none\" functions call add1.default drop1.default. argument test = \"permutation\" functions add test results anova.cca. Function drop1.cca call anova.cca argument = \"margin\". Function add1.cca implement test single term additions directly available anova.cca. Functions used implicitly step, ordiR2step ordistep. deviance.cca deviance.rda used step firm basis, setting argument test = \"permutation\" may help getting useful insight validity model building. Function ordistep calls alternately drop1.cca add1.cca argument test = \"permutation\" selects variables permutation \\(P\\)-values. Meticulous use add1.cca drop1.cca allow judicious model building. default number permutations set low value, permutation tests can take long time. sufficient give impression significances terms, higher values permutations used \\(P\\) values really important.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Returns similar object add1 drop1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"data(dune) data(dune.env) ## Automatic model building based on AIC but with permutation tests step(cca(dune ~ 1, dune.env), reformulate(names(dune.env)), test=\"perm\") #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 86.608 2.2536 0.005 ** #> + Management 3 86.935 2.1307 0.005 ** #> + A1 1 87.411 2.1400 0.045 * #> 87.657 #> + Manure 4 88.832 1.5251 0.030 * #> + Use 2 89.134 1.1431 0.200 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC F Pr(>F) #> 86.608 #> + Management 3 86.813 1.4565 0.045 * #> + A1 1 86.992 1.2624 0.175 #> + Use 2 87.259 1.2760 0.085 . #> + Manure 4 87.342 1.3143 0.050 * #> - Moisture 3 87.657 2.2536 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #> ## see ?ordistep to do the same, but based on permutation P-values if (FALSE) { ordistep(cca(dune ~ 1, dune.env), reformulate(names(dune.env))) } ## Manual model building ## -- define the maximal model for scope mbig <- rda(dune ~ ., dune.env) ## -- define an empty model to start with m0 <- rda(dune ~ 1, dune.env) ## -- manual selection and updating add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 89.620 #> A1 1 89.591 1.9217 0.015 * #> Moisture 3 87.707 2.5883 0.005 ** #> Management 3 87.082 2.8400 0.005 ** #> Use 2 91.032 1.1741 0.225 #> Manure 4 89.232 1.9539 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Management) add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 87.082 #> A1 1 87.424 1.2965 0.215 #> Moisture 3 85.567 1.9764 0.010 ** #> Use 2 88.284 1.0510 0.400 #> Manure 3 87.517 1.3902 0.105 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Moisture) ## -- included variables still significant? drop1(m0, test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> Management 3 87.707 2.1769 0.010 ** #> Moisture 3 87.082 1.9764 0.015 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> A1 1 86.220 0.8359 0.605 #> Use 2 86.842 0.8027 0.765 #> Manure 3 85.762 1.1225 0.390"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"additive diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set (gamma diversity). hierarchical null model testing, statistic returned function evaluated according nested hierarchical sampling design (hiersimu).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"adipart(...) # S3 method for default adipart(y, x, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula adipart(formula, data, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) hiersimu(...) # S3 method for default hiersimu(y, x, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula hiersimu(formula, data, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, function performs overall decomposition alpha, beta gamma diversities. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variables referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, diversity index calculated (see Details). weights Character, \"unif\" uniform weights, \"prop\" weighting proportional sample abundances use weighted averaging individual alpha values within strata given level sampling hierarchy. relative Logical, TRUE alpha beta diversity values given relative value gamma function adipart. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. FUN function used hiersimu. must fully specified, currently arguments passed function via .... location Character, identifies function (mean median) used calculate location samples. drop.highest Logical, drop highest level . FUN evaluates arrays least 2 dimensions, highest level dropped, selected . ... arguments passed functions, e.g. base logarithm Shannon diversity, method, thin burnin arguments oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Additive diversity partitioning means mean alpha beta diversities add gamma diversity, thus beta diversity measured dimensions alpha gamma (Lande 1996). additive procedure extended across multiple scales hierarchical sampling design \\(= 1, 2, 3, \\ldots, m\\) levels sampling (Crist et al. 2003). Samples lower hierarchical levels nested within higher level units, thus \\(=1\\) \\(=m\\) grain size increasing constant survey extent. level \\(\\), \\(\\alpha_i\\) denotes average diversity found within samples. highest sampling level, diversity components calculated $$\\beta_m = \\gamma - \\alpha_m$$ lower sampling level $$\\beta_i = \\alpha_{+1} - \\alpha_i$$ , additive partition diversity $$\\gamma = \\alpha_1 + \\sum_{=1}^m \\beta_i$$ Average alpha components can weighted uniformly (weight=\"unif\") calculate simple average, proportionally sample abundances (weight=\"prop\") calculate weighted average follows $$\\alpha_i = \\sum_{j=1}^{n_i} D_{ij} w_{ij}$$ \\(D_{ij}\\) diversity index \\(w_{ij}\\) weight calculated \\(j\\)th sample \\(\\)th sampling level. implementation additive diversity partitioning adipart follows Crist et al. 2003. based species richness (\\(S\\), \\(S-1\\)), Shannon's Simpson's diversity indices stated index argument. expected diversity components calculated nsimul times individual based randomisation community data matrix. done \"r2dtable\" method oecosimu default. hiersimu works almost way adipart, without comparing actual statistic values returned FUN highest possible value (cf. gamma diversity). , cases, difficult ensure additive properties mean statistic values along hierarchy.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"object class \"adipart\" \"hiersimu\" structure oecosimu objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Crist, T.O., Veech, J.., Gering, J.C. Summerville, K.S. (2003). Partitioning species diversity across landscapes regions: hierarchical analysis \\(\\alpha\\), \\(\\beta\\), \\(\\gamma\\)-diversity. . Nat., 162, 734--743. Lande, R. (1996). Statistics partitioning species diversity, similarity among multiple communities. Oikos, 76, 5--13.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l1=1:nrow(mite), l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)), l4=rep(1, nrow(mite)))) ## Let's see in a map par(mfrow=c(1,3)) plot(mite.xy, main=\"l1\", col=as.numeric(levsm$l1)+1, asp = 1) plot(mite.xy, main=\"l2\", col=as.numeric(levsm$l2)+1, asp = 1) plot(mite.xy, main=\"l3\", col=as.numeric(levsm$l3)+1, asp = 1) par(mfrow=c(1,1)) ## Additive diversity partitioning adipart(mite, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -28.979 22.393 21.978 22.429 22.765 0.05 * #> gamma 35.000 0.000 35.000 35.000 35.000 35.000 1.00 #> beta.1 19.886 28.979 12.607 12.235 12.571 13.022 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## the next two define identical models adipart(mite, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, x = levsm, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -54.1889 22.35038 22.10429 22.37143 22.584 0.05 * #> alpha.2 29.750 -25.1503 34.81579 34.50000 34.75000 35.000 0.05 * #> alpha.3 33.000 0.0000 35.00000 35.00000 35.00000 35.000 0.05 * #> gamma 35.000 0.0000 35.00000 35.00000 35.00000 35.000 1.00 #> beta.1 14.636 9.9124 12.46541 12.10000 12.45000 12.779 0.05 * #> beta.2 3.250 15.2208 0.18421 0.00000 0.25000 0.500 0.05 * #> beta.3 2.000 0.0000 0.00000 0.00000 0.00000 0.000 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adipart(mite ~ l2 + l3, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -51.771 22.378947 22.100714 22.400000 22.587 0.05 * #> alpha.2 29.750 -25.578 34.736842 34.500000 34.750000 35.000 0.05 * #> alpha.3 33.000 -17.206 34.973684 34.725000 35.000000 35.000 0.05 * #> gamma 35.000 0.000 35.000000 35.000000 35.000000 35.000 1.00 #> beta.1 14.636 11.165 12.357895 12.025714 12.428571 12.686 0.05 * #> beta.2 3.250 15.455 0.236842 0.000000 0.250000 0.500 0.05 * #> beta.3 2.000 17.206 0.026316 0.000000 0.000000 0.275 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical null model testing ## diversity analysis (similar to adipart) hiersimu(mite, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(y = mite, FUN = diversity, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> level_1 0.76064 -52.201 0.93892 0.93378 0.93866 0.9444 0.05 * #> leve_2 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 hiersimu(mite ~ l2 + l3, levsm, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ l2 + l3, data = levsm, FUN = diversity, #> relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> unit 0.76064 -60.783 0.93927 0.93425 0.93883 0.9435 0.05 * #> l2 0.89736 -116.072 0.99795 0.99617 0.99808 0.9991 0.05 * #> l3 0.92791 -421.205 0.99932 0.99902 0.99935 0.9995 0.05 * #> all 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical testing with the Morisita index morfun <- function(x) dispindmorisita(x)$imst hiersimu(mite ~., levsm, morfun, drop.highest=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ ., data = levsm, FUN = morfun, #> drop.highest = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> l1 0.52070 6.4968 0.35648 0.31140 0.35449 0.3978 0.05 * #> l2 0.60234 12.5142 0.15359 0.08900 0.14525 0.2039 0.05 * #> l3 0.67509 17.6125 -0.19189 -0.26300 -0.20619 -0.1149 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Analysis variance using distance matrices --- partitioning distance matrices among sources variation fitting linear models (e.g., factors, polynomial regression) distance matrices; uses permutation test pseudo-\\(F\\) ratios.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"adonis2(formula, data, permutations = 999, method = \"bray\", sqrt.dist = FALSE, add = FALSE, by = \"terms\", parallel = getOption(\"mc.cores\"), na.action = na.fail, strata = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"formula Model formula. left-hand side (LHS) formula must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist used find dissimilarities. right-hand side (RHS) formula defines independent variables. can continuous variables factors, can transformed within formula, can interactions typical formula. data data frame independent variables, rows order community data matrix dissimilarity matrix named LHS formula. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. method name method used vegdist calculate pairwise distances left hand side formula data frame matrix. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" analyse one-degree--freedom contrasts sequentially, = NULL assess overall significance terms together. arguments passed anova.cca. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. na.action Handling missing values right-hand-side formula (see na.fail explanation alternatives). Missing values allowed left-hand-side. NB, argument subset implemented. strata Groups within constrain permutations. traditional non-movable strata set Blocks permute package, flexible alternatives may appropriate. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"adonis2 function analysis partitioning sums squares using dissimilarities. function based principles McArdle & Anderson (2001) can perform sequential, marginal overall tests. function also allows using additive constants squareroot dissimilarities avoid negative eigenvalues, can also handle semimetric indices (Bray-Curtis) produce negative eigenvalues. adonis2 tests identical anova.cca dbrda. Euclidean distances, tests also identical anova.cca rda. function partitions sums squares multivariate data set, directly analogous MANOVA (multivariate analysis variance). McArdle Anderson (2001) Anderson (2001) refer method “permutational MANOVA” (formerly “nonparametric MANOVA”). , inputs linear predictors, response matrix arbitrary number columns, robust alternative parametric MANOVA ordination methods describing variation attributed different experimental treatments uncontrolled covariates. method also analogous distance-based redundancy analysis algorithmically similar dbrda (Legendre Anderson 1999), provides alternative AMOVA (nested analysis molecular variance, Excoffier, Smouse, Quattro, 1992; amova ade4 package) crossed nested factors.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"function returns anova.cca result object new column partial \\(R^2\\): proportion sum squares total, marginal models (= \"margin\") \\(R^2\\) terms add 1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson (2001, Fig. 4) warns method may confound location dispersion effects: significant differences may caused different within-group variation (dispersion) instead different mean values groups (see Warton et al. 2012 general analysis). However, seems adonis2 less sensitive dispersion effects alternatives (anosim, mrpp). Function betadisper sister function adonis2 study differences dispersion within geometric framework.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson, M.J. 2001. new method non-parametric multivariate analysis variance. Austral Ecology, 26: 32--46. Excoffier, L., P.E. Smouse, J.M. Quattro. 1992. Analysis molecular variance inferred metric distances among DNA haplotypes: Application human mitochondrial DNA restriction data. Genetics, 131:479--491. Legendre, P. M.J. Anderson. 1999. Distance-based redundancy analysis: Testing multispecies responses multifactorial ecological experiments. Ecological Monographs, 69:1--24. McArdle, B.H. M.J. Anderson. 2001. Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology, 82: 290--297. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Martin Henry H. Stevens Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"data(dune) data(dune.env) ## default test by terms adonis2(dune ~ Management*A1, data = dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.2629 0.002 ** #> A1 1 0.4409 0.10256 2.9387 0.012 * #> Management:A1 3 0.5892 0.13705 1.3090 0.217 #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## overall tests adonis2(dune ~ Management*A1, data = dune.env, by = NULL) #> Permutation test for adonis under reduced model #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env, by = NULL) #> Df SumOfSqs R2 F Pr(>F) #> Model 7 2.4987 0.58122 2.3792 0.004 ** #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ### Example of use with strata, for nested (e.g., block) designs. dat <- expand.grid(rep=gl(2,1), NO3=factor(c(0,10)),field=gl(3,1) ) dat #> rep NO3 field #> 1 1 0 1 #> 2 2 0 1 #> 3 1 10 1 #> 4 2 10 1 #> 5 1 0 2 #> 6 2 0 2 #> 7 1 10 2 #> 8 2 10 2 #> 9 1 0 3 #> 10 2 0 3 #> 11 1 10 3 #> 12 2 10 3 Agropyron <- with(dat, as.numeric(field) + as.numeric(NO3)+2) +rnorm(12)/2 Schizachyrium <- with(dat, as.numeric(field) - as.numeric(NO3)+2) +rnorm(12)/2 total <- Agropyron + Schizachyrium dotplot(total ~ NO3, dat, jitter.x=TRUE, groups=field, type=c('p','a'), xlab=\"NO3\", auto.key=list(columns=3, lines=TRUE) ) Y <- data.frame(Agropyron, Schizachyrium) mod <- metaMDS(Y, trace = FALSE) plot(mod) ### Ellipsoid hulls show treatment with(dat, ordiellipse(mod, NO3, kind = \"ehull\", label = TRUE)) ### Spider shows fields with(dat, ordispider(mod, field, lty=3, col=\"red\", label = TRUE)) ### Incorrect (no strata) adonis2(Y ~ NO3, data = dat, permutations = 199) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.06 . #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Correct with strata with(dat, adonis2(Y ~ NO3, data = dat, permutations = 199, strata = field)) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Blocks: strata #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199, strata = field) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.005 ** #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":null,"dir":"Reference","previous_headings":"","what":"Analysis of Similarities — anosim","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Analysis of Similarities — anosim","text":"","code":"anosim(x, grouping, permutations = 999, distance = \"bray\", strata = NULL, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Analysis of Similarities — anosim","text":"x Data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. distance Choice distance metric measures dissimilarity two observations. See vegdist options. used x dissimilarity structure symmetric square matrix. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units. Function anosim operates directly dissimilarity matrix. suitable dissimilarity matrix produced functions dist vegdist. method philosophically allied NMDS ordination (monoMDS), uses rank order dissimilarity values. two groups sampling units really different species composition, compositional dissimilarities groups greater within groups. anosim statistic \\(R\\) based difference mean ranks groups (\\(r_B\\)) within groups (\\(r_W\\)): $$R = (r_B - r_W)/(N (N-1) / 4)$$ divisor chosen \\(R\\) interval \\(-1 \\dots +1\\), value \\(0\\) indicating completely random grouping. statistical significance observed \\(R\\) assessed permuting grouping vector obtain empirical distribution \\(R\\) null-model. See permutations additional details permutation tests Vegan. distribution simulated values can inspected permustats function. function summary plot methods. show valuable information assess validity method: function assumes ranked dissimilarities within groups equal median range. plot method uses boxplot options notch=TRUE varwidth=TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Analysis of Similarities — anosim","text":"function returns list class \"anosim\" following items: call Function call. statistic value ANOSIM statistic \\(R\\) signif Significance permutation. perm Permutation values \\(R\\). distribution permutation values can inspected function permustats. class.vec Factor value dissimilarities classes class name corresponding dissimilarity within class. dis.rank Rank dissimilarity entry. dissimilarity name dissimilarity index: \"method\" entry dist object. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Analysis of Similarities — anosim","text":"Clarke, K. R. (1993). Non-parametric multivariate analysis changes community structure. Australian Journal Ecology 18, 117--143. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Analysis of Similarities — anosim","text":"Jari Oksanen, help Peter R. Minchin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Analysis of Similarities — anosim","text":"anosim function can confound differences groups dispersion within groups results can difficult interpret (cf. Warton et al. 2012). function returns lot information ease studying performance. anosim models analysed adonis2 seems robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Analysis of Similarities — anosim","text":"","code":"data(dune) data(dune.env) dune.dist <- vegdist(dune) dune.ano <- with(dune.env, anosim(dune.dist, Management)) summary(dune.ano) #> #> Call: #> anosim(x = dune.dist, grouping = Management) #> Dissimilarity: bray #> #> ANOSIM statistic R: 0.2579 #> Significance: 0.009 #> #> Permutation: free #> Number of permutations: 999 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.121 0.164 0.205 0.242 #> #> Dissimilarity ranks between and within classes: #> 0% 25% 50% 75% 100% N #> Between 4 58.50 104.00 145.500 188.0 147 #> BF 5 15.25 25.50 41.250 57.0 3 #> HF 1 7.25 46.25 68.125 89.5 10 #> NM 6 64.75 124.50 156.250 181.0 15 #> SF 3 32.75 53.50 99.250 184.0 15 #> plot(dune.ano) #> Warning: some notches went outside hinges ('box'): maybe set notch=FALSE"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function performs ANOVA like permutation test Constrained Correspondence Analysis (cca), Redundancy Analysis (rda) distance-based Redundancy Analysis (dbRDA, dbrda) assess significance constraints.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"# S3 method for cca anova(object, ..., permutations = how(nperm=999), by = NULL, model = c(\"reduced\", \"direct\", \"full\"), parallel = getOption(\"mc.cores\"), strata = NULL, cutoff = 1, scope = NULL) # S3 method for cca permutest(x, permutations = how(nperm = 99), model = c(\"reduced\", \"direct\", \"full\"), by = NULL, first = FALSE, strata = NULL, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"object One several result objects cca, rda, dbrda capscale. several result objects, compared order supplied. single object, test specified overall test given. x single ordination result object. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Setting = \"axis\" assess significance constrained axis, setting = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" assess sequentially one-degree--freedom contrasts split factors. model Permutation model: model=\"direct\" permutes community data, model=\"reduced\" permutes residuals community data Conditions (partial model), model = \"full\" permutes residuals Conditions Constraints. parallel Use parallel processing given number cores. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. error use permutations matrix, defines blocks. legacy argument deprecated future: use permutations = (..., blocks) instead. cutoff effective =\"axis\" stops permutations axis equals exceeds cutoff \\(p\\)-value. scope effective =\"margin\" can used select marginal terms testing. default test marginal terms drop.scope. first Analyse significance first axis. ... Parameters passed functions. anova.cca passes arguments permutest.cca. anova = \"axis\" can use argument cutoff (defaults 1) stops permutations exceeding given level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Functions anova.cca permutest.cca implement ANOVA like permutation tests joint effect constraints cca, rda, dbrda capscale. Function anova intended user-friendly alternative permutest (real workhorse). Function anova can analyse sequence constrained ordination models. analysis based differences residual deviance permutations nested models. default test sum constrained eigenvalues. Setting first = TRUE perform test first constrained eigenvalue. Argument first can set either anova.cca permutest.cca. also possible perform significance tests axis term (constraining variable) using argument anova.cca. Setting = \"axis\" perform separate significance tests constrained axis. previous constrained axes used conditions (“partialled ”) test first constrained eigenvalues performed (Legendre et al. 2011). can stop permutation tests exceeding given significance level argument cutoff speed calculations large models. Setting = \"terms\" perform separate significance test term (constraining variable). terms assessed sequentially first last, order terms influence significances. Setting = \"onedf\" perform similar sequential test one-degree--freedom effects, multi-level factors split contrasts. Setting = \"margin\" perform separate significance test marginal term model terms. marginal test also accepts scope argument drop.scope can character vector term labels analysed, fitted model lower scope. marginal effects also known “Type III” effects, current function evaluates marginal terms. , instance, ignore main effects included interaction terms. calculating pseudo-\\(F\\), terms compared residual full model. Community data permuted choice model=\"direct\", residuals partial CCA/ RDA/ dbRDA choice model=\"reduced\" (default). partial CCA/ RDA/ dbRDA stage, model=\"reduced\" simply permutes data equivalent model=\"direct\". test statistic “pseudo-\\(F\\)”, ratio constrained unconstrained total Inertia (Chi-squares, variances something similar), divided respective degrees freedom. conditions (“partial” terms), sum eigenvalues remains constant, pseudo-\\(F\\) eigenvalues give equal results. partial CCA/ RDA/ dbRDA, effect conditioning variables (“covariables”) removed permutation, total Chi-square fixed, test based pseudo-\\(F\\) differ test based plain eigenvalues.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function anova.cca calls permutest.cca fills anova table. Additional attributes Random.seed (random seeds used), control (permutation design, see ) F.perm (permuted test statistics).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Legendre, P. Legendre, L. (2012). Numerical Ecology. 3rd English ed. Elsevier. Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"data(dune, dune.env) mod <- cca(dune ~ Moisture + Management, dune.env) ## overall test anova(mod) #> Permutation test for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Model 6 1.0024 1.9515 0.003 ** #> Residual 13 1.1129 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## tests for individual terms anova(mod, by=\"term\") #> Permutation test for cca under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.62831 2.4465 0.001 *** #> Management 3 0.37407 1.4565 0.049 * #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 anova(mod, by=\"margin\") #> Permutation test for cca under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.39854 1.5518 0.036 * #> Management 3 0.37407 1.4565 0.053 . #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential test for contrasts anova(mod, by = \"onedf\") #> Permutation test for cca under reduced model #> Sequential test for contrasts #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture.L 1 0.41081 4.7988 0.001 *** #> Moisture.Q 1 0.11261 1.3154 0.167 #> Moisture.C 1 0.10489 1.2253 0.226 #> ManagementHF 1 0.08849 1.0337 0.359 #> ManagementNM 1 0.20326 2.3744 0.011 * #> ManagementSF 1 0.08231 0.9615 0.455 #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## test for adding all environmental variables anova(mod, cca(dune ~ ., dune.env)) #> Permutation tests for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model 1: dune ~ Moisture + Management #> Model 2: dune ~ A1 + Moisture + Management + Use + Manure #> ResDf ResChiSquare Df ChiSquare F Pr(>F) #> 1 13 1.1129 #> 2 7 0.6121 6 0.50079 0.9545 0.532"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Averaged Subsampled Dissimilarity Matrices — avgdist","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function computes dissimilarity matrix dataset multiple times using vegdist randomly subsampling dataset time. subsampled iterations averaged (mean) provide distance matrix represents average multiple subsampling iterations. emulates behavior distance matrix calculator within Mothur microbial ecology toolkit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"avgdist(x, sample, distfun = vegdist, meanfun = mean, transf = NULL, iterations = 100, dmethod = \"bray\", diag = TRUE, upper = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"x Community data matrix. sample subsampling depth used iteration. Samples meet threshold removed analysis, identity returned user stdout. distfun dissimilarity matrix function used. Default vegan vegdist meanfun calculation use average (mean median). transf Option transforming count data calculating distance matrix. base transformation option can used (e.g. sqrt) iterations number random iterations perform averaging. Default 100 iterations. dmethod Dissimilarity index used specified dissimilarity matrix function. Default Bray-Curtis diag, upper Return dissimilarities diagonal upper triangle. NB. default differs vegdist returns symmetric \"dist\" structure instead lower diagonal. However, object accessed matrix indices unless cast matrix .matrix. ... additional arguments add distance function mean/median function specified.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"Geoffrey Hannigan, minor tweaks Gavin L. Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function builds function rrarefy additional distance matrix function (e.g. vegdist) add meaningful representations distances among randomly subsampled datasets presenting average multiple random iterations. function runs using vegdist. functionality utilized Mothur standalone microbial ecology toolkit .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"# Import an example count dataset data(BCI) # Test the base functionality mean.avg.dist <- avgdist(BCI, sample = 50, iterations = 10) # Test the transformation function mean.avg.dist.t <- avgdist(BCI, sample = 50, iterations = 10, transf = sqrt) # Test the median functionality median.avg.dist <- avgdist(BCI, sample = 50, iterations = 10, meanfun = median) # Print the resulting tables head(as.matrix(mean.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 #> 1 0.000 0.556 0.606 0.648 0.620 0.624 0.610 0.606 0.654 0.624 0.638 0.648 0.708 #> 2 0.556 0.000 0.568 0.568 0.632 0.568 0.532 0.554 0.576 0.578 0.588 0.560 0.678 #> 3 0.606 0.568 0.000 0.582 0.578 0.618 0.600 0.568 0.610 0.574 0.610 0.622 0.738 #> 4 0.648 0.568 0.582 0.000 0.624 0.654 0.610 0.572 0.594 0.594 0.596 0.636 0.672 #> 5 0.620 0.632 0.578 0.624 0.000 0.602 0.662 0.626 0.670 0.606 0.678 0.716 0.754 #> 6 0.624 0.568 0.618 0.654 0.602 0.000 0.598 0.566 0.616 0.644 0.576 0.568 0.706 #> 14 15 16 17 18 19 20 21 22 23 24 25 26 #> 1 0.630 0.648 0.600 0.646 0.744 0.682 0.626 0.600 0.646 0.700 0.658 0.620 0.620 #> 2 0.600 0.592 0.554 0.564 0.660 0.584 0.598 0.606 0.544 0.642 0.574 0.624 0.592 #> 3 0.628 0.612 0.616 0.646 0.730 0.660 0.646 0.568 0.636 0.722 0.652 0.672 0.660 #> 4 0.602 0.602 0.618 0.634 0.666 0.622 0.636 0.656 0.568 0.642 0.614 0.654 0.638 #> 5 0.610 0.642 0.630 0.760 0.766 0.698 0.654 0.642 0.738 0.702 0.652 0.688 0.652 #> 6 0.618 0.696 0.626 0.618 0.668 0.634 0.664 0.648 0.630 0.694 0.608 0.682 0.674 #> 27 28 29 30 31 32 33 34 35 36 37 38 39 #> 1 0.702 0.684 0.674 0.634 0.680 0.724 0.692 0.684 0.744 0.666 0.694 0.716 0.700 #> 2 0.632 0.586 0.568 0.582 0.646 0.606 0.618 0.634 0.722 0.638 0.590 0.590 0.616 #> 3 0.668 0.660 0.654 0.676 0.660 0.674 0.694 0.700 0.764 0.684 0.628 0.684 0.708 #> 4 0.644 0.596 0.612 0.608 0.662 0.594 0.610 0.632 0.730 0.648 0.562 0.588 0.636 #> 5 0.712 0.752 0.728 0.734 0.676 0.710 0.698 0.768 0.830 0.678 0.690 0.752 0.760 #> 6 0.628 0.640 0.630 0.668 0.668 0.636 0.662 0.682 0.768 0.678 0.658 0.664 0.700 #> 40 41 42 43 44 45 46 47 48 49 50 #> 1 0.714 0.686 0.624 0.688 0.654 0.668 0.668 0.650 0.676 0.694 0.676 #> 2 0.638 0.670 0.644 0.654 0.626 0.664 0.648 0.638 0.648 0.678 0.672 #> 3 0.716 0.702 0.634 0.672 0.668 0.646 0.720 0.730 0.712 0.698 0.704 #> 4 0.680 0.652 0.618 0.680 0.698 0.696 0.658 0.666 0.658 0.696 0.708 #> 5 0.828 0.688 0.654 0.658 0.690 0.654 0.772 0.720 0.704 0.692 0.656 #> 6 0.724 0.686 0.678 0.670 0.666 0.698 0.734 0.692 0.714 0.734 0.704 head(as.matrix(mean.avg.dist.t)) #> 1 2 3 4 5 6 7 #> 1 0.0000000 0.5073247 0.5451162 0.5511308 0.5863431 0.5693314 0.5561033 #> 2 0.5073247 0.0000000 0.5076177 0.5454488 0.5590474 0.5361407 0.5075526 #> 3 0.5451162 0.5076177 0.0000000 0.5169507 0.5442375 0.5536933 0.5580432 #> 4 0.5511308 0.5454488 0.5169507 0.0000000 0.5556915 0.6029668 0.5894129 #> 5 0.5863431 0.5590474 0.5442375 0.5556915 0.0000000 0.5730661 0.5904097 #> 6 0.5693314 0.5361407 0.5536933 0.6029668 0.5730661 0.0000000 0.5161496 #> 8 9 10 11 12 13 14 #> 1 0.5673559 0.5627535 0.5792425 0.5729159 0.6020539 0.7016723 0.5490011 #> 2 0.5366592 0.5380985 0.5623811 0.5321790 0.5515688 0.6779465 0.5572373 #> 3 0.5073755 0.5536056 0.5389882 0.5801096 0.5868026 0.7260347 0.5816036 #> 4 0.5259962 0.5706750 0.5613498 0.5523447 0.5753112 0.6804780 0.5838035 #> 5 0.5484180 0.5855322 0.5881431 0.6281267 0.6276212 0.7377994 0.5875698 #> 6 0.5529963 0.5552535 0.6215621 0.5618841 0.5336650 0.6716372 0.5641578 #> 15 16 17 18 19 20 21 #> 1 0.6050878 0.5605091 0.6262520 0.6790944 0.6158975 0.5894641 0.6238942 #> 2 0.5492945 0.5571279 0.5683865 0.6719160 0.5864794 0.5817164 0.5785775 #> 3 0.5651808 0.5907658 0.5986803 0.7180656 0.6208023 0.5831755 0.6141896 #> 4 0.5846365 0.5995058 0.6019378 0.6723126 0.6185437 0.5788246 0.6204008 #> 5 0.5886468 0.6313116 0.6908350 0.7310610 0.6738026 0.6056946 0.6248085 #> 6 0.6245763 0.5734551 0.5897524 0.6553604 0.5945132 0.6253476 0.6236726 #> 22 23 24 25 26 27 28 #> 1 0.6087932 0.6864876 0.5674543 0.6124061 0.6000094 0.6287013 0.6246696 #> 2 0.5851407 0.6506247 0.5554757 0.5519726 0.6143138 0.5810440 0.5887911 #> 3 0.6301892 0.7101043 0.5608707 0.5964616 0.6536689 0.6015372 0.6254962 #> 4 0.6116276 0.6570546 0.5806733 0.5942875 0.6756654 0.6170196 0.6088559 #> 5 0.6692645 0.7281868 0.6019467 0.6240673 0.6717579 0.6824310 0.6781349 #> 6 0.5779917 0.6752783 0.5474248 0.6048853 0.6297599 0.6032439 0.6025756 #> 29 30 31 32 33 34 35 #> 1 0.5877126 0.5931557 0.6122546 0.6435203 0.6114740 0.6243358 0.7012874 #> 2 0.5493477 0.5852872 0.5988185 0.6153332 0.5921568 0.6244034 0.6905708 #> 3 0.5678376 0.6210073 0.6158757 0.6377790 0.6371270 0.6555867 0.7323038 #> 4 0.5759931 0.5912841 0.5983615 0.6066556 0.6016788 0.6087876 0.6950808 #> 5 0.6276522 0.6379576 0.6034350 0.6457655 0.6563073 0.6866576 0.7235042 #> 6 0.5816252 0.6242943 0.6152092 0.6137642 0.6116231 0.6246531 0.7312678 #> 36 37 38 39 40 41 42 #> 1 0.5974168 0.6467898 0.6716840 0.6643855 0.6683070 0.6353745 0.6004874 #> 2 0.5903695 0.5752574 0.6187077 0.6192344 0.6573943 0.6571387 0.6027516 #> 3 0.5687077 0.6126409 0.6617885 0.6959650 0.6909293 0.6525653 0.6014759 #> 4 0.6194344 0.5761296 0.6102164 0.6295049 0.6560977 0.5989171 0.6281164 #> 5 0.6056532 0.6441953 0.6593381 0.7110348 0.7252049 0.6568578 0.6156351 #> 6 0.6109414 0.6067715 0.6656711 0.6870593 0.6686455 0.6769139 0.6426746 #> 43 44 45 46 47 48 49 #> 1 0.6310309 0.5900500 0.5774603 0.6474425 0.6488217 0.6157111 0.6070289 #> 2 0.6028820 0.6126007 0.6208170 0.6652427 0.6134982 0.6243098 0.6319083 #> 3 0.6162330 0.6398741 0.6039060 0.7021972 0.6677691 0.6678864 0.6696487 #> 4 0.6197236 0.6379939 0.6156272 0.6582549 0.6062615 0.6139683 0.6466872 #> 5 0.6203780 0.6542059 0.6040162 0.7417980 0.6946591 0.6954255 0.6537119 #> 6 0.6386277 0.6418793 0.6306854 0.6948586 0.6326248 0.6678749 0.6692019 #> 50 #> 1 0.5867459 #> 2 0.6227903 #> 3 0.6365131 #> 4 0.6260672 #> 5 0.6250031 #> 6 0.6423431 head(as.matrix(median.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 #> 1 0.00 0.56 0.60 0.62 0.66 0.63 0.58 0.62 0.65 0.61 0.59 0.65 0.74 0.64 0.64 #> 2 0.56 0.00 0.53 0.59 0.64 0.63 0.60 0.58 0.58 0.60 0.59 0.62 0.68 0.58 0.61 #> 3 0.60 0.53 0.00 0.61 0.63 0.58 0.56 0.57 0.60 0.55 0.60 0.64 0.76 0.66 0.61 #> 4 0.62 0.59 0.61 0.00 0.64 0.66 0.62 0.58 0.57 0.58 0.63 0.62 0.66 0.55 0.57 #> 5 0.66 0.64 0.63 0.64 0.00 0.57 0.65 0.64 0.62 0.62 0.68 0.67 0.76 0.66 0.65 #> 6 0.63 0.63 0.58 0.66 0.57 0.00 0.58 0.62 0.61 0.66 0.60 0.59 0.74 0.69 0.70 #> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 #> 1 0.63 0.65 0.76 0.67 0.65 0.65 0.66 0.68 0.66 0.66 0.62 0.66 0.65 0.66 0.66 #> 2 0.62 0.59 0.69 0.61 0.56 0.64 0.62 0.67 0.60 0.58 0.62 0.64 0.63 0.56 0.63 #> 3 0.61 0.64 0.72 0.69 0.64 0.62 0.62 0.74 0.65 0.65 0.66 0.64 0.61 0.60 0.64 #> 4 0.63 0.62 0.67 0.61 0.60 0.68 0.59 0.67 0.62 0.60 0.66 0.60 0.62 0.59 0.62 #> 5 0.65 0.73 0.81 0.70 0.65 0.64 0.72 0.69 0.71 0.67 0.71 0.68 0.72 0.70 0.67 #> 6 0.63 0.60 0.68 0.63 0.66 0.68 0.63 0.70 0.61 0.64 0.66 0.63 0.65 0.62 0.69 #> 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 #> 1 0.65 0.69 0.66 0.66 0.74 0.65 0.68 0.69 0.70 0.68 0.67 0.63 0.63 0.64 0.67 #> 2 0.62 0.61 0.61 0.63 0.75 0.66 0.60 0.60 0.67 0.68 0.68 0.62 0.64 0.64 0.69 #> 3 0.63 0.61 0.69 0.68 0.78 0.67 0.61 0.65 0.68 0.72 0.71 0.65 0.66 0.64 0.65 #> 4 0.59 0.65 0.61 0.66 0.79 0.61 0.57 0.61 0.65 0.66 0.68 0.63 0.66 0.68 0.71 #> 5 0.65 0.71 0.70 0.75 0.82 0.70 0.65 0.73 0.79 0.76 0.72 0.64 0.73 0.70 0.68 #> 6 0.68 0.68 0.67 0.68 0.82 0.72 0.71 0.69 0.71 0.75 0.74 0.69 0.67 0.69 0.72 #> 46 47 48 49 50 #> 1 0.69 0.66 0.66 0.66 0.67 #> 2 0.69 0.62 0.67 0.63 0.65 #> 3 0.73 0.70 0.70 0.68 0.72 #> 4 0.68 0.65 0.65 0.69 0.64 #> 5 0.76 0.70 0.75 0.72 0.70 #> 6 0.72 0.71 0.73 0.72 0.71 # Run example to illustrate low variance of mean, median, and stdev results # Mean and median std dev are around 0.05 sdd <- avgdist(BCI, sample = 50, iterations = 100, meanfun = sd) summary(mean.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4640 0.6180 0.6540 0.6543 0.6900 0.8360 summary(median.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4600 0.6100 0.6500 0.6501 0.6900 0.8600 summary(sdd) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.03849 0.05617 0.05921 0.05919 0.06208 0.07661 # Test for when subsampling depth excludes some samples # Return samples that are removed for not meeting depth filter depth.avg.dist <- avgdist(BCI, sample = 450, iterations = 10) #> Warning: The following sampling units were removed because they were below sampling depth: 1, 2, 6, 7, 8, 9, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 31, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 # Print the result depth.avg.dist #> 3 4 5 10 15 30 32 #> 3 0.0000000 0.3282222 0.3635556 0.2942222 0.3562222 0.4555556 0.4726667 #> 4 0.3282222 0.0000000 0.3775556 0.3173333 0.3480000 0.3908889 0.4273333 #> 5 0.3635556 0.3775556 0.0000000 0.3900000 0.3995556 0.4968889 0.5231111 #> 10 0.2942222 0.3173333 0.3900000 0.0000000 0.3135556 0.4266667 0.4200000 #> 15 0.3562222 0.3480000 0.3995556 0.3135556 0.0000000 0.4535556 0.4657778 #> 30 0.4555556 0.3908889 0.4968889 0.4266667 0.4535556 0.0000000 0.3811111 #> 32 0.4726667 0.4273333 0.5231111 0.4200000 0.4657778 0.3811111 0.0000000 #> 35 0.6591111 0.6266667 0.6966667 0.6817778 0.6573333 0.5157778 0.5893333 #> 40 0.5624444 0.5286667 0.6413333 0.5240000 0.5628889 0.4540000 0.4057778 #> 35 40 #> 3 0.6591111 0.5624444 #> 4 0.6266667 0.5286667 #> 5 0.6966667 0.6413333 #> 10 0.6817778 0.5240000 #> 15 0.6573333 0.5628889 #> 30 0.5157778 0.4540000 #> 32 0.5893333 0.4057778 #> 35 0.0000000 0.4280000 #> 40 0.4280000 0.0000000"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":null,"dir":"Reference","previous_headings":"","what":"Beals Smoothing and Degree of Absence — beals","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing replaces entry community data probability target species occurring particular site, based joint occurrences target species species actually occur site. Swan's (1970) degree absence applies Beals smoothing zero items long zeros replaced smoothed values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"beals(x, species = NA, reference = x, type = 0, include = TRUE) swan(x, maxit = Inf, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Beals Smoothing and Degree of Absence — beals","text":"x Community data frame matrix. species Column index used compute Beals function single species. default (NA) indicates function computed species. reference Community data frame matrix used compute joint occurrences. default, x used reference compute joint occurrences. type Numeric. Specifies abundance values used function beals. See details explanation. include logical flag indicates whether target species included computing mean conditioned probabilities. original Beals (1984) definition equivalent include=TRUE, formulation Münzbergová Herben equal include=FALSE. maxit Maximum number iterations. default Inf means iterations continued zeros number zeros change. Probably maxit = 1 makes sense addition default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing estimated probability \\(p_{ij}\\) species \\(j\\) occurs site \\(\\). defined \\(p_{ij} = \\frac{1}{S_i} \\sum_k \\frac{N_{jk} I_{ik}}{N_k}\\), \\(S_i\\) number species site \\(\\), \\(N_{jk}\\) number joint occurrences species \\(j\\) \\(k\\), \\(N_k\\) number occurrences species \\(k\\), \\(\\) incidence (0 1) species (last term usually omitted equation, necessary). \\(N_{jk}\\) can interpreted mean conditional probability, beals function can interpreted mean conditioned probabilities (De Cáceres & Legendre 2008). present function generalized abundance values (De Cáceres & Legendre 2008). type argument specifies abundance values used. type = 0 presence/absence mode. type = 1 abundances reference (x) used compute conditioned probabilities. type = 2 abundances x used compute weighted averages conditioned probabilities. type = 3 abundances used compute conditioned probabilities weighted averages. Beals smoothing originally suggested method data transformation remove excessive zeros (Beals 1984, McCune 1994). However, suitable method purpose since maintain information species presences: species may higher probability occurrence site occur sites occurs. Moreover, regularizes data strongly. method may useful identifying species belong species pool (Ewald 2002) identify suitable unoccupied patches metapopulation analysis (Münzbergová & Herben 2004). case, function called include=FALSE cross-validation smoothing species; argument species can used one species studied. Swan (1970) suggested replacing zero values degrees absence species community data matrix. Swan expressed method terms similarity matrix, equivalent applying Beals smoothing zero values, step shifting smallest initially non-zero item value one, repeating many times zeros left data. actually similar extended dissimilarities (implemented function stepacross), rarely used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Beals Smoothing and Degree of Absence — beals","text":"function returns transformed data matrix vector Beals smoothing requested single species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals, E.W. 1984. Bray-Curtis ordination: effective strategy analysis multivariate ecological data. Pp. 1--55 : MacFadyen, . & E.D. Ford [eds.] Advances Ecological Research, 14. Academic Press, London. De Cáceres, M. & Legendre, P. 2008. Beals smoothing revisited. Oecologia 156: 657--669. Ewald, J. 2002. probabilistic approach estimating species pools large compositional matrices. J. Veg. Sci. 13: 191--198. McCune, B. 1994. Improving community ordination Beals smoothing function. Ecoscience 1: 82--86. Münzbergová, Z. & Herben, T. 2004. Identification suitable unoccupied habitats metapopulation studies using co-occurrence species. Oikos 105: 408--414. Swan, J.M.. 1970. examination ordination problems use simulated vegetational data. Ecology 51: 89--102.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Beals Smoothing and Degree of Absence — beals","text":"Miquel De Cáceres Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"data(dune) ## Default x <- beals(dune) ## Remove target species x <- beals(dune, include = FALSE) ## Smoothed values against presence or absence of species pa <- decostand(dune, \"pa\") boxplot(as.vector(x) ~ unlist(pa), xlab=\"Presence\", ylab=\"Beals\") ## Remove the bias of tarbet species: Yields lower values. beals(dune, type =3, include = FALSE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord #> 1 0.49590853 0.38333415 0.01157407 0.4923280 0.30827883 0.4935662 0.43263047 #> 2 0.47083676 0.39501120 0.03361524 0.4718807 0.34723984 0.4917791 0.42000984 #> 3 0.34063019 0.52738394 0.01520046 0.5309152 0.21609954 0.4033301 0.33010938 #> 4 0.30816435 0.51198853 0.02876960 0.5971801 0.21542662 0.4398775 0.35732610 #> 5 0.59949785 0.27622698 0.06632771 0.3349203 0.48876285 0.4322142 0.44309579 #> 6 0.58819821 0.26299306 0.05967771 0.2700508 0.53154426 0.3696613 0.39760652 #> 7 0.56496165 0.29412293 0.05329633 0.3403047 0.48010987 0.4051777 0.40471531 #> 8 0.21230502 0.66906674 0.02588333 0.5187956 0.16247716 0.2720122 0.21219877 #> 9 0.30323659 0.59744543 0.02213662 0.5792855 0.21896113 0.3292320 0.28613526 #> 10 0.54083871 0.26902092 0.07349127 0.3372958 0.42671693 0.4705094 0.42934344 #> 11 0.40509331 0.31656550 0.10259239 0.3185489 0.38766111 0.3713794 0.31413659 #> 12 0.21008725 0.66278454 0.03625297 0.5753377 0.20078932 0.2802946 0.22974415 #> 13 0.21850759 0.68239707 0.02191119 0.6404427 0.16737280 0.2939740 0.24942466 #> 14 0.13570397 0.76284476 0.02298398 0.4107645 0.12128973 0.1682755 0.13757552 #> 15 0.09168815 0.79412733 0.02538032 0.4505613 0.10117099 0.1420251 0.09794548 #> 16 0.06335463 0.87877202 0.00742115 0.5232448 0.05538377 0.1516354 0.09458531 #> 17 0.55254140 0.07330247 0.29233391 0.1013889 0.69331132 0.3129358 0.34982363 #> 18 0.37751017 0.34451209 0.08535723 0.2838834 0.36918166 0.3676424 0.30478244 #> 19 0.29826049 0.25952255 0.35137675 0.1934048 0.51929869 0.2237843 0.18074796 #> 20 0.05429986 0.76675441 0.06144615 0.4063662 0.10738280 0.1450721 0.06706410 #> Chenalbu Cirsarve Comapalu Eleopalu Elymrepe Empenigr #> 1 0.025132275 0.09504980 0.000000000 0.05592045 0.4667439 0.00000000 #> 2 0.043866562 0.08570299 0.026548839 0.08656209 0.4407282 0.01829337 #> 3 0.065338638 0.08967477 0.031898812 0.16099072 0.4137888 0.01074444 #> 4 0.057970906 0.12920228 0.039859621 0.16112450 0.4399661 0.02527165 #> 5 0.026434737 0.05520104 0.015892090 0.05419613 0.3575948 0.03029752 #> 6 0.021256367 0.03223112 0.030347896 0.08784329 0.3138879 0.03093489 #> 7 0.038467708 0.04706743 0.017083997 0.06694311 0.3586644 0.02304603 #> 8 0.063278453 0.06688407 0.100703044 0.29777644 0.3046956 0.02102222 #> 9 0.069879277 0.07647268 0.045830682 0.19018562 0.3523460 0.01838883 #> 10 0.025686639 0.06037513 0.029746617 0.07787078 0.3736128 0.03425596 #> 11 0.021234732 0.05778318 0.035740922 0.11146095 0.2884798 0.07310076 #> 12 0.103543341 0.07799259 0.045375827 0.19518888 0.3354080 0.03413656 #> 13 0.122547745 0.07905124 0.056084315 0.22437598 0.3511708 0.01840390 #> 14 0.042990591 0.03618335 0.241811837 0.55982776 0.1428372 0.01989756 #> 15 0.035609053 0.04022968 0.198176675 0.53973883 0.1462975 0.02215971 #> 16 0.056246994 0.05184498 0.201352298 0.51523810 0.1832397 0.00742115 #> 17 0.007716049 0.01049383 0.009876543 0.02777778 0.1929470 0.21968254 #> 18 0.014640428 0.04454602 0.042890320 0.17341352 0.2651538 0.06763669 #> 19 0.019591245 0.03668466 0.031845637 0.12592768 0.1422725 0.26011417 #> 20 0.037623741 0.03453783 0.185726965 0.58476297 0.1168700 0.05905666 #> Hyporadi Juncarti Juncbufo Lolipere Planlanc Poaprat Poatriv #> 1 0.07702746 0.14794933 0.1987270 0.9226190 0.40103107 0.9863946 0.8826329 #> 2 0.07454127 0.13017869 0.2070478 0.8272395 0.40700777 0.8972046 0.8288385 #> 3 0.05562332 0.22291082 0.2544828 0.7205525 0.27933493 0.8083020 0.8383185 #> 4 0.06985986 0.21320122 0.2318440 0.7197924 0.25797285 0.7940926 0.8197302 #> 5 0.10245961 0.10406655 0.2164230 0.8380779 0.52628928 0.9035899 0.8094632 #> 6 0.11463153 0.11631772 0.2166255 0.8000021 0.58765018 0.8666677 0.7782619 #> 7 0.10837376 0.11293676 0.2110045 0.8053380 0.51905808 0.8925059 0.8018775 #> 8 0.06550319 0.33219882 0.2323566 0.5403355 0.20596764 0.6160461 0.7101299 #> 9 0.05343787 0.23134366 0.2675624 0.6874068 0.25274756 0.7523318 0.8247374 #> 10 0.13692492 0.09080902 0.1678040 0.8102783 0.52588347 0.8915882 0.7543592 #> 11 0.18108995 0.13478872 0.1656396 0.7180948 0.47012501 0.8062720 0.6404351 #> 12 0.06777311 0.27306206 0.3231724 0.5875943 0.22110550 0.6932541 0.8199960 #> 13 0.04250245 0.28204736 0.3339728 0.5714581 0.18153869 0.7063028 0.7993754 #> 14 0.04665747 0.46685537 0.1206518 0.3356311 0.14342002 0.3817081 0.5090703 #> 15 0.05040404 0.51561767 0.1370235 0.3689922 0.13523214 0.4078219 0.5263520 #> 16 0.01731602 0.54304667 0.1776781 0.3561752 0.07269979 0.4124222 0.6071083 #> 17 0.36492870 0.03333333 0.1038156 0.5858415 0.59641331 0.7434618 0.5036834 #> 18 0.17491099 0.18956922 0.1376386 0.7124388 0.45087176 0.7368632 0.5859071 #> 19 0.39145281 0.13543701 0.1127832 0.4289185 0.40784415 0.5548077 0.3605827 #> 20 0.07795311 0.53056145 0.1192488 0.3262685 0.13059496 0.3662817 0.4523029 #> Ranuflam Rumeacet Sagiproc Salirepe Scorautu Trifprat Trifrepe #> 1 0.08105273 0.3160963 0.3371121 0.02729885 0.8898317 0.21701279 0.8782576 #> 2 0.13042865 0.3031318 0.3302063 0.05983781 0.9349640 0.20673650 0.9125666 #> 3 0.22632936 0.2909068 0.4204104 0.06065155 0.9036443 0.14654749 0.8817430 #> 4 0.21909541 0.2610006 0.4191908 0.07579199 0.9204237 0.12896524 0.8943213 #> 5 0.08063087 0.3979230 0.2612828 0.07589611 0.9576838 0.34808957 0.9142360 #> 6 0.10909966 0.4330705 0.2539380 0.08921540 0.9590466 0.35423465 0.9110822 #> 7 0.10541081 0.4113622 0.2954682 0.07094548 0.9550487 0.32489503 0.9171688 #> 8 0.40134447 0.2331043 0.4009544 0.11569906 0.8755515 0.09897600 0.8002526 #> 9 0.26006489 0.3464870 0.4531178 0.07351827 0.9145996 0.16269563 0.8714833 #> 10 0.10355742 0.3226025 0.2732735 0.09037489 0.9568824 0.26807372 0.9003730 #> 11 0.13269569 0.2753878 0.3673397 0.16465286 0.9442707 0.19982976 0.8979262 #> 12 0.29873222 0.3507140 0.5122033 0.08041977 0.9377963 0.13849854 0.9079979 #> 13 0.33309468 0.3107471 0.5131337 0.06572594 0.9255312 0.11199114 0.8841739 #> 14 0.64674225 0.1241545 0.2528665 0.15917563 0.8477706 0.06176123 0.6485949 #> 15 0.64449081 0.1459458 0.3151199 0.17750323 0.8430677 0.05831084 0.7170446 #> 16 0.66893881 0.1508409 0.3480368 0.15783292 0.8131968 0.03916718 0.6776273 #> 17 0.03549383 0.2913631 0.3292030 0.25651777 0.9839744 0.27593101 0.8141660 #> 18 0.18805395 0.2668100 0.3154533 0.17191937 0.9554011 0.20461193 0.8600701 #> 19 0.14551892 0.1831168 0.4798245 0.36429493 0.9902041 0.11966159 0.8147968 #> 20 0.62796060 0.1098600 0.3105433 0.21674174 0.8457313 0.03708580 0.6350394 #> Vicilath Bracruta Callcusp #> 1 0.17244420 0.7476589 0.003527337 #> 2 0.18494940 0.7415172 0.034597921 #> 3 0.12833142 0.7666969 0.075789630 #> 4 0.12550967 0.7919786 0.081110164 #> 5 0.16693075 0.8079786 0.023129027 #> 6 0.18035860 0.8387650 0.040981168 #> 7 0.19027523 0.8089116 0.024070054 #> 8 0.10213052 0.8109194 0.201958942 #> 9 0.08630413 0.7972178 0.092982775 #> 10 0.23383453 0.7660374 0.033527777 #> 11 0.24317802 0.8182692 0.043950322 #> 12 0.08049055 0.8061715 0.100954127 #> 13 0.06604026 0.7465509 0.122856392 #> 14 0.07857237 0.7238162 0.347514804 #> 15 0.07370069 0.7997141 0.381379395 #> 16 0.03353260 0.8029953 0.364128496 #> 17 0.20728700 0.7635487 0.009876543 #> 18 0.26222869 0.8397471 0.109959916 #> 19 0.18188455 0.8161275 0.082361157 #> 20 0.09111967 0.8397124 0.397924041 ## Uses abundance information. ## Vector with beals smoothing values corresponding to the first species ## in dune. beals(dune, species=1, include=TRUE) #> 1 2 3 4 5 6 7 8 #> 0.5923077 0.5032372 0.3499038 0.3306953 0.5944041 0.5928780 0.5824352 0.2082532 #> 9 10 11 12 13 14 15 16 #> 0.2960799 0.5462492 0.3659392 0.2610043 0.1982372 0.0922619 0.1140625 0.1066506 #> 17 18 19 20 #> 0.6020408 0.3844577 0.2865741 0.0750000"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Multivariate homogeneity of groups dispersions (variances) — betadisper","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Implements Marti Anderson's PERMDISP2 procedure analysis multivariate homogeneity group dispersions (variances). betadisper multivariate analogue Levene's test homogeneity variances. Non-euclidean distances objects group centres (centroids medians) handled reducing original distances principal coordinates. procedure latterly used means assessing beta diversity. anova, scores, plot boxplot methods. TukeyHSD.betadisper creates set confidence intervals differences mean distance--centroid levels grouping factor specified family-wise probability coverage. intervals based Studentized range statistic, Tukey's 'Honest Significant Difference' method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"betadisper(d, group, type = c(\"median\",\"centroid\"), bias.adjust = FALSE, sqrt.dist = FALSE, add = FALSE) # S3 method for betadisper anova(object, ...) # S3 method for betadisper scores(x, display = c(\"sites\", \"centroids\"), choices = c(1,2), ...) # S3 method for betadisper eigenvals(x, ...) # S3 method for betadisper plot(x, axes = c(1,2), cex = 0.7, pch = seq_len(ng), col = NULL, lty = \"solid\", lwd = 1, hull = TRUE, ellipse = FALSE, conf, segments = TRUE, seg.col = \"grey\", seg.lty = lty, seg.lwd = lwd, label = TRUE, label.cex = 1, ylab, xlab, main, sub, ...) # S3 method for betadisper boxplot(x, ylab = \"Distance to centroid\", ...) # S3 method for betadisper TukeyHSD(x, which = \"group\", ordered = FALSE, conf.level = 0.95, ...) # S3 method for betadisper print(x, digits = max(3, getOption(\"digits\") - 3), neigen = 8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"d distance structure returned dist, betadiver vegdist. group vector describing group structure, usually factor object can coerced factor using .factor. Can consist factor single level (.e., one group). type type analysis perform. Use spatial median group centroid? spatial median now default. bias.adjust logical: adjust small sample bias beta diversity estimates? sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. display character; partial match access scores \"sites\" \"species\". object, x object class \"betadisper\", result call betadisper. choices, axes principal coordinate axes wanted. hull logical; convex hull group plotted? ellipse logical; standard deviation data ellipse group plotted? conf Expected fractions data coverage data ellipses, e.g. 0.95. default draw 1 standard deviation data ellipse, supplied, conf multiplied corresponding value found Chi-squared distribution 2df provide requested coverage (probability contour). pch plot symbols groups, vector length equal number groups. col colors plot symbols centroid labels groups, vector length equal number groups. lty, lwd linetype, linewidth convex hulls confidence ellipses. segments logical; segments joining points centroid drawn? seg.col colour draw segments points centroid. Can vector, case one colour per group. seg.lty, seg.lwd linetype line width segments. label logical; centroids labelled respective factor label? label.cex numeric; character expansion centroid labels. cex, ylab, xlab, main, sub graphical parameters. details, see plot.default. character vector listing terms fitted model intervals calculated. Defaults grouping factor. ordered logical; see TukeyHSD. conf.level numeric value zero one giving family-wise confidence level use. digits, neigen numeric; print method, sets number digits use (per print.default) maximum number axes display eigenvalues , repsectively. ... arguments, including graphical parameters ( plot.betadisper boxplot.betadisper), passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"One measure multivariate dispersion (variance) group samples calculate average distance group members group centroid spatial median (referred 'centroid' now unless stated otherwise) multivariate space. test dispersions (variances) one groups different, distances group members group centroid subject ANOVA. multivariate analogue Levene's test homogeneity variances distances group members group centroids Euclidean distance. However, better measures distance Euclidean distance available ecological data. can accommodated reducing distances produced using dissimilarity coefficient principal coordinates, embeds within Euclidean space. analysis proceeds calculating Euclidean distances group members group centroid basis principal coordinate axes rather original distances. Non-metric dissimilarity coefficients can produce principal coordinate axes negative Eigenvalues. correspond imaginary, non-metric part distance objects. negative Eigenvalues produced, must correct imaginary distances. distance centroid point $$z_{ij}^c = \\sqrt{\\Delta^2(u_{ij}^+, c_i^+) - \\Delta^2(u_{ij}^-, c_i^-)},$$ \\(\\Delta^2\\) squared Euclidean distance \\(u_{ij}\\), principal coordinate \\(j\\)th point \\(\\)th group, \\(c_i\\), coordinate centroid \\(\\)th group. super-scripted ‘\\(+\\)’ ‘\\(-\\)’ indicate real imaginary parts respectively. equation (3) Anderson (2006). imaginary part greater magnitude real part, taking square root negative value, resulting NaN, cases changed zero distances (warning). line behaviour Marti Anderson's PERMDISP2 programme. test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance \\(F\\). alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution \\(F\\) Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can also performed using permutest.betadisper. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper. simple wrapper TukeyHSD. user directed read help file TukeyHSD using function. particular, note statement using function unbalanced designs. results analysis can visualised using plot boxplot methods. One additional use functions assessing beta diversity (Anderson et al 2006). Function betadiver provides popular dissimilarity measures purpose. noted passing Anderson (2006) related context O'Neill (2000), estimates dispersion around central location (median centroid) calculated data biased downward. bias matters comparing diversity among treatments small, unequal numbers samples. Setting bias.adjust=TRUE using betadisper imposes \\(\\sqrt{n/(n-1)}\\) correction (Stier et al. 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"anova method returns object class \"anova\" inheriting class \"data.frame\". scores method returns list one components \"sites\" \"centroids\". plot function invisibly returns object class \"ordiplot\", plotting structure can used identify.ordiplot (identify points) functions ordiplot family. boxplot function invisibly returns list whose components documented boxplot. eigenvals.betadisper returns named vector eigenvalues. TukeyHSD.betadisper returns list. See TukeyHSD details. betadisper returns list class \"betadisper\" following components: eig numeric; eigenvalues principal coordinates analysis. vectors matrix; eigenvectors principal coordinates analysis. distances numeric; Euclidean distances principal coordinate space samples respective group centroid median. group factor; vector describing group structure centroids matrix; locations group centroids medians principal coordinates. group.distances numeric; mean distance group centroid median. call matched function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"group consists single level group, anova permutest methods appropriate used data stop error. Missing values either d group removed prior performing analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Stewart Schultz noticed permutation test type=\"centroid\" wrong type error anti-conservative. , default type changed \"median\", uses spatial median group centroid. Tests suggests permutation test type analysis gives correct error rates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62, 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. O'Neill, M.E. (2000) Weighted Least Squares Approach Levene's Test Homogeneity Variance. Australian & New Zealand Journal Statistics 42, 81-–100. Stier, .C., Geange, S.W., Hanson, K.M., & Bolker, B.M. (2013) Predator density timing arrival affect reef fish community assembly. Ecology 94, 1057--1068.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Gavin L. Simpson; bias correction Adrian Stier Ben Bolker.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F permutest(mod, pairwise = TRUE, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 99 0.07 . #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Pairwise comparisons: #> (Observed p-value below diagonal, permuted p-value above diagonal) #> grazed ungrazed #> grazed 0.08 #> ungrazed 0.04295 ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Plot the groups and distances to centroids on the ## first two PCoA axes plot(mod) ## with data ellipses instead of hulls plot(mod, ellipse = TRUE, hull = FALSE) # 1 sd data ellipse plot(mod, ellipse = TRUE, hull = FALSE, conf = 0.90) # 90% data ellipse # plot with manual colour specification my_cols <- c(\"#1b9e77\", \"#7570b3\") plot(mod, col = my_cols, pch = c(16,17), cex = 1.1) ## can also specify which axes to plot, ordering respected plot(mod, axes = c(3,1), seg.col = \"forestgreen\", seg.lty = \"dashed\") ## Draw a boxplot of the distances to centroid for each group boxplot(mod) ## `scores` and `eigenvals` also work scrs <- scores(mod) str(scrs) #> List of 2 #> $ sites : num [1:24, 1:2] 0.0946 -0.3125 -0.3511 -0.3291 -0.1926 ... #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:24] \"18\" \"15\" \"24\" \"27\" ... #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" #> $ centroids: num [1:2, 1:2] -0.1455 0.2786 0.0758 -0.2111 #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:2] \"grazed\" \"ungrazed\" #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" head(scores(mod, 1:4, display = \"sites\")) #> PCoA1 PCoA2 PCoA3 PCoA4 #> 18 0.09459373 0.15914576 0.074400844 -0.202466025 #> 15 -0.31248809 0.10032751 -0.062243360 0.110844864 #> 24 -0.35106507 -0.05954096 -0.038079447 0.095060928 #> 27 -0.32914546 -0.17019348 0.231623720 0.019110623 #> 23 -0.19259443 -0.01459250 -0.005679372 -0.209718312 #> 19 -0.06794575 -0.14501690 -0.085645653 0.002431355 # group centroids/medians scores(mod, 1:4, display = \"centroids\") #> PCoA1 PCoA2 PCoA3 PCoA4 #> grazed -0.1455200 0.07584572 -0.01366220 -0.0178990 #> ungrazed 0.2786095 -0.21114993 -0.03475586 0.0220129 # eigenvalues from the underlying principal coordinates analysis eigenvals(mod) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 #> 1.7552165 1.1334455 0.4429018 0.3698054 0.2453532 0.1960921 0.1751131 #> PCoA8 PCoA9 PCoA10 PCoA11 PCoA12 PCoA13 PCoA14 #> 0.1284467 0.0971594 0.0759601 0.0637178 0.0583225 0.0394934 0.0172699 #> PCoA15 PCoA16 PCoA17 PCoA18 PCoA19 PCoA20 PCoA21 #> 0.0051011 -0.0004131 -0.0064654 -0.0133147 -0.0253944 -0.0375105 -0.0480069 #> PCoA22 PCoA23 #> -0.0537146 -0.0741390 ## try out bias correction; compare with mod3 (mod3B <- betadisper(dis, groups, type = \"median\", bias.adjust=TRUE)) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"median\", bias.adjust #> = TRUE) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.4055 0.2893 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 anova(mod3B) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 0.06468 . #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 permutest(mod3B, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 99 0.05 * #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## should always work for a single group group <- factor(rep(\"grazed\", NROW(varespec))) (tmp <- betadisper(dis, group, type = \"median\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"median\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed #> 0.4255 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 (tmp <- betadisper(dis, group, type = \"centroid\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"centroid\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to centroid: #> grazed #> 0.4261 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## simulate missing values in 'd' and 'group' ## using spatial medians groups[c(2,20)] <- NA dis[c(2, 20)] <- NA mod2 <- betadisper(dis, groups) ## messages #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod2 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to median: #> grazed ungrazed #> 0.3984 0.3008 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod2, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 99 0.16 #> Residuals 18 0.296910 0.016495 anova(mod2) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 0.1369 #> Residuals 18 0.296910 0.016495 plot(mod2) boxplot(mod2) plot(TukeyHSD(mod2)) ## Using group centroids mod3 <- betadisper(dis, groups, type = \"centroid\") #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod3 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"centroid\") #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to centroid: #> grazed ungrazed #> 0.4001 0.3108 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod3, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 99 0.12 #> Residuals 18 0.189749 0.010542 anova(mod3) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 0.09166 . #> Residuals 18 0.189749 0.010542 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod3) boxplot(mod3) plot(TukeyHSD(mod3))"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of beta Diversity — betadiver","title":"Indices of beta Diversity — betadiver","text":"function estimates 24 indices beta diversity reviewed Koleff et al. (2003). Alternatively, finds co-occurrence frequencies triangular plots (Koleff et al. 2003).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of beta Diversity — betadiver","text":"","code":"betadiver(x, method = NA, order = FALSE, help = FALSE, ...) # S3 method for betadiver plot(x, ...) # S3 method for betadiver scores(x, triangular = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of beta Diversity — betadiver","text":"x Community data matrix, betadiver result plot scores functions. method index beta diversity defined Koleff et al. (2003), Table 1. can use either subscript \\(\\beta\\) number index. See argument help . order Order sites increasing number species. influence configuration triangular plot non-symmetric indices. help Show numbers, subscript names defining equations indices exit. triangular Return scores suitable triangular plotting proportions. FALSE, returns 3-column matrix raw counts. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of beta Diversity — betadiver","text":"commonly used index beta diversity \\(\\beta_w = S/\\alpha - 1\\), \\(S\\) total number species, \\(\\alpha\\) average number species per site (Whittaker 1960). drawback model \\(S\\) increases sample size, expectation \\(\\alpha\\) remains constant, beta diversity increases sample size. solution problem study beta diversity pairs sites (Marion et al. 2017). denote number species shared two sites \\(\\) numbers unique species (shared) \\(b\\) \\(c\\), \\(S = + b + c\\) \\(\\alpha = (2 + b + c)/2\\) \\(\\beta_w = (b+c)/(2 + b + c)\\). Sørensen dissimilarity defined vegan function vegdist argument binary = TRUE. Many indices dissimilarity indices well. Function betadiver finds indices reviewed Koleff et al. (2003). indices found function designdist, current function provides conventional shortcut. function finds indices. proper analysis must done functions betadisper, adonis2 mantel. indices directly taken Table 1 Koleff et al. (2003), can selected either index number subscript name used Koleff et al. numbers, names defining equations can seen using betadiver(help = TRUE). cases two alternative forms, one term \\(-1\\) used. several duplicate indices, number distinct alternatives much lower 24 formally provided. formulations used functions differ occasionally Koleff et al. (2003), still mathematically equivalent. method = NA, index calculated, instead object class betadiver returned. list elements , b c. Function plot can used display proportions elements triangular plot suggested Koleff et al. (2003), scores extracts triangular coordinates raw scores. Function plot returns invisibly triangular coordinates \"ordiplot\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of beta Diversity — betadiver","text":"method = NA, function returns object class \"betadisper\" elements , b, c. method specified, function returns \"dist\" object can used function analysing dissimilarities. beta diversity, particularly useful functions betadisper study betadiversity groups, adonis2 model, mantel compare beta diversities dissimilarities distances (including geographical distances). Although betadiver returns \"dist\" object, indices similarities used place dissimilarities, user error. Functions 10 (\"j\"), 11 (\"sor\") 21 (\"rlb\") similarity indices. Function sets argument \"maxdist\" similarly vegdist, using NA fixed upper limit, 0 similarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of beta Diversity — betadiver","text":"Baselga, . (2010) Partitioning turnover nestedness components beta diversity. Global Ecology Biogeography 19, 134--143. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence-absence data. Journal Animal Ecology 72, 367--382. Marion, Z.H., Fordyce, J.. Fitzpatrick, B.M. (2017) Pairwise beta diversity resolves underappreciated source confusion calculating species turnover. Ecology 98, 933--939. Whittaker, R.H. (1960) Vegetation Siskiyou mountains, Oregon California. Ecological Monographs 30, 279--338.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of beta Diversity — betadiver","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Indices of beta Diversity — betadiver","text":"indices return similarities instead dissimilarities.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of beta Diversity — betadiver","text":"","code":"## Raw data and plotting data(sipoo) m <- betadiver(sipoo) plot(m) ## The indices betadiver(help=TRUE) #> 1 \"w\" = (b+c)/(2*a+b+c) #> 2 \"-1\" = (b+c)/(2*a+b+c) #> 3 \"c\" = (b+c)/2 #> 4 \"wb\" = b+c #> 5 \"r\" = 2*b*c/((a+b+c)^2-2*b*c) #> 6 \"I\" = log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c) #> 7 \"e\" = exp(log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c))-1 #> 8 \"t\" = (b+c)/(2*a+b+c) #> 9 \"me\" = (b+c)/(2*a+b+c) #> 10 \"j\" = a/(a+b+c) #> 11 \"sor\" = 2*a/(2*a+b+c) #> 12 \"m\" = (2*a+b+c)*(b+c)/(a+b+c) #> 13 \"-2\" = pmin.int(b,c)/(pmax.int(b,c)+a) #> 14 \"co\" = (a*c+a*b+2*b*c)/(2*(a+b)*(a+c)) #> 15 \"cc\" = (b+c)/(a+b+c) #> 16 \"g\" = (b+c)/(a+b+c) #> 17 \"-3\" = pmin.int(b,c)/(a+b+c) #> 18 \"l\" = (b+c)/2 #> 19 \"19\" = 2*(b*c+1)/(a+b+c)/(a+b+c-1) #> 20 \"hk\" = (b+c)/(2*a+b+c) #> 21 \"rlb\" = a/(a+c) #> 22 \"sim\" = pmin.int(b,c)/(pmin.int(b,c)+a) #> 23 \"gl\" = 2*abs(b-c)/(2*a+b+c) #> 24 \"z\" = (log(2)-log(2*a+b+c)+log(a+b+c))/log(2) ## The basic Whittaker index d <- betadiver(sipoo, \"w\") ## This should be equal to Sorensen index (binary Bray-Curtis in ## vegan) range(d - vegdist(sipoo, binary=TRUE)) #> [1] 0 0"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":null,"dir":"Reference","previous_headings":"","what":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function computes coefficients dispersal direction geographically connected areas, defined Legendre Legendre (1984), also described Legendre Legendre (2012, section 13.3.4).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"bgdispersal(mat, PAonly = FALSE, abc = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"mat Data frame matrix containing community composition data table (species presence-absence abundance data). PAonly FALSE four types coefficients, DD1 DD4, requested; TRUE DD1 DD2 sought (see Details). abc TRUE, return tables , b c used DD1 DD2.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"signs DD coefficients indicate direction dispersal, provided asymmetry significant. positive sign indicates dispersal first (row DD tables) second region (column); negative sign indicates opposite. McNemar test asymmetry computed presence-absence data test hypothesis significant asymmetry two areas comparison. input data table, rows sites areas, columns taxa. often, taxa species, coefficients can computed genera families well. DD1 DD2 computed presence-absence data. four types coefficients computed quantitative data, converted presence-absence computation DD1 DD2. PAonly = FALSE indicates four types coefficients requested. PAonly = TRUE DD1 DD2 sought.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Function bgdispersal returns list containing following matrices: DD1 \\(DD1_{j,k} = ((b - c))/((+ b + c)^2)\\) DD2 \\(DD2_{j,k} = (2 (b - c))/((2a + b + c) (+ b + c))\\) \\(\\), \\(b\\), \\(c\\) meaning computation binary similarity coefficients. DD3 \\(DD3_{j,k} = {W(-B) / (+B-W)^2} \\) DD4 \\(DD4_{j,k} = 2W(-B) / ((+B)(+B-W))\\) W = sum(pmin(vector1, vector2)), = sum(vector1), B = sum(vector2) McNemar McNemar chi-square statistic asymmetry (Sokal Rohlf 1995): \\(2(b \\log(b) + c \\log(c) - (b+c) \\log((b+c)/2)) / q\\), \\(q = 1 + 1/(2(b+c))\\) (Williams correction continuity) prob.McNemar probabilities associated McNemar statistics, chi-square test. H0: asymmetry \\((b-c)\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Legendre, P. V. Legendre. 1984. Postglacial dispersal freshwater fishes Québec peninsula. Can. J. Fish. Aquat. Sci. 41: 1781-1802. Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Sokal, R. R. F. J. Rohlf. 1995. Biometry. principles practice statistics biological research. 3rd edn. W. H. Freeman, New York.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function uses powerful alternative McNemar test classical formula. classical formula constructed spirit Pearson's Chi-square, formula function constructed spirit Wilks Chi-square \\(G\\) statistic. Function mcnemar.test uses classical formula. new formula introduced vegan version 1.10-11, older implementations bgdispersal used classical formula.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"mat <- matrix(c(32,15,14,10,70,30,100,4,10,30,25,0,18,0,40, 0,0,20,0,0,0,0,4,0,30,20,0,0,0,0,25,74,42,1,45,89,5,16,16,20), 4, 10, byrow=TRUE) bgdispersal(mat) #> $DD1 #> [,1] [,2] [,3] [,4] #> [1,] 0.00 0.24 0.21 0.00 #> [2,] -0.24 0.00 0.08 -0.24 #> [3,] -0.21 -0.08 0.00 -0.21 #> [4,] 0.00 0.24 0.21 0.00 #> #> $DD2 #> [,1] [,2] [,3] [,4] #> [1,] 0.0000000 0.3428571 0.3230769 0.0000000 #> [2,] -0.3428571 0.0000000 0.1142857 -0.3428571 #> [3,] -0.3230769 -0.1142857 0.0000000 -0.3230769 #> [4,] 0.0000000 0.3428571 0.3230769 0.0000000 #> #> $DD3 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.1567922 0.1420408 -0.01325831 #> [2,] -0.15679216 0.0000000 0.1101196 -0.20049485 #> [3,] -0.14204082 -0.1101196 0.0000000 -0.13586560 #> [4,] 0.01325831 0.2004949 0.1358656 0.00000000 #> #> $DD4 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.2513176 0.2425087 -0.01960102 #> [2,] -0.25131757 0.0000000 0.1725441 -0.30993929 #> [3,] -0.24250871 -0.1725441 0.0000000 -0.23381521 #> [4,] 0.01960102 0.3099393 0.2338152 0.00000000 #> #> $McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 7.677938 9.0571232 0.000000 #> [2,] NA NA 0.2912555 7.677938 #> [3,] NA NA NA 9.057123 #> [4,] NA NA NA NA #> #> $prob.McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 0.005590001 0.002616734 1.000000000 #> [2,] NA NA 0.589417103 0.005590001 #> [3,] NA NA NA 0.002616734 #> [4,] NA NA NA NA #>"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":null,"dir":"Reference","previous_headings":"","what":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Function finds best subset environmental variables, Euclidean distances scaled environmental variables maximum (rank) correlation community dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# S3 method for default bioenv(comm, env, method = \"spearman\", index = \"bray\", upto = ncol(env), trace = FALSE, partial = NULL, metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), parallel = getOption(\"mc.cores\"), ...) # S3 method for formula bioenv(formula, data, ...) bioenvdist(x, which = \"best\")"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"comm Community data frame dissimilarity object square matrix can interpreted dissimilarities. env Data frame continuous environmental variables. method correlation method used cor. index dissimilarity index used community data (comm) vegdist. ignored comm dissimilarities. upto Maximum number parameters studied subsets. formula, data Model formula data. trace Trace calculations partial Dissimilarities partialled inspecting variables env. metric Metric used distances environmental distances. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. x bioenv result object. number model environmental distances evaluated, \"best\" model. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function calculates community dissimilarity matrix using vegdist. selects possible subsets environmental variables, scales variables, calculates Euclidean distances subset using dist. function finds correlation community dissimilarities environmental distances, size subsets, saves best result. \\(2^p-1\\) subsets \\(p\\) variables, exhaustive search may take , , long time (parameter upto offers partial relief). argument metric defines distances given set environmental variables. metric = \"euclidean\", variables scaled unit variance Euclidean distances calculated. metric = \"mahalanobis\", Mahalanobis distances calculated: addition scaling unit variance, matrix current set environmental variables also made orthogonal (uncorrelated). metric = \"manhattan\", variables scaled unit range Manhattan distances calculated, distances sums differences environmental variables. metric = \"gower\", Gower distances calculated using function daisy. allows also using factor variables, continuous variables results equal metric = \"manhattan\". function can called model formula LHS data matrix RHS lists environmental variables. formula interface practical selecting transforming environmental variables. argument partial can perform “partial” analysis. partializing item must dissimilarity object class dist. partial item can used correlation method, strictly correct Pearson. Function bioenvdist recalculates environmental distances used within function. default calculate distances best model, number model can given. Clarke & Ainsworth (1993) suggested method used selecting best subset environmental variables interpreting results nonmetric multidimensional scaling (NMDS). recommended parallel display NMDS community dissimilarities NMDS Euclidean distances best subset scaled environmental variables. warned use Procrustes analysis, looks like good way comparing two ordinations. Clarke & Ainsworth wrote computer program BIO-ENV giving name current function. Presumably BIO-ENV later incorporated Clarke's PRIMER software (available Windows). addition, Clarke & Ainsworth suggested novel method rank correlation available current function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function returns object class bioenv summary method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Clarke, K. R & Ainsworth, M. 1993. method linking multivariate community structure environmental variables. Marine Ecology Progress Series, 92, 205--219.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"want study ‘significance’ bioenv results, can use function mantel mantel.partial use definition correlation. However, bioenv standardizes environmental variables depending used metric, must mantel comparable results (standardized data returned item x result object). safest use bioenvdist extract environmental distances really used within bioenv. NB., bioenv selects variables maximize Mantel correlation, significance tests based priori selection variables biased.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# The method is very slow for large number of possible subsets. # Therefore only 6 variables in this example. data(varespec) data(varechem) sol <- bioenv(wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, varechem) sol #> #> Call: #> bioenv(formula = wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, data = varechem) #> #> Subset of environmental variables with best correlation to community data. #> #> Correlations: spearman #> Dissimilarities: bray #> Metric: euclidean #> #> Best model has 3 parameters (max. 6 allowed): #> P Ca Al #> with correlation 0.4004806 #> ## IGNORE_RDIFF_BEGIN summary(sol) #> size correlation #> P 1 0.2513 #> P Al 2 0.4004 #> P Ca Al 3 0.4005 #> P Ca pH Al 4 0.3619 #> log(N) P Ca pH Al 5 0.3216 #> log(N) P K Ca pH Al 6 0.2822 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA biplot — biplot.rda","title":"PCA biplot — biplot.rda","text":"Draws PCA biplot species scores indicated biplot arrows","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA biplot — biplot.rda","text":"","code":"# S3 method for rda biplot(x, choices = c(1, 2), scaling = \"species\", display = c(\"sites\", \"species\"), type, xlim, ylim, col = c(1,2), const, correlation = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA biplot — biplot.rda","text":"x rda result object. choices Axes show. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. negative scaling values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. display Scores shown. must alternatives \"species\" species scores, /\"sites\" site scores. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. Can length 2 (e.g. type = c(\"text\", \"points\")), case first element describes species scores handled, second site scores drawn. xlim, ylim x y limits (min, max) plot. col Colours used sites species (order). one colour given, used . const General scaling constant scores.rda. ... parameters plotting functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA biplot — biplot.rda","text":"Produces plot biplot results call rda. common \"species\" scores PCA drawn biplot arrows point direction increasing values variable. biplot.rda function provides wrapper plot.cca allow easy production plot. biplot.rda suitable unconstrained models. used ordination object constraints, error issued. species scores drawn using \"text\", arrows drawn origin 0.85 * species score, whilst labels drawn species score. type used \"points\", labels drawn therefore arrows drawn origin actual species score.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"PCA biplot — biplot.rda","text":"plot function returns invisibly plotting structure can used identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"PCA biplot — biplot.rda","text":"Gavin Simpson, based plot.cca Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA biplot — biplot.rda","text":"","code":"data(dune) mod <- rda(dune, scale = TRUE) biplot(mod, scaling = \"symmetric\") ## different type for species and site scores biplot(mod, scaling = \"symmetric\", type = c(\"text\", \"points\")) ## We can use ordiplot pipes in R 4.1 to build similar plots with ## flexible control if (FALSE) { if (getRversion() >= \"4.1\") { plot(mod, scaling = \"symmetric\", type=\"n\") |> text(\"sites\", cex=0.8) |> text(\"species\", arrows=TRUE, length=0.02, col=\"red\", cex=0.6) } }"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":null,"dir":"Reference","previous_headings":"","what":"K-means partitioning using a range of values of K — cascadeKM","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function wrapper kmeans function. creates several partitions forming cascade small large number groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"cascadeKM(data, inf.gr, sup.gr, iter = 100, criterion = \"calinski\", parallel = getOption(\"mc.cores\")) cIndexKM(y, x, index = \"all\") # S3 method for cascadeKM plot(x, min.g, max.g, grpmts.plot = TRUE, sortg = FALSE, gridcol = NA, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"K-means partitioning using a range of values of K — cascadeKM","text":"data data matrix. objects (samples) rows. inf.gr number groups partition smallest number groups cascade (min). sup.gr number groups partition largest \t number groups cascade (max). iter number random starting configurations value \\(K\\). criterion criterion used select best partition. default value \"calinski\", refers Calinski-Harabasz (1974) criterion. simple structure index (\"ssi\") also available. indices available package cclust. experience, two indices work best likely return maximum value near optimal number clusters \"calinski\" \"ssi\". y Object class \"kmeans\" returned clustering algorithm kmeans x Data matrix columns correspond variables rows observations, plotting object plot index available indices : \"calinski\" \"ssi\". Type \"\" obtain indices. Abbreviations names also accepted. min.g, max.g minimum maximum numbers groups displayed. grpmts.plot Show plot (TRUE FALSE). sortg Sort objects function group membership produce easily interpretable graph. See Details. original object names kept; used labels output table x, although graph. row names, sequential row numbers used keep track original order objects. gridcol colour grid lines plots. NA, default value, removes grid lines. ... parameters functions (ignored). parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function creates several partitions forming cascade small large number groups formed kmeans. work performed function cIndex based clustIndex package cclust). criteria removed version computation errors generated one object found group. default value \"calinski\", refers well-known Calinski-Harabasz (1974) criterion. available index simple structure index \"ssi\" (Dolnicar et al. 1999). case groups equal sizes, \"calinski\" generally good criterion indicate correct number groups. Users take indications literally groups equal size. Type \"\" obtain indices. indices defined : calinski: \\((SSB/(K-1))/(SSW/(n-K))\\), \\(n\\) number data points \\(K\\) number clusters. \\(SSW\\) sum squares within clusters \\(SSB\\) sum squares among clusters. index simply \\(F\\) (ANOVA) statistic. ssi: “Simple Structure Index” multiplicatively combines several elements influence interpretability partitioning solution. best partition indicated highest SSI value. simulation study, Milligan Cooper (1985) found Calinski-Harabasz criterion recovered correct number groups often. recommend criterion , groups equal sizes, maximum value \"calinski\" usually indicates correct number groups. Another available index simple structure index \"ssi\". Users take indications indices literally groups equal size explore groups corresponding values \\(K\\). Function cascadeKM plot method. Two plots produced. graph left objects abscissa number groups ordinate. groups represented colours. graph right shows values criterion (\"calinski\" \"ssi\") determining best partition. highest value criterion marked red. Points marked orange, , indicate partitions producing increase criterion value number groups increases; may represent interesting partitions. sortg=TRUE, objects reordered following procedure: (1) simple matching distance matrix computed among objects, based table K-means assignments groups, \\(K\\) = min.g \\(K\\) = max.g. (2) principal coordinate analysis (PCoA, Gower 1966) computed centred distance matrix. (3) first principal coordinate used new order objects graph. simplified algorithm used compute first principal coordinate , using iterative algorithm described Legendre & Legendre (2012). full distance matrix among objects never computed; avoids problem storing number objects large. Distance values computed needed algorithm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Function cascadeKM returns object class cascadeKM items: partition Table partitions found different numbers groups \\(K\\), \\(K\\) = inf.gr \\(K\\) = sup.gr. results Values criterion select best partition. criterion name criterion used. size number objects found group, partitions (columns). Function cIndex returns vector index values. maximum value indices supposed indicate best partition. indices work best groups equal sizes. groups equal sizes, one put much faith maximum indices, also explore groups corresponding values \\(K\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Calinski, T. J. Harabasz. 1974. dendrite method cluster analysis. Commun. Stat. 3: 1--27. Dolnicar, S., K. Grabler J. . Mazanec. 1999. tale three cities: perceptual charting analyzing destination images. Pp. 39-62 : Woodside, . et al. [eds.] Consumer psychology tourism, hospitality leisure. CAB International, New York. Gower, J. C. 1966. distance properties latent root vector methods used multivariate analysis. Biometrika 53: 325--338. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Milligan, G. W. & M. C. Cooper. 1985. examination procedures determining number clusters data set. Psychometrika 50: 159--179. Weingessel, ., Dimitriadou, . Dolnicar, S. 2002. examination indexes determining number clusters binary data sets. Psychometrika 67: 137--160.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Marie-Helene Ouellette Marie-Helene.Ouellette@UMontreal.ca, Sebastien Durand Sebastien.Durand@UMontreal.ca Pierre Legendre Pierre.Legendre@UMontreal.ca. Parallel processing Virgilio Gómez-Rubio. Edited vegan Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"# Partitioning a (10 x 10) data matrix of random numbers mat <- matrix(runif(100),10,10) res <- cascadeKM(mat, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning an autocorrelated time series vec <- sort(matrix(runif(30),30,1)) res <- cascadeKM(vec, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning a large autocorrelated time series # Note that we remove the grid lines vec <- sort(matrix(runif(1000),1000,1)) res <- cascadeKM(vec, 2, 7, iter = 10, criterion = 'calinski') toto <- plot(res, gridcol=NA)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca performs correspondence analysis, optionally constrained correspondence analysis (.k.. canonical correspondence analysis), optionally partial constrained correspondence analysis. Function rda performs redundancy analysis, optionally principal components analysis. popular ordination techniques community ecology.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"# S3 method for formula cca(formula, data, na.action = na.fail, subset = NULL, ...) # S3 method for formula rda(formula, data, scale=FALSE, na.action = na.fail, subset = NULL, ...) # S3 method for default cca(X, Y, Z, ...) # S3 method for default rda(X, Y, Z, scale=FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"formula Model formula, left hand side gives community data matrix, right hand side gives constraining variables, conditioning variables can given within special function Condition. data Data frame containing variables right hand side model formula. X Community data matrix. Y Constraining matrix, typically environmental variables. Can missing. data.frame, expanded model.matrix factors expanded contrasts (“dummy variables”). better use formula instead argument, analyses work formula used. Z Conditioning matrix, effect removed (“partialled ”) next step. Can missing. data.frame, expanded similarly constraining matrix. scale Scale species unit variance (like correlations). na.action Handling missing values constraints conditions. default (na.fail) stop missing value. Choice na.omit removes rows missing values. Choice na.exclude keeps observations gives NA results calculated. WA scores rows may found also missing values constraints. Missing values never allowed dependent community data. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data. ... arguments print plot functions (ignored functions).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Since introduction (ter Braak 1986), constrained, canonical, correspondence analysis spin-, redundancy analysis, popular ordination methods community ecology. Functions cca rda similar popular proprietary software Canoco, although implementation completely different. functions based Legendre & Legendre's (2012) algorithm: cca Chi-square transformed data matrix subjected weighted linear regression constraining variables, fitted values submitted correspondence analysis performed via singular value decomposition (svd). Function rda similar, uses ordinary, unweighted linear regression unweighted SVD. Legendre & Legendre (2012), Table 11.5 (p. 650) give skeleton RDA algorithm vegan. algorithm CCA similar, involves standardization row column weights. functions can called either matrix-like entries community data constraints, formula interface. general, formula interface preferred, allows better control model allows factor constraints. analyses ordination results possible model fitted formula (e.g., cases anova.cca, automatic model building). following sections, X, Y Z, although referred matrices, commonly data frames. matrix interface, community data matrix X must given, data matrices may omitted, corresponding stage analysis skipped. matrix Z supplied, effects removed community matrix, residual matrix submitted next stage. called partial correspondence redundancy analysis. matrix Y supplied, used constrain ordination, resulting constrained canonical correspondence analysis, redundancy analysis. Finally, residual submitted ordinary correspondence analysis (principal components analysis). matrices Z Y missing, data matrix analysed ordinary correspondence analysis ( principal components analysis). Instead separate matrices, model can defined using model formula. left hand side must community data matrix (X). right hand side defines constraining model. constraints can contain ordered unordered factors, interactions among variables functions variables. defined contrasts honoured factor variables. constraints can also matrices (data frames). formula can include special term Condition conditioning variables (“covariables”) partialled analysis. following commands equivalent: cca(X, Y, Z), cca(X ~ Y + Condition(Z)), Y Z refer constraints conditions matrices respectively. Constrained correspondence analysis indeed constrained method: CCA try display variation data, part can explained used constraints. Consequently, results strongly dependent set constraints transformations interactions among constraints. shotgun method use environmental variables constraints. However, exploratory problems better analysed unconstrained methods correspondence analysis (decorana, corresp) non-metric multidimensional scaling (metaMDS) environmental interpretation analysis (envfit, ordisurf). CCA good choice user clear strong priori hypotheses constraints interested major structure data set. CCA able correct curve artefact commonly found correspondence analysis forcing configuration linear constraints. However, curve artefact can avoided low number constraints curvilinear relation . curve can reappear even two badly chosen constraints single factor. Although formula interface makes easy include polynomial interaction terms, terms often produce curved artefacts (difficult interpret), probably avoided. According folklore, rda used ``short gradients'' rather cca. However, based research finds methods based Euclidean metric uniformly weaker based Chi-squared metric. However, standardized Euclidean distance may appropriate measures (see Hellinger standardization decostand particular). Partial CCA (pCCA; alternatively partial RDA) can used remove effect conditioning background random variables covariables CCA proper. fact, pCCA compares models cca(X ~ Z) cca(X ~ Y + Z) attributes difference effect Y cleansed effect Z. people used method extracting “components variance” CCA. However, effect variables together stronger sum separately, can increase total Chi-square partialling variation, give negative “components variance”. general, components “variance” trusted due interactions two sets variables. functions summary plot methods documented separately (see plot.cca, summary.cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca returns huge object class cca, described separately cca.object. Function rda returns object class rda inherits class cca described cca.object. scaling used rda scores described separate vignette package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"original method ter Braak, current implementation follows Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. McCune, B. (1997) Influence noisy environmental data canonical correspondence analysis. Ecology 78, 2617-2623. Palmer, M. W. (1993) Putting things even better order: advantages canonical correspondence analysis. Ecology 74,2215-2230. Ter Braak, C. J. F. (1986) Canonical Correspondence Analysis: new eigenvector technique multivariate direct gradient analysis. Ecology 67, 1167-1179.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"responsible author Jari Oksanen, code borrows heavily Dave Roberts (Montana State University, USA).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"data(varespec) data(varechem) ## Common but bad way: use all variables you happen to have in your ## environmental data matrix vare.cca <- cca(varespec, varechem) vare.cca #> Call: cca(X = varespec, Y = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Constrained 1.4415 0.6920 14 #> Unconstrained 0.6417 0.3080 9 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4389 0.2918 0.1628 0.1421 0.1180 0.0890 0.0703 0.0584 0.0311 0.0133 0.0084 #> CCA12 CCA13 CCA14 #> 0.0065 0.0062 0.0047 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 #> 0.19776 0.14193 0.10117 0.07079 0.05330 0.03330 0.01887 0.01510 0.00949 #> plot(vare.cca) ## Formula interface and a better model vare.cca <- cca(varespec ~ Al + P*(K + Baresoil), data=varechem) vare.cca #> Call: cca(formula = varespec ~ Al + P * (K + Baresoil), data = #> varechem) #> #> Inertia Proportion Rank #> Total 2.083 1.000 #> Constrained 1.046 0.502 6 #> Unconstrained 1.038 0.498 17 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 0.3756 0.2342 0.1407 0.1323 0.1068 0.0561 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.27577 0.15411 0.13536 0.11803 0.08887 0.05511 0.04919 0.03781 #> (Showing 8 of 17 unconstrained eigenvalues) #> plot(vare.cca) ## Partialling out and negative components of variance cca(varespec ~ Ca, varechem) #> Call: cca(formula = varespec ~ Ca, data = varechem) #> #> Inertia Proportion Rank #> Total 2.08320 1.00000 #> Constrained 0.15722 0.07547 1 #> Unconstrained 1.92598 0.92453 22 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.15722 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.4745 0.2939 0.2140 0.1954 0.1748 0.1171 0.1121 0.0880 #> (Showing 8 of 22 unconstrained eigenvalues) #> cca(varespec ~ Ca + Condition(pH), varechem) #> Call: cca(formula = varespec ~ Ca + Condition(pH), data = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Conditional 0.1458 0.0700 1 #> Constrained 0.1827 0.0877 1 #> Unconstrained 1.7547 0.8423 21 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.18269 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3834 0.2749 0.2123 0.1760 0.1701 0.1161 0.1089 0.0880 #> (Showing 8 of 21 unconstrained eigenvalues) #> ## RDA data(dune) data(dune.env) dune.Manure <- rda(dune ~ Manure, dune.env) plot(dune.Manure)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":null,"dir":"Reference","previous_headings":"","what":"Result Object from Constrained Ordination — cca.object","title":"Result Object from Constrained Ordination — cca.object","text":"Ordination methods cca, rda, dbrda capscale return similar result objects. methods use internal function ordConstrained. differ (1) initial transformation data defining inertia, (2) weighting, (3) use rectangular rows \\(\\times\\) columns data symmetric rows \\(\\times\\) rows dissimilarities: rda initializes data give variance correlations inertia, cca based double-standardized data give Chi-square inertia uses row column weights, capscale maps real part dissimilarities rectangular data performs RDA, dbrda performs RDA-like analysis directly symmetric dissimilarities. Function ordConstrained returns result components methods, calling function may add components final result. However, access result components directly (using $): internal structure regarded stable application interface (API), can change release. access results components directly, take risk breakage vegan release. vegan provides wide set accessor functions components, functions updated result object changes. documentation gives overview accessor functions cca result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Result Object from Constrained Ordination — cca.object","text":"","code":"ordiYbar(x, model = c(\"CCA\", \"CA\", \"pCCA\", \"partial\", \"initial\")) # S3 method for cca model.frame(formula, ...) # S3 method for cca model.matrix(object, ...) # S3 method for cca weights(object, display = \"sites\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Result Object from Constrained Ordination — cca.object","text":"object, x, formula result object cca, rda, dbrda, capscale. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. ordiYbar value can also \"initial\" internal working input data, \"partial\" internal working input data removing partial effects. display Display either \"sites\" \"species\". ... arguments passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Result Object from Constrained Ordination — cca.object","text":"internal (“working”) form dependent (community) data can accessed function ordiYbar. form depends ordination method: instance, cca data weighted Chi-square transformed, dbrda Gower-centred dissimilarities. input data original (“response”) form can accessed fitted.cca residuals.cca. Function predict.cca can return either working response data, also lower-rank approximations. model matrix independent data (“Constraints” “Conditions”) can extracted model.matrix. partial analysis, function returns list design matrices called Conditions Constraints. either component missing, single matrix returned. redundant (aliased) terms appear model matrix. terms can found alias.cca. Function model.frame tries reconstruct data frame model matrices derived. possible original model fitted formula data arguments, still fails data unavailable. number observations can accessed nobs.cca, residual degrees freedom df.residual.cca. information observations missing values can accessed na.action. terms formula fitted model can accessed formula terms. weights used cca can accessed weights. unweighted methods (rda) weights equal. ordination results saved separate components partial terms, constraints residual unconstrained ordination. guarantee components internal names currently, cautious developing scripts functions directly access components. constrained ordination algorithm based QR decomposition constraints conditions (environmental data), QR component saved separately partial constrained components. QR decomposition constraints can accessed qr.cca. also include residual effects partial terms (Conditions), used together ordiYbar(x, \"partial\"). environmental data first centred rda weighted centred cca. QR decomposition used many functions access cca results, can used find many items directly stored object. examples, see coef.cca, coef.rda, vif.cca, permutest.cca, predict.cca, predict.rda, calibrate.cca. See qr possible uses component. instance, rank constraints can found QR decomposition. eigenvalues solution can accessed eigenvals.cca. Eigenvalues evaluated partial component, available constrained residual components. ordination scores internally stored (weighted) orthonormal scores matrices. results can accessed scores.cca scores.rda functions. ordination scores scaled accessed scores functions, internal (weighted) orthonormal scores can accessed setting scaling = FALSE. Unconstrained residual component species site scores, constrained component also fitted site scores linear combination scores sites biplot scores centroids constraint variables. biplot scores correspond model.matrix, centroids calculated factor variables used. scores can selected defining axes, direct way accessing scores certain component. number dimensions can assessed eigenvals. addition, types can derived results although saved results. instance, regression scores model coefficients can accessed scores coef functions. Partial component scores. Distance-based methods (dbrda, capscale) can negative eigenvalues associated imaginary axis scores. addition, species scores initially missing dbrda accessory found analysis capscale (may misleading). Function sppscores can used add species scores replace meaningful ones.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Result Object from Constrained Ordination — cca.object","text":"latest large change result object made release 2.5-1 2016. can modernize ancient stray results modernobject <- update(ancientobject).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Result Object from Constrained Ordination — cca.object","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Result Object from Constrained Ordination — cca.object","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":null,"dir":"Reference","previous_headings":"","what":"Multinomial Species Classification Method (CLAM) — clamtest","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"CLAM statistical approach classifying generalists specialists two distinct habitats described Chazdon et al. (2011).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"clamtest(comm, groups, coverage.limit = 10, specialization = 2/3, npoints = 20, alpha = 0.05/20) # S3 method for clamtest summary(object, ...) # S3 method for clamtest plot(x, xlab, ylab, main, pch = 21:24, col.points = 1:4, col.lines = 2:4, lty = 1:3, position = \"bottomright\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"comm Community matrix, consisting counts. groups vector identifying two habitats. Must exactly two unique values levels. Habitat IDs grouping vector must match corresponding rows community matrix comm. coverage.limit Integer, sample coverage based correction applied rare species counts limit. Sample coverage calculated separately two habitats. Sample relative abundances used species higher equal coverage.limit total counts per habitat. specialization Numeric, specialization threshold value 0 1. value \\(2/3\\) represents ‘supermajority’ rule, value \\(1/2\\) represents ‘simple majority’ rule assign shared species habitat specialists. npoints Integer, number points used determine boundary lines plots. alpha Numeric, nominal significance level individual tests. default value reduces conventional limit \\(0.05\\) account overdispersion multiple testing several species simultaneously. However, firm reason exactly limit. x, object Fitted model object class \"clamtest\". xlab, ylab Labels plot axes. main Main title plot. pch, col.points Symbols colors used plotting species groups. lty, col.lines Line types colors boundary lines plot separate species groups. position Position figure legend, see legend specification details. Legend shown position = NULL. ... Additional arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"method uses multinomial model based estimated species relative abundance two habitats (, B). minimizes bias due differences sampling intensities two habitat types well bias due insufficient sampling within habitat. method permits robust statistical classification habitat specialists generalists, without excluding rare species priori (Chazdon et al. 2011). Based user-defined specialization threshold, model classifies species one four groups: (1) generalists; (2) habitat specialists; (3) habitat B specialists; (4) rare classify confidence.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"data frame (class attribute \"clamtest\"), columns: Species: species name (column names comm), Total_**: total count habitat , Total_*B*: total count habitat B, Classes: species classification, factor levels Generalist, Specialist_**, Specialist_*B*, Too_rare. ** *B* placeholders habitat names/labels found data. summary method returns descriptive statistics results. plot method returns values invisibly produces bivariate scatterplot species total abundances two habitats. Symbols boundary lines shown species groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Chazdon, R. L., Chao, ., Colwell, R. K., Lin, S.-Y., Norden, N., Letcher, S. G., Clark, D. B., Finegan, B. Arroyo J. P.(2011). novel statistical method classifying habitat generalists specialists. Ecology 92, 1332--1343.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Peter Solymos solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"code tested standalone CLAM software provided website Anne Chao (http://chao.stat.nthu.edu.tw/wordpress); minor inconsistencies found, especially finding threshold 'rare' species. inconsistencies probably due numerical differences two implementation. current R implementation uses root finding iso-lines instead iterative search. original method (Chazdon et al. 2011) two major problems: assumes error distribution multinomial. justified choice individuals freely distributed, -dispersion clustering individuals. ecological data, variance much higher multinomial assumption, therefore test statistic optimistic. original authors suggest multiple testing adjustment multiple testing based number points (npoints) used draw critical lines plot, whereas adjustment based number tests (.e., tested species). function uses numerical values original paper, automatic connection npoints alpha arguments, must work adjustment .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"data(mite) data(mite.env) sol <- with(mite.env, clamtest(mite, Shrub==\"None\", alpha=0.005)) summary(sol) #> Two Groups Species Classification Method (CLAM) #> #> Specialization threshold = 0.6666667 #> Alpha level = 0.005 #> #> Estimated sample coverage: #> FALSE TRUE #> 1.0000 0.9996 #> #> Minimum abundance for classification: #> FALSE TRUE #> 27 9 #> #> Species Proportion #> Generalist 10 0.286 #> Specialist_FALSE 14 0.400 #> Specialist_TRUE 4 0.114 #> Too_rare 7 0.200 head(sol) #> Species Total_FALSE Total_TRUE Classes #> 1 Brachy 534 77 Generalist #> 2 PHTH 89 0 Specialist_FALSE #> 3 HPAV 389 207 Generalist #> 4 RARD 85 0 Specialist_FALSE #> 5 SSTR 22 0 Too_rare #> 6 Protopl 26 0 Too_rare plot(sol)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an Object for Null Model Algorithms — commsim","title":"Create an Object for Null Model Algorithms — commsim","text":"commsim function can used feed Null Model algorithms nullmodel analysis. make.commsim function returns various predefined algorithm types (see Details). functions represent low level interface community null model infrastructure vegan intent extensibility, less emphasis direct use users.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"commsim(method, fun, binary, isSeq, mode) make.commsim(method) # S3 method for commsim print(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create an Object for Null Model Algorithms — commsim","text":"method Character, name algorithm. fun function. possible formal arguments function see Details. binary Logical, algorithm applies presence-absence count matrices. isSeq Logical, algorithm sequential (needs burnin thinning) . mode Character, storage mode community matrix, either \"integer\" \"double\". x object class commsim. ... Additional arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create an Object for Null Model Algorithms — commsim","text":"function fun must return array dim(nr, nc, n), must take following arguments: x: input matrix, n: number permuted matrices output, nr: number rows, nc: number columns, rs: vector row sums, cs: vector column sums, rf: vector row frequencies (non-zero cells), cf: vector column frequencies (non-zero cells), s: total sum x, fill: matrix fill (non-zero cells), thin: thinning value sequential algorithms, ...: additional arguments. can define null model, several null model algorithm pre-defined can called name. predefined algorithms described detail following chapters. binary null models produce matrices zeros (absences) ones (presences) also input matrix quantitative. two types quantitative data: Counts integers natural unit individuals can shuffled, abundances can real (floating point) values natural subunit shuffling. quantitative models can handle counts, able handle real values. null models sequential next matrix derived current one. makes models dependent previous models, usually must thin matrices study sequences stability: see oecosimu details instructions. See Examples structural constraints imposed algorithm defining null model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"binary-null-models","dir":"Reference","previous_headings":"","what":"Binary null models","title":"Create an Object for Null Model Algorithms — commsim","text":"binary null models preserve fill: number presences conversely number absences. classic models may also preserve column (species) frequencies (c0) row frequencies species richness site (r0) take account commonness rarity species (r1, r2). Algorithms swap, tswap, curveball, quasiswap backtracking preserve row column frequencies. Three first ones sequential two latter non-sequential produce independent matrices. Basic algorithms reviewed Wright et al. (1998). \"r00\": non-sequential algorithm binary matrices preserves number presences (fill). \"r0\": non-sequential algorithm binary matrices preserves site (row) frequencies. \"r1\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses column marginal frequencies probabilities selecting species. \"r2\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses squared column marginal frequencies probabilities selecting species. \"c0\": non-sequential algorithm binary matrices preserves species frequencies (Jonsson 2001). \"swap\": sequential algorithm binary matrices changes matrix structure, influence marginal sums (Gotelli & Entsminger 2003). inspects \\(2 \\times 2\\) submatrices long swap can done. \"tswap\": sequential algorithm binary matrices. \"swap\" algorithm, tries fixed number times performs zero many swaps one step (according thin argument call). approach suggested Miklós & Podani (2004) found ordinary swap may lead biased sequences, since columns rows easily swapped. \"curveball\": sequential method binary matrices implements ‘Curveball’ algorithm Strona et al. (2014). algorithm selects two random rows finds set unique species occur one rows. algorithm distributes set unique species rows preserving original row frequencies. Zero several species swapped one step, usually matrix perturbed strongly sequential methods. \"quasiswap\": non-sequential algorithm binary matrices implements method matrix first filled honouring row column totals, integers may larger one. method inspects random \\(2 \\times 2\\) matrices performs quasiswap . addition ordinary swaps, quasiswap can reduce numbers one ones preserving marginal totals (Miklós & Podani 2004). method non-sequential, accepts thin argument: convergence checked every thin steps. allows performing several ordinary swaps addition fill changing swaps helps reducing removing bias. \"greedyqswap\": greedy variant quasiswap. greedy step, one element \\(2 \\times 2\\) matrix taken \\(> 1\\) elements. greedy steps biased, method can thinned, first thin steps greedy. Even modest thinning (say thin = 20) removes reduces bias, thin = 100 (1% greedy steps) looks completely safe still speeds simulation. code experimental provided scrutiny, tested bias use. \"backtracking\": non-sequential algorithm binary matrices implements filling method constraints row column frequencies (Gotelli & Entsminger 2001). matrix first filled randomly, typically row column sums reached incidences filled . begins \"backtracking\", incidences removed, filling started , backtracking done many times incidences filled matrix. results may biased inspected carefully use.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-models-for-counts-with-fixed-marginal-sums","dir":"Reference","previous_headings":"","what":"Quantitative Models for Counts with Fixed Marginal Sums","title":"Create an Object for Null Model Algorithms — commsim","text":"models shuffle individuals counts keep marginal sums fixed, marginal frequencies preserved. Algorithm r2dtable uses standard R function r2dtable also used simulated \\(P\\)-values chisq.test. Algorithm quasiswap_count uses , preserves original fill. Typically means increasing numbers zero cells result zero-inflated respect r2dtable. \"r2dtable\": non-sequential algorithm count matrices. algorithm keeps matrix sum row/column sums constant. Based r2dtable. \"quasiswap_count\": non-sequential algorithm count matrices. algorithm similar Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function preserving row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping \\(2 \\times 2\\) submatrices (see \"swap_count\" algorithm details) maintain row column totals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative swap models similar binary swap, swap largest permissible value. models section maintain fill perform quantitative swap can done without changing fill. Single step swap often changes matrix little. particular, cell counts variable, high values change slowly. Checking chain stability independence even crucial binary swap, strong thinning often needed. models never used without inspecting properties current data. null models can also defined using permatswap function. \"swap_count\": sequential algorithm count matrices. algorithm find \\(2 \\times 2\\) submatrices can swapped leaving column row totals fill unchanged. algorithm finds largest value submatrix can swapped (\\(d\\)). Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. \"abuswap_r\": sequential algorithm count nonnegative real valued matrices fixed row frequencies (see also permatswap). algorithm similar swap_count, uses different swap value row \\(2 \\times 2\\) submatrix. step changes corresponding column sums, honours matrix fill, row sums, row/column frequencies (Hardy 2008; randomization scheme 2x). \"abuswap_c\": sequential algorithm count nonnegative real valued matrices fixed column frequencies (see also permatswap). algorithm similar previous one, operates columns. step changes corresponding row sums, honours matrix fill, column sums, row/column frequencies (Hardy 2008; randomization scheme 3x).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-and-shuffle-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap and Shuffle Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative Swap Shuffle methods (swsh methods) preserve fill column row frequencies, also either row column sums. methods first perform binary quasiswap shuffle original quantitative data non-zero cells. samp methods shuffle original non-zero cell values can used also non-integer data. methods redistribute individuals randomly among non-zero cells can used integer data. shuffling either free whole matrix, within rows (r methods) within columns (c methods). Shuffling within row preserves row sums, shuffling within column preserves column sums. models can also defined permatswap. \"swsh_samp\": non-sequential algorithm quantitative data (either integer counts non-integer values). Original non-zero values values shuffled. \"swsh_both\": non-sequential algorithm count data. Individuals shuffled freely non-zero cells. \"swsh_samp_r\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately row. \"swsh_samp_c\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately column. \"swsh_both_r\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values within row. \"swsh_both_c\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values column.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-shuffle-methods","dir":"Reference","previous_headings":"","what":"Quantitative Shuffle Methods","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative shuffle methods generalizations binary models r00, r0 c0. _ind methods shuffle individuals grand sum, row sum column sums preserved. methods similar r2dtable still slacker constraints marginal sums. _samp _both methods first apply corresponding binary model similar restriction marginal frequencies distribute quantitative values non-zero cells. _samp models shuffle original cell values can therefore handle also non-count real values. _both models shuffle individuals among non-zero values. shuffling whole matrix r00_, within row r0_ within column c0_ cases. \"r00_ind\": non-sequential algorithm count matrices. algorithm preserves grand sum individuals shuffled among cells matrix. \"r0_ind\": non-sequential algorithm count matrices. algorithm preserves row sums individuals shuffled among cells row matrix. \"c0_ind\": non-sequential algorithm count matrices. algorithm preserves column sums individuals shuffled among cells column matrix. \"r00_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves grand sum cells matrix shuffled. \"r0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves row sums cells within row shuffled. \"c0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves column sums constant cells within column shuffled. \"r00_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells matrix shuffled. \"r0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells row shuffled. \"c0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells column shuffled.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create an Object for Null Model Algorithms — commsim","text":"object class commsim elements corresponding arguments (method, binary, isSeq, mode, fun). input make.comsimm commsim object, returned without evaluation. case, character method argument matched predefined algorithm names. error message issued none found. method argument missing, function returns names currently available null model algorithms character vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Create an Object for Null Model Algorithms — commsim","text":"Gotelli, N.J. & Entsminger, N.J. (2001). Swap fill algorithms null model analysis: rethinking knight's tour. Oecologia 129, 281--291. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97. Strona, G., Nappo, D., Boccacci, F., Fattorini, S. & San-Miguel-Ayanz, J. (2014). fast unbiased procedure randomize ecological binary matrices fixed row column totals. Nature Communications 5:4114 doi:10.1038/ncomms5114 . Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an Object for Null Model Algorithms — commsim","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"## write the r00 algorithm f <- function(x, n, ...) array(replicate(n, sample(x)), c(dim(x), n)) (cs <- commsim(\"r00\", fun=f, binary=TRUE, isSeq=FALSE, mode=\"integer\")) #> An object of class “commsim” #> ‘r00’ method (binary, non-sequential, integer mode) #> ## retrieving the sequential swap algorithm (cs <- make.commsim(\"swap\")) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## feeding a commsim object as argument make.commsim(cs) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## making the missing c1 model using r1 as a template ## non-sequential algorithm for binary matrices ## that preserves the species (column) frequencies, ## but uses row marginal frequencies ## as probabilities of selecting sites f <- function (x, n, nr, nc, rs, cs, ...) { out <- array(0L, c(nr, nc, n)) J <- seq_len(nc) storage.mode(rs) <- \"double\" for (k in seq_len(n)) for (j in J) out[sample.int(nr, cs[j], prob = rs), j, k] <- 1L out } cs <- make.commsim(\"r1\") cs$method <- \"c1\" cs$fun <- f ## structural constraints diagfun <- function(x, y) { c(sum = sum(y) == sum(x), fill = sum(y > 0) == sum(x > 0), rowSums = all(rowSums(y) == rowSums(x)), colSums = all(colSums(y) == colSums(x)), rowFreq = all(rowSums(y > 0) == rowSums(x > 0)), colFreq = all(colSums(y > 0) == colSums(x > 0))) } evalfun <- function(meth, x, n) { m <- nullmodel(x, meth) y <- simulate(m, nsim=n) out <- rowMeans(sapply(1:dim(y)[3], function(i) diagfun(attr(y, \"data\"), y[,,i]))) z <- as.numeric(c(attr(y, \"binary\"), attr(y, \"isSeq\"), attr(y, \"mode\") == \"double\")) names(z) <- c(\"binary\", \"isSeq\", \"double\") c(z, out) } x <- matrix(rbinom(10*12, 1, 0.5)*rpois(10*12, 3), 12, 10) algos <- make.commsim() a <- t(sapply(algos, evalfun, x=x, n=10)) print(as.table(ifelse(a==1,1,0)), zero.print = \".\") #> binary isSeq double sum fill rowSums colSums rowFreq colFreq #> r00 1 . . 1 1 . . . . #> c0 1 . . 1 1 . 1 . 1 #> r0 1 . . 1 1 1 . 1 . #> r1 1 . . 1 1 1 . 1 . #> r2 1 . . 1 1 1 . 1 . #> quasiswap 1 . . 1 1 1 1 1 1 #> greedyqswap 1 . . 1 1 1 1 1 1 #> swap 1 1 . 1 1 1 1 1 1 #> tswap 1 1 . 1 1 1 1 1 1 #> curveball 1 1 . 1 1 1 1 1 1 #> backtrack 1 . . 1 1 1 1 1 1 #> r2dtable . . . 1 . 1 1 . . #> swap_count . 1 . 1 1 1 1 . . #> quasiswap_count . . . 1 1 1 1 . . #> swsh_samp . . 1 1 1 . . 1 1 #> swsh_both . . . 1 1 . . 1 1 #> swsh_samp_r . . 1 1 1 1 . 1 1 #> swsh_samp_c . . 1 1 1 . 1 1 1 #> swsh_both_r . . . 1 1 1 . 1 1 #> swsh_both_c . . . 1 1 . 1 1 1 #> abuswap_r . 1 1 1 1 1 . 1 1 #> abuswap_c . 1 1 1 1 . 1 1 1 #> r00_samp . . 1 1 1 . . . . #> c0_samp . . 1 1 1 . 1 . 1 #> r0_samp . . 1 1 1 1 . 1 . #> r00_ind . . . 1 . . . . . #> c0_ind . . . 1 . . 1 . . #> r0_ind . . . 1 . 1 . . . #> r00_both . . . 1 1 . . . . #> c0_both . . . 1 1 . 1 . 1 #> r0_both . . . 1 1 1 . 1 ."},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":null,"dir":"Reference","previous_headings":"","what":"Contribution Diversity Approach — contribdiv","title":"Contribution Diversity Approach — contribdiv","text":"contribution diversity approach based differentiation within-unit among-unit diversity using additive diversity partitioning unit distinctiveness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Contribution Diversity Approach — contribdiv","text":"","code":"contribdiv(comm, index = c(\"richness\", \"simpson\"), relative = FALSE, scaled = TRUE, drop.zero = FALSE) # S3 method for contribdiv plot(x, sub, xlab, ylab, ylim, col, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Contribution Diversity Approach — contribdiv","text":"comm community data matrix samples rows species column. index Character, diversity index calculated. relative Logical, TRUE contribution diversity values expressed signed deviation mean. See details. scaled Logical, TRUE relative contribution diversity values scaled sum gamma values (index = \"richness\") sum gamma values times number rows comm (index = \"simpson\"). See details. drop.zero Logical, empty rows dropped result? empty rows dropped, corresponding results NAs. x object class \"contribdiv\". sub, xlab, ylab, ylim, col Graphical arguments passed plot. ... arguments passed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Contribution Diversity Approach — contribdiv","text":"approach proposed Lu et al. (2007). Additive diversity partitioning (see adipart references) deals relation mean alpha total (gamma) diversity. Although alpha diversity values often vary considerably. Thus, contributions sites total diversity uneven. site specific contribution measured contribution diversity components. unit e.g. many unique species contribute higher level (gamma) diversity another unit number species, common. Distinctiveness species \\(j\\) can defined number sites occurs (\\(n_j\\)), sum relative frequencies (\\(p_j\\)). Relative frequencies computed sitewise \\(sum_j{p_ij}\\)s site \\(\\) sum \\(1\\). contribution site \\(\\) total diversity given \\(alpha_i = sum_j(1 / n_ij)\\) dealing richness \\(alpha_i = sum(p_{ij} * (1 - p_{ij}))\\) Simpson index. unit distinctiveness site \\(\\) average species distinctiveness, averaging species occur site \\(\\). species richness: \\(alpha_i = mean(n_i)\\) (paper, second equation contains typo, \\(n\\) without index). Simpson index: \\(alpha_i = mean(n_i)\\). Lu et al. (2007) gives -depth description different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Contribution Diversity Approach — contribdiv","text":"object class \"contribdiv\" inheriting data frame. Returned values alpha, beta gamma components sites (rows) community matrix. \"diff.coef\" attribute gives differentiation coefficient (see Examples).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Contribution Diversity Approach — contribdiv","text":"Lu, H. P., Wagner, H. H. Chen, X. Y. 2007. contribution diversity approach evaluate species diversity. Basic Applied Ecology, 8, 1--12.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Contribution Diversity Approach — contribdiv","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Contribution Diversity Approach — contribdiv","text":"","code":"## Artificial example given in ## Table 2 in Lu et al. 2007 x <- matrix(c( 1/3,1/3,1/3,0,0,0, 0,0,1/3,1/3,1/3,0, 0,0,0,1/3,1/3,1/3), 3, 6, byrow = TRUE, dimnames = list(LETTERS[1:3],letters[1:6])) x #> a b c d e f #> A 0.3333333 0.3333333 0.3333333 0.0000000 0.0000000 0.0000000 #> B 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 0.0000000 #> C 0.0000000 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 ## Compare results with Table 2 contribdiv(x, \"richness\") #> alpha beta gamma #> A 1 1.5 2.5 #> B 1 0.5 1.5 #> C 1 1.0 2.0 contribdiv(x, \"simpson\") #> alpha beta gamma #> A 0.6666667 0.1851852 0.8518519 #> B 0.6666667 0.1111111 0.7777778 #> C 0.6666667 0.1481481 0.8148148 ## Relative contribution (C values), compare with Table 2 (cd1 <- contribdiv(x, \"richness\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.5 0.5 #> B 0 -0.5 -0.5 #> C 0 0.0 0.0 (cd2 <- contribdiv(x, \"simpson\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.03703704 0.03703704 #> B 0 -0.03703704 -0.03703704 #> C 0 0.00000000 0.00000000 ## Differentiation coefficients attr(cd1, \"diff.coef\") # D_ST #> [1] 0.5 attr(cd2, \"diff.coef\") # D_DT #> [1] 0.1818182 ## BCI data set data(BCI) opar <- par(mfrow=c(2,2)) plot(contribdiv(BCI, \"richness\"), main = \"Absolute\") plot(contribdiv(BCI, \"richness\", relative = TRUE), main = \"Relative\") plot(contribdiv(BCI, \"simpson\")) plot(contribdiv(BCI, \"simpson\", relative = TRUE)) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] Distance-based Redundancy Analysis — dbrda","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Distance-based redundancy analysis (dbRDA) ordination method similar Redundancy Analysis (rda), allows non-Euclidean dissimilarity indices, Manhattan Bray-Curtis distance. Despite non-Euclidean feature, analysis strictly linear metric. called Euclidean distance, results identical rda, dbRDA less efficient. Functions dbrda constrained versions metric scaling, .k.. principal coordinates analysis, based Euclidean distance can used, useful, dissimilarity measures. Function capscale simplified version based Euclidean approximation dissimilarities. functions can also perform unconstrained principal coordinates analysis, optionally using extended dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"dbrda(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...) capscale(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, comm = NULL, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"formula Model formula. function can called formula interface. usual features formula hold, especially defined cca rda. LHS must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist function given dfun used find dissimilarities. RHS defines constraints. constraints can continuous variables factors, can transformed within formula, can interactions typical formula. RHS can special term Condition defines variables “partialled ” constraints, just like rda cca. allows use partial dbRDA. data Data frame containing variables right hand side model formula. distance name dissimilarity (distance) index LHS formula data frame instead dissimilarity matrix. sqrt.dist Take square roots dissimilarities. See section Details . comm Community data frame used finding species scores LHS formula dissimilarity matrix. used LHS data frame. supplied, “species scores” unavailable dissimilarities supplied. N.B., available capscale: dbrda return species scores. Function sppscores can used add species scores missing. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). \"lingoes\" (TRUE) uses recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. latter one cmdscale. dfun Distance dissimilarity function used. function returning standard \"dist\" taking index name first argument can used. metaMDSdist Use metaMDSdist similarly metaMDS. means automatic data transformation using extended flexible shortest path dissimilarities (function stepacross) many dissimilarities based shared species. na.action Handling missing values constraints conditions. default (na.fail) stop missing values. Choices na.omit na.exclude delete rows missing values, differ representation results. na.omit non-missing site scores shown, na.exclude gives NA scores missing observations. Unlike rda, WA scores available missing constraints conditions. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data (given formula comm argument). ... parameters passed underlying functions (e.g., metaMDSdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Functions dbrda capscale provide two alternative implementations dbRDA. Function dbrda based McArdle & Anderson (2001) directly decomposes dissimilarities. Euclidean distances results identical rda. Non-Euclidean dissimilarities may give negative eigenvalues associated imaginary axes. Function capscale based Legendre & Anderson (1999): dissimilarity data first ordinated using metric scaling, ordination results analysed rda. capscale ignores imaginary component give negative eigenvalues (report magnitude imaginary component). user supplied community data frame instead dissimilarities, functions find dissimilarities using vegdist distance function given dfun specified distance. functions accept distance objects vegdist, dist, method producing compatible objects. constraining variables can continuous factors , can interaction terms, can transformed call. Moreover, can special term Condition just like rda cca “partial” analysis can performed. Function dbrda return species scores, can also missing capscale, can added analysis using function sppscores. Non-Euclidean dissimilarities can produce negative eigenvalues (Legendre & Anderson 1999, McArdle & Anderson 2001). negative eigenvalues, printed output capscale add column sums positive eigenvalues item sum negative eigenvalues, dbrda add column giving number real dimensions positive eigenvalues. negative eigenvalues disturbing, functions let distort dissimilarities non-negative eigenvalues produced argument add = TRUE. Alternatively, sqrt.dist = TRUE, square roots dissimilarities can used may help avoiding negative eigenvalues (Legendre & Anderson 1999). functions can also used perform ordinary metric scaling .k.. principal coordinates analysis using formula constant right hand side, comm ~ 1. metaMDSdist = TRUE, function can automatic data standardization use extended dissimilarities using function stepacross similarly non-metric multidimensional scaling metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"functions return object class dbrda capscale inherit rda. See cca.object description result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Anderson, M.J. & Willis, T.J. (2003). Canonical analysis principal coordinates: useful method constrained ordination ecology. Ecology 84, 511--525. Gower, J.C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. Legendre, P. & Legendre, L. (2012). Numerical Ecology. 3rd English Edition. Elsevier. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290--297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Function dbrda implements real distance-based RDA preferred capscale. capscale originally developed variant constrained analysis proximities (Anderson & Willis 2003), developments made similar dbRDA. However, discards imaginary dimensions negative eigenvalues ordination significance tests area based real dimensions positive eigenvalues. capscale may removed vegan future. vegan since 2003 (CRAN release 1.6-0) dbrda first released 2016 (version 2.4-0), removal capscale may disruptive historical examples scripts, modern times dbrda used. inertia named dissimilarity index defined dissimilarity data, unknown distance information missing. largest original dissimilarity larger 4, capscale handles input similarly rda bases analysis variance instead sum squares. Keyword mean added inertia cases, e.g. Euclidean Manhattan distances. Inertia based squared index, keyword squared added name distance, unless data square root transformed (argument sqrt.dist=TRUE). additive constant used argument add, Lingoes Cailliez adjusted added name inertia, value constant printed.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"data(varespec, varechem) ## dbrda dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\") #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\") #> #> Inertia Proportion Rank RealDims #> Total 4.5444 1.0000 #> Conditional 0.9726 0.2140 1 #> Constrained 0.9731 0.2141 3 3 #> Unconstrained 2.5987 0.5718 19 13 #> Inertia is squared Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5362 0.3198 0.1171 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9054 0.5070 0.3336 0.2581 0.2027 0.1605 0.1221 0.0825 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues with sqrt distances dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\", sqrt.dist = TRUE) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\", sqrt.dist = TRUE) #> #> Inertia Proportion Rank #> Total 6.9500 1.0000 #> Conditional 0.9535 0.1372 1 #> Constrained 1.2267 0.1765 3 #> Unconstrained 4.7698 0.6863 19 #> Inertia is Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5817 0.4086 0.2365 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9680 0.6100 0.4469 0.3837 0.3371 0.3012 0.2558 0.2010 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues also with Jaccard distances (m <- dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"jaccard\")) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"jaccard\") #> #> Inertia Proportion Rank #> Total 6.5044 1.0000 #> Conditional 1.0330 0.1588 1 #> Constrained 1.2068 0.1855 3 #> Unconstrained 4.2646 0.6557 19 #> Inertia is squared Jaccard distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5992 0.3994 0.2082 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.0388 0.6441 0.4518 0.3759 0.3239 0.2785 0.2279 0.1644 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## add species scores sppscores(m) <- wisconsin(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":null,"dir":"Reference","previous_headings":"","what":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Performs detrended correspondence analysis basic reciprocal averaging orthogonal correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"decorana(veg, iweigh=0, iresc=4, ira=0, mk=26, short=0, before=NULL, after=NULL) # S3 method for decorana plot(x, choices=c(1,2), origin=TRUE, display=c(\"both\",\"sites\",\"species\",\"none\"), cex = 0.8, cols = c(1,2), type, xlim, ylim, ...) # S3 method for decorana text(x, display = c(\"sites\", \"species\"), labels, choices = 1:2, origin = TRUE, select, ...) # S3 method for decorana points(x, display = c(\"sites\", \"species\"), choices=1:2, origin = TRUE, select, ...) # S3 method for decorana scores(x, display=\"sites\", choices=1:4, origin=TRUE, tidy=FALSE, ...) downweight(veg, fraction = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"veg Community data, matrix-like object. iweigh Downweighting rare species (0: ). iresc Number rescaling cycles (0: rescaling). ira Type analysis (0: detrended, 1: basic reciprocal averaging). mk Number segments rescaling. short Shortest gradient rescaled. Hill's piecewise transformation: values transformation. Hill's piecewise transformation: values transformation -- must correspond values . x decorana result object. choices Axes shown. origin Use true origin even detrended correspondence analysis. display Display sites, species, neither. cex Plot character size. cols Colours used sites species. type Type plots, partial match \"text\", \"points\" \"none\". labels Optional text used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. xlim, ylim x y limits (min,max) plot. fraction Abundance fraction downweighting begins. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score (\"sites\", \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. ... arguments plot function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"late 1970s, correspondence analysis became method choice ordination vegetation science, since seemed better able cope non-linear species responses principal components analysis. However, even correspondence analysis can produce arc-shaped configuration single gradient. Mark Hill developed detrended correspondence analysis correct two assumed ‘faults’ correspondence analysis: curvature straight gradients packing sites ends gradient. curvature removed replacing orthogonalization axes detrending. orthogonalization successive axes made non-correlated, detrending remove systematic dependence axes. Detrending performed using smoothing window mk segments. packing sites ends gradient undone rescaling axes extraction. rescaling, axis supposed scaled ‘SD’ units, average width Gaussian species responses supposed one whole axis. innovations piecewise linear transformation species abundances downweighting rare species regarded unduly high influence ordination axes. seems detrending actually works twisting ordination space, results look non-curved two-dimensional projections (‘lolly paper effect’). result, points usually easily recognized triangular diamond shaped pattern, obviously artefact detrending. Rescaling works differently commonly presented, . decorana use, even evaluate, widths species responses. Instead, tries equalize weighted standard deviation species scores axis segments (parameter mk effect, since decorana finds segments internally). Function tolerance returns internal criterion can used assess success rescaling. plot method plots species site scores. Classical decorana scaled axes smallest site score 0 ( smallest species score negative), summary, plot scores use true origin, unless origin = FALSE. addition proper eigenvalues, function reports ‘decorana values’ detrended analysis. ‘decorana values’ values legacy code decorana returns eigenvalues. estimated iteration, describe joint effects axes detrending. ‘decorana values’ estimated rescaling show effect eigenvalues. proper eigenvalues estimated extraction axes ratio weighted sum squares site species scores even detrended rescaled solutions. eigenvalues estimated axis separately, additive, higher decorana axes can show effects already explained prior axes. ‘Additive eigenvalues’ cleansed effects prior axes, can assumed add total inertia (scaled Chi-square). proportions cumulative proportions explained can use eigenvals.decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana returns object class \"decorana\", print, summary, scores, plot, points text methods, support functions eigenvals, bstick, screeplot, predict tolerance. downweight independent function can also used methods decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Hill, M.O. Gauch, H.G. (1980). Detrended correspondence analysis: improved ordination technique. Vegetatio 42, 47--58. Oksanen, J. Minchin, P.R. (1997). Instability ordination results changes input data order: explanations remedies. Journal Vegetation Science 8, 447--454.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Mark O. Hill wrote original Fortran code, R port Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana uses central numerical engine original Fortran code (public domain), 1/3 original program. tried implement original behaviour, although great part preparatory steps written R language, may differ somewhat original code. However, well-known bugs corrected strict criteria used (Oksanen & Minchin 1997). Please note really need piecewise transformation even downweighting within decorana, since powerful extensive alternatives R, options included compliance original software. different fraction abundance needed downweighting, function downweight must applied decorana. Function downweight indeed can applied prior correspondence analysis, can used together cca, . Github package natto R implementation decorana allows easier inspection algorithm also easier development function. vegan 2.6-6 earlier summary method, nothing useful now defunct. former information can extracted scores weights.decorana.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"data(varespec) vare.dca <- decorana(varespec) vare.dca #> #> Call: #> decorana(veg = varespec) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.0832 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.5235 0.3253 0.20010 0.19176 #> Additive Eigenvalues 0.5235 0.3217 0.17919 0.11922 #> Decorana values 0.5249 0.1572 0.09669 0.06075 #> Axis lengths 2.8161 2.2054 1.54650 1.64864 #> plot(vare.dca) ### the detrending rationale: gaussresp <- function(x,u) exp(-(x-u)^2/2) x <- seq(0,6,length=15) ## The gradient u <- seq(-2,8,len=23) ## The optima pack <- outer(x,u,gaussresp) matplot(x, pack, type=\"l\", main=\"Species packing\") opar <- par(mfrow=c(2,2)) plot(scores(prcomp(pack)), asp=1, type=\"b\", main=\"PCA\") plot(scores(decorana(pack, ira=1)), asp=1, type=\"b\", main=\"CA\") plot(scores(decorana(pack)), asp=1, type=\"b\", main=\"DCA\") plot(scores(cca(pack ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") ### Let's add some noise: noisy <- (0.5 + runif(length(pack)))*pack par(mfrow=c(2,1)) matplot(x, pack, type=\"l\", main=\"Ideal model\") matplot(x, noisy, type=\"l\", main=\"Noisy model\") par(mfrow=c(2,2)) plot(scores(prcomp(noisy)), type=\"b\", main=\"PCA\", asp=1) plot(scores(decorana(noisy, ira=1)), type=\"b\", main=\"CA\", asp=1) plot(scores(decorana(noisy)), type=\"b\", main=\"DCA\", asp=1) plot(scores(cca(noisy ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":null,"dir":"Reference","previous_headings":"","what":"Standardization Methods for Community Ecology — decostand","title":"Standardization Methods for Community Ecology — decostand","text":"function provides popular (effective) standardization methods community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"decostand(x, method, MARGIN, range.global, logbase = 2, na.rm=FALSE, ...) wisconsin(x) decobackstand(x, zap = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Standardization Methods for Community Ecology — decostand","text":"x Community data, matrix-like object. decobackstand standardized data. method Standardization method. See Details available options. MARGIN Margin, default acceptable. 1 = rows, 2 = columns x. range.global Matrix range found method = \"range\". allows using ranges across subsets data. dimensions MARGIN must match x. logbase logarithm base used method = \"log\". na.rm Ignore missing values row column standardizations. zap Make near-zero values exact zeros avoid negative values exaggerated estimates species richness. ... arguments function (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Standardization Methods for Community Ecology — decostand","text":"function offers following standardization methods community data: total: divide margin total (default MARGIN = 1). max: divide margin maximum (default MARGIN = 2). frequency: divide margin total multiply number non-zero items, average non-zero entries one (Oksanen 1983; default MARGIN = 2). normalize: make margin sum squares equal one (default MARGIN = 1). range: standardize values range 0 ... 1 (default MARGIN = 2). values constant, transformed 0. rank, rrank: rank replaces abundance values increasing ranks leaving zeros unchanged, rrank similar uses relative ranks maximum 1 (default MARGIN = 1). Average ranks used tied values. standardize: scale x zero mean unit variance (default MARGIN = 2). pa: scale x presence/absence scale (0/1). chi.square: divide row sums square root column sums, adjust square root matrix total (Legendre & Gallagher 2001). used Euclidean distance, distances similar Chi-square distance used correspondence analysis. However, results cmdscale still differ, since CA weighted ordination method (default MARGIN = 1). hellinger: square root method = \"total\" (Legendre & Gallagher 2001). log: logarithmic transformation suggested Anderson et al. (2006): \\(\\log_b (x) + 1\\) \\(x > 0\\), \\(b\\) base logarithm; zeros left zeros. Higher bases give less weight quantities presences, logbase = Inf gives presence/absence scaling. Please note \\(\\log(x+1)\\). Anderson et al. (2006) suggested (strongly) modified Gower distance (implemented method = \"altGower\" vegdist), standardization can used independently distance indices. alr: Additive log ratio (\"alr\") transformation (Aitchison 1986) reduces data skewness compositionality bias. transformation assumes positive values, pseudocounts can added argument pseudocount. One rows/columns reference can given reference (name index). first row/column used default (reference = 1). Note transformation drops one row column transformed output data. alr transformation defined formally follows: $$alr = [log\\frac{x_1}{x_D}, ..., log\\frac{x_{D-1}}{x_D}]$$ denominator sample \\(x_D\\) can chosen arbitrarily. transformation often used pH chemistry measurenments. also commonly used multinomial logistic regression. Default MARGIN = 1 uses row reference. clr: centered log ratio (\"clr\") transformation proposed Aitchison (1986) used reduce data skewness compositionality bias. transformation frequent applications microbial ecology (see e.g. Gloor et al., 2017). clr transformation defined : $$clr = log\\frac{x}{g(x)} = log x - log g(x)$$ \\(x\\) single value, g(x) geometric mean \\(x\\). method can operate positive data; common way deal zeroes add pseudocount (e.g. smallest positive value data), either adding manually input data, using argument pseudocount decostand(x, method = \"clr\", pseudocount = 1). Adding pseudocount inevitably introduce bias; see rclr method one available solution. rclr: robust clr (\"rclr\") similar regular clr (see ) allows data contains zeroes. method use pseudocounts, unlike standard clr. robust clr (rclr) divides values geometric mean observed features; zero values kept zeroes, taken account. high dimensional data, geometric mean rclr approximates true geometric mean; see e.g. Martino et al. (2019) rclr transformation defined formally follows: $$rclr = log\\frac{x}{g(x > 0)}$$ \\(x\\) single value, \\(g(x > 0)\\) geometric mean sample-wide values \\(x\\) positive (> 0). Standardization, contrasted transformation, means entries transformed relative entries. methods default margin. MARGIN=1 means rows (sites normal data set) MARGIN=2 means columns (species normal data set). Command wisconsin shortcut common Wisconsin double standardization species (MARGIN=2) first standardized maxima (max) sites (MARGIN=1) site totals (tot). standardization methods give nonsense results negative data entries normally occur community data. empty sites species (constant method = \"range\"), many standardization change NaN. Function decobackstand can used transform standardized data back original. possible standardization may implemented cases possible. round-errors back-transformation exact, wise overwrite original data. zap=TRUE original zeros exact.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Standardization Methods for Community Ecology — decostand","text":"Returns standardized data frame, adds attribute \"decostand\" giving name applied standardization \"method\" attribute \"parameters\" appropriate transformation parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Standardization Methods for Community Ecology — decostand","text":"Jari Oksanen, Etienne Laliberté (method = \"log\"), Leo Lahti (alr, \"clr\" \"rclr\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Standardization Methods for Community Ecology — decostand","text":"Common transformations can made standard R functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Standardization Methods for Community Ecology — decostand","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Egozcue, J.J., Pawlowsky-Glahn, V., Mateu-Figueras, G., Barcel'o-Vidal, C. (2003) Isometric logratio transformations compositional data analysis. Mathematical Geology 35, 279--300. Gloor, G.B., Macklaim, J.M., Pawlowsky-Glahn, V. & Egozcue, J.J. (2017) Microbiome Datasets Compositional: Optional. Frontiers Microbiology 8, 2224. Legendre, P. & Gallagher, E.D. (2001) Ecologically meaningful transformations ordination species data. Oecologia 129, 271--280. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Oksanen, J. (1983) Ordination boreal heath-like vegetation principal component analysis, correspondence analysis multidimensional scaling. Vegetatio 52, 181--189.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"data(varespec) sptrans <- decostand(varespec, \"max\") apply(sptrans, 2, max) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv Descflex Betupube #> 1 1 1 1 1 1 1 1 #> Vacculig Diphcomp Dicrsp Dicrfusc Dicrpoly Hylosple Pleuschr Polypili #> 1 1 1 1 1 1 1 1 #> Polyjuni Polycomm Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 1 1 1 1 1 1 1 1 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris Cladchlo Cladbotr #> 1 1 1 1 1 1 1 1 #> Cladamau Cladsp Cetreric Cetrisla Flavniva Nepharct Stersp Peltapht #> 1 1 1 1 1 1 1 1 #> Icmaeric Cladcerv Claddefo Cladphyl #> 1 1 1 1 sptrans <- wisconsin(varespec) # CLR transformation for rows, with pseudocount varespec.clr <- decostand(varespec, \"clr\", pseudocount=1) # ALR transformation for rows, with pseudocount and reference sample varespec.alr <- decostand(varespec, \"alr\", pseudocount=1, reference=1) ## Chi-square: PCA similar but not identical to CA. ## Use wcmdscale for weighted analysis and identical results. sptrans <- decostand(varespec, \"chi.square\") plot(procrustes(rda(sptrans), cca(varespec)))"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Design your own Dissimilarities — designdist","title":"Design your own Dissimilarities — designdist","text":"Function designdist lets define dissimilarities using terms shared total quantities, number rows number columns. shared total quantities can binary, quadratic minimum terms. binary terms, shared component number shared species, totals numbers species sites. quadratic terms cross-products sums squares, minimum terms sums parallel minima row totals. Function designdist2 similar, finds dissimilarities among two data sets. Function chaodist lets define dissimilarities using terms supposed take account “unseen species” (see Chao et al., 2005 Details vegdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Design your own Dissimilarities — designdist","text":"","code":"designdist(x, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) designdist2(x, y, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) chaodist(x, method = \"1 - 2*U*V/(U+V)\", name)"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Design your own Dissimilarities — designdist","text":"x Input data. y Another input data set: dissimilarities calculated among rows x rows y. method Equation dissimilarities. can use terms J shared quantity, B totals, N number rows (sites) P number columns (species) chaodist can use terms U V. equation can also contain R functions accepts vector arguments returns vectors length. terms shared total components found. vectors x y \"quadratic\" terms J = sum(x*y), = sum(x^2), B = sum(y^2), \"minimum\" terms J = sum(pmin(x,y)), = sum(x) B = sum(y), \"binary\" terms either transforming data binary form (shared number species, number species row). abcd Use 2x2 contingency table notation binary data: \\(\\) number shared species, \\(b\\) \\(c\\) numbers species occurring one sites , \\(d\\) number species occur neither sites. alphagamma Use beta diversity notation terms alpha average alpha diversity compared sites, gamma diversity pooled sites, delta absolute value difference average alpha alpha diversities compared sites. Terms B refer alpha diversities compared sites. name name want use index. default combine method equation terms argument. maxdist Theoretical maximum dissimilarity, NA index open absolute maximum. necessary argument, used vegan functions, certain maximum, better supply value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Design your own Dissimilarities — designdist","text":"popular dissimilarity measures ecology can expressed help terms J, B, also involve matrix dimensions N P. examples can define designdist : function designdist can implement dissimilarity indices vegdist elsewhere, can also used implement many indices, amongst , described Legendre & Legendre (2012). can also used implement indices beta diversity described Koleff et al. (2003), also specific function betadiver purpose. want implement binary dissimilarities based 2x2 contingency table notation, can set abcd = TRUE. notation = J, b = -J, c = B-J, d = P--B+J. notation often used instead tangible default notation reasons opaque . alphagamma = TRUE possible use beta diversity notation terms alpha average alpha diversity gamma gamma diversity two compared sites. terms calculated alpha = (+B)/2, gamma = +B-J delta = abs(-B)/2. Terms B also available give alpha diversities individual compared sites. beta diversity terms may make sense binary terms (diversities expressed numbers species), calculated quadratic minimum terms well ( warning). Function chaodist similar designgist, uses terms U V Chao et al. (2005). terms supposed take account effects unseen species. U V scaled range \\(0 \\dots 1\\). take place B product U*V used place J designdist. Function chaodist can implement commonly used Chao et al. (2005) style dissimilarity: Function vegdist implements Jaccard-type Chao distance, documentation contains complete discussion calculation terms.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Design your own Dissimilarities — designdist","text":"designdist returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Design your own Dissimilarities — designdist","text":"Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005) new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence--absence data. J. Animal Ecol. 72, 367--382. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Design your own Dissimilarities — designdist","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Design your own Dissimilarities — designdist","text":"designdist use compiled code, based vectorized R code. designdist function can much faster vegdist, although latter uses compiled code. However, designdist skip missing values uses much memory calculations. use sum terms can numerically unstable. particularly, terms large, precision may lost. risk large number columns high, particularly large quadratic terms. precise calculations better use functions like dist vegdist robust numerical problems.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Design your own Dissimilarities — designdist","text":"","code":"data(BCI) ## Four ways of calculating the same Sørensen dissimilarity d0 <- vegdist(BCI, \"bray\", binary = TRUE) d1 <- designdist(BCI, \"(A+B-2*J)/(A+B)\") d2 <- designdist(BCI, \"(b+c)/(2*a+b+c)\", abcd = TRUE) d3 <- designdist(BCI, \"gamma/alpha - 1\", alphagamma = TRUE) ## Arrhenius dissimilarity: the value of z in the species-area model ## S = c*A^z when combining two sites of equal areas, where S is the ## number of species, A is the area, and c and z are model parameters. ## The A below is not the area (which cancels out), but number of ## species in one of the sites, as defined in designdist(). dis <- designdist(BCI, \"(log(A+B-J)-log(A+B)+log(2))/log(2)\") ## This can be used in clustering or ordination... ordiplot(cmdscale(dis)) #> species scores not available ## ... or in analysing beta diversity (without gradients) summary(dis) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2733 0.3895 0.4192 0.4213 0.4537 0.5906"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions extract statistics resemble deviance AIC result constrained correspondence analysis cca redundancy analysis rda. functions rarely needed directly, called step automatic model building. Actually, cca rda AIC functions certainly wrong.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# S3 method for cca deviance(object, ...) # S3 method for cca extractAIC(fit, scale = 0, k = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"object result constrained ordination (cca rda). fit fitted model constrained ordination. scale optional numeric specifying scale parameter model, see scale step. k numeric specifying \"weight\" equivalent degrees freedom (=:edf) part AIC formula. ... arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions find statistics resemble deviance AIC constrained ordination. Actually, constrained ordination methods log-Likelihood, means AIC deviance. Therefore use functions, use , trust . use functions, remains responsibility check adequacy result. deviance cca equal Chi-square residual data matrix fitting constraints. deviance rda defined residual sum squares. deviance function rda also used distance-based RDA dbrda. Function extractAIC mimics extractAIC.lm translating deviance AIC. little need call functions directly. However, called implicitly step function used automatic selection constraining variables. check resulting model criteria, statistics used unfounded. particular, penalty k properly defined, default k = 2 justified theoretically. continuous covariates, step function base model building magnitude eigenvalues, value k influences stopping point ( variables highest eigenvalues necessarily significant permutation tests anova.cca). also multi-class factors, value k capricious effect model building. step function pass arguments add1.cca drop1.cca, setting test = \"permutation\" provide permutation tests deletion addition can help judging validity model building.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"deviance functions return “deviance”, extractAIC returns effective degrees freedom “AIC”.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Godínez-Domínguez, E. & Freire, J. (2003) Information-theoretic approach selection spatial temporal models community organization. Marine Ecology Progress Series 253, 17--24.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions unfounded untested used directly implicitly. Moreover, usual caveats using step valid.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# The deviance of correspondence analysis equals Chi-square data(dune) data(dune.env) chisq.test(dune) #> Warning: Chi-squared approximation may be incorrect #> #> \tPearson's Chi-squared test #> #> data: dune #> X-squared = 1449, df = 551, p-value < 2.2e-16 #> deviance(cca(dune)) #> [1] 1448.956 # Stepwise selection (forward from an empty model \"dune ~ 1\") ord <- cca(dune ~ ., dune.env) step(cca(dune ~ 1, dune.env), scope = formula(ord)) #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC #> + Moisture 3 86.608 #> + Management 3 86.935 #> + A1 1 87.411 #> 87.657 #> + Manure 4 88.832 #> + Use 2 89.134 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC #> 86.608 #> + Management 3 86.813 #> + A1 1 86.992 #> + Use 2 87.259 #> + Manure 4 87.342 #> - Moisture 3 87.657 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #>"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":null,"dir":"Reference","previous_headings":"","what":"Morisita index of intraspecific aggregation — dispindmorisita","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Calculates Morisita index dispersion, standardized index values, called clumpedness uniform indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"dispindmorisita(x, unique.rm = FALSE, crit = 0.05, na.rm = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"x community data matrix, sites (samples) rows species columns. unique.rm logical, TRUE, unique species (occurring one sample) removed result. crit two-sided p-value used calculate critical Chi-squared values. na.rm logical. missing values (including NaN) omitted calculations?","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita index dispersion defined (Morisita 1959, 1962): Imor = n * (sum(xi^2) - sum(xi)) / (sum(xi)^2 - sum(xi)) \\(xi\\) count individuals sample \\(\\), \\(n\\) number samples (\\(= 1, 2, \\ldots, n\\)). \\(Imor\\) values 0 \\(n\\). uniform (hyperdispersed) patterns value falls 0 1, clumped patterns falls 1 \\(n\\). increasing sample sizes (.e. joining neighbouring quadrats), \\(Imor\\) goes \\(n\\) quadrat size approaches clump size. random patterns, \\(Imor = 1\\) counts samples follow Poisson frequency distribution. deviation random expectation (null hypothesis) can tested using critical values Chi-squared distribution \\(n-1\\) degrees freedom. Confidence intervals around 1 can calculated clumped \\(Mclu\\) uniform \\(Muni\\) indices (Hairston et al. 1971, Krebs 1999) (Chi2Lower Chi2Upper refers e.g. 0.025 0.975 quantile values Chi-squared distribution \\(n-1\\) degrees freedom, respectively, crit = 0.05): Mclu = (Chi2Lower - n + sum(xi)) / (sum(xi) - 1) Muni = (Chi2Upper - n + sum(xi)) / (sum(xi) - 1) Smith-Gill (1975) proposed scaling Morisita index [0, n] interval [-1, 1], setting -0.5 0.5 values confidence limits around random distribution rescaled value 0. rescale Morisita index, one following four equations apply calculate standardized index \\(Imst\\): () Imor >= Mclu > 1: Imst = 0.5 + 0.5 (Imor - Mclu) / (n - Mclu), (b) Mclu > Imor >= 1: Imst = 0.5 (Imor - 1) / (Mclu - 1), (c) 1 > Imor > Muni: Imst = -0.5 (Imor - 1) / (Muni - 1), (d) 1 > Muni > Imor: Imst = -0.5 + 0.5 (Imor - Muni) / Muni.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Returns data frame many rows number columns input data, four columns. Columns : imor unstandardized Morisita index, mclu clumpedness index, muni uniform index, imst standardized Morisita index, pchisq Chi-squared based probability null hypothesis random expectation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita, M. 1959. Measuring dispersion individuals analysis distributional patterns. Mem. Fac. Sci. Kyushu Univ. Ser. E 2, 215--235. Morisita, M. 1962. Id-index, measure dispersion individuals. Res. Popul. Ecol. 4, 1--7. Smith-Gill, S. J. 1975. Cytophysiological basis disruptive pigmentary patterns leopard frog, Rana pipiens. II. Wild type mutant cell specific patterns. J. Morphol. 146, 35--54. Hairston, N. G., Hill, R. Ritte, U. 1971. interpretation aggregation patterns. : Patil, G. P., Pileou, E. C. Waters, W. E. eds. Statistical Ecology 1: Spatial Patterns Statistical Distributions. Penn. State Univ. Press, University Park. Krebs, C. J. 1999. Ecological Methodology. 2nd ed. Benjamin Cummings Publishers.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"common error found several papers standardizing case (b), denominator given Muni - 1. results hiatus [0, 0.5] interval standardized index. root typo book Krebs (1999), see Errata book (Page 217, https://www.zoology.ubc.ca/~krebs/downloads/errors_2nd_printing.pdf).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"data(dune) x <- dispindmorisita(dune) x #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Chenalbu NaN Inf -Inf NaN NaN #> Cirsarve 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Empenigr 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 y <- dispindmorisita(dune, unique.rm = TRUE) y #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 dim(x) ## with unique species #> [1] 30 5 dim(y) ## unique species removed #> [1] 27 5"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":null,"dir":"Reference","previous_headings":"","what":"Dispersion-based weighting of species counts — dispweight","title":"Dispersion-based weighting of species counts — dispweight","text":"Transform abundance data downweighting species overdispersed Poisson error.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"dispweight(comm, groups, nsimul = 999, nullmodel = \"c0_ind\", plimit = 0.05) gdispweight(formula, data, plimit = 0.05) # S3 method for dispweight summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dispersion-based weighting of species counts — dispweight","text":"comm Community data matrix. groups Factor describing group structure. missing, sites regarded belonging one group. NA values allowed. nsimul Number simulations. nullmodel nullmodel used commsim within groups. default follows Clarke et al. (2006). plimit Downweight species \\(p\\)-value limit. formula, data Formula left-hand side community data frame right-hand side gives explanatory variables. explanatory variables found data frame given data parent frame. object Result object dispweight gdispweight. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dispersion-based weighting of species counts — dispweight","text":"dispersion index (\\(D\\)) calculated ratio variance expected value species. species abundances follow Poisson distribution, expected dispersion \\(E(D) = 1\\), \\(D > 1\\), species overdispersed. inverse \\(1/D\\) can used downweight species abundances. Species downweighted overdispersion judged statistically significant (Clarke et al. 2006). Function dispweight implements original procedure Clarke et al. (2006). one factor can used group sites find species means. significance overdispersion assessed freely distributing individuals species within factor levels. achieved using nullmodel \"c0_ind\" (accords Clarke et al. 2006), nullmodels can used, though may meaningful (see commsim alternatives). species absent factor level, whole level ignored calculation overdispersion, number degrees freedom can vary among species. reduced number degrees freedom used divisor overdispersion \\(D\\), species higher dispersion hence lower weights transformation. Function gdispweight generalized parametric version dispweight. function based glm quasipoisson error family. glm model can used, including several factors continuous covariates. Function gdispweight uses test statistic dispweight (Pearson Chi-square), ignore factor levels species absent, number degrees freedom equal species. Therefore transformation weights can higher dispweight. gdispweight function evaluates significance overdispersion parametrically Chi-square distribution (pchisq). Functions dispweight gdispweight transform data, add information overdispersion weights attributes result. summary can used extract print information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dispersion-based weighting of species counts — dispweight","text":"Function returns transformed data following new attributes: D Dispersion statistic. df Degrees freedom species. p \\(p\\)-value Dispersion statistic \\(D\\). weights weights applied community data. nsimul Number simulations used assess \\(p\\)-value, NA simulations performed. nullmodel name commsim null model, NA simulations performed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dispersion-based weighting of species counts — dispweight","text":"Clarke, K. R., M. G. Chapman, P. J. Somerfield, H. R. Needham. 2006. Dispersion-based weighting species counts assemblage analyses. Marine Ecology Progress Series, 320, 11–27.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dispersion-based weighting of species counts — dispweight","text":"Eduard Szöcs eduardszoesc@gmail.com wrote original dispweight, Jari Oksanen significantly modified code, provided support functions developed gdispweight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"data(mite, mite.env) ## dispweight and its summary mite.dw <- with(mite.env, dispweight(mite, Shrub, nsimul = 99)) ## IGNORE_RDIFF_BEGIN summary(mite.dw) #> Dispersion Weight Df Pr(Disp.) #> Brachy 9.6908 0.1031909 67 0.01 ** #> PHTH 3.2809 0.3047900 49 0.01 ** #> HPAV 6.5263 0.1532264 67 0.01 ** #> RARD 6.0477 0.1653525 49 0.01 ** #> SSTR 2.2619 0.4421053 49 0.01 ** #> Protopl 5.4229 0.1844031 49 0.01 ** #> MEGR 4.5354 0.2204860 67 0.01 ** #> MPRO 1.2687 0.7882353 67 0.04 * #> TVIE 2.5956 0.3852706 67 0.01 ** #> HMIN 10.0714 0.0992906 67 0.01 ** #> HMIN2 7.5674 0.1321466 49 0.01 ** #> NPRA 2.6743 0.3739344 67 0.01 ** #> TVEL 9.6295 0.1038474 49 0.01 ** #> ONOV 11.3628 0.0880064 67 0.01 ** #> SUCT 8.7372 0.1144533 67 0.01 ** #> LCIL 129.4436 0.0077254 67 0.01 ** #> Oribatl1 4.1250 0.2424248 67 0.01 ** #> Ceratoz1 1.7150 0.5830768 67 0.01 ** #> PWIL 2.2943 0.4358538 67 0.01 ** #> Galumna1 2.8777 0.3474943 49 0.01 ** #> Stgncrs2 3.8242 0.2614953 49 0.01 ** #> HRUF 1.7575 0.5690021 67 0.01 ** #> Trhypch1 14.9225 0.0670128 67 0.01 ** #> PPEL 1.3628 1.0000000 49 0.07 . #> NCOR 2.5875 0.3864771 67 0.01 ** #> SLAT 2.7857 0.3589744 49 0.01 ** #> FSET 4.8901 0.2044944 49 0.01 ** #> Lepidzts 1.6577 0.6032360 49 0.02 * #> Eupelops 1.4611 0.6844033 67 0.04 * #> Miniglmn 1.6505 0.6058733 49 0.01 ** #> LRUG 12.0658 0.0828790 67 0.01 ** #> PLAG2 3.2403 0.3086090 67 0.01 ** #> Ceratoz3 3.5125 0.2846947 67 0.01 ** #> Oppiminu 3.1680 0.3156525 67 0.01 ** #> Trimalc2 10.5927 0.0944046 67 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Based on 99 simulations on 'c0_ind' nullmodel ## IGNORE_RDIFF_END ## generalized dispersion weighting mite.dw <- gdispweight(mite ~ Shrub + WatrCont, data = mite.env) rda(mite.dw ~ Shrub + WatrCont, data = mite.env) #> Call: rda(formula = mite.dw ~ Shrub + WatrCont, data = mite.env) #> #> Inertia Proportion Rank #> Total 38.1640 1.0000 #> Constrained 9.2129 0.2414 3 #> Unconstrained 28.9511 0.7586 35 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 7.986 0.748 0.480 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 5.886 3.634 2.791 2.592 1.932 1.573 1.210 1.078 #> (Showing 8 of 35 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":null,"dir":"Reference","previous_headings":"","what":"Connectedness of Dissimilarities — distconnected","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected finds groups connected disregarding dissimilarities threshold NA. function can used find groups can ordinated together transformed stepacross. Function .shared returns logical dissimilarity object, TRUE means sites species common. minimal structure distconnected can used set missing values dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"distconnected(dis, toolong = 1, trace = TRUE) no.shared(x)"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Connectedness of Dissimilarities — distconnected","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. trace Summarize results distconnected x Community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Connectedness of Dissimilarities — distconnected","text":"Data sets disconnected sample plots groups sample plots share species sites groups sites. data sets sensibly ordinated unconstrained method subsets related . instance, correspondence analysis polarize subsets eigenvalue 1. Neither can dissimilarities transformed stepacross, path points, result contain NAs. Function distconnected find subsets dissimilarity matrices. function return grouping vector can used sub-setting data. data connected, result vector \\(1\\)s. connectedness two points can defined either threshold toolong using input dissimilarities NAs. Function .shared returns dist structure value TRUE two sites nothing common, value FALSE least one shared species. minimal structure can analysed distconnected. function can used select dissimilarities shared species indices fixed upper limit. Function distconnected uses depth-first search (Sedgewick 1990).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected returns vector observations using integers identify connected groups. data connected, values 1. Function .shared returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Connectedness of Dissimilarities — distconnected","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Connectedness of Dissimilarities — distconnected","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"## There are no disconnected data in vegan, and the following uses an ## extremely low threshold limit for connectedness. This is for ## illustration only, and not a recommended practice. data(dune) dis <- vegdist(dune) gr <- distconnected(dis, toolong=0.4) #> Connectivity of distance matrix with threshold dissimilarity 0.4 #> Data are disconnected: 6 groups #> Groups sizes #> 1 2 3 4 5 6 #> 1 11 2 4 1 1 # Make sites with no shared species as NA in Manhattan dissimilarities dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":null,"dir":"Reference","previous_headings":"","what":"Ecological Diversity Indices — diversity","title":"Ecological Diversity Indices — diversity","text":"Shannon, Simpson, Fisher diversity indices species richness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ecological Diversity Indices — diversity","text":"","code":"diversity(x, index = \"shannon\", groups, equalize.groups = FALSE, MARGIN = 1, base = exp(1)) simpson.unb(x, inverse = FALSE) fisher.alpha(x, MARGIN = 1, ...) specnumber(x, groups, MARGIN = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ecological Diversity Indices — diversity","text":"x Community data, matrix-like object vector. index Diversity index, one \"shannon\", \"simpson\" \"invsimpson\". MARGIN Margin index computed. base logarithm base used shannon. inverse Use inverse Simpson similarly diversity(x, \"invsimpson\"). groups grouping factor: given, finds diversity communities pooled groups. equalize.groups Instead observed abundances, standardize communities unit total. ... Parameters passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ecological Diversity Indices — diversity","text":"Shannon Shannon--Weaver (Shannon--Wiener) index defined \\(H' = -\\sum_i p_i \\log_{b} p_i\\), \\(p_i\\) proportional abundance species \\(\\) \\(b\\) base logarithm. popular use natural logarithms, argue base \\(b = 2\\) (makes sense, real difference). variants Simpson's index based \\(D = \\sum p_i^2\\). Choice simpson returns \\(1-D\\) invsimpson returns \\(1/D\\). simpson.unb finds unbiased Simpson indices discrete samples (Hurlbert 1971, eq. 5). less sensitive sample size basic Simpson indices. unbiased indices can calculated data integer counts. diversity function can find total (gamma) diversity pooled communities argument groups. average alpha diversity can found mean diversities groups, difference ratio estimate beta diversity (see Examples). pooling can based either observed abundancies, communities can equalized unit total pooling; see Jost (2007) discussion. Functions adipart multipart provide canned alternatives estimating alpha, beta gamma diversities hierarchical settings. fisher.alpha estimates \\(\\alpha\\) parameter Fisher's logarithmic series (see fisherfit). estimation possible genuine counts individuals. None diversity indices usable empty sampling units without species, indices can give numeric value. Filtering cases left user. Function specnumber finds number species. MARGIN = 2, finds frequencies species. groups given, finds total number species group (see example finding one kind beta diversity option). Better stories can told Simpson's index Shannon's index, still grander narratives rarefaction (Hurlbert 1971). However, indices closely related (Hill 1973), reason despise one others (graduate student, drag , obey Professor's orders). particular, exponent Shannon index linearly related inverse Simpson (Hill 1973) although former may sensitive rare species. Moreover, inverse Simpson asymptotically equal rarefied species richness sample two individuals, Fisher's \\(\\alpha\\) similar inverse Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ecological Diversity Indices — diversity","text":"vector diversity indices numbers species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ecological Diversity Indices — diversity","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12, 42--58. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ecological Diversity Indices — diversity","text":"Jari Oksanen Bob O'Hara (fisher.alpha).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ecological Diversity Indices — diversity","text":"","code":"data(BCI, BCI.env) H <- diversity(BCI) simp <- diversity(BCI, \"simpson\") invsimp <- diversity(BCI, \"inv\") ## Unbiased Simpson unbias.simp <- simpson.unb(BCI) ## Fisher alpha alpha <- fisher.alpha(BCI) ## Plot all pairs(cbind(H, simp, invsimp, unbias.simp, alpha), pch=\"+\", col=\"blue\") ## Species richness (S) and Pielou's evenness (J): S <- specnumber(BCI) ## rowSums(BCI > 0) does the same... J <- H/log(S) ## beta diversity defined as gamma/alpha - 1: ## alpha is the average no. of species in a group, and gamma is the ## total number of species in the group (alpha <- with(BCI.env, tapply(specnumber(BCI), Habitat, mean))) #> OldHigh OldLow OldSlope Swamp Young #> 85.75000 91.76923 91.58333 94.00000 90.00000 (gamma <- with(BCI.env, specnumber(BCI, Habitat))) #> OldHigh OldLow OldSlope Swamp Young #> 158 210 183 128 117 gamma/alpha - 1 #> OldHigh OldLow OldSlope Swamp Young #> 0.8425656 1.2883487 0.9981802 0.3617021 0.3000000 ## similar calculations with Shannon diversity (alpha <- with(BCI.env, tapply(diversity(BCI), Habitat, mean))) # average #> OldHigh OldLow OldSlope Swamp Young #> 3.638598 3.876413 3.887122 4.003780 3.246729 (gamma <- with(BCI.env, diversity(BCI, groups=Habitat))) # pooled #> OldHigh OldLow OldSlope Swamp Young #> 3.873186 4.284972 4.212098 4.164335 3.387536 ## additive beta diversity based on Shannon index gamma-alpha #> OldHigh OldLow OldSlope Swamp Young #> 0.2345878 0.4085595 0.3249760 0.1605548 0.1408068"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and Environment in Dutch Dune Meadows. — dune","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune meadow vegetation data, dune, cover class values 30 species 20 sites. corresponding environmental data frame dune.env following entries:","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune data frame observations 30 species 20 sites. species names abbreviated 4+4 letters (see make.cepnames). following names changed original source (Jongman et al. 1987): Leontodon autumnalis Scorzoneroides, Potentilla palustris Comarum. dune.env data frame 20 observations following 5 variables: A1: numeric vector thickness soil A1 horizon. Moisture: ordered factor levels: 1 < 2 < 4 < 5. Management: factor levels: BF (Biological farming), HF (Hobby farming), NM (Nature Conservation Management), SF (Standard Farming). Use: ordered factor land-use levels: Hayfield < Haypastu < Pasture. Manure: ordered factor levels: 0 < 1 < 2 < 3 < 4.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"Jongman, R.H.G, ter Braak, C.J.F & van Tongeren, O.F.R. (1987). Data Analysis Community Landscape Ecology. Pudoc, Wageningen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":null,"dir":"Reference","previous_headings":"","what":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"Classification table species dune data set.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"dune.taxon data frame 30 species (rows) classified five taxonomic levels (columns). dune.phylodis dist object estimated coalescence ages extracted doi:10.5061/dryad.63q27 (Zanne et al. 2014) using tools packages ape phylobase.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"families orders based APG IV (2016) vascular plants Hill et al. (2006) mosses. higher levels (superorder subclass) based Chase & Reveal (2009). Chase & Reveal (2009) treat Angiosperms mosses subclasses class Equisetopsida (land plants), brylogists traditionally used much inflated levels adjusted match Angiosperm classification.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"APG IV [Angiosperm Phylogeny Group] (2016) update Angiosperm Phylogeny Group classification orders families flowering plants: APG IV. Bot. J. Linnean Soc. 181: 1--20. Chase, M.W. & Reveal, J. L. (2009) phylogenetic classification land plants accompany APG III. Bot. J. Linnean Soc. 161: 122--127. Hill, M.O et al. (2006) annotated checklist mosses Europe Macaronesia. J. Bryology 28: 198--267. Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506: 89--92.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Eigenvalues from an Ordination Object — eigenvals","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Function extracts eigenvalues object . Many multivariate methods return objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"eigenvals(x, ...) # S3 method for cca eigenvals(x, model = c(\"all\", \"unconstrained\", \"constrained\"), constrained = NULL, ...) # S3 method for decorana eigenvals(x, kind = c(\"additive\", \"axiswise\", \"decorana\"), ...) # S3 method for eigenvals summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"x object extract eigenvalues. object eigenvals result object. model eigenvalues return objects inherit class \"cca\" . constrained Return constrained eigenvalues. Deprecated vegan 2.5-0. Use model instead. kind Kind eigenvalues returned decorana. \"additive\" eigenvalues can used reporting importances components summary. \"axiswise\" gives non-additive eigenvalues, \"decorana\" decorana values (see decorana details). ... arguments functions (usually ignored)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"generic function methods cca, wcmdscale, pcnm, prcomp, princomp, dudi (ade4), pca pco ( labdsv) result objects. default method also extracts eigenvalues result looks like eigen svd. Functions prcomp princomp contain square roots eigenvalues called standard deviations, eigenvals function returns squares. Function svd contains singular values, function eigenvals returns squares. constrained ordination methods cca, rda capscale function returns constrained unconstrained eigenvalues concatenated one vector, partial component ignored. However, argument constrained = TRUE constrained eigenvalues returned. summary eigenvals result returns eigenvalues, proportion explained cumulative proportion explained. result object can negative eigenvalues (wcmdscale, dbrda, pcnm) correspond imaginary axes Euclidean mapping non-Euclidean distances (Gower 1985). case real axes (corresponding positive eigenvalues) \"explain\" proportion >1 total variation, negative eigenvalues bring cumulative proportion 1. capscale find positive eigenvalues used finding proportions. decorana importances cumulative proportions reported kind = \"additive\", alternatives add total inertia input data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"object class \"eigenvals\", vector eigenvalues. summary method returns object class \"summary.eigenvals\", matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Gower, J. C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ev <- eigenvals(mod) ev #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 CA5 #> 0.3615566 0.1699600 0.1126167 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 #> CA6 CA7 CA8 CA9 CA10 CA11 CA12 CA13 #> 0.1002670 0.0772991 0.0536938 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 #> CA14 CA15 CA16 CA17 CA18 CA19 CA20 #> 0.0101910 0.0094701 0.0055090 0.0030529 0.0025118 0.0019485 0.0005178 summary(ev) #> Importance of components: #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 #> Eigenvalue 0.3616 0.16996 0.11262 0.3500 0.2201 0.18507 0.15512 #> Proportion Explained 0.1736 0.08159 0.05406 0.1680 0.1056 0.08884 0.07446 #> Cumulative Proportion 0.1736 0.25514 0.30920 0.4772 0.5829 0.67172 0.74618 #> CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> Eigenvalue 0.13511 0.10027 0.07730 0.05369 0.03656 0.03509 0.02823 #> Proportion Explained 0.06485 0.04813 0.03711 0.02577 0.01755 0.01684 0.01355 #> Cumulative Proportion 0.81104 0.85917 0.89627 0.92205 0.93960 0.95644 0.96999 #> CA12 CA13 CA14 CA15 CA16 CA17 #> Eigenvalue 0.017065 0.012247 0.010191 0.009470 0.005509 0.003053 #> Proportion Explained 0.008192 0.005879 0.004892 0.004546 0.002644 0.001465 #> Cumulative Proportion 0.978183 0.984062 0.988954 0.993500 0.996145 0.997610 #> CA18 CA19 CA20 #> Eigenvalue 0.002512 0.0019485 0.0005178 #> Proportion Explained 0.001206 0.0009353 0.0002486 #> Cumulative Proportion 0.998816 0.9997514 1.0000000 ## choose which eignevalues to return eigenvals(mod, model = \"unconstrained\") #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 0.1002670 0.0772991 0.0536938 #> CA9 CA10 CA11 CA12 CA13 CA14 CA15 CA16 #> 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 0.0101910 0.0094701 0.0055090 #> CA17 CA18 CA19 CA20 #> 0.0030529 0.0025118 0.0019485 0.0005178"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fits an Environmental Vector or Factor onto an Ordination — envfit","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"function fits environmental vectors factors onto ordination. projections points onto vectors maximum correlation corresponding environmental variables, factors show averages factor levels. continuous varaibles equal fitting linear trend surface (plane 2D) variable (see ordisurf); trend surface can presented showing gradient (direction steepest increase) using arrow. environmental variables dependent variables explained ordination scores, dependent variable analysed separately.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"# S3 method for default envfit(ord, env, permutations = 999, strata = NULL, choices=c(1,2), display = \"sites\", w = weights(ord, display), na.rm = FALSE, ...) # S3 method for formula envfit(formula, data, ...) # S3 method for envfit plot(x, choices = c(1,2), labels, arrow.mul, at = c(0,0), axis = FALSE, p.max = NULL, col = \"blue\", bg, add = TRUE, ...) # S3 method for envfit scores(x, display, choices, arrow.mul=1, tidy = FALSE, ...) vectorfit(X, P, permutations = 0, strata = NULL, w, ...) factorfit(X, P, permutations = 0, strata = NULL, w, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"ord ordination object structure ordination scores can extracted (including data frame matrix scores). env Data frame, matrix vector environmental variables. variables can mixed type (factors, continuous variables) data frames. X Matrix data frame ordination scores. P Data frame, matrix vector environmental variable(s). must continuous vectorfit factors characters factorfit. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Set permutations = 0 skip permutations. formula, data Model formula data. na.rm Remove points missing values ordination scores environmental variables. operation casewise: whole row data removed missing value na.rm = TRUE. x result object envfit. ordiArrowMul ordiArrowTextXY must two-column matrix ( matrix-like object) containing coordinates arrow heads two plot axes, methods extract structure envfit results. choices Axes plotted. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable scores (\"vectors\" \"factors\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Change plotting labels. argument list elements vectors factors give new plotting labels. either elements omitted, default labels used. one type elements (vectors factors), labels can given vector. default labels can displayed labels command. arrow.mul Multiplier vector lengths. arrows automatically scaled similarly plot.cca given plot add = TRUE. However, scores can used adjust arrow lengths plot function used. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. axis Plot axis showing scaling fitted arrows. p.max Maximum estimated \\(P\\) value displayed variables. must calculate \\(P\\) values setting permutations use option. col Colour plotting. bg Background colour labels. bg set, labels displayed ordilabel instead text. See Examples using semitransparent background. add Results added existing ordination plot. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. display fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). scores function either \"vectors\" \"factors\" (synonyms \"bp\" \"cn\", resp.). w Weights used fitting (concerns mainly cca decorana results nonconstant weights). ... Parameters passed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Function envfit finds vectors factor averages environmental variables. Function plot.envfit adds ordination diagram. X data.frame, envfit uses factorfit factor variables vectorfit variables. X matrix vector, envfit uses vectorfit. Alternatively, model can defined simplified model formula, left hand side must ordination result object matrix ordination scores, right hand side lists environmental variables. formula interface can used easier selection /transformation environmental variables. main effects analysed even interaction terms defined formula. ordination results extracted scores extra arguments passed scores. fitted models apply results defined extracting scores using envfit. instance, scaling constrained ordination (see scores.rda, scores.cca) must set way envfit plot ordination results (see Examples). printed output continuous variables (vectors) gives direction cosines coordinates heads unit length vectors. plot scaled correlation (square root column r2) “weak” predictors shorter arrows “strong” predictors. can see scaled relative lengths using command scores. plotted (scaled) arrows adjusted current graph using constant multiplier: keep relative r2-scaled lengths arrows tries fill current plot. can see multiplier using ordiArrowMul(result_of_envfit), set argument arrow.mul. Functions vectorfit factorfit can called directly. Function vectorfit finds directions ordination space towards environmental vectors change rapidly maximal correlations ordination configuration. Function factorfit finds averages ordination scores factor levels. Function factorfit treats ordered unordered factors similarly. permutations \\(> 0\\), significance fitted vectors factors assessed using permutation environmental variables. goodness fit statistic squared correlation coefficient (\\(r^2\\)). factors defined \\(r^2 = 1 - ss_w/ss_t\\), \\(ss_w\\) \\(ss_t\\) within-group total sums squares. See permutations additional details permutation tests Vegan. User can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. fitted vectors similar biplot arrows constrained ordination fitted LC scores (display = \"lc\") set scaling = \"species\" (see scores.cca). weighted fitting gives similar results biplot arrows class centroids cca. lengths arrows fitted vectors automatically adjusted physical size plot, arrow lengths compared across plots. similar scaling arrows, must explicitly set arrow.mul argument plot command; see ordiArrowMul ordiArrowTextXY. results can accessed scores.envfit function returns either fitted vectors scaled correlation coefficient centroids fitted environmental variables, named list .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Functions vectorfit factorfit return lists classes vectorfit factorfit print method. result object following items: arrows Arrow endpoints vectorfit. arrows scaled unit length. centroids Class centroids factorfit. r Goodness fit statistic: Squared correlation coefficient permutations Number permutations. control list control values permutations returned function . pvals Empirical P-values variable. Function envfit returns list class envfit results vectorfit envfit items. Function plot.envfit scales vectors correlation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Jari Oksanen. permutation test derives code suggested Michael Scroggie.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Fitted vectors become method choice displaying environmental variables ordination. Indeed, optimal way presenting environmental variables Constrained Correspondence Analysis cca, since linear constraints. unconstrained ordination relation external variables ordination configuration may less linear, therefore methods arrows may useful. simplest adjust plotting symbol sizes (cex, symbols) environmental variables. Fancier methods involve smoothing regression methods abound R, ordisurf provides wrapper .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"data(varespec, varechem) library(MASS) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2276827 #> Run 2 stress 0.2204286 #> Run 3 stress 0.1967393 #> Run 4 stress 0.2066032 #> Run 5 stress 0.2234314 #> Run 6 stress 0.1869637 #> Run 7 stress 0.18584 #> Run 8 stress 0.18458 #> ... Procrustes: rmse 0.04935318 max resid 0.1575045 #> Run 9 stress 0.2178486 #> Run 10 stress 0.195049 #> Run 11 stress 0.2300388 #> Run 12 stress 0.1948413 #> Run 13 stress 0.1955836 #> Run 14 stress 0.2390981 #> Run 15 stress 0.2229031 #> Run 16 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04162605 max resid 0.1518104 #> Run 17 stress 0.2390411 #> Run 18 stress 0.1825658 #> ... Procrustes: rmse 1.358437e-05 max resid 2.641802e-05 #> ... Similar to previous best #> Run 19 stress 0.2092456 #> Run 20 stress 0.1982376 #> *** Best solution repeated 1 times (fit <- envfit(ord, varechem, perm = 999)) #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05732 -0.99836 0.2536 0.046 * #> P 0.61973 0.78481 0.1938 0.093 . #> K 0.76646 0.64229 0.1809 0.121 #> Ca 0.68520 0.72835 0.4119 0.008 ** #> Mg 0.63253 0.77454 0.4270 0.004 ** #> S 0.19140 0.98151 0.1752 0.123 #> Al -0.87159 0.49023 0.5269 0.001 *** #> Fe -0.93602 0.35195 0.4451 0.002 ** #> Mn 0.79871 -0.60172 0.5231 0.002 ** #> Zn 0.61756 0.78652 0.1879 0.107 #> Mo -0.90306 0.42951 0.0609 0.539 #> Baresoil 0.92488 -0.38025 0.2508 0.041 * #> Humdepth 0.93283 -0.36032 0.5201 0.002 ** #> pH -0.64798 0.76165 0.2308 0.070 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> scores(fit, \"vectors\") #> NMDS1 NMDS2 #> N -0.02886657 -0.5027884 #> P 0.27284738 0.3455254 #> K 0.32602389 0.2732093 #> Ca 0.43973450 0.4674260 #> Mg 0.41334179 0.5061403 #> S 0.08011287 0.4108193 #> Al -0.63269483 0.3558584 #> Fe -0.62444176 0.2347947 #> Mn 0.57765881 -0.4351900 #> Zn 0.26771371 0.3409563 #> Mo -0.22293623 0.1060315 #> Baresoil 0.46316234 -0.1904201 #> Humdepth 0.67271099 -0.2598473 #> pH -0.31130599 0.3659163 plot(ord) plot(fit) plot(fit, p.max = 0.05, col = \"red\") ## Adding fitted arrows to CCA. We use \"lc\" scores, and hope ## that arrows are scaled similarly in cca and envfit plots ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, type=\"p\") fit <- envfit(ord, varechem, perm = 999, display = \"lc\") plot(fit, p.max = 0.05, col = \"red\") ## 'scaling' must be set similarly in envfit and in ordination plot plot(ord, type = \"p\", scaling = \"sites\") fit <- envfit(ord, varechem, perm = 0, display = \"lc\", scaling = \"sites\") plot(fit, col = \"red\") ## Class variables, formula interface, and displaying the ## inter-class variability with ordispider, and semitransparent ## white background for labels (semitransparent colours are not ## supported by all graphics devices) data(dune) data(dune.env) ord <- cca(dune) fit <- envfit(ord ~ Moisture + A1, dune.env, perm = 0) plot(ord, type = \"n\") with(dune.env, ordispider(ord, Moisture, col=\"skyblue\")) with(dune.env, points(ord, display = \"sites\", col = as.numeric(Moisture), pch=16)) plot(fit, cex=1.2, axis=TRUE, bg = rgb(1, 1, 1, 0.5)) ## Use shorter labels for factor centroids labels(fit) #> $vectors #> [1] \"A1\" #> #> $factors #> [1] \"Moisture1\" \"Moisture2\" \"Moisture4\" \"Moisture5\" #> plot(ord) plot(fit, labels=list(factors = paste(\"M\", c(1,2,4,5), sep = \"\")), bg = rgb(1,1,0,0.5))"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds minimum (\\(q^*\\)) evenness profile based Tsallis entropy. scale factor entropy represents specific weighting species relative frequencies leads minimum evenness community (Mendes et al. 2008).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"eventstar(x, qmax = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"x community matrix numeric vector. qmax Maximum scale parameter Tsallis entropy used finding minimum Tsallis based evenness range c(0, qmax).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds characteristic value scale parameter \\(q\\) Tsallis entropy corresponding minimum evenness (equitability) profile based Tsallis entropy. value proposed Mendes et al. (2008) \\(q^*\\). \\(q^\\ast\\) index represents scale parameter one parameter Tsallis diversity family leads greatest deviation maximum equitability given relative abundance vector community. value \\(q^\\ast\\) found identifying minimum evenness profile scaling factor \\(q\\) one-dimensional minimization. evenness profile known convex function, guaranteed underlying optimize function find unique solution range c(0, qmax). scale parameter value \\(q^\\ast\\) used find corresponding values diversity (\\(H_{q^\\ast}\\)), evenness (\\(H_{q^\\ast}(\\max)\\)), numbers equivalent (\\(D_{q^\\ast}\\)). calculation details, see tsallis Examples . Mendes et al. (2008) advocated use \\(q^\\ast\\) corresponding diversity, evenness, Hill numbers, unique value representing diversity profile, positively associated rare species community, thus potentially useful indicator certain relative abundance distributions communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"data frame columns: qstar scale parameter value \\(q\\ast\\) corresponding minimum value Tsallis based evenness profile. Estar Value evenness based normalized Tsallis entropy \\(q^\\ast\\). Hstar Value Tsallis entropy \\(q^\\ast\\). Dstar Value Tsallis entropy \\(q^\\ast\\) converted numbers equivalents (also called Hill numbers, effective number species, ‘true’ diversity; cf. Jost 2007). See tsallis calculation details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Mendes, R.S., Evangelista, L.R., Thomaz, S.M., Agostinho, .. Gomes, L.C. (2008) unified index measure ecological diversity species rarity. Ecography 31, 450--456. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439. Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Values \\(q^\\ast\\) found Mendes et al. (2008) ranged 0.56 1.12 presenting low variability, interval 0 5 safely encompass possibly expected \\(q^\\ast\\) values practice, profiling evenness changing value qmax argument advised output values near range limits found.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Eduardo Ribeiro Cunha edurcunha@gmail.com Heloisa Beatriz Antoniazi Evangelista helobeatriz@gmail.com, technical input Péter Sólymos.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"data(BCI) (x <- eventstar(BCI[1:5,])) #> qstar Estar Hstar Dstar #> 1 0.6146389 0.4263687 10.524584 67.03551 #> 2 0.6249249 0.4080263 9.534034 57.66840 #> 3 0.6380858 0.4062032 9.225458 57.69174 #> 4 0.6245808 0.4062213 10.140189 65.50247 #> 5 0.6404825 0.4219957 9.828138 66.96440 ## profiling y <- as.numeric(BCI[10,]) (z <- eventstar(y)) #> qstar Estar Hstar Dstar #> 1 0.6372529 0.4117557 9.546332 61.77715 q <- seq(0, 2, 0.05) Eprof <- tsallis(y, scales=q, norm=TRUE) Hprof <- tsallis(y, scales=q) Dprof <- tsallis(y, scales=q, hill=TRUE) opar <- par(mfrow=c(3,1)) plot(q, Eprof, type=\"l\", main=\"Evenness\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, norm=TRUE), col=2) plot(q, Hprof, type=\"l\", main=\"Diversity\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar), col=2) plot(q, Dprof, type=\"l\", main=\"Effective number of species\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, hill=TRUE), col=2) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Function fisherfit fits Fisher's logseries abundance data. Function prestonfit groups species frequencies doubling octave classes fits Preston's lognormal model, function prestondistr fits truncated lognormal model without pooling data octaves.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"fisherfit(x, ...) prestonfit(x, tiesplit = TRUE, ...) prestondistr(x, truncate = -1, ...) # S3 method for prestonfit plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", line.col = \"red\", lwd = 2, ...) # S3 method for prestonfit lines(x, line.col = \"red\", lwd = 2, ...) veiledspec(x, ...) as.fisher(x, ...) # S3 method for fisher plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", kind = c(\"bar\", \"hiplot\", \"points\", \"lines\"), add = FALSE, ...) as.preston(x, tiesplit = TRUE, ...) # S3 method for preston plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", ...) # S3 method for preston lines(x, xadjust = 0.5, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"x Community data vector fitting functions result object plot functions. tiesplit Split frequencies \\(1, 2, 4, 8\\) etc adjacent octaves. truncate Truncation point log-Normal model, log2 units. Default value \\(-1\\) corresponds left border zero Octave. choice strongly influences fitting results. xlab, ylab Labels x y axes. bar.col Colour data bars. line.col Colour fitted line. lwd Width fitted line. kind Kind plot drawn: \"bar\" similar bar plot plot.fisherfit, \"hiplot\" draws vertical lines plot(..., type=\"h\"), \"points\" \"lines\" obvious. add Add existing plot. xadjust Adjustment horizontal positions octaves. ... parameters passed functions. Ignored prestonfit tiesplit passed .preston prestondistr.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher's logarithmic series expected number species \\(f\\) \\(n\\) observed individuals \\(f_n = \\alpha x^n / n\\) (Fisher et al. 1943). estimation possible genuine counts individuals. parameter \\(\\alpha\\) used diversity index can estimated separate function fisher.alpha. parameter \\(x\\) taken nuisance parameter estimated separately taken \\(n/(n+\\alpha)\\). Helper function .fisher transforms abundance data Fisher frequency table. Diversity given NA communities one (zero) species: reliable way estimating diversity, even equations return bogus numeric value cases. Preston (1948) satisfied Fisher's model seemed imply infinite species richness, postulated rare species diminishing class species middle frequency scale. achieved collapsing higher frequency classes wider wider “octaves” doubling class limits: 1, 2, 3--4, 5--8, 9--16 etc. occurrences. seems Preston regarded frequencies 1, 2, 4, etc.. “tied” octaves (Williamson & Gaston 2005). means half species frequency 1 shown lowest octave, rest transferred second octave. Half species second octave transferred higher one well, usually large number species. practise makes data look lognormal reducing usually high lowest octaves. can achieved setting argument tiesplit = TRUE. tiesplit = FALSE frequencies split, ones lowest octave, twos second, etc. Williamson & Gaston (2005) discuss alternative definitions detail, consulted critical review log-Normal model. logseries data look like lognormal plotted Preston's way. expected frequency \\(f\\) abundance octave \\(o\\) defined \\(f_o = S_0 \\exp(-(\\log_2(o) - \\mu)^2/2/\\sigma^2)\\), \\(\\mu\\) location mode \\(\\sigma\\) width, \\(\\log_2\\) scale, \\(S_0\\) expected number species mode. lognormal model usually truncated left rare species observed. Function prestonfit fits truncated lognormal model second degree log-polynomial octave pooled data using Poisson ( tiesplit = FALSE) quasi-Poisson (tiesplit = TRUE) error. Function prestondistr fits left-truncated Normal distribution \\(\\log_2\\) transformed non-pooled observations direct maximization log-likelihood. Function prestondistr modelled function fitdistr can used alternative distribution models. functions common print, plot lines methods. lines function adds fitted curve octave range line segments showing location mode width (sd) response. Function .preston transforms abundance data octaves. Argument tiesplit influence fit prestondistr, influence barplot octaves. total extrapolated richness fitted Preston model can found function veiledspec. function accepts results prestonfit prestondistr. veiledspec called species count vector, internally use prestonfit. Function specpool provides alternative ways estimating number unseen species. fact, Preston's lognormal model seems truncated ends, may main reason result differ lognormal models fitted Rank--Abundance diagrams functions rad.lognormal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"function prestonfit returns object fitted coefficients, observed (freq) fitted (fitted) frequencies, string describing fitting method. Function prestondistr omits entry fitted. function fisherfit returns result nlm, item estimate \\(\\alpha\\). result object amended nuisance parameter item fisher observed data .fisher","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12: 42--58. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Williamson, M. & Gaston, K.J. (2005). lognormal distribution appropriate null hypothesis species--abundance distribution. Journal Animal Ecology 74, 409--422.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Bob O'Hara Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"data(BCI) mod <- fisherfit(BCI[5,]) mod #> #> Fisher log series model #> No. of species: 101 #> Fisher alpha: 37.96423 #> # prestonfit seems to need large samples mod.oct <- prestonfit(colSums(BCI)) mod.ll <- prestondistr(colSums(BCI)) mod.oct #> #> Preston lognormal model #> Method: Quasi-Poisson fit to octaves #> No. of species: 225 #> #> mode width S0 #> 4.885798 2.932690 32.022923 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.500000 16.00000 18.00000 19.000 30.00000 35.00000 31.00000 26.50000 #> Fitted 7.994154 13.31175 19.73342 26.042 30.59502 31.99865 29.79321 24.69491 #> 8 9 10 11 #> Observed 18.00000 13.00000 7.000000 2.0000 #> Fitted 18.22226 11.97021 7.000122 3.6443 #> mod.ll #> #> Preston lognormal model #> Method: maximized likelihood to log2 abundances #> No. of species: 225 #> #> mode width S0 #> 4.365002 2.753531 33.458185 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.50000 16.00000 18.00000 19.00000 30.00000 35.00000 31.00000 26.50000 #> Fitted 9.52392 15.85637 23.13724 29.58961 33.16552 32.58022 28.05054 21.16645 #> 8 9 10 11 #> Observed 18.00000 13.000000 7.000000 2.00000 #> Fitted 13.99829 8.113746 4.121808 1.83516 #> plot(mod.oct) lines(mod.ll, line.col=\"blue3\") # Different ## Smoothed density den <- density(log2(colSums(BCI))) lines(den$x, ncol(BCI)*den$y, lwd=2) # Fairly similar to mod.oct ## Extrapolated richness veiledspec(mod.oct) #> Extrapolated Observed Veiled #> 235.40577 225.00000 10.40577 veiledspec(mod.ll) #> Extrapolated Observed Veiled #> 230.931018 225.000000 5.931018"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Functions goodness inertcomp can used assess goodness fit individual sites species. Function vif.cca alias.cca can used analyse linear dependencies among constraints conditions. addition, diagnostic tools (see 'Details').","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"# S3 method for cca goodness(object, choices, display = c(\"species\", \"sites\"), model = c(\"CCA\", \"CA\"), summarize = FALSE, addprevious = FALSE, ...) inertcomp(object, display = c(\"species\", \"sites\"), unity = FALSE, proportional = FALSE) spenvcor(object) intersetcor(object) vif.cca(object) # S3 method for cca alias(object, names.only = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"object result object cca, rda, dbrda capscale. display Display \"species\" \"sites\". Species available dbrda capscale. choices Axes shown. Default show axes \"model\". model Show constrained (\"CCA\") unconstrained (\"CA\") results. summarize Show accumulated total. addprevious Add variation explained previous components statistic=\"explained\". model = \"CCA\" add conditioned (partialled ) variation, model = \"CA\" add conditioned constrained variation. give cumulative explanation previous components. unity Scale inertia components unit sum (sum items 1). proportional Give inertia components proportional corresponding total item (sum row 1). option takes precedence unity. names.Return names aliased variable(s) instead defining equations. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Function goodness gives cumulative proportion inertia accounted species chosen axes. proportions can assessed either species sites depending argument display, species available distance-based dbrda. function implemented capscale. Function inertcomp decomposes inertia partial, constrained unconstrained components site species. Legendre & De Cáceres (2012) called inertia components local contributions beta-diversity (LCBD) species contributions beta-diversity (SCBD), give relative contributions summing unity (argument unity = TRUE). interpretation, appropriate dissimilarity measures used dbrda appropriate standardization rda (Legendre & De Cáceres 2012). function implemented capscale. Function spenvcor finds -called “species -- environment correlation” (weighted) correlation weighted average scores linear combination scores. bad measure goodness ordination, sensitive extreme scores (like correlations ), sensitive overfitting using many constraints. Better models often poorer correlations. Function ordispider can show graphically. Function intersetcor finds -called “interset correlation” (weighted) correlation weighted averages scores constraints. defined contrasts used factor variables. bad measure since correlation. , focuses correlations single contrasts single axes instead looking multivariate relationship. Fitted vectors (envfit) provide better alternative. Biplot scores (see scores.cca) multivariate alternative (weighted) correlation linear combination scores constraints. Function vif.cca gives variance inflation factors constraint contrast factor constraints. partial ordination, conditioning variables analysed together constraints. Variance inflation diagnostic tool identify useless constraints. common rule values 10 indicate redundant constraints. later constraints complete linear combinations conditions previous constraints, completely removed estimation, biplot scores centroids calculated aliased constraints. note printed default output aliased constraints. Function alias give linear coefficients defining aliased constraints, names argument names.= TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"functions return matrices vectors appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London. Gross, J. (2003). Variance inflation factors. R News 3(1), 13--15. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Jari Oksanen. vif.cca relies heavily code W. N. Venables. alias.cca simplified version alias.lm.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) goodness(mod, addprevious = TRUE) #> CCA1 CCA2 CCA3 CCA4 #> Achimill 0.36630013 0.3822685 0.3838616 0.4934158 #> Agrostol 0.67247051 0.6724758 0.6779597 0.7773267 #> Airaprae 0.36213737 0.3698100 0.3816619 0.3908018 #> Alopgeni 0.61547145 0.6966105 0.7042650 0.7212918 #> Anthodor 0.24619147 0.2795001 0.3509172 0.3609709 #> Bellpere 0.41185412 0.4179432 0.4847618 0.4849622 #> Bromhord 0.33487622 0.3397416 0.3870032 0.5505037 #> Chenalbu 0.23594716 0.2684323 0.2828928 0.2885321 #> Cirsarve 0.29041563 0.3013655 0.3080671 0.3591280 #> Comapalu 0.16338257 0.6836790 0.7390659 0.7963425 #> Eleopalu 0.55132024 0.6099415 0.6193301 0.6259818 #> Elymrepe 0.25239595 0.2710266 0.2761491 0.2882666 #> Empenigr 0.27089495 0.3132399 0.3153052 0.3154203 #> Hyporadi 0.31349648 0.3371809 0.3387669 0.3388716 #> Juncarti 0.43923609 0.4492937 0.4871043 0.5224072 #> Juncbufo 0.70439967 0.7226263 0.7228786 0.7257471 #> Lolipere 0.48141171 0.5720410 0.5727299 0.6034007 #> Planlanc 0.54969676 0.6084389 0.6802195 0.6826265 #> Poaprat 0.40267189 0.4944813 0.5014516 0.5326546 #> Poatriv 0.49694972 0.5409439 0.5468830 0.5594817 #> Ranuflam 0.68677962 0.6983001 0.7020461 0.7064850 #> Rumeacet 0.44788204 0.5211145 0.7673956 0.7691199 #> Sagiproc 0.27039747 0.3497634 0.3553109 0.3613746 #> Salirepe 0.64788354 0.7264891 0.7276110 0.7639711 #> Scorautu 0.54312496 0.5510319 0.6078931 0.6140593 #> Trifprat 0.37328840 0.4101104 0.6624199 0.6625703 #> Trifrepe 0.03048149 0.2115857 0.3300132 0.4207437 #> Vicilath 0.17824132 0.1784611 0.3762406 0.4279428 #> Bracruta 0.15585567 0.1641095 0.1672797 0.2449864 #> Callcusp 0.30771429 0.3143582 0.3308502 0.3518027 goodness(mod, addprevious = TRUE, summ = TRUE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 0.4934158 0.7773267 0.3908018 0.7212918 0.3609709 0.4849622 0.5505037 0.2885321 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 0.3591280 0.7963425 0.6259818 0.2882666 0.3154203 0.3388716 0.5224072 0.7257471 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 0.6034007 0.6826265 0.5326546 0.5594817 0.7064850 0.7691199 0.3613746 0.7639711 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 0.6140593 0.6625703 0.4207437 0.4279428 0.2449864 0.3518027 # Inertia components inertcomp(mod, prop = TRUE) #> pCCA CCA CA #> Achimill 0.34271900 0.15069678 0.5065842 #> Agrostol 0.55602406 0.22130269 0.2226733 #> Airaprae 0.06404726 0.32675457 0.6091982 #> Alopgeni 0.34238968 0.37890210 0.2787082 #> Anthodor 0.10259139 0.25837947 0.6390291 #> Bellpere 0.40972447 0.07523776 0.5150378 #> Bromhord 0.33046684 0.22003683 0.4494963 #> Chenalbu 0.11064346 0.17788865 0.7114679 #> Cirsarve 0.26649913 0.09262886 0.6408720 #> Comapalu 0.16096277 0.63537969 0.2036575 #> Eleopalu 0.53954819 0.08643366 0.3740182 #> Elymrepe 0.22234322 0.06592337 0.7117334 #> Empenigr 0.10361994 0.21180040 0.6845797 #> Hyporadi 0.03889627 0.29997533 0.6611284 #> Juncarti 0.43439190 0.08801527 0.4775928 #> Juncbufo 0.66622672 0.05952038 0.2742529 #> Lolipere 0.46273045 0.14067027 0.3965993 #> Planlanc 0.51993753 0.16268893 0.3173735 #> Poaprat 0.39408053 0.13857406 0.4673454 #> Poatriv 0.05598349 0.50349824 0.4405183 #> Ranuflam 0.68509904 0.02138594 0.2935150 #> Rumeacet 0.40125987 0.36786003 0.2308801 #> Sagiproc 0.26050435 0.10087025 0.6386254 #> Salirepe 0.12527838 0.63869277 0.2360289 #> Scorautu 0.10895437 0.50510492 0.3859407 #> Trifprat 0.34544815 0.31712212 0.3374297 #> Trifrepe 0.02132183 0.39942191 0.5792563 #> Vicilath 0.12125433 0.30668844 0.5720572 #> Bracruta 0.07222706 0.17275938 0.7550136 #> Callcusp 0.29447422 0.05732850 0.6481973 inertcomp(mod) #> pCCA CCA CA #> Achimill 0.0173766015 0.007640656 0.02568493 #> Agrostol 0.0456558521 0.018171449 0.01828399 #> Airaprae 0.0066672285 0.034014687 0.06341666 #> Alopgeni 0.0325977567 0.036073980 0.02653486 #> Anthodor 0.0096274015 0.024246897 0.05996790 #> Bellpere 0.0154640710 0.002839669 0.01943887 #> Bromhord 0.0180126793 0.011993496 0.02450059 #> Chenalbu 0.0031913088 0.005130874 0.02052099 #> Cirsarve 0.0110663060 0.003846389 0.02661204 #> Comapalu 0.0127652351 0.050389111 0.01615116 #> Eleopalu 0.0797827194 0.012780901 0.05530588 #> Elymrepe 0.0193932154 0.005749967 0.06207879 #> Empenigr 0.0063826176 0.013046147 0.04216766 #> Hyporadi 0.0046669914 0.035992710 0.07932587 #> Juncarti 0.0359126341 0.007276518 0.03948420 #> Juncbufo 0.0494087668 0.004414156 0.02033917 #> Lolipere 0.0368344271 0.011197683 0.03157023 #> Planlanc 0.0366139947 0.011456552 0.02234944 #> Poaprat 0.0142991623 0.005028142 0.01695757 #> Poatriv 0.0028845344 0.025942611 0.02269759 #> Ranuflam 0.0446783229 0.001394671 0.01914141 #> Rumeacet 0.0288221948 0.026423110 0.01658394 #> Sagiproc 0.0151161507 0.005853146 0.03705718 #> Salirepe 0.0142756439 0.072779924 0.02689581 #> Scorautu 0.0030643984 0.014206339 0.01085478 #> Trifprat 0.0228613139 0.020986733 0.02233067 #> Trifrepe 0.0008339368 0.015622139 0.02265580 #> Vicilath 0.0049088357 0.012415912 0.02315905 #> Bracruta 0.0032317812 0.007730074 0.03378289 #> Callcusp 0.0319130878 0.006212868 0.07024716 # vif.cca vif.cca(mod) #> Moisture.L Moisture.Q Moisture.C A1 ManagementHF ManagementNM #> 1.504327 1.284489 1.347660 1.367328 2.238653 2.570972 #> ManagementSF #> 2.424444 # Aliased constraints mod <- cca(dune ~ ., dune.env) mod #> Call: cca(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 1.5032 0.7106 12 #> Unconstrained 0.6121 0.2894 7 #> Inertia is scaled Chi-square #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4671 0.3410 0.1761 0.1532 0.0953 0.0703 0.0589 0.0499 0.0318 0.0260 0.0228 #> CCA12 #> 0.0108 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 #> 0.27237 0.10876 0.08975 0.06305 0.03489 0.02529 0.01798 #> vif.cca(mod) #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> 2.208249 2.858927 3.072715 3.587087 6.608315 142.359372 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> 12.862713 2.642718 3.007238 80.828330 49.294455 21.433337 #> Manure^4 #> NA alias(mod) #> Model : #> dune ~ A1 + Moisture + Management + Use + Manure #> #> Complete : #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> Manure^4 8.366600 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> Manure^4 5.291503 -4.472136 2.645751 #> with(dune.env, table(Management, Manure)) #> Manure #> Management 0 1 2 3 4 #> BF 0 2 1 0 0 #> HF 0 1 2 2 0 #> NM 6 0 0 0 0 #> SF 0 0 1 2 3 # The standard correlations (not recommended) ## IGNORE_RDIFF_BEGIN spenvcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 #> 0.9636709 0.9487249 0.9330741 0.8734876 0.9373716 0.8362687 0.9748793 0.8392720 #> CCA9 CCA10 CCA11 CCA12 #> 0.8748741 0.6087512 0.6633248 0.7581210 intersetcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 #> A1 -0.5332506 0.13691202 -0.47996401 -0.259859587 -0.09894964 #> Moisture.L -0.8785505 0.17867589 0.03714134 0.181952935 -0.09826534 #> Moisture.Q -0.1956664 -0.33044917 -0.27321286 -0.180333890 0.26609291 #> Moisture.C -0.2023782 -0.09698397 0.28596824 -0.261712720 -0.49103002 #> ManagementHF 0.3473460 0.01680324 -0.51205769 0.194144965 0.30752664 #> ManagementNM -0.5699549 -0.61111645 0.14751127 -0.013777789 0.04571982 #> ManagementSF -0.1197499 0.64084416 0.19780650 0.134892908 -0.09679992 #> Use.L -0.1871999 0.32990444 -0.30941161 -0.372747011 0.09586963 #> Use.Q -0.1820298 -0.48874152 -0.01997442 -0.009812946 0.04812588 #> Manure.L 0.3175126 0.65945634 0.03724864 -0.025383543 -0.04077470 #> Manure.Q -0.4075615 -0.21149073 0.49297244 -0.176686201 0.11973190 #> Manure.C 0.4676279 0.11376054 0.29132473 -0.173382982 0.14219924 #> Manure^4 0.2222349 -0.12789494 -0.12921227 0.108367170 -0.02559567 #> CCA6 CCA7 CCA8 CCA9 CCA10 #> A1 -0.15225816 0.25788462 0.19247720 -0.27694466 -0.1158449480 #> Moisture.L -0.02923342 0.07858647 -0.10772510 0.07101300 0.0952517164 #> Moisture.Q -0.11211675 0.05062810 -0.48302647 0.06138704 -0.2053304965 #> Moisture.C -0.23581275 -0.38693407 -0.10144580 -0.21907160 0.1875632770 #> ManagementHF -0.24278705 0.16364055 -0.14053438 0.31066725 0.1310215145 #> ManagementNM -0.06430101 0.23917584 0.14375754 -0.27103732 0.0002768613 #> ManagementSF -0.01611984 -0.49726250 0.08073472 -0.30235728 -0.1381281272 #> Use.L 0.19127262 -0.44624831 -0.18450714 0.12950951 0.0452826749 #> Use.Q 0.13485545 0.10367354 -0.11020112 0.41245485 -0.0766932005 #> Manure.L -0.22265819 -0.49627772 -0.16971786 -0.03943343 -0.0045229147 #> Manure.Q -0.19402211 -0.11937394 0.17611673 -0.44002593 0.0903998202 #> Manure.C 0.14760330 0.07842345 0.37774417 0.10181374 0.1055057288 #> Manure^4 -0.36683782 0.05953330 0.40927409 -0.06054381 -0.1500198368 #> CCA11 CCA12 #> A1 -0.03550223 -0.08881387 #> Moisture.L 0.06404776 -0.08587882 #> Moisture.Q -0.21810558 0.16917878 #> Moisture.C 0.13701079 -0.14260914 #> ManagementHF 0.17283125 0.13296499 #> ManagementNM -0.01358436 0.09533598 #> ManagementSF -0.01468592 -0.06614834 #> Use.L -0.08584883 0.32559307 #> Use.Q 0.41893616 0.04881247 #> Manure.L 0.02396993 0.13049087 #> Manure.Q 0.12987366 0.07137031 #> Manure.C 0.05176927 -0.41550238 #> Manure^4 -0.41603287 0.01661279 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS find goodness fit measure points nonmetric multidimensional scaling, function stressplot makes Shepard diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"# S3 method for metaMDS goodness(object, dis, ...) # S3 method for default stressplot(object, dis, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"object result object metaMDS, monoMDS isoMDS. dis Dissimilarities. used metaMDS monoMDS, must used isoMDS. pch Plotting character points. Default dependent number points. p.col, l.col Point line colours. lwd Line width. monoMDS default lwd = 1 two lines drawn, lwd = 2 otherwise. ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS finds goodness fit statistic observations (points). defined sum squared values equal squared stress. Large values indicate poor fit. absolute values goodness statistic depend definition stress: isoMDS expresses stress percents, therefore goodness values 100 times higher monoMDS expresses stress proportion. Function stressplot draws Shepard diagram plot ordination distances monotone linear fit line original dissimilarities. addition, displays two correlation-like statistics goodness fit graph. nonmetric fit based stress \\(S\\) defined \\(R^2 = 1-S^2\\). “linear fit” squared correlation fitted values ordination distances. monoMDS, “linear fit” \\(R^2\\) “stress type 2” equal. functions can used metaMDS, monoMDS isoMDS. original dissimilarities given monoMDS metaMDS results (latter tries reconstruct dissimilarities using metaMDSredist isoMDS used engine). isoMDS dissimilarities must given. either case, functions inspect dissimilarities consistent current ordination, refuse analyse inconsistent dissimilarities. Function goodness.metaMDS generic vegan, must spell name completely isoMDS class.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness returns vector values. Function stressplot returns invisibly object items original dissimilarities, ordination distances fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"data(varespec) mod <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2410643 #> Run 2 stress 0.1948413 #> Run 3 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.0416189 max resid 0.1517616 #> Run 4 stress 0.2383056 #> Run 5 stress 0.2096851 #> Run 6 stress 0.2329653 #> Run 7 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 2.732084e-05 max resid 8.096006e-05 #> ... Similar to previous best #> Run 8 stress 0.1825658 #> ... Procrustes: rmse 1.167373e-05 max resid 3.019892e-05 #> ... Similar to previous best #> Run 9 stress 0.2092456 #> Run 10 stress 0.1976154 #> Run 11 stress 0.2331577 #> Run 12 stress 0.2330486 #> Run 13 stress 0.1948413 #> Run 14 stress 0.2356726 #> Run 15 stress 0.1955838 #> Run 16 stress 0.2339637 #> Run 17 stress 0.2088293 #> Run 18 stress 0.2138946 #> Run 19 stress 0.2136761 #> Run 20 stress 0.2352729 #> *** Best solution repeated 2 times stressplot(mod) gof <- goodness(mod) gof #> [1] 0.02984516 0.03513715 0.04189194 0.04598243 0.04003134 0.03441442 #> [7] 0.03294898 0.03050145 0.03060754 0.02994058 0.03526332 0.02621426 #> [13] 0.03830984 0.02980913 0.03369611 0.02225897 0.03561569 0.03505254 #> [19] 0.06577496 0.03268352 0.03503121 0.02956628 0.05168013 0.04601957 plot(mod, display = \"sites\", type = \"n\") points(mod, display = \"sites\", cex = 2*gof/mean(gof))"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":null,"dir":"Reference","previous_headings":"","what":"Indicator Power of Species — indpower","title":"Indicator Power of Species — indpower","text":"Indicator power calculation Halme et al. (2009) congruence indicator target species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indicator Power of Species — indpower","text":"","code":"indpower(x, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indicator Power of Species — indpower","text":"x Community data frame matrix. type type statistic returned. See Details explanation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indicator Power of Species — indpower","text":"Halme et al. (2009) described index indicator power defined \\(IP_I = \\sqrt{\\times b}\\), \\(= S / O_I\\) \\(b = 1 - (O_T - S) / (N - O_I)\\). \\(N\\) number sites, \\(S\\) number shared occurrences indicator (\\(\\)) target (\\(T\\)) species. \\(O_I\\) \\(O_T\\) number occurrences indicator target species. type argument function call enables choose statistic return. type = 0 returns \\(IP_I\\), type = 1 returns \\(\\), type = 2 returns \\(b\\). Total indicator power (TIP) indicator species column mean (without value, see examples). Halme et al. (2009) explain calculate confidence intervals statistics, see Examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indicator Power of Species — indpower","text":"matrix indicator species rows target species columns (indicated first letters row/column names).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indicator Power of Species — indpower","text":"Halme, P., Mönkkönen, M., Kotiaho, J. S, Ylisirniö, -L. 2009. Quantifying indicator power indicator species. Conservation Biology 23: 1008--1016.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indicator Power of Species — indpower","text":"Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indicator Power of Species — indpower","text":"","code":"data(dune) ## IP values ip <- indpower(dune) ## and TIP values diag(ip) <- NA (TIP <- rowMeans(ip, na.rm=TRUE)) #> i.Achimill i.Agrostol i.Airaprae i.Alopgeni i.Anthodor i.Bellpere i.Bromhord #> 0.3186250 0.3342800 0.2168133 0.3416198 0.3567884 0.3432281 0.3665632 #> i.Chenalbu i.Cirsarve i.Comapalu i.Eleopalu i.Elymrepe i.Empenigr i.Hyporadi #> 0.2095044 0.2781640 0.1713273 0.2414787 0.3263516 0.2016196 0.2378197 #> i.Juncarti i.Juncbufo i.Lolipere i.Planlanc i.Poaprat i.Poatriv i.Ranuflam #> 0.2915850 0.3331330 0.3998442 0.3426064 0.4094319 0.3929520 0.2663080 #> i.Rumeacet i.Sagiproc i.Salirepe i.Scorautu i.Trifprat i.Trifrepe i.Vicilath #> 0.3484684 0.3788905 0.2898512 0.4362493 0.3145854 0.4503764 0.2605349 #> i.Bracruta i.Callcusp #> 0.4252676 0.2070766 ## p value calculation for a species ## from Halme et al. 2009 ## i is ID for the species i <- 1 fun <- function(x, i) indpower(x)[i,-i] ## 'c0' randomizes species occurrences os <- oecosimu(dune, fun, \"c0\", i=i, nsimul=99) #> Warning: nullmodel transformed 'comm' to binary data ## get z values from oecosimu output z <- os$oecosimu$z ## p-value (p <- sum(z) / sqrt(length(z))) #> [1] -1.471628 ## 'heterogeneity' measure (chi2 <- sum((z - mean(z))^2)) #> [1] 86.33867 pchisq(chi2, df=length(z)-1) #> [1] 0.9999999 ## Halme et al.'s suggested output out <- c(TIP=TIP[i], significance=p, heterogeneity=chi2, minIP=min(fun(dune, i=i)), varIP=sd(fun(dune, i=i)^2)) out #> TIP.i.Achimill significance heterogeneity minIP varIP #> 0.3186250 -1.4716280 86.3386705 0.0000000 0.2142097"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Linear Model Diagnostics for Constrained Ordination — influence.cca","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"set function extracts influence statistics linear model statistics directly constrained ordination result object cca, rda, capscale dbrda. constraints linear model functions support functions return identical results corresponding linear models (lm), can use documentation. main functions normal usage leverage values (hatvalues), standardized residuals (rstandard), studentized leave-one-residuals (rstudent), Cook's distance (cooks.distance). addition, vcov returns variance-covariance matrix coefficients, diagonal values variances coefficients. functions mainly support functions , can used directly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"# S3 method for cca hatvalues(model, ...) # S3 method for cca rstandard(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca rstudent(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca cooks.distance(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca sigma(object, type = c(\"response\", \"canoco\"), ...) # S3 method for cca vcov(object, type = \"canoco\", ...) # S3 method for cca SSD(object, type = \"canoco\", ...) # S3 method for cca qr(x, ...) # S3 method for cca df.residual(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"model, object, x constrained ordination result object. type Type statistics used extracting raw residuals residual standard deviation (sigma). Either \"response\" species data difference WA LC scores \"canoco\". ... arguments functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"vegan algorithm constrained ordination uses linear model (weighted linear model cca) find fitted values dependent community data, constrained ordination based fitted response (Legendre & Legendre 2012). hatvalues give leverage values constraints, leverage independent response data. influence statistics (rstandard, rstudent, cooks.distance) based leverage, raw residuals residual standard deviation (sigma). type = \"response\" raw residuals given unconstrained component constrained ordination, influence statistics matrix dimensions . observations times . species. cca statistics obtained lm model using Chi-square standardized species data (see decostand) dependent variable, row sums community data weights, rda lm model uses non-modified community data weights. algorithm CANOCO software constraints results iteration performing linear regression weighted averages (WA) scores constraints taking fitted values regression linear combination (LC) scores (ter Braak 1984). WA scores directly found species scores, LC scores linear combinations constraints regression. type = \"canoco\" raw residuals differences WA LC scores, residual standard deviation (sigma) taken axis sum squared WA scores minus one. quantities relationship residual component ordination, rather methodological artefacts algorithm used vegan. result matrix dimensions . observations times . constrained axes. Function vcov returns matrix variances covariances regression coefficients. diagonal values matrix variances, square roots give standard errors regression coefficients. function based SSD extracts sum squares crossproducts residuals. residuals defined similarly influence measures type similar properties limitations, define dimensions result matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Function .mlm casts ordination object multiple linear model class \"mlm\" (see lm), similar statistics can derived modified object set functions. However, problems R implementation analysis multiple linear model objects. results differ, current set functions probable correct. use .mlm objects avoided.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"data(varespec, varechem) mod <- cca(varespec ~ Al + P + K, varechem) ## leverage hatvalues(mod) #> 18 15 24 27 23 19 22 #> 0.06904416 0.06666628 0.15245083 0.18944882 0.09291510 0.05122338 0.15309307 #> 16 28 13 14 20 25 7 #> 0.09605909 0.27139695 0.75889765 0.04958141 0.06582891 0.10590183 0.20630888 #> 5 6 3 4 2 9 12 #> 0.19797654 0.16280522 0.22738889 0.30915530 0.15557066 0.14855598 0.09046701 #> 10 11 21 #> 0.12745850 0.10984996 0.14195559 plot(hatvalues(mod), type = \"h\") ## ordination plot with leverages: points with high leverage have ## similar LC and WA scores plot(mod, type = \"n\") ordispider(mod) # segment from LC to WA scores points(mod, dis=\"si\", cex=5*hatvalues(mod), pch=21, bg=2) # WA scores text(mod, dis=\"bp\", col=4) ## deviation and influence head(rstandard(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.4084518 0.9442480 -0.68178124 -0.798241724 0.9883838 -0.3086868 #> 15 -1.3902462 -1.5717947 -0.70784872 -0.645563228 0.2353736 -0.1679226 #> 24 0.9622453 -0.9520875 -0.08884556 -0.654099911 0.2420416 0.4832198 #> 27 -1.1080099 1.0938951 1.70146427 -0.196668562 -0.3937467 -0.7424140 #> 23 0.3979939 1.3218254 -0.63872221 -1.003315524 1.8996365 -0.4495408 #> 19 -1.5874575 0.7894087 -0.59609083 -0.006142973 0.1334143 -0.1060450 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 -0.5785258 -0.4585683 0.7640788 4.3748349 -0.39301720 -0.656213958 #> 15 -0.5416812 -0.4594716 -0.3410155 -0.2704388 -0.05769657 0.406022095 #> 24 -0.6409619 0.1908003 0.0198320 -0.2175720 3.75416938 -0.009140093 #> 27 4.2976822 -0.2704153 0.9211453 -0.2037075 -0.84410200 -0.712968237 #> 23 -0.8356637 -0.2779318 -0.0206861 -0.2899895 -0.67773316 -0.294674408 #> 19 -0.4453843 -0.3739569 -0.3174207 -0.2429740 -0.14281646 -0.819230368 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 -0.4933634 -0.6030042 -1.3812122 -0.07430854 -0.5332731 -0.7140513 #> 15 -0.3630434 -0.4055353 1.9930791 0.02126658 -0.2274864 -0.6730241 #> 24 2.2312025 -1.3907968 0.5792314 -0.45358405 -0.4628096 -0.2153846 #> 27 -0.5715008 1.6310289 0.8124329 -0.34628172 -0.8630615 0.9423113 #> 23 -0.4684368 -1.1995321 -0.8241477 -0.04953929 0.7890327 -0.6774675 #> 19 -0.3186606 -0.3948224 0.6180114 0.05035990 0.8260214 2.3626483 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 18 0.2918311 -0.42027512 -0.40191333 1.5303054 0.85056948 -0.2218833 #> 15 -0.6168627 -0.47096984 -0.42715787 0.1758993 -0.45409579 -0.3739920 #> 24 1.9274063 0.09984384 0.21066100 0.1668363 -0.08508535 -1.2835676 #> 27 -1.5729552 -0.31944598 -0.09167396 -0.1864165 0.51080623 -0.1083797 #> 23 0.4875575 0.61798970 -0.18848086 0.5574309 0.28079468 -0.4026464 #> 19 -0.1141521 -0.25344373 -0.31179839 -0.2391669 -0.65577452 0.6584127 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris #> 18 -0.39796095 0.93836573 -0.2564543 0.33864378 1.1572584 -0.2083305 #> 15 0.06761526 0.24340663 -0.1767166 0.27491203 1.0721632 1.9724410 #> 24 1.24902375 -0.98471253 -0.4801382 2.51184311 -1.4063518 -0.3084304 #> 27 -0.59021669 -1.25354423 -0.2460447 -1.09351514 -1.1681499 -1.0337232 #> 23 -0.34604539 -0.10730202 3.9477300 2.51924664 0.3536280 3.3882402 #> 19 -0.33866721 0.02698153 0.1776632 0.03968833 -0.7512944 -0.4763562 #> Cladchlo Cladbotr Cladamau Cladsp Cetreric Cetrisla #> 18 -0.5711604 -0.4914716 4.0852019 0.2489284 -0.4428064 -0.5834462 #> 15 -0.4347061 -0.6517740 -0.3155708 -0.2568784 0.2369559 -0.3468005 #> 24 0.6013607 0.4603779 -0.1576900 -0.5995616 2.7826114 0.3931826 #> 27 -0.5436659 -0.2788962 -0.1606997 0.1557879 -0.7298364 -0.5867612 #> 23 0.1029999 0.6494142 -0.3415411 -0.2330698 -0.6309632 -0.5301977 #> 19 0.2793712 -0.1315438 -0.2705486 -0.2865837 -0.4664078 -0.5056846 #> Flavniva Nepharct Stersp Peltapht Icmaeric Cladcerv #> 18 0.30428187 -0.3624631 -0.23665431 -0.1571633 -0.630333375 0.12058739 #> 15 0.17345018 -0.1919943 0.05469573 -0.3233311 -0.561177494 0.08518455 #> 24 -0.74154401 -0.4146848 -0.05500461 -0.7609417 0.255058737 -0.92761801 #> 27 0.11301489 -0.5064006 -0.08681568 -0.1247151 -0.001277338 -0.04038189 #> 23 0.09411988 -0.4627811 0.47668055 3.5826478 -0.274664798 -0.05517988 #> 19 0.07211309 -0.1693122 -0.17244475 -0.3155345 -0.461532920 -0.02709075 #> Claddefo Cladphyl #> 18 -0.43581630 -0.2098378 #> 15 0.94176661 -0.1028102 #> 24 -0.07508682 -1.0479632 #> 27 -1.06110299 -0.4908554 #> 23 2.66430575 -0.4244333 #> 19 -0.08427954 -0.1692474 head(cooks.distance(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.003093283 0.01653142 0.0086184263 1.181427e-02 0.0181129462 0.0017667454 #> 15 0.034513793 0.04411649 0.0089472619 7.441951e-03 0.0009892926 0.0005035324 #> 24 0.041636714 0.04076229 0.0003549575 1.923947e-02 0.0026344196 0.0105001237 #> 27 0.071736260 0.06992022 0.1691597848 2.260067e-03 0.0090591037 0.0322065174 #> 23 0.004056312 0.04474315 0.0104472601 2.577825e-02 0.0924100906 0.0051750754 #> 19 0.034013281 0.00841101 0.0047958896 5.093326e-07 0.0002402422 0.0001517834 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 0.006205594 0.003898934 1.082466e-02 0.3548634058 2.863921e-03 7.984152e-03 #> 15 0.005239584 0.003769873 2.076622e-03 0.0013060122 5.944416e-05 2.943802e-03 #> 24 0.018474359 0.001637053 1.768633e-05 0.0021286828 6.337714e-01 3.756697e-06 #> 27 1.079245024 0.004272814 4.958014e-02 0.0024247421 4.163335e-02 2.970242e-02 #> 23 0.017883042 0.001978130 1.095811e-05 0.0021534900 1.176240e-02 2.223634e-03 #> 19 0.002677405 0.001887502 1.359924e-03 0.0007968265 2.752966e-04 9.058502e-03 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 0.004513066 0.006741841 0.035371924 1.023801e-04 0.0052727512 0.009453588 #> 15 0.002353565 0.002936747 0.070934636 8.076156e-06 0.0009241023 0.008088546 #> 24 0.223863289 0.086982577 0.015087217 9.251676e-03 0.0096318471 0.002086096 #> 27 0.019084693 0.155444288 0.038567940 7.006653e-03 0.0435246182 0.051884812 #> 23 0.005619275 0.036846999 0.017393559 6.284604e-05 0.0159429407 0.011753178 #> 19 0.001370570 0.002104010 0.005155103 3.423056e-05 0.0092093061 0.075342921 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang #> 18 0.0015790689 0.0032749544 0.0029950405 0.0434204290 0.0134139742 #> 15 0.0067949512 0.0039609143 0.0032582625 0.0005525065 0.0036821731 #> 24 0.1670519112 0.0004482780 0.0019955986 0.0012516596 0.0003255476 #> 27 0.1445719869 0.0059627459 0.0004910701 0.0020305802 0.0152462776 #> 23 0.0060873754 0.0097800450 0.0009097308 0.0079572035 0.0020190921 #> 19 0.0001758783 0.0008669766 0.0013121758 0.0007720518 0.0058043470 #> Cladstel Cladunci Cladcocc Cladcorn Cladgrac Cladfimb #> 18 0.0009128242 2.936424e-03 1.632609e-02 0.0012194323 2.126298e-03 0.024831249 #> 15 0.0024976637 8.163919e-05 1.057972e-03 0.0005576530 1.349574e-03 0.020527262 #> 24 0.0740870642 7.015301e-02 4.360375e-02 0.0103666250 2.837200e-01 0.088939118 #> 27 0.0006863532 2.035516e-02 9.181862e-02 0.0035373629 6.987166e-02 0.079734929 #> 23 0.0041517010 3.066511e-03 2.948453e-04 0.3990922140 1.625248e-01 0.003202372 #> 19 0.0058511426 1.548070e-03 9.826017e-06 0.0004260291 2.126033e-05 0.007618415 #> Cladcris Cladchlo Cladbotr Cladamau Cladsp Cetreric #> 18 0.0008047178 0.0060485887 0.0044785217 0.3094317912 0.001148912 0.003635513 #> 15 0.0694732020 0.0033744318 0.0075858327 0.0017782910 0.001178323 0.001002639 #> 24 0.0042777886 0.0162620357 0.0095308986 0.0011181833 0.016164882 0.348184969 #> 27 0.0624395810 0.0172709295 0.0045450274 0.0015089732 0.001418140 0.031124511 #> 23 0.2939860714 0.0002716767 0.0107999564 0.0029872002 0.001391075 0.010194981 #> 19 0.0030627243 0.0010534342 0.0002335528 0.0009879496 0.001108529 0.002936134 #> Cetrisla Flavniva Nepharct Stersp Peltapht Icmaeric #> 18 0.006311601 1.716683e-03 0.0024359336 1.038405e-03 0.0004579733 7.366793e-03 #> 15 0.002147676 5.372282e-04 0.0006582422 5.342149e-05 0.0018668272 5.623539e-03 #> 24 0.006951741 2.472742e-02 0.0077328775 1.360514e-04 0.0260380002 2.925400e-03 #> 27 0.020117510 7.463161e-04 0.0149844191 4.404005e-04 0.0009088445 9.533741e-08 #> 23 0.007198701 2.268512e-04 0.0054844061 5.818798e-03 0.3286900860 1.931899e-03 #> 19 0.003451467 7.018956e-05 0.0003869197 4.013694e-04 0.0013438101 2.875078e-03 #> Cladcerv Claddefo Cladphyl #> 18 2.696135e-04 3.521639e-03 0.0008164040 #> 15 1.295779e-04 1.583784e-02 0.0001887477 #> 24 3.869397e-02 2.535317e-04 0.0493852158 #> 27 9.528504e-05 6.579101e-02 0.0140785723 #> 23 7.797221e-05 1.817802e-01 0.0046131462 #> 19 9.905729e-06 9.587131e-05 0.0003866237 ## Influence measures from lm y <- decostand(varespec, \"chi.square\") # needed in cca y1 <- with(y, Cladstel) # take one species for lm lmod1 <- lm(y1 ~ Al + P + K, varechem, weights = rowSums(varespec)) ## numerically identical within 2e-15 all(abs(cooks.distance(lmod1) - cooks.distance(mod)[, \"Cladstel\"]) < 1e-8) #> [1] TRUE ## t-values of regression coefficients based on type = \"canoco\" ## residuals coef(mod) #> CCA1 CCA2 CCA3 #> Al 0.007478556 -0.001883637 0.003380774 #> P -0.006491081 -0.102189737 -0.022306682 #> K -0.006755568 0.015343662 0.017067351 coef(mod)/sqrt(diag(vcov(mod, type = \"canoco\"))) #> CCA1 CCA2 CCA3 #> Al 6.5615451 -1.397643 3.313629 #> P -0.4576132 -6.092557 -1.756774 #> K -2.0862129 4.007159 5.887926"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":null,"dir":"Reference","previous_headings":"","what":"Isometric Feature Mapping Ordination — isomap","title":"Isometric Feature Mapping Ordination — isomap","text":"function performs isometric feature mapping consists three simple steps: (1) retain shortest dissimilarities among objects, (2) estimate dissimilarities shortest path distances, (3) perform metric scaling (Tenenbaum et al. 2000).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"isomap(dist, ndim=10, ...) isomapdist(dist, epsilon, k, path = \"shortest\", fragmentedOK =FALSE, ...) # S3 method for isomap summary(object, ...) # S3 method for isomap plot(x, net = TRUE, n.col = \"gray\", type = \"points\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Isometric Feature Mapping Ordination — isomap","text":"dist Dissimilarities. ndim Number axes metric scaling (argument k cmdscale). epsilon Shortest dissimilarity retained. k Number shortest dissimilarities retained point. epsilon k given, epsilon used. path Method used stepacross estimate shortest path, alternatives \"shortest\" \"extended\". fragmentedOK dissimilarity matrix fragmented. TRUE, analyse largest connected group, otherwise stop error. x, object isomap result object. net Draw net retained dissimilarities. n.col Colour drawn net segments. can also vector recycled points, colour net segment mixture joined points. type Plot observations either \"points\", \"text\" use \"none\" plot observations. \"text\" use ordilabel net = TRUE ordiplot net = FALSE, pass extra arguments functions. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Isometric Feature Mapping Ordination — isomap","text":"function isomap first calls function isomapdist dissimilarity transformation, performs metric scaling result. arguments isomap passed isomapdist. functions separate isompadist transformation easily used functions simple linear mapping cmdscale. Function isomapdist retains either dissimilarities equal shorter epsilon, epsilon given, least k shortest dissimilarities point. complete dissimilarity matrix reconstructed using stepacross using either flexible shortest paths extended dissimilarities (details, see stepacross). De'ath (1999) actually published essentially method Tenenbaum et al. (2000), De'ath's function available function xdiss non-CRAN package mvpart. differences isomap introduced k criterion, whereas De'ath used epsilon criterion. practice, De'ath also retains higher proportion dissimilarities typical isomap. plot function uses internally ordiplot, except adds text net using ordilabel. plot function passes extra arguments functions. addition, vegan3d package function rgl.isomap make dynamic 3D plots can rotated screen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Isometric Feature Mapping Ordination — isomap","text":"Function isomapdist returns dissimilarity object similar dist. Function isomap returns object class isomap plot summary methods. plot function returns invisibly object class ordiplot. Function scores can extract ordination scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Isometric Feature Mapping Ordination — isomap","text":"De'ath, G. (1999) Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecology 144, 191--199 Tenenbaum, J.B., de Silva, V. & Langford, J.C. (2000) global network framework nonlinear dimensionality reduction. Science 290, 2319--2323.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Isometric Feature Mapping Ordination — isomap","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Isometric Feature Mapping Ordination — isomap","text":"Tenenbaum et al. (2000) justify isomap tool unfolding manifold (e.g. 'Swiss Roll'). Even manifold structure, sampling must even dense dissimilarities along manifold shorter across folds. data manifold structure, results sensitive parameter values.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"## The following examples also overlay minimum spanning tree to ## the graphics in red. op <- par(mar=c(4,4,1,1)+0.2, mfrow=c(2,2)) data(BCI) dis <- vegdist(BCI) tr <- spantree(dis) pl <- ordiplot(cmdscale(dis), main=\"cmdscale\") #> species scores not available lines(tr, pl, col=\"red\") ord <- isomap(dis, k=3) ord #> #> Isometric Feature Mapping (isomap) #> #> Call: #> isomap(dist = dis, k = 3) #> #> Distance method: bray shortest isomap #> Criterion: k = 3 pl <- plot(ord, main=\"isomap k=3\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, k=5), main=\"isomap k=5\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, epsilon=0.45), main=\"isomap epsilon=0.45\") lines(tr, pl, col=\"red\") par(op) ## colour points and web by the dominant species dom <- apply(BCI, 1, which.max) ## need nine colours, but default palette has only eight op <- palette(c(palette(\"default\"), \"sienna\")) plot(ord, pch = 16, col = dom, n.col = dom) palette(op)"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":null,"dir":"Reference","previous_headings":"","what":"Kendall coefficient of concordance — kendall.global","title":"Kendall coefficient of concordance — kendall.global","text":"Function kendall.global computes tests coefficient concordance among several judges (variables, species) permutation test. Function kendall.post carries posteriori tests contributions individual judges (variables, species) overall concordance group permutation tests. several groups judges identified data table, coefficients concordance (kendall.global) posteriori tests (kendall.post) computed group separately. Use ecology: identify significant species associations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"kendall.global(Y, group, nperm = 999, mult = \"holm\") kendall.post(Y, group, nperm = 999, mult = \"holm\")"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kendall coefficient of concordance — kendall.global","text":"Y Data file (data frame matrix) containing quantitative semiquantitative data. Rows objects columns judges (variables). community ecology, table often site--species table. group vector defining judges divided groups. See example . groups explicitly defined, judges data file considered forming single group. nperm Number permutations performed. Default 999. mult Correct P-values multiple testing using alternatives described p.adjust addition \"sidak\" (see Details). Bonferroni correction overly conservative; recommended. included allow comparisons methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Kendall coefficient of concordance — kendall.global","text":"Y must contain quantitative data. transformed ranks within column computation coefficient concordance. search species associations described Legendre (2005) proceeds 3 steps: (1) Correlation analysis species. possible method compute Ward's agglomerative clustering matrix correlations among species. detail: (1.1) compute Pearson Spearman correlation matrix (correl.matrix) among species; (1.2) turn distance matrix: mat.D = .dist(1-correl.matrix); (1.3) carry Ward's hierarchical clustering matrix using hclust: clust.ward = hclust(mat.D, \"ward\"); (1.4) plot dendrogram: plot(clust.ward, hang=-1); (1.5) cut dendrogram two groups, retrieve vector species membership: group.2 = cutree(clust.ward, k=2). (1.6) steps 2 3 , may come back try divisions species k = \\(3, 4, 5, \\dots\\) groups. (2) Compute global tests significance 2 () groups using function kendall.global vector defining groups. Groups globally significant must refined abandoned. (3) Compute posteriori tests contribution individual species concordance group using function kendall.post vector defining groups. species negative values \"Spearman.mean\", means species clearly belong group, hence group inclusive. Go back (1.5) cut dendrogram finely. left right groups can cut separately, independently levels along dendrogram; write vector group membership cutree produce desired groups. corrections used multiple testing applied list P-values (P); take account number tests (k) carried simultaneously (number groups kendall.global, number species kendall.post). corrections performed using function p.adjust; see function description correction methods. addition, Šidák correction defined \\(P_{corr} = 1 -(1 - P)^k\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Kendall coefficient of concordance — kendall.global","text":"table containing following information rows. columns correspond groups \"judges\" defined vector \"group\". function Kendall.post used, many tables number predefined groups. W Kendall's coefficient concordance, W. F F statistic. F = W*(m-1)/(1-W) m number judges. Prob.F Probability associated F statistic, computed F distribution nu1 = n-1-(2/m) nu2 = nu1*(m-1); n number objects. Corrected prob.F Probabilities associated F, corrected using method selected parameter mult. Shown one group. Chi2 Friedman's chi-square statistic (Friedman 1937) used permutation test W. Prob.perm Permutational probabilities, uncorrected. Corrected prob.perm Permutational probabilities corrected using method selected parameter mult. Shown one group. Spearman.mean Mean Spearman correlations judge test judges group. W.per.species Contribution judge test overall concordance statistic group.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Kendall coefficient of concordance — kendall.global","text":"Friedman, M. 1937. use ranks avoid assumption normality implicit analysis variance. Journal American Statistical Association 32: 675-701. Kendall, M. G. B. Babington Smith. 1939. problem m rankings. Annals Mathematical Statistics 10: 275-287. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245. Legendre, P. 2009. Coefficient concordance. : Encyclopedia Research Design. SAGE Publications (press). Siegel, S. N. J. Castellan, Jr. 1988. Nonparametric statistics behavioral sciences. 2nd edition. McGraw-Hill, New York.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Kendall coefficient of concordance — kendall.global","text":"F. Guillaume Blanchet, University Alberta, Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"data(mite) mite.hel <- decostand(mite, \"hel\") # Reproduce the results shown in Table 2 of Legendre (2005), a single group mite.small <- mite.hel[c(4,9,14,22,31,34,45,53,61,69),c(13:15,23)] kendall.global(mite.small, nperm=49) #> $Concordance_analysis #> Group.1 #> W 0.44160305 #> F 2.37252221 #> Prob.F 0.04403791 #> Chi2 15.89770992 #> Prob.perm 0.12000000 #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.small, mult=\"holm\", nperm=49) #> $A_posteriori_tests #> TVEL ONOV SUCT Trhypch1 #> Spearman.mean 0.3265678 0.3965503 0.4570402 -0.1681251 #> W.per.species 0.4949258 0.5474127 0.5927802 0.1239061 #> Prob 0.0800000 0.0200000 0.0200000 0.7200000 #> Corrected prob 0.1600000 0.0800000 0.0800000 0.7200000 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # Reproduce the results shown in Tables 3 and 4 of Legendre (2005), 2 groups group <-c(1,1,2,1,1,1,1,1,2,1,1,1,1,1,1,2,1,2,1,1,1,1,2,1,2,1,1,1,1,1,2,2,2,2,2) kendall.global(mite.hel, group=group, nperm=49) #> $Concordance_analysis #> Group.1 Group.2 #> W 3.097870e-01 2.911888e-01 #> F 1.032305e+01 4.108130e+00 #> Prob.F 1.177138e-85 4.676566e-22 #> Corrected prob.F 2.354275e-85 4.676566e-22 #> Chi2 5.130073e+02 2.210123e+02 #> Prob.perm 2.000000e-02 2.000000e-02 #> Corrected prob.perm 4.000000e-02 4.000000e-02 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.hel, group=group, mult=\"holm\", nperm=49) #> $A_posteriori_tests_Group #> $A_posteriori_tests_Group[[1]] #> Brachy PHTH RARD SSTR Protopl MEGR #> Spearman.mean 0.1851177 0.4258111 0.359058 0.2505486 0.1802160 0.2833298 #> W.per.species 0.2190711 0.4497357 0.385764 0.2817757 0.2143736 0.3131911 #> Prob 0.0200000 0.0200000 0.020000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 #> MPRO HMIN HMIN2 NPRA TVEL ONOV #> Spearman.mean 0.09248024 0.2444656 0.4138494 0.1263751 0.4177343 0.3301159 #> W.per.species 0.13029357 0.2759462 0.4382723 0.1627761 0.4419954 0.3580278 #> Prob 0.14000000 0.0200000 0.0200000 0.0400000 0.0200000 0.0200000 #> Corrected prob 0.70000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> SUCT Oribatl1 PWIL Galumna1 Stgncrs2 HRUF #> Spearman.mean 0.2185421 0.421216 0.2574779 0.4180699 0.3623428 0.1250230 #> W.per.species 0.2511028 0.445332 0.2884163 0.4423170 0.3889118 0.1614804 #> Prob 0.0200000 0.020000 0.0200000 0.0200000 0.0200000 0.0600000 #> Corrected prob 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 0.7000000 #> PPEL SLAT FSET Lepidzts Eupelops Miniglmn #> Spearman.mean 0.2188216 0.3016159 0.4217606 0.2577037 0.1108022 0.2301430 #> W.per.species 0.2513707 0.3307153 0.4458539 0.2886327 0.1478521 0.2622203 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0600000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> $A_posteriori_tests_Group[[2]] #> HPAV TVIE LCIL Ceratoz1 Trhypch1 NCOR #> Spearman.mean 0.1222579 0.2712078 0.1906408 0.1375601 0.1342409 0.3342345 #> W.per.species 0.2020527 0.3374616 0.2642189 0.2159637 0.2129463 0.3947586 #> Prob 0.0400000 0.0200000 0.0200000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> LRUG PLAG2 Ceratoz3 Oppiminu Trimalc2 #> Spearman.mean 0.3446561 0.1833099 0.3188922 0.1764232 0.2498877 #> W.per.species 0.4042328 0.2575544 0.3808111 0.2512938 0.3180797 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Function linestack plots vertical one-dimensional plots numeric vectors. plots always labelled, labels moved vertically avoid overwriting.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"linestack(x, labels, cex = 0.8, side = \"right\", hoff = 2, air = 1.1, at = 0, add = FALSE, axis = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"x Numeric vector plotted. labels Labels used instead default (names x). May expressions drawn plotmath. cex Size labels. side Put labels \"right\" \"left\" axis. hoff Distance vertical axis label units width letter “m”. air Multiplier string height leave empty space labels. Position plot horizontal axis. add Add existing plot. axis Add axis plot. ... graphical parameters labels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function returns invisibly shifted positions labels user coordinates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Jari Oksanen modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function always draws labelled diagrams. want unlabelled diagrams, can use, e.g., plot, stripchart rug.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"## First DCA axis data(dune) ord <- decorana(dune) linestack(scores(ord, choices=1, display=\"sp\")) linestack(scores(ord, choices=1, display=\"si\"), side=\"left\", add=TRUE) title(main=\"DCA axis 1\") ## Expressions as labels N <- 10 # Number of sites df <- data.frame(Ca = rlnorm(N, 2), NO3 = rlnorm(N, 4), SO4 = rlnorm(N, 10), K = rlnorm(N, 3)) ord <- rda(df, scale = TRUE) ### vector of expressions for labels labs <- expression(Ca^{2+phantom()}, NO[3]^{-phantom()}, SO[4]^{2-phantom()}, K^{+phantom()}) scl <- \"sites\" linestack(scores(ord, choices = 1, display = \"species\", scaling = scl), labels = labs, air = 2) linestack(scores(ord, choices = 1, display = \"site\", scaling = scl), side = \"left\", add = TRUE) title(main = \"PCA axis 1\")"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":null,"dir":"Reference","previous_headings":"","what":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"standard CEP name four first letters generic name four first letters specific epithet Latin name. last epithet, may subspecific name, used current function. name one component, abbreviated eight characters (see abbreviate). returned names made unique function make.unique adds numbers end CEP names needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(names, seconditem = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"names names formatted CEP names. seconditem Take always second item original name abbreviated name instead last original item (default).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Cornell Ecology Programs (CEP) used eight-letter abbreviations species site names. species, names formed taking four first letters generic name four first letters specific subspecific epithet. current function first makes valid R names using make.names, splits elements. CEP name made taking four first letters first element, four first letters last (default) second element ( seconditem = TRUE). one name element, abbreviated eight letters. Finally, names made unique may add numbers duplicated names. CEP names originally used, old FORTRAN IV CHARACTER data type, text stored numerical variables, popular computers hold four characters. modern times, reason limitation, ecologists used names, may practical avoid congestion ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Function returns CEP names.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"function simpleminded rigid. must write better one need.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(c(\"Aa maderoi\", \"Poa sp.\", \"Cladina rangiferina\", \"Cladonia cornuta\", \"Cladonia cornuta var. groenlandica\", \"Cladonia rangiformis\", \"Bryoerythrophyllum\")) #> [1] \"Aamade\" \"Poasp\" \"Cladrang\" \"Cladcorn\" \"Cladgroe\" #> [6] \"Cladrang.1\" \"Bryrythr\" data(BCI) colnames(BCI) <- make.cepnames(colnames(BCI))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel Correlogram — mantel.correlog","title":"Mantel Correlogram — mantel.correlog","text":"Function mantel.correlog computes multivariate Mantel correlogram. Proposed Sokal (1986) Oden Sokal (1986), method also described Legendre Legendre (2012, pp. 819--821).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel Correlogram — mantel.correlog","text":"","code":"mantel.correlog(D.eco, D.geo=NULL, XY=NULL, n.class=0, break.pts=NULL, cutoff=TRUE, r.type=\"pearson\", nperm=999, mult=\"holm\", progressive=TRUE) # S3 method for mantel.correlog plot(x, alpha=0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel Correlogram — mantel.correlog","text":"D.eco ecological distance matrix, class either dist matrix. D.geo geographic distance matrix, class either dist matrix. Provide either D.geo XY. Default: D.geo=NULL. XY file Cartesian geographic coordinates points. Default: XY=NULL. n.class Number classes. n.class=0, Sturges equation used unless break points provided. break.pts Vector containing break points distance distribution. Provide (n.class+1) breakpoints, , list beginning ending point. Default: break.pts=NULL. cutoff second half distance classes, cutoff = TRUE limits correlogram distance classes include points. cutoff = FALSE, correlogram includes distance classes. r.type Type correlation calculation Mantel statistic. Default: r.type=\"pearson\". choices r.type=\"spearman\" r.type=\"kendall\", functions cor mantel. nperm Number permutations tests significance. Default: nperm=999. large data files, permutation tests rather slow. mult Correct P-values multiple testing. correction methods \"holm\" (default), \"hochberg\", \"sidak\", methods available p.adjust function: \"bonferroni\" (best known, recommended overly conservative), \"hommel\", \"BH\", \"\", \"fdr\", \"none\". progressive Default: progressive=TRUE progressive correction multiple-testing, described Legendre Legendre (1998, p. 721). Test first distance class: correction; second distance class: correct 2 simultaneous tests; distance class k: correct k simultaneous tests. progressive=FALSE: correct tests n.class simultaneous tests. x Output mantel.correlog. alpha Significance level points drawn black symbols correlogram. Default: alpha=0.05. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel Correlogram — mantel.correlog","text":"correlogram graph spatial correlation values plotted, ordinate, function geographic distance classes among study sites along abscissa. Mantel correlogram, Mantel correlation (Mantel 1967) computed multivariate (e.g. multi-species) distance matrix user's choice design matrix representing geographic distance classes turn. Mantel statistic tested permutational Mantel test performed vegan's mantel function. correction multiple testing applied, permutations necessary -correction case, obtain significant p-values higher correlogram classes. print.mantel.correlog function prints correlogram. See examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel Correlogram — mantel.correlog","text":"mantel.res table distance classes rows class indices, number distances per class, Mantel statistics (computed using Pearson's r, Spearman's r, Kendall's tau), p-values columns. positive Mantel statistic indicates positive spatial correlation. additional column p-values corrected multiple testing added unless mult=\"none\". n.class n umber distance classes. break.pts break points provided user computed program. mult name correction multiple testing. correction: mult=\"none\". progressive logical (TRUE, FALSE) value indicating whether progressive correction multiple testing requested. n.tests number distance classes Mantel tests computed tested significance. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel Correlogram — mantel.correlog","text":"Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel Correlogram — mantel.correlog","text":"Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Mantel, N. 1967. detection disease clustering generalized regression approach. Cancer Res. 27: 209-220. Oden, N. L. R. R. Sokal. 1986. Directional autocorrelation: extension spatial correlograms two dimensions. Syst. Zool. 35: 608-617. Sokal, R. R. 1986. Spatial data analysis historical processes. 29-43 : E. Diday et al. [eds.] Data analysis informatics, IV. North-Holland, Amsterdam. Sturges, H. . 1926. choice class interval. Journal American Statistical Association 21: 65–66.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel Correlogram — mantel.correlog","text":"","code":"# Mite data available in \"vegan\" data(mite) data(mite.xy) mite.hel <- decostand(mite, \"hellinger\") # Detrend the species data by regression on the site coordinates mite.hel.resid <- resid(lm(as.matrix(mite.hel) ~ ., data=mite.xy)) # Compute the detrended species distance matrix mite.hel.D <- dist(mite.hel.resid) # Compute Mantel correlogram with cutoff, Pearson statistic mite.correlog <- mantel.correlog(mite.hel.D, XY=mite.xy, nperm=49) summary(mite.correlog) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 4 -none- call mite.correlog #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.135713 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.118174 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.037820 0.04 0.06 . #> D.cl.4 2.699274 696.000000 -0.098605 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.112682 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.107603 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.022264 0.12 0.14 #> D.cl.8 5.612730 326.000000 NA NA NA #> D.cl.9 6.341094 260.000000 NA NA NA #> D.cl.10 7.069458 184.000000 NA NA NA #> D.cl.11 7.797822 130.000000 NA NA NA #> D.cl.12 8.526186 66.000000 NA NA NA #> D.cl.13 9.254550 32.000000 NA NA NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # or: print(mite.correlog) # or: print.mantel.correlog(mite.correlog) plot(mite.correlog) # Compute Mantel correlogram without cutoff, Spearman statistic mite.correlog2 <- mantel.correlog(mite.hel.D, XY=mite.xy, cutoff=FALSE, r.type=\"spearman\", nperm=49) summary(mite.correlog2) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 6 -none- call mite.correlog2 #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, cutoff = FALSE, r.type = \"spearman\", nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.134229 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.121270 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.035413 0.02 0.06 . #> D.cl.4 2.699274 696.000000 -0.095899 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.118692 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.117148 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.031123 0.08 0.14 #> D.cl.8 5.612730 326.000000 0.026064 0.14 0.16 #> D.cl.9 6.341094 260.000000 0.050573 0.08 0.24 #> D.cl.10 7.069458 184.000000 0.057017 0.04 0.20 #> D.cl.11 7.797822 130.000000 0.036195 0.10 0.32 #> D.cl.12 8.526186 66.000000 -0.054242 0.02 0.24 #> D.cl.13 9.254550 32.000000 -0.066677 0.04 0.26 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mite.correlog2) # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Function mantel finds Mantel statistic matrix correlation two dissimilarity matrices, function mantel.partial finds partial Mantel statistic partial matrix correlation three dissimilarity matrices. significance statistic evaluated permuting rows columns first dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"mantel(xdis, ydis, method=\"pearson\", permutations=999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\")) mantel.partial(xdis, ydis, zdis, method = \"pearson\", permutations = 999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"xdis, ydis, zdis Dissimilarity matrices ordist objects. first object xdis permuted permutation tests. method Correlation method, accepted cor: \"pearson\", \"spearman\" \"kendall\". permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. na.rm Remove missing values calculation Mantel correlation. Use option care: Permutation tests can biased, particular two matrices missing values matching positions. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Mantel statistic simply correlation entries two dissimilarity matrices (use cross products, linearly related). However, significance directly assessed, \\(N(N-1)/2\\) entries just \\(N\\) observations. Mantel developed asymptotic test, use permutations \\(N\\) rows columns dissimilarity matrix. first matrix (xdist) permuted, second kept constant. See permutations additional details permutation tests Vegan. Partial Mantel statistic uses partial correlation conditioned third matrix. first matrix permuted correlation structure second first matrices kept constant. Although mantel.partial silently accepts methods \"pearson\", partial correlations probably wrong methods. function uses cor, accept alternatives pearson product moment correlations spearman kendall rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"function returns list class mantel following components: Call Function call. method Correlation method used, returned cor.test. statistic Mantel statistic. signif Empirical significance level permutations. perm vector permuted values. distribution permuted values can inspected permustats function. permutations Number permutations. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"test due Mantel, course, current implementation based Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English Edition. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Legendre & Legendre (2012, Box 10.4) warn using partial Mantel correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"## Is vegetation related to environment? data(varespec) data(varechem) veg.dist <- vegdist(varespec) # Bray-Curtis env.dist <- vegdist(scale(varechem), \"euclid\") mantel(veg.dist, env.dist) #> #> Mantel statistic based on Pearson's product-moment correlation #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist) #> #> Mantel statistic r: 0.3047 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.116 0.147 0.176 0.210 #> Permutation: free #> Number of permutations: 999 #> mantel(veg.dist, env.dist, method=\"spear\") #> #> Mantel statistic based on Spearman's rank correlation rho #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist, method = \"spear\") #> #> Mantel statistic r: 0.2838 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.123 0.162 0.182 0.208 #> Permutation: free #> Number of permutations: 999 #>"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS performs Nonmetric Multidimensional Scaling (NMDS), tries find stable solution using several random starts. addition, standardizes scaling result, configurations easier interpret, adds species scores site ordination. metaMDS function provide actual NMDS, calls another function purpose. Currently monoMDS default choice, also possible call isoMDS (MASS package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"metaMDS(comm, distance = \"bray\", k = 2, try = 20, trymax = 20, engine = c(\"monoMDS\", \"isoMDS\"), autotransform =TRUE, noshare = (engine == \"isoMDS\"), wascores = TRUE, expand = TRUE, trace = 1, plot = FALSE, previous.best, ...) # S3 method for metaMDS plot(x, display = c(\"sites\", \"species\"), choices = c(1, 2), type = \"p\", shrink = FALSE, ...) # S3 method for metaMDS points(x, display = c(\"sites\", \"species\"), choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS text(x, display = c(\"sites\", \"species\"), labels, choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS scores(x, display = c(\"sites\", \"species\"), shrink = FALSE, choices, tidy = FALSE, ...) metaMDSdist(comm, distance = \"bray\", autotransform = TRUE, noshare = TRUE, trace = 1, commname, zerodist = \"ignore\", distfun = vegdist, ...) metaMDSiter(dist, k = 2, try = 20, trymax = 20, trace = 1, plot = FALSE, previous.best, engine = \"monoMDS\", maxit = 200, parallel = getOption(\"mc.cores\"), ...) initMDS(x, k=2) postMDS(X, dist, pc=TRUE, center=TRUE, halfchange, threshold=0.8, nthreshold=10, plot=FALSE, ...) metaMDSredist(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"comm Community data. Alternatively, dissimilarities either dist structure symmetric square matrix. latter case stages skipped except random starts centring pc rotation axes. distance Dissimilarity index used vegdist. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher global non-metric MDS, still higher local NMDS. try, trymax Minimum maximum numbers random starts search stable solution. try reached, iteration stop similar solutions repeated trymax reached. engine function used MDS. default use monoMDS function vegan, backward compatibility also possible use isoMDS MASS. autotransform Use simple heuristics possible data transformation typical community data (see ). community data, probably set autotransform = FALSE. noshare Triggering calculation step-across extended dissimilarities function stepacross. argument can logical numerical value greater zero less one. TRUE, extended dissimilarities used always shared species sites, FALSE, never used. noshare numerical value, stepacross used proportion site pairs shared species exceeds noshare. number pairs shared species found .shared function, noshare effect input data dissimilarities instead community data. wascores Calculate species scores using function wascores. expand Expand weighted averages species wascores. trace Trace function; trace = 2 higher voluminous. plot Graphical tracing: plot interim results. may want set par(ask = TRUE) option. previous.best Start searches previous solution. x metaMDS result (dissimilarity structure initMDS). choices Axes shown. type Plot type: \"p\" points, \"t\" text, \"n\" axes . display Display \"sites\" \"species\". shrink Shrink back species scores expanded originally. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable code (\"sites\" \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Optional test used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. X Configuration multidimensional scaling. commname name comm: given function called directly. zerodist Handling zero dissimilarities: either \"fail\" \"add\" small positive value, \"ignore\". monoMDS accepts zero dissimilarities default zerodist = \"ignore\", isoMDS may need set zerodist = \"add\". distfun Dissimilarity function. function returning dist object accepting argument method can used (extra arguments may cause name conflicts). maxit Maximum number iterations single NMDS run; passed engine function monoMDS isoMDS. parallel Number parallel processes predefined socket cluster. use pre-defined socket clusters (say, clus), must issue clusterEvalQ(clus, library(vegan)) make available internal vegan functions. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist Dissimilarity matrix used multidimensional scaling. pc Rotate principal components. center Centre configuration. halfchange Scale axes half-change units. defaults TRUE dissimilarities known theoretical maximum value (ceiling). Function vegdist information attribute maxdist, distfun interpreted simple test (can fail), information may available input data distances. FALSE, ordination dissimilarities scaled range input dissimilarities. threshold Largest dissimilarity used half-change scaling. dissimilarities known (inferred) ceiling, threshold relative ceiling (see halfchange). nthreshold Minimum number points half-change scaling. object result object metaMDS. ... parameters passed functions. Function metaMDS passes arguments component functions metaMDSdist, metaMDSiter, postMDS, distfun engine.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-metric Multidimensional Scaling (NMDS) commonly regarded robust unconstrained ordination method community ecology (Minchin 1987). Function metaMDS wrapper function calls several functions combine Minchin's (1987) recommendations one command. complete steps metaMDS : Transformation: data values larger common abundance class scales, function performs Wisconsin double standardization (wisconsin). values look large, function also performs sqrt transformation. standardizations generally found improve results. However, limits completely arbitrary (present, data maximum 50 triggers sqrt \\(>9\\) triggers wisconsin). want full control analysis, set autotransform = FALSE standardize transform data independently. autotransform intended community data, data types, set autotransform = FALSE. step perfomed using metaMDSdist, step skipped input dissimilarities. Choice dissimilarity: good result, use dissimilarity indices good rank order relation ordering sites along gradients (Faith et al. 1987). default Bray-Curtis dissimilarity, often test winner. However, dissimilarity index vegdist can used. Function rankindex can used finding test winner data gradients. default choice may bad analyse community data, probably select appropriate index using argument distance. step performed using metaMDSdist, step skipped input dissimilarities. Step-across dissimilarities: Ordination may difficult large proportion sites shared species. case, results may improved stepacross dissimilarities, flexible shortest paths among sites. default NMDS engine monoMDS able break tied values maximum dissimilarity, often sufficient handle cases shared species, therefore default use stepacross monoMDS. Function isoMDS handle tied values adequately, therefore default use stepacross always sites shared species engine = \"isoMDS\". stepacross triggered option noshare. like manipulation original distances, set noshare = FALSE. step skipped input data dissimilarities instead community data. step performed using metaMDSdist, step skipped always input dissimilarities. NMDS random starts: NMDS easily gets trapped local optima, must start NMDS several times random starts confident found global solution. strategy metaMDS first run NMDS starting metric scaling (cmdscale usually finds good solution often close local optimum), use previous.best solution supplied, take solution standard (Run 0). metaMDS starts NMDS several random starts (minimum number given try maximum number trymax). random starts generated initMDS. solution better (lower stress) previous standard, taken new standard. solution better close standard, metaMDS compares two solutions using Procrustes analysis (function procrustes option symmetric = TRUE). solutions similar Procrustes rmse largest residual small, solutions regarded repeated better one taken new standard. conditions stringent, may found good relatively similar solutions although function yet satisfied. Setting trace = TRUE monitor final stresses, plot = TRUE display Procrustes overlay plots comparison. step performed using metaMDSiter. first step performed input data (comm) dissimilarities. Random starts can run parallel processing (argument parallel). Scaling results: metaMDS run postMDS final result. Function postMDS provides following ways “fixing” indeterminacy scaling orientation axes NMDS: Centring moves origin average axes; Principal components rotate configuration variance points maximized first dimension (function MDSrotate can alternatively rotate configuration first axis parallel environmental variable); Half-change scaling scales configuration one unit means halving community similarity replicate similarity. Half-change scaling based closer dissimilarities relation ordination distance community dissimilarity rather linear (limit set argument threshold). enough points threshold (controlled parameter nthreshold), dissimilarities regressed distances. intercept regression taken replicate dissimilarity, half-change distance similarity halves according linear regression. Obviously method applicable dissimilarity indices scaled \\(0 \\ldots 1\\), Kulczynski, Bray-Curtis Canberra indices. half-change scaling used, ordination scaled range original dissimilarities. Half-change scaling skipped default input dissimilarities, can turned argument halfchange = TRUE. NB., PC rotation changes directions reference axes, influence configuration solution general. Species scores: Function adds species scores final solution weighted averages using function wascores given value parameter expand. expansion weighted averages can undone shrink = TRUE plot scores functions, calculation species scores can suppressed wascores = FALSE. step skipped input dissimilarities community data unavailable. However, species scores can added replaced sppscores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"results-could-not-be-repeated","dir":"Reference","previous_headings":"","what":"Results Could Not Be Repeated","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-linear optimization hard task, best possible solution (“global optimum”) may found random starting configuration. software solve starting result metric scaling (cmdscale). probably give good result, necessarily “global optimum”. Vegan , metaMDS tries verify improve first solution (“try 0”) using several random starts seeing result can repeated improved improved solution repeated. succeed, get message result repeated. However, result least good usual standard strategy starting metric scaling may improved. may need anything message, can satisfied result. want sure probably “global optimum” may try following instructions. default engine = \"monoMDS\" function tabulate stopping criteria used, can see criterion made stringent. criteria can given arguments metaMDS current values described monoMDS. particular, reach maximum number iterations, increase value maxit. may ask larger number random starts without losing old ones giving previous solution argument previous.best. addition slack convergence criteria low number random starts, wrong number dimensions (argument k) common reason able repeat similar solutions. NMDS usually run low number dimensions (k=2 k=3), complex data increasing k one may help. run NMDS much higher number dimensions (say, k=10 ), reconsider drastically reduce k. heterogeneous data sets partial disjunctions, may help set stepacross, data sets default weakties = TRUE sufficient. Please note can give arguments metaMDS* functions NMDS engine (default monoMDS) metaMDS command, check documentation functions details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"common-wrong-claims","dir":"Reference","previous_headings":"","what":"Common Wrong Claims","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"NMDS often misunderstood wrong claims properties common Web even publications. often claimed NMDS configuration non-metric means fit environmental variables species onto space. false statement. fact, result configuration NMDS metric, can used like ordination result. NMDS rank orders Euclidean distances among points ordination non-metric monotone relationship observed dissimilarities. transfer function observed dissimilarities ordination distances non-metric (Kruskal 1964a, 1964b), ordination result configuration metric observed dissimilarities can kind (metric non-metric). ordination configuration usually rotated principal components metaMDS. rotation performed finding result, changes direction reference axes. important feature NMDS solution ordination distances, change rotation. Similarly, rank order distances change uniform scaling centring configuration points. can also rotate NMDS solution external environmental variables MDSrotate. rotation also change orientation axes, change configuration points distances points ordination space. Function stressplot displays method graphically: plots observed dissimilarities distances ordination space, also shows non-metric monotone regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS returns object class metaMDS. final site ordination stored item points, species ordination item species, stress item stress (NB, scaling stress depends engine: isoMDS uses percents, monoMDS proportions range \\(0 \\ldots 1\\)). items store information steps taken items returned engine function. object print, plot, points text methods. Functions metaMDSdist metaMDSredist return vegdist objects. Function initMDS returns random configuration intended used within isoMDS . Functions metaMDSiter postMDS returns result NMDS updated configuration.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Kruskal, J.B. (1964a). Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. (1964b). Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. (1987). evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS simple wrapper NMDS engine (either monoMDS isoMDS) support functions (metaMDSdist, stepacross, metaMDSiter, initMDS, postMDS, wascores). can call support functions separately better control results. Data transformation, dissimilarities possible stepacross made function metaMDSdist returns dissimilarity result. Iterative search ( starting values initMDS monoMDS) made metaMDSiter. Processing result configuration done postMDS, species scores added wascores. want certain reaching global solution, can compare results several independent runs. can also continue analysis previous results configuration. Function may save used dissimilarity matrix (monoMDS ), metaMDSredist tries reconstruct used dissimilarities original data transformation possible stepacross. metaMDS function designed used community data. type data, probably set arguments non-default values: probably least wascores, autotransform noshare FALSE. negative data entries, metaMDS set previous FALSE warning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"metaMDS uses monoMDS NMDS engine vegan version 2.0-0, replaced isoMDS function. can set argument engine select old engine.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"## The recommended way of running NMDS (Minchin 1987) ## data(dune) ## IGNORE_RDIFF_BEGIN ## Global NMDS using monoMDS sol <- metaMDS(dune) #> Run 0 stress 0.1192678 #> Run 1 stress 0.3016052 #> Run 2 stress 0.1192678 #> ... New best solution #> ... Procrustes: rmse 9.804396e-06 max resid 2.993113e-05 #> ... Similar to previous best #> Run 3 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 0.02027462 max resid 0.06499128 #> Run 4 stress 0.1192679 #> Run 5 stress 0.1183186 #> ... Procrustes: rmse 3.964007e-06 max resid 1.2573e-05 #> ... Similar to previous best #> Run 6 stress 0.1192679 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 6.675204e-06 max resid 1.892118e-05 #> ... Similar to previous best #> Run 8 stress 0.1192678 #> Run 9 stress 0.1183186 #> ... Procrustes: rmse 2.31839e-06 max resid 6.973206e-06 #> ... Similar to previous best #> Run 10 stress 0.1889645 #> Run 11 stress 0.1183186 #> ... Procrustes: rmse 4.550156e-06 max resid 1.4618e-05 #> ... Similar to previous best #> Run 12 stress 0.1192678 #> Run 13 stress 0.1192678 #> Run 14 stress 0.1192678 #> Run 15 stress 0.1192678 #> Run 16 stress 0.1192679 #> Run 17 stress 0.1808912 #> Run 18 stress 0.1183186 #> ... Procrustes: rmse 1.954831e-05 max resid 6.115607e-05 #> ... Similar to previous best #> Run 19 stress 0.1808911 #> Run 20 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 1.3197e-06 max resid 3.321941e-06 #> ... Similar to previous best #> *** Best solution repeated 1 times sol #> #> Call: #> metaMDS(comm = dune) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1183186 #> Stress type 1, weak ties #> Best solution was repeated 1 time in 20 tries #> The best solution was from try 20 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> plot(sol, type=\"t\") ## Start from previous best solution sol <- metaMDS(dune, previous.best = sol) #> Starting from 2-dimensional configuration #> Run 0 stress 0.1183186 #> Run 1 stress 0.1183186 #> ... Procrustes: rmse 5.279158e-06 max resid 1.797511e-05 #> ... Similar to previous best #> Run 2 stress 0.1183186 #> ... Procrustes: rmse 2.460953e-06 max resid 5.877957e-06 #> ... Similar to previous best #> Run 3 stress 0.1192678 #> Run 4 stress 0.1886532 #> Run 5 stress 0.1192678 #> Run 6 stress 0.1808911 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 7.828309e-06 max resid 2.405585e-05 #> ... Similar to previous best #> Run 8 stress 0.1183186 #> ... Procrustes: rmse 6.32253e-06 max resid 2.037397e-05 #> ... Similar to previous best #> Run 9 stress 0.1808911 #> Run 10 stress 0.1183186 #> ... Procrustes: rmse 1.809694e-06 max resid 5.978548e-06 #> ... Similar to previous best #> Run 11 stress 0.1901491 #> Run 12 stress 0.1183186 #> ... Procrustes: rmse 1.546268e-05 max resid 4.957148e-05 #> ... Similar to previous best #> Run 13 stress 0.1812932 #> Run 14 stress 0.1183186 #> ... Procrustes: rmse 5.112711e-06 max resid 1.098817e-05 #> ... Similar to previous best #> Run 15 stress 0.1183186 #> ... Procrustes: rmse 6.246131e-06 max resid 2.000824e-05 #> ... Similar to previous best #> Run 16 stress 0.1192679 #> Run 17 stress 0.1192678 #> Run 18 stress 0.1886532 #> Run 19 stress 0.1192679 #> Run 20 stress 0.1192678 #> *** Best solution repeated 9 times ## Local NMDS and stress 2 of monoMDS sol2 <- metaMDS(dune, model = \"local\", stress=2) #> Run 0 stress 0.1928478 #> Run 1 stress 0.1928482 #> ... Procrustes: rmse 0.0006537695 max resid 0.001899655 #> ... Similar to previous best #> Run 2 stress 0.1928476 #> ... New best solution #> ... Procrustes: rmse 0.0004471422 max resid 0.001294366 #> ... Similar to previous best #> Run 3 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 0.0002885977 max resid 0.0008320155 #> ... Similar to previous best #> Run 4 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 5.59458e-05 max resid 0.0001616763 #> ... Similar to previous best #> Run 5 stress 0.1928475 #> ... Procrustes: rmse 4.864069e-05 max resid 0.0001571056 #> ... Similar to previous best #> Run 6 stress 0.1928477 #> ... Procrustes: rmse 0.0002420844 max resid 0.0006909829 #> ... Similar to previous best #> Run 7 stress 0.1928475 #> ... Procrustes: rmse 0.0001346077 max resid 0.0003887498 #> ... Similar to previous best #> Run 8 stress 0.1928475 #> ... Procrustes: rmse 0.0001625818 max resid 0.0004729866 #> ... Similar to previous best #> Run 9 stress 0.1928475 #> ... Procrustes: rmse 8.000053e-05 max resid 0.0002277773 #> ... Similar to previous best #> Run 10 stress 0.192849 #> ... Procrustes: rmse 0.0006301494 max resid 0.001813666 #> ... Similar to previous best #> Run 11 stress 0.1928475 #> ... Procrustes: rmse 0.0001014964 max resid 0.0002751379 #> ... Similar to previous best #> Run 12 stress 0.192848 #> ... Procrustes: rmse 0.0002964116 max resid 0.0008527077 #> ... Similar to previous best #> Run 13 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 6.461832e-05 max resid 0.0001901839 #> ... Similar to previous best #> Run 14 stress 0.1928482 #> ... Procrustes: rmse 0.0003439501 max resid 0.001017732 #> ... Similar to previous best #> Run 15 stress 0.1928479 #> ... Procrustes: rmse 0.0002821687 max resid 0.0008172892 #> ... Similar to previous best #> Run 16 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 3.13617e-05 max resid 8.909983e-05 #> ... Similar to previous best #> Run 17 stress 0.1928475 #> ... Procrustes: rmse 4.709197e-05 max resid 0.0001417594 #> ... Similar to previous best #> Run 18 stress 0.1928477 #> ... Procrustes: rmse 0.0002414341 max resid 0.0007100825 #> ... Similar to previous best #> Run 19 stress 0.1928479 #> ... Procrustes: rmse 0.0003023992 max resid 0.0008897382 #> ... Similar to previous best #> Run 20 stress 0.1928476 #> ... Procrustes: rmse 0.0002063278 max resid 0.0006048887 #> ... Similar to previous best #> *** Best solution repeated 5 times sol2 #> #> Call: #> metaMDS(comm = dune, model = \"local\", stress = 2) #> #> local Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1928475 #> Stress type 2, weak ties #> Best solution was repeated 5 times in 20 tries #> The best solution was from try 16 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## Use Arrhenius exponent 'z' as a binary dissimilarity measure sol <- metaMDS(dune, distfun = betadiver, distance = \"z\") #> Run 0 stress 0.1067169 #> Run 1 stress 0.1073148 #> Run 2 stress 0.1073148 #> Run 3 stress 0.1073148 #> Run 4 stress 0.1073148 #> Run 5 stress 0.1073148 #> Run 6 stress 0.1067169 #> ... New best solution #> ... Procrustes: rmse 2.925047e-06 max resid 7.049929e-06 #> ... Similar to previous best #> Run 7 stress 0.1069781 #> ... Procrustes: rmse 0.006690411 max resid 0.02349817 #> Run 8 stress 0.1067169 #> ... Procrustes: rmse 6.432458e-06 max resid 1.886945e-05 #> ... Similar to previous best #> Run 9 stress 0.1069786 #> ... Procrustes: rmse 0.006774522 max resid 0.02386629 #> Run 10 stress 0.1067169 #> ... Procrustes: rmse 2.346391e-06 max resid 5.306599e-06 #> ... Similar to previous best #> Run 11 stress 0.1067169 #> ... Procrustes: rmse 2.816635e-06 max resid 7.563776e-06 #> ... Similar to previous best #> Run 12 stress 0.1069784 #> ... Procrustes: rmse 0.006735205 max resid 0.02369698 #> Run 13 stress 0.1067169 #> ... Procrustes: rmse 3.36442e-06 max resid 1.127286e-05 #> ... Similar to previous best #> Run 14 stress 0.2355185 #> Run 15 stress 0.1644741 #> Run 16 stress 0.1067169 #> ... Procrustes: rmse 8.389288e-06 max resid 2.174836e-05 #> ... Similar to previous best #> Run 17 stress 0.1073148 #> Run 18 stress 0.1067169 #> ... Procrustes: rmse 5.169725e-06 max resid 1.633663e-05 #> ... Similar to previous best #> Run 19 stress 0.1067169 #> ... Procrustes: rmse 3.667758e-06 max resid 9.091241e-06 #> ... Similar to previous best #> Run 20 stress 0.1073148 #> *** Best solution repeated 8 times sol #> #> Call: #> metaMDS(comm = dune, distance = \"z\", distfun = betadiver) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: beta.z #> #> Dimensions: 2 #> Stress: 0.1067169 #> Stress type 1, weak ties #> Best solution was repeated 8 times in 20 tries #> The best solution was from try 6 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":null,"dir":"Reference","previous_headings":"","what":"Oribatid Mite Data with Explanatory Variables — mite","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Oribatid mite data. 70 soil cores collected Daniel Borcard 1989. See Borcard et al. (1992, 1994) details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) data(mite.xy)"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"three linked data sets: mite contains data 35 species Oribatid mites, mite.env contains environmental data sampling sites, mite.xy contains geographic coordinates, mite.pcnm contains 22 PCNM base functions (columns) computed geographic coordinates 70 sampling sites (Borcard & Legendre 2002). whole sampling area 2.5 m x 10 m size. fields environmental data : SubsDens Substrate density (g/L) WatrCont Water content substrate (g/L) Substrate Substrate type, factor levels Sphagn1, \tSphagn2 Sphagn3 Sphagn Litter Barepeat Interface Shrub Shrub density, ordered factor levels 1 < 2 < 3 Topo Microtopography, factor levels Blanket Hummock","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Pierre Legendre","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Borcard, D., P. Legendre P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045-1055. Borcard, D. P. Legendre. 1994. Environmental control spatial structure ecological communities: example using Oribatid mites (Acari, Oribatei). Environmental Ecological Statistics 1: 37-61. Borcard, D. P. Legendre. 2002. -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153: 51-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Function implements Kruskal's (1964a,b) non-metric multidimensional scaling (NMDS) using monotone regression primary (“weak”) treatment ties. addition traditional global NMDS, function implements local NMDS, linear hybrid multidimensional scaling.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"monoMDS(dist, y, k = 2, model = c(\"global\", \"local\", \"linear\", \"hybrid\"), threshold = 0.8, maxit = 200, weakties = TRUE, stress = 1, scaling = TRUE, pc = TRUE, smin = 1e-4, sfgrmin = 1e-7, sratmax=0.999999, ...) # S3 method for monoMDS scores(x, choices = NA, ...) # S3 method for monoMDS plot(x, choices = c(1,2), type = \"t\", ...) # S3 method for monoMDS points(x, choices = c(1,2), select, ...) # S3 method for monoMDS text(x, labels, choices = c(1,2), select, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"dist Input dissimilarities. y Starting configuration. random configuration generated missing. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher non-metric MDS. model MDS model: \"global\" normal non-metric MDS monotone regression, \"local\" non-metric MDS separate regressions point, \"linear\" uses linear regression, \"hybrid\" uses linear regression dissimilarities threshold addition monotone regression. See Details. threshold Dissimilarity linear regression used alternately monotone regression. maxit Maximum number iterations. weakties Use primary weak tie treatment, equal observed dissimilarities allowed different fitted values. FALSE, secondary (strong) tie treatment used, tied values broken. stress Use stress type 1 2 (see Details). scaling Scale final scores unit root mean squares. pc Rotate final scores principal components. smin, sfgrmin, sratmax Convergence criteria: iterations stop stress drops smin, scale factor gradient drops sfgrmin, stress ratio two iterations goes sratmax (still \\(< 1\\)). x monoMDS result. choices Dimensions returned plotted. default NA returns dimensions. type type plot: \"t\" text, \"p\" points, \"n\" none. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. labels Labels use used instead row names. ... parameters functions (ignored monoMDS, passed graphical functions plot.).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"several versions non-metric multidimensional scaling R, monoMDS offers following unique combination features: “Weak” treatment ties (Kruskal 1964a,b), tied dissimilarities can broken monotone regression. especially important cases compared sites share species dissimilarities tied maximum value one. Breaking ties allows points different distances can help recovering long coenoclines (gradients). Functions smacof package also hav adequate tie treatment. Handles missing values meaningful way. Offers “local” “hybrid” scaling addition usual “global” NMDS (see ). Uses fast compiled code (isoMDS MASS package also uses compiled code). Function monoMDS uses Kruskal's (1964b) original monotone regression minimize stress. two alternatives stress: Kruskal's (1964a,b) original “stress 1” alternative version “stress 2” (Sibson 1972). stresses can expressed general formula $$s^2 = \\frac{\\sum (d - \\hat d)^2}{\\sum(d - d_0)^2}$$ \\(d\\) distances among points ordination configuration, \\(\\hat d\\) fitted ordination distances, \\(d_0\\) ordination distances null model. “stress 1” \\(d_0 = 0\\), “stress 2” \\(d_0 = \\bar{d}\\) mean distances. “Stress 2” can expressed \\(s^2 = 1 - R^2\\), \\(R^2\\) squared correlation fitted values ordination distances, related “linear fit” stressplot. Function monoMDS can fit several alternative NMDS variants can selected argument model. default model = \"global\" fits global NMDS, Kruskal's (1964a,b) original NMDS similar isoMDS (MASS). Alternative model = \"local\" fits local NMDS independent monotone regression used point (Sibson 1972). Alternative model = \"linear\" fits linear MDS. fits linear regression instead monotone, identical metric scaling principal coordinates analysis (cmdscale) performs eigenvector decomposition dissimilarities (Gower 1966). Alternative model = \"hybrid\" implements hybrid MDS uses monotone regression points linear regression dissimilarities threshold dissimilarity alternating steps (Faith et al. 1987). Function stressplot can used display kind regression model. Scaling, orientation direction axes arbitrary. However, function always centres axes, default scaling scale configuration unit root mean square rotate axes (argument pc) principal components first dimension shows major variation. possible rotate solution first axis parallel given environmental variable using function MDSrotate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"convergence-criteria","dir":"Reference","previous_headings":"","what":"Convergence Criteria","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"NMDS iterative, function stops convergence criteria met. actually criterion assured convergence, solution can local optimum. compare several random starts (use monoMDS via metaMDS) assess solutions likely global optimum. stopping criteria : maxit: Maximum number iterations. Reaching criterion means solutions almost certainly found, maxit increased. smin: Minimum stress. stress nearly zero, fit almost perfect. Usually means data set small requested analysis, may several different solutions almost perfect. reduce number dimensions (k), get data ( observations) use method, metric scaling (cmdscale, wcmdscale). sratmax: Change stress. Values close one mean almost unchanged stress. may mean solution, can also signal stranding suboptimal solution flat stress surface. sfgrmin: Minimum scale factor. Values close zero mean almost unchanged configuration. may mean solution, also happen local optima.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Returns object class \"monoMDS\". final scores returned item points (function scores extracts results), stress item stress. addition, large number items (may change without notice future releases).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Faith, D.P., Minchin, P.R Belbin, L. 1987. Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J.C. (1966). distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Kruskal, J.B. 1964a. Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. 1964b. Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. 1987. evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107. Sibson, R. 1972. Order invariant methods data analysis. Journal Royal Statistical Society B 34, 311--349.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Peter R. Michin (Fortran core) Jari Oksanen (R interface).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"default NMDS function used metaMDS. Function metaMDS adds support functions NMDS can run like recommended Minchin (1987).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"data(dune) dis <- vegdist(dune) m <- monoMDS(dis, model = \"loc\") m #> #> Call: #> monoMDS(dist = dis, model = \"loc\") #> #> Local non-metric Multidimensional Scaling #> #> 20 points, dissimilarity ‘bray’, call ‘vegdist(x = dune)’ #> #> Dimensions: 2 #> Stress: 0.07596745 #> Stress type 1, weak ties #> Scores scaled to unit root mean square, rotated to principal components #> Stopped after 62 iterations: Stress nearly unchanged (ratio > sratmax) plot(m)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":null,"dir":"Reference","previous_headings":"","what":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. Function meandist finds mean within block dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"mrpp(dat, grouping, permutations = 999, distance = \"euclidean\", weight.type = 1, strata = NULL, parallel = getOption(\"mc.cores\")) meandist(dist, grouping, ...) # S3 method for meandist summary(object, ...) # S3 method for meandist plot(x, kind = c(\"dendrogram\", \"histogram\"), cluster = \"average\", ylim, axes = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"dat data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor numeric index grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. used assess significance MRPP statistic, \\(delta\\). distance Choice distance metric measures dissimilarity two observations . See vegdist options. used dat dissimilarity structure symmetric square matrix. weight.type choice group weights. See Details options. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist dist object dissimilarities, produced functions dist, vegdist designdist. . object, x meandist result object. kind Draw dendrogram histogram; see Details. cluster clustering method hclust function kind = \"dendrogram\". hclust method can used, perhaps \"average\" \"single\" make sense. ylim Limits vertical axes (optional). axes Draw scale vertical axis. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. difference may one location (differences mean) one spread (differences within-group distance; cf. Warton et al. 2012). Function mrpp operates data.frame matrix rows observations responses data matrix. response(s) may uni- multivariate. method philosophically mathematically allied analysis variance, compares dissimilarities within among groups. two groups sampling units really different (e.g. species composition), average within-group compositional dissimilarities less average dissimilarities two random collection sampling units drawn entire population. mrpp statistic \\(\\delta\\) overall weighted mean within-group means pairwise dissimilarities among sampling units. choice group weights currently clear. mrpp function offers three choices: (1) group size (\\(n\\)), (2) degrees--freedom analogue (\\(n-1\\)), (3) weight number unique distances calculated among \\(n\\) sampling units (\\(n(n-1)/2\\)). mrpp algorithm first calculates pairwise distances entire dataset, calculates \\(\\delta\\). permutes sampling units associated pairwise distances, recalculates \\(\\delta\\) based permuted data. repeats permutation step permutations times. significance test fraction permuted deltas less observed delta, small sample correction. function also calculates change-corrected within-group agreement \\(= 1 -\\delta/E(\\delta)\\), \\(E(\\delta)\\) expected \\(\\delta\\) assessed average dissimilarities. first argument dat can interpreted dissimilarities, used directly. cases function treats dat observations, uses vegdist find dissimilarities. default distance Euclidean traditional use method, dissimilarities vegdist also available. Function meandist calculates matrix mean within-cluster dissimilarities (diagonal) -cluster dissimilarities (-diagonal elements), attribute n grouping counts. Function summary finds within-class, -class overall means dissimilarities, MRPP statistics weight.type options Classification Strength, CS (Van Sickle Hughes, 2000). CS defined dissimilarities \\(\\bar{B} - \\bar{W}\\), \\(\\bar{B}\\) mean cluster dissimilarity \\(\\bar{W}\\) mean within cluster dissimilarity weight.type = 1. function perform significance tests statistics, must use mrpp appropriate weight.type. currently significance test CS, mrpp weight.type = 1 gives correct test \\(\\bar{W}\\) good approximation CS. Function plot draws dendrogram histogram result matrix based within-group group dissimilarities. dendrogram found method given cluster argument using function hclust. terminal segments hang within-cluster dissimilarity. clusters heterogeneous combined class, leaf segment reversed. histograms based dissimilarities, ore otherwise similar Van Sickle Hughes (2000): horizontal line drawn level mean -cluster dissimilarity vertical lines connect within-cluster dissimilarities line.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"function returns list class mrpp following items: call Function call. delta overall weighted mean group mean distances. E.delta expected delta, null hypothesis group structure. mean original dissimilarities. CS Classification strength (Van Sickle Hughes, 2000). Currently implemented always NA. n Number observations class. classdelta Mean dissimilarities within classes. overall \\(\\delta\\) weighted average values given weight.type . Pvalue Significance test. chance-corrected estimate proportion distances explained group identity; value analogous coefficient determination linear model. distance Choice distance metric used; \"method\" entry dist object. weight.type choice group weights used. boot.deltas vector \"permuted deltas,\" deltas calculated permuted datasets. distribution item can inspected permustats function. permutations number permutations used. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"B. McCune J. B. Grace. 2002. Analysis Ecological Communities. MjM Software Design, Gleneden Beach, Oregon, USA. P. W. Mielke K. J. Berry. 2001. Permutation Methods: Distance Function Approach. Springer Series Statistics. Springer. J. Van Sickle R. M. Hughes 2000. Classification strengths ecoregions, catchments, geographic clusters aquatic vertebrates Oregon. J. N. . Benthol. Soc. 19:370--384. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"M. Henry H. Stevens HStevens@muohio.edu Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"difference may one location (differences mean) one spread (differences within-group distance). , may find significant difference two groups simply one groups greater dissimilarities among sampling units. mrpp models can analysed adonis2 seems suffer problems mrpp robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"data(dune) data(dune.env) dune.mrpp <- with(dune.env, mrpp(dune, Management)) dune.mrpp #> #> Call: #> mrpp(dat = dune, grouping = Management) #> #> Dissimilarity index: euclidean #> Weights for groups: n #> #> Class means and counts: #> #> BF HF NM SF #> delta 10.03 11.08 10.66 12.27 #> n 3 5 6 6 #> #> Chance corrected within-group agreement A: 0.1246 #> Based on observed delta 11.15 and expected delta 12.74 #> #> Significance of delta: 0.001 #> Permutation: free #> Number of permutations: 999 #> # Save and change plotting parameters def.par <- par(no.readonly = TRUE) layout(matrix(1:2,nr=1)) plot(dune.ord <- metaMDS(dune, trace=0), type=\"text\", display=\"sites\" ) with(dune.env, ordihull(dune.ord, Management)) with(dune.mrpp, { fig.dist <- hist(boot.deltas, xlim=range(c(delta,boot.deltas)), main=\"Test of Differences Among Groups\") abline(v=delta); text(delta, 2*mean(fig.dist$counts), adj = -0.5, expression(bold(delta)), cex=1.5 ) } ) par(def.par) ## meandist dune.md <- with(dune.env, meandist(vegdist(dune), Management)) dune.md #> BF HF NM SF #> BF 0.4159972 0.4736637 0.7296979 0.6247169 #> HF 0.4736637 0.4418115 0.7217933 0.5673664 #> NM 0.7296979 0.7217933 0.6882438 0.7723367 #> SF 0.6247169 0.5673664 0.7723367 0.5813015 #> attr(,\"class\") #> [1] \"meandist\" \"matrix\" #> attr(,\"n\") #> grouping #> BF HF NM SF #> 3 5 6 6 summary(dune.md) #> #> Mean distances: #> Average #> within groups 0.5746346 #> between groups 0.6664172 #> overall 0.6456454 #> #> Summary statistics: #> Statistic #> MRPP A weights n 0.1423836 #> MRPP A weights n-1 0.1339124 #> MRPP A weights n(n-1) 0.1099842 #> Classification strength 0.1127012 plot(dune.md) plot(dune.md, kind=\"histogram\")"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":null,"dir":"Reference","previous_headings":"","what":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso adds attribute vario object class \"cca\" describes spatial partitioning cca object performs optional permutation test spatial independence residuals. function plot.mso creates diagnostic plot spatial partitioning \"cca\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"mso(object.cca, object.xy, grain = 1, round.up = FALSE, permutations = 0) msoplot(x, alpha = 0.05, explained = FALSE, ylim = NULL, legend = \"topleft\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"object.cca object class cca, created cca rda function. object.xy vector, matrix data frame spatial coordinates data represented object.cca. number rows must match number observations (given nobs) cca.object. Alternatively, interpoint distances can supplied dist object. grain Interval size distance classes. round.Determines choice breaks. false, distances rounded nearest multiple grain. true, distances rounded upper multiple grain. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x result object mso. alpha Significance level two-sided permutation test Mantel statistic spatial independence residual inertia point-wise envelope variogram total variance. Bonferroni-type correction can achieved dividing overall significance value (e.g. 0.05) number distance classes. explained false, suppresses plotting variogram explained variance. ylim Limits y-axis. legend x y co-ordinates used position legend. can specified keyword way accepted legend. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Mantel test adaptation function mantel vegan package parallel testing several distance classes. compares mean inertia distance class pooled mean inertia distance classes. explanatory variables (RDA, CCA, pRDA, pCCA) significance test residual autocorrelation performed running function mso, function plot.mso print estimate much autocorrelation (based significant distance classes) causes global error variance regression analysis underestimated","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso returns amended cca rda object additional attributes grain, H, H.test vario. grain grain attribute defines interval size distance classes . H H object class 'dist' contains geographic distances observations. H.test H.test contains set dummy variables describe pairs observations (rows = elements object$H) fall distance class (columns). vario vario attribute data frame contains following components rda case (cca case brackets): H Distance class multiples grain. Dist Average distance pairs observations distance class H. n Number unique pairs observations distance class \tH. Empirical (chi-square) variogram total variance \t(inertia). Sum Sum empirical (chi-square) variograms explained \tresidual variance (inertia). CA Empirical (chi-square) variogram residual variance \t(inertia). CCA Empirical (chi-square) variogram explained variance \t(inertia). pCCA Empirical (chi-square) variogram conditioned \tvariance (inertia). se Standard error empirical (chi-square) variogram \ttotal variance (inertia). CA.signif P-value permutation test spatial \tindependence residual variance (inertia).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Wagner, H.H. 2004. Direct multi-scale ordination canonical correspondence analysis. Ecology 85: 342--351.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"responsible author Helene Wagner.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function based code published Ecological Archives E085-006 (doi:10.1890/02-0738 ).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"## Reconstruct worked example of Wagner (submitted): X <- matrix(c(1, 2, 3, 2, 1, 0), 3, 2) Y <- c(3, -1, -2) tmat <- c(1:3) ## Canonical correspondence analysis (cca): Example.cca <- cca(X, Y) Example.cca <- mso(Example.cca, tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.cca) Example.cca$vario #> H Dist n All Sum CA CCA se #> 1 1 1 2 0.25 0.3456633 0.07461735 0.2710459 0 #> 2 2 2 1 1.00 0.8086735 0.01147959 0.7971939 NA ## Correspondence analysis (ca): Example.ca <- mso(cca(X), tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.ca) ## Unconstrained ordination with test for autocorrelation ## using oribatid mite data set as in Wagner (2004) data(mite) data(mite.env) data(mite.xy) mite.cca <- cca(log(mite + 1)) mite.cca <- mso(mite.cca, mite.xy, grain = 1, permutations = 99) msoplot(mite.cca) mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, grain = 1, #> permutations = 99) #> #> Inertia Rank #> Total 1.164 #> Unconstrained 1.164 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3662 0.1328 0.0723 0.0658 0.0559 0.0481 0.0418 0.0391 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All CA CA.signif #> 0 0 0.3555 63 0.6250 0.6250 0.01 #> 1 1 1.0659 393 0.7556 0.7556 0.01 #> 2 2 2.0089 534 0.8931 0.8931 0.01 #> 3 3 2.9786 417 1.0988 1.0988 0.01 #> 4 4 3.9817 322 1.3321 1.3321 0.01 #> 5 5 5.0204 245 1.5109 1.5109 0.01 #> 10 10 6.8069 441 1.7466 1.7466 0.01 #> #> Permutation: free #> Number of permutations: 99 #> ## Constrained ordination with test for residual autocorrelation ## and scale-invariance of species-environment relationships mite.cca <- cca(log(mite + 1) ~ SubsDens + WatrCont + Substrate + Shrub + Topo, mite.env) mite.cca <- mso(mite.cca, mite.xy, permutations = 99) msoplot(mite.cca) #> Error variance of regression model underestimated by 0.4 percent mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, permutations = #> 99) #> #> Inertia Proportion Rank #> Total 1.1638 1.0000 #> Constrained 0.5211 0.4478 11 #> Unconstrained 0.6427 0.5522 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 #> 0.31207 0.06601 0.04117 0.02938 0.02438 0.01591 0.01201 0.00752 0.00612 0.00373 #> CCA11 #> 0.00284 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.07888 0.06752 0.05457 0.04023 0.03855 0.03491 0.03233 0.02692 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All Sum CA CCA se CA.signif #> 0 0 0.3555 63 0.6250 0.7479 0.5512 0.1967 0.03506 0.01 #> 1 1 1.0659 393 0.7556 0.8820 0.6339 0.2482 0.01573 0.18 #> 2 2 2.0089 534 0.8931 0.9573 0.6473 0.3100 0.01487 0.68 #> 3 3 2.9786 417 1.0988 1.1010 0.6403 0.4607 0.01858 0.46 #> 4 4 3.9817 322 1.3321 1.2548 0.6521 0.6027 0.02439 0.98 #> 5 5 5.0204 245 1.5109 1.4564 0.6636 0.7928 0.02801 0.41 #> 10 10 6.8069 441 1.7466 1.6266 0.6914 0.9351 0.02052 0.19 #> #> Permutation: free #> Number of permutations: 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Multiplicative Diversity Partitioning — multipart","title":"Multiplicative Diversity Partitioning — multipart","text":"multiplicative diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set pooled samples (gamma diversity).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"multipart(...) # S3 method for default multipart(y, x, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula multipart(formula, data, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multiplicative Diversity Partitioning — multipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, two levels assumed: row group first level, rows group second level. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variable(s) referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, entropy index calculated (see Details). relative Logical, TRUE beta diversity standardized maximum (see Details). scales Numeric, length 1, order generalized diversity index used. global Logical, indicates calculation beta diversity values, see Details. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. ... arguments passed oecosimu, .e. method, thin burnin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multiplicative Diversity Partitioning — multipart","text":"Multiplicative diversity partitioning based Whittaker's (1972) ideas, recently generalised one parametric diversity families (.e. Rényi Tsallis) Jost (2006, 2007). Jost recommends use numbers equivalents (Hill numbers), instead pure diversities, proofs, satisfies multiplicative partitioning requirements. current implementation multipart calculates Hill numbers based functions renyi tsallis (provided index argument). values one scales desired, done separate runs, adds extra dimensionality implementation, resolved efficiently. Alpha diversities averages Hill numbers hierarchy levels, global gamma diversity alpha value calculated highest hierarchy level. global = TRUE, beta calculated relative global gamma value: $$\\beta_i = \\gamma / \\alpha_{}$$ global = FALSE, beta calculated relative local gamma values (local gamma means diversity calculated particular cluster based pooled abundance vector): $$\\beta_ij = \\alpha_{(+1)j} / mean(\\alpha_{ij})$$ \\(j\\) particular cluster hierarchy level \\(\\). beta diversity value level \\(\\) mean beta values clusters level, \\(\\beta_{} = mean(\\beta_{ij})\\). relative = TRUE, respective beta diversity values standardized maximum possible values (\\(mean(\\beta_{ij}) / \\beta_{max,ij}\\)) given \\(\\beta_{max,ij} = n_{j}\\) (number lower level units given cluster \\(j\\)). expected diversity components calculated nsimul times individual based randomization community data matrix. done \"r2dtable\" method oecosimu default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multiplicative Diversity Partitioning — multipart","text":"object class \"multipart\" structure \"oecosimu\" objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multiplicative Diversity Partitioning — multipart","text":"Jost, L. (2006). Entropy diversity. Oikos, 113, 363--375. Jost, L. (2007). Partitioning diversity independent alpha beta components. Ecology, 88, 2427--2439. Whittaker, R. (1972). Evolution measurement species diversity. Taxon, 21, 213--251.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multiplicative Diversity Partitioning — multipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)))) ## Multiplicative diversity partitioning multipart(mite, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(y = mite, x = levsm, index = \"renyi\", scales = 1, #> nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 11.235 -81.754 14.07974 14.02539 14.07261 14.148 0.05 * #> gamma 12.006 -256.950 14.13219 14.11654 14.13205 14.146 0.05 * #> beta.1 1.071 29.363 1.00373 0.99935 1.00399 1.008 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ l2 + l3, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"renyi\", scales = 1, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -46.004 12.17819 12.02119 12.15824 12.3415 0.05 * #> alpha.2 11.2353 -87.290 14.08463 14.03637 14.07817 14.1383 0.05 * #> alpha.3 12.0064 -490.481 14.13774 14.13229 14.13634 14.1450 0.05 * #> gamma 14.1603 0.000 14.16027 14.16027 14.16027 14.1603 1.00 #> beta.1 1.3568 20.034 1.16000 1.14399 1.16150 1.1779 0.05 * #> beta.2 1.0710 29.394 1.00379 0.99989 1.00410 1.0073 0.05 * #> beta.3 1.1794 577.616 1.00159 1.00108 1.00169 1.0020 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, relative=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.055481 -46.949 12.221716 12.073934 12.244586 12.3654 0.05 * #> alpha.2 11.235261 -123.496 14.087872 14.043345 14.090654 14.1218 0.05 * #> alpha.3 12.006443 -323.402 14.135613 14.124838 14.135306 14.1443 0.05 * #> gamma 14.160271 0.000 14.160271 14.160271 14.160271 14.1603 1.00 #> beta.1 0.078594 17.624 0.068099 0.067172 0.068222 0.0691 0.05 * #> beta.2 0.535514 42.418 0.501697 0.500646 0.501814 0.5031 0.05 * #> beta.3 0.589695 380.700 0.500872 0.500566 0.500883 0.5013 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, global=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, global = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global TRUE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -74.528 12.1813 12.0860 12.1748 12.2676 0.05 * #> alpha.2 11.2353 -92.796 14.0859 14.0459 14.0812 14.1484 0.05 * #> alpha.3 12.0064 -331.084 14.1370 14.1256 14.1378 14.1455 0.05 * #> gamma 14.1603 0.000 14.1603 14.1603 14.1603 14.1603 1.00 #> beta.1 1.7578 112.658 1.1625 1.1543 1.1631 1.1716 0.05 * #> beta.2 1.2603 116.533 1.0053 1.0008 1.0056 1.0081 0.05 * #> beta.3 1.1794 389.764 1.0016 1.0010 1.0016 1.0025 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":null,"dir":"Reference","previous_headings":"","what":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Patches local communities regarded nested subsets community. general, species poor communities subsets species rich communities, rare species occur species rich communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"nestedchecker(comm) nestedn0(comm) nesteddisc(comm, niter = 200) nestedtemp(comm, ...) nestednodf(comm, order = TRUE, weighted = FALSE, wbinary = FALSE) nestedbetasor(comm) nestedbetajac(comm) # S3 method for nestedtemp plot(x, kind = c(\"temperature\", \"incidence\"), col=rev(heat.colors(100)), names = FALSE, ...) # S3 method for nestednodf plot(x, col = \"red\", names = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"comm Community data. niter Number iterations reorder tied columns. x Result object plot. col Colour scheme matrix temperatures. kind kind plot produced. names Label columns rows plot using names comm. logical vector length 2, row column labels returned accordingly. order Order rows columns frequencies. weighted Use species abundances weights interactions. wbinary Modify original method binary data give result weighted unweighted analysis. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"nestedness functions evaluate alternative indices nestedness. functions intended used together Null model communities used argument oecosimu analyse non-randomness results. Function nestedchecker gives number checkerboard units, 2x2 submatrices species occur different sites (Stone & Roberts 1990). Function nestedn0 implements nestedness measure N0 number absences sites richer pauperate site species occurs (Patterson & Atmar 1986). Function nesteddisc implements discrepancy index number ones shifted fill row ones table arranged species frequencies (Brualdi & Sanderson 1999). original definition arranges species (columns) frequencies, method handling tied frequencies. nesteddisc function tries order tied columns minimize discrepancy statistic rather slow, large number tied columns guarantee best ordering found (argument niter gives maximum number tried orders). case warning tied columns issued. Function nestedtemp finds matrix temperature defined sum “surprises” arranged matrix. arranged unsurprising matrix species within proportion given matrix fill upper left corner matrix, surprise absence presences diagonal distance fill line (Atmar & Patterson 1993). Function tries pack species sites low temperature (Rodríguez-Gironés & Santamaria 2006), iterative procedure, temperatures usually vary among runs. Function nestedtemp also plot method can display either incidences temperatures surprises. Matrix temperature rather vaguely described (Atmar & Patterson 1993), Rodríguez-Gironés & Santamaria (2006) explicit description used . However, results probably differ implementations, users cautious interpreting results. details calculations explained vignette Design decisions implementation can read using functions browseVignettes. Function nestedness bipartite package direct port BINMATNEST programme Rodríguez-Gironés & Santamaria (2006). Function nestednodf implements nestedness metric based overlap decreasing fill (Almeida-Neto et al., 2008). Two basic properties required matrix maximum degree nestedness according metric: (1) complete overlap 1's right left columns rows, (2) decreasing marginal totals pairs columns pairs rows. nestedness statistic evaluated separately columns (N columns) rows (N rows) combined whole matrix (NODF). set order = FALSE, statistic evaluated current matrix ordering allowing tests meaningful hypothesis matrix structure default ordering row column totals (breaking ties total abundances weighted = TRUE) (see Almeida-Neto et al. 2008). weighted = TRUE, function finds weighted version index (Almeida-Neto & Ulrich, 2011). However, requires quantitative null models adequate testing. Almeida-Neto & Ulrich (2011) say positive nestedness values first row/column higher second. condition, weighted analysis binary data always give zero nestedness. argument wbinary = TRUE, equality rows/columns also indicates nestedness, binary data give identical results weighted unweighted analysis. However, can also influence results weighted analysis results may differ Almeida-Neto & Ulrich (2011). Functions nestedbetasor nestedbetajac find multiple-site dissimilarities decompose components turnover nestedness following Baselga (2012); pairwise dissimilarities can found designdist. can seen decomposition beta diversity (see betadiver). Function nestedbetasor uses Sørensen dissimilarity turnover component Simpson dissimilarity (Baselga 2012), nestedbetajac uses analogous methods Jaccard index. functions return vector three items: turnover, nestedness sum multiple Sørensen Jaccard dissimilarity. last one total beta diversity (Baselga 2012). functions treat data presence/absence (binary) can used binary nullmodel. overall dissimilarity constant nullmodels fix species (column) frequencies (\"c0\"), components constant row columns also fixed (e.g., model \"quasiswap\"), functions meaningful null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"result returned nestedness function contains item called statistic, components differ among functions. functions constructed can handled oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Almeida-Neto, M., Guimarães, P., Guimarães, P.R., Loyola, R.D. & Ulrich, W. (2008). consistent metric nestedness analysis ecological systems: reconciling concept measurement. Oikos 117, 1227--1239. Almeida-Neto, M. & Ulrich, W. (2011). straightforward computational approach measuring nestedness using quantitative matrices. Env. Mod. Software 26, 173--178. Atmar, W. & Patterson, B.D. (1993). measurement order disorder distribution species fragmented habitat. Oecologia 96, 373--382. Baselga, . (2012). relationship species replacement, dissimilarity derived nestedness, nestedness. Global Ecol. Biogeogr. 21, 1223--1232. Brualdi, R.. & Sanderson, J.G. (1999). Nested species subsets, gaps, discrepancy. Oecologia 119, 256--264. Patterson, B.D. & Atmar, W. (1986). Nested subsets structure insular mammalian faunas archipelagos. Biol. J. Linnean Soc. 28, 65--82. Rodríguez-Gironés, M.. & Santamaria, L. (2006). new algorithm calculate nestedness temperature presence-absence matrices. J. Biogeogr. 33, 924--935. Stone, L. & Roberts, . (1990). checkerboard score species distributions. Oecologia 85, 74--79. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Jari Oksanen Gustavo Carvalho (nestednodf).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"data(sipoo) ## Matrix temperature out <- nestedtemp(sipoo) out #> nestedness temperature: 10.23506 #> with matrix fill 0.2233333 plot(out) plot(out, kind=\"incid\") ## Use oecosimu to assess the non-randomness of checker board units nestedchecker(sipoo) #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 oecosimu(sipoo, nestedchecker, \"quasiswap\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = #> \"quasiswap\") #> #> nullmodel method ‘quasiswap’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.55254 2715.3 2564.9 2713.0 2911.2 0.57 ## Another Null model and standardized checkerboard score oecosimu(sipoo, nestedchecker, \"r00\", statistic = \"C.score\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r00\", #> statistic = \"C.score\") #> #> nullmodel method ‘r00’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> C.score 2.2588 -28.215 9.2285 8.7036 9.2294 9.6735 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract the Number of Observations from a vegan Fit. — nobs.cca","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Extract number ‘observations’ vegan model fit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"","code":"# S3 method for cca nobs(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"object fitted model object. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Function nobs generic R, vegan provides methods objects betadisper, cca related methods, CCorA, decorana, isomap, metaMDS, pcnm, procrustes, radfit, varpart wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"single number, normally integer, giving number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Null Model and Simulation — nullmodel","title":"Null Model and Simulation — nullmodel","text":"nullmodel function creates object, can serve basis Null Model simulation via simulate method. update method updates nullmodel object without sampling (effective sequential algorithms). smbind binds together multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Null Model and Simulation — nullmodel","text":"","code":"nullmodel(x, method) # S3 method for nullmodel print(x, ...) # S3 method for nullmodel simulate(object, nsim = 1, seed = NULL, burnin = 0, thin = 1, ...) # S3 method for nullmodel update(object, nsim = 1, seed = NULL, ...) # S3 method for simmat print(x, ...) smbind(object, ..., MARGIN, strict = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Null Model and Simulation — nullmodel","text":"x community matrix. print method, object printed. method Character, specifying one null model algorithms listed help page commsim. can user supplied object class commsim. object object class nullmodel returned function nullmodel. case smbind simmat object returned update simulate methods. nsim Positive integer, number simulated matrices return. update method, number burnin steps made sequential algorithms update status input model object. seed object specifying random number generator initialized (\"seeded\"). Either NULL integer used call set.seed simulating matrices. set, value saved \"seed\" attribute returned value. default, NULL change random generator state, return .Random.seed \"seed\" attribute, see Value. burnin Nonnegative integer, specifying number steps discarded starting simulation. Active sequential null model algorithms. Ignored non-sequential null model algorithms. thin Positive integer, number simulation steps made returned matrix. Active sequential null model algorithms. Ignored non-sequential null model algorithms. MARGIN Integer, indicating dimension multiple simmat objects bound together smbind. 1: matrices stacked (row bound), 2: matrices column bound, 3: iterations combined. Needs length 1. dimensions expected match across objects. strict Logical, consistency time series attributes (\"start\", \"end\", \"thin\", number simulated matrices) simmat objects strictly enforced binding multiple objects together using smbind. Applies input objects based sequential null model algorithms. ... Additional arguments supplied algorithms. case smbind can contain multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Null Model and Simulation — nullmodel","text":"purpose nullmodel function create object, necessary statistics input matrix calculated . information reused, recalculated step simulation process done simulate method. simulate method carries simulation, simulated matrices stored array. sequential algorithms, method updates state input nullmodel object. Therefore, possible diagnostic tests returned simmat object, make simulations, use increased thinning value desired. update method makes burnin steps case sequential algorithms update status input model without attempt return matrices. non-sequential algorithms method nothing. update preferred way making burnin iterations without sampling. Alternatively, burnin can done via simulate method. convergence diagnostics, recommended use simulate method without burnin. input nullmodel object updated, samples can simulated desired without start process . See Examples. smbind function can used combine multiple simmat objects. comes handy null model simulations stratified sites (MARGIN = 1) species (MARGIN = 2), case multiple objects returned identical/consistent settings e.g. parallel computations (MARGIN = 3). Sanity checks made ensure combining multiple objects sensible, user's responsibility check independence simulated matrices null distribution converged case sequential null model algorithms. strict = FALSE setting can relax checks regarding start, end, thinning values sequential null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Null Model and Simulation — nullmodel","text":"function nullmodel returns object class nullmodel. set objects sharing environment: data: original matrix integer mode. nrow: number rows. ncol: number columns. rowSums: row sums. colSums: column sums. rowFreq: row frequencies (number nonzero cells). colFreq: column frequencies (number nonzero cells). totalSum: total sum. fill: number nonzero cells matrix. commsim: commsim object result method argument. state: current state permutations, matrix similar original. NULL non-sequential algorithms. iter: current number iterations sequential algorithms. NULL non-sequential algorithms. simulate method returns object class simmat. array simulated matrices (third dimension corresponding nsim argument). update method returns current state (last updated matrix) invisibly, update input object sequential algorithms. non sequential algorithms, returns NULL. smbind function returns object class simmat.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Null Model and Simulation — nullmodel","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Null Model and Simulation — nullmodel","text":"Care must taken input matrix contains single row column. input invalid swapping hypergeometric distribution (calling r2dtable) based algorithms. also applies cases input stratified subsets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Null Model and Simulation — nullmodel","text":"","code":"data(mite) x <- as.matrix(mite)[1:12, 21:30] ## non-sequential nullmodel (nm <- nullmodel(x, \"r00\")) #> An object of class “nullmodel” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> (sm <- simulate(nm, nsim=10)) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## sequential nullmodel (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 55, End = 100, Thin = 5 #> ## sequential nullmodel with burnin and extra updating (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, burnin=10, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> ## sequential nullmodel with separate initial burnin (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> nm <- update(nm, nsim=10) (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> ## combining multiple simmat objects ## stratification nm1 <- nullmodel(x[1:6,], \"r00\") sm1 <- simulate(nm1, nsim=10) nm2 <- nullmodel(x[7:12,], \"r00\") sm2 <- simulate(nm2, nsim=10) smbind(sm1, sm2, MARGIN=1) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## binding subsequent samples from sequential algorithms ## start, end, thin retained nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=10) sm1 <- simulate(nm, nsim=10, thin=5) sm2 <- simulate(nm, nsim=20, thin=5) sm3 <- simulate(nm, nsim=10, thin=5) smbind(sm3, sm2, sm1, MARGIN=3) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 40 #> Start = 15, End = 210, Thin = 5 #> ## 'replicate' based usage which is similar to the output ## of 'parLapply' or 'mclapply' in the 'parallel' package ## start, end, thin are set, also noting number of chains smfun <- function(x, burnin, nsim, thin) { nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=burnin) simulate(nm, nsim=nsim, thin=thin) } smlist <- replicate(3, smfun(x, burnin=50, nsim=10, thin=5), simplify=FALSE) smbind(smlist, MARGIN=3) # Number of permuted matrices = 30 #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 30 #> Start = 55, End = 100, Thin = 5 (3 chains) #> if (FALSE) { ## parallel null model calculations library(parallel) if (.Platform$OS.type == \"unix\") { ## forking on Unix systems smlist <- mclapply(1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) smbind(smlist, MARGIN=3) } ## socket type cluster, works on all platforms cl <- makeCluster(3) clusterEvalQ(cl, library(vegan)) clusterExport(cl, c(\"smfun\", \"x\")) smlist <- parLapply(cl, 1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) stopCluster(cl) smbind(smlist, MARGIN=3) }"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function evaluates statistic vector statistics community evaluates significance series simulated random communities. approach used traditionally analysis nestedness, function general can used statistics evaluated simulated communities. Function oecosimu collects evaluates statistics. Null model communities described make.commsim permatfull/ permatswap, definition Null models nullmodel, nestedness statistics nestednodf (describes several alternative statistics, including nestedness temperature, \\(N0\\), checker board units, nestedness discrepancy NODF).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"oecosimu(comm, nestfun, method, nsimul = 99, burnin = 0, thin = 1, statistic = \"statistic\", alternative = c(\"two.sided\", \"less\", \"greater\"), batchsize = NA, parallel = getOption(\"mc.cores\"), ...) # S3 method for oecosimu as.ts(x, ...) # S3 method for oecosimu toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"comm Community data, Null model object generated nullmodel object class simmat (array permuted matrices simulate.nullmodel). comm community data, null model simulation method must specified. comm nullmodel, simulation method ignored, comm simmat object, arguments ignored except nestfun, statistic alternative. nestfun Function analysed. nestedness functions provided vegan (see nestedtemp), function can used accepts community first argument, returns either plain number vector result list item name defined argument statistic. See Examples defining functions. method Null model method: either name (character string) method defined make.commsim commsim function. argument ignored comm nullmodel simmat object. See Details Examples. nsimul Number simulated null communities (ignored comm simmat object). burnin Number null communities discarded proper analysis sequential methods (\"tswap\") (ignored non-sequential methods comm simmat object). thin Number discarded null communities two evaluations nestedness statistic sequential methods (ignored non-sequential methods comm simmat object). statistic name statistic returned nestfun. alternative character string specifying alternative hypothesis, must one \"two.sided\" (default), \"greater\" \"less\". Please note \\(p\\)-value two-sided test approximately two times higher corresponding one-sided test (\"greater\" \"less\" depending sign difference). batchsize Size Megabytes largest simulation object. larger structure produced, analysis broken internally batches. default NA analysis broken batches. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. define nestfun Windows needs R packages vegan permute, must set socket cluster call. x oecosimu result object. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu wrapper evaluates statistic using function given nestfun, simulates series null models based nullmodel, evaluates statistic null models. vegan packages contains nestedness functions described separately (nestedchecker, nesteddisc, nestedn0, nestedtemp, nestednodf), many functions can used long meaningful simulated communities. applicable function must return either statistic plain number vector, list element \"statistic\" (like chisq.test), item whose name given argument statistic. statistic can single number (like typical nestedness index), can vector. vector indices can used analyse site (row) species (column) properties, see treedive example. Raup-Crick index (raupcrick) gives example using dissimilarities. Null model type can given name (quoted character string) used define Null model make.commsim. include binary models described Wright et al. (1998), Jonsson (2001), Gotelli & Entsminger (2003), Miklós & Podani (2004), others. several quantitative Null models, discussed Hardy (2008), several unpublished (see make.commsim, permatfull, permatswap discussion). user can also define commsim function (see Examples). Function works first defining nullmodel given commsim, generating series simulated communities simulate.nullmodel. shortcut can used stages input can Community data (comm), Null model function (nestfun) number simulations (nsimul). nullmodel object number simulations, argument method ignored. three-dimensional array simulated communities generated simulate.nullmodel, arguments method nsimul ignored. last case allows analysing several statistics simulations. function first generates simulations given nullmodel analyses using nestfun. large data sets /large number simulations, generated objects can large, memory exhausted, analysis can become slow system can become unresponsive. simulation broken several smaller batches simulated nullmodel objective set batchsize avoid memory problems (see object.size estimating size current data set). parallel processing still increases memory needs. parallel processing used evaluating nestfun. main load may simulation nullmodel, parallel argument help . Function .ts transforms simulated results sequential methods time series ts object. allows using analytic tools time series studying sequences (see examples). Function toCoda transforms simulated results sequential methods \"mcmc\" object coda package. coda package provides functions analysis stationarity, adequacy sample size, autocorrelation, need burn-much sequential methods, summary results. Please consult documentation coda package. Function permustats provides support standard density, densityplot, qqnorm qqmath functions simulated values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu returns object class \"oecosimu\". result object items statistic oecosimu. statistic contains complete object returned nestfun original data. oecosimu component contains following items: statistic Observed values statistic. simulated Simulated values statistic. means Mean values statistic simulations. z Standardized effect sizes (SES, .k.. \\(z\\)-values) observed statistic based simulations. pval \\(P\\)-values statistic based simulations. alternative type testing given argument alternative. method method used nullmodel. isSeq TRUE method sequential.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Jari Oksanen Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"wonder name oecosimu, look journal names References (nestedtemp). internal structure function radically changed vegan 2.2-0 introduction commsim nullmodel deprecation commsimulator.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"## Use the first eigenvalue of correspondence analysis as an index ## of structure: a model for making your own functions. data(sipoo) ## Traditional nestedness statistics (number of checkerboard units) oecosimu(sipoo, nestedchecker, \"r0\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r0\") #> #> nullmodel method ‘r0’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 -18.742 8017.0 7469.0 8021.0 8450.5 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential model, one-sided test, a vector statistic out <- oecosimu(sipoo, decorana, \"swap\", burnin=100, thin=10, statistic=\"evals\", alt = \"greater\") out #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = decorana, method = \"swap\", #> burnin = 100, thin = 10, statistic = \"evals\", alternative = \"greater\") #> #> nullmodel method ‘swap’ with 99 simulations #> options: thin 10, burnin 100 #> alternative hypothesis: statistic is greater than simulated values #> #> #> Call: #> nestfun(veg = comm) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.4436 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.3822 0.2612 0.1668 0.08723 #> Additive Eigenvalues 0.3822 0.2609 0.1631 0.07650 #> Decorana values 0.4154 0.2465 0.1391 0.04992 #> Axis lengths 2.9197 2.5442 2.7546 1.78074 #> #> #> statistic SES mean 50% 95% Pr(sim.) #> DCA1 0.382249 2.04544 0.32960 0.33249 0.3677 0.01 ** #> DCA2 0.261208 1.77368 0.21549 0.21404 0.2587 0.05 * #> DCA3 0.166788 0.63257 0.15363 0.15405 0.1907 0.22 #> DCA4 0.087226 -1.69622 0.12533 0.12788 0.1636 0.96 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Inspect the swap sequence as a time series object plot(as.ts(out)) lag.plot(as.ts(out)) acf(as.ts(out)) ## Density plot densityplot(permustats(out), as.table = TRUE, layout = c(1,4)) ## Use quantitative null models to compare ## mean Bray-Curtis dissimilarities data(dune) meandist <- function(x) mean(vegdist(x, \"bray\")) mbc1 <- oecosimu(dune, meandist, \"r2dtable\") mbc1 #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = \"r2dtable\") #> #> nullmodel method ‘r2dtable’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 13.84 0.46601 0.44155 0.46756 0.4928 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Define your own null model as a 'commsim' function: shuffle cells ## in each row foo <- function(x, n, nr, nc, ...) { out <- array(0, c(nr, nc, n)) for (k in seq_len(n)) out[,,k] <- apply(x, 2, function(z) sample(z, length(z))) out } cf <- commsim(\"myshuffle\", foo, isSeq = FALSE, binary = FALSE, mode = \"double\") oecosimu(dune, meandist, cf) #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = cf) #> #> nullmodel method ‘myshuffle’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 3.8862 0.63505 0.62995 0.63508 0.64 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Use pre-built null model nm <- simulate(nullmodel(sipoo, \"curveball\"), 99) oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘curveball’ with 99 simulations #> options: thin 1, burnin 0 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 1.4459 2710.6 2635.0 2723.0 2762.7 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Several chains of a sequential model -- this can be generalized ## for parallel processing (see ?smbind) nm <- replicate(5, simulate(nullmodel(sipoo, \"swap\"), 99, thin=10, burnin=100), simplify = FALSE) ## nm is now a list of nullmodels: use smbind to combine these into one ## nullmodel with several chains ## IGNORE_RDIFF_BEGIN nm <- smbind(nm, MARGIN = 3) nm #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 18 x 50 matrix #> Number of permuted matrices = 495 #> Start = 110, End = 1090, Thin = 10 (5 chains) #> oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘swap’ with 495 simulations #> options: thin 10, burnin 100, chains 5 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.71842 2698.1 2572.0 2676.0 2904.7 0.4577 ## IGNORE_RDIFF_END ## After this you can use toCoda() and tools in the coda package to ## analyse the chains (these will show that thin, burnin and nsimul are ## all too low for real analysis)."},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":null,"dir":"Reference","previous_headings":"","what":"Support Functions for Drawing Vectors — ordiArrowTextXY","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Support functions assist drawing vectors (arrows) ordination plots. ordiArrowMul finds multiplier coordinates head vector occupy fill proportion plot region. ordiArrowTextXY finds coordinates locations labels drawn just beyond head vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"ordiArrowTextXY(x, labels, display, choices = c(1,2), rescale = TRUE, fill = 0.75, at = c(0,0), ...) ordiArrowMul(x, at = c(0,0), fill = 0.75, display, choices = c(1,2), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"x R object, scores can determine suitable ordination scores object created envfit, two-column matrix coordinates arrow heads two plot axes. labels Change plotting labels. character vector labels label coordinates sought. supplied, determined row names x, scores(x, ...) required. either defined, suitable labels generated. display character string known scores one methods indicates type scores extract. fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). x created envfit display can set user takes value \"vectors\". Ignored x matrix. choices Axes plotted. rescale logical; coordinates extracted x rescaled fill fill proportion plot region? default always rescale coordinates usually desired objects x coordinates retrieved. supplying x 2-column matrix already rescaled, set FALSE. fill numeric; proportion plot fill span arrows. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. ... Parameters passed scores, strwidth strheight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowMul finds multiplier scale bunch arrows fill ordination plot, ordiArrowTextXY finds coordinates labels arrows. NB., ordiArrowTextXY draw labels; simply returns coordinates labels drawn use another function, text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowTextXY, 2-column matrix coordinates label centres coordinate system currently active plotting device. ordiArrowMul, length-1 vector containing scaling factor.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Jari Oksanen, modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"## Scale arrows by hand to fill 80% of the plot ## Biplot arrows by hand data(varespec, varechem) ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, display = c(\"species\",\"sites\")) ## biplot scores bip <- scores(ord, choices = 1:2, display = \"bp\") ## scaling factor for arrows to fill 80% of plot (mul <- ordiArrowMul(bip, fill = 0.8)) #> [1] 2.092173 bip.scl <- bip * mul # Scale the biplot scores labs <- rownames(bip) # Arrow labels ## calculate coordinate of labels for arrows (bip.lab <- ordiArrowTextXY(bip.scl, rescale = FALSE, labels = labs)) #> [,1] [,2] #> Al 1.9098765 -0.3562415 #> P -0.9298005 -1.6652122 #> K -1.0069931 -0.3764923 ## draw arrows and text labels arrows(0, 0, bip.scl[,1], bip.scl[,2], length = 0.1) text(bip.lab, labels = labs) ## Handling of ordination objects directly mul2 <- ordiArrowMul(ord, display = \"bp\", fill = 0.8) stopifnot(all.equal(mul, mul2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Functions add arrows, line segments, regular grids points. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"ordiarrows(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, startmark, label = FALSE, length = 0.1, ...) ordisegments(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, label = FALSE, ...) ordigrid(ord, levels, replicates, display = \"sites\", lty = c(1,1), col = c(1,1), lwd = c(1,1), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. levels, replicates Alternatively, regular groups can defined arguments levels replicates, levels gives number groups, replicates number successive items group. order.Order points increasing order variable within groups. Reverse sign variable decreasing ordering. display Item displayed. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names. ordiellipse, ordihull ordispider group name centroid object, ordiarrows start arrow, ordisegments ends. ordiellipse ordihull use standard text, others use ordilabel. startmark plotting character used mark first item. default use mark, instance, startmark = 1 draw circle. plotting characters, see pch points. col Colour lines, label borders startmark ordiarrows ordisegments. can vector recycled groups. ordigrid can vector length 2 used levels replicates. length Length edges arrow head (inches). lty, lwd Line type, line width used levels replicates ordigrid. ... Parameters passed graphical functions lines, segments, arrows, scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Function ordiarrows draws arrows ordisegments draws line segments successive items groups. Function ordigrid draws line segments within groups corresponding items among groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"functions add graphical items ordination graph: must draw graph first.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"example(pyrifos) #> #> pyrifs> data(pyrifos) #> #> pyrifs> ditch <- gl(12, 1, length=132) #> #> pyrifs> week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) #> #> pyrifs> dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) mod <- rda(pyrifos) plot(mod, type = \"n\") ## Annual succession by ditches, colour by dose ordiarrows(mod, ditch, label = TRUE, col = as.numeric(dose)) legend(\"topright\", levels(dose), lty=1, col=1:5, title=\"Dose\") ## Show only control and highest Pyrifos treatment plot(mod, type = \"n\") ordiarrows(mod, ditch, label = TRUE, show.groups = c(\"2\", \"3\", \"5\", \"11\")) ordiarrows(mod, ditch, label = TRUE, show = c(\"6\", \"9\"), col = 2) legend(\"topright\", c(\"Control\", \"Pyrifos 44\"), lty = 1, col = c(1,2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions add convex hulls, “spider” graphs, ellipses cluster dendrogram ordination diagrams. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"ordihull(ord, groups, display = \"sites\", draw = c(\"lines\",\"polygon\", \"none\"), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd = NULL, ...) ordiellipse(ord, groups, display=\"sites\", kind = c(\"sd\",\"se\", \"ehull\"), conf, draw = c(\"lines\",\"polygon\", \"none\"), w = weights(ord, display), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd=NULL, ...) ordibar(ord, groups, display = \"sites\", kind = c(\"sd\", \"se\"), conf, w = weights(ord, display), col = 1, show.groups, label = FALSE, lwd = NULL, length = 0, ...) ordispider(ord, groups, display=\"sites\", w = weights(ord, display), spiders = c(\"centroid\", \"median\"), show.groups, label = FALSE, col = NULL, lty = NULL, lwd = NULL, ...) ordicluster(ord, cluster, prune = 0, display = \"sites\", w = weights(ord, display), col = 1, draw = c(\"segments\", \"none\"), ...) # S3 method for ordihull summary(object, ...) # S3 method for ordiellipse summary(object, ...) ordiareatest(ord, groups, area = c(\"hull\", \"ellipse\"), kind = \"sd\", permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. display Item displayed. draw character; objects represented plot? ordihull ordiellipse use either lines polygon draw lines. ordicluster, line segments drawn using segments. suppress plotting, use \"none\". Graphical parameters passed . main difference polygons may filled non-transparent. none nothing drawn, function returns invisible plotting. col Colour hull ellipse lines (draw = \"lines\") fills (draw = \"polygon\") ordihull ordiellipse. draw = \"polygon\", colour bordering lines can set argument border polygon function. functions effect depends underlining functions argument passed . multiple values col specified used element names(table(groups)) (order), shorter vectors recycled. Function ordicluster groups, argument recycled points, colour connecting lines mixture point s cluster. alpha Transparency fill colour draw = \"polygon\" ordihull ordiellipse. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names centroid object. ordiellipse ordihull use standard text, others use ordilabel. w Weights used find average within group. Weights used automatically cca decorana results, unless undone user. w=NULL sets equal weights points. kind Draw standard deviations points (sd), standard errors (se) ellipsoid hulls enclose points group (ehull). conf Confidence limit ellipses, e.g. 0.95. given, corresponding sd se multiplied corresponding value found Chi-squared distribution 2df. spiders centres spider bodies calculated either centroids (averages) spatial medians. cluster Result hierarchic cluster analysis, hclust agnes. prune Number upper level hierarchies removed dendrogram. prune \\(>0\\), dendrogram disconnected. object result object ordihull ordiellipse. result invisible, can saved, used summaries (areas etc. hulls ellipses). area Evaluate area convex hulls ordihull, ellipses ordiellipse. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. lty, lwd, border Vectors parameters can supplied applied (appropriate) element names(table(groups)) (order). Shorter vectors recycled. length Width (inches) small (“caps”) ends bar segment (passed arrows). ... Parameters passed graphical functions scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Function ordihull draws lines polygons convex hulls found function chull encircling items groups. Function ordiellipse draws lines polygons ellipses groups. function can either draw standard deviation points (kind=\"sd\") standard error (weighted) centroids (kind=\"se\"), (weighted) correlation defines direction principal axis ellipse. kind = \"se\" used together argument conf, ellipses show confidence regions locations group centroids. kind=\"ehull\" function draws ellipse encloses points group using ellipsoidhull (cluster package). Function ordibar draws crossed “error bars” using either either standard deviation point scores standard error (weighted) average scores. principal axes corresponding ordiellipse, found principal component analysis (weighted) covariance matrix. Functions ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups. Argument kind can used select kind ellipse, effect convex hulls. Function ordispider draws ‘spider’ diagram point connected group centroid segments. Weighted centroids used correspondence analysis methods cca decorana user gives weights call. ordispider called cca rda result without groups argument, function connects ‘WA’ scores corresponding ‘LC’ score. argument (invisible) ordihull object, function connect points hull centroid. Function ordicluster overlays cluster dendrogram onto ordination. needs result hierarchic clustering hclust agnes, similar structure. Function ordicluster connects cluster centroids line segments. Function uses centroids points clusters, therefore similar average linkage methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"functions add graphical items ordination graph: must draw graph first. draw line segments, grids arrows, see ordisegments, ordigrid andordiarrows.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions ordihull, ordiellipse ordispider return invisible plotting structure. Function ordispider return coordinates point connected (centroids ‘LC’ scores). Function ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ Management, dune.env) plot(mod, type=\"n\", scaling = \"symmetric\") ## Catch the invisible result of ordihull... pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", label = TRUE)) ## ... and find centres and areas of the hulls summary(pl) #> BF HF NM SF #> CCA1 0.2917476 0.36826105 -1.3505642 0.2762936 #> CCA2 0.8632208 0.09419919 0.2681515 -0.8139398 #> Area 0.1951715 0.59943363 1.7398193 1.0144372 ## use more colours and add ellipsoid hulls plot(mod, type = \"n\") pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", col = 1:4, draw=\"polygon\", label =TRUE)) with(dune.env, ordiellipse(mod, Management, scaling = \"symmetric\", kind = \"ehull\", col = 1:4, lwd=3)) ## ordispider to connect WA and LC scores plot(mod, dis=c(\"wa\",\"lc\"), type=\"p\") ordispider(mod) ## Other types of plots plot(mod, type = \"p\", display=\"sites\") cl <- hclust(vegdist(dune)) ordicluster(mod, cl, prune=3, col = cutree(cl, 4)) ## confidence ellipse: location of the class centroids plot(mod, type=\"n\", display = \"sites\") with(dune.env, text(mod, display=\"sites\", labels = as.character(Management), col=as.numeric(Management))) pl <- with(dune.env, ordiellipse(mod, Management, kind=\"se\", conf=0.95, lwd=2, draw = \"polygon\", col=1:4, border=1:4, alpha=63)) summary(pl) #> BF HF NM SF #> CCA1 0.4312652 0.5583211 -1.87848340 0.5601499 #> CCA2 1.3273917 0.6373120 -0.05503211 -1.3859924 #> Area 1.4559842 1.3806668 2.73667419 1.5559135 ## add confidence bars with(dune.env, ordibar(mod, Management, kind=\"se\", conf=0.95, lwd=2, col=1:4, label=TRUE))"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Function ordilabel similar text, text opaque label. can help crowded ordination plots: still see text labels, least uppermost readable. Argument priority helps make important labels visible.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"ordilabel(x, display, labels, choices = c(1, 2), priority, select, cex = 0.8, fill = \"white\", border = NULL, col = NULL, xpd = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"x ordination object object known scores. display Kind scores displayed (passed scores). labels Optional text used plots. given, text found ordination object. choices Axes shown (passed scores). priority Vector length number labels. items high priority plotted uppermost. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. cex Character expansion text (passed text). fill Background colour labels (col argument polygon). border colour visibility border label defined polygon. col Text colour. Default NULL give value border par(\"fg\") border NULL. xpd Draw labels also outside plot region (see par). ... arguments (passed text).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"function may useful crowded ordination plots, particular together argument priority. see text labels, least readable. alternatives crowded plots identify.ordiplot, orditorp orditkplot ( vegan3d package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"data(dune) ord <- cca(dune) plot(ord, type = \"n\") ordilabel(ord, dis=\"sites\", cex=1.2, font=3, fill=\"hotpink\", col=\"blue\") ## You may prefer separate plots, but here species as well ordilabel(ord, dis=\"sp\", font=2, priority=colSums(dune))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Alternative plot and identify Functions for Ordination — ordiplot","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot alternative plotting function can worked vegan ordination result many non-vegan results. addition, plot functions vegan ordinations return invisibly \"ordiplot\" result object, allows using ordiplot support functions result: identify can used add labels selected site, species constraint points, points text can add elements plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"ordiplot(ord, choices = c(1, 2), type=\"points\", display, xlim, ylim, cex = 0.7, ...) # S3 method for ordiplot identify(x, what, labels, ...) # S3 method for ordiplot points(x, what, select, arrows = FALSE, ...) # S3 method for ordiplot text(x, what, labels, select, arrows = FALSE, length = 0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"ord result ordination. choices Axes shown. type type graph may \"points\", \"text\" \"none\" ordination method. display Display \"sites\" \"species\". default methods display , cca, rda, dbrda capscale plot.cca. xlim, ylim x y limits (min,max) plot. cex Character expansion factor points text. ... graphical parameters. x result object ordiplot. Items identified ordination plot. types depend kind plot used. methods know sites species, functions cca rda know addition constraints (LC scores), centroids biplot, plot.procrustes ordination plot heads points. labels Optional text used labels. Row names used missing. arrows Draw arrows origin. always TRUE biplot scores value ignored. Setting TRUE draw arrows type scores. allows, e.g, using biplot arrows species. arrow head value scores, possible text moved outwards. length Length arrow heads (see arrows). select Items displayed. can either logical vector TRUE displayed items vector indices displayed items.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot draws ordination diagram using black circles sites red crosses species. returns invisibly object class ordiplot can used identify.ordiplot label selected sites species, constraints cca rda. function can handle output several alternative ordination methods. cca, rda decorana uses plot method option type = \"points\". addition, plot functions methods return invisibly ordiplot object can used identify.ordiplot label points. ordinations relies scores extract scores. full user control plots, best call ordiplot type = \"none\" save result, add sites species using points.ordiplot text.ordiplot pass arguments corresponding default graphical functions. functions can chained pipes allows alternative intuitive way building plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot returns invisibly object class ordiplot used scores. general, vegan plot functions ordination results also return invisible ordiplot object. plot(..., type = \"n\") used originally, plot empty, items can added invisible object. Functions points text return input object without modification, allows chaining commands pipes. Function identify.ordiplot uses object label point.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"purpose functions provide similar functionality plot, plotid specid methods library labdsv. functions somewhat limited parametrization, can call directly standard identify plot functions better user control.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"## Draw a plot for a non-vegan ordination (cmdscale). data(dune) dune.dis <- vegdist(wisconsin(dune)) dune.mds <- cmdscale(dune.dis, eig = TRUE) dune.mds$species <- wascores(dune.mds$points, dune, expand = TRUE) pl <- ordiplot(dune.mds, type = \"none\") points(pl, \"sites\", pch=21, col=\"red\", bg=\"yellow\") text(pl, \"species\", col=\"blue\", cex=0.9) if (FALSE) { ## same plot using pipes (|>) ordiplot(dune.mds, type=\"n\") |> points(\"sites\", pch=21, col=\"red\", bg=\"yellow\") |> text(\"species\", col=\"blue\", cex=0.9) ## Some people think that species should be shown with arrows in PCA. ## Other ordination methods also return an invisible ordiplot object and ## we can use pipes to draw those arrows. mod <- rda(dune) plot(mod, type=\"n\") |> points(\"sites\", pch=16, col=\"red\") |> text(\"species\", arrows = TRUE, length=0.05, col=\"blue\") } ## Default plot of the previous using identify to label selected points if (FALSE) { pl <- ordiplot(dune.mds) identify(pl, \"spec\")}"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function ordipointlabel produces ordination plots points text label points. points exact location given ordination, function tries optimize location text labels minimize overplotting text. function may useful moderately crowded ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"ordipointlabel(x, display = c(\"sites\", \"species\"), choices = c(1, 2), col = c(1, 2), pch = c(\"o\", \"+\"), font = c(1, 1), cex = c(0.8, 0.8), add = FALSE, select, ...) # S3 method for ordipointlabel plot(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"x ordipointlabel() result object ordination function. plot.ordipointlabel object resulting call ordipointlabel(). display Scores displayed plot. choices Axes shown. col, pch, font, cex Colours, point types, font style character expansion kind scores displayed plot. vectors length number items display. add Add existing plot. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. select used single set scores plotted (.e. length(display) == 1), otherwise ignored warning issued. logical vector used, must length scores plotted. ... arguments passed points text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function uses simulated annealing (optim, method = \"SANN\") optimize location text labels points. eight possible locations: , , sides corners. weak preference text right point, weak avoidance corner positions. exact locations goodness solution varies runs, guarantee finding global optimum. optimization can take long time difficult cases high number potential overlaps. Several sets scores can displayed one plot. function modelled pointLabel maptools package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function returns invisibly object class ordipointlabel items xy coordinates points, labels coordinates labels, items pch, cex font graphical parameters point label. addition, returns result optim attribute \"optim\". unit overlap area character \"m\", variable cex smallest alternative. plot method based orditkplot alter reset graphical parameters via par. result object ordipointlabel inherits orditkplot vegan3d package, may possible edit result object orditkplot, good results necessary points span whole horizontal axis without empty margins.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function designed ordination graphics, optimization works properly plots isometric aspect ratio.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"data(dune) ord <- cca(dune) plt <- ordipointlabel(ord) ## set scaling - should be no warnings! ordipointlabel(ord, scaling = \"sites\") ## plot then add plot(ord, scaling = \"symmetric\", type = \"n\") ordipointlabel(ord, display = \"species\", scaling = \"symm\", add = TRUE) ordipointlabel(ord, display = \"sites\", scaling = \"symm\", add = TRUE) ## redraw plot without rerunning SANN optimisation plot(plt)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function provides plot.lm style diagnostic plots results constrained ordination cca, rda, dbrda capscale. Normally need plots, ordination descriptive make assumptions distribution residuals. However, permute residuals significance tests (anova.cca), may interested inspecting residuals really exchangeable independent fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"ordiresids(x, kind = c(\"residuals\", \"scale\", \"qqmath\"), residuals = \"working\", type = c(\"p\", \"smooth\", \"g\"), formula, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"x Ordination result cca, rda, dbrda, capscale. kind type plot: \"residuals\" plot residuals fitted values, \"scale\" square root absolute residuals fitted values, \"qqmath\" residuals expected distribution (defaults qnorm), unless defined differently formula argument. residuals kind residuals fitted values, alternatives \"working\", \"response\", \"standardized\" \"studentized\" (see Details). type type plot. argument passed lattice functions. formula Formula override default plot. formula can contain items Fitted, Residuals, Species Sites (provided names species sites available ordination result). ... arguments passed lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"default plots similar plot.lm, use Lattice functions xyplot qqmath. alternatives default formulae can replaced user. elements available formula groups argument Fitted, Residuals, Species Sites. residuals = \"response\" residuals = \"working\" fitted values residuals found functions fitted.cca residuals.cca. residuals = \"standardized\" residuals found rstandard.cca, residuals = \"studentized\" found rstudent.cca, cases fitted values standardized sigma.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function returns Lattice object can displayed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ordiresids(mod) ordiresids(mod, formula = Residuals ~ Fitted | Species, residuals=\"standard\", cex = 0.5)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":null,"dir":"Reference","previous_headings":"","what":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Automatic stepwise model building constrained ordination methods (cca, rda, dbrda, capscale). function ordistep modelled step can forward, backward stepwise model selection using permutation tests. Function ordiR2step performs forward model choice solely adjusted \\(R^2\\) \\(P\\)-value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"ordistep(object, scope, direction = c(\"both\", \"backward\", \"forward\"), Pin = 0.05, Pout = 0.1, permutations = how(nperm = 199), steps = 50, trace = TRUE, ...) ordiR2step(object, scope, Pin = 0.05, R2scope = TRUE, permutations = how(nperm = 499), trace = TRUE, R2permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"object ordistep, ordination object inheriting cca rda. scope Defines range models examined stepwise search. can list containing components upper lower, formulae. single item, interpreted target scope, depending direction. direction \"forward\", single item interpreted upper scope formula input object lower scope. See step details. ordiR2step, defines upper scope; can also ordination object model extracted. direction mode stepwise search, can one \"\", \"backward\", \"forward\", default \"\". scope argument missing, default direction \"backward\" ordistep ( ordiR2step argument, works forward). Pin, Pout Limits permutation \\(P\\)-values adding (Pin) term model, dropping (Pout) model. Term added \\(P \\le\\) Pin, removed \\(P >\\) Pout. R2scope Use adjusted \\(R^2\\) stopping criterion: models lower adjusted \\(R^2\\) scope accepted. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. passed anova.cca: see details. steps Maximum number iteration steps dropping adding terms. trace positive, information printed model building. Larger values may give information. R2permutations Number permutations used estimation adjusted \\(R^2\\) cca using RsquareAdj. ... additional arguments add1.cca drop1.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"basic functions model choice constrained ordination add1.cca drop1.cca. functions, ordination models can chosen standard R function step bases term choice AIC. AIC-like statistics ordination provided functions deviance.cca extractAIC.cca ( similar functions rda). Actually, constrained ordination methods AIC, therefore step may trusted. function provides alternative using permutation \\(P\\)-values. Function ordistep defines model, scope models considered, direction procedure similarly step. function alternates drop add steps stops model changed one step. - + signs summary table indicate stage performed. often sensible Pout \\(>\\) Pin stepwise models avoid cyclic adds drops single terms. Function ordiR2step builds model forward maximizes adjusted \\(R^2\\) (function RsquareAdj) every step, stopping adjusted \\(R^2\\) starts decrease, adjusted \\(R^2\\) scope exceeded, selected permutation \\(P\\)-value exceeded (Blanchet et al. 2008). second criterion ignored option R2scope = FALSE, third criterion can ignored setting Pin = 1 (higher). function used adjusted \\(R^2\\) calculated. number predictors higher number observations, adjusted \\(R^2\\) also unavailable. models can analysed R2scope = FALSE, variable selection stop models become overfitted adjusted \\(R^2\\) calculated, adjusted \\(R^2\\) reported zero. \\(R^2\\) cca based simulations (see RsquareAdj) different runs ordiR2step can give different results. Functions ordistep (based \\(P\\) values) ordiR2step (based adjusted \\(R^2\\) hence eigenvalues) can select variables different order.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Functions return selected model one additional component, anova, contains brief information steps taken. can suppress voluminous output model building setting trace = FALSE, find summary model history anova item.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Blanchet, F. G., Legendre, P. & Borcard, D. (2008) Forward selection explanatory variables. Ecology 89, 2623--2632.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"## See add1.cca for another example ### Dune data data(dune) data(dune.env) mod0 <- rda(dune ~ 1, dune.env) # Model with intercept only mod1 <- rda(dune ~ ., dune.env) # Model with all explanatory variables ## With scope present, the default direction is \"both\" mod <- ordistep(mod0, scope = formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.035 * #> + Use 2 91.032 1.1741 0.300 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.085 . #> + A1 1 87.424 1.2965 0.170 #> + Use 2 88.284 1.0510 0.355 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.010 ** #> - Management 3 87.707 2.1769 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.535 #> + Use 2 86.842 0.8027 0.785 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## summary table of steps mod$anova #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 85.567 1.9764 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Example of ordistep, forward ordistep(mod0, scope = formula(mod1), direction=\"forward\") #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.020 * #> + Use 2 91.032 1.1741 0.290 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.055 . #> + A1 1 87.424 1.2965 0.225 #> + Use 2 88.284 1.0510 0.410 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.595 #> + Use 2 86.842 0.8027 0.730 #> #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## Example of ordiR2step (always forward) ## stops because R2 of 'mod1' exceeded ordiR2step(mod0, mod1) #> Step: R2.adj= 0 #> Call: dune ~ 1 #> #> R2.adjusted #> 0.32508817 #> + Management 0.22512409 #> + Moisture 0.20050225 #> + Manure 0.16723149 #> + A1 0.04626579 #> + Use 0.01799755 #> 0.00000000 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.84 0.002 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: R2.adj= 0.2251241 #> Call: dune ~ Management #> #> R2.adjusted #> + Moisture 0.3450334 #> 0.3250882 #> + Manure 0.2779515 #> + A1 0.2392216 #> + Use 0.2300349 #> 0.2251241 #> #> Call: rda(formula = dune ~ Management, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 29.2307 0.3475 3 #> Unconstrained 54.8930 0.6525 16 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 14.865 10.690 3.675 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 #> 15.270 8.428 6.899 5.675 3.988 3.121 2.588 2.380 1.818 1.376 0.995 #> PC12 PC13 PC14 PC15 PC16 #> 0.785 0.661 0.467 0.283 0.159 #>"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface given variable plots result ordination diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"# S3 method for default ordisurf(x, y, choices = c(1, 2), knots = 10, family = \"gaussian\", col = \"red\", isotropic = TRUE, thinplate = TRUE, bs = \"tp\", fx = FALSE, add = FALSE, display = \"sites\", w = weights(x, display), main, nlevels = 10, levels, npoints = 31, labcex = 0.6, bubble = FALSE, cex = 1, select = TRUE, method = \"REML\", gamma = 1, plot = TRUE, lwd.cl = par(\"lwd\"), ...) # S3 method for formula ordisurf(formula, data, ...) # S3 method for ordisurf calibrate(object, newdata, ...) # S3 method for ordisurf plot(x, what = c(\"contour\",\"persp\",\"gam\"), add = FALSE, bubble = FALSE, col = \"red\", cex = 1, nlevels = 10, levels, labcex = 0.6, lwd.cl = par(\"lwd\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"x ordisurf ordination configuration, either matrix result known scores. plot.ordisurf object class \"ordisurf\" returned ordisurf. y Variable plotted / modelled function ordination scores. choices Ordination axes. knots Number initial knots gam (one degrees freedom). knots = 0 knots = 1 function fit linear trend surface, knots = 2 function fit quadratic trend surface instead smooth surface. vector length 2 allowed isotropic = FALSE, first second elements knots referring first second ordination dimensions (indicated choices) respectively. family Error distribution gam. col Colour contours. isotropic, thinplate Fit isotropic smooth surface (.e. smoothness ordination dimensions) via gam. Use thinplate deprecated removed future version package. bs two letter character string indicating smoothing basis use. (e.g. \"tp\" thin plate regression spline, \"cr\" cubic regression spline). One c(\"tp\", \"ts\", \"cr\", \"cs\", \"ds\", \"ps\", \"ad\"). See smooth.terms view refer . default use thin plate splines: bs = \"tp\". fx indicates whether smoothers fixed degree freedom regression splines (fx = FALSE) penalised regression splines (fx = TRUE). Can vector length 2 anisotropic surfaces (isotropic = FALSE). make sense use fx = TRUE select = TRUE error . warning issued specify fx = TRUE forget use select = FALSE though fitting continues using select = FALSE. add Add contours existing diagram draw new plot? display Type scores known scores: typically \"sites\" ordinary site scores \"lc\" linear combination scores. w Prior weights data. Concerns mainly cca decorana results nonconstant weights. main main title plot, default name plotted variable new plot. nlevels, levels Either vector levels contours drawn, suggested number contours nlevels levels supplied. npoints numeric; number locations evaluate fitted surface. represents number locations dimension. labcex Label size contours. Setting zero suppress labels. bubble Use “bubble plot” points, vary point diameter value plotted variable. bubble numeric, value used maximum symbol size ( cex), bubble = TRUE, value cex gives maximum. minimum size always cex = 0.4. option effect add = FALSE. cex Character expansion plotting symbols. select Logical; specify gam argument \"select\". TRUE gam can add extra penalty term can penalized zero. means smoothing parameter estimation part fitting can completely remove terms model. corresponding smoothing parameter estimated zero extra penalty effect. method character; smoothing parameter estimation method. Options allowed : \"GCV.Cp\" uses GCV models unknown scale parameter Mallows' Cp/UBRE/AIC models known scale; \"GACV.Cp\" \"GCV.Cp\" uses GACV (Generalised Approximate CV) instead GCV; \"REML\" \"ML\" use restricted maximum likelihood maximum likelihood estimation known unknown scale; \"P-REML\" \"P-ML\" use REML ML estimation use Pearson estimate scale. gamma Multiplier inflate model degrees freedom GCV UBRE/AIC score . effectively places extra penalty complex models. oft-used value gamma = 1.4. plot logical; plotting done ordisurf? Useful want fitted response surface model. lwd.cl numeric; lwd (line width) parameter use drawing contour lines. formula, data Alternative definition fitted model x ~ y, left-hand side ordination x right-hand side single fitted continuous variable y. variable y must working environment data frame environment given data. arguments passed default method. object ordisurf result object. newdata Coordinates two-dimensional ordination new points. character; type plot produce. \"contour\" produces contour plot response surface, see contour details. \"persp\" produces perspective plot , see persp details. \"gam\" plots fitted GAM model, object inherits class \"gam\" returned ordisurf, see plot.gam. ... parameters passed scores, graphical functions. See Note exceptions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface using penalised splines (Wood 2003) gam, uses predict.gam find fitted values regular grid. smooth surface can fitted extra penalty allows entire smoother penalized back 0 degrees freedom, effectively removing term model (see Marra & Wood, 2011). addition extra penalty invoked setting argument select TRUE. alternative use spline basis includes shrinkage (bs = \"ts\" bs = \"cs\"). ordisurf() exposes large number options gam specifying basis functions used surface. stray defaults, read Notes section relevant documentation s smooth.terms. function plots fitted contours convex hull data points either existing ordination diagram draws new plot. select = TRUE smooth effectively penalised model, contours plotted. gam determines degree smoothness fitted response surface model fitting, unless fx = TRUE. Argument method controls gam performs smoothness selection. See gam details available options. Using \"REML\" \"ML\" yields p-values smooths best coverage properties things matter . function uses scores extract ordination scores, x can result object known function. user can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. behaviour consistent envfit. complete accordance constrained cca, set display = \"lc\". Function calibrate returns fitted values response variable. newdata must coordinates points fitted values desired. function based predict.gam pass extra arguments function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"ordisurf usually called side effect drawing contour plot. function returns result object class \"ordisurf\" inherits gam used internally fit surface, adds item grid contains data grid surface. item grid elements x y vectors axis coordinates, element z matrix fitted values contour. values outside convex hull observed points indicated NA z. gam component result can used analysis like predicting new values (see predict.gam).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Dave Roberts, Jari Oksanen Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"default use isotropic smoother via s employing thin plate regression splines (bs = \"tp\"). make sense ordination equal smoothing directions rotation invariant. However, different degrees smoothness along dimensions required, anisotropic smooth surface may applicable. can achieved use isotropic = FALSE, wherein surface fitted via tensor product smoother via te (unless bs = \"ad\", case separate splines dimension fitted using s). Cubic regression splines P splines can used isotropic = FALSE. Adaptive smooths (bs = \"ad\"), especially two dimensions, require large number observations; without many hundreds observations, default complexities smoother exceed number observations fitting fail. get old behaviour ordisurf use select = FALSE, method = \"GCV.Cp\", fx = FALSE, bs = \"tp\". latter two options current defaults. Graphical arguments supplied plot.ordisurf passed underlying plotting functions, contour, persp, plot.gam. exception arguments col cex can currently passed plot.gam bug way function evaluates arguments arranging plot. work-around call plot.gam directly result call ordisurf. See Examples illustration .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"fitted GAM regression model usual assumptions models. particular note assumption independence residuals. observations independent (e.g. repeat measures set objects, experimental design, inter alia) trust p-values GAM output. need control (.e. add additional fixed effects model, use complex smoothers), extract ordination scores using scores function generate gam call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Marra, G.P & Wood, S.N. (2011) Practical variable selection generalized additive models. Comput. Stat. Data Analysis 55, 2372--2387. Wood, S.N. (2003) Thin plate regression splines. J. R. Statist. Soc. B 65, 95--114.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(varespec) vare.mds <- monoMDS(vare.dist) ## IGNORE_RDIFF_BEGIN ordisurf(vare.mds ~ Baresoil, varechem, bubble = 5) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## as above but without the extra penalties on smooth terms, ## and using GCV smoothness selection (old behaviour of `ordisurf()`): ordisurf(vare.mds ~ Baresoil, varechem, col = \"blue\", add = TRUE, select = FALSE, method = \"GCV.Cp\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.34 total = 8.34 #> #> GCV score: 154.2897 ## Cover of Cladina arbuscula fit <- ordisurf(vare.mds ~ Cladarbu, varespec, family=quasipoisson) ## Get fitted values calibrate(fit) #> 18 15 24 27 23 19 22 #> 21.6245333 8.0344376 3.8614798 2.4595115 6.3958275 5.4260377 6.7158154 #> 16 28 13 14 20 25 7 #> 11.6759768 0.8041457 31.1726268 16.1942820 9.5897933 5.4219853 29.8214589 #> 5 6 3 4 2 9 12 #> 22.8500283 29.9936019 7.1653707 15.5309226 2.8467735 0.9051372 3.5541639 #> 10 11 21 #> 1.3099294 10.7783687 0.9177922 ## Variable selection via additional shrinkage penalties ## This allows non-significant smooths to be selected out ## of the model not just to a linear surface. There are 2 ## options available: ## - option 1: `select = TRUE` --- the *default* ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", select = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## - option 2: use a basis with shrinkage ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", bs = \"ts\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ts\", fx = FALSE) #> #> Estimated degrees of freedom: #> 6.26 total = 7.26 #> #> REML score: 98.83943 ## or bs = \"cs\" with `isotropic = FALSE` ## IGNORE_RDIFF_END ## Plot method plot(fit, what = \"contour\") ## Plotting the \"gam\" object plot(fit, what = \"gam\") ## 'col' and 'cex' not passed on ## or via plot.gam directly library(mgcv) #> Loading required package: nlme #> This is mgcv 1.9-1. For overview type 'help(\"mgcv-package\")'. plot.gam(fit, cex = 2, pch = 1, col = \"blue\") ## 'col' effects all objects drawn... ### controlling the basis functions used ## Use Duchon splines ordisurf(vare.mds ~ Baresoil, varechem, bs = \"ds\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ds\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94117 ## A fixed degrees of freedom smooth, must use 'select = FALSE' ordisurf(vare.mds ~ Baresoil, varechem, knots = 4, fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 4, bs = \"tp\", fx = TRUE) #> #> Estimated degrees of freedom: #> 3 total = 4 #> #> REML score: 85.55126 ## An anisotropic smoother with cubic regression spline bases ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cr\", knots = 4) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(4, 4), bs = c(\"cr\", \"cr\"), fx = c(FALSE, #> FALSE)) #> #> Estimated degrees of freedom: #> 3.08 total = 4.08 #> #> REML score: 93.02782 ## An anisotropic smoother with cubic regression spline with ## shrinkage bases & different degrees of freedom in each dimension ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cs\", knots = c(3,4), fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(3, 4), bs = c(\"cs\", \"cs\"), fx = c(TRUE, #> TRUE)) #> #> Estimated degrees of freedom: #> 11 total = 12 #> #> REML score: 42.9834"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text or Points to Ordination Plots — orditorp","title":"Add Text or Points to Ordination Plots — orditorp","text":"function adds text points ordination plots. Text used can done without overwriting text labels, points used otherwise. function can help reducing clutter ordination graphics, manual editing may still necessary.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"orditorp(x, display, labels, choices = c(1, 2), priority, select, cex = 0.7, pcex, col = par(\"col\"), pcol, pch = par(\"pch\"), air = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text or Points to Ordination Plots — orditorp","text":"x result object ordination ordiplot result. display Items displayed plot. one alternative allowed. Typically \"sites\" \"species\". labels Optional text used labels. Row names used missing. choices Axes shown. priority Text used items higher priority labels overlap. vector length number items plotted. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. logical vector used, must length scores plotted. cex, pcex Text point sizes, see plot.default.. col, pcol Text point colours, see plot.default. pch Plotting character, see points. air Amount empty space text labels. Values <1 allow overlapping text. ... arguments scores (various methods), text points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text or Points to Ordination Plots — orditorp","text":"Function orditorp add either text points existing plot. items high priority added first text used can done without overwriting previous labels,points used otherwise. priority missing, labels added outskirts centre. Function orditorp can used ordination results, plotting results ordiplot ordination plot functions (plot.cca, plot.decorana, plot.metaMDS). Arguments can passed relevant scores method ordination object (x) drawn. See relevant scores help page arguments can used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add Text or Points to Ordination Plots — orditorp","text":"function returns invisibly logical vector TRUE means item labelled text FALSE means marked point. returned vector can used select argument ordination text points functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text or Points to Ordination Plots — orditorp","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"## A cluttered ordination plot : data(BCI) mod <- cca(BCI) plot(mod, dis=\"sp\", type=\"t\") # Now with orditorp and abbreviated species names cnam <- make.cepnames(names(BCI)) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, pch=\"+\", pcol=\"grey\") ## show select in action set.seed(1) take <- sample(ncol(BCI), 50) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, select = take, pch=\"+\", pcol=\"grey\") # \\dontshow{ ## example(orditorp) should not set random seed in the user session rm(.Random.seed) #> Warning: object '.Random.seed' not found # }"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Trellis (Lattice) Plots for Ordination — ordixyplot","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Functions ordicloud, ordisplom ordixyplot provide interface plot ordination results using Trellis functions cloud, splom xyplot package lattice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"ordixyplot(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", aspect = \"iso\", envfit, type = c(\"p\", \"biplot\"), ...) ordisplom(x, data=NULL, formula = NULL, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", type = \"p\", ...) ordicloud(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi3d\", prepanel = \"prepanel.ordi3d\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"x ordination result scores knows: ordination result vegan many others. data Optional data amend ordination results. ordination results found x, may give data variables needed plots. Typically environmental data. formula Formula define plots. default formula used omitted. ordination axes must called names ordination results (names vary among methods). ordisplom, special character . refers ordination result. display kind scores: argument passed scores. choices axes selected: argument passed scores. panel, prepanel names panel prepanel functions. aspect aspect plot (passed lattice function). envfit Result envfit function displayed ordixyplot. Please note needs choices ordixyplot. type type plot. knows alternatives panel.xyplot. addition ordixyplot alternatives \"biplot\", \"arrows\" \"polygon\". first displays fitted vectors factor centroids envfit, constrained ordination, biplot arrows factor centroids envfit given. second (type = \"arrows\") trellis variant ordiarrows draws arrows groups. line parameters controlled trellis.par.set superpose.line, user can set length, angle ends parameters panel.arrows. last one (type = \"polygon\") draws polygon enclosing points panel polygon enclosing points data. overall polygon controlled trellis.par.set plot.polygon, panel polygon controlled superpose.polygon. ... Arguments passed scores methods lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"functions provide interface corresponding lattice functions. graphical parameters passed lattice function graphs extremely configurable. See Lattice xyplot, splom cloud details, usage possibilities. argument x must always ordination result. scores extracted vegan function scores functions work vegan ordinations many others. formula used define models. functions simple default formulae used formula missing. formula omitted ordisplom produces pairs plot ordination axes variables data. formula given, ordination results must referred . variables names. functions, formula must use names ordination scores names data. ordination scores found x, data optional. data contain variables ordination scores used plots. Typically, environmental variables (typically factors) define panels plot symbols. proper work done panel function. layout can changed defining panel functions. See panel.xyplot, panel.splom panel.cloud details survey possibilities. Ordination graphics always isometric: scale used axes. controlled (can changed) argument aspect ordixyplot. ordicloud isometric scaling defined panel prepanel functions. must replace functions want non-isometric scaling graphs. select isometric scaling ordisplom.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"function return Lattice objects class \"trellis\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"data(dune, dune.env) ord <- cca(dune) ## Pairs plots ordisplom(ord) ordisplom(ord, data=dune.env, choices=1:2) ordisplom(ord, data=dune.env, form = ~ . | Management, groups=Manure) ## Scatter plot with polygons ordixyplot(ord, data=dune.env, form = CA1 ~ CA2 | Management, groups=Manure, type = c(\"p\",\"polygon\")) ## Choose a different scaling ordixyplot(ord, scaling = \"symmetric\") ## ... Slices of third axis ordixyplot(ord, form = CA1 ~ CA2 | equal.count(CA3, 4), type = c(\"g\",\"p\", \"polygon\")) ## Display environmental variables ordixyplot(ord, envfit = envfit(ord ~ Management + A1, dune.env, choices=1:3)) ## 3D Scatter plots ordicloud(ord, form = CA2 ~ CA3*CA1, groups = Manure, data = dune.env) ordicloud(ord, form = CA2 ~ CA3*CA1 | Management, groups = Manure, data = dune.env, auto.key = TRUE, type = c(\"p\",\"h\"))"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Coordinates of Neighbourhood Matrix — pcnm","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"function computed classical PCNM principal coordinate analysis truncated distance matrix. commonly used transform (spatial) distances rectangular data suitable constrained ordination regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"pcnm(dis, threshold, w, dist.ret = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"dis distance matrix. threshold threshold value truncation distance. missing, minimum distance giving connected network used. found longest distance minimum spanning tree dis. w Prior weights rows. dist.ret Return distances used calculate PCNMs.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Principal Coordinates Neighbourhood Matrix (PCNM) map distances rows onto rectangular matrix rows using truncation threshold long distances (Borcard & Legendre 2002). original distances Euclidean distances two dimensions (like normal spatial distances), mapped onto two dimensions truncation distances. truncation, higher number principal coordinates. selection truncation distance huge influence PCNM vectors. default use longest distance keep data connected. distances truncation threshold given arbitrary value 4 times threshold. regular data, first PCNM vectors show wide scale variation later PCNM vectors show smaller scale variation (Borcard & Legendre 2002), irregular data interpretation clear. PCNM functions used express distances rectangular form similar normal explanatory variables used , e.g., constrained ordination (rda, cca dbrda) univariate regression (lm) together environmental variables (row weights supplied cca; see Examples). regarded powerful method forcing rectangular environmental data distances using partial mantel analysis (mantel.partial) together geographic distances (Legendre et al. 2008, see Tuomisto & Ruokolainen 2008). function based pcnm function Dray's unreleased spacemakeR package. differences current function uses spantree internal support function. current function also can use prior weights rows using weighted metric scaling wcmdscale. use row weights allows finding orthonormal PCNMs also correspondence analysis (e.g., cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"list following elements: values Eigenvalues obtained principal coordinates analysis. vectors Eigenvectors obtained principal coordinates analysis. scaled unit norm. vectors can extracted scores function. default return PCNM vectors, argument choices selects given vectors. threshold Truncation distance. dist distance matrix values threshold replaced arbitrary value four times threshold. String \"pcnm\" added method attribute, new attribute threshold added distances. returned dist.ret = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Borcard D. Legendre P. (2002) -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153, 51--68. Legendre, P., Borcard, D Peres-Neto, P. (2008) Analyzing explaining beta diversity? Comment. Ecology 89, 3238--3244. Tuomisto, H. & Ruokolainen, K. (2008) Analyzing explaining beta diversity? reply. Ecology 89, 3244--3256.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Jari Oksanen, based code Stephane Dray.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"## Example from Borcard & Legendre (2002) data(mite.xy) pcnm1 <- pcnm(dist(mite.xy)) op <- par(mfrow=c(1,3)) ## Map of PCNMs in the sample plot ordisurf(mite.xy, scores(pcnm1, choi=1), bubble = 4, main = \"PCNM 1\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.71 total = 9.71 #> #> REML score: -120.7705 ordisurf(mite.xy, scores(pcnm1, choi=2), bubble = 4, main = \"PCNM 2\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.18 total = 8.18 #> #> REML score: -103.4662 ordisurf(mite.xy, scores(pcnm1, choi=3), bubble = 4, main = \"PCNM 3\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.32 total = 9.32 #> #> REML score: -94.19053 par(op) ## Plot first PCNMs against each other ordisplom(pcnm1, choices=1:4) ## Weighted PCNM for CCA data(mite) rs <- rowSums(mite)/sum(mite) pcnmw <- pcnm(dist(mite.xy), w = rs) ord <- cca(mite ~ scores(pcnmw)) ## Multiscale ordination: residual variance should have no distance ## trend msoplot(mso(ord, mite.xy))"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":null,"dir":"Reference","previous_headings":"","what":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Individual (count data) incidence (presence-absence data) based null models can generated community level simulations. Options preserving characteristics original matrix (rows/columns sums, matrix fill) restricted permutations (based strata) discussed Details section.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"permatfull(m, fixedmar = \"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, ...) permatswap(m, method = \"quasiswap\", fixedmar=\"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, burnin = 0, thin = 1, ...) # S3 method for permat print(x, digits = 3, ...) # S3 method for permat summary(object, ...) # S3 method for summary.permat print(x, digits = 2, ...) # S3 method for permat plot(x, type = \"bray\", ylab, xlab, col, lty, lowess = TRUE, plot = TRUE, text = TRUE, ...) # S3 method for permat lines(x, type = \"bray\", ...) # S3 method for permat as.ts(x, type = \"bray\", ...) # S3 method for permat toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"m community data matrix plots (samples) rows species (taxa) columns. fixedmar character, stating row/column sums preserved (\"none\", \"rows\", \"columns\", \"\"). strata Numeric vector factor length nrow(m) grouping rows within strata restricted permutations. Unique values levels used. mtype Matrix data type, either \"count\" count data, \"prab\" presence-absence type incidence data. times Number permuted matrices. method Character method used swap algorithm (\"swap\", \"tswap\", \"quasiswap\", \"backtrack\") described function make.commsim. mtype=\"count\" \"quasiswap\", \"swap\", \"swsh\" \"abuswap\" methods available (see details). shuffle Character, indicating whether individuals (\"ind\"), samples (\"samp\") (\"\") shuffled, see details. burnin Number null communities discarded proper analysis sequential (\"swap\", \"tswap\") methods. thin Number discarded permuted matrices two evaluations sequential (\"swap\", \"tswap\") methods. x, object Object class \"permat\" digits Number digits used rounding. ylab, xlab, col, lty graphical parameters plot method. type Character, type plot displayed: \"bray\" Bray-Curtis dissimilarities, \"chisq\" Chi-squared values. lowess, plot, text Logical arguments plot method, whether locally weighted regression curve drawn, plot drawn, statistic values printed plot. ... arguments passed simulate.nullmodel methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"function permatfull useful matrix fill allowed vary, matrix type count. fixedmar argument used set constraints permutation. none margins fixed, cells randomised within matrix. rows columns fixed, cells within rows columns randomised, respectively. margins fixed, r2dtable function used based Patefield's (1981) algorithm. presence absence data, matrix fill necessarily fixed, permatfull wrapper function make.commsim. r00, r0, c0, quasiswap algorithms make.commsim used \"none\", \"rows\", \"columns\", \"\" values fixedmar argument, respectively shuffle argument effect mtype = \"count\" permatfull function used \"none\", \"rows\", \"columns\" values fixedmar. cases count data individual based randomisations. \"samp\" \"\" options result fixed matrix fill. \"\" option means individuals shuffled among non zero cells ensuring cell zeros result, cell (zero new valued cells) shuffled. function permatswap useful matrix fill (.e. proportion empty cells) row/columns sums kept constant. permatswap uses different kinds swap algorithms, row columns sums fixed cases. presence-absence data, swap tswap methods make.commsim can used. count data, special swap algorithm ('swapcount') implemented results permuted matrices fixed marginals matrix fill time. 'quasiswapcount' algorithm (method=\"quasiswap\" mtype=\"count\") uses trick Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function retaining row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping 2x2 submatrices (see 'swapcount' algorithm details) maintain row column totals. algorithm generates independent matrices step, burnin thin arguments considered. default method, sequential (swapcount ) independence subsequent matrices checked. swapcount algorithm (method=\"swap\" mtype=\"count\") tries find 2x2 submatrices (identified 2 random row 2 random column indices), can swapped order leave column row totals fill unchanged. First, algorithm finds largest value submatrix can swapped (\\(d\\)) whether diagonal antidiagonal way. Submatrices contain values larger zero either diagonal antidiagonal position can swapped. Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. algorithm sequential, subsequent matrices independent, swaps modify little matrix large. cases many burnin steps thinning needed get independent random matrices. Although algorithm implemented C, large burnin thin values can slow considerably. WARNING: according simulations, algorithm seems biased non random, thus use avoided! algorithm \"swsh\" function permatswap hybrid algorithm. First, makes binary quasiswaps keep row column incidences constant, non-zero values modified according shuffle argument (\"samp\" \"\" available case, applied non-zero values). also recognizes fixedmar argument \"\" (vegan versions <= 2.0 algorithm fixedmar = \"none\"). algorithm \"abuswap\" produces two kinds null models (based fixedmar=\"columns\" fixedmar=\"rows\") described Hardy (2008; randomization scheme 2x 3x, respectively). preserve column row occurrences, column row sums time. (Note similar constraints can achieved non sequential \"swsh\" algorithm fixedmar argument set \"columns\" \"rows\", respectively.) Constraints row/column sums, matrix fill, total sum sums within strata can checked summary method. plot method visually testing randomness permuted matrices, especially sequential swap algorithms. tendency graph, higher burnin thin values can help sequential methods. New lines can added existing plot lines method. Unrestricted restricted permutations: strata NULL, functions perform unrestricted permutations. Otherwise, used restricted permutations. strata contain least 2 rows order perform randomization (case low row numbers, swap algorithms can rather slow). design well balanced (.e. number observations within stratum), permuted matrices may biased constraints forced submatrices different dimensions. often means, number potential permutations decrease dimensions. constraints put, less randomness can expected. plot method useful graphically testing trend independence permuted matrices. especially important using sequential algorithms (\"swap\", \"tswap\", \"abuswap\"). .ts method can used extract Bray-Curtis dissimilarities Chi-squared values time series. can used testing independence (see Examples). method toCoda useful accessing diagnostic tools available coda package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Functions permatfull permatswap return object class \"permat\" containing function call (call), original data matrix used permutations (orig) list permuted matrices length times (perm). summary method returns various statistics list (including mean Bray-Curtis dissimilarities calculated pairwise among original permuted matrices, Chi-square statistics, check results constraints; see Examples). Note strata used original call, summary calculation may take longer. plot creates plot side effect. .ts method returns object class \"ts\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Original references presence-absence algorithms given help page make.commsim. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Péter Sólymos, solymos@ualberta.ca Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"## A simple artificial community data matrix. m <- matrix(c( 1,3,2,0,3,1, 0,2,1,0,2,1, 0,0,1,2,0,3, 0,0,0,1,4,3 ), 4, 6, byrow=TRUE) ## Using the quasiswap algorithm to create a ## list of permuted matrices, where ## row/columns sums and matrix fill are preserved: x1 <- permatswap(m, \"quasiswap\") summary(x1) #> Summary of object of class 'permat' #> #> Call: permatswap(m = m, method = \"quasiswap\") #> #> Matrix type: count #> Permutation type: swap #> Method: quasiswap_count, burnin: 0, thin: 1 #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 13.13 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3667 0.4333 0.4145 0.4667 0.6000 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 16.27 19.59 21.10 21.51 23.36 31.69 ## Unrestricted permutation retaining ## row/columns sums but not matrix fill: x2 <- permatfull(m) summary(x2) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 16.16 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 0 % #> Column incidences retained: 1.01 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.3667 0.3865 0.4333 0.6333 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 7.824 12.046 15.660 16.015 19.413 28.132 ## Unrestricted permutation of presence-absence type ## not retaining row/columns sums: x3 <- permatfull(m, \"none\", mtype=\"prab\") x3$orig ## note: original matrix is binarized! #> [,1] [,2] [,3] [,4] [,5] [,6] #> [1,] 1 1 1 0 1 1 #> [2,] 0 1 1 0 1 1 #> [3,] 0 0 1 1 0 1 #> [4,] 0 0 0 1 1 1 summary(x3) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, fixedmar = \"none\", mtype = \"prab\") #> #> Matrix type: prab #> Permutation type: full #> Method: r00 #> Restricted: FALSE #> Fixed margins: none #> Individuals and samples are shuffled #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 15 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 4.04 % #> Column sums retained: 0 % #> Row incidences retained: 4.04 % #> Column incidences retained: 0 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.4000 0.3852 0.4000 0.5333 #> #> Chi-squared for original matrix: 8.4 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 8.812 13.583 15.208 15.295 17.083 21.896 ## Restricted permutation, ## check sums within strata: x4 <- permatfull(m, strata=c(1,1,2,2)) summary(x4) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, strata = c(1, 1, 2, 2)) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: TRUE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 38.38 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 2.02 % #> Sums within strata retained: 100 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.06667 0.20000 0.26667 0.23502 0.26667 0.46667 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 14.21 18.68 21.05 22.09 25.26 36.50 ## NOTE: 'times' argument usually needs to be >= 99 ## here much lower value is used for demonstration ## Not sequential algorithm data(BCI) a <- permatswap(BCI, \"quasiswap\", times=19) ## Sequential algorithm b <- permatswap(BCI, \"abuswap\", fixedmar=\"col\", burnin=0, thin=100, times=19) opar <- par(mfrow=c(2,2)) plot(a, main=\"Not sequential\") plot(b, main=\"Sequential\") plot(a, \"chisq\") plot(b, \"chisq\") par(opar) ## Extract Bray-Curtis dissimilarities ## as time series bc <- as.ts(b) ## Lag plot lag.plot(bc) ## First order autoregressive model mar <- arima(bc, c(1,0,0)) mar #> #> Call: #> arima(x = bc, order = c(1, 0, 0)) #> #> Coefficients: #> ar1 intercept #> 0.9915 0.1850 #> s.e. 0.0120 0.1374 #> #> sigma^2 estimated as 0.000346: log likelihood = 46.71, aic = -87.42 ## Ljung-Box test of residuals Box.test(residuals(mar)) #> #> \tBox-Pierce test #> #> data: residuals(mar) #> X-squared = 0.35725, df = 1, p-value = 0.55 #> ## Graphical diagnostics tsdiag(mar)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract, Analyse and Display Permutation Results — permustats","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results vegan functions. support functions can find quantiles standardized effect sizes, plot densities Q-Q plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"permustats(x, ...) # S3 method for permustats summary(object, interval = 0.95, alternative, ...) # S3 method for permustats densityplot(x, data, xlab = \"Permutations\", ...) # S3 method for permustats density(x, observed = TRUE, ...) # S3 method for permustats qqnorm(y, observed = TRUE, ...) # S3 method for permustats qqmath(x, data, observed = TRUE, sd.scale = FALSE, ylab = \"Permutations\", ...) # S3 method for permustats boxplot(x, scale = FALSE, names, ...) # S3 method for permustats pairs(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract, Analyse and Display Permutation Results — permustats","text":"object, x, y object handled. interval numeric; coverage interval reported. alternative character string specifying limits used interval direction test evaluating \\(p\\)-values. Must one \"two.sided\" (upper lower limit), \"greater\" (upper limit), \"less\" (lower limit). Usually alternative given result object, can specified argument. xlab, ylab Arguments densityplot qqmath functions. observed Add observed statistic among permutations. sd.scale Scale permutations unit standard deviation observed statistic standardized effect size. data Ignored. scale Use standardized effect size (SES). names Names boxes (default: names statistics). ... arguments passed function. density passed density.default, boxplot boxplot.default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results observed statistics several vegan functions perform permutations simulations. summary method permustats estimates standardized effect sizes (SES) difference observed statistic mean permutations divided standard deviation permutations (also known \\(z\\)-values). also prints mean, median, limits contain interval percent permuted values. default (interval = 0.95), two-sided test (2.5%, 97.5%) one-sided tests either 5% 95% quantile \\(p\\)-value depending test direction. mean, quantiles \\(z\\) values evaluated permuted values without observed statistic, \\(p\\)-value evaluated observed statistic. intervals \\(p\\)-value evaluated test direction original test, can changed argument alternative. Several permustats objects can combined c function. c function checks statistics equal, performs sanity tests. density densityplot methods display kernel density estimates permuted values. observed value statistic included permuted values, densityplot method marks observed statistic vertical line. However density method uses standard plot method mark observed value. qqnorm qqmath display Q-Q plots permutations, optionally together observed value (default) shown horizontal line plots. qqnorm plots permutation values standard Normal variate. qqmath defaults standard Normal well, can accept alternatives (see standard qqmath). qqmath function can also plot observed statistic standardized effect size (SES) standandized permutations (argument sd.scale). permutations standardized without observed statistic, similarly summary. Functions density qqnorm based standard R methods accept arguments. handle one statistic, used several test statistic evaluated. densityplot qqmath lattice graphics, can used either one several statistics. functions pass arguments underlying functions; see documentation. Functions qqmath densityplot default use axis scaling subplots lattice. can use argument scales set independent scaling subplots appropriate (see xyplot exhaustive list arguments). Function boxplot draws box--whiskers plots effect size, difference permutations observed statistic. scale = TRUE, permutations standardized unit standard deviation, plot show standardized effect sizes. Function pairs plots permutation values statistics . function passes extra arguments pairs. permustats can extract permutation statistics results adonis2, anosim, anova.cca, mantel, mantel.partial, mrpp, oecosimu, ordiareatest, permutest.cca, protest, permutest.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function returns object class \"permustats\". list items \"statistic\" observed statistics, permutations contains permuted values, alternative contains text defining character test (\"two.sided\", \"less\" \"greater\"). qqnorm density methods return standard result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract, Analyse and Display Permutation Results — permustats","text":"Jari Oksanen contributions Gavin L. Simpson (permustats.permutest.betadisper method related modifications summary.permustats print method) Eduard Szöcs (permustats.anova.cca).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"data(dune, dune.env) mod <- adonis2(dune ~ Management + A1, data = dune.env) ## use permustats perm <- permustats(mod) summary(perm) #> #> statistic SES mean lower median upper Pr(perm) #> Management 3.0730 4.6870 1.0387 0.9565 1.8217 0.004 ** #> A1 2.7676 2.7175 1.0022 0.8484 2.2436 0.028 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.95) densityplot(perm) qqmath(perm) boxplot(perm, scale=TRUE, lty=1, pch=16, cex=0.6, col=\"hotpink\", ylab=\"SES\") abline(h=0, col=\"skyblue\") ## example of multiple types of statistic mod <- with(dune.env, betadisper(vegdist(dune), Management)) pmod <- permutest(mod, nperm = 99, pairwise = TRUE) perm <- permustats(pmod) summary(perm, interval = 0.90) #> #> statistic SES mean lower median upper Pr(perm) #> Overall (F) 1.9506 0.7173 1.1427 0.8211 2.4909 0.154 #> BF-HF (t) -0.5634 -0.4124 -0.0443 -2.0202 -0.0293 1.8851 0.591 #> BF-NM (t) -2.2387 -1.8672 -0.0045 -1.8423 0.0074 2.0628 0.067 . #> BF-SF (t) -1.1675 -0.9341 -0.0086 -1.9337 -0.0450 1.9486 0.283 #> HF-NM (t) -2.1017 -1.9328 0.0277 -1.6716 0.0346 1.7582 0.067 . #> HF-SF (t) -0.8789 -0.7872 0.0321 -1.8598 0.0284 1.8394 0.379 #> NM-SF (t) 0.9485 0.8265 0.0121 -1.9118 0.0690 1.7827 0.379 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.9)"},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation tests in Vegan — permutations","title":"Permutation tests in Vegan — permutations","text":"version 2.2-0, vegan significantly improved access restricted permutations brings line offered Canoco. permutation designs modelled permutation schemes Canoco 3.1 (ter Braak, 1990). vegan currently provides following features within permutation tests: Free permutation DATA, also known randomisation, Free permutation DATA within levels grouping variable, Restricted permutations line transects time series, Permutation groups samples whilst retaining within-group ordering, Restricted permutations spatial grids, Blocking, samples never permuted blocks, Split-plot designs, permutation whole plots, split plots, . , use DATA mean either observed data function data, example residuals ordination model presence covariables. capabilities provided functions permute package. user can request particular type permutation supplying permutations argument function object returned , defines samples permuted. Alternatively, user can simply specify required number permutations simple randomisation procedure performed. Finally, user can supply matrix permutations ( number rows equal number permutations number columns equal number observations data) vegan use permutations instead generating new permutations. majority functions vegan allow full range possibilities outlined . Exceptions include kendall.post kendall.global. Null hypothesis first two types permutation test listed assumes free exchangeability DATA (within levels grouping variable, specified). Dependence observations, arises due spatial temporal autocorrelation, -complicated experimental designs, split-plot designs, violates fundamental assumption test requires complex restricted permutation test designs. designs available via permute package vegan provides access version 2.2-0 onwards. Unless otherwise stated help pages specific functions, permutation tests vegan follow format/structure: appropriate test statistic chosen. statistic chosen described help pages individual functions. value test statistic evaluate observed data analysis/model recorded. Denote value \\(x_0\\). DATA randomly permuted according one schemes, value test statistic permutation evaluated recorded. Step 3 repeated total \\(n\\) times, \\(n\\) number permutations requested. Denote values \\(x_i\\), \\(= 1, ..., n\\) Count number values test statistic, \\(x_i\\), Null distribution extreme test statistic observed data \\(x_0\\). Denote count \\(N\\). use phrase extreme include cases two-sided test performed large negative values test statistic considered. permutation p-value computed $$p = \\frac{N + 1}{n + 1}$$ description illustrates default number permutations specified vegan functions takes values 199 999 example. Pretty p values achieved \\(+ 1\\) denominator results division 200 1000, 199 999 random permutations used test. simple intuition behind presence \\(+ 1\\) numerator denominator represent inclusion observed value statistic Null distribution (e.g. Manly 2006). Phipson & Smyth (2010) present compelling explanation inclusion \\(+ 1\\) numerator denominator p value calculation. Fisher (1935) mind permutation test involve enumeration possible permutations data yielding exact test. However, complete enumeration may feasible practice owing potentially vast number arrangements data, even modestly-sized data sets free permutation samples. result evaluate p value tail probability Null distribution test statistic directly random sample possible permutations. Phipson & Smyth (2010) show naive calculation permutation p value $$p = \\frac{N}{n}$$ leads invalid test incorrect type error rate. go show replacing unknown tail probability ( p value) Null distribution biased estimator $$p = \\frac{N + 1}{n + 1}$$ positive bias induced just right size account uncertainty estimation tail probability set randomly sampled permutations yield test correct type error rate. estimator described correct situation permutations data samples randomly without replacement. strictly happens vegan permutations drawn pseudo-randomly independent one another. Note actual chance happening practice small functions permute guarantee generate unique set permutations unless complete enumeration permutations requested. feasible smallest data sets restrictive permutation designs, cases chance drawing set permutations repeats lessened sample size, thence size set possible permutations, increases. situation sampling permutations replacement , tail probability \\(p\\) calculated biased estimator described somewhat conservative, large amount depends number possible values test statistic can take permutation data (Phipson & Smyth, 2010). represents slight loss statistical power conservative p value calculation used . However, unless sample sizes small permutation design set values test statistic can take also small, loss power unlikely critical. minimum achievable p-value $$p_{\\mathrm{min}} = \\frac{1}{n + 1}$$ hence depends number permutations evaluated. However, one simply increase number permutations (\\(n\\)) achieve potentially lower p-value unless number observations available permits number permutations. unlikely problem smallest data sets free permutation (randomisation) valid, restricted permutation designs low number observations, may many unique permutations data might desire reach required level significance. currently responsibility user determine total number possible permutations DATA. number possible permutations allowed specified design can calculated using numPerms permute package. Heuristics employed within shuffleSet function used vegan can triggered generate entire set permutations instead random set. settings controlling triggering complete enumeration step contained within permutation design created using link[permute]{} can set user. See details. Limits total number permutations DATA severe temporally spatially ordered data experimental designs low replication. example, time series \\(n = 100\\) observations just 100 possible permutations including observed ordering. situations low number permutations possible due nature DATA experimental design, enumeration permutations becomes important achievable computationally. , provided brief overview capabilities vegan permute. get best new functionality details set permutation designs using , consult vignette Restricted permutations; using permute package supplied permute accessible via vignette(\"permutations\", package = \"permute\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"random-number-generation","dir":"Reference","previous_headings":"","what":"Random Number Generation","title":"Permutation tests in Vegan — permutations","text":"permutations based random number generator provided R. may change R releases change permutations vegan test results. One change R release 3.6.0. new version clearly better permutation tests use . However, need reproduce old results, can set R random number generator previous version RNGversion.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation tests in Vegan — permutations","text":"Manly, B. F. J. (2006). Randomization, Bootstrap Monte Carlo Methods Biology, Third Edition. Chapman Hall/CRC. Phipson, B., & Smyth, G. K. (2010). Permutation P-values never zero: calculating exact P-values permutations randomly drawn. Statistical Applications Genetics Molecular Biology, 9, Article 39. DOI: 10.2202/1544-6115.1585 ter Braak, C. J. F. (1990). Update notes: CANOCO version 3.1. Wageningen: Agricultural Mathematics Group. (UR). See also: Davison, . C., & Hinkley, D. V. (1997). Bootstrap Methods Application. Cambridge University Press.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation tests in Vegan — permutations","text":"Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Implements permutation-based test multivariate homogeneity group dispersions (variances) results call betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"# S3 method for betadisper permutest(x, pairwise = FALSE, permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"x object class \"betadisper\", result call betadisper. pairwise logical; perform pairwise comparisons group means? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. ... Arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance F. alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution F Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can performed setting argument pairwise TRUE. classical t test performed pairwise group dispersions. combined permutation test based t statistic calculated pairwise group dispersions. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"permutest.betadisper returns list class \"permutest.betadisper\" following components: tab ANOVA table object inheriting class \"data.frame\". pairwise list components observed permuted containing observed permuted p-values pairwise comparisons group mean distances (dispersions variances). groups character; levels grouping factor. control list, result call .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62(1), 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9(6), 683--693.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F pmod <- permutest(mod, permutations = 99, pairwise = TRUE) ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Has permustats() method pstat <- permustats(pmod) densityplot(pstat, scales = list(x = list(relation = \"free\"))) qqmath(pstat, scales = list(relation = \"free\"))"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Functions plot extract results constrained correspondence analysis (cca), redundancy analysis (rda), distance-based redundancy analysis (dbrda) constrained analysis principal coordinates (capscale).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"# S3 method for cca plot(x, choices = c(1, 2), display = c(\"sp\", \"wa\", \"cn\"), scaling = \"species\", type, xlim, ylim, const, correlation = FALSE, hill = FALSE, ...) # S3 method for cca text(x, display = \"sites\", labels, choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca points(x, display = \"sites\", choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", hill = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for rda scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", const, correlation = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for cca summary(object, digits = max(3, getOption(\"digits\") - 3), ...) # S3 method for cca labels(object, display, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"x, object cca result object. choices Axes shown. display Scores shown. must include alternatives \"species\" \"sp\" species scores, sites \"wa\" site scores, \"lc\" linear constraints LC scores, \"bp\" biplot arrows \"cn\" centroids factor constraints instead arrow, \"reg\" regression coefficients (.k.. canonical coefficients). alternative \"\" selects available scores. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. Corresponding negative values can used cca additionally multiply results \\(\\sqrt(1/(1-\\lambda))\\). scaling know Hill scaling (although nothing Hill's rescaling decorana). corresponding negative values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Arguments correlation hill scores.rda scores.cca respectively can used combination character descriptions get corresponding negative value. correlation, hill logical; scaling character description scaling type, correlation hill used select corresponding negative scaling type; either correlation-like scores Hill's scaling PCA/RDA CA/CCA respectively. See argument scaling details. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score, names variable label, weights (CCA) variable weight. possible values score species, sites (WA scores), constraints (LC scores sites calculated directly constraining variables), biplot (biplot arrows), centroids (levels factor variables), factorbiplot (biplot arrows model centroids), regression (regression coefficients find LC scores constraints). scores used conventional plot, directly suitable used ggplot2 package. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. xlim, ylim x y limits (min,max) plot. labels Optional text used instead row names. use , good check default labels order using labels command. arrow.mul Factor expand arrows graph. Arrows scaled automatically fit graph missing. head.arrow Default length arrow heads. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. const General scaling constant rda scores. default use constant gives biplot scores, , scores approximate original data (see vignette ‘Design Decisions’ browseVignettes(\"vegan\") details discussion). const vector two items, first used species, second item site scores. droplist Return matrix instead named list one kind scores requested. axis.bp Draw axis biplot arrows. digits Number digits output. ... Parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function used cca rda. produces quick, standard plot current scaling. plot function sets colours (col), plotting characters (pch) character sizes (cex) certain standard values. fuller control produced plot, best call plot type=\"none\" first, add plotting item separately using text.cca points.cca functions. use default settings standard text points functions accept parameters, allowing full user control produced plots. Environmental variables receive special treatment. display=\"bp\", arrows drawn. labelled text unlabelled points. arrows basically unit scaling, sites scaled (scaling \"sites\" \"symmetric\"), scores requested axes adjusted relative axis highest eigenvalue. scaling = \"species\" scaling = \"none\", arrows consistent vectors fitted linear combination scores (display = \"lc\" function envfit), scaling alternatives differ. basic plot function uses simple heuristics adjusting unit-length arrows current plot area, user can give expansion factor mul.arrow. display=\"cn\" centroids levels factor variables displayed (available factors formula interface used cca rda). option continuous variables still presented arrows ordered factors arrows centroids. display = \"reg\" arrows drawn regression coefficients (.k.. canonical coefficients) constraints conditions. Biplot arrows can interpreted individually, regression coefficients must interpreted together: LC score site sum regressions displayed arrows. partialled conditions zero shown biplot arrows, shown regressions, show effect must partialled get LC scores. biplot arrows standard easily interpreted, regression arrows used know need . want better control plots, best construct plot text points commands accept graphical parameters. important remember use scaling, correlation hill arguments calls. plot.cca command returns invisibly ordiplot result object, consistent scaling elements. easiest way full control graphics first set plot frame using plot type = \"n\" needed scores display save result. points text commands ordiplot allow full graphical control (see section Examples). Utility function labels returns default labels order applied text. Palmer (1993) suggested using linear constraints (“LC scores”) ordination diagrams, gave better results simulations site scores (“WA scores”) step constrained unconstrained analysis. However, McCune (1997) showed noisy environmental variables (environmental measurements noisy) destroy “LC scores” whereas “WA scores” little affected. Therefore plot function uses site scores (“WA scores”) default. consistent usage statistics functions R (lda, cancor).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function returns invisibly plotting structure can used function identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Moisture + Management, dune.env) ## better control -- remember to set scaling etc identically plot(mod, type=\"n\", scaling=\"sites\") text(mod, dis=\"cn\", scaling=\"sites\") points(mod, pch=21, col=\"red\", bg=\"yellow\", cex=1.2, scaling=\"sites\") text(mod, \"species\", col=\"blue\", cex=0.8, scaling=\"sites\") ## catch the invisible result and use ordiplot support - the example ## will make a biplot with arrows for species and correlation scaling pca <- rda(dune) pl <- plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) with(dune.env, points(pl, \"site\", pch=21, col=1, bg=Management)) text(pl, \"sp\", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE) with(dune.env, legend(\"bottomleft\", levels(Management), pch=21, pt.bg=1:4, bty=\"n\")) ## Scaling can be numeric or more user-friendly names ## e.g. Hill's scaling for (C)CA scrs <- scores(mod, scaling = \"sites\", hill = TRUE) ## or correlation-based scores in PCA/RDA scrs <- scores(rda(dune ~ A1 + Moisture + Management, dune.env), scaling = \"sites\", correlation = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Response Curves for Treatments with Repeated Observations — prc","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Principal Response Curves (PRC) special case Redundancy Analysis (rda) multivariate responses repeated observation design. originally suggested ecological communities. easier interpret traditional constrained ordination. can also used study effects factor depend levels factor B, + :B, multivariate response experiment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"prc(response, treatment, time, ...) # S3 method for prc summary(object, axis = 1, scaling = \"sites\", const, digits = 4, correlation = FALSE, ...) # S3 method for prc plot(x, species = TRUE, select, scaling = \"symmetric\", axis = 1, correlation = FALSE, const, type = \"l\", xlab, ylab, ylim, lty = 1:5, col = 1:6, pch, legpos, cex = 0.8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"response Multivariate response data. Typically community (species) data. data counts, probably log transformed prior analysis. treatment factor treatments. time unordered factor defining observations times repeated design. object, x prc result object. axis Axis shown (one axis can selected). scaling Scaling species scores, identical scaling scores.rda. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. const General scaling constant species scores (see scores.rda details). Lower values reduce range species scores, influence regression coefficients. digits Number significant digits displayed. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. species Display species scores. select Vector select displayed species. can vector indices logical vector TRUE selected species type Type plot: \"l\" lines, \"p\" points \"b\" . xlab, ylab Text replace default axis labels. ylim Limits vertical axis. lty, col, pch Line type, colour plotting characters (defaults supplied). legpos position legend. guess made supplied, NA suppress legend. cex Character expansion symbols species labels. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"PRC special case rda single factor treatment single factor time points repeated observations. vegan, corresponding rda model defined rda(response ~ treatment * time + Condition(time)). Since time appears twice model formula, main effects aliased, main effect treatment interaction terms available, used PRC. Instead usual multivariate ordination diagrams, PRC uses canonical (regression) coefficients species scores single axis. current functions provide special summary plot methods display rda results PRC fashion. current version works default contrasts (contr.treatment) coefficients contrasts first level, levels must arranged first level control ( baseline). necessary, must change baseline level function relevel. Function summary prints species scores coefficients. Function plot plots coefficients time using matplot, similar defaults. graph (PRC) meaningful first treatment level control, results contrasts first level unordered factors used. plot also displays species scores right vertical axis using function linestack. Typically number species high can displayed default settings, users can reduce character size padding (air) linestack, select subset species. legend displayed unless suppressed legpos = NA, functions tries guess put legend legpos supplied.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"function special case rda returns result object (see cca.object). However, special summary plot methods display returns differently rda.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Jari Oksanen Cajo ter Braak","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"first level treatment must control: use function relevel guarantee correct reference level. current version ignore user setting contrasts always use treatment contrasts (contr.treatment). time must unordered factor.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"## Chlorpyrifos experiment and experimental design: Pesticide ## treatment in ditches (replicated) and followed over from 4 weeks ## before to 24 weeks after exposure data(pyrifos) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) ditch <- gl(12, 1, length=132) ## IGNORE_RDIFF_BEGIN ## PRC mod <- prc(pyrifos, dose, week) mod # RDA #> Call: prc(response = pyrifos, treatment = dose, time = week) #> #> Inertia Proportion Rank #> Total 288.9920 1.0000 #> Conditional 63.3493 0.2192 10 #> Constrained 96.6837 0.3346 44 #> Unconstrained 128.9589 0.4462 77 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 25.282 8.297 6.044 4.766 4.148 3.857 3.587 3.334 3.087 2.551 2.466 #> RDA12 RDA13 RDA14 RDA15 RDA16 RDA17 RDA18 RDA19 RDA20 RDA21 RDA22 #> 2.209 2.129 1.941 1.799 1.622 1.579 1.440 1.398 1.284 1.211 1.133 #> RDA23 RDA24 RDA25 RDA26 RDA27 RDA28 RDA29 RDA30 RDA31 RDA32 RDA33 #> 1.001 0.923 0.862 0.788 0.750 0.712 0.685 0.611 0.584 0.537 0.516 #> RDA34 RDA35 RDA36 RDA37 RDA38 RDA39 RDA40 RDA41 RDA42 RDA43 RDA44 #> 0.442 0.417 0.404 0.368 0.340 0.339 0.306 0.279 0.271 0.205 0.179 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 17.156 9.189 7.585 6.064 5.730 4.843 4.518 4.105 #> (Showing 8 of 77 unconstrained eigenvalues) #> summary(mod) # PRC #> #> Call: #> prc(response = pyrifos, treatment = dose, time = week) #> Species scores: #> Simve Daplo Cerpu Alogu Aloco Alore Aloaf Copsp #> 2.688099 1.464566 0.542739 0.280040 0.177019 0.315038 0.426524 1.169368 #> Ostsp Slyla Acrha Aloex Chysp Alona Plead Oxyte #> 2.312186 -0.556899 0.105535 0.228092 0.095042 0.063689 0.138397 0.025401 #> Grate Copdi NauLa CilHa Strvi amosp Ascmo Synsp #> 0.096840 1.428854 4.847070 0.895241 3.069709 -1.357663 0.069736 -0.026494 #> Squro Squmu Polar Kerqu Anufi Mytve Mytvi Mytmu #> 0.264390 -0.452667 0.461989 0.495348 0.432767 0.074372 0.090928 0.105891 #> Lepsp Leppa Colob Colbi Colun Lecsp Lecqu Lecco #> 0.998286 0.084809 -0.723051 -0.139569 -0.828338 0.472866 -0.088860 -0.290531 #> Leclu Lecfl Tripo Cepsp Monlo Monae Scalo Trilo #> 0.049788 -0.408035 0.215234 -0.809597 -0.527913 -0.089948 -0.077192 -0.039086 #> Tripo.1 Tricy Trisp Tepat Rotne Notla Filsp Lopox #> 0.246435 0.335400 0.078423 0.007719 0.143730 -0.114301 -0.168356 -0.030990 #> hydrspec bothrosp olchaeta erpoocto glsicomp alglhete hebdstag sphidae #> -0.048698 0.398665 -1.165154 -0.901030 -0.144389 -0.073049 0.902223 1.463655 #> ansuvote armicris bathcont binitent gyraalbu hippcomp lymnstag lymnaes7 #> 0.140685 1.680010 0.073282 -1.950500 0.033051 0.404473 -0.263679 0.135150 #> physfont plbacorn popyanti radiovat radipere valvcris valvpisc hycarina #> -0.026383 0.084761 1.272223 -0.019815 -0.625468 0.010579 -0.267577 1.044034 #> gammpule aselaqua proameri collembo caenhora caenluct caenrobu cloedipt #> 1.526450 1.578743 0.116577 0.029906 5.767844 2.376188 0.126181 4.734035 #> cloesimi aeshniae libellae conagrae corident coripanz coripunc cymabons #> 1.242207 0.212699 -0.081867 1.630574 0.013761 0.120439 -0.176746 0.046360 #> hesplinn hespsahl notoglau notomacu notoobli notoviri pacoconc pleaminu #> -0.069465 0.033240 0.555201 0.050060 0.082356 0.215863 0.016709 0.071168 #> sigadist sigafall sigastri sigarasp colyfusc donacis6 gyrimari haliconf #> 0.076479 -0.018364 0.060206 -0.277312 0.036493 -0.078608 0.010579 0.446911 #> haliflav haligruf haliobli herubrev hya_herm hyglpusi hyhyovat hypoplan #> 0.044538 0.450146 0.131472 -0.128486 -0.322379 -0.011775 0.024196 0.014976 #> hyporusp hytuinae hytuvers laphminu noteclav rhantusp sialluta ablalong #> 0.232042 2.316179 1.772351 0.632897 0.007912 0.067618 1.109341 0.014976 #> ablaphmo cltanerv malopisp mopetenu prdiussp pstavari chironsp crchirsp #> 2.992695 0.075631 0.047658 0.008716 0.554911 0.082842 1.889916 0.016709 #> crclglat ditendsp mitegchl pachgarc pachgvit popegnub popedisp acriluce #> 0.028953 0.083481 0.230629 -0.012187 -0.029907 0.224272 -0.069649 -0.007950 #> chclpige conescut cricotsp liesspec psclbarb psclgsli psclobvi psclplat #> -0.008744 0.821036 0.121530 0.107387 -0.028639 0.601568 -0.362378 -0.052054 #> psclpsil pscladsp cladotsp laa_spec patanysp tatarssp zaa_spec anopmacu #> 0.007339 0.005674 0.539894 0.034105 0.146807 0.669430 0.049949 -0.163731 #> cepogoae chaoobsc cucidae4 tabanusp agdasphr athrater cyrncren holodubi #> 2.555403 2.442310 0.033240 -0.011601 0.271815 0.067618 0.071168 0.094754 #> holopici leceriae lilurhom monaangu mystazur mystloni oecefurv oecelacu #> 0.611618 0.298633 0.009063 0.644295 0.033240 2.998460 0.536628 0.259064 #> triabico paponysp #> 0.088915 0.097788 #> #> Coefficients for dose + week:dose interaction #> which are contrasts to dose 0 #> rows are dose, columns are week #> -4 -1 0.1 1 2 4 8 12 15 #> 0.1 -0.07218 -0.1375 -0.1020 -0.04068 -0.2101 -0.1364 -0.08077 -0.1536 -0.1123 #> 0.9 -0.08106 -0.1935 -0.1936 -0.47699 -0.4977 -0.4306 -0.13532 -0.3548 -0.2408 #> 6 -0.16616 -0.1232 -0.4539 -1.15638 -1.0835 -1.1511 -0.56112 -0.4698 -0.3078 #> 44 -0.13979 -0.1958 -0.7308 -1.26088 -1.2978 -1.4627 -1.29139 -1.0081 -0.7819 #> 19 24 #> 0.1 -0.2163 -0.07835 #> 0.9 -0.1756 -0.15442 #> 6 -0.3293 -0.18227 #> 44 -0.5768 -0.31022 logabu <- colSums(pyrifos) plot(mod, select = logabu > 100) ## IGNORE_RDIFF_END ## Ditches are randomized, we have a time series, and are only ## interested in the first axis ctrl <- how(plots = Plots(strata = ditch,type = \"free\"), within = Within(type = \"series\"), nperm = 99) anova(mod, permutations = ctrl, first=TRUE) #> Permutation test for rda under reduced model #> Plots: ditch, plot permutation: free #> Permutation: series #> Number of permutations: 99 #> #> Model: prc(response = pyrifos, treatment = dose, time = week) #> Df Variance F Pr(>F) #> RDA1 1 25.282 15.096 0.01 ** #> Residual 77 128.959 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function predict can used find site species scores estimates response data new data sets, Function calibrate estimates values constraints new data set. Functions fitted residuals return estimates response data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"# S3 method for cca fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\"), type = c(\"response\", \"working\"), ...) # S3 method for capscale fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\", \"Imaginary\"), type = c(\"response\", \"working\"), ...) # S3 method for cca residuals(object, ...) # S3 method for cca predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", hill = FALSE, ...) # S3 method for rda predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", correlation = FALSE, const, ...) # S3 method for dbrda predict(object, newdata, type = c(\"response\", \"lc\", \"wa\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", const, ...) # S3 method for cca calibrate(object, newdata, rank = \"full\", ...) # S3 method for cca coef(object, norm = FALSE, ...) # S3 method for decorana predict(object, newdata, type = c(\"response\", \"sites\", \"species\"), rank = 4, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"object result object cca, rda, dbrda, capscale decorana. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. fitted method capscale can also \"Imaginary\" imaginary components negative eigenvalues newdata New data frame used prediction calibration. Usually new community data frame, type = \"lc\" constrained component type = \"response\" type = \"working\" must data frame constraints. newdata must number rows original community data cca result type = \"response\" type = \"working\". original model row column names, new data must contain rows columns names (row names species scores, column names \"wa\" scores constraint names \"lc\" scores). cases rows columns must match directly. argument implemented \"wa\" scores dbrda. type type prediction, fitted values residuals: \"response\" scales results ordination gives results, \"working\" gives values used internally, Chi-square standardization cca scaling centring rda. capscale dbrda \"response\" gives dissimilarities, \"working\" internal data structure analysed ordination. Alternative \"wa\" gives site scores weighted averages community data, \"lc\" site scores linear combinations environmental data, \"sp\" species scores. predict.decorana alternatives scores \"sites\" \"species\". rank rank number axes used approximation. default use axes (full rank) \"model\" available four axes predict.decorana. scaling logical, character, numeric; Scaling predicted scores meaning cca, rda, dbrda, capscale. See scores.cca details acceptable values. correlation, hill logical; correlation-like scores Hill's scaling appropriate RDA CCA respectively. See scores.cca additional details. const Constant multiplier RDA scores. used scaling FALSE, default value give similar scaling scores.rda. norm Coefficients variables centred scaled unit norm. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function fitted gives approximation original data matrix dissimilarities ordination result either scale response scaled internally function. Function residuals gives approximation original data unconstrained ordination. argument type = \"response\" fitted.cca residuals.cca function give marginal totals original data matrix, fitted residuals add original data. Functions fitted residuals dbrda capscale give dissimilarities type = \"response\", additive. However, \"working\" scores additive capscale ( dbrda). fitted residuals capscale dbrda include additive constant requested function call. variants fitted residuals defined model mod <- cca(y ~ x), cca(fitted(mod)) equal constrained ordination, cca(residuals(mod)) equal unconstrained part ordination. Function predict can find estimate original data matrix dissimilarities (type = \"response\") rank. rank = \"full\" identical fitted. addition, function can find species scores site scores community data matrix cca rda. function can used new data, can used add new species site scores existing ordinations. function returns (weighted) orthonormal scores default, must specify explicit scaling add scores ordination diagrams. type = \"wa\" function finds site scores species scores. case, new data can contain new sites, species must match original new data. type=\"sp\" function finds species scores site constraints (linear combination scores). case new data can contain new species, sites must match original new data. type = \"lc\" function finds linear combination scores sites environmental data. case new data frame must contain constraining conditioning environmental variables model formula. type = \"response\" type = \"working\" new data must contain environmental variables constrained component desired, community data matrix residual unconstrained component desired. types, function uses newdata find new \"lc\" (constrained) \"wa\" scores (unconstrained) finds response working data new row scores species scores. original site (row) species (column) weights used type = \"response\" type = \"working\" correspondence analysis (cca) therefore number rows must match original data newdata. completely new data frame created, extreme care needed defining variables similarly original model, particular (ordered) factors. ordination performed formula interface, newdata can data frame matrix, extreme care needed columns match original newdata. Function calibrate.cca finds estimates constraints community ordination \"wa\" scores cca, rda capscale. often known calibration, bioindication environmental reconstruction. Basically, method similar projecting site scores onto biplot arrows, uses regression coefficients. function can called newdata cross-validation possible. newdata may contain new sites, species must match original new data. function work ‘partial’ models Condition term, used newdata capscale dbrda results. results may interpretable continuous variables. Function coef give regression coefficients centred environmental variables (constraints conditions) linear combination scores. coefficients unstandardized environmental variables. coefficients NA aliased effects. Function predict.decorana similar predict.cca. However, type = \"species\" available detrended correspondence analysis (DCA), detrending destroys mutual reciprocal averaging (except first axis rescaling used). Detrended CA attempt approximate original data matrix, type = \"response\" meaning detrended analysis (except rank = 1).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"functions return matrices, vectors dissimilarities appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) # Definition of the concepts 'fitted' and 'residuals' mod #> Call: cca(formula = dune ~ A1 + Management + Condition(Moisture), data #> = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Conditional 0.6283 0.2970 3 #> Constrained 0.5109 0.2415 4 #> Unconstrained 0.9761 0.4615 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 #> 0.24932 0.12090 0.08160 0.05904 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> cca(fitted(mod)) #> Call: cca(X = fitted(mod)) #> #> Inertia Rank #> Total 0.5109 #> Unconstrained 0.5109 4 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 #> 0.24932 0.12090 0.08160 0.05904 #> cca(residuals(mod)) #> Call: cca(X = residuals(mod)) #> #> Inertia Rank #> Total 0.9761 #> Unconstrained 0.9761 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> # Remove rare species (freq==1) from 'cca' and find their scores # 'passively'. freq <- specnumber(dune, MARGIN=2) freq #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 7 10 2 8 6 6 5 1 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 1 2 5 6 1 3 5 4 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 12 7 14 13 6 5 7 3 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 18 3 16 3 15 3 mod <- cca(dune[, freq>1] ~ A1 + Management + Condition(Moisture), dune.env) ## IGNORE_RDIFF_BEGIN predict(mod, type=\"sp\", newdata=dune[, freq==1], scaling=\"species\") #> CCA1 CCA2 CCA3 CCA4 #> Chenalbu 1.5737337 0.7842538 0.5503660 -0.35108333 #> Cirsarve 0.5945146 0.3714228 -0.2862647 -0.88373727 #> Empenigr -1.8771953 0.9904299 -0.2446222 -0.04858656 # New sites predict(mod, type=\"lc\", new=data.frame(A1 = 3, Management=\"NM\", Moisture=\"2\"), scal=2) #> CCA1 CCA2 CCA3 CCA4 #> 1 -2.38829 1.230652 -0.2363485 0.3338258 # Calibration and residual plot mod <- cca(dune ~ A1 + Moisture, dune.env) pred <- calibrate(mod) pred #> A1 Moisture.L Moisture.Q Moisture.C #> 1 2.2630533 -0.62633470 -0.20456759 0.220761764 #> 2 4.0510042 -0.47341146 -0.36986691 0.474939409 #> 3 4.2752294 -0.07214500 -0.60797514 0.303213289 #> 4 4.5398659 0.03192745 -1.12417368 0.932223234 #> 5 5.0409406 -0.84235946 0.43000738 -0.291599200 #> 6 5.1962100 -0.91316862 1.11354235 -0.804453944 #> 7 4.2452549 -0.76452556 0.60464291 -0.484842066 #> 8 5.0208369 0.43886340 0.08169514 0.132995916 #> 9 4.2663219 0.10720486 -0.34067849 -0.675151598 #> 10 4.0411356 -0.65472729 0.02832164 0.558402684 #> 11 2.8280051 -0.45762457 0.63079135 -0.089977975 #> 12 5.1204137 0.36328912 -0.69118581 -0.665622948 #> 13 4.9034218 0.47069541 -0.54378271 -0.118643453 #> 14 11.6455841 0.60920550 0.78341426 0.532852308 #> 15 10.7829689 0.69208513 0.82190786 0.237311062 #> 16 7.9892176 0.96421599 0.46793089 0.373647014 #> 17 0.9218684 -0.15822891 0.14593271 1.189161582 #> 18 3.1680733 -0.41737900 1.03352732 -0.236938282 #> 19 -1.2003506 0.57033354 0.72777285 0.509955590 #> 20 4.7876770 1.00324330 1.49898460 0.009202396 ## IGNORE_RDIFF_END with(dune.env, plot(A1, pred[,\"A1\"] - A1, ylab=\"Prediction Error\")) abline(h=0)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":null,"dir":"Reference","previous_headings":"","what":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes rotates configuration maximum similarity another configuration. Function protest tests non-randomness (significance) two configurations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"procrustes(X, Y, scale = TRUE, symmetric = FALSE, scores = \"sites\", ...) # S3 method for procrustes summary(object, digits = getOption(\"digits\"), ...) # S3 method for procrustes plot(x, kind=1, choices=c(1,2), to.target = TRUE, type = \"p\", xlab, ylab, main, ar.col = \"blue\", length=0.05, cex = 0.7, ...) # S3 method for procrustes points(x, display = c(\"target\", \"rotated\"), choices = c(1,2), truemean = FALSE, ...) # S3 method for procrustes text(x, display = c(\"target\", \"rotated\"), choices = c(1,2), labels, truemean = FALSE, ...) # S3 method for procrustes lines(x, type = c(\"segments\", \"arrows\"), choices = c(1, 2), truemean = FALSE, ...) # S3 method for procrustes residuals(object, ...) # S3 method for procrustes fitted(object, truemean = TRUE, ...) # S3 method for procrustes predict(object, newdata, truemean = TRUE, ...) protest(X, Y, scores = \"sites\", permutations = how(nperm = 999), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"X Target matrix Y Matrix rotated. scale Allow scaling axes Y. symmetric Use symmetric Procrustes statistic (rotation still non-symmetric). scores Kind scores used. display argument used corresponding scores function: see scores, scores.cca scores.cca alternatives. x, object object class procrustes. digits Number digits output. kind plot function, kind plot produced: kind = 1 plots shifts two configurations, kind = 0 draws corresponding empty plot, kind = 2 plots impulse diagram residuals. choices Axes (dimensions) plotted. xlab, ylab Axis labels, defaults unacceptable. main Plot title, default unacceptable. display Show \"target\" \"rotated\" matrix points. .target Draw arrows point target. type type plot drawn. plot, type can \"points\" \"text\" select marker tail arrow, \"none\" drawing empty plot. lines type selects either arrows line segments connect target rotated configuration. truemean Use original range target matrix instead centring fitted values. Function plot.procrustes needs truemean = FALSE, adding graphical items plots original results may need truemean = TRUE. newdata Matrix coordinates rotated translated target. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ar.col Arrow colour. length Width arrow head. labels Character vector text labels. Rownames result object used default. cex Character expansion points text. ... parameters passed functions. procrustes protest parameters passed scores, graphical functions underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Procrustes rotation rotates matrix maximum similarity target matrix minimizing sum squared differences. Procrustes rotation typically used comparison ordination results. particularly useful comparing alternative solutions multidimensional scaling. scale=FALSE, function rotates matrix Y. scale=TRUE, scales linearly configuration Y maximum similarity. Since Y scaled fit X, scaling non-symmetric. However, symmetric=TRUE, configurations scaled equal dispersions symmetric version Procrustes statistic computed. Instead matrix, X Y can results ordination scores can extract results. Function procrustes passes extra arguments scores, scores.cca etc. can specify arguments scaling. Function plot plots procrustes object returns invisibly ordiplot object function identify.ordiplot can used identifying points. items ordiplot object called heads points kind=1 (ordination diagram) sites kind=2 (residuals). ordination diagrams, arrow heads point target configuration .target = TRUE, rotated configuration .target = FALSE. Target original rotated axes shown cross hairs two-dimensional Procrustes analysis, higher number dimensions, rotated axes projected onto plot scaled centred range. Function plot passes parameters underlying plotting functions. full control plots, can draw axes using plot kind = 0, add items points lines. functions pass parameters underlying functions can select plotting characters, size, colours etc., can select width, colour type line segments arrows, can select orientation head width arrows. Function residuals returns pointwise residuals, fitted fitted values, either centred zero mean (truemean=FALSE) original scale ( hardly make sense symmetric = TRUE). addition, summary print methods. matrix X lower number columns matrix Y, matrix X filled zero columns match dimensions. means function can used rotate ordination configuration environmental variable ( practically extracting result fitted function). Function predict can used add new rotated coordinates target. predict function always translate coordinates original non-centred matrix. function used newdata symmetric analysis. Function protest performs symmetric Procrustes analysis repeatedly estimate significance Procrustes statistic. Function protest uses correlation-like statistic derived symmetric Procrustes sum squares \\(ss\\) \\(r =\\sqrt{1-ss}\\), also prints sum squares symmetric analysis, sometimes called \\(m_{12}^2\\). Function protest print method, otherwise uses procrustes methods. Thus plot protest object yields Procrustean superimposition plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes returns object class procrustes items. Function protest inherits procrustes, amends new items: Yrot Rotated matrix Y. X Target matrix. ss Sum squared differences X Yrot. rotation Orthogonal rotation matrix. translation Translation origin. scale Scaling factor. xmean centroid target. symmetric Type ss statistic. call Function call. t0 following items class protest: Procrustes correlation non-permuted solution. t Procrustes correlations permutations. distribution correlations can inspected permustats function. signif Significance t permutations Number permutations. control list control values permutations returned function . control list passed argument control describing permutation design.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate Analysis. Academic Press. Peres-Neto, P.R. Jackson, D.. (2001). well multivariate data sets match? advantages Procrustean superimposition approach Mantel test. Oecologia 129: 169-178.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"function protest follows Peres-Neto & Jackson (2001), implementation still Mardia et al. (1979).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"## IGNORE_RDIFF_BEGIN data(varespec) vare.dist <- vegdist(wisconsin(varespec)) mds.null <- monoMDS(vare.dist, y = cmdscale(vare.dist)) mds.alt <- monoMDS(vare.dist) vare.proc <- procrustes(mds.alt, mds.null) vare.proc #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Procrustes sum of squares: #> 11.17 #> summary(vare.proc) #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Number of objects: 24 Number of dimensions: 2 #> #> Procrustes sum of squares: #> 11.17448 #> Procrustes root mean squared error: #> 0.6823512 #> Quantiles of Procrustes errors: #> Min 1Q Median 3Q Max #> 0.1642438 0.2425785 0.2783603 0.4983976 2.4447632 #> #> Rotation matrix: #> [,1] [,2] #> [1,] 0.99937107 -0.03546065 #> [2,] 0.03546065 0.99937107 #> #> Translation of averages: #> [,1] [,2] #> [1,] -1.713781e-17 1.56769e-17 #> #> Scaling of target: #> [1] 0.7310245 #> plot(vare.proc) plot(vare.proc, kind=2) residuals(vare.proc) #> 18 15 24 27 23 19 22 16 #> 0.2734040 0.2032392 0.4708118 0.4420710 0.3547337 0.1642438 0.2515286 0.2611623 #> 28 13 14 20 25 7 5 6 #> 0.7773604 0.3075051 0.2833167 0.1749943 0.2684784 0.5167965 0.9747233 0.2437827 #> 3 4 2 9 12 10 11 21 #> 0.2252071 0.7586954 2.4447632 0.2389659 0.2093104 0.2597721 0.4922646 1.0884414 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":null,"dir":"Reference","previous_headings":"","what":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data log transformed abundances aquatic invertebrate twelve ditches studied eleven times insecticide treatment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos)"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data frame 132 observations log-transformed (log(10*x + 1)) abundances 178 species. twelve sites (ditches, mesocosms), studied repeatedly eleven occasions. treatment levels, treatment times, ditch ID's data frame, data regular, example shows obtain external variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data set obtained experiment outdoor experimental ditches. Twelve mesocosms allocated random treatments; four served controls, remaining eight treated insecticide chlorpyrifos, nominal dose levels 0.1, 0.9, 6, 44 \\(\\mu\\)g/ L two mesocosms . example data set invertebrates. Sampling done 11 times, week -4 pre-treatment week 24 post-treatment, giving total 132 samples (12 mesocosms times 11 sampling dates), see van den Brink & ter Braak (1999) details. data set contains species data, example shows obtain treatment, time ditch ID variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"CANOCO 4 example data, permission Cajo J. F. ter Braak.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos) ditch <- gl(12, 1, length=132) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank -- Abundance or Dominance / Diversity Models — radfit","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions construct rank -- abundance dominance / diversity Whittaker plots fit brokenstick, preemption, log-Normal, Zipf Zipf-Mandelbrot models species abundance.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"# S3 method for default radfit(x, ...) rad.null(x, family=poisson, ...) rad.preempt(x, family = poisson, ...) rad.lognormal(x, family = poisson, ...) rad.zipf(x, family = poisson, ...) rad.zipfbrot(x, family = poisson, ...) # S3 method for radline predict(object, newdata, total, ...) # S3 method for radfit plot(x, BIC = FALSE, legend = TRUE, ...) # S3 method for radfit.frame plot(x, order.by, BIC = FALSE, model, legend = TRUE, as.table = TRUE, ...) # S3 method for radline plot(x, xlab = \"Rank\", ylab = \"Abundance\", type = \"b\", ...) radlattice(x, BIC = FALSE, ...) # S3 method for radfit lines(x, ...) # S3 method for radfit points(x, ...) as.rad(x) # S3 method for rad plot(x, xlab = \"Rank\", ylab = \"Abundance\", log = \"y\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"x Data frame, matrix vector giving species abundances, object plotted. family Error distribution (passed glm). alternatives accepting link = \"log\" family can used, although make sense. object fitted result object. newdata Ranks used ordinations. models can interpolate non-integer “ranks” (although may approximate), extrapolation may fail total new total used predicting abundance. Observed total count used omitted. order.vector used ordering sites plots. BIC Use Bayesian Information Criterion, BIC, instead Akaike's AIC. penalty BIC \\(k = \\log(S)\\) \\(S\\) number species, whereas AIC uses \\(k = 2\\). model Show specified model. missing, AIC used select model. model names (can abbreviated) Null, Preemption, Lognormal, Zipf, Mandelbrot. legend Add legend line colours. .table Arrange panels starting upper left corner (passed xyplot). xlab,ylab Labels x y axes. type Type plot, \"b\" plotting observed points fitted lines, \"p\" points, \"l\" fitted lines, \"n\" setting frame. log Use logarithmic scale given axis. default log = \"y\" gives traditional plot community ecology preemption model straight line, log = \"xy\" Zipf model straight line. log = \"\" axes original arithmetic scale. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Rank--Abundance Dominance (RAD) Dominance/Diversity plots (Whittaker 1965) display logarithmic species abundances species rank order. plots supposed effective analysing types abundance distributions communities. functions fit popular models mainly following Wilson (1991). Functions rad.null, rad.preempt, rad.lognormal, rad.zipf zipfbrot fit individual models (described ) single vector (row data frame), function radfit fits models. argument function radfit can either vector single community data frame row represents distinct community. Function rad.null fits brokenstick model expected abundance species rank \\(r\\) \\(a_r = (J/S) \\sum_{x=r}^S (1/x)\\) (Pielou 1975), \\(J\\) total number individuals (site total) \\(S\\) total number species community. gives Null model individuals randomly distributed among observed species, fitted parameters. Function rad.preempt fits niche preemption model, .k.. geometric series Motomura model, expected abundance \\(\\) species rank \\(r\\) \\(a_r = J \\alpha (1 - \\alpha)^{r-1}\\). estimated parameter preemption coefficient \\(\\alpha\\) gives decay rate abundance per rank. niche preemption model straight line RAD plot. Function rad.lognormal fits log-Normal model assumes logarithmic abundances distributed Normally, \\(a_r = \\exp( \\log \\mu + \\log \\sigma N)\\), \\(N\\) Normal deviate. Function rad.zipf fits Zipf model \\(a_r = J p_1 r^\\gamma\\) \\(p_1\\) fitted proportion abundant species, \\(\\gamma\\) decay coefficient. Zipf--Mandelbrot model (rad.zipfbrot) adds one parameter: \\(a_r = J c (r + \\beta)^\\gamma\\) \\(p_1\\) Zipf model changes meaningless scaling constant \\(c\\). Log-Normal Zipf models generalized linear models (glm) logarithmic link function. Zipf--Mandelbrot adds one nonlinear parameter Zipf model, fitted using nlm nonlinear parameter estimating parameters log-Likelihood glm. Preemption model fitted purely nonlinear model. estimated parameters Null model. default family poisson appropriate genuine counts (integers), families accept link = \"log\" can used. Families Gamma gaussian may appropriate abundance data, cover. best model selected AIC. Therefore ‘quasi’ families quasipoisson used: AIC log-Likelihood needed non-linear models. functions plot functions. radfit applied data frame, plot uses Lattice graphics, plot functions use ordinary graphics. ordinary graphics functions return invisibly ordiplot object observed points, function identify.ordiplot can used label selected species. Alternatively, radlattice uses Lattice graphics display radfit model single site separate panel together AIC BIC values. Function .rad base function construct ordered RAD data. plot used RAD plot functions pass extra arguments (xlab log) function. function returns ordered vector taxa occurring site, corresponding attribute \"index\" included taxa.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions rad.null, rad.preempt, rad.lognormal, zipf zipfbrot fit single RAD model single site. result object class \"radline\" inherits glm, can handled ( ) glm methods. Function radfit fits models either single site rows data frame matrix. fitted single site, function returns object class \"radfit\" items y (observed values), family, models list fitted \"radline\" models. applied data frame matrix, radfit function returns object class \"radfit.frame\" list \"radfit\" objects, item names corresponding row name. result objects (\"radline\", \"radfit\", \"radfit.frame\") can accessed method functions. following methods available: AIC, coef, deviance, logLik. addition fit results can accessed fitted, predict residuals (inheriting residuals.glm). graphical functions discussed Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Pielou, E.C. (1975) Ecological Diversity. Wiley & Sons. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Whittaker, R. H. (1965) Dominance diversity plant communities. Science 147, 250--260. Wilson, J. B. (1991) Methods fitting dominance/diversity curves. Journal Vegetation Science 2, 35--46.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"RAD models usually fitted proportions instead original abundances. However, nothing models seems require division abundances site totals, original observations used functions. wish use proportions, must standardize data site totals, e.g. decostand use appropriate family Gamma. lognormal model fitted standard way, think quite correct -- least equivalent fitting Normal density log abundances like originally suggested (Preston 1948). models may fail. particular, estimation Zipf-Mandelbrot model difficult. fitting fails, NA returned. Wilson (1991) defined preemption model \\(a_r = J p_1 (1 - \\alpha)^{r-1}\\), \\(p_1\\) fitted proportion first species. However, parameter \\(p_1\\) completely defined \\(\\alpha\\) since fitted proportions must add one, therefore handle preemption one-parameter model. Veiled log-Normal model included earlier releases function, removed flawed: implicit veil line also appears ordinary log-Normal. latest release version rad.veil 1.6-10.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"data(BCI) mod <- rad.lognormal(BCI[5,]) mod #> #> RAD model: Log-Normal #> Family: poisson #> No. of species: 101 #> Total abundance: 505 #> #> log.mu log.sigma Deviance AIC BIC #> 0.951926 1.165929 17.077549 317.656487 322.886728 plot(mod) mod <- radfit(BCI[1,]) ## Standard plot overlaid for all models ## Preemption model is a line plot(mod) ## log for both axes: Zipf model is a line plot(mod, log = \"xy\") ## Lattice graphics separately for each model radlattice(mod) # Take a subset of BCI to save time and nerves mod <- radfit(BCI[3:5,]) mod #> #> Deviance for RAD models: #> #> 3 4 5 #> Null 86.1127 49.8111 80.855 #> Preemption 58.9295 39.7817 76.311 #> Lognormal 29.2719 16.6588 17.078 #> Zipf 50.1262 47.9108 30.936 #> Mandelbrot 5.7342 5.5665 10.573 plot(mod, pch=\".\")"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":null,"dir":"Reference","previous_headings":"","what":"Compares Dissimilarity Indices for Gradient Detection — rankindex","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Rank correlations dissimilarity indices gradient separation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"rankindex(grad, veg, indices = c(\"euc\", \"man\", \"gow\", \"bra\", \"kul\"), stepacross = FALSE, method = \"spearman\", metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"grad gradient variable matrix. veg community data matrix. indices Dissimilarity indices compared, partial matches alternatives vegdist. Alternatively, can (named) list functions returning objects class 'dist'. stepacross Use stepacross find shorter path dissimilarity. dissimilarities site pairs shared species set NA using .shared indices fixed upper limit can also analysed. method Correlation method used. metric Metric evaluate gradient separation. See Details. ... parameters stepacross.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"good dissimilarity index multidimensional scaling high rank-order similarity gradient separation. function compares indices vegdist gradient separation using rank correlation coefficients cor. gradient separation point assessed using given metric. default use Euclidean distance continuous variables scaled unit variance, use Gower metric mixed data using function daisy grad factors. alternatives Mahalanabis distances based grad matrix scaled columns orthogonal (uncorrelated) unit variance, Manhattan distances grad variables scaled unit range. indices argument can accept dissimilarity indices besides ones calculated vegdist function. , argument value (possibly named) list functions. function must return valid 'dist' object dissimilarities, similarities accepted converted dissimilarities beforehand.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Returns named vector rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Faith, F.P., Minchin, P.R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Jari Oksanen, additions Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"several problems using rank correlation coefficients. Typically many ties \\(n(n-1)/2\\) gradient separation values derived just \\(n\\) observations. Due floating point arithmetics, many tied values differ machine epsilon arbitrarily ranked differently rank used cor.test. Two indices identical certain transformation standardization may differ slightly (magnitude \\(10^{-15}\\)) may lead third fourth decimal instability rank correlations. Small differences rank correlations taken seriously. Probably method replaced sounder method, yet know ... may experiment mantel, anosim even protest. Earlier version function used method = \"kendall\", far slow large data sets. functions returning dissimilarity objects self contained, ... argument passes additional parameters stepacross functions supplied via indices argument.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"data(varespec) data(varechem) ## The variables are automatically scaled rankindex(varechem, varespec) #> euc man gow bra kul #> 0.2396330 0.2735087 0.2288358 0.2837910 0.2839834 rankindex(varechem, wisconsin(varespec)) #> euc man gow bra kul #> 0.4200990 0.4215642 0.3708606 0.4215642 0.4215642 ## Using non vegdist indices as functions funs <- list(Manhattan=function(x) dist(x, \"manhattan\"), Gower=function(x) cluster:::daisy(x, \"gower\"), Ochiai=function(x) designdist(x, \"1-J/sqrt(A*B)\")) rankindex(scale(varechem), varespec, funs) #> Manhattan Gower Ochiai #> 0.2735087 0.2288358 0.1696862"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":null,"dir":"Reference","previous_headings":"","what":"Rarefaction Species Richness — rarefy","title":"Rarefaction Species Richness — rarefy","text":"Rarefied species richness community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rarefaction Species Richness — rarefy","text":"","code":"rarefy(x, sample, se = FALSE, MARGIN = 1) rrarefy(x, sample) drarefy(x, sample) rarecurve(x, step = 1, sample, xlab = \"Sample Size\", ylab = \"Species\", label = TRUE, col, lty, tidy = FALSE, ...) rareslope(x, sample)"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rarefaction Species Richness — rarefy","text":"x Community data, matrix-like object vector. MARGIN Margin index computed. sample Subsample size rarefying community, either single value vector. se Estimate standard errors. step Step size sample sizes rarefaction curves. xlab, ylab Axis labels plots rarefaction curves. label Label rarefaction curves rownames x (logical). col, lty plotting colour line type, see par. Can vector length nrow(x), one per sample, extended length internally. tidy Instead drawing plot, return “tidy” data frame can used ggplot2 graphics. data frame variables Site (factor), Sample Species. ... Parameters passed nlm, plot, lines ordilabel rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rarefaction Species Richness — rarefy","text":"Function rarefy gives expected species richness random subsamples size sample community. size sample smaller total community size, function work larger sample well (warning) return non-rarefied species richness (standard error = 0). sample vector, rarefaction observations performed sample size separately. Rarefaction can performed genuine counts individuals. function rarefy based Hurlbert's (1971) formulation, standard errors Heck et al. (1975). Function rrarefy generates one randomly rarefied community data frame vector given sample size. sample can vector giving sample sizes row. sample size equal larger observed number individuals, non-rarefied community returned. random rarefaction made without replacement variance rarefied communities rather related rarefaction proportion size sample. Random rarefaction sometimes used remove effects different sample sizes. usually bad idea: random rarefaction discards valid data, introduces random error reduces quality data (McMurdie & Holmes 2014). better use normalizing transformations (decostand vegan) possible variance stabilization (decostand dispweight vegan) methods sensitive sample sizes. Function drarefy returns probabilities species occur rarefied community size sample. sample can vector giving sample sizes row. sample equal larger observed number individuals, observed species sampling probability 1. Function rarecurve draws rarefaction curve row input data. rarefaction curves evaluated using interval step sample sizes, always including 1 total sample size. sample specified, vertical line drawn sample horizontal lines rarefied species richnesses. Function rareslope calculates slope rarecurve (derivative rarefy) given sample size; sample need integer. Rarefaction functions used observed counts. think necessary use multiplier data, rarefy first multiply. Removing rare species rarefaction can also give biased results. Observed count data normally include singletons (species count 1), missing, functions issue warnings. may false positives, recommended check observed counts multiplied rare taxa removed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rarefaction Species Richness — rarefy","text":"vector rarefied species richness values. single sample se = TRUE, function rarefy returns 2-row matrix rarefied richness (S) standard error (se). sample vector rarefy, function returns matrix column sample size, se = TRUE, rarefied richness standard error consecutive lines. Function rarecurve returns invisible list rarefy results corresponding drawn curve. Alternatively, tidy = TRUE returns data frame can used ggplot2 graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rarefaction Species Richness — rarefy","text":"Heck, K.L., van Belle, G. & Simberloff, D. (1975). Explicit calculation rarefaction diversity measurement determination sufficient sample size. Ecology 56, 1459--1461. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. McMurdie, P.J. & Holmes, S. (2014). Waste , want : rarefying microbiome data inadmissible. PLoS Comput Biol 10(4): e1003531. doi:10.1371/journal.pcbi.1003531","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rarefaction Species Richness — rarefy","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rarefaction Species Richness — rarefy","text":"","code":"data(BCI) S <- specnumber(BCI) # observed number of species (raremax <- min(rowSums(BCI))) #> [1] 340 Srare <- rarefy(BCI, raremax) plot(S, Srare, xlab = \"Observed No. of Species\", ylab = \"Rarefied No. of Species\") abline(0, 1) rarecurve(BCI, step = 20, sample = raremax, col = \"blue\", cex = 0.6)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":null,"dir":"Reference","previous_headings":"","what":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Function finds Raup-Crick dissimilarity probability number co-occurring species species occurrence probabilities proportional species frequencies.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"raupcrick(comm, null = \"r1\", nsimul = 999, chase = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"comm Community data treated presence/absence data. null Null model used method oecosimu. nsimul Number null communities assessing dissimilarity index. chase Use Chase et al. (2011) method tie handling ( recommended except comparing results Chase script). ... parameters passed oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Raup-Crick index probability compared sampling units non-identical species composition. probability can regarded dissimilarity, although metric: identical sampling units can dissimilarity slightly \\(0\\), dissimilarity can nearly zero range shared species, sampling units shared species can dissimilarity slightly \\(1\\). Moreover, communities sharing rare species appear similar (lower probability finding rare species together), communities sharing number common species. function always treat data binary (presence/ absence). probability assessed using simulation oecosimu test statistic observed number shared species sampling units evaluated community null model (see Examples). default null model \"r1\" probability selecting species proportional species frequencies. vegdist function implements variant Raup-Crick index equal sampling probabilities species using exact analytic equations without simulation. corresponds null model \"r0\" also can used current function. null model methods oecosimu can used current function, new unpublished methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function returns object inheriting dist can interpreted dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function developed Brian Inouye contacted us informed us method Chase et al. (2011), function takes idea code published paper. current function written Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"test statistic number shared species, typically tied large number simulation results. tied values handled differently current function function published Chase et al. (2011). vegan, index number simulated values smaller equal observed value, smaller observed value used Chase et al. (2011) option split = FALSE script; can achieved chase = TRUE vegan. Chase et al. (2011) script split = TRUE uses half tied simulation values calculate distance measure, choice directly reproduced vegan ( average vegan raupcrick results chase = TRUE chase = FALSE).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"## data set with variable species richness data(sipoo) ## default raupcrick dr1 <- raupcrick(sipoo) ## use null model \"r0\" of oecosimu dr0 <- raupcrick(sipoo, null = \"r0\") ## vegdist(..., method = \"raup\") corresponds to 'null = \"r0\"' d <- vegdist(sipoo, \"raup\") op <- par(mfrow=c(2,1), mar=c(4,4,1,1)+.1) plot(dr1 ~ d, xlab = \"Raup-Crick with Null R1\", ylab=\"vegdist\") plot(dr0 ~ d, xlab = \"Raup-Crick with Null R0\", ylab=\"vegdist\") par(op) ## The calculation is essentially as in the following oecosimu() call, ## except that designdist() is replaced with faster code if (FALSE) oecosimu(sipoo, function(x) designdist(x, \"J\", \"binary\"), method = \"r1\")"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":null,"dir":"Reference","previous_headings":"","what":"Reads a CEP (Canoco) data file — read.cep","title":"Reads a CEP (Canoco) data file — read.cep","text":"read.cep reads file formatted relaxed strict CEP format used Canoco software, among others.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"read.cep(file, positive=TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reads a CEP (Canoco) data file — read.cep","text":"file File name (character variable). positive positive entries, like community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reads a CEP (Canoco) data file — read.cep","text":"Cornell Ecology Programs (CEP) introduced several data formats designed punched cards. One ‘condensed strict’ format adopted popular software DECORANA TWINSPAN. relaxed variant format later adopted Canoco software (ter Braak 1984). Function read.cep reads legacy files written format. condensed CEP CANOCO formats : Two three title cards, importantly specifying format number items per record. Data condensed format: First number line site identifier (integer), followed pairs (‘couplets’) numbers identifying species abundance (integer floating point number). Species site names, given Fortran format (10A8): Ten names per line, eight columns . option positive = TRUE function removes rows columns zero negative marginal sums. community data positive entries, removes empty sites species. data entries can negative, ruins data, data sets read option positive = FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reads a CEP (Canoco) data file — read.cep","text":"Returns data frame, columns species rows sites. Column row names taken CEP file, changed unique R names make.names stripping blanks.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Reads a CEP (Canoco) data file — read.cep","text":"ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reads a CEP (Canoco) data file — read.cep","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reads a CEP (Canoco) data file — read.cep","text":"Function read.cep used Fortran read data vegan 2.4-5 earlier, Fortran /O longer allowed CRAN packages, function re-written R. original Fortran code robust, several legacy data sets may fail current version, read previous Fortran version. CRAN package cepreader makes available original Fortran-based code run separate subprocess. cepreader package can also read ‘free’ ‘open’ Canoco formats handled function. function based read.fortran. REAL format defines decimal part species abundances (F5.1), read.fortran divides input corresponding power 10 even input data explicit decimal separator. F5.1, 100 become 10, 0.1 become 0.01. Function read.cep tries undo division, check scaling results reading data, necessary, multiply results original scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"## Provided that you have the file \"dune.spe\" if (FALSE) { theclassic <- read.cep(\"dune.spe\")}"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":null,"dir":"Reference","previous_headings":"","what":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi find Rényi diversities scale corresponding Hill number (Hill 1973). Function renyiaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"renyi(x, scales = c(0, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64, Inf), hill = FALSE) # S3 method for renyi plot(x, ...) renyiaccum(x, scales = c(0, 0.5, 1, 2, 4, Inf), permutations = 100, raw = FALSE, collector = FALSE, subset, ...) # S3 method for renyiaccum plot(x, what = c(\"Collector\", \"mean\", \"Qnt 0.025\", \"Qnt 0.975\"), type = \"l\", ...) # S3 method for renyiaccum persp(x, theta = 220, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"x Community data matrix plotting object. scales Scales Rényi diversity. hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. collector Accumulate diversities order sites data set, collector curve can plotted summary permutations. argument ignored raw = TRUE. subset logical expression indicating sites (rows) keep: missing values taken FALSE. Items plotted. type Type plot, type = \"l\" means lines. theta Angle defining viewing direction (azimuthal) persp. col Colours used surface. Single colour passed , vector colours selected midpoint rectangle persp. zlim Limits vertical axis. ... arguments passed renyi graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Common diversity indices special cases Rényi diversity $$H_a = \\frac{1}{1-} \\log \\sum p_i^$$ \\(\\) scale parameter, Hill (1975) suggested use -called ‘Hill numbers’ defined \\(N_a = \\exp(H_a)\\). Hill numbers number species \\(= 0\\), \\(\\exp(H')\\) exponent Shannon diversity \\(= 1\\), inverse Simpson \\(= 2\\) \\(1/ \\max(p_i)\\) \\(= \\infty\\). According theory diversity ordering, one community can regarded diverse another Rényi diversities higher (Tóthmérész 1995). plot method renyi uses lattice graphics, displays diversity values scale separate panel site together minimum, maximum median values complete data. Function renyiaccum similar specaccum finds Rényi Hill diversities given scales random permutations accumulated sites. plot function uses lattice function xyplot display accumulation curves value scales separate panel. addition, persp method plot diversity surface scale number sites. Similar dynamic graphics can made rgl.renyiaccum vegan3d package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi returns data frame selected indices. Function renyiaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Hill, M.O. (1973). Diversity evenness: unifying notation consequences. Ecology 54, 427--473. Kindt, R., Van Damme, P., Simons, .J. (2006). Tree diversity western Kenya: using profiles characterise richness evenness. Biodiversity Conservation 15, 1253--1270. Tóthmérész, B. (1995). Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Roeland Kindt Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) mod <- renyi(BCI[i,]) plot(mod) mod <- renyiaccum(BCI[i,]) plot(mod, as.table=TRUE, col = c(1, 2, 2)) persp(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":null,"dir":"Reference","previous_headings":"","what":"Reorder a Hierarchical Clustering Tree — reorder.hclust","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Function takes hierarchical clustering tree hclust vector values reorders clustering tree order supplied vector, maintaining constraints tree. method generic function reorder alternative reordering \"dendrogram\" object reorder.dendrogram","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"# S3 method for hclust reorder(x, wts, agglo.FUN = c(\"mean\", \"min\", \"max\", \"sum\", \"uwmean\"), ...) # S3 method for hclust rev(x) # S3 method for hclust scores(x, display = \"internal\", ...) cutreeord(tree, k = NULL, h = NULL)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"x, tree hierarchical clustering hclust. wts numeric vector reordering. agglo.FUN function weights agglomeration, see . display return \"internal\" nodes \"terminal\" nodes (also called \"leaves\"). k, h scalars vectors giving numbers desired groups heights tree cut (passed function cutree). ... additional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Dendrograms can ordered many ways. reorder function reorders hclust tree provides alternative reorder.dendrogram can reorder dendrogram. current function also work differently agglo.FUN \"mean\": reorder.dendrogram always take direct mean member groups ignoring sizes, function used weighted.mean weighted group sizes, group mean always mean member leaves (terminal nodes). want ignore group sizes, can use unweighted mean \"uwmean\". function accepts limited list agglo.FUN functions assessing value wts groups. ordering always ascending, order leaves can reversed rev. Function scores finds coordinates nodes two-column matrix. terminal nodes (leaves) value item merged tree, labels can still hang level (see plot.hclust). Function cutreeord cuts tree groups numbered left right tree. based standard function cutree numbers groups order appear input data instead order tree.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Reordered hclust result object added item value gives value statistic merge level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"functions really base R.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"## reorder by water content of soil data(mite, mite.env) hc <- hclust(vegdist(wisconsin(sqrt(mite)))) ohc <- with(mite.env, reorder(hc, WatrCont)) plot(hc) plot(ohc) ## label leaves by the observed value, and each branching point ## (internal node) by the cluster mean with(mite.env, plot(ohc, labels=round(WatrCont), cex=0.7)) ordilabel(scores(ohc), label=round(ohc$value), cex=0.7) ## Slightly different from reordered 'dendrogram' which ignores group ## sizes in assessing means. den <- as.dendrogram(hc) den <- with(mite.env, reorder(den, WatrCont, agglo.FUN = mean)) plot(den)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Species or Site Scores from an Ordination — scores","title":"Get Species or Site Scores from an Ordination — scores","text":"Function access either species site scores specified axes ordination methods. scores function generic vegan, vegan ordination functions scores functions documented separately method (see e.g. scores.cca, scores.metaMDS, scores.decorana). help file documents default scores method used non-vegan ordination objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"# S3 method for default scores(x, choices, display=c(\"sites\", \"species\", \"both\"), tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Species or Site Scores from an Ordination — scores","text":"x ordination result. choices Ordination axes. missing, default method returns axes. display Partial match access scores \"sites\" \"species\" \"\". tidy Return \"\" scores data frame compatible ggplot2, variable score labelling scores \"sites\" \"species\". ... parameters (unused).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Species or Site Scores from an Ordination — scores","text":"Function scores generic method vegan. Several vegan functions scores methods defaults new arguments. help page describes default method. methods, see, e.g., scores.cca, scores.rda, scores.decorana. vegan ordination functions scores method used extract scores instead directly accessing . Scaling transformation scores also happen scores function. scores function available, results can plotted using ordiplot, ordixyplot etc., ordination results can compared procrustes analysis. scores.default function used extract scores non-vegan ordination results. Many standard ordination methods libraries specific class, specific method can written . However, scores.default guesses commonly used functions keep site scores possible species scores. x matrix, scores.default returns chosen columns matrix, ignoring whether species sites requested (regard bug feature, please). Currently function seems work least isoMDS, prcomp, princomp ade4 objects. may work cases fail mysteriously.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Species or Site Scores from an Ordination — scores","text":"function returns matrix scores one type requested, named list matrices display = \"\", ggplot2 compatible data frame tidy = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get Species or Site Scores from an Ordination — scores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"data(varespec) vare.pca <- prcomp(varespec) scores(vare.pca, choices=c(1,2)) #> PC1 PC2 #> 18 -10.7847878 18.7094315 #> 15 -27.8036826 -11.7414745 #> 24 -25.6919559 -14.5399684 #> 27 -31.7820166 -31.2216800 #> 23 -19.6315869 -2.5541193 #> 19 -0.2413294 -11.4974077 #> 22 -26.6771373 -12.3140897 #> 16 -21.9230366 0.4449159 #> 28 -39.6083051 -41.8877392 #> 13 -4.0664328 20.4191153 #> 14 -18.4416245 5.4406988 #> 20 -17.3999191 2.3653380 #> 25 -25.1673547 -13.2508067 #> 7 -11.4065430 41.7356300 #> 5 -8.4243752 45.3805255 #> 6 -2.0759474 36.9311222 #> 3 39.8617580 8.0590041 #> 4 13.1065901 12.8377217 #> 2 57.6827011 -4.8983565 #> 9 63.3138332 -22.4481549 #> 12 44.1073111 -10.1653935 #> 10 64.9418975 -16.7633564 #> 11 11.5313633 3.9720890 #> 21 -3.4194194 -3.0130455"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Screeplot methods plotting variances ordination axes/components overlaying broken stick distributions. Also, provides alternative screeplot methods princomp prcomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"# S3 method for cca screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, if (is.null(x$CCA) || x$CCA$rank == 0) x$CA$rank else x$CCA$rank), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for decorana screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = 4, ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for prcomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for princomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) bstick(n, ...) # S3 method for default bstick(n, tot.var = 1, ...) # S3 method for cca bstick(n, ...) # S3 method for prcomp bstick(n, ...) # S3 method for princomp bstick(n, ...) # S3 method for decorana bstick(n, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"x object component variances can determined. bstick logical; broken stick distribution drawn? npcs number components plotted. type type plot. ptype type == \"lines\" bstick = TRUE, character indicating type plotting used lines; actually types plot.default. bst.col, bst.lty colour line type used draw broken stick distribution. xlab, ylab, main graphics parameters. legend logical; draw legend? n object variances can extracted number variances (components) case bstick.default. tot.var total variance split. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"functions provide screeplots ordination methods vegan enhanced versions broken stick prcomp princomp. Function bstick gives brokenstick values ordered random proportions, defined \\(p_i = (tot/n) \\sum_{x=}^n (1/x)\\) (Legendre & Legendre 2012), \\(tot\\) total \\(n\\) number brokenstick components (cf. radfit). Broken stick recommended stopping rule principal component analysis (Jackson 1993): principal components retained long observed eigenvalues higher corresponding random broken stick components. bstick function generic. default needs number components total, specific methods extract information ordination results. also bstick method cca. However, broken stick model strictly valid correspondence analysis (CA), eigenvalues CA defined \\(\\leq 1\\), whereas brokenstick components restrictions. brokenstick components detrended correspondence analysis (DCA) assume input data full rank, additive eigenvalues used screeplot (see decorana).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Function screeplot draws plot currently active device, returns invisibly xy.coords points bars eigenvalues. Function bstick returns numeric vector broken stick components.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Jackson, D. . (1993). Stopping rules principal components analysis: comparison heuristical statistical approaches. Ecology 74, 2204--2214. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"data(varespec) vare.pca <- rda(varespec, scale = TRUE) bstick(vare.pca) #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 7.1438620 5.2308185 4.2742968 3.6366156 3.1583548 2.7757461 2.4569055 2.1836136 #> PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 #> 1.9444831 1.7319228 1.5406184 1.3667054 1.2072851 1.0601279 0.9234819 0.7959457 #> PC17 PC18 PC19 PC20 PC21 PC22 PC23 #> 0.6763805 0.5638485 0.4575683 0.3568818 0.2612296 0.1701323 0.0831758 screeplot(vare.pca, bstick = TRUE, type = \"lines\")"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":null,"dir":"Reference","previous_headings":"","what":"Similarity Percentages — simper","title":"Similarity Percentages — simper","text":"Discriminating species two groups using Bray-Curtis dissimilarities","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Similarity Percentages — simper","text":"","code":"simper(comm, group, permutations = 999, parallel = 1, ...) # S3 method for simper summary(object, ordered = TRUE, digits = max(3,getOption(\"digits\") - 3), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Similarity Percentages — simper","text":"comm Community data. group Factor describing group structure. missing one level, contributions estimated non-grouped data dissimilarities show overall heterogeneity species abundances. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. object object returned simper. ordered Logical; species ordered average contribution? digits Number digits output. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. (yet implemented). ... Parameters passed functions. simper extra parameters passed shuffleSet permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Similarity Percentages — simper","text":"Similarity percentage, simper (Clarke 1993) based decomposition Bray-Curtis dissimilarity index (see vegdist, designdist). contribution individual species \\(\\) overall Bray-Curtis dissimilarity \\(d_{jk}\\) given $$d_{ijk} = \\frac{|x_{ij}-x_{ik}|}{\\sum_{=1}^S (x_{ij}+x_{ik})}$$ \\(x\\) abundance species \\(\\) sampling units \\(j\\) \\(k\\). overall index sum individual contributions \\(S\\) species \\(d_{jk}=\\sum_{=1}^S d_{ijk}\\). simper functions performs pairwise comparisons groups sampling units finds contribution species average -group Bray-Curtis dissimilarity. Although method called “Similarity Percentages”, really studied dissimilarities instead similarities (Clarke 1993). function displays important species pair groups. species contribute least 70 % differences groups. function returns much extensive results (including species) can accessed directly result object (see section Value). Function summary transforms result list data frames. argument ordered = TRUE data frames also include cumulative contributions ordered species contribution. results simper can difficult interpret often misunderstood even publications. method gives contribution species overall dissimilarities, caused variation species abundances, partly differences among groups. Even make groups copies , method single species high contribution, contributions non-existing -group differences random noise variation species abundances. abundant species usually highest variances, high contributions even differ among groups. Permutation tests study differences among groups, can used find species differences among groups important component contribution dissimilarities. Analysis without group argument find species contributions average overall dissimilarity among sampling units. non-grouped contributions can compared grouped contributions see much added value grouping species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Similarity Percentages — simper","text":"list class \"simper\" following items: species species names. average Species contribution average -group dissimilarity. overall average -group dissimilarity. sum item average. sd Standard deviation contribution. ratio Average sd ratio. ava, avb Average abundances per group. ord index vector order vectors contribution order cusum back original data order. cusum Ordered cumulative contribution. based item average, sum total 1. p Permutation \\(p\\)-value. Probability getting larger equal average contribution random permutation group factor. area available permutations used (default: calculated).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Similarity Percentages — simper","text":"Eduard Szöcs Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Similarity Percentages — simper","text":"Clarke, K.R. 1993. Non-parametric multivariate analyses changes community structure. Australian Journal Ecology, 18, 117–143.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Similarity Percentages — simper","text":"","code":"data(dune) data(dune.env) (sim <- with(dune.env, simper(dune, Management, permutations = 99))) #> cumulative contributions of most influential species: #> #> $SF_BF #> Agrostol Alopgeni Lolipere Trifrepe Poatriv Scorautu Bromhord #> 0.09824271 0.18254830 0.25956958 0.33367870 0.40734444 0.47729205 0.53120026 #> Achimill Planlanc Elymrepe Bracruta #> 0.57946526 0.62522255 0.67016196 0.71098133 #> #> $SF_HF #> Agrostol Alopgeni Lolipere Planlanc Rumeacet Elymrepe Poatriv #> 0.08350879 0.16534834 0.23934930 0.30843624 0.37716139 0.43334492 0.48351753 #> Bracruta Eleopalu Poaprat Anthodor Sagiproc Trifprat #> 0.52804045 0.57205850 0.61423981 0.65549838 0.69628951 0.73696831 #> #> $SF_NM #> Poatriv Alopgeni Agrostol Lolipere Eleopalu Poaprat Bracruta Elymrepe #> 0.1013601 0.1935731 0.2667383 0.3377578 0.3999419 0.4526707 0.5044725 0.5505643 #> Scorautu Trifrepe Sagiproc Salirepe #> 0.5926117 0.6320111 0.6712478 0.7091528 #> #> $BF_HF #> Rumeacet Poatriv Planlanc Bromhord Lolipere Elymrepe Trifrepe #> 0.08163219 0.15193797 0.21918333 0.27967181 0.33969561 0.39843338 0.45298204 #> Anthodor Achimill Bracruta Alopgeni Trifprat Juncarti #> 0.50276849 0.55222648 0.60021994 0.64584333 0.69126471 0.73366621 #> #> $BF_NM #> Lolipere Poatriv Poaprat Trifrepe Bromhord Bracruta Eleopalu Agrostol #> 0.1242718 0.1992126 0.2711756 0.3414609 0.3958520 0.4448077 0.4910724 0.5369083 #> Achimill Scorautu Anthodor Planlanc #> 0.5823926 0.6253645 0.6638182 0.7012577 #> #> $HF_NM #> Poatriv Lolipere Rumeacet Poaprat Planlanc Bracruta Eleopalu #> 0.09913221 0.17468460 0.23917190 0.29701331 0.35469313 0.40365488 0.44804851 #> Agrostol Trifrepe Elymrepe Anthodor Juncarti Trifprat Salirepe #> 0.49226546 0.53434466 0.57564661 0.61543243 0.65341300 0.68921695 0.72432408 #> ## IGNORE_RDIFF_BEGIN summary(sim) #> #> Contrast: SF_BF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.06137 0.03419 1.79490 4.66700 0.00000 0.098 0.05 * #> Alopgeni 0.05267 0.03648 1.44390 4.33300 0.66700 0.182 0.14 #> Lolipere 0.04812 0.03945 1.21980 3.00000 6.00000 0.260 0.39 #> Trifrepe 0.04630 0.02553 1.81380 1.33300 4.66700 0.334 0.09 . #> Poatriv 0.04602 0.03380 1.36150 4.66700 3.66700 0.407 0.46 #> Scorautu 0.04370 0.02492 1.75340 1.33300 4.33300 0.477 0.04 * #> Bromhord 0.03368 0.02586 1.30230 0.50000 2.66700 0.531 0.02 * #> Achimill 0.03015 0.02082 1.44820 0.16700 2.33300 0.580 0.04 * #> Planlanc 0.02859 0.02155 1.32650 0.00000 2.00000 0.625 0.49 #> Elymrepe 0.02807 0.02978 0.94280 2.00000 1.33300 0.670 0.52 #> Bracruta 0.02550 0.02390 1.06690 2.00000 2.00000 0.711 0.83 #> Poaprat 0.02513 0.02397 1.04850 2.50000 4.00000 0.751 0.82 #> Sagiproc 0.02433 0.02215 1.09830 1.83300 0.66700 0.790 0.39 #> Bellpere 0.01986 0.01709 1.16220 0.66700 1.66700 0.822 0.10 . #> Eleopalu 0.01861 0.04296 0.43330 1.33300 0.00000 0.852 0.82 #> Anthodor 0.01754 0.02580 0.67980 0.00000 1.33300 0.880 0.75 #> Juncbufo 0.01603 0.02371 0.67620 1.16700 0.00000 0.905 0.57 #> Vicilath 0.01467 0.01331 1.10260 0.00000 1.00000 0.929 0.04 * #> Hyporadi 0.01029 0.01520 0.67680 0.00000 0.66700 0.945 0.62 #> Ranuflam 0.00931 0.01360 0.68450 0.66700 0.00000 0.960 0.93 #> Juncarti 0.00698 0.01611 0.43330 0.50000 0.00000 0.972 0.95 #> Callcusp 0.00698 0.01611 0.43330 0.50000 0.00000 0.983 0.79 #> Rumeacet 0.00453 0.01044 0.43330 0.33300 0.00000 0.990 0.95 #> Cirsarve 0.00398 0.00918 0.43360 0.33300 0.00000 0.996 0.37 #> Chenalbu 0.00233 0.00537 0.43330 0.16700 0.00000 1.000 0.41 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_HF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.04738 0.03127 1.51510 4.66700 1.40000 0.084 0.35 #> Alopgeni 0.04643 0.03290 1.41150 4.33300 1.60000 0.165 0.20 #> Lolipere 0.04199 0.02701 1.55460 3.00000 4.00000 0.239 0.75 #> Planlanc 0.03920 0.03321 1.18040 0.00000 3.00000 0.308 0.02 * #> Rumeacet 0.03899 0.02737 1.42470 0.33300 3.20000 0.377 0.01 ** #> Elymrepe 0.03188 0.02955 1.07870 2.00000 2.00000 0.433 0.30 #> Poatriv 0.02847 0.02152 1.32270 4.66700 4.80000 0.484 1.00 #> Bracruta 0.02526 0.02104 1.20040 2.00000 2.80000 0.528 0.92 #> Eleopalu 0.02497 0.03888 0.64240 1.33300 0.80000 0.572 0.77 #> Poaprat 0.02393 0.01918 1.24780 2.50000 3.40000 0.614 0.98 #> Anthodor 0.02341 0.02143 1.09230 0.00000 1.80000 0.655 0.67 #> Sagiproc 0.02314 0.02048 1.13010 1.83300 0.80000 0.696 0.40 #> Trifprat 0.02308 0.02343 0.98500 0.00000 1.80000 0.737 0.01 ** #> Juncarti 0.02285 0.02568 0.88990 0.50000 1.60000 0.777 0.51 #> Trifrepe 0.02238 0.01949 1.14860 1.33300 2.80000 0.817 0.94 #> Juncbufo 0.02164 0.02224 0.97330 1.16700 1.20000 0.855 0.24 #> Scorautu 0.02051 0.01642 1.24890 1.33300 2.80000 0.891 0.79 #> Achimill 0.01518 0.01139 1.33260 0.16700 1.20000 0.918 0.75 #> Bromhord 0.01338 0.01450 0.92220 0.50000 0.80000 0.941 0.79 #> Ranuflam 0.01066 0.01339 0.79640 0.66700 0.40000 0.960 0.86 #> Bellpere 0.00999 0.01257 0.79480 0.66700 0.40000 0.978 0.84 #> Callcusp 0.00662 0.01508 0.43930 0.50000 0.00000 0.989 0.92 #> Cirsarve 0.00381 0.00867 0.43940 0.33300 0.00000 0.996 0.54 #> Chenalbu 0.00221 0.00503 0.43930 0.16700 0.00000 1.000 0.51 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Hyporadi 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Vicilath 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07828 0.04095 1.91180 4.66700 0.00000 0.101 0.01 ** #> Alopgeni 0.07122 0.04696 1.51670 4.33300 0.00000 0.194 0.01 ** #> Agrostol 0.05651 0.04418 1.27920 4.66700 2.16700 0.267 0.05 * #> Lolipere 0.05485 0.05991 0.91550 3.00000 0.33300 0.338 0.15 #> Eleopalu 0.04803 0.04717 1.01820 1.33300 2.16700 0.400 0.04 * #> Poaprat 0.04072 0.03179 1.28100 2.50000 0.66700 0.453 0.06 . #> Bracruta 0.04001 0.03440 1.16310 2.00000 2.83300 0.504 0.08 . #> Elymrepe 0.03560 0.03852 0.92430 2.00000 0.00000 0.551 0.12 #> Scorautu 0.03247 0.03481 0.93280 1.33300 3.16700 0.593 0.13 #> Trifrepe 0.03043 0.03163 0.96190 1.33300 1.83300 0.632 0.59 #> Sagiproc 0.03030 0.03048 0.99430 1.83300 0.50000 0.671 0.02 * #> Salirepe 0.02928 0.03201 0.91440 0.00000 1.83300 0.709 0.02 * #> Anthodor 0.02454 0.03669 0.66880 0.00000 1.33300 0.741 0.56 #> Callcusp 0.02276 0.02944 0.77310 0.50000 1.16700 0.770 0.08 . #> Ranuflam 0.02257 0.02282 0.98890 0.66700 1.33300 0.800 0.08 . #> Juncarti 0.02254 0.02860 0.78830 0.50000 1.16700 0.829 0.53 #> Hyporadi 0.02011 0.03129 0.64260 0.00000 1.16700 0.855 0.21 #> Juncbufo 0.01986 0.02903 0.68400 1.16700 0.00000 0.881 0.21 #> Planlanc 0.01542 0.02277 0.67720 0.00000 0.83300 0.900 0.98 #> Airaprae 0.01488 0.02188 0.68020 0.00000 0.83300 0.920 0.06 . #> Bellpere 0.01232 0.01592 0.77370 0.66700 0.33300 0.936 0.72 #> Comapalu 0.01188 0.01741 0.68260 0.00000 0.66700 0.951 0.05 * #> Achimill 0.00929 0.01493 0.62240 0.16700 0.33300 0.963 0.98 #> Bromhord 0.00717 0.01633 0.43910 0.50000 0.00000 0.972 0.98 #> Rumeacet 0.00559 0.01275 0.43840 0.33300 0.00000 0.980 0.98 #> Empenigr 0.00523 0.01200 0.43540 0.00000 0.33300 0.986 0.29 #> Cirsarve 0.00478 0.01089 0.43910 0.33300 0.00000 0.993 0.02 * #> Chenalbu 0.00289 0.00660 0.43820 0.16700 0.00000 0.996 0.02 * #> Vicilath 0.00279 0.00642 0.43450 0.00000 0.16700 1.000 0.81 #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_HF #> #> average sd ratio ava avb cumsum p #> Rumeacet 0.03867 0.02606 1.48380 0.00000 3.20000 0.082 0.03 * #> Poatriv 0.03330 0.02579 1.29110 3.66700 4.80000 0.152 0.96 #> Planlanc 0.03185 0.01830 1.74010 2.00000 3.00000 0.219 0.34 #> Bromhord 0.02865 0.01799 1.59260 2.66700 0.80000 0.280 0.07 . #> Lolipere 0.02843 0.02215 1.28340 6.00000 4.00000 0.340 1.00 #> Elymrepe 0.02782 0.02959 0.94040 1.33300 2.00000 0.398 0.56 #> Trifrepe 0.02584 0.01656 1.56030 4.66700 2.80000 0.453 0.78 #> Anthodor 0.02358 0.02042 1.15470 1.33300 1.80000 0.503 0.55 #> Achimill 0.02343 0.01474 1.58930 2.33300 1.20000 0.552 0.24 #> Bracruta 0.02273 0.01802 1.26170 2.00000 2.80000 0.600 0.89 #> Alopgeni 0.02161 0.02308 0.93630 0.66700 1.60000 0.646 0.91 #> Trifprat 0.02151 0.02207 0.97470 0.00000 1.80000 0.691 0.12 #> Juncarti 0.02008 0.02555 0.78600 0.00000 1.60000 0.734 0.59 #> Scorautu 0.01932 0.01357 1.42410 4.33300 2.80000 0.774 0.78 #> Bellpere 0.01829 0.01486 1.23050 1.66700 0.40000 0.813 0.22 #> Agrostol 0.01761 0.02284 0.77080 0.00000 1.40000 0.850 1.00 #> Juncbufo 0.01500 0.02066 0.72600 0.00000 1.20000 0.882 0.67 #> Vicilath 0.01285 0.01140 1.12740 1.00000 0.00000 0.909 0.04 * #> Sagiproc 0.01168 0.01297 0.90080 0.66700 0.80000 0.934 0.92 #> Eleopalu 0.01017 0.02111 0.48170 0.00000 0.80000 0.955 0.93 #> Hyporadi 0.00895 0.01312 0.68240 0.66700 0.00000 0.974 0.64 #> Poaprat 0.00720 0.01010 0.71330 4.00000 3.40000 0.989 1.00 #> Ranuflam 0.00508 0.01055 0.48170 0.00000 0.40000 1.000 0.97 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Callcusp 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_NM #> #> average sd ratio ava avb cumsum p #> Lolipere 0.09068 0.02644 3.42900 6.00000 0.33300 0.124 0.04 * #> Poatriv 0.05468 0.04465 1.22500 3.66700 0.00000 0.199 0.28 #> Poaprat 0.05251 0.01813 2.89700 4.00000 0.66700 0.271 0.02 * #> Trifrepe 0.05129 0.02756 1.86100 4.66700 1.83300 0.342 0.04 * #> Bromhord 0.03969 0.02920 1.35900 2.66700 0.00000 0.396 0.01 ** #> Bracruta 0.03572 0.02869 1.24500 2.00000 2.83300 0.445 0.40 #> Eleopalu 0.03376 0.03573 0.94500 0.00000 2.16700 0.491 0.43 #> Agrostol 0.03345 0.03473 0.96300 0.00000 2.16700 0.537 0.87 #> Achimill 0.03319 0.02338 1.42000 2.33300 0.33300 0.582 0.02 * #> Scorautu 0.03136 0.02026 1.54800 4.33300 3.16700 0.625 0.29 #> Anthodor 0.02806 0.03295 0.85200 1.33300 1.33300 0.664 0.37 #> Planlanc 0.02732 0.02193 1.24600 2.00000 0.83300 0.701 0.62 #> Salirepe 0.02677 0.02927 0.91400 0.00000 1.83300 0.738 0.11 #> Bellpere 0.02353 0.01909 1.23200 1.66700 0.33300 0.770 0.04 * #> Hyporadi 0.02172 0.02450 0.88600 0.66700 1.16700 0.800 0.26 #> Ranuflam 0.02031 0.02275 0.89300 0.00000 1.33300 0.828 0.27 #> Elymrepe 0.01999 0.02926 0.68300 1.33300 0.00000 0.855 0.79 #> Callcusp 0.01783 0.02681 0.66500 0.00000 1.16700 0.880 0.44 #> Juncarti 0.01769 0.02600 0.68100 0.00000 1.16700 0.904 0.73 #> Vicilath 0.01577 0.01447 1.09000 1.00000 0.16700 0.925 0.01 ** #> Sagiproc 0.01543 0.01857 0.83100 0.66700 0.50000 0.947 0.79 #> Airaprae 0.01341 0.01969 0.68100 0.00000 0.83300 0.965 0.30 #> Comapalu 0.01074 0.01571 0.68400 0.00000 0.66700 0.980 0.42 #> Alopgeni 0.01000 0.01463 0.68300 0.66700 0.00000 0.993 0.99 #> Empenigr 0.00479 0.01105 0.43300 0.00000 0.33300 1.000 0.41 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Juncbufo 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Rumeacet 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: HF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07155 0.01368 5.23000 4.80000 0.00000 0.099 0.01 ** #> Lolipere 0.05453 0.02962 1.84100 4.00000 0.33300 0.175 0.22 #> Rumeacet 0.04655 0.03081 1.51100 3.20000 0.00000 0.239 0.01 ** #> Poaprat 0.04175 0.01885 2.21500 3.40000 0.66700 0.297 0.06 . #> Planlanc 0.04163 0.02956 1.40800 3.00000 0.83300 0.355 0.03 * #> Bracruta 0.03534 0.02010 1.75800 2.80000 2.83300 0.404 0.40 #> Eleopalu 0.03204 0.03231 0.99200 0.80000 2.16700 0.448 0.51 #> Agrostol 0.03192 0.02889 1.10500 1.40000 2.16700 0.492 0.99 #> Trifrepe 0.03037 0.02287 1.32800 2.80000 1.83300 0.534 0.71 #> Elymrepe 0.02981 0.03868 0.77100 2.00000 0.00000 0.576 0.53 #> Anthodor 0.02872 0.02480 1.15800 1.80000 1.33300 0.615 0.24 #> Juncarti 0.02741 0.02854 0.96100 1.60000 1.16700 0.653 0.19 #> Trifprat 0.02584 0.02597 0.99500 1.80000 0.00000 0.689 0.01 ** #> Salirepe 0.02534 0.02729 0.92900 0.00000 1.83300 0.724 0.18 #> Alopgeni 0.02446 0.03240 0.75500 1.60000 0.00000 0.758 0.92 #> Scorautu 0.02070 0.01412 1.46600 2.80000 3.16700 0.787 0.85 #> Ranuflam 0.01928 0.01994 0.96700 0.40000 1.33300 0.814 0.36 #> Juncbufo 0.01818 0.02465 0.73800 1.20000 0.00000 0.839 0.43 #> Hyporadi 0.01714 0.02655 0.64600 0.00000 1.16700 0.863 0.42 #> Callcusp 0.01683 0.02490 0.67600 0.00000 1.16700 0.886 0.42 #> Achimill 0.01656 0.01490 1.11100 1.20000 0.33300 0.909 0.59 #> Sagiproc 0.01528 0.01653 0.92400 0.80000 0.50000 0.930 0.89 #> Airaprae 0.01261 0.01824 0.69100 0.00000 0.83300 0.947 0.30 #> Bromhord 0.01209 0.01517 0.79700 0.80000 0.00000 0.964 0.84 #> Comapalu 0.01011 0.01456 0.69400 0.00000 0.66700 0.978 0.25 #> Bellpere 0.00880 0.01373 0.64100 0.40000 0.33300 0.990 0.94 #> Empenigr 0.00454 0.01033 0.43900 0.00000 0.33300 0.997 0.60 #> Vicilath 0.00240 0.00546 0.43900 0.00000 0.16700 1.000 0.86 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 99 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Function simulates response data frame adds Gaussian error fitted responses Redundancy Analysis (rda), Constrained Correspondence Analysis (cca) distance-based RDA (capscale). function special case generic simulate, works similarly simulate.lm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"# S3 method for rda simulate(object, nsim = 1, seed = NULL, indx = NULL, rank = \"full\", correlated = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"object object representing fitted rda, cca capscale model. nsim number response matrices simulated. one dissimilarity matrix returned capscale, larger nsim error. seed object specifying random number generator initialized (‘seeded’). See simulate details. indx Index residuals added fitted values, produced shuffleSet sample. index can duplicate entries bootstrapping allowed. nsim \\(>1\\), output compliant shuffleSet one line simulation. nsim missing, number rows indx used define number simulations, nsim given, match number rows indx. null, parametric simulation used Gaussian error added fitted values. rank rank constrained component: passed predict.rda predict.cca. correlated species regarded correlated parametric simulation indx given? correlated = TRUE, multivariate Gaussian random error generated, FALSE, Gaussian random error generated separately species. argument effect capscale information species. ... additional optional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"implementation follows \"lm\" method simulate, adds Gaussian (Normal) error fitted values (fitted.rda) using function rnorm correlated = FALSE mvrnorm correlated = TRUE. standard deviations (rnorm) covariance matrices species (mvrnorm) estimated residuals fitting constraints. Alternatively, function can take permutation index used add permuted residuals (unconstrained component) fitted values. Raw data used rda. Internal Chi-square transformed data used cca within function, returned matrix similar original input data. simulation performed internal metric scaling data capscale, function returns Euclidean distances calculated simulated data. simulation uses real components, imaginary dimensions ignored.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"nsim = 1, returns matrix dissimilarities ( capscale) similar additional arguments random number seed simulate. nsim > 1, returns similar array returned simulate.nullmodel similar attributes.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"data(dune) data(dune.env) mod <- rda(dune ~ Moisture + Management, dune.env) ## One simulation update(mod, simulate(mod) ~ .) #> Call: rda(formula = simulate(mod) ~ Moisture + Management, data = #> dune.env) #> #> Inertia Proportion Rank #> Total 79.3906 1.0000 #> Constrained 52.2955 0.6587 6 #> Unconstrained 27.0951 0.3413 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 24.007 14.238 5.712 3.314 2.879 2.145 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 6.128 4.391 3.331 3.143 2.121 1.757 1.488 1.152 1.115 1.044 0.682 0.539 0.205 #> ## An impression of confidence regions of site scores plot(mod, display=\"sites\") for (i in 1:5) lines(procrustes(mod, update(mod, simulate(mod) ~ .)), col=\"blue\") ## Simulate a set of null communities with permutation of residuals simulate(mod, indx = shuffleSet(nrow(dune), 99)) #> An object of class “simulate.rda” #> ‘simulate index’ method (abundance, non-sequential) #> 20 x 30 matrix #> Number of permuted matrices = 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":null,"dir":"Reference","previous_headings":"","what":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Land birds islands covered coniferous forest Sipoo Archipelago, southern Finland.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map)"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"sipoo data frame contains data occurrences 50 land bird species 18 islands Sipoo Archipelago (Simberloff & Martin, 1991, Appendix 3). species referred 4+4 letter abbreviation Latin names (using five letters two species names make unique). sipoo.map data contains geographic coordinates islands ETRS89-TM35FIN coordinate system (EPSG:3067) areas islands hectares.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Simberloff, D. & Martin, J.-L. (1991). Nestedness insular avifaunas: simple summary statistics masking complex species patterns. Ornis Fennica 68:178--192.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map) plot(N ~ E, data=sipoo.map, asp = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":null,"dir":"Reference","previous_headings":"","what":"Minimum Spanning Tree — spantree","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree connecting points, disregarding dissimilarities threshold NA.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Minimum Spanning Tree — spantree","text":"","code":"spantree(d, toolong = 0) # S3 method for spantree as.hclust(x, ...) # S3 method for spantree cophenetic(x) spandepth(x) # S3 method for spantree plot(x, ord, cex = 0.7, type = \"p\", labels, dlim, FUN = sammon, ...) # S3 method for spantree lines(x, ord, display=\"sites\", col = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Minimum Spanning Tree — spantree","text":"d Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. x spantree result object. ord ordination configuration, ordination result known scores. cex Character expansion factor. type Observations plotted points type=\"p\" type=\"b\", text label type=\"t\". tree (lines) always plotted. labels Text used type=\"t\" node names missing. dlim ceiling value used highest cophenetic dissimilarity. FUN Ordination function find configuration cophenetic dissimilarities. supplied FUN work, supply ordination result argument ord. display Type scores used ord. col Colour line segments. can vector recycled points, line colour mixture two joined points. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree dissimilarities (may several minimum spanning trees, function finds one). Dissimilarities threshold toolong NAs disregarded, spanning tree found dissimilarities. data disconnected, function return disconnected tree (forest), corresponding link NA. Connected subtrees can identified using distconnected. Minimum spanning tree closely related single linkage clustering, .k.. nearest neighbour clustering, genetics neighbour joining tree available hclust agnes functions. important practical difference minimum spanning tree concept cluster membership, always joins individual points . Function .hclust can change spantree result corresponding hclust object. Function cophenetic finds distances points along tree segments. Function spandepth returns depth node. nodes tree either leaves (one link) internal nodes (one link). leaves recursively removed tree, depth layer leaf removed. disconnected spantree object (forest) tree analysed separately disconnected nodes tree depth zero. Function plot displays tree supplied ordination configuration, lines adds spanning tree ordination graph. configuration supplied plot, function ordinates cophenetic dissimilarities spanning tree overlays tree result. default ordination function sammon (package MASS), Sammon scaling emphasizes structure neighbourhood nodes may able beautifully represent tree (may need set dlim, sometimes results remain twisted). ordination methods work disconnected trees, must supply ordination configuration. Function lines overlay tree existing plot. Function spantree uses Prim's method implemented priority-first search dense graphs (Sedgewick 1990). Function cophenetic uses function stepacross option path = \"extended\". spantree fast, cophenetic slow large data sets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Minimum Spanning Tree — spantree","text":"Function spantree returns object class spantree list two vectors, length \\(n-1\\). number links tree one less number observations, first item omitted. items kid child node parent, starting parent number two. link parent, value NA tree disconnected node. dist Corresponding distance. kid = NA, dist = 0. labels Names nodes found input dissimilarities. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Minimum Spanning Tree — spantree","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Minimum Spanning Tree — spantree","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Minimum Spanning Tree — spantree","text":"principle, minimum spanning tree equivalent single linkage clustering can performed using hclust agnes. However, functions combine clusters information actually connected points (“single link”) recovered result. graphical output single linkage clustering plotted ordicluster look different equivalent spanning tree plotted lines.spantree.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Minimum Spanning Tree — spantree","text":"","code":"data(dune) dis <- vegdist(dune) tr <- spantree(dis) ## Add tree to a metric scaling plot(tr, cmdscale(dis), type = \"t\") ## Find a configuration to display the tree neatly plot(tr, type = \"t\") #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Depths of nodes depths <- spandepth(tr) plot(tr, type = \"t\", label = depths) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Plot as a dendrogram cl <- as.hclust(tr) plot(cl) ## cut hclust tree to classes and show in colours in spantree plot(tr, col = cutree(cl, 5), pch=16) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":null,"dir":"Reference","previous_headings":"","what":"Species Accumulation Curves — specaccum","title":"Species Accumulation Curves — specaccum","text":"Function specaccum finds species accumulation curves number species certain number sampled sites individuals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species Accumulation Curves — specaccum","text":"","code":"specaccum(comm, method = \"exact\", permutations = 100, conditioned =TRUE, gamma = \"jack1\", w = NULL, subset, ...) # S3 method for specaccum plot(x, add = FALSE, random = FALSE, ci = 2, ci.type = c(\"bar\", \"line\", \"polygon\"), col = par(\"fg\"), lty = 1, ci.col = col, ci.lty = 1, ci.length = 0, xlab, ylab = x$method, ylim, xvar = c(\"sites\", \"individuals\", \"effort\"), ...) # S3 method for specaccum boxplot(x, add = FALSE, ...) fitspecaccum(object, model, method = \"random\", ...) # S3 method for fitspecaccum plot(x, col = par(\"fg\"), lty = 1, xlab = \"Sites\", ylab = x$method, ...) # S3 method for specaccum predict(object, newdata, interpolation = c(\"linear\", \"spline\"), ...) # S3 method for fitspecaccum predict(object, newdata, ...) specslope(object, at)"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species Accumulation Curves — specaccum","text":"comm Community data set. method Species accumulation method (partial match). Method \"collector\" adds sites order happen data, \"random\" adds sites random order, \"exact\" finds expected (mean) species richness, \"coleman\" finds expected richness following Coleman et al. 1982, \"rarefaction\" finds mean accumulating individuals instead sites. permutations Number permutations method = \"random\". Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. conditioned Estimation standard deviation conditional empirical dataset exact SAC gamma Method estimating total extrapolated number species survey area function specpool w Weights giving sampling effort. subset logical expression indicating sites (rows) keep: missing values taken FALSE. x specaccum result object add Add existing graph. random Draw random simulation separately instead drawing average confidence intervals. ci Multiplier used get confidence intervals standard deviation (standard error estimate). Value ci = 0 suppresses drawing confidence intervals. ci.type Type confidence intervals graph: \"bar\" draws vertical bars, \"line\" draws lines, \"polygon\" draws shaded area. col Colour drawing lines. lty line type (see par). ci.col Colour drawing lines filling \"polygon\". ci.lty Line type confidence intervals border \"polygon\". ci.length Length horizontal bars (inches) end vertical bars ci.type = \"bar\". xlab,ylab Labels x (defaults xvar) y axis. ylim y limits plot. xvar Variable used horizontal axis: \"individuals\" can used method = \"rarefaction\". object Either community data set fitted specaccum model. model Nonlinear regression model (nls). See Details. newdata Optional data used prediction interpreted number sampling units (sites). missing, fitted values returned. interpolation Interpolation method used newdata. Number plots slope evaluated. Can real number. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species Accumulation Curves — specaccum","text":"Species accumulation curves (SAC) used compare diversity properties community data sets using different accumulator functions. classic method \"random\" finds mean SAC standard deviation random permutations data, subsampling without replacement (Gotelli & Colwell 2001). \"exact\" method finds expected SAC using sample-based rarefaction method independently developed numerous times (Chiarucci et al. 2008) often known Mao Tau estimate (Colwell et al. 2012). unconditional standard deviation exact SAC represents moment-based estimation conditioned empirical data set (sd samples > 0). unconditional standard deviation based estimation extrapolated number species survey area (.k.. gamma diversity), estimated function specpool. conditional standard deviation developed Jari Oksanen ( published, sd=0 samples). Method \"coleman\" finds expected SAC standard deviation following Coleman et al. (1982). methods based sampling sites without replacement. contrast, method = \"rarefaction\" finds expected species richness standard deviation sampling individuals instead sites. achieves applying function rarefy number individuals corresponding average number individuals per site. Methods \"random\" \"collector\" can take weights (w) give sampling effort site. weights w influence order sites accumulated, value sampling effort sites equal. summary results expressed sites even accumulation uses weights (methods \"random\", \"collector\"), based individuals (\"rarefaction\"). actual sampling effort given item Effort Individuals printed result. weighted \"random\" method effort refers average effort per site, sum weights per number sites. weighted method = \"random\", averaged species richness found linear interpolation single random permutations. Therefore least first value (often several first) NA richness, values interpolated cases extrapolated. plot function defaults display results scaled sites, can changed selecting xvar = \"effort\" (weighted methods) xvar = \"individuals\" (method = \"rarefaction\"). summary boxplot methods available method = \"random\". Function predict specaccum can return values corresponding newdata. method \"exact\", \"rarefaction\" \"coleman\" function uses analytic equations interpolated non-integer values, methods linear (approx) spline (spline) interpolation. newdata given, function returns values corresponding data. NB., fitted values method=\"rarefaction\" based rounded integer counts, predict can use fractional non-integer counts newdata give slightly different results. Function fitspecaccum fits nonlinear (nls) self-starting species accumulation model. input object can result specaccum community data frame. latter case function first fits specaccum model proceeds fitting nonlinear model. function can apply limited set nonlinear regression models suggested species-area relationship (Dengler 2009). selfStart models. permissible alternatives \"arrhenius\" (SSarrhenius), \"gleason\" (SSgleason), \"gitay\" (SSgitay), \"lomolino\" (SSlomolino) vegan package. addition following standard R models available: \"asymp\" (SSasymp), \"gompertz\" (SSgompertz), \"michaelis-menten\" (SSmicmen), \"logis\" (SSlogis), \"weibull\" (SSweibull). See functions model specification details. weights w used fit based accumulated effort model = \"rarefaction\" accumulated number individuals. plot still based sites, unless alternative selected xvar. Function predict fitspecaccum uses predict.nls, can pass arguments function. addition, fitted, residuals, nobs, coef, AIC, logLik deviance work result object. Function specslope evaluates derivative species accumulation curve given number sample plots, gives rate increase number species. function works specaccum result object based analytic models \"exact\", \"rarefaction\" \"coleman\", non-linear regression results fitspecaccum. Nonlinear regression may fail reason, fitspecaccum models fragile may succeed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species Accumulation Curves — specaccum","text":"Function specaccum returns object class \"specaccum\", fitspecaccum model class \"fitspecaccum\" adds items \"specaccum\" (see end list ): call Function call. method Accumulator method. sites Number sites. method = \"rarefaction\" number sites corresponding certain number individuals generally integer, average number individuals also returned item individuals. effort Average sum weights corresponding number sites model fitted argument w richness number species corresponding number sites. method = \"collector\" observed richness, methods average expected richness. sd standard deviation SAC (standard error). NULL method = \"collector\", estimated permutations method = \"random\", analytic equations methods. perm Permutation results method = \"random\" NULL cases. column perm holds one permutation. weights Matrix accumulated weights corresponding columns perm matrix model fitted argument w. fitted, residuals, coefficients fitspecacum: fitted values, residuals nonlinear model coefficients. method = \"random\" matrices column random accumulation. models fitspecaccum: list fitted nls models (see Examples accessing models).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Species Accumulation Curves — specaccum","text":"Chiarucci, ., Bacaro, G., Rocchini, D. & Fattorini, L. (2008). Discovering rediscovering sample-based rarefaction formula ecological literature. Commun. Ecol. 9: 121--123. Coleman, B.D, Mares, M.., Willis, M.R. & Hsieh, Y. (1982). Randomness, area species richness. Ecology 63: 1121--1133. Colwell, R.K., Chao, ., Gotelli, N.J., Lin, S.Y., Mao, C.X., Chazdon, R.L. & Longino, J.T. (2012). Models estimators linking individual-based sample-based rarefaction, extrapolation comparison assemblages. J. Plant Ecol. 5: 3--21. Dengler, J. (2009). function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity: procedures pitfalls measurement comparison species richness. Ecol. Lett. 4, 379--391.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species Accumulation Curves — specaccum","text":"Roeland Kindt r.kindt@cgiar.org Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Species Accumulation Curves — specaccum","text":"SAC method = \"exact\" developed Roeland Kindt, standard deviation Jari Oksanen (unpublished). method = \"coleman\" underestimates SAC handle properly sampling without replacement. , standard deviation take account species correlations, generally low.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species Accumulation Curves — specaccum","text":"","code":"data(BCI) sp1 <- specaccum(BCI) #> Warning: the standard deviation is zero sp2 <- specaccum(BCI, \"random\") sp2 #> Species Accumulation Curve #> Accumulation method: random, with 100 permutations #> Call: specaccum(comm = BCI, method = \"random\") #> #> #> Sites 1.00000 2.00000 3.00000 4.00000 5.00000 6.00000 7.00000 #> Richness 90.42000 121.29000 138.34000 150.46000 158.46000 165.38000 170.34000 #> sd 7.23931 7.43361 7.38305 6.80377 5.92959 5.07475 4.58196 #> #> Sites 8.00000 9.00000 10.00000 11.0000 12.00000 13.00000 14.0000 #> Richness 175.02000 178.93000 182.15000 184.8500 187.47000 189.88000 191.9000 #> sd 4.43808 4.74662 4.58671 4.4093 4.23419 4.08565 4.1451 #> #> Sites 15.00000 16.00000 17.00000 18.00000 19.00000 20.00000 21.00000 #> Richness 193.75000 195.50000 197.32000 198.94000 200.32000 201.68000 202.95000 #> sd 4.13503 4.01889 3.54447 3.53859 3.62617 3.62895 3.41528 #> #> Sites 22.00000 23.00000 24.00000 25.00000 26.00000 27.000 28.00000 #> Richness 204.17000 205.39000 206.65000 207.73000 208.91000 209.880 210.90000 #> sd 3.47009 3.28417 3.22357 3.06448 3.06856 3.217 3.14787 #> #> Sites 29.00000 30.00000 31.00000 32.00000 33.00000 34.00000 35.00000 #> Richness 211.64000 212.74000 213.57000 214.43000 215.15000 215.93000 216.75000 #> sd 3.12862 3.01719 3.01931 2.92069 2.76111 2.77163 2.69446 #> #> Sites 36.00000 37.00000 38.00000 39.00000 40.00000 41.00000 42.00000 #> Richness 217.54000 218.20000 219.00000 219.61000 220.14000 220.77000 221.28000 #> sd 2.62629 2.64384 2.45361 2.22427 2.11307 1.93769 1.82618 #> #> Sites 43.00000 44.00000 45.00000 46.00000 47.00000 48.0000 49.00000 #> Richness 221.86000 222.33000 222.85000 223.28000 223.69000 224.1400 224.60000 #> sd 1.79235 1.49784 1.45904 1.36389 1.07961 0.8764 0.58603 #> #> Sites 50 #> Richness 225 #> sd 0 summary(sp2) #> 1 sites 2 sites 3 sites 4 sites #> Min. : 77.00 Min. :105.0 Min. :119.0 Min. :132.0 #> 1st Qu.: 85.00 1st Qu.:115.8 1st Qu.:134.0 1st Qu.:146.0 #> Median : 88.00 Median :121.5 Median :139.0 Median :151.0 #> Mean : 90.42 Mean :121.3 Mean :138.3 Mean :150.5 #> 3rd Qu.: 94.25 3rd Qu.:126.0 3rd Qu.:143.2 3rd Qu.:155.0 #> Max. :109.00 Max. :144.0 Max. :155.0 Max. :166.0 #> 5 sites 6 sites 7 sites 8 sites 9 sites #> Min. :140.0 Min. :149.0 Min. :157.0 Min. :161 Min. :166.0 #> 1st Qu.:155.0 1st Qu.:162.0 1st Qu.:168.0 1st Qu.:173 1st Qu.:176.0 #> Median :159.0 Median :165.0 Median :171.0 Median :175 Median :179.0 #> Mean :158.5 Mean :165.4 Mean :170.3 Mean :175 Mean :178.9 #> 3rd Qu.:163.0 3rd Qu.:170.0 3rd Qu.:174.0 3rd Qu.:178 3rd Qu.:182.0 #> Max. :172.0 Max. :176.0 Max. :181.0 Max. :186 Max. :190.0 #> 10 sites 11 sites 12 sites 13 sites #> Min. :170.0 Min. :174.0 Min. :174.0 Min. :179.0 #> 1st Qu.:179.0 1st Qu.:182.0 1st Qu.:184.0 1st Qu.:187.0 #> Median :183.0 Median :185.0 Median :188.0 Median :190.5 #> Mean :182.2 Mean :184.8 Mean :187.5 Mean :189.9 #> 3rd Qu.:185.0 3rd Qu.:188.0 3rd Qu.:191.0 3rd Qu.:193.0 #> Max. :191.0 Max. :193.0 Max. :200.0 Max. :202.0 #> 14 sites 15 sites 16 sites 17 sites #> Min. :181.0 Min. :183.0 Min. :185.0 Min. :188.0 #> 1st Qu.:189.0 1st Qu.:191.0 1st Qu.:193.0 1st Qu.:195.0 #> Median :193.0 Median :194.0 Median :196.0 Median :197.5 #> Mean :191.9 Mean :193.8 Mean :195.5 Mean :197.3 #> 3rd Qu.:194.0 3rd Qu.:196.0 3rd Qu.:198.0 3rd Qu.:200.0 #> Max. :203.0 Max. :204.0 Max. :205.0 Max. :205.0 #> 18 sites 19 sites 20 sites 21 sites #> Min. :188.0 Min. :188.0 Min. :190.0 Min. :195.0 #> 1st Qu.:196.8 1st Qu.:198.0 1st Qu.:199.0 1st Qu.:201.0 #> Median :199.0 Median :200.0 Median :202.0 Median :203.0 #> Mean :198.9 Mean :200.3 Mean :201.7 Mean :202.9 #> 3rd Qu.:201.0 3rd Qu.:203.0 3rd Qu.:204.0 3rd Qu.:205.0 #> Max. :208.0 Max. :209.0 Max. :210.0 Max. :211.0 #> 22 sites 23 sites 24 sites 25 sites #> Min. :195.0 Min. :199.0 Min. :199.0 Min. :200.0 #> 1st Qu.:202.0 1st Qu.:203.0 1st Qu.:204.0 1st Qu.:206.0 #> Median :204.0 Median :205.0 Median :207.0 Median :207.5 #> Mean :204.2 Mean :205.4 Mean :206.7 Mean :207.7 #> 3rd Qu.:206.2 3rd Qu.:207.0 3rd Qu.:209.0 3rd Qu.:210.0 #> Max. :214.0 Max. :214.0 Max. :214.0 Max. :215.0 #> 26 sites 27 sites 28 sites 29 sites #> Min. :203.0 Min. :203.0 Min. :204.0 Min. :204.0 #> 1st Qu.:207.0 1st Qu.:208.0 1st Qu.:209.0 1st Qu.:210.0 #> Median :209.0 Median :210.0 Median :211.0 Median :212.0 #> Mean :208.9 Mean :209.9 Mean :210.9 Mean :211.6 #> 3rd Qu.:211.0 3rd Qu.:212.0 3rd Qu.:213.0 3rd Qu.:214.0 #> Max. :217.0 Max. :218.0 Max. :220.0 Max. :220.0 #> 30 sites 31 sites 32 sites 33 sites #> Min. :206.0 Min. :206.0 Min. :206.0 Min. :207.0 #> 1st Qu.:211.0 1st Qu.:212.0 1st Qu.:213.0 1st Qu.:214.0 #> Median :212.5 Median :213.0 Median :214.0 Median :215.0 #> Mean :212.7 Mean :213.6 Mean :214.4 Mean :215.2 #> 3rd Qu.:215.0 3rd Qu.:215.0 3rd Qu.:216.0 3rd Qu.:216.0 #> Max. :221.0 Max. :221.0 Max. :222.0 Max. :222.0 #> 34 sites 35 sites 36 sites 37 sites #> Min. :208.0 Min. :208.0 Min. :208.0 Min. :208.0 #> 1st Qu.:214.0 1st Qu.:215.0 1st Qu.:216.0 1st Qu.:217.0 #> Median :216.0 Median :217.0 Median :218.0 Median :218.0 #> Mean :215.9 Mean :216.8 Mean :217.5 Mean :218.2 #> 3rd Qu.:218.0 3rd Qu.:219.0 3rd Qu.:219.0 3rd Qu.:220.0 #> Max. :222.0 Max. :222.0 Max. :223.0 Max. :224.0 #> 38 sites 39 sites 40 sites 41 sites #> Min. :212.0 Min. :212.0 Min. :212.0 Min. :212.0 #> 1st Qu.:217.0 1st Qu.:218.0 1st Qu.:219.0 1st Qu.:220.0 #> Median :219.0 Median :220.0 Median :220.0 Median :221.0 #> Mean :219.0 Mean :219.6 Mean :220.1 Mean :220.8 #> 3rd Qu.:220.2 3rd Qu.:221.0 3rd Qu.:221.0 3rd Qu.:222.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 42 sites 43 sites 44 sites 45 sites #> Min. :215.0 Min. :215.0 Min. :218.0 Min. :219.0 #> 1st Qu.:220.0 1st Qu.:221.0 1st Qu.:221.0 1st Qu.:222.0 #> Median :221.0 Median :222.0 Median :222.0 Median :223.0 #> Mean :221.3 Mean :221.9 Mean :222.3 Mean :222.8 #> 3rd Qu.:222.2 3rd Qu.:223.0 3rd Qu.:223.2 3rd Qu.:224.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 46 sites 47 sites 48 sites 49 sites 50 sites #> Min. :220.0 Min. :221.0 Min. :222.0 Min. :223.0 Min. :225 #> 1st Qu.:222.8 1st Qu.:223.0 1st Qu.:224.0 1st Qu.:224.0 1st Qu.:225 #> Median :223.0 Median :224.0 Median :224.0 Median :225.0 Median :225 #> Mean :223.3 Mean :223.7 Mean :224.1 Mean :224.6 Mean :225 #> 3rd Qu.:224.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 Max. :225 plot(sp1, ci.type=\"poly\", col=\"blue\", lwd=2, ci.lty=0, ci.col=\"lightblue\") boxplot(sp2, col=\"yellow\", add=TRUE, pch=\"+\") ## Fit Lomolino model to the exact accumulation mod1 <- fitspecaccum(sp1, \"lomolino\") coef(mod1) #> Asym xmid slope #> 258.440682 2.442061 1.858694 fitted(mod1) #> [1] 94.34749 121.23271 137.45031 148.83053 157.45735 164.31866 169.95946 #> [8] 174.71115 178.78954 182.34254 185.47566 188.26658 190.77402 193.04337 #> [15] 195.11033 197.00350 198.74606 200.35705 201.85227 203.24499 204.54643 #> [22] 205.76612 206.91229 207.99203 209.01150 209.97609 210.89054 211.75903 #> [29] 212.58527 213.37256 214.12386 214.84180 215.52877 216.18692 216.81820 #> [36] 217.42437 218.00703 218.56767 219.10762 219.62811 220.13027 220.61514 #> [43] 221.08369 221.53679 221.97528 222.39991 222.81138 223.21037 223.59747 #> [50] 223.97327 plot(sp1) ## Add Lomolino model using argument 'add' plot(mod1, add = TRUE, col=2, lwd=2) ## Fit Arrhenius models to all random accumulations mods <- fitspecaccum(sp2, \"arrh\") plot(mods, col=\"hotpink\") boxplot(sp2, col = \"yellow\", border = \"blue\", lty=1, cex=0.3, add= TRUE) ## Use nls() methods to the list of models sapply(mods$models, AIC) #> [1] 329.2456 321.8614 331.1568 352.1309 354.8606 323.3767 322.0897 347.7218 #> [9] 327.6283 327.3510 335.4843 308.5300 373.4959 334.3691 320.6975 338.1084 #> [17] 330.2018 343.5560 348.5827 300.2513 315.7909 310.7837 305.5217 295.1005 #> [25] 359.2639 342.7111 339.7456 328.4175 338.4335 368.4988 354.7097 328.8704 #> [33] 342.1528 292.9875 328.3752 299.7201 336.5796 364.3153 336.4822 339.6666 #> [41] 354.7445 320.7841 304.1628 314.9640 349.6753 350.7169 307.6840 296.7971 #> [49] 359.6018 302.9863 319.6832 339.5021 328.8450 329.0553 298.0970 334.0182 #> [57] 291.9027 349.7161 325.1376 347.7797 340.6111 338.6148 352.0292 308.8538 #> [65] 315.1890 289.7947 351.6732 339.3587 320.3515 300.9377 323.3054 290.9419 #> [73] 350.8587 322.6970 340.7948 334.1965 358.2890 300.7034 338.5079 338.8941 #> [81] 341.7687 331.4553 337.5975 349.7867 331.8581 270.3068 354.5083 326.0026 #> [89] 337.8555 322.7217 302.5019 324.7469 322.1415 324.6922 324.5427 347.2050 #> [97] 309.1766 316.1194 343.9958 350.5935"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":null,"dir":"Reference","previous_headings":"","what":"Extrapolated Species Richness in a Species Pool — specpool","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions estimate extrapolated species richness species pool, number unobserved species. Function specpool based incidences sample sites, gives single estimate collection sample sites (matrix). Function estimateR based abundances (counts) single sample site.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"specpool(x, pool, smallsample = TRUE) estimateR(x, ...) specpool2vect(X, index = c(\"jack1\",\"jack2\", \"chao\", \"boot\",\"Species\")) poolaccum(x, permutations = 100, minsize = 3) estaccumR(x, permutations = 100, parallel = getOption(\"mc.cores\")) # S3 method for poolaccum summary(object, display, alpha = 0.05, ...) # S3 method for poolaccum plot(x, alpha = 0.05, type = c(\"l\",\"g\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"x Data frame matrix species data analysis result plot function. pool vector giving classification pooling sites species data. missing, sites pooled together. smallsample Use small sample correction \\((N-1)/N\\), \\(N\\) number sites within pool. X, object specpool result object. index selected index extrapolated richness. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. minsize Smallest number sampling units reported. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. display Indices displayed. alpha Level quantiles shown. proportion left outside symmetric limits. type Type graph produced xyplot. ... parameters (used).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Many species always remain unseen undetected collection sample plots. function uses popular ways estimating number unseen species adding observed species richness (Palmer 1990, Colwell & Coddington 1994). incidence-based estimates specpool use frequencies species collection sites. following, \\(S_P\\) extrapolated richness pool, \\(S_0\\) observed number species collection, \\(a_1\\) \\(a_2\\) number species occurring one two sites collection, \\(p_i\\) frequency species \\(\\), \\(N\\) number sites collection. variants extrapolated richness specpool : specpool normally uses basic Chao equation, doubletons (\\(a2=0\\)) switches bias-corrected version. case Chao equation simplifies \\(S_0 + \\frac{1}{2} a_1 (a_1-1) \\frac{N-1}{N}\\). abundance-based estimates estimateR use counts (numbers individuals) species single site. called matrix data frame, function give separate estimates site. two variants extrapolated richness estimateR bias-corrected Chao ACE (O'Hara 2005, Chiu et al. 2014). Chao estimate similar bias corrected one , \\(a_i\\) refers number species abundance \\(\\) instead number sites, small-sample correction used. ACE estimate defined : \\(a_i\\) refers number species abundance \\(\\) \\(S_{rare}\\) number rare species, \\(S_{abund}\\) number abundant species, arbitrary threshold abundance 10 rare species, \\(N_{rare}\\) number individuals rare species. Functions estimate standard errors estimates. concern number added species, assume variance observed richness. equations standard errors complicated reproduced help page, can studied R source code function discussed vignette can read browseVignettes(\"vegan\"). standard error based following sources: Chiu et al. (2014) Chao estimates Smith van Belle (1984) first-order Jackknife bootstrap (second-order jackknife still missing). variance estimator \\(S_{ace}\\) see O'Hara (2005). Functions poolaccum estaccumR similar specaccum, estimate extrapolated richness indices specpool estimateR addition number species random ordering sampling units. Function specpool uses presence data estaccumR count data. functions share summary plot methods. summary returns quantile envelopes permutations corresponding given level alpha standard deviation permutations sample size. NB., based standard deviations estimated within specpool estimateR, based permutations. plot function shows mean envelope permutations given alpha models. selection models can restricted order changes using display argument summary plot. configuration plot command, see xyplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Function specpool returns data frame entries observed richness indices class pool vector. utility function specpool2vect maps pooled values vector giving value selected index original site. Function estimateR returns estimates standard errors site. Functions poolaccum estimateR return matrices permutation results richness estimator, vector sample sizes table means permutations estimator.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Chao, . (1987). Estimating population size capture-recapture data unequal catchability. Biometrics 43, 783--791. Chiu, C.H., Wang, Y.T., Walther, B.. & Chao, . (2014). Improved nonparametric lower bound species richness via modified Good-Turing frequency formula. Biometrics 70, 671--682. Colwell, R.K. & Coddington, J.. (1994). Estimating terrestrial biodiversity extrapolation. Phil. Trans. Roy. Soc. London B 345, 101--118. O'Hara, R.B. (2005). Species richness estimators: many species can dance head pin? J. Anim. Ecol. 74, 375--386. Palmer, M.W. (1990). estimation species richness extrapolation. Ecology 71, 1195--1198. Smith, E.P & van Belle, G. (1984). Nonparametric estimation species richness. Biometrics 40, 119--129.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Bob O'Hara (estimateR) Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions based assumption species pool: community closed fixed pool size \\(S_P\\). general, functions give lower limit species richness: real richness \\(S >= S_P\\), consistent bias estimates. Even bias-correction Chao reduces bias, remove completely (Chiu et al. 2014). Optional small sample correction added specpool vegan 2.2-0. used older literature (Chao 1987), recommended recently (Chiu et al. 2014).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"data(dune) data(dune.env) pool <- with(dune.env, specpool(dune, Management)) pool #> Species chao chao.se jack1 jack1.se jack2 boot boot.se n #> BF 16 17.19048 1.5895675 19.33333 2.211083 19.83333 17.74074 1.646379 3 #> HF 21 21.51429 0.9511693 23.40000 1.876166 22.05000 22.56864 1.821518 5 #> NM 21 22.87500 2.1582871 26.00000 3.291403 25.73333 23.77696 2.300982 6 #> SF 21 29.88889 8.6447967 27.66667 3.496029 31.40000 23.99496 1.850288 6 op <- par(mfrow=c(1,2)) boxplot(specnumber(dune) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") boxplot(specnumber(dune)/specpool2vect(pool) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") par(op) data(BCI) ## Accumulation model pool <- poolaccum(BCI) summary(pool, display = \"chao\") #> $chao #> N Chao 2.5% 97.5% Std.Dev #> [1,] 3 162.3374 142.5272 186.2227 11.217082 #> [2,] 4 176.1243 156.4704 206.3243 12.158304 #> [3,] 5 183.8507 162.9868 209.2821 12.488675 #> [4,] 6 188.9018 165.6732 214.2779 12.888148 #> [5,] 7 193.8679 175.8712 216.1384 11.602711 #> [6,] 8 199.0315 180.2949 227.8918 12.604562 #> [7,] 9 202.0946 183.3317 228.8381 11.958432 #> [8,] 10 204.9843 185.3556 227.5898 12.637762 #> [9,] 11 206.6160 186.7205 234.5167 11.679759 #> [10,] 12 209.1286 189.2824 232.6359 12.351073 #> [11,] 13 211.3655 191.7918 232.0423 11.991967 #> [12,] 14 212.8128 195.8592 239.3318 11.537817 #> [13,] 15 215.5275 198.2739 239.7978 11.422201 #> [14,] 16 218.7094 198.3891 243.3826 12.165145 #> [15,] 17 221.6814 197.6029 256.5328 14.464302 #> [16,] 18 222.7418 201.4964 257.9494 13.227892 #> [17,] 19 224.8619 206.6236 251.9979 13.165360 #> [18,] 20 226.3324 210.2832 249.5616 12.858116 #> [19,] 21 228.4284 210.9920 257.9636 13.784663 #> [20,] 22 228.8425 211.1903 254.3500 11.629250 #> [21,] 23 230.6429 214.1737 251.1681 10.510649 #> [22,] 24 231.9109 213.0874 252.7553 10.925851 #> [23,] 25 232.9972 214.3610 258.4917 11.207306 #> [24,] 26 233.8067 216.9697 263.2733 11.546339 #> [25,] 27 235.6347 217.3831 266.9230 12.800245 #> [26,] 28 235.8942 218.6913 261.4115 11.761873 #> [27,] 29 235.9612 218.6591 258.5682 11.133700 #> [28,] 30 236.6706 219.4315 259.9315 10.291499 #> [29,] 31 236.8530 220.2338 259.2903 9.810390 #> [30,] 32 237.6018 221.4868 262.0864 10.195825 #> [31,] 33 237.4291 222.2990 256.2778 9.227891 #> [32,] 34 237.6712 222.7844 258.2268 8.979536 #> [33,] 35 237.7856 222.2750 257.2540 8.537867 #> [34,] 36 238.4325 223.6473 255.5610 8.566487 #> [35,] 37 238.1834 224.9892 253.5811 7.776992 #> [36,] 38 238.4685 225.2802 254.0155 7.626532 #> [37,] 39 238.2082 226.0390 252.7975 7.642443 #> [38,] 40 238.7991 227.1971 254.1619 7.155802 #> [39,] 41 239.2246 227.3368 258.5178 7.926574 #> [40,] 42 238.4935 228.5195 256.2137 7.287824 #> [41,] 43 238.3856 227.8327 255.8470 7.300153 #> [42,] 44 238.0092 228.5279 252.9488 6.520961 #> [43,] 45 238.1537 228.7980 253.3757 6.443894 #> [44,] 46 238.0532 230.5990 250.9539 5.466587 #> [45,] 47 237.7542 231.3089 251.0192 4.821564 #> [46,] 48 237.2842 231.9015 248.6399 4.038862 #> [47,] 49 236.8407 233.3115 245.4082 2.803888 #> [48,] 50 236.3732 236.3732 236.3732 0.000000 #> #> attr(,\"class\") #> [1] \"summary.poolaccum\" plot(pool) ## Quantitative model estimateR(BCI[1:5,]) #> 1 2 3 4 5 #> S.obs 93.000000 84.000000 90.000000 94.000000 101.000000 #> S.chao1 117.473684 117.214286 141.230769 111.550000 136.000000 #> se.chao1 11.583785 15.918953 23.001405 8.919663 15.467344 #> S.ACE 122.848959 117.317307 134.669844 118.729941 137.114088 #> se.ACE 5.736054 5.571998 6.191618 5.367571 5.848474"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distance-based ordination (dbrda, capscale, metaMDS) information species, methods may add species scores community data available. However, species scores may missing ( always dbrda), may close relation used dissimilarity index. function add species scores replace existing species scores distance-based methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"sppscores(object) <- value"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"object Ordination result. value Community data find species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distances information species (columns, variables), hence distance-based ordination information species scores. However, species scores can added supplementary information analysis help interpretation results. ordination methods (capscale, metaMDS) can supplement species scores analysis community data available analysis. capscale species scores found projecting community data site ordination (linear combination scores), scores accurate analysis used Euclidean distances. dissimilarity index can expressed Euclidean distances transformed data (instance, Chord Hellinger Distances), species scores based transformed data accurate, function still finds dissimilarities untransformed data. Usually community dissimilarities differ two significant ways Euclidean distances: bound maximum 1, use absolute differences instead squared differences. cases, may better use species scores transformed Euclidean distances good linear relation used dissimilarities. often useful standardize data row unit total, perform squareroot transformation damp effect squared differences (see Examples). Function dbrda never finds species scores, mathematically similar capscale, similar rules followed supplementing species scores. Function metaMDS uses weighted averages (wascores) find species scores. better relationship dissimilarities projection scores used metric ordination, similar transformation community data used dissimilarities species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Replacement function adds species scores replaces old scores ordination object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"data(BCI, BCI.env) mod <- dbrda(vegdist(BCI) ~ Habitat, BCI.env) ## add species scores sppscores(mod) <- BCI ## Euclidean distances of BCI differ from used dissimilarity plot(vegdist(BCI), dist(BCI)) ## more linear relationship plot(vegdist(BCI), dist(sqrt(decostand(BCI, \"total\")))) ## better species scores sppscores(mod) <- sqrt(decostand(BCI, \"total\"))"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":null,"dir":"Reference","previous_headings":"","what":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function stepacross tries replace dissimilarities shortest paths stepping across intermediate sites regarding dissimilarities threshold missing data (NA). path = \"shortest\" flexible shortest path (Williamson 1978, Bradfield & Kenkel 1987), path = \"extended\" approximation known extended dissimilarities (De'ath 1999). use stepacross improve ordination high beta diversity, many sites species common.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"stepacross(dis, path = \"shortest\", toolong = 1, trace = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. path method stepping across (partial match) Alternative \"shortest\" finds shortest paths, \"extended\" approximation known extended dissimilarities. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . trace Trace calculations. ... parameters (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Williamson (1978) suggested using flexible shortest paths estimate dissimilarities sites nothing common, shared species. path = \"shortest\" function stepacross replaces dissimilarities toolong longer NA, tries find shortest paths sites using remaining dissimilarities. Several dissimilarity indices semi-metric means obey triangle inequality \\(d_{ij} \\leq d_{ik} + d_{kj}\\), shortest path algorithm can replace dissimilarities well, even shorter toolong. De'ath (1999) suggested simplified method known extended dissimilarities, calculated path = \"extended\". method, dissimilarities toolong longer first made NA, function tries replace NA dissimilarities path single stepping stone points. NA replaced one pass, function make new passes updated dissimilarities long NA replaced extended dissimilarities. mean second passes, remaining NA dissimilarities allowed one stepping stone site, previously replaced dissimilarities updated. , function consider dissimilarities shorter toolong, although replaced shorter path semi-metric indices, used part paths. optimal cases, extended dissimilarities equal shortest paths, may longer. alternative defining long dissimilarities parameter toolong, input dissimilarities can contain NAs. toolong zero negative, function make dissimilarities NA. NAs input toolong = 0, path = \"shortest\" find shorter paths semi-metric indices, path = \"extended\" nothing. Function .shared can used set dissimilarities NA. data disconnected path points, result contain NAs warning issued. Several methods handle NA dissimilarities, warning taken seriously. Function distconnected can used find connected groups remove rare outlier observations groups observations. Alternative path = \"shortest\" uses Dijkstra's method finding flexible shortest paths, implemented priority-first search dense graphs (Sedgewick 1990). Alternative path = \"extended\" follows De'ath (1999), implementation simpler code.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function returns object class dist extended dissimilarities (see functions vegdist dist). value path appended method attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Bradfield, G.E. & Kenkel, N.C. (1987). Nonlinear ordination using flexible shortest path adjustment ecological distances. Ecology 68, 750--753. De'ath, G. (1999). Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecol. 144, 191--199. Sedgewick, R. (1990). Algorithms C. Addison Wesley. Williamson, M.H. (1978). ordination incidence data. J. Ecol. 66, 911-920.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"function changes original dissimilarities, like . may best use function really must: extremely high beta diversity large proportion dissimilarities upper limit (species common). Semi-metric indices vary degree violating triangle inequality. Morisita Horn--Morisita indices vegdist may strongly semi-metric, shortest paths can change indices much. Mountford index violates basic rules dissimilarities: non-identical sites zero dissimilarity species composition poorer site subset richer. Mountford index, can find three sites \\(, j, k\\) \\(d_{ik} = 0\\) \\(d_{jk} = 0\\), \\(d_{ij} > 0\\). results stepacross Mountford index can weird. stepacross needed, best try use metric indices .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"# There are no data sets with high beta diversity in vegan, but this # should give an idea. data(dune) dis <- vegdist(dune) edis <- stepacross(dis) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities... plot(edis, dis, xlab = \"Shortest path\", ylab = \"Original\") ## Manhattan distance have no fixed upper limit. dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune) dis <- stepacross(dis, toolong=0) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities..."},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions plot ordination distances given number dimensions observed distances distances full space eigenvector methods. display similar Shepard diagram (stressplot non-metric multidimensional scaling metaMDS monoMDS), shows linear relationship eigenvector ordinations. stressplot methods available wcmdscale, rda, cca, capscale, dbrda, prcomp princomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"# S3 method for wcmdscale stressplot(object, k = 2, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"object Result object eigenvector ordination (wcmdscale, rda, cca, dbrda, capscale) k Number dimensions ordination distances displayed. pch, p.col, l.col, lwd Plotting character, point colour line colour like default stressplot ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"functions offer similar display eigenvector ordinations standard Shepard diagram (stressplot) non-metric multidimensional scaling. ordination distances given number dimensions plotted observed distances. metric distances, ordination distances full space (ordination axes) equal observed distances, fit line shows equality. general, fit line go points, points observed distances approach fit line . However, non-Euclidean distances ( wcmdscale, dbrda capscale) negative eigenvalues ordination distances can exceed observed distances real dimensions; imaginary dimensions negative eigenvalues correct excess distances. used dbrda, capscale wcmdscale argument add avoid negative eigenvalues, ordination distances exceed observed dissimilarities. partial ordination (cca, rda, capscale Condition formula), distances partial component included observed distances ordination distances. k=0, ordination distances refer partial ordination. exception dbrda distances partial, constrained residual components additive, first components can shown, partial models shown .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions draw graph return invisibly ordination distances ordination distances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"data(dune, dune.env) mod <- rda(dune) stressplot(mod) mod <- rda(dune ~ Management, dune.env) stressplot(mod, k=3)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of Taxonomic Diversity and Distinctness — taxondive","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function finds indices taxonomic diversity distinctness, averaged taxonomic distances among species individuals community (Clarke & Warwick 1998, 2001)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"taxondive(comm, dis, match.force = FALSE) taxa2dist(x, varstep = FALSE, check = TRUE, labels)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"comm Community data. dis Taxonomic distances among taxa comm. dist object symmetric square matrix. match.force Force matching column names comm labels dis. FALSE, matching happens dimensions differ, case species must identical order . x Classification table row species basic taxon, columns identifiers classification higher levels. varstep Vary step lengths successive levels relative proportional loss number distinct classes. check TRUE, remove redundant levels different rows constant rows regard row different basal taxon (species). FALSE levels retained basal taxa (species) also must coded variables (columns). get warning species coded, can ignore intention. labels labels attribute taxonomic distances. Row names used given. Species matched labels comm dis taxondive different dimensions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke & Warwick (1998, 2001) suggested several alternative indices taxonomic diversity distinctness. Two basic indices called taxonomic diversity (\\(\\Delta\\)) distinctness (\\(\\Delta^*\\)): equations give index value single site, summation goes species \\(\\) \\(j\\). \\(\\omega\\) taxonomic distances among taxa, \\(x\\) species abundances, \\(n\\) total abundance site. presence/absence data indices reduce index \\(\\Delta^+\\), index Clarke & Warwick (1998) also estimate standard deviation. Clarke & Warwick (2001) presented two new indices: \\(s\\Delta^+\\) product species richness \\(\\Delta^+\\), index variation taxonomic distinctness (\\(\\Lambda^+\\)) defined dis argument must species dissimilarities. must similar dissimilarities produced dist. customary integer steps taxonomic hierarchies, kind dissimilarities can used, phylogenetic trees genetic differences. , dis need taxonomic, species classifications can used. Function taxa2dist can produce suitable dist object classification table. species (basic taxon) corresponds row classification table, columns give classification different levels. varstep = FALSE successive levels separated equal steps, varstep = TRUE step length relative proportional decrease number classes (Clarke & Warwick 1999). check = TRUE, function removes classes distinct species combine species one class, assumes row presents distinct basic taxon. function scales distances longest path length taxa 100 (necessarily check = FALSE). Function plot.taxondive plots \\(\\Delta^+\\) Number species, together expectation approximate 2*sd limits. Function summary.taxondive finds \\(z\\) values significances Normal distribution \\(\\Delta^+\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function returns object class taxondive following items: Species Number species site. D, Dstar, Dplus, SDplus, Lambda \\(\\Delta\\), \\(\\Delta^*\\), \\(\\Delta^+\\), \\(s\\Delta^+\\) \\(\\Lambda^+\\) site. sd.Dplus Standard deviation \\(\\Delta^+\\). ED, EDstar, EDplus Expected values corresponding statistics. Function taxa2dist returns object class \"dist\", attribute \"steps\" step lengths successive levels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke, K.R & Warwick, R.M. (1998) taxonomic distinctness index statistical properties. Journal Applied Ecology 35, 523--531. Clarke, K.R. & Warwick, R.M. (1999) taxonomic distinctness measure biodiversity: weighting step lengths hierarchical levels. Marine Ecology Progress Series 184: 21--29. Clarke, K.R. & Warwick, R.M. (2001) biodiversity index applicable species lists: variation taxonomic distinctness. Marine Ecology Progress Series 216, 265--278.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"function still preliminary may change. scaling taxonomic dissimilarities influences results. multiply taxonomic distances (step lengths) constant, values Deltas multiplied constant, value \\(\\Lambda^+\\) square constant.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"## Preliminary: needs better data and some support functions data(dune) data(dune.taxon) # Taxonomic distances from a classification table with variable step lengths. taxdis <- taxa2dist(dune.taxon, varstep=TRUE) plot(hclust(taxdis), hang = -1) # Indices mod <- taxondive(dune, taxdis) mod #> Species Delta Delta* Lambda+ Delta+ S Delta+ #> 1 5.000 22.736 29.232 900.298 43.364 216.82 #> 2 10.000 51.046 55.988 822.191 56.232 562.32 #> 3 10.000 41.633 46.194 1025.471 62.869 628.69 #> 4 13.000 50.795 55.140 888.244 64.837 842.88 #> 5 14.000 63.498 67.856 715.393 69.211 968.95 #> 6 11.000 70.201 76.361 628.743 73.281 806.09 #> 7 13.000 61.605 66.187 679.337 69.918 908.94 #> 8 12.000 52.544 56.374 756.375 66.729 800.74 #> 9 13.000 50.526 54.108 849.448 63.205 821.67 #> 10 12.000 60.068 64.960 730.736 69.291 831.49 #> 11 9.000 69.589 77.740 404.609 77.803 700.23 #> 12 9.000 62.405 69.795 552.129 74.470 670.23 #> 13 10.000 47.316 53.842 536.429 66.657 666.57 #> 14 7.000 71.383 82.091 239.543 82.013 574.09 #> 15 8.000 68.564 77.097 334.889 79.010 632.08 #> 16 8.000 55.984 64.400 978.014 69.708 557.66 #> 17 7.000 53.913 60.222 632.990 59.286 415.00 #> 18 9.000 73.235 81.865 438.355 76.288 686.59 #> 19 9.000 68.727 76.091 336.364 78.636 707.73 #> 20 8.000 72.343 80.670 444.915 82.078 656.62 #> Expected 65.330 62.560 71.031 summary(mod) #> Delta Delta* Delta+ sd(Delta+) z(Delta+) Pr(>|z|) #> 1 22.7362 29.2322 43.3636 10.0499 -2.7530 0.005905 ** #> 2 51.0458 55.9878 56.2323 5.7727 -2.5636 0.010359 * #> 3 41.6334 46.1936 62.8687 5.7727 -1.4140 0.157360 #> 4 50.7952 55.1396 64.8368 4.5677 -1.3562 0.175047 #> 5 63.4979 67.8564 69.2108 4.2482 -0.4285 0.668251 #> 6 70.2011 76.3614 73.2810 5.3189 0.4229 0.672332 #> 7 61.6049 66.1871 69.9184 4.5677 -0.2437 0.807499 #> 8 52.5437 56.3743 66.7287 4.9215 -0.8743 0.381975 #> 9 50.5258 54.1079 63.2051 4.5677 -1.7134 0.086640 . #> 10 60.0680 64.9597 69.2906 4.9215 -0.3537 0.723567 #> 11 69.5894 77.7396 77.8030 6.3011 1.0747 0.282517 #> 12 62.4049 69.7949 74.4697 6.3011 0.5457 0.585290 #> 13 47.3158 53.8421 66.6566 5.7727 -0.7578 0.448548 #> 14 71.3834 82.0909 82.0130 7.7069 1.4249 0.154186 #> 15 68.5645 77.0970 79.0097 6.9314 1.1510 0.249714 #> 16 55.9840 64.3999 69.7078 6.9314 -0.1909 0.848565 #> 17 53.9134 60.2224 59.2857 7.7069 -1.5240 0.127501 #> 18 73.2349 81.8645 76.2879 6.3011 0.8342 0.404155 #> 19 68.7273 76.0909 78.6364 6.3011 1.2069 0.227457 #> 20 72.3431 80.6704 82.0779 6.9314 1.5937 0.111005 #> Expected 65.3302 62.5603 71.0313 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":null,"dir":"Reference","previous_headings":"","what":"Species tolerances and sample heterogeneities — tolerance","title":"Species tolerances and sample heterogeneities — tolerance","text":"Species tolerances sample heterogeneities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"tolerance(x, ...) # S3 method for cca tolerance(x, choices = 1:2, which = c(\"species\",\"sites\"), scaling = \"species\", useN2 = TRUE, hill = FALSE, ...) # S3 method for decorana tolerance(x, data, choices = 1:4, which = c(\"sites\", \"species\"), useN2 = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species tolerances and sample heterogeneities — tolerance","text":"Function compute species tolerances site heterogeneity measures unimodal ordinations (CCA & CA). Implements Eq 6.47 6.48 Canoco 4.5 Reference Manual (pages 178--179).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species tolerances and sample heterogeneities — tolerance","text":"Matrix tolerances/heterogeneities additional attributes: , scaling, N2, latter NA useN2 = FALSE N2 estimated.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species tolerances and sample heterogeneities — tolerance","text":"Gavin L. Simpson Jari Oksanen (decorana method).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species tolerances and sample heterogeneities — tolerance","text":"x object class \"cca\". choices numeric; ordination axes compute tolerances heterogeneities . Defaults axes 1 2. character; one \"species\" \"sites\", indicating whether species tolerances sample heterogeneities respectively computed. scaling character numeric; ordination scaling use. See scores.cca details. hill logical; scaling character, control whether Hill's scaling used (C)CA respectively. See scores.cca details. useN2 logical; bias tolerances / heterogeneities reduced via scaling Hill's N2? data Original input data used decorana. missing, function tries get data used decorana call. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ ., data = dune.env) ## defaults to species tolerances tolerance(mod) #> #> Species Tolerance #> #> Scaling: 2 #> #> CCA1 CCA2 #> Achimill 0.32968099 0.9241988 #> Agrostol 0.93670069 0.9238455 #> Airaprae 1.04694096 0.5889849 #> Alopgeni 0.72227472 0.3760138 #> Anthodor 1.00596787 0.8338212 #> Bellpere 0.32891011 0.9962790 #> Bromhord 0.27740999 0.6236199 #> Chenalbu 0.00000000 0.0000000 #> Cirsarve 0.00000000 0.0000000 #> Comapalu 0.47185632 0.8029414 #> Eleopalu 0.50344134 0.9384960 #> Elymrepe 0.35119963 0.5642491 #> Empenigr 0.00000000 0.0000000 #> Hyporadi 1.05840696 0.7523003 #> Juncarti 0.78397702 1.0686743 #> Juncbufo 0.69275956 0.6180830 #> Lolipere 0.51006235 0.8278177 #> Planlanc 0.36040676 0.6962294 #> Poaprat 0.58184277 0.9547104 #> Poatriv 0.78695928 0.7433503 #> Ranuflam 0.56576326 1.1725628 #> Rumeacet 0.58715663 0.8751491 #> Sagiproc 0.70922180 1.1153129 #> Salirepe 0.98530179 0.1077917 #> Scorautu 1.04355761 1.0724439 #> Trifprat 0.03045846 0.3651949 #> Trifrepe 1.21543364 0.9115613 #> Vicilath 0.24853962 0.6194084 #> Bracruta 1.03787313 1.0958331 #> Callcusp 0.57882025 1.0418623 #> ## sample heterogeneities for CCA axes 1:6 tolerance(mod, which = \"sites\", choices = 1:6) #> #> Sample Heterogeneity #> #> Scaling: 2 #> #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 1 0.2350112 0.8611530 1.7964571 0.4445499 2.4235732 0.5496289 #> 2 0.7100754 0.4136311 0.8151643 0.6311751 1.0467901 0.2514646 #> 3 0.5076492 0.7279717 0.8306874 0.5590739 0.3904998 0.9162012 #> 4 0.5955037 0.6901907 0.7931255 0.4873638 0.3966068 0.8700581 #> 5 0.6001048 0.5614830 1.1481560 0.3569604 0.4423909 1.9420043 #> 6 0.7272637 0.6867342 1.6068628 0.7778498 0.9187843 0.4938865 #> 7 0.6478967 0.4993262 0.7207318 0.3817131 0.4130713 0.7228173 #> 8 0.8563491 0.5498552 0.4217718 0.3370226 0.3013276 0.9535190 #> 9 0.5599722 0.7399384 0.4170304 1.0535541 1.4612437 0.7626183 #> 10 0.5210280 0.5806978 0.5856634 0.4174860 1.8559344 0.8890262 #> 11 0.4489323 0.6016877 0.3317371 1.8780211 1.2965939 2.1953737 #> 12 0.4948094 1.1084494 0.5226746 1.5064446 0.5703077 1.1561020 #> 13 0.6998985 0.8859365 0.4215474 0.8582272 0.5673698 0.5186678 #> 14 1.5925779 0.6747926 0.8927360 1.6798300 0.3480218 0.1575892 #> 15 1.0107648 0.5294221 1.0975629 1.7632888 0.2240900 0.3727240 #> 16 0.8031479 0.6058313 0.4871527 0.4227451 0.5341256 0.6990815 #> 17 0.5936276 1.5142792 0.5137979 1.0224938 1.7931775 0.6261853 #> 18 0.5689409 1.4067575 0.6398557 0.4983399 0.4364791 0.6590394 #> 19 1.1330387 0.9816332 1.1242398 0.7238920 0.5577662 0.7036044 #> 20 0.6737757 1.4458326 1.4380928 1.0959027 0.4142423 0.5332460 #> ## average should be 1 with scaling = \"sites\", hill = TRUE tol <- tolerance(mod, which = \"sites\", scaling = \"sites\", hill = TRUE, choices = 1:4) colMeans(tol) #> CCA1 CCA2 CCA3 CCA4 #> 1.059199 1.048823 1.000551 1.077612 apply(tol, 2, sd) #> CCA1 CCA2 CCA3 CCA4 #> 0.3174462 0.2793521 0.3714540 0.2681931 ## Rescaling tries to set all tolerances to 1 tol <- tolerance(decorana(dune)) colMeans(tol) #> DCA1 DCA2 DCA3 DCA4 #> 0.9817657 0.9249544 0.9444811 0.9821666 apply(tol, 2, sd) #> DCA1 DCA2 DCA3 DCA4 #> 0.1977777 0.3204058 0.2646872 0.1210543"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":null,"dir":"Reference","previous_headings":"","what":"Functional Diversity and Community Distances from Species Trees — treedive","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Functional diversity defined total branch length trait dendrogram connecting species, excluding unnecessary root segments tree (Petchey Gaston 2006). Tree distance increase total branch length combining two sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"treedive(comm, tree, match.force = TRUE, verbose = TRUE) treeheight(tree) treedist(x, tree, relative = TRUE, match.force = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"comm, x Community data frame matrix. tree dendrogram treedive must species (columns). match.force Force matching column names data (comm, x) labels tree. FALSE, matching happens dimensions differ (warning message). order data must match order tree matching names done. verbose Print diagnostic messages warnings. relative Use distances relative height combined tree. ... arguments passed functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Function treeheight finds sum lengths connecting segments dendrogram produced hclust, dendrogram can coerced correct type using .hclust. applied clustering species traits, measure functional diversity (Petchey Gaston 2002, 2006), applied phylogenetic trees phylogenetic diversity. Function treedive finds treeheight site (row) community matrix. function uses subset dendrogram species occur site, excludes tree root needed connect species (Petchey Gaston 2006). subset dendrogram found first calculating cophenetic distances input dendrogram, reconstructing dendrogram subset cophenetic distance matrix species occurring site. Diversity 0 one species, NA empty communities. Function treedist finds dissimilarities among trees. Pairwise dissimilarity two trees found combining species common tree seeing much tree height shared much unique. relative = FALSE dissimilarity defined \\(2 (\\cup B) - - B\\), \\(\\) \\(B\\) heights component trees \\(\\cup B\\) height combined tree. relative = TRUE dissimilarity \\((2(\\cup B)--B)/(\\cup B)\\). Although latter formula similar Jaccard dissimilarity (see vegdist, designdist), range \\(0 \\ldots 1\\), since combined tree can add new root. two zero-height trees combined tree zero height, relative index attains maximum value \\(2\\). dissimilarity zero combined zero-height tree. functions need dendrogram species traits phylogenies input. species traits contain factor ordered factor variables, recommended use Gower distances mixed data (function daisy package cluster), usually recommended clustering method UPGMA (method = \"average\" function hclust) (Podani Schmera 2006). Phylogenetic trees can changed dendrograms using function .hclust.phylo ape package. possible analyse non-randomness tree diversity using oecosimu. needs specifying adequate Null model, results change choice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"vector diversity values single tree height, dissimilarity structure inherits dist can used similarly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Lozupone, C. Knight, R. 2005. UniFrac: new phylogenetic method comparing microbial communities. Applied Environmental Microbiology 71, 8228--8235. Petchey, O.L. Gaston, K.J. 2002. Functional diversity (FD), species richness community composition. Ecology Letters 5, 402--411. Petchey, O.L. Gaston, K.J. 2006. Functional diversity: back basics looking forward. Ecology Letters 9, 741--758. Podani J. Schmera, D. 2006. dendrogram-based methods functional diversity. Oikos 115, 179--185.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"## There is no data set on species properties yet, and we demonstrate ## the methods using phylogenetic trees data(dune) data(dune.phylodis) cl <- hclust(dune.phylodis) treedive(dune, cl) #> forced matching of 'tree' labels and 'comm' names #> 1 2 3 4 5 6 7 8 #> 384.0913 568.8791 1172.9455 1327.9317 1426.9067 1391.1628 1479.5062 1523.0792 #> 9 10 11 12 13 14 15 16 #> 1460.0423 1316.4832 1366.9960 1423.5582 895.1120 1457.2705 1505.9501 1187.5165 #> 17 18 19 20 #> 517.6920 1394.5162 1470.4671 1439.5571 ## Significance test using Null model communities. ## The current choice fixes numbers of species and picks species ## proportionally to their overall frequency oecosimu(dune, treedive, \"r1\", tree = cl, verbose = FALSE) #> Warning: nullmodel transformed 'comm' to binary data #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = treedive, method = \"r1\", tree = #> cl, verbose = FALSE) #> #> nullmodel method ‘r1’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> 1 384.09 -1.238698 773.72 383.47 628.63 1237.7 0.11 #> 2 568.88 -2.353877 1222.61 663.93 1337.50 1561.4 0.01 ** #> 3 1172.95 -0.136353 1210.76 679.53 1327.94 1581.5 0.63 #> 4 1327.93 -0.418624 1427.25 886.63 1505.15 1733.1 0.45 #> 5 1426.91 -0.322462 1496.97 926.98 1559.25 1734.4 0.43 #> 6 1391.16 0.254066 1326.70 753.19 1408.30 1634.7 0.87 #> 7 1479.51 0.280433 1407.90 880.79 1476.23 1735.1 0.99 #> 8 1523.08 0.567886 1389.63 792.74 1428.01 1655.3 0.65 #> 9 1460.04 0.065245 1448.21 899.67 1472.92 1654.4 0.87 #> 10 1316.48 -0.192836 1362.43 708.60 1441.63 1608.0 0.47 #> 11 1367.00 0.768160 1150.05 621.65 1266.73 1486.4 0.43 #> 12 1423.56 1.101039 1088.13 622.24 1208.95 1474.2 0.25 #> 13 895.11 -0.972053 1188.10 656.34 1325.53 1558.5 0.61 #> 14 1457.27 1.477151 980.91 491.11 1101.65 1459.3 0.07 . #> 15 1505.95 1.532058 1052.35 575.91 1171.88 1460.7 0.03 * #> 16 1187.52 0.306608 1089.58 543.75 1238.26 1490.7 0.89 #> 17 517.69 -1.446211 959.50 504.26 1101.16 1367.8 0.09 . #> 18 1394.52 0.961778 1099.50 618.22 1249.49 1493.8 0.33 #> 19 1470.47 1.134795 1121.18 564.70 1240.26 1562.8 0.15 #> 20 1439.56 1.160566 1104.59 629.08 1244.20 1467.4 0.17 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Phylogenetically ordered community table dtree <- treedist(dune, cl) tabasco(dune, hclust(dtree), cl) ## Use tree distances in distance-based RDA dbrda(dtree ~ 1) #> Call: dbrda(formula = dtree ~ 1) #> #> Inertia Rank RealDims #> Total 2.183 #> Unconstrained 2.183 19 10 #> Inertia is squared Treedist distance #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.1971 0.4546 0.2967 0.1346 0.1067 0.0912 0.0391 0.0190 #> (Showing 8 of 19 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":null,"dir":"Reference","previous_headings":"","what":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis find Tsallis diversities scale corresponding evenness measures. Function tsallisaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"tsallis(x, scales = seq(0, 2, 0.2), norm = FALSE, hill = FALSE) tsallisaccum(x, scales = seq(0, 2, 0.2), permutations = 100, raw = FALSE, subset, ...) # S3 method for tsallisaccum persp(x, theta = 220, phi = 15, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"x Community data matrix plotting object. scales Scales Tsallis diversity. norm Logical, TRUE diversity values normalized maximum (diversity value equiprobability conditions). hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. subset logical expression indicating sites (rows) keep: missing values taken FALSE. theta, phi angles defining viewing direction. theta gives azimuthal direction phi colatitude. col Colours used surface. zlim Limits vertical axis. ... arguments passed tsallis graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis diversity (also equivalent Patil Taillie diversity) one-parametric generalised entropy function, defined : $$H_q = \\frac{1}{q-1} (1-\\sum_{=1}^S p_i^q)$$ \\(q\\) scale parameter, \\(S\\) number species sample (Tsallis 1988, Tothmeresz 1995). diversity concave \\(q>0\\), non-additive (Keylock 2005). \\(q=0\\) gives number species minus one, \\(q\\) tends 1 gives Shannon diversity, \\(q=2\\) gives Simpson index (see function diversity). norm = TRUE, tsallis gives values normalized maximum: $$H_q(max) = \\frac{S^{1-q}-1}{1-q}$$ \\(S\\) number species. \\(q\\) tends 1, maximum defined \\(ln(S)\\). hill = TRUE, tsallis gives Hill numbers (numbers equivalents, see Jost 2007): $$D_q = (1-(q-1) H)^{1/(1-q)}$$ Details plotting methods accumulating values can found help pages functions renyi renyiaccum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis returns data frame selected indices. Function tsallisaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487. Tothmeresz, B. (1995) Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290. Patil, G. P. Taillie, C. (1982) Diversity concept measurement. J. . Stat. Ass. 77, 548--567. Keylock, C. J. (2005) Simpson diversity Shannon-Wiener index special cases generalized entropy. Oikos 109, 203--207. Jost, L (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Péter Sólymos, solymos@ualberta.ca, based code Roeland Kindt Jari Oksanen written renyi","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) x1 <- tsallis(BCI[i,]) x1 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 #> 8 87 39.98200 19.77567 10.60391 6.181934 3.908381 2.659907 1.928610 1.473408 #> 10 93 41.50369 20.10059 10.63988 6.164157 3.889803 2.648176 1.922207 1.470121 #> 5 100 44.20598 21.17686 11.08126 6.350040 3.969940 2.683412 1.937940 1.477217 #> 17 92 40.88788 19.62643 10.29183 5.926929 3.736897 2.553271 1.864739 1.435868 #> 19 108 47.27018 22.32602 11.49839 6.492802 4.013094 2.692466 1.936696 1.473828 #> 13 92 41.98171 20.60707 10.96297 6.342501 3.982373 2.694847 1.945425 1.481616 #> 46 85 38.71044 19.06049 10.22455 5.987209 3.810489 2.611431 1.904913 1.461976 #> 16 92 41.45601 20.19256 10.71573 6.210870 3.916821 2.663870 1.931567 1.475861 #> 36 91 40.76595 19.77916 10.48193 6.081803 3.846109 2.625076 1.910150 1.463936 #> 7 81 37.47840 18.73414 10.17269 6.004908 3.836811 2.631757 1.918091 1.469869 #> 24 94 42.55654 20.75558 10.99284 6.343926 3.979427 2.692702 1.944478 1.481404 #> 22 90 40.16368 19.41416 10.25630 5.939677 3.755413 2.566788 1.872586 1.439722 #> 1.8 2 #> 8 1.173986 0.9671998 #> 10 1.172341 0.9663808 #> 5 1.175553 0.9678267 #> 17 1.152122 0.9545126 #> 19 1.172481 0.9655820 #> 13 1.178033 0.9692075 #> 46 1.168556 0.9646728 #> 16 1.175935 0.9686598 #> 36 1.169232 0.9648567 #> 7 1.173083 0.9672014 #> 24 1.178152 0.9694268 #> 22 1.153640 0.9548316 diversity(BCI[i,],\"simpson\") == x1[[\"2\"]] #> 8 10 5 17 19 13 46 16 36 7 24 22 #> TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE plot(x1) x2 <- tsallis(BCI[i,],norm=TRUE) x2 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 #> 8 1 0.9154260 0.8674322 0.8491531 0.8535754 0.8729254 0.8992501 0.9258854 #> 10 1 0.9001004 0.8450759 0.8255446 0.8324355 0.8561634 0.8872568 0.9180261 #> 5 1 0.9038026 0.8502581 0.8308757 0.8372445 0.8602030 0.8904867 0.9204832 #> 17 1 0.8945733 0.8308281 0.8026305 0.8032747 0.8244489 0.8566981 0.8913214 #> 19 1 0.9079031 0.8537683 0.8315799 0.8347890 0.8554245 0.8846675 0.9147428 #> 13 1 0.9185048 0.8723405 0.8549707 0.8595970 0.8786069 0.9042012 0.9298883 #> 46 1 0.9032439 0.8485357 0.8278693 0.8331492 0.8554539 0.8856795 0.9162035 #> 16 1 0.9070033 0.8547937 0.8356888 0.8417570 0.8641446 0.8938075 0.9232641 #> 36 1 0.8998669 0.8431295 0.8216927 0.8272622 0.8505725 0.8820846 0.9137994 #> 7 1 0.9094899 0.8600339 0.8427936 0.8492653 0.8706729 0.8985570 0.9261475 #> 24 1 0.9149445 0.8667059 0.8486326 0.8536824 0.8738548 0.9008891 0.9279024 #> 22 1 0.8945746 0.8334045 0.8082215 0.8109045 0.8325271 0.8637842 0.8965990 #> 1.6 1.8 2 #> 8 0.9486740 0.9660687 0.9783170 #> 10 0.9438798 0.9632975 0.9767719 #> 5 0.9456409 0.9644767 0.9775050 #> 17 0.9223038 0.9469046 0.9648877 #> 19 0.9406572 0.9605041 0.9745226 #> 13 0.9516896 0.9681999 0.9797424 #> 46 0.9422698 0.9621112 0.9760219 #> 16 0.9479928 0.9664763 0.9791887 #> 36 0.9407648 0.9611954 0.9754595 #> 7 0.9493990 0.9669335 0.9791421 #> 24 0.9507028 0.9678509 0.9797399 #> 22 0.9256369 0.9486075 0.9654409 plot(x2) mod1 <- tsallisaccum(BCI[i,]) plot(mod1, as.table=TRUE, col = c(1, 2, 2)) persp(mod1) mod2 <- tsallisaccum(BCI[i,], norm=TRUE) persp(mod2,theta=100,phi=30)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and environment in lichen pastures — varespec","title":"Vegetation and environment in lichen pastures — varespec","text":"varespec data frame 24 rows 44 columns. Columns estimated cover values 44 species. variable names formed scientific names, self explanatory anybody familiar vegetation type. varechem data frame 24 rows 14 columns, giving soil characteristics sites varespec data frame. chemical measurements obvious names. Baresoil gives estimated cover bare soil, Humdepth thickness humus layer.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varechem) data(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Vegetation and environment in lichen pastures — varespec","text":"Väre, H., Ohtonen, R. Oksanen, J. (1995) Effects reindeer grazing understorey vegetation dry Pinus sylvestris forests. Journal Vegetation Science 6, 523--530.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varespec) data(varechem)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":null,"dir":"Reference","previous_headings":"","what":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"function partitions variation community data community dissimilarities respect two, three, four explanatory tables, using adjusted \\(R^2\\) redundancy analysis ordination (RDA) distance-based redundancy analysis. response single vector, partitioning partial regression. Collinear variables explanatory tables removed prior partitioning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"varpart(Y, X, ..., data, chisquare = FALSE, transfo, scale = FALSE, add = FALSE, sqrt.dist = FALSE, permutations) # S3 method for varpart summary(object, ...) showvarparts(parts, labels, bg = NULL, alpha = 63, Xnames, id.size = 1.2, ...) # S3 method for varpart234 plot(x, cutoff = 0, digits = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Y Data frame matrix containing response data table dissimilarity structure inheriting dist. community ecology, table often site--species table dissimilarity object. X Two four explanatory models, variables tables. can defined three alternative ways: (1) one-sided model formulae beginning ~ defining model, (2) name single numeric factor variable, (3) name matrix numeric data frame numeric factor variables. model formulae can factors, interaction terms transformations variables. names variables model formula found data frame given data argument, found , user environment. Single variables, data frames matrices found user environment. entries till next argument (data transfo) interpreted explanatory models, names extra arguments abbreviated omitted. ... parameters passed functions. NB, arguments dots abbreviated must spelt completely. data data frame variables used formulae X. chisquare Partition Chi-square inertia Correspondence Analysis (cca). transfo Transformation Y (community data) using decostand. alternatives decostand can used, preserving Euclidean metric include \"hellinger\", \"chi.square\", \"total\", \"norm\". Ignored Y dissimilarities. scale columns Y standardized unit variance. Ignored Y dissimilarities. add Add constant non-diagonal values euclidify dissimilarities (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. argument effect Y dissimilarities. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. NB., argument name abbreviated. argument effect Y dissimilarities. permutations chisquare = TRUE, adjusted \\(R^2\\) estimated permutations, paramater can list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parts Number explanatory tables (circles) displayed. labels Labels used displayed fractions. Default use letters printed output. bg Fill colours circles ellipses. alpha Transparency fill colour. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. Xnames Names sources variation. Default names X1, X2, X3 X4. Xnames=NA, Xnames=NULL Xnames=\"\" produce names. names can changed names. often best use short names. id.size numerical value giving character expansion factor names circles ellipses. x, object varpart result. cutoff values cutoff displayed. digits number significant digits; number decimal places least one higher.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"functions partition variation Y components accounted two four explanatory tables combined effects. Y multicolumn data frame matrix, partitioning based redundancy analysis (RDA, see rda) constrained correspondence analysis chisquare = TRUE (CCA, see cca). Y single variable, partitioning based linear regression. Y dissimilarities, decomposition based distance-based redundancy analysis (db-RDA, see dbrda) following McArdle & Anderson (2001). input dissimilarities must compatible results dist. Vegan functions vegdist, designdist, raupcrick betadiver produce objects, many dissimilarity functions R packages. Partitioning made squared dissimilarities analogously using variance rectangular data -- unless sqrt.dist = TRUE specified. function primarily uses adjusted \\(R^2\\) assess partitions explained explanatory tables combinations (see RsquareAdj), unbiased method (Peres-Neto et al., 2006). raw \\(R^2\\) basic fractions also displayed, biased estimates variation explained explanatory table. correspondence analysis (chisquare = TRUE), adjusted \\(R^2\\) found permutation vary repeated analyses. identifiable fractions designated lower case alphabets. meaning symbols can found separate document (use browseVignettes(\"vegan\")), can displayed graphically using function showvarparts. fraction testable can directly expressed RDA db-RDA model. cases printed output also displays corresponding RDA model using notation explanatory tables | conditions (partialled ; see rda details). Although single fractions can testable, mean fractions simultaneously can tested, since number testable fractions higher number estimated models. non-testable components found differences testable components. testable components permutation variance correspondence analysis (chisquare = TRUE), non-testable components even higher variance. abridged explanation alphabetic symbols individual fractions follows, computational details checked vignette (readable browseVignettes(\"vegan\")) source code. two explanatory tables, fractions explained uniquely two tables [] [b], joint effect [c]. three explanatory tables, fractions explained uniquely three tables [] [c], joint fractions two tables [d] [f], joint fraction three tables [g]. four explanatory tables, fractions explained uniquely four tables [] [d], joint fractions two tables [e] [j], joint fractions three variables [k] [n], joint fraction four tables [o]. summary give overview unique overall contribution group variables. overall contribution (labelled “Contributed”) consists unique contribution variable equal shares fraction variable contributes. summary tabulates fraction divided variables, contributed component sum divided fractions. summary based idea Lai et al. (2022), similar output rdacca.hp package. plot function displays Venn diagram labels intersection (individual fraction) adjusted R squared higher cutoff. helper function showvarpart displays fraction labels. circles ellipses labelled short default names names defined user argument Xnames. Longer explanatory file names can written varpart output plot follows: use option Xnames=NA, add new names using text function. bit fiddling coordinates (see locator) character size allow users place names reasonably short lengths varpart plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Function varpart returns object class \"varpart\" items scale transfo (can missing) hold information standardizations, tables contains names explanatory tables, call function call. function varpart calls function varpart2, varpart3 varpart4 return object class \"varpart234\" saves result item part. items object : SS.Y Sum squares matrix Y. n Number observations (rows). nsets Number explanatory tables bigwarning Warnings collinearity. fract Basic fractions estimated constrained models. indfract Individual fractions possible subsections Venn diagram (see showvarparts). contr1 Fractions can found conditioning single explanatory table models three four explanatory tables. contr2 Fractions can found conditioning two explanatory tables models four explanatory tables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"fraction-data-frames","dir":"Reference","previous_headings":"","what":"Fraction Data Frames","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Items fract, indfract, contr1 contr2 data frames items: Df: Degrees freedom numerator \\(F\\)-statistic fraction. R.square: Raw \\(R^2\\). calculated fract NA items. Adj.R.square: Adjusted \\(R^2\\). Testable: fraction can expressed (partial) RDA model, directly Testable, field TRUE. case fraction label also gives specification testable RDA model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"() References variation partitioning Borcard, D., P. Legendre & P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045--1055. Lai J., Y. Zou, J. Zhang & P. Peres-Neto. 2022. Generalizing hierarchical variation partitioning multiple regression canonical analysis using rdacca.hp R package. Methods Ecology Evolution, 13: 782--788. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. (b) Reference transformations species data Legendre, P. E. D. Gallagher. 2001. Ecologically meaningful transformations ordination species data. Oecologia 129: 271--280. (c) Reference adjustment bimultivariate redundancy statistic Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87: 2614--2625. (d) References partitioning dissimilarities Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290-297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal, Canada. developed Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"can use command browseVignettes(\"vegan\") display document presents Venn diagrams showing fraction names partitioning variation Y respect 2, 3, 4 tables explanatory variables, well equations used variation partitioning. functions frequently give negative estimates variation. Adjusted \\(R^2\\) can negative fraction; unadjusted \\(R^2\\) testable fractions variances non-negative. Non-testable fractions found directly, subtracting different models, subtraction results can negative. fractions orthogonal, linearly independent, complicated nonlinear dependencies can cause negative non-testable fractions. fraction can negative non-Euclidean dissimilarities underlying db-RDA model can yield negative eigenvalues (see dbrda). negative eigenvalues underlying analysis can avoided arguments sqrt.dist add similar effect dbrda: square roots several dissimilarities negative eigenvalues, negative eigenvalues produced Lingoes Cailliez adjustment, effect add random variation dissimilarities. simplified, fast version RDA, CCA adn dbRDA used (functions simpleRDA2, simpleCCA simpleDBRDA). actual calculations done functions varpart2 varpart4, intended called directly user.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) # Two explanatory data frames -- Hellinger-transform Y mod <- varpart(mite, mite.env, mite.pcnm, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = mite.env, mite.pcnm, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.52650 0.43670 TRUE #> [b+c] = X2 22 0.62300 0.44653 TRUE #> [a+b+c] = X1+X2 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09141 TRUE #> [b] = X2|X1 22 0.10124 TRUE #> [c] 0 0.34530 FALSE #> [d] = Residuals 0.46206 FALSE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0914 0.264 mite.env #> X2 0.1012 0.274 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 #> [a] 0.0914 #> [b] 0.1012 #> [c] 0.1726 0.1726 ## Use fill colours showvarparts(2, bg = c(\"hotpink\",\"skyblue\")) plot(mod, bg = c(\"hotpink\",\"skyblue\")) ## Test fraction [a] using partial RDA, '~ .' in formula tells to use ## all variables of data mite.env. aFrac <- rda(decostand(mite, \"hel\"), mite.env, mite.pcnm) anova(aFrac) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(X = decostand(mite, \"hel\"), Y = mite.env, Z = mite.pcnm) #> Df Variance F Pr(>F) #> Model 11 0.053592 1.8453 0.001 *** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## RsquareAdj gives the same result as component [a] of varpart RsquareAdj(aFrac) #> $r.squared #> [1] 0.1359251 #> #> $adj.r.squared #> [1] 0.09140797 #> ## Partition Bray-Curtis dissimilarities varpart(vegdist(mite), mite.env, mite.pcnm) #> #> Partition of squared Bray distance in dbRDA #> #> Call: varpart(Y = vegdist(mite), X = mite.env, mite.pcnm) #> #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 14.696 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.50512 0.41127 TRUE #> [b+c] = X2 22 0.60144 0.41489 TRUE #> [a+b+c] = X1+X2 33 0.74631 0.51375 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09887 TRUE #> [b] = X2|X1 22 0.10249 TRUE #> [c] 0 0.31240 FALSE #> [d] = Residuals 0.48625 FALSE #> --- #> Use function ‘dbrda’ to test significance of fractions of interest ## Three explanatory tables with formula interface mod <- varpart(mite, ~ SubsDens + WatrCont, ~ Substrate + Shrub + Topo, mite.pcnm, data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm, data = mite.env, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm #> #> No. of explanatory tables: 3 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [a+d+f+g] = X1 2 0.32677 0.30667 TRUE #> [b+d+e+g] = X2 9 0.40395 0.31454 TRUE #> [c+e+f+g] = X3 22 0.62300 0.44653 TRUE #> [a+b+d+e+f+g] = X1+X2 11 0.52650 0.43670 TRUE #> [a+c+d+e+f+g] = X1+X3 24 0.67372 0.49970 TRUE #> [b+c+d+e+f+g] = X2+X3 31 0.72400 0.49884 TRUE #> [a+b+c+d+e+f+g] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3 2 0.03910 TRUE #> [b] = X2 | X1+X3 9 0.03824 TRUE #> [c] = X3 | X1+X2 22 0.10124 TRUE #> [d] 0 0.01407 FALSE #> [e] 0 0.09179 FALSE #> [f] 0 0.08306 FALSE #> [g] 0 0.17045 FALSE #> [h] = Residuals 0.46206 FALSE #> Controlling 1 table X #> [a+d] = X1 | X3 2 0.05317 TRUE #> [a+f] = X1 | X2 2 0.12216 TRUE #> [b+d] = X2 | X3 9 0.05231 TRUE #> [b+e] = X2 | X1 9 0.13003 TRUE #> [c+e] = X3 | X1 22 0.19303 TRUE #> [c+f] = X3 | X2 22 0.18429 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.144 ~SubsDens + WatrCont #> X2 0.0382 0.148 ~Substrate + Shrub + Topo #> X3 0.1012 0.245 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 X3 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10124 #> [d] 0.00703 0.00703 #> [e] 0.04590 0.04590 #> [f] 0.04153 0.04153 #> [g] 0.05682 0.05682 0.05682 showvarparts(3, bg=2:4) plot(mod, bg=2:4) ## Use RDA to test fraction [a] ## Matrix can be an argument in formula rda.result <- rda(decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) anova(rda.result) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) #> Df Variance F Pr(>F) #> Model 2 0.013771 2.6079 0.005 ** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Four explanatory tables mod <- varpart(mite, ~ SubsDens + WatrCont, ~Substrate + Shrub + Topo, mite.pcnm[,1:11], mite.pcnm[,12:22], data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm[, 1:11], mite.pcnm[, 12:22], data = mite.env, transfo = #> \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm[, 1:11] #> X4: mite.pcnm[, 12:22] #> #> No. of explanatory tables: 4 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [aeghklno] = X1 2 0.32677 0.30667 TRUE #> [befiklmo] = X2 9 0.40395 0.31454 TRUE #> [cfgjlmno] = X3 11 0.53231 0.44361 TRUE #> [dhijkmno] = X4 11 0.09069 -0.08176 TRUE #> [abefghiklmno] = X1+X2 11 0.52650 0.43670 TRUE #> [acefghjklmno] = X1+X3 13 0.59150 0.49667 TRUE #> [adeghijklmno] = X1+X4 13 0.40374 0.26533 TRUE #> [bcefgijklmno] = X2+X3 20 0.63650 0.48813 TRUE #> [bdefhijklmno] = X2+X4 20 0.53338 0.34292 TRUE #> [cdfghijklmno] = X3+X4 22 0.62300 0.44653 TRUE #> [abcefghijklmno] = X1+X2+X3 22 0.67947 0.52944 TRUE #> [abdefghijklmno] = X1+X2+X4 22 0.61553 0.43557 TRUE #> [acdefghijklmno] = X1+X3+X4 24 0.67372 0.49970 TRUE #> [bcdefghijklmno] = X2+X3+X4 31 0.72400 0.49884 TRUE #> [abcdefghijklmno] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3+X4 2 0.03910 TRUE #> [b] = X2 | X1+X3+X4 9 0.03824 TRUE #> [c] = X3 | X1+X2+X4 11 0.10237 TRUE #> [d] = X4 | X1+X2+X3 11 0.00850 TRUE #> [e] 0 0.01407 FALSE #> [f] 0 0.13200 FALSE #> [g] 0 0.05355 FALSE #> [h] 0 0.00220 FALSE #> [i] 0 -0.00547 FALSE #> [j] 0 -0.00963 FALSE #> [k] 0 -0.00231 FALSE #> [l] 0 0.24037 FALSE #> [m] 0 -0.03474 FALSE #> [n] 0 0.02730 FALSE #> [o] 0 -0.06761 FALSE #> [p] = Residuals 0 0.46206 FALSE #> Controlling 2 tables X #> [ae] = X1 | X3+X4 2 0.05317 TRUE #> [ag] = X1 | X2+X4 2 0.09265 TRUE #> [ah] = X1 | X2+X3 2 0.04131 TRUE #> [be] = X2 | X3+X4 9 0.05231 TRUE #> [bf] = X2 | X1+X4 9 0.17024 TRUE #> [bi] = X2 | X1+X3 9 0.03277 TRUE #> [cf] = X3 | X1+X4 11 0.23437 TRUE #> [cg] = X3 | X2+X4 11 0.15592 TRUE #> [cj] = X3 | X1+X2 11 0.09274 TRUE #> [dh] = X4 | X2+X3 11 0.01071 TRUE #> [di] = X4 | X1+X3 11 0.00303 TRUE #> [dj] = X4 | X1+X2 11 -0.00113 TRUE #> Controlling 1 table X #> [aghn] = X1 | X2 2 0.12216 TRUE #> [aehk] = X1 | X3 2 0.05306 TRUE #> [aegl] = X1 | X4 2 0.34709 TRUE #> [bfim] = X2 | X1 9 0.13003 TRUE #> [beik] = X2 | X3 9 0.04452 TRUE #> [befl] = X2 | X4 9 0.42468 TRUE #> [cfjm] = X3 | X1 11 0.19000 TRUE #> [cgjn] = X3 | X2 11 0.17359 TRUE #> [cfgl] = X3 | X4 11 0.52830 TRUE #> [dijm] = X4 | X1 11 -0.04134 TRUE #> [dhjn] = X4 | X2 11 0.02837 TRUE #> [dhik] = X4 | X3 11 0.00292 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.1456 ~SubsDens + WatrCont #> X2 0.0382 0.1594 ~Substrate + Shrub + Topo #> X3 0.1024 0.2511 mite.pcnm[, 1:11] #> X4 0.0085 -0.0181 mite.pcnm[, 12:22] #> #> Contributions of fractions to sets: #> #> X1 X2 X3 X4 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10237 #> [d] 0.00850 #> [e] 0.00703 0.00703 #> [f] 0.06600 0.06600 #> [g] 0.02678 0.02678 #> [h] 0.00110 0.00110 #> [i] -0.00274 -0.00274 #> [j] -0.00482 -0.00482 #> [k] -0.00077 -0.00077 -0.00077 #> [l] 0.08012 0.08012 0.08012 #> [m] -0.01158 -0.01158 -0.01158 #> [n] 0.00910 0.00910 0.00910 #> [o] -0.01690 -0.01690 -0.01690 -0.01690 plot(mod, bg=2:5) ## Show values for all partitions by putting 'cutoff' low enough: plot(mod, cutoff = -Inf, cex = 0.7, bg=2:5)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":null,"dir":"Reference","previous_headings":"","what":"Defunct Functions in Package vegan — vegan-defunct","title":"Defunct Functions in Package vegan — vegan-defunct","text":"functions variables listed longer part vegan longer needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Defunct Functions in Package vegan — vegan-defunct","text":"","code":"## defunct in vegan 2.7-0 (development 2.6-7) adonis(formula, data, permutations = 999, method = \"bray\", strata = NULL, contr.unordered = \"contr.sum\", contr.ordered = \"contr.poly\", parallel = getOption(\"mc.cores\"), ...) ## defunct in vegan 2.6-0 as.mlm(x) humpfit(mass, spno, family = poisson, start) vegandocs(doc = c(\"NEWS\", \"ONEWS\", \"FAQ-vegan\", \"intro-vegan\", \"diversity-vegan\", \"decision-vegan\", \"partitioning\", \"permutations\")) ## defunct in vegan 2.5-0 commsimulator(x, method, thin=1) ## defunct in vegan 2.4-0 # S3 method for adonis density(x, ...) # S3 method for vegandensity plot(x, main = NULL, xlab = NULL, ylab = \"Density\", type = \"l\", zero.line = TRUE, obs.line = TRUE, ...) # S3 method for adonis densityplot(x, data, xlab = \"Null\", ...) ## defunct in vegan 2.2-0 metaMDSrotate(object, vec, na.rm = FALSE, ...) ## defunct in vegan 2.0-0 getNumObs(object, ...) permuted.index2(n, control = permControl())"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Defunct Functions in Package vegan — vegan-defunct","text":"adonis2 replaces adonis extended functionality completely new internal design. shared arguments adonis similar adonis2, arguments contr.unordered contr.ordered can set contrasts within adonis. .mlm function replaced set functions can find statistics directly ordination result object: see hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca, vcov.cca. Function humpfit transferred natto package still available https://github.com/jarioksa/natto/. R functions news used read vegan NEWS (news(package = \"vegan\")), browseVignettes better tool reading vignettes vegandocs. Function commsimulator replaced make.commsim defines Null models, functions nullmodel simulate.nullmodel check input data generate Null model communities. deprecated density densityplot methods replaced similar methods permustats. permustats offers powerful analysis tools permutations, including summary.permustats giving \\(z\\) values (.k.. standardized effect sizes, SES), Q-Q plots (qqnorm.permustats, qqmath.permustats). Function metaMDSrotate replaced MDSrotate can handle monoMDS results addition metaMDS. permutation functions moved permute package, documented . permute package replaces permuted.index permuted.index2 shuffle getNumObs specific nobs-methods.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions in vegan package — vegan-deprecated","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"functions provided compatibility older versions vegan , may defunct soon next release.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"","code":"## moved to vegan3d package: install from CRAN orditkplot(...) ## use toCoda instead as.mcmc.oecosimu(x) as.mcmc.permat(x)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"x object transformed. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"orditkplot moved vegan3d (version 1.3-0). Install package CRAN use old way. .mcmc functions replaced toCoda.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":null,"dir":"Reference","previous_headings":"","what":"Internal vegan functions — vegan-internal","title":"Internal vegan functions — vegan-internal","text":"Internal vegan functions intended called directly, within functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Internal vegan functions — vegan-internal","text":"","code":"ordiParseFormula(formula, data, xlev = NULL, na.action = na.fail, subset = NULL, X) ordiTerminfo(d, data) ordiNAexclude(x, excluded) ordiNApredict(omit, x) ordiArgAbsorber(..., shrink, origin, scaling, triangular, display, choices, const, truemean, FUN) centroids.cca(x, mf, wt) getPermuteMatrix(perm, N, strata = NULL) howHead(x, ...) pasteCall(call, prefix = \"Call:\") veganCovEllipse(cov, center = c(0, 0), scale = 1, npoints = 100) veganMahatrans(x, s2, tol = sqrt(.Machine$double.eps)) hierParseFormula(formula, data) GowerDblcen(x, na.rm = TRUE) addLingoes(d) addCailliez(d)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Internal vegan functions — vegan-internal","text":"description intended vegan developers: functions intended users, used within functions. general, functions exported namespace, must use get ::: directly call functions. ordiParseFormula returns list three matrices (dependent variables, model.matrix constraints conditions, possibly NULL) needed constrained ordination. Argument xlev passed model.frame. left-hand-side already evaluated calling code, can given argument X re-evaluated. ordiTermInfo finds term information constrained ordination described cca.object. ordiNAexclude implements na.action = na.exclude constrained ordination finding WA scores CCA components site scores unconstrained component excluded rows observations. Function ordiNApredict pads result object WA scores similarly napredict. ordiArgAbsorber absorbs arguments scores function vegan cause superfluous warnings graphical function FUN. implement scores functions new arguments, update ordiArgAbsorber. centroids.cca finds weighted centroids variables. getPermuteMatrix interprets user input returns permutation matrix row gives indices observations permutation. input perm can single number number simple permutations, result defining permutation scheme permutation matrix. howHead formats permutation scheme display. formatting compact one used print permute package, shows non-default choices. output normally used printing results vegan permutations. pasteCall prints function call nicely wrapped Sweave output. veganCovEllipse finds coordinates drawing covariance ellipse. veganMahatrans transforms data matrix Euclidean distances Mahalanobis distances. input data x must matrix centred columns, s2 covariance matrix. s2 given, covariance matrix found x within function. hierParseFormula returns list one matrix (left hand side) model frame factors representing hierarchy levels (right hand side) used adipart, multipart hiersimu. GowerDblcen performs Gower double centring matrix dissimilarities. Similar function earlier available compiled code stats, part official API, therefore poorer replacement. addLingoes addCailliez find constant added non-diagonal (squared) dissimilarities make eigenvalues non-negative Principal Co-ordinates Analysis (wcmdscale, dbrda, capscale). Function cmdscale implements Cailliez method. argument matrix dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":null,"dir":"Reference","previous_headings":"","what":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan package provides tools descriptive community ecology. basic functions diversity analysis, community ordination dissimilarity analysis. multivariate tools can used data types well.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"functions vegan package contain tools diversity analysis, ordination methods tools analysis dissimilarities. Together labdsv package, vegan package provides standard tools descriptive community analysis. Package ade4 provides alternative comprehensive package, several packages complement vegan provide tools deeper analysis specific fields. Package BiodiversityR provides GUI large subset vegan functionality. vegan package developed GitHub (https://github.com/vegandevs/vegan/). GitHub provides --date information forums bug reports. important changes vegan documents can read news(package=\"vegan\") vignettes can browsed browseVignettes(\"vegan\"). vignettes include vegan FAQ, discussion design decisions, short introduction ordination discussion diversity methods. see preferable citation package, type citation(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan development team Jari Oksanen, F. Guillaume Blanchet, Roeland Kindt, Pierre Legendre, Peter R. Minchin, R. B. O'Hara, Gavin L. Simpson, Peter Solymos, M. Henry H. Stevens, Helene Wagner. Many people contributed individual functions: see credits function help pages.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"","code":"### Example 1: Unconstrained ordination ## NMDS data(varespec) data(varechem) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.184583 #> ... Procrustes: rmse 0.04902295 max resid 0.1551746 #> Run 2 stress 0.1948413 #> Run 3 stress 0.2136761 #> Run 4 stress 0.2079056 #> Run 5 stress 0.2087937 #> Run 6 stress 0.196245 #> Run 7 stress 0.2069725 #> Run 8 stress 0.1948413 #> Run 9 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04165046 max resid 0.1519318 #> Run 10 stress 0.1843196 #> Run 11 stress 0.2109853 #> Run 12 stress 0.2177543 #> Run 13 stress 0.2224267 #> Run 14 stress 0.2044974 #> Run 15 stress 0.195049 #> Run 16 stress 0.1969805 #> Run 17 stress 0.22911 #> Run 18 stress 0.2397062 #> Run 19 stress 0.18584 #> Run 20 stress 0.2295494 #> *** Best solution was not repeated -- monoMDS stopping criteria: #> 1: no. of iterations >= maxit #> 19: stress ratio > sratmax plot(ord, type = \"t\") ## Fit environmental variables ef <- envfit(ord, varechem) ef #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05721 -0.99836 0.2537 0.038 * #> P 0.61964 0.78489 0.1938 0.102 #> K 0.76634 0.64244 0.1809 0.127 #> Ca 0.68509 0.72846 0.4119 0.005 ** #> Mg 0.63243 0.77462 0.4271 0.003 ** #> S 0.19127 0.98154 0.1752 0.110 #> Al -0.87166 0.49011 0.5269 0.001 *** #> Fe -0.93612 0.35169 0.4451 0.001 *** #> Mn 0.79870 -0.60173 0.5231 0.001 *** #> Zn 0.61752 0.78655 0.1879 0.102 #> Mo -0.90302 0.42960 0.0609 0.521 #> Baresoil 0.92500 -0.37996 0.2508 0.050 * #> Humdepth 0.93288 -0.36018 0.5200 0.002 ** #> pH -0.64804 0.76161 0.2308 0.074 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> plot(ef, p.max = 0.05) ### Example 2: Constrained ordination (RDA) ## The example uses formula interface to define the model data(dune) data(dune.env) ## No constraints: PCA mod0 <- rda(dune ~ 1, dune.env) mod0 #> Call: rda(formula = dune ~ 1, data = dune.env) #> #> Inertia Rank #> Total 84.12 #> Unconstrained 84.12 19 #> Inertia is variance #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 24.795 18.147 7.629 7.153 5.695 4.333 3.199 2.782 #> (Showing 8 of 19 unconstrained eigenvalues) #> plot(mod0) ## All environmental variables: Full model mod1 <- rda(dune ~ ., dune.env) mod1 #> Call: rda(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 63.2062 0.7513 12 #> Unconstrained 20.9175 0.2487 7 #> Inertia is variance #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 22.396 16.208 7.039 4.038 3.760 2.609 2.167 1.803 1.404 0.917 0.582 #> RDA12 #> 0.284 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 #> 6.627 4.309 3.549 2.546 2.340 0.934 0.612 #> plot(mod1) ## Automatic selection of variables by permutation P-values mod <- ordistep(mod0, scope=formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.015 * #> + A1 1 89.591 1.9217 0.065 . #> + Use 2 91.032 1.1741 0.240 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.095 . #> + A1 1 87.424 1.2965 0.200 #> + Use 2 88.284 1.0510 0.360 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.015 * #> - Management 3 87.707 2.1769 0.010 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.325 #> + A1 1 86.220 0.8359 0.585 #> + Use 2 86.842 0.8027 0.770 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> plot(mod) ## Permutation test for all variables anova(mod) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Model 6 46.425 2.6682 0.001 *** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test of \"type III\" effects, or significance when a term ## is added to the model after all other terms anova(mod, by = \"margin\") #> Permutation test for rda under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Management 3 18.938 2.1769 0.005 ** #> Moisture 3 17.194 1.9764 0.008 ** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Plot only sample plots, use different symbols and draw SD ellipses ## for Managemenet classes plot(mod, display = \"sites\", type = \"n\") with(dune.env, points(mod, disp = \"si\", pch = as.numeric(Management))) with(dune.env, legend(\"topleft\", levels(Management), pch = 1:4, title = \"Management\")) with(dune.env, ordiellipse(mod, Management, label = TRUE)) ## add fitted surface of diversity to the model ordisurf(mod, diversity(dune), add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 1.28 total = 2.28 #> #> REML score: 3.00623 ### Example 3: analysis of dissimilarites a.k.a. non-parametric ### permutational anova adonis2(dune ~ ., dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ ., data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> A1 1 0.7230 0.16817 5.2038 0.002 ** #> Moisture 3 1.1871 0.27613 2.8482 0.006 ** #> Management 3 0.9036 0.21019 2.1681 0.028 * #> Use 2 0.0921 0.02143 0.3315 0.983 #> Manure 3 0.4208 0.09787 1.0096 0.439 #> Residual 7 0.9725 0.22621 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adonis2(dune ~ Management + Moisture, dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management + Moisture, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.7907 0.001 *** #> Moisture 3 1.1516 0.26788 2.9726 0.002 ** #> Residual 13 1.6788 0.39051 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Dissimilarity Indices for Community Ecologists — vegdist","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function computes dissimilarity indices useful popular community ecologists. indices use quantitative data, although named corresponding binary index, can calculate binary index using appropriate argument. find favourite index , can see can implemented using designdist. Gower, Bray--Curtis, Jaccard Kulczynski indices good detecting underlying ecological gradients (Faith et al. 1987). Morisita, Horn--Morisita, Binomial, Cao Chao indices able handle different sample sizes (Wolda 1981, Krebs 1999, Anderson & Millar 2004), Mountford (1962) Raup-Crick indices presence--absence data able handle unknown (variable) sample sizes. indices discussed Krebs (1999) Legendre & Legendre (2012), properties compared Wolda (1981) Legendre & De Cáceres (2012). Aitchison (1986) distance equivalent Euclidean distance CLR-transformed samples (\"clr\") deals positive compositional data. Robust Aitchison distance Martino et al. (2019) uses robust CLR (\"rlcr\"), making applicable non-negative data including zeroes (unlike standard Aitchison).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"vegdist(x, method=\"bray\", binary=FALSE, diag=FALSE, upper=FALSE, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"x Community data matrix. method Dissimilarity index, partial match \"manhattan\", \"euclidean\", \"canberra\", \"clark\", \"bray\", \"kulczynski\", \"jaccard\", \"gower\", \"altGower\", \"morisita\", \"horn\", \"mountford\", \"raup\", \"binomial\", \"chao\", \"cao\", \"mahalanobis\", \"chisq\", \"chord\", \"hellinger\", \"aitchison\", \"robust.aitchison\". binary Perform presence/absence standardization analysis using decostand. diag Compute diagonals. upper Return upper diagonal. na.rm Pairwise deletion missing observations computing dissimilarities. ... parameters. ignored, except method =\"gower\" accepts range.global parameter decostand, method=\"aitchison\", accepts pseudocount parameter decostand used clr transformation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jaccard (\"jaccard\"), Mountford (\"mountford\"), Raup--Crick (\"raup\"), Binomial Chao indices discussed later section. function also finds indices presence/ absence data setting binary = TRUE. following overview gives first quantitative version, \\(x_{ij}\\) \\(x_{ik}\\) refer quantity species (column) \\(\\) sites (rows) \\(j\\) \\(k\\). binary versions \\(\\) \\(B\\) numbers species compared sites, \\(J\\) number species occur compared sites similarly designdist (many indices produce identical binary versions): Jaccard index computed \\(2B/(1+B)\\), \\(B\\) Bray--Curtis dissimilarity. Binomial index derived Binomial deviance null hypothesis two compared communities equal. able handle variable sample sizes. index fixed upper limit, can vary among sites shared species. discussion, see Anderson & Millar (2004). Cao index CYd index (Cao et al. 1997) suggested minimally biased index high beta diversity variable sampling intensity. Cao index fixed upper limit, can vary among sites shared species. index intended count (integer) data, undefined zero abundances; replaced arbitrary value \\(0.1\\) following Cao et al. (1997). Cao et al. (1997) used \\(\\log_{10}\\), current function uses natural logarithms values approximately \\(2.30\\) times higher 10-based logarithms. Anderson & Thompson (2004) give alternative formulation Cao index highlight relationship Binomial index (). Mountford index defined \\(M = 1/\\alpha\\) \\(\\alpha\\) parameter Fisher's logseries assuming compared communities samples community (cf. fisherfit, fisher.alpha). index \\(M\\) found positive root equation \\(\\exp() + \\exp(bM) = 1 + \\exp[(+b-j)M]\\), \\(j\\) number species occurring communities, \\(\\) \\(b\\) number species separate community (index uses presence--absence information). Mountford index usually misrepresented literature: indeed Mountford (1962) suggested approximation used starting value iterations, proper index defined root equation . function vegdist solves \\(M\\) Newton method. Please note either \\(\\) \\(b\\) equal \\(j\\), one communities subset , dissimilarity \\(0\\) meaning non-identical objects may regarded similar index non-metric. Mountford index range \\(0 \\dots \\log(2)\\). Raup--Crick dissimilarity (method = \"raup\") probabilistic index based presence/absence data. defined \\(1 - prob(j)\\), based probability observing least \\(j\\) species shared compared communities. current function uses analytic result hypergeometric distribution (phyper) find probabilities. probability (index) dependent number species missing sites, adding -zero species data removing missing species data influence index. probability ( index) may almost zero almost one wide range parameter values. index nonmetric: two communities shared species may dissimilarity slightly one, two identical communities may dissimilarity slightly zero. index uses equal occurrence probabilities species, Raup Crick originally suggested sampling probabilities proportional species frequencies (Chase et al. 2011). simulation approach unequal species sampling probabilities implemented raupcrick function following Chase et al. (2011). index can also used transposed data give probabilistic dissimilarity index species co-occurrence (identical Veech 2013). Chao index tries take account number unseen species pairs, similarly method = \"chao\" specpool. Function vegdist implements Jaccard, index defined \\(1-\\frac{U \\times V}{U + V - U \\times V}\\); types can defined function chaodist. Chao equation, \\(U = C_j/N_j + (N_k - 1)/N_k \\times a_1/(2 a_2) \\times S_j/N_j\\), \\(V\\) similar except site index \\(k\\). \\(C_j\\) total number individuals species site \\(j\\) shared site \\(k\\), \\(N_j\\) total number individuals site \\(j\\), \\(a_1\\) (\\(a_2\\)) number species occurring site \\(j\\) one (two) individuals site \\(k\\), \\(S_j\\) total number individuals species present site \\(j\\) occur one individual site \\(k\\) (Chao et al. 2005). Morisita index can used genuine count data (integers) . Horn--Morisita variant able handle abundance data. Mahalanobis distances Euclidean distances matrix columns centred, unit variance, uncorrelated. index commonly used community data, sometimes used environmental variables. calculation based transforming data matrix using Euclidean distances following Mardia et al. (1979). Mahalanobis transformation usually fails number columns larger number rows (sampling units). transformation fails, distances nearly constant except small numeric noise. Users must check returned Mahalanobis distances meaningful. Euclidean Manhattan dissimilarities good gradient separation without proper standardization still included comparison special needs. Chi-square distances (\"chisq\") Euclidean distances Chi-square transformed data (see decostand). internal standardization used correspondence analysis (cca, decorana). Weighted principal coordinates analysis distances row sums weights equal correspondence analysis (see Example wcmdscale). Chi-square distance intended non-negative data, typical community data. However, can calculated long margin sums positive, warning issued negative data entries. Chord distances (\"chord\") Euclidean distance matrix rows standardized unit norm (sums squares 1) using decostand. Geometrically standardization moves row points surface multidimensional unit sphere, distances chords across hypersphere. Hellinger distances (\"hellinger\") related Chord distances, data standardized unit total (row sums 1) using decostand, square root transformed. distances upper limit \\(\\sqrt{2}\\). Bray--Curtis Jaccard indices rank-order similar, indices become identical rank-order similar standardizations, especially presence/absence transformation equalizing site totals decostand. Jaccard index metric, probably preferred instead default Bray-Curtis semimetric. Aitchison distance (1986) robust Aitchison distance (Martino et al. 2019) metrics deal compositional data. Aitchison distance said outperform Jensen-Shannon divergence Bray-Curtis dissimilarity, due better stability subsetting aggregation, proper distance (Aitchison et al., 2000). naming conventions vary. one adopted traditional rather truthful priority. function finds either quantitative binary variants indices name, correctly may refer one alternatives instance, Bray index known also Steinhaus, Czekanowski Sørensen index. quantitative version Jaccard probably called Ružička index. abbreviation \"horn\" Horn--Morisita index misleading, since separate Horn index. abbreviation changed index implemented vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Function drop-replacement dist function returns distance object type. result object adds attribute maxdist gives theoretical maximum index sampling units share species, NA maximum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Aitchison, J., Barceló-Vidal, C., Martín-Fernández, J.., Pawlowsky-Glahn, V. (2000). Logratio analysis compositional distance. Math. Geol. 32, 271–275. Anderson, M.J. Millar, R.B. (2004). Spatial variation effects habitat temperate reef fish assemblages northeastern New Zealand. Journal Experimental Marine Biology Ecology 305, 191--221. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006). Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Anderson, M.J & Thompson, .. (2004). Multivariate control charts ecological environmental monitoring. Ecological Applications 14, 1921--1935. Cao, Y., Williams, W.P. & Bark, .W. (1997). Similarity measure bias river benthic Auswuchs community analysis. Water Environment Research 69, 95--106. Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005). new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1 Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J. C. (1971). general coefficient similarity properties. Biometrics 27, 623--637. Krebs, C. J. (1999). Ecological Methodology. Addison Wesley Longman. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141 Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate analysis. Academic Press. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Mountford, M. D. (1962). index similarity application classification problems. : P.W.Murphy (ed.), Progress Soil Zoology, 43--50. Butterworths. Veech, J. . (2013). probabilistic model analysing species co-occurrence. Global Ecology Biogeography 22, 252--260. Wolda, H. (1981). Similarity indices, sample size diversity. Oecologia 50, 296--302.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jari Oksanen, contributions Tyler Smith (Gower index), Michael Bedward (Raup--Crick index), Leo Lahti (Aitchison robust Aitchison distance).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function alternative dist adding ecologically meaningful indices. methods produce similar types objects can interchanged method accepting either. Manhattan Euclidean dissimilarities identical methods. Canberra index divided number variables vegdist, dist. differ constant multiplier, alternative vegdist range (0,1). Function daisy (package cluster) provides alternative implementation Gower index also can handle mixed data numeric class variables. two versions Gower distance (\"gower\", \"altGower\") differ scaling: \"gower\" divides distances number observations (rows) scales column unit range, \"altGower\" omits double-zeros divides number pairs least one -zero value, scale columns (Anderson et al. 2006). can use decostand add range standardization \"altGower\" (see Examples). Gower (1971) suggested omitting double zeros presences, often taken general feature Gower distances. See Examples implementing Anderson et al. (2006) variant Gower index. dissimilarity indices vegdist designed community data, give misleading values negative data entries. results may also misleading NA NaN empty sites. principle, study species composition without species remove empty sites community data.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"data(varespec) vare.dist <- vegdist(varespec) # Orlóci's Chord distance: range 0 .. sqrt(2) vare.dist <- vegdist(decostand(varespec, \"norm\"), \"euclidean\") # Anderson et al. (2006) version of Gower vare.dist <- vegdist(decostand(varespec, \"log\"), \"altGower\") #> Warning: non-integer data: divided by smallest positive value # Range standardization with \"altGower\" (that excludes double-zeros) vare.dist <- vegdist(decostand(varespec, \"range\"), \"altGower\")"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Compact Ordered Community Tables — vegemite","title":"Display Compact Ordered Community Tables — vegemite","text":"Functions vegemite tabasco display compact community tables. Function vegemite prints text tables species rows, site takes one column without spaces. Function tabasco provides interface heatmap colour image data. community table can ordered explicit indexing, environmental variables results ordination cluster analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"vegemite(x, use, scale, sp.ind, site.ind, zero=\".\", select ,...) tabasco(x, use, sp.ind = NULL, site.ind = NULL, select, Rowv = TRUE, Colv = TRUE, labRow = NULL, labCol = NULL, scale, col = heat.colors(12), ...) coverscale(x, scale=c(\"Braun.Blanquet\", \"Domin\", \"Hult\", \"Hill\", \"fix\",\"log\"), maxabund, character = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Compact Ordered Community Tables — vegemite","text":"x Community data. use Either vector, object cca, decorana etc. hclust dendrogram ordering sites species. sp.ind, site.ind Species site indices. tabasco, can also hclust tree, agnes clusterings dendrograms. zero Character used zeros. select Select subset sites. can logical vector (TRUE selected sites), vector indices selected sites. order indices influence results, must specify use site.ind reorder sites. Rowv, Colv Re-order dendrograms rows (sites) columns (species) x. Rowv = TRUE, row dendrograms ordered first axis correspondence analysis, Colv = TRUE column dendrograms weighted average (wascores) row order. Alternatively, arguments can vectors used reorder dendrogram. labRow, labCol character vectors row column labels used heatmap instead default. NB., input matrix transposed row labels used data columns. scale vegemite coverscale: cover scale used (can abbreviated). tabasco: scaling colours heatmap. alternatives coverscale can used tabasco, addition \"column\" \"row\" scale columns rows equal maxima (NB., refer transposed data heatmap), \"none\" uses original values. col vector colours used -zero abundance values. maxabund Maximum abundance used scale = \"log\". Data maximum selected subset used missing. character Return character codes suitable vegemite. FALSE, returns corresponding integers. ... Arguments passed coverscale (.e., maxabund) vegemite heatmap tabasco.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Compact Ordered Community Tables — vegemite","text":"function vegemite prints traditional community table. display transposed, species rows sites columns. table printed compact form: one character can used abundance, spaces columns. Species occurrences dropped table. Function tabasco produces similar table vegemite using heatmap, abundances coded colours. function scales abundances equal intervals colour palette, either rows columns can scaled equal maxima, coverscale class systems can used. function can also display dendrograms sites (columns) species given argument (use sites, sp.ind species). parameter use used re-order output. use can vector object hclust agnes, dendrogram ordination result recognized scores (ordination methods vegan vegan). hclust, agnes dendrogram must sites. dendrogram displayed sites tabasco, shown vegemite. dendrogram species displayed, except given sp.ind. use vector, used ordering sites. use object ordination, sites species arranged first axis (provided results available also species). use object hclust, agnes dendrogram, sites ordered similarly cluster dendrogram. Function tabasco re-orders dendrogram Rowv = TRUE Rowv vector. re-ordering available vegemite, can done hand using reorder.dendrogram reorder.hclust. Please note dendrogram hclust reordering can differ: unweighted means merged branches used dendrogram, weighted means (= means leaves cluster) used reorder.hclust. cases species scores missing, species ordered weighted averages (wascores) site order. Species sites can ordered explicitly giving indices names parameters sp.ind site.ind. given, take precedence use. subset sites can displayed using argument select, used order sites, still must give use site.ind. However, tabasco makes two exceptions: site.ind select used use dendrogram (clustering result). addition, sp.ind can hclust tree, agnes clustering dendrogram, case dendrogram plotted left side heatmap. Phylogenetic trees directly used, package ape tools transform hclust trees. scale given, vegemite calls coverscale transform percent cover scale scales traditional class scales used vegetation science (coverscale can called directly, ). Function tabasco can also use traditional class scales, treats transformed values corresponding integers. Braun-Blanquet Domin scales actually strict cover scales, limits used codes r + arbitrary. Scale Hill may inappropriately named, since Mark O. Hill probably never intended cover scale. However, used default “cut levels” TWINSPAN, surprisingly many users stick default, de facto standard publications. traditional scales assume values cover percentages maximum 100. However, non-traditional alternative log can used scale range. class limits integer powers 1/2 maximum (argument maxabund), + used non-zero entries less 1/512 maximum (log stands alternatively logarithmic logical). Scale fix intended “fixing” 10-point scales: truncates scale values integers, replaces 10 X positive values 1 +.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Compact Ordered Community Tables — vegemite","text":"functions used mainly display table, return (invisibly) list items species ordered species index, sites ordered site index, table final ordered community table. items can used arguments sp.ind site.ind reproduce table, table can edited. addition table, vegemite prints numbers species sites name used cover scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Display Compact Ordered Community Tables — vegemite","text":"cover scales presented many textbooks vegetation science; used: Shimwell, D.W. (1971) Description Classification Vegetation. Sidgwick & Jackson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Compact Ordered Community Tables — vegemite","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Compact Ordered Community Tables — vegemite","text":"name vegemite chosen output compact, tabasco just compact, uses heat colours.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"data(varespec) ## Print only more common species freq <- apply(varespec > 0, 2, sum) vegemite(varespec, scale=\"Hult\", sp.ind = freq > 10) #> #> 1122212121122 1112 #> 854739268340575634292011 #> Callvulg 111...11.311...1111.111. #> Empenigr 211332121112211111212213 #> Vaccviti 323332212113221211233234 #> Pinusylv 111.111111111.11.1111111 #> Dicrfusc 12.111441121211111111111 #> Dicrpoly .11.11.11..1.11....1.111 #> Pleuschr 144533435123411111111131 #> Polyjuni 111.11.111112111.111.111 #> Pohlnuta 111111111111.1..11.11111 #> Ptilcili 1111111111..1..11.11..12 #> Cladarbu 321122121332143423111121 #> Cladrang 321221131312145443313241 #> Cladstel 11111311.211.11254555542 #> Cladunci 112111111131111111111111 #> Cladcocc 11..11111111111111.1.11. #> Cladcorn 111111111111111111111111 #> Cladgrac 111111111111111111111.11 #> Cladfimb 11.1111111111111111111.1 #> Cladcris 111111111111111111111111 #> Cladchlo ..1.111..1...1.11.11.1.1 #> Cetreric 111...11.111111111.1.1.. #> Cetrisla .11....1111.1....1.11111 #> Stersp 111.11.1111.112111...11. #> Claddefo 1111111111111111111111.1 #> 24 sites, 24 species #> scale: Hult ## Order by correspondence analysis, use Hill scaling and layout: dca <- decorana(varespec) vegemite(varespec, dca, \"Hill\", zero=\"-\") #> #> 1 1 1 11122211122222 #> 203942561738913046572458 #> Flavniva -1114-11-1-11-1--------- #> Cladstel 5555551451425411111211-- #> Cladphyl -1-1----1-------1------- #> Cladcerv 1---1-----------------1- #> Cladsp ---11--1--11---1-1-11-1- #> Cladamau --1---1----1------------ #> Cladchlo 1111---1-11-111-----11-- #> Cladrang 535254555555223414332321 #> Diphcomp --11-----112----------1- #> Stersp -11-1-4111111-1-111--111 #> Pinusylv 11-111111-111111111-1111 #> Polypili ------111--11-11--1----- #> Cetrisla -1-111--1-1--1--111--111 #> Cladcocc -1111-1111111-11111-1-11 #> Cladarbu 113142453555313343413231 #> Vacculig --1-1----3-1---1---12-11 #> Pohlnuta -11111--11111111111111-1 #> Cladfimb 11111111-111111111111-11 #> Callvulg 111-21-12-51---1221-21-- #> Icmaeric ------1--1------11------ #> Empenigr 342214131314344333143131 #> Vaccviti 342514244334455432444443 #> Cladgrac 1-1111111111111111111111 #> Cetreric -1111-11-111---1111-111- #> Cladcorn 111111111111111111111111 #> Cladcris 111111111111111111111111 #> Peltapht --------1-11--1----1--1- #> Ptilcili 1-11---11-11141--1111111 #> Barbhatc ----------1-121----1---- #> Claddefo 11111111-111111111111111 #> Cladbotr ----------1-1111---1-1-1 #> Betupube -------------1------1-1- #> Dicrpoly -1-1--1-11--1211-11--1-1 #> Cladunci 111111122111111251212311 #> Polycomm ------------11--1--1--1- #> Polyjuni 11-11-1111111121111--131 #> Rhodtome ----------1--2-----21--1 #> Dicrfusc 111111111121112145425-41 #> Pleuschr 111213114132524434555555 #> Vaccmyrt ---1------1-24--12133--4 #> Descflex ----1----1--11-----21-11 #> Nepharct --1------1-1---1------2- #> Dicrsp -----1----1--1-1-11-1541 #> Hylosple ---------------1---3--13 #> 24 sites, 44 species #> scale: Hill ## Show one class from cluster analysis, but retain the ordering above clus <- hclust(vegdist(varespec)) cl <- cutree(clus, 3) sel <- vegemite(varespec, use=dca, select = cl == 3, scale=\"Br\") #> #> 1 12 #> 20921 #> Flavniva .++.. #> Cladstel 55542 #> Cladphyl .++.. #> Cladcerv r.... #> Cladsp ..+.. #> Cladchlo r++.+ #> Cladrang 22121 #> Diphcomp ..+.. #> Stersp .+... #> Pinusylv r++1+ #> Cetrisla .++++ #> Cladcocc .++.. #> Cladarbu +1+1+ #> Pohlnuta .++++ #> Cladfimb r++++ #> Callvulg r+.+. #> Empenigr 22122 #> Vaccviti 22223 #> Cladgrac r.+++ #> Cetreric .++.. #> Cladcorn r+++r #> Cladcris r++++ #> Ptilcili r.+.2 #> Barbhatc ....1 #> Claddefo r++++ #> Cladbotr ....+ #> Betupube ....+ #> Dicrpoly .++.1 #> Cladunci +++1+ #> Polycomm ....+ #> Polyjuni r++.+ #> Rhodtome ....1 #> Dicrfusc r++++ #> Pleuschr ++121 #> Vaccmyrt ..+.2 #> Descflex ....+ #> Dicrsp ...++ #> 5 sites, 37 species #> scale: Braun.Blanquet ## Re-create previous vegemite(varespec, sp=sel$sp, site=sel$site, scale=\"Hult\") #> #> 1 12 #> 20921 #> Flavniva .11.. #> Cladstel 55552 #> Cladphyl .11.. #> Cladcerv 1.... #> Cladsp ..1.. #> Cladchlo 111.1 #> Cladrang 32131 #> Diphcomp ..1.. #> Stersp .1... #> Pinusylv 11111 #> Cetrisla .1111 #> Cladcocc .11.. #> Cladarbu 11111 #> Pohlnuta .1111 #> Cladfimb 11111 #> Callvulg 11.1. #> Empenigr 22123 #> Vaccviti 22334 #> Cladgrac 1.111 #> Cetreric .11.. #> Cladcorn 11111 #> Cladcris 11111 #> Ptilcili 1.1.2 #> Barbhatc ....1 #> Claddefo 11111 #> Cladbotr ....1 #> Betupube ....1 #> Dicrpoly .11.1 #> Cladunci 11111 #> Polycomm ....1 #> Polyjuni 111.1 #> Rhodtome ....1 #> Dicrfusc 11111 #> Pleuschr 11111 #> Vaccmyrt ..1.3 #> Descflex ....1 #> Dicrsp ...11 #> 5 sites, 37 species #> scale: Hult ## Re-order clusters by ordination clus <- as.dendrogram(clus) clus <- reorder(clus, scores(dca, choices=1, display=\"sites\"), agglo.FUN = mean) vegemite(varespec, clus, scale = \"Hult\") #> #> 1 111 1211221212222 #> 431567380922149306254578 #> Flavniva 21.111.111....11........ #> Cladamau .1.1...1................ #> Stersp 111211111....111.1.111.1 #> Polypili ..111..1......111..1.... #> Diphcomp .1...111.1...........1.. #> Cladphyl ..1.....11...1.......... #> Cladrang 344544332133111223121121 #> Cladcerv 1.........1..........1.. #> Cladstel 454121215555213111111.1. #> Cladarbu 322344331111132222121111 #> Vacculig 11...2.1........1.1..111 #> Callvulg 111.1.311.11.1..11111... #> Icmaeric ...1.1.......1...1...... #> Cladsp 1...1.11.1......111..11. #> Cladcocc 1111111111...1111111.1.1 #> Pinusylv 1.111.1111111111111111.1 #> Cladchlo .1..111.111.1.11..1.1... #> Cetrisla 1.1...1.11.111...1.111.1 #> Cladfimb 11.11111111111111111.111 #> Peltapht ..1...11.......1.....11. #> Cetreric 11.1111111...1..111111.. #> Cladgrac 11111111.111111111111111 #> Pohlnuta 111..11111.1111111111.11 #> Ptilcili .11.1.11.11.2.11.1111111 #> Barbhatc ......1.....1.11......1. #> Cladcorn 111111111111111111111111 #> Vaccviti 113122132323412331223232 #> Cladcris 111111111111111111111111 #> Empenigr 111111122122312322111231 #> Cladbotr ......1.....1.111...1.11 #> Betupube ............1.....1..1.. #> Cladunci 111111111111131111112111 #> Claddefo 11.111111111111111111111 #> Dicrpoly ..11.1..11..1.1111.11..1 #> Polycomm ............111......11. #> Rhodtome ......1.....1.....1...11 #> Polyjuni 1.111111111.111111.112.1 #> Dicrfusc 11111111111112111442.211 #> Pleuschr 113111111111123333444455 #> Vaccmyrt ......1..1..311..111..13 #> Nepharct .1...1.1........1....1.. #> Dicrsp ......1....11...111133.1 #> Descflex 1....1......1.1...1..111 #> Hylosple ................1....122 #> 24 sites, 44 species #> scale: Hult ## Abundance values have such a wide range that they must be rescaled tabasco(varespec, dca, scale=\"Braun\") ## Classification trees for species data(dune, dune.taxon) taxontree <- hclust(taxa2dist(dune.taxon)) plotree <- hclust(vegdist(dune), \"average\") ## Automatic reordering of clusters tabasco(dune, plotree, sp.ind = taxontree) ## No reordering of taxonomy tabasco(dune, plotree, sp.ind = taxontree, Colv = FALSE) ## Species cluster: most dissimilarity indices do a bad job when ## comparing rare and common species, but Raup-Crick makes sense sptree <- hclust(vegdist(t(dune), \"raup\"), \"average\") tabasco(dune, plotree, sptree)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Averages Scores for Species — wascores","title":"Weighted Averages Scores for Species — wascores","text":"Computes Weighted Averages scores species ordination configuration environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Averages Scores for Species — wascores","text":"","code":"wascores(x, w, expand=FALSE) eigengrad(x, w)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Averages Scores for Species — wascores","text":"x Environmental variables ordination scores. w Weights: species abundances. expand Expand weighted averages weighted variance corresponding environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores computes weighted averages. Weighted averages “shrink”: extreme values used calculating averages. expand = TRUE, function “deshrinks” weighted averages making biased weighted variance equal biased weighted variance corresponding environmental variable. Function eigengrad returns inverses squared expansion factors attribute shrinkage wascores result environmental gradient. equal constrained eigenvalue cca one gradient used constraint, describes strength gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores returns matrix species define rows ordination axes environmental variables define columns. expand = TRUE, attribute shrinkage inverses squared expansion factors cca eigenvalues variable. Function eigengrad returns shrinkage attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Weighted Averages Scores for Species — wascores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Averages Scores for Species — wascores","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(wisconsin(varespec)) vare.mds <- monoMDS(vare.dist) vare.points <- postMDS(vare.mds$points, vare.dist) vare.wa <- wascores(vare.points, varespec) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa, rownames(vare.wa), cex=0.8, col=\"blue\") ## Omit rare species (frequency <= 4) freq <- apply(varespec>0, 2, sum) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa[freq > 4,], rownames(vare.wa)[freq > 4],cex=0.8,col=\"blue\") ## Works for environmental variables, too. wascores(varechem, varespec) #> N P K Ca Mg S Al #> Callvulg 25.12401 41.66188 246.92383 572.3431 99.40828 49.10552 245.15751 #> Empenigr 21.61371 44.17350 158.92517 580.8403 89.20628 35.75327 107.05670 #> Rhodtome 22.34553 40.35530 162.31108 643.8819 95.07712 30.84303 28.83159 #> Vaccmyrt 24.96352 49.80649 189.70177 656.4179 96.75529 34.55190 32.75705 #> Vaccviti 21.14028 45.86984 162.12372 613.2126 94.30084 37.26975 116.14656 #> Pinusylv 18.37299 44.24818 163.60195 670.9387 93.52214 37.02238 150.45523 #> Descflex 22.24089 54.74804 212.20357 771.0159 114.15179 38.00000 24.64143 #> Betupube 21.51034 28.86552 112.28276 513.5586 75.15172 23.55172 33.12414 #> Vacculig 28.00729 33.48758 114.90230 372.9653 70.89954 29.32378 202.89008 #> Diphcomp 22.34228 39.71049 127.44259 446.1565 80.16173 32.32963 122.31574 #> Dicrsp 21.33007 60.03758 185.04563 828.0544 148.67509 46.75427 90.42294 #> Dicrfusc 23.45681 39.14575 162.91954 578.2309 77.81648 33.48086 60.66890 #> Dicrpoly 20.65446 43.87409 150.51485 665.5845 115.22112 36.17079 90.16733 #> Hylosple 26.10599 67.88980 245.78681 779.6520 111.96685 42.27433 24.92738 #> Pleuschr 22.60476 54.22534 199.96241 712.6278 109.23425 40.01132 70.43900 #> Polypili 23.17377 43.75902 144.82623 724.9738 85.42623 30.58525 145.73115 #> Polyjuni 22.89480 47.98022 154.81906 643.3864 87.27819 33.63863 53.24888 #> Polycomm 21.73521 41.17042 154.91549 631.8704 101.84789 32.17324 46.80986 #> Pohlnuta 19.99885 48.59198 169.68855 678.3813 104.31641 39.94427 132.05458 #> Ptilcili 21.27880 33.44211 127.08522 564.5652 85.96417 27.11720 56.60692 #> Barbhatc 21.17461 27.93323 113.13542 497.9138 77.50564 23.72288 42.09749 #> Cladarbu 23.56127 38.04952 142.03073 454.9019 74.00779 33.81002 173.12698 #> Cladrang 24.28421 38.60534 135.31177 463.2750 70.54209 32.53349 183.79979 #> Cladstel 19.28049 46.71060 158.00576 540.4904 80.19153 40.29106 225.89526 #> Cladunci 21.41240 45.49844 163.40402 621.9100 98.35538 40.00734 119.59481 #> Cladcocc 21.72473 42.80681 156.32330 557.9007 80.95448 36.25161 149.82616 #> Cladcorn 22.11640 47.06656 160.36881 623.5185 95.17781 36.75273 104.71463 #> Cladgrac 22.51887 44.06576 156.50214 583.1558 94.10292 36.93930 134.13424 #> Cladfimb 21.77980 41.82652 153.29444 512.4646 78.28232 35.62323 128.96061 #> Cladcris 20.88795 44.12262 171.04016 574.5672 92.52169 38.24003 116.03507 #> Cladchlo 19.51207 45.39655 150.93190 571.0233 95.77586 39.50862 156.81983 #> Cladbotr 22.97660 38.89574 167.20000 590.8021 99.57234 34.79362 87.75957 #> Cladamau 25.07143 35.84286 105.07857 395.2214 68.18571 27.11429 95.91429 #> Cladsp 19.21923 47.37308 168.49231 526.7654 79.54423 45.15385 215.33846 #> Cetreric 21.00944 47.76972 165.07972 579.6322 99.14944 42.25472 163.46000 #> Cetrisla 18.36552 42.73695 151.78374 626.2813 89.77833 35.33498 132.68227 #> Flavniva 18.56110 61.18194 207.67705 502.9203 60.91755 50.22532 396.82405 #> Nepharct 23.33099 49.10019 146.84715 618.1601 64.27319 29.95760 31.72300 #> Stersp 28.19743 32.84800 94.33459 389.5143 53.25377 24.22175 95.39326 #> Peltapht 21.08553 54.45395 193.38816 886.5487 119.35132 37.92500 106.16447 #> Icmaeric 28.88636 27.00000 87.86818 307.0500 40.48182 22.17273 89.94091 #> Cladcerv 20.25000 56.79000 192.36000 519.2300 62.10000 45.18000 314.92000 #> Claddefo 22.19198 45.22981 167.73069 583.7983 92.01320 38.51369 100.46139 #> Cladphyl 15.73750 54.56875 180.39375 775.4500 99.65625 43.35000 208.55000 #> Fe Mn Zn Mo Baresoil Humdepth pH #> Callvulg 75.457843 52.38247 8.281074 0.4734635 27.241036 2.187819 2.845108 #> Empenigr 38.146102 53.49357 7.159938 0.3289657 27.324317 2.367439 2.888078 #> Rhodtome 5.560906 70.48260 7.444100 0.2251490 37.325030 2.689154 2.895352 #> Vaccmyrt 5.589213 75.17221 7.838533 0.2666732 31.404171 2.798935 2.855216 #> Vaccviti 37.586067 51.81515 7.617213 0.3680289 26.307701 2.307879 2.923128 #> Pinusylv 39.121898 35.22311 7.733333 0.3485401 17.762968 1.996350 3.049148 #> Descflex 6.066429 110.87232 9.526607 0.2316071 22.740179 2.834821 2.822857 #> Betupube 5.417241 37.53448 5.637931 0.2068966 51.496552 2.527586 2.979310 #> Vacculig 93.963929 37.73062 4.593824 0.3780552 21.410710 2.041196 3.006965 #> Diphcomp 73.281173 46.88025 4.593827 0.3725309 31.836574 2.103704 2.856790 #> Dicrsp 22.504617 65.35020 13.060765 0.5610370 23.182889 2.232247 2.954272 #> Dicrfusc 13.922252 61.40958 6.922859 0.3218816 26.918674 2.484399 2.806431 #> Dicrpoly 20.973927 33.28779 9.110561 0.3892739 37.304043 2.228713 3.015842 #> Hylosple 4.729157 115.14606 9.885976 0.2851996 20.956264 2.925000 2.807594 #> Pleuschr 19.113811 77.12277 9.007860 0.3405945 24.584979 2.596881 2.858446 #> Polypili 51.993443 36.74754 8.045902 0.2204918 17.368852 1.493443 3.227869 #> Polyjuni 12.885704 82.56274 7.945126 0.2760289 28.116303 2.615523 2.874729 #> Polycomm 7.895775 68.22535 7.843662 0.2591549 38.687324 2.926761 2.860563 #> Pohlnuta 33.089313 42.02290 8.452290 0.3935115 24.709351 2.147328 2.985496 #> Ptilcili 14.036188 33.93547 5.906924 0.2303712 48.941884 2.502498 2.973376 #> Barbhatc 8.199687 31.11379 5.550784 0.2084639 54.331975 2.514734 2.986834 #> Cladarbu 65.470394 38.28429 6.387500 0.4464046 22.592997 2.048540 2.937879 #> Cladrang 76.612752 34.86010 6.616452 0.3903501 17.270158 1.799128 3.022346 #> Cladstel 84.639467 33.27903 7.287216 0.4054879 9.854042 1.851973 3.052167 #> Cladunci 27.463504 40.10322 9.108102 0.5120114 28.312376 2.362687 2.858564 #> Cladcocc 46.653763 43.53584 7.269176 0.3698925 19.972222 2.025090 2.974194 #> Cladcorn 32.916238 53.38441 7.518489 0.3639871 26.620868 2.399518 2.890997 #> Cladgrac 45.758366 44.63911 7.702724 0.4244163 25.546654 2.230350 2.933658 #> Cladfimb 41.710354 47.83687 6.815152 0.3887626 24.763889 2.239394 2.902778 #> Cladcris 34.729585 44.75984 6.933735 0.3746988 29.711352 2.400000 2.841633 #> Cladchlo 42.089655 35.13276 7.908621 0.3814655 22.319655 2.075000 3.022414 #> Cladbotr 23.374468 46.54255 7.263830 0.3106383 45.725532 2.580851 2.904255 #> Cladamau 68.971429 41.98571 4.928571 0.3214286 27.592857 1.857143 2.914286 #> Cladsp 47.913462 49.61923 8.421154 0.6173077 16.864423 2.213462 2.921154 #> Cetreric 42.614167 36.96694 9.516389 0.5687500 21.452639 2.058056 2.923889 #> Cetrisla 29.617734 33.24138 7.518227 0.3192118 26.417980 2.027586 3.065025 #> Flavniva 94.339916 37.03232 9.116371 0.9987764 19.692312 1.799241 2.923629 #> Nepharct 12.910837 115.15684 7.743536 0.2180608 23.135932 2.541065 2.918251 #> Stersp 30.226998 31.97061 7.635502 0.2828767 15.844007 1.477740 3.038756 #> Peltapht 37.598684 56.64079 7.652632 0.2046053 28.321053 2.286842 3.026316 #> Icmaeric 24.236364 23.95909 6.618182 0.2863636 18.727273 1.568182 2.968182 #> Cladcerv 111.090000 52.04000 8.530000 0.6800000 15.393000 1.870000 2.900000 #> Claddefo 25.116325 48.81105 7.599609 0.4096285 33.814545 2.468133 2.823656 #> Cladphyl 50.475000 35.28125 8.568750 0.2812500 7.728125 1.575000 3.231250 ## And the strengths of these variables are: eigengrad(varechem, varespec) #> N P K Ca Mg S Al #> 0.13000842 0.18880078 0.16246365 0.15722067 0.16359171 0.13391967 0.29817406 #> Fe Mn Zn Mo Baresoil Humdepth pH #> 0.20766831 0.27254480 0.16783834 0.09542514 0.20931501 0.25051326 0.14583161"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Weighted classical multidimensional scaling, also known weighted principal coordinates analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"wcmdscale(d, k, eig = FALSE, add = FALSE, x.ret = FALSE, w) # S3 method for wcmdscale plot(x, choices = c(1, 2), type = \"t\", ...) # S3 method for wcmdscale scores(x, choices = NA, tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"d distance structure returned dist full symmetric matrix containing dissimilarities. k dimension space data represented ; must \\(\\{1,2,\\ldots,n-1\\}\\). missing, dimensions zero eigenvalue. eig indicates whether eigenvalues returned. add additive constant \\(c\\) added non-diagonal dissimilarities \\(n-1\\) eigenvalues non-negative. Alternatives \"lingoes\" (default, also used TRUE) \"cailliez\" ( alternative cmdscale). See Legendre & Anderson (1999). x.ret indicates whether doubly centred symmetric distance matrix returned. w Weights points. x wcmdscale result object function called options eig = TRUE x.ret = TRUE (See Details). choices Axes returned; NA returns real axes. type Type graph may \"t\"ext, \"p\"oints \"n\"one. tidy Return scores compatible ggplot2: scores data.frame, score type variable score labelled \"sites\", weights variable weigth, names variable label. ... arguments passed graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Function wcmdscale based function cmdscale (package stats base R), uses point weights. Points high weights stronger influence result low weights. Setting equal weights w = 1 give ordinary multidimensional scaling. default options, function returns matrix scores scaled eigenvalues real axes. function called eig = TRUE x.ret = TRUE, function returns object class \"wcmdscale\" print, plot, scores, eigenvals stressplot methods, described section Value. method Euclidean, non-Euclidean dissimilarities eigenvalues can negative. disturbs , can avoided adding constant non-diagonal dissimilarities making eigenvalues non-negative. function implements methods discussed Legendre & Anderson (1999): method Lingoes (add=\"lingoes\") adds constant \\(c\\) squared dissimilarities \\(d\\) using \\(\\sqrt{d^2 + 2 c}\\) method Cailliez (add=\"cailliez\") dissimilarities using \\(d + c\\). Legendre & Anderson (1999) recommend method Lingoes, base R function cmdscale implements method Cailliez.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"eig = FALSE x.ret = FALSE (default), matrix k columns whose rows give coordinates points corresponding positive eigenvalues. Otherwise, object class wcmdscale containing components mostly similar cmdscale: points matrix k columns whose rows give coordinates points chosen represent dissimilarities. eig \\(n-1\\) eigenvalues computed scaling process eig true. x doubly centred weighted distance matrix x.ret true. ac, add additive constant adjustment method used avoid negative eigenvalues. NA FALSE adjustment done. GOF Goodness fit statistics k axes. first value based sum absolute values eigenvalues, second value based sum positive eigenvalues weights Weights. negaxes matrix scores axes negative eigenvalues scaled absolute eigenvalues similarly points. NULL negative eigenvalues k specified, include negative eigenvalues. call Function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Gower, J. C. (1966) distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecology 69, 1--24. Mardia, K. V., Kent, J. T. Bibby, J. M. (1979). Chapter 14 Multivariate Analysis, London: Academic Press.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"## Correspondence analysis as a weighted principal coordinates ## analysis of Euclidean distances of Chi-square transformed data data(dune) rs <- rowSums(dune)/sum(dune) d <- dist(decostand(dune, \"chi\")) ord <- wcmdscale(d, w = rs, eig = TRUE) ## Ordinary CA ca <- cca(dune) ## IGNORE_RDIFF_BEGIN ## Eigevalues are numerically similar ca$CA$eig - ord$eig #> CA1 CA2 CA3 CA4 CA5 #> 6.661338e-16 4.440892e-16 0.000000e+00 1.942890e-16 4.163336e-16 #> CA6 CA7 CA8 CA9 CA10 #> -2.359224e-16 -2.081668e-16 1.110223e-16 -1.387779e-17 -2.775558e-17 #> CA11 CA12 CA13 CA14 CA15 #> 7.632783e-17 1.387779e-17 1.387779e-17 -3.122502e-17 6.938894e-18 #> CA16 CA17 CA18 CA19 #> -3.295975e-17 -1.214306e-17 -1.647987e-17 6.028164e-17 ## Configurations are similar when site scores are scaled by ## eigenvalues in CA procrustes(ord, ca, choices=1:19, scaling = \"sites\") #> #> Call: #> procrustes(X = ord, Y = ca, choices = 1:19, scaling = \"sites\") #> #> Procrustes sum of squares: #> -4.263e-14 #> ## IGNORE_RDIFF_END plot(procrustes(ord, ca, choices=1:2, scaling=\"sites\")) ## Reconstruction of non-Euclidean distances with negative eigenvalues d <- vegdist(dune) ord <- wcmdscale(d, eig = TRUE) ## Only positive eigenvalues: cor(d, dist(ord$points)) #> [1] 0.9975185 ## Correction with negative eigenvalues: cor(d, sqrt(dist(ord$points)^2 - dist(ord$negaxes)^2)) #> [1] 1"},{"path":[]},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-7","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-7 (in development)","text":"vegan longer suggests tcltk, suggests vegan3d (version 1.3-0). See orditkplot section Deprecated Defunct.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-6-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.6-7 (in development)","text":"set functions add new points existing NMDS ordination metaMDS monoMDS. serves purpose adding new points existing eigenvector ordination (instance, predict.rda). main function MDSaddpoints. needs input rectangular matrix dissimilarities new points (rows) old points (columns). Support function dist2xy can extract needed matrices dissimilarities (old new) points, function designdist2 can directly find needed dissimilarities two data matrices. addition, analogue package can calculate rectangular dissimilarities, including many indices defined designdist2. function still experimental. particular user interface may need development. Comments welcome.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-7","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-7 (in development)","text":"adonis defunct: use adonis2 improved functionality. Disabled use summary get ordination scores: use scores! summary.cca see #644. summary.decorana defunct. nothing useful, can extract information scores weights. orditkplot code removed, try launch vegan3d::orditkplot loudly (deprecation message go vegan3d). experimental, may go direct error, depending user experience: please comment!","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-61","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6.1","title":"vegan 2.6-6.1","text":"CRAN release: 2024-05-21","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6.1","text":"C function do_wcentre (weighted centring) can segfault due protection error. problem found automatic CRAN checks. do_wcentre internal function called envfit (vectorfit), wcmdscale varpart (simpleCCA) Fixes bug #653.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-6","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6","title":"vegan 2.6-6","text":"CRAN release: 2024-05-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-6","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-6","text":"vegan depends R version 4.1.0. possible build vegan webR/wasm Fortran compiler. Issue #623.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-6","text":"Permutation tests CCA completely redesigned follow C.J.F ter Braak & D.E. te Beest: Environ Ecol Stat 29, 849–868 (2022) (https://doi.org/10.1007/s10651-022-00545-4). constraints now re-weighted permuted response data, partial model also residualized conditions (partial terms). vegan (release 2.4-6) tests identical Canoco, ter Braak & te Beest demonstrated results biased. old vegan (release 2.4-2 earlier) predictors re-weighted residualized. Re-weighting sufficient remove bias moderate variation weights, residualizing predictors necessary strongly varying weights. See discussion issue #542. new scheme concerns CCA weighted method, RDA dbRDA permutation unchanged. summary ordination results longer prints ordination scores often voluminous hide real summary; see issue #203. Ordination scores extracted scores function. breaks CRAN packages use summary.cca extract scores. switch use scores. maintainers contacted patch files suggested adapt change. See instructions fix packages. scores function constrained ordination (CCA, RDA,dbRDA) default return types scores (display = \"\"). Function can optionally return single type scores list one matrix instead returning matrix (new argument droplist). Constrained ordination objects (cca, rda, dbrda) fitted without formula interface can permutation tests (anova) \"axis\" \"onedf\". Models \"terms\" \"margin\" possible formula interface. Permutation tests constrained ordination objects (cca, rda, dbrda) = \"axis\" stop permutations later axis cutoff limit reached. Earlier cutoff exceeded. default stop permutations P-value 1 reached. analysis takes care P-values axes non-decreasing similarly Canoco. Coefficients effects prc models scaled similarly scaled vegan pre 2.5-1. change suggested Cajo ter Braak. Handling negative eigenvalues changed summary eigenvals. Negative eigenvalues given negative “explanation”, accumulated proportions add 1 last non-negative eigenvalue, 1 last negative eigenvalue. printed output capscale shows proportions real components ignores imaginary dimensions. consistent summary support methods. Issue #636. RsquareAdj capscale based positive eigenvalues, imaginary components ignored. stressplot.dbrda refuses handle partial models. first component variation can displayed dbrda internal (“working”) data structures additive. unconstrained model \"CA\", constrained \"CCA\" partial none. predict dbrda return actual type = \"working\". Earlier returned \"lc\" scores weighted eigenvalues. generated distances eigenvalues, though.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6","text":"Parallel processing inefficiently implemented slower non-parallel permutation tests constrained ordination adonis2. plot scores cca rda family methods gave error non-existing axes requested. Now ignores requests axes numbers higher result object. summary prc ignored extra parameters (const). -fitted models high number aliased variables caused rare failure adonis2 permutation tests constrained ordination methods (cca, rda, dbrda, capscale) arguments = \"margin\" = \"axis\". also concerned vif.cca intersetcor. Typically occurred high-order interactions factor variables. See issues #452 #622 methods accept rectangular raw data input alternative distances, pass arguments distance functions. arguments vegdist binary = TRUE pseudocount Aitchison distance. concerns dbrda, capscale bioenv. See issue #631 simper gave arbitrary p-values species occur subset. Now given NA. See https://stackoverflow.com/questions/77881877/ Rsquare.adj gave arbitrary p-values -fitted models residual variation. Now returns NA R2 adjusted. Automatic model building proceed cases, fixed ordiR2step returns R2 = 0 overfitted cases. constrained ordination methods issue warning model residual component. See issue #610 inertcomp(..., display = \"sites\", proportional = TRUE) gave wrong values.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"data-sets-2-6-6","dir":"Changelog","previous_headings":"","what":"Data Sets","title":"vegan 2.6-6","text":"Extended description BCI data sets avoid confusion. complete BCI survey includes stems 1 cm DBH, BCI data set vegan subset stems DBH 10 cm published Science 295, 666—669, 2002. data set intended demonstrate methods vegan ecological research suggest contacting BCI team using complete surveys made available Dryad.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-6","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-6","text":"adonis deprecated: use adonis2. several CRAN packages still use adonis although contacted authors June 2022 April 2024, printed message forthcoming deprecation since vegan 2.6-2. See issue #523. See instructions adapt packages functions use adonis2. orditkplot moved CRAN package vegan3d deprecated vegan. See issue #585 announcement #632 use summary extract ordination scores deprecated: use scores extract scores. version still allows extracting scores summary, fail next versions. summary.cca see instructions change package. Support removed ancient cca objects (results cca, rda, dbrda capscale) generated CRAN release 2.5 (2016). still stray relics, use newobject <- update(ancientobject) modernize result. .mcmc.oecosimu .mcmc.permat defunct: use toCoda. Code defunct functions completely removed.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-4","dir":"Changelog","previous_headings":"","what":"vegan 2.6-4","title":"vegan 2.6-4","text":"CRAN release: 2022-10-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-4","text":"Support scores ggplot2 graphics improved extended ordination functions. Suitable scores can requested argument tidy = TRUE, general available types scores returned data frame variable score labelling type. option implemented default method scores structured wcmdscale objects, glitches fixed rda family decorana. Previously tidy scores implemented cca, rda, dbrda family methods, metaMDS, envfit rarecurve. adonis2 anova constrained ordination results can perform sequential test one-degree--freedom effects multi-level factors split contrasts. Previously test available permutest. New summary function varpart brief overview. summary shows unique overall contributed variation set variables. fractions shared several sets variables divided equally contributing sets following Lai J, Zou Y, Zhang J, Peres-Neto P (2022) Methods Ecology Evolution, 13: 782–788. decorana estimates orthogonalized eigenvalues total inertia (scaled Chi-square). Orthogonalized eigenvalues can add total inertia. Together enabled implementing eigenvals, bstick screeplot methods decorana. Axis lengths reported decorana methods. Implemented tolerance method decorana. returns criterion used rescaling DCA, can used inspect success rescaling: constant 1 whole axis. New toCoda function transform sequential null model results oecosimu object can analysed coda convergence independence MCMC model. Function replaces .mcmc.oecosimu .mcmc.permat. metaMDS informative finding similar repeated results random starts uses less confusing language reporting results. Hellinger distance directly available vegdist. vegdist, betadiver raupcrick set attribute maxdist giving numeric value theoretical maximum dissimilarity index. many dissimilarities 1, √2 Chord Hellinger distances, instance. attribute NA open indices ceiling. betadiver three similarity indices set maxdist 0. metaMDS defaults halfchange scaling dissimilarities numeric maxdist attribute, adapt threshold ceiling value. open indices without ceiling, threshold scale dissimilarities. metaMDS used simple test detect index ceiling 1, test now robust can also find maximum values. inference made, function broadcast message assumed value ceiling. Mountford index vegdist now scaled maximum value log(2). Earlier Mountford distances scaled maximum 1. hatvalues constrained ordination objects can sometimes practically 1 1, now cases exactly 1. cases rstandard, rstudent cooks.distance NaN. behaviour similar stats::lm.influence functions. .rad can handle multi-row data frames matrices return list Rank-Abundance data row. Earlier one site handled. decostand returns attribute parameters settings variables used standardization. New function decobackstand can use parameters reconstruct original non-standardized data. Back-transformation exact round-errors, although attempt keep original zeros exact. Back-transformation possible methods pa, rank rrank implemented alr. Back-transformation queried https://stackoverflow.com/questions/73263526/ Rarefaction rarefaction-based methods make sense original observed counts give misleading results data multiplied rare species removed. Observed counts usually singletons (species count one), method issue warning minimum count higher one (may false positive, inspect data). Concerns functions rarefy, drarefy, rrarefy, rarecurve, specaccum(..., method=\"rarefy\"), rareslope avgdist. See github discussion #537. avgdist exposes .dist arguments can return \"dist\"ance objects appear lower triangles instead appearing symmetric matrices. betadisper plots accept col argument (PR #300).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-4","text":"decorana returned wrong results Hill’s piecewise transformation (arguments /) used, unless downweighting also used. scores failed metaMDS result species scores. Bug introduced release 2.6-2. Issue raised https://stackoverflow.com/questions/72483924/ tolerance.cca failed one axis (choice) requested. decostand(..., method=\"alr\") accept name reference, fail cases. CRAN package proxy interfered simper caused obscure error (github issue #528).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-4","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-4","text":"adonis way deprecation. Use adonis2 instead. .mcmc.oecosimu .mcmc.permat deprecated: used S3 methods without depending coda package. Use toCoda instead.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-2","dir":"Changelog","previous_headings":"","what":"vegan 2.6-2","title":"vegan 2.6-2","text":"CRAN release: 2022-04-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-2","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-2","text":"Compiled code adapted changes R 4.2.0. See issues #447, #507. Cross-references function packages adapted stringent tests CRAN","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-2","text":"Aitchison robust Aitchison distances added vegdist. Similar data transformations added decostand. Several functions can return “tidy” data structures can used ggplot2 graphics: rarecurve, scores functions constrained ordination (cca etc.), decorana, envfit, metaMDS. scores.envfit gained argument arrow.mul. vegan plot functions used automatically, now easier use envfit non-vegan plotting. Added function simpson.unb unbiased Simpson diversity robust variation sample sizes. diversity gained argument group calculate indices pooled data. Discussed issue #393. simper much faster even though parallel processing implemented new code. pairs function added plot permustats variables . varpart accepts dissimilarities given symmetric square matrix instead \"dist\" object per wish issue #497. metaMDS adopted user-friendly policy, trymax always maximum number tries. See dicussion https://stackoverflow.com/questions/66748605/. adonis2 accepts strata. adonis2 new main function replaces old adonis. See issue #427. Fisher alpha (fisherfit) badly suited extreme communities follow Fisher’s model. Now fisherfit returns NA communities 0 1 species, issues warning communities consisting singletons extreme Fisher alpha. adipart multipart formulae automatically add unique id constant. always sandwich requested grouping alpha gamma diversities, change results requested groupings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-2","text":"anova function failed marginal tests constrained partial ordination model (cca, rda etc.) interaction terms. Issue #463. Constrained ordination (cca etc.) gave misleading results external variables (constraints, condition) constant explained nothing. decorana fail axes zero eigenvalues. Issue #401. Species accumulation (specaccum) failed one species, several “communities”. Issue #501. Parallel processing failed Windows socket clusters permutest betadisper. Issue #369. orditorp failed numeric labels supplied. Reported https://stackoverflow.com/questions/69272366/. Argument summarize accidentally dropped goodness.cca 2017. taxa2dist failed one usable taxonomic level. See https://stackoverflow.com/questions/67231431/.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-2","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-2","text":"Function adonis2 replace adonis. humpfit functions defunct removed. available non-CRAN package natto https://github.com/jarioksa/natto. commisimulator defunct. Use simulate nullmodel objects. permuted.index finally defunct (deprecated vegan 2.2-0). .mlm defunct. Use functions documented influence.cca, hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca others.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-7","dir":"Changelog","previous_headings":"","what":"vegan 2.5-7","title":"vegan 2.5-7","text":"CRAN release: 2020-11-28","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-7","text":"Several distance-based functions failed distances zero (betadisper, capscale, isomap, monoMDS, pcnm, wcmdscale). Reported github issue #372. Non-linear self-starting regression models SSarrhenius, SSgitay, SSgleason SSlomolino failed future R. failure caused internal changes R-devel. Github issue #382. Arrow labels wrong position plot.envfit(..., add = FALSE). rarecurve added unnecessary names results. Github issue #352. permutest betadisper failed parallel processing Windows systems socket clusters used. Github issue #369.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-7","text":"Chi-square Chord distances added vegdist. distances can calculated Euclidean distances transformed data, actually available earlier, many users notice . monoMDS (hence metaMDS) uses stricter convergence criteria. improves possibilities find stable solutions. However, users may still need tweak convergence criteria data. See discussion Github issue #354. text functions constrained ordination plots (cca, rda, dbrda, capscale) accept now expression labels. allows using subscripts, superscripts mathematical expressions. New support function labels.cca returns current text labels authors can change desired ones. See github issue #374. vegemite returns invisibly final formatted table allowing processing. ordiplot passes cex argument linestack decorana plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-6","dir":"Changelog","previous_headings":"","what":"vegan 2.5-6","title":"vegan 2.5-6","text":"CRAN release: 2019-09-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-6","text":"vegdist silently accepted missing values (NA) removed analysis also option na.rm = FALSE. behaviour introduced vegan version 2.5-1. See GitHub issue #319. labels displaced bunch arrows drawn origin ordination graph envfit. See GitHub issue #315. Hill scale coverscale open-ended limited percent data, unlike traditional cover class scales undefined 100% cover.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-6","text":"results .rad longer print index attribute: attribute still object, printing made output messy.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-5","dir":"Changelog","previous_headings":"","what":"vegan 2.5-5","title":"vegan 2.5-5","text":"CRAN release: 2019-05-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-5","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-5","text":"vegan depends R 3.4.0 higher. next vegan release may increase dependence R 3.6.0. R 3.6.0 improved method find random indices permuting sampling data. Vegan relies now R functions ecological null models (functions nullmodel, oecosimu, commsim, permatfull, permatswap others). Technically change compatible R 3.4.0 later, can gain benefits improved code current release R. null models may change due change, certainly change R 3.6.0. See NEWS R 3.6.0 release discussion github issue #312. vegan permutation routines rely permute, gain similar benefits improved randomness upgrade R. Thanks new R dependence, sigma constrained ordination results works without workarounds vegan 2.5-2. fixes completely issue discussed #274. Vegan test results reproduced older versions R 3.6.0. worried , upgrade R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-5","text":"metaMDS failed scaling results engine monoMDS used. However, recommend use monoMDS. See github issue #310.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-5","text":"betadisper changed interpretation negative squared distances give complex-valued distances. Now regarded zero-distances whereas earlier used modulus. change results cases negative squared distances. discussion, see github issue #306.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-4","dir":"Changelog","previous_headings":"","what":"vegan 2.5-4","title":"vegan 2.5-4","text":"CRAN release: 2019-02-04","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-testing-2-5-4","dir":"Changelog","previous_headings":"","what":"Installation and Testing","title":"vegan 2.5-4","text":"code interpreting formula change R 3.6.0, makes constrained ordination methods (cca, rda, dbrda, capscale) fail. See github issue #299. R 3.6.0 introduces new environment variable _R_CHECK_LENGTH_1_LOGIC2_, several functions fail variable set. Changes concern ordiplot, plot summary constrained ordination objects, ordixyplot. See github issue #305.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-4","text":"decorana gave incorrect results downweighting used (argument iweigh = 1). bug introduced vegan 2.5-1 reported github issue #303. goodness constrained ordination methods failed constraints rank = 1 (one constraining variable). Reported Pierre Legendre.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-4","text":"Adjusted R2 enabled partial RDA models (functions rda dbrda) partial CCA models (function cca) function RsquareAdj. feature disabled vegan 2.5-1 . RDA, calculation similar vegan 2.4-6 earlier. Partial CCA now consistent RDA differs earlier implementation. methods, partial models consistent varpart. See github issue #295.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-3","dir":"Changelog","previous_headings":"","what":"vegan 2.5-3","title":"vegan 2.5-3","text":"CRAN release: 2018-10-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-3","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-3","text":"Tests numerical analysis written robustly give similar results alternative platforms versions R BLAS/Lapack libraries. See github issue #282.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-3","text":"Constrained ordination gave misleading results constraints conditions data NULL variables. rarely happens normal usage, happen marginal anova reported github issue #291. Several functions numerical analysis wrongly accepted non-numeric data (instance, factors) gave either meaningless results confusing error messages. Fixed functions include beals, designdist, diversity, gdispweight, indpower, spantree, specpool, tsallis, tsallisaccum vegdist. See github issue #292. envfit vectors fail missing data. original data scaled centred similarly simulations simulate.rda several simulations returned simmat object (compatible nullmodel simulations can used oecosimu).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-3","text":"anosim checks input avoid confusing error messages like reported StackOverflow question 52082743. Broken-stick distribution (function bstick) longer calculated distance-based Redundancy Analysis (dbrda) negative eigenvalues, clear done. Now dbrda capscale similar respect. print function betadisper results gained new argument neigen select number eigenvalues shown. print robust number eigenvalues lower requested neigen.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-5-3","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.5-3","text":"Function humpfit moved natto package still available https://github.com/jarioksa/natto. scheduled complete removal vegan 2.6-0.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-2","dir":"Changelog","previous_headings":"","what":"vegan 2.5-2","title":"vegan 2.5-2","text":"CRAN release: 2018-05-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-compatibility-2-5-2","dir":"Changelog","previous_headings":"","what":"Installation and Compatibility","title":"vegan 2.5-2","text":"Vegan declares dependence R version 3.2.0. dependence yet noticed previous vegan release. However, generic sigma function defined R-3.3.0, therefore sigma.cca vegan must spelt completely using R-3.2.x. See discussion issue #274. CRAN package klaR function rda, loaded together vegan clashes vegan rda Redundancy Analysis. Vegan tries mitigate problem. cases vegan functions used vegan loaded klaR, error message issued klaR objects handled vegan functions. klaR also tricked print informative message handles vegan objects. However, vegan namespace can attached automatically start-klaR functions take precedence. reported issue #277. Bioconductor package phyloseq problem vegdist function dissimilarities. problem can fixed re-installing phyloseq source package. , must either downgrade vegan version 2.4-6 wait till Bioconductor binary packages upgraded. reported Stackoverflow, vegan issue #272, phyloseq issues #918 #921.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-2","text":"Plotting betadisper failed groups one member. Reported Stackoverflow “Error: Incorrect .dimensions” plotting multivariate data Vegan. Permutation tests constrained ordination (anova.cca, permutest.cca) fail parallel processing socket clusters. Socket clusters always used Windows can also used operating systems created makeCluster. See issue #276.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-1","dir":"Changelog","previous_headings":"","what":"vegan 2.5-1","title":"vegan 2.5-1","text":"CRAN release: 2018-04-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-5-1","dir":"Changelog","previous_headings":"","what":"GENERAL","title":"vegan 2.5-1","text":"major new release changes package: Nearly 40% program files changed previous release. Please report regressions issues https://github.com/vegandevs/vegan/issues/. Compiled code used much extensively, compiled functions use .Call interface. gives smaller memory footprint also faster. wall clock time, greatest gains permutation tests constrained ordination methods (anova.cca) binary null models (nullmodel). Constrained ordination functions (cca, rda, dbrda, capscale) completely rewritten share code. makes consistent robust. internal structure changed constrained ordination objects, scripts may fail try access result object directly. never guarantee unchanged internal structure, scripts changed use provided support functions access result object (see documentation cca.object github issue #262). support analysis functions may longer work result objects created previous vegan versions. use update(old.result.object) fix old result objects. See github issues #218, #227. vegan includes tests run checking package installation. See github issues #181, #271. informative messages (warnings, notes error messages) cleaned unified also makes possible provide translations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.5-1","text":"avgdist: new function find averaged dissimilarities several random rarefactions communities. Code Geoffrey Hannigan. See github issues #242, #243, #246. chaodist: new function similar designdist, uses Chao terms supposed take account effects unseen species (Chao et al., Ecology Letters 8, 148-159; 2005). Earlier Jaccard-type Chao dissimilarity vegdist, new code allows defining kind Chao dissimilarity. New functions find influence statistics constrained ordination objects: hatvalues, sigma, rstandard, rstudent, cooks.distance, SSD, vcov, df.residual. earlier found via .mlm function deprecated. See github issue #234. boxplot added permustats results display (standardized) effect sizes. sppscores: new function add replace species scores distance-based ordination dbrda, capscale metaMDS. Earlier dbrda species scores, species scores capscale metaMDS based raw input data may consistent used dissimilarity measure. See github issue #254. cutreeord: new function similar stats::cutree, numbers cluster order appear dendrogram (left right) instead labelling order appeared data. sipoo.map: new data set locations sizes islands Sipoo archipelago bird data set sipoo.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-constrained-ordination-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Constrained Ordination","title":"vegan 2.5-1","text":"inertia Correspondence Analysis (cca) called “scaled Chi-square” instead using name little known statistic. elements Constraints Conditions data frames non-formula call rda cca, automatically expanded model matrices can contain factor variables. Earlier numerical model matrices factors used formula interface. Regression scores constraints can extracted plotted constrained ordination methods. See github issue #226. Full model (model = \"full\") enabled permutations tests constrained ordination results anova.cca permutest.cca. permutest.cca gained new option = \"onedf\" perform tests sequential one degree--freedom contrasts factors. option (yet) enabled anova.cca. permutation tests robust, scoping issues fixed. Permutation tests use compiled C code much faster. See github issue #211. permutest printed layout similar anova.cca. eigenvals gained new argument (model) select either constrained unconstrained scores. old argument (constrained) deprecated. See github issue #207. summary.eigenvals returns matrix instead list containing matrix. Adjusted R2 calculated partial ordination, unclear done (function RsquareAdj). ordiresids can display standardized studentized residuals. Function construct model.frame model.matrix constrained ordination robust fail fewer cases. goodness inertcomp constrained ordination result object longer option find distances: explained variation available. inertcomp gained argument unity. give “local contributions beta-diversity” (LCBD) “species contribution beta-diversity” (SCBD) Legendre & De Cáceres (Ecology Letters 16, 951-963; 2012). goodness disabled capscale. prc gained argument const general scaling results similarly rda. prc uses regression scores Canoco-compatibility.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-null-model-communities-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Null Model Communities","title":"vegan 2.5-1","text":"C code swap-based binary null models made efficients, models faster. Many models selected 2 times 2 submatrix, generated four random numbers (two rows, two columns). Now skip selecting third fourth random number obvious matrix swapped. Since time used generating random numbers functions, candidates rejected, speeds functions. However, also means random number sequences change previous vegan versions, old binary model results replicated exactly. See github issues #197, #255 details timing. Ecological null models (nullmodel, simulate, make.commsim, oecosimu) gained new null model \"greedyqswap\" can radically speed quasi-swap models minimal risk introducing bias. Backtracking written C much faster. However, backtracking models biased, provided classic legacy models.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-other-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Other Functions","title":"vegan 2.5-1","text":"adonis2 gained column R2 similarly old adonis. Great part R code decorana written C makes faster reduces memory footprint. metaMDS results gained new points text methods. ordiplot ordination plot functions can chained points text functions allowing use magrittr pipes. points text functions gained argument draw arrows allowing use drawing biplots adding vectors environmental variables ordiplot. Since many ordination plot methods return invisible \"ordiplot\" object, points text methods also work . See github issue #257. Lattice graphics (ordixyplot) ordination can add polygons enclose points panel complete data. ordicluster gained option suppress drawing plots can easily embedded functions calculations. .rad returns index included taxa attribute. Random rarefaction (function rrarefy) uses compiled C code much faster. plot specaccum can draw short horizontal bars vertical error bars. See StackOverflow question 45378751. decostand gained new standardization methods rank rrank replace abundance values ranks relative ranks. See github issue #225. Clark dissimilarity added vegdist (calculated designdist). designdist evaluates minimum terms compiled code, function faster vegdist also dissimilarities using minimum terms. Although designdist usually faster vegdist, numerically less stable, particular large data sets. swan passes type argument beals. tabasco can use traditional cover scale values function coverscale. Function coverscale can return scaled values integers numerical analysis instead returning characters printing. varpart can partition Chi-squared inertia correspondence analysis new argument chisquare. adjusted R2 based permutation tests, replicate analysis random variation. explanatory tables can data frames factors single factors varpart automatically expanded model matrices. Earlier factors used one-sided model formulae. Based code suggested Daniel Borcard, Univ. Montréal.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-1","text":"long Condition() statements (> 500 characters) failed partial constrained ordination models (cca, rda, dbrda, capscale). problem detected StackOverflow question 49249816. Labels adjusted arrows rescaled envfit plots. See StackOverflow question 49259747. ordiArrowMul failed one arrow plotted envfit.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-5-1","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.5-1","text":".mlm function constrained correspondence analysis deprecated favour new functions directly give influence statistics. See github issue #234. commsimulator now defunct: use simulate nullmodel objects. ade4 cca objects longer handled vegan: ade4 cca since version 1.7-8 (August 9, 2017).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-6","dir":"Changelog","previous_headings":"","what":"vegan 2.4-6","title":"vegan 2.4-6","text":"CRAN release: 2018-01-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-6","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-6","text":"CRAN packages longer allowed use FORTRAN input, read.cep function used FORTRAN format read legacy CEP Canoco files. avoid NOTEs WARNINGs, function re-written R. new read.cep less powerful fragile, can read data “condensed” format, can fail several cases successful old code. old FORTRAN-based function still available cepreader. See github issue #263. cepreader package developed https://github.com/vegandevs/cepreader.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-6","text":"functions rarefaction (rrarefy), species abundance distribution (preston) species pool (estimateR) need exact integer data, test allowed small fuzz. functions worked correctly original data, data transformed back-transformed, pass integer test fuzz give wrong results. instance, sqrt(3)^2 pass test 3, interpreted strictly integer 2. See github issue #259.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-6","text":"ordiresids uses now weighted residuals cca results.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-5","dir":"Changelog","previous_headings":"","what":"vegan 2.4-5","title":"vegan 2.4-5","text":"CRAN release: 2017-12-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-5","text":"Several “Swap & Shuffle” null models generated wrong number initial matrices. Usually generated many, dangerous, slow. However, random sequences change fix. Lattice graphics ordination (ordixyplot friends) colour arrows groups instead randomly mixed colours. Information constant mirrored permutations omitted reporting permutation tests (e.g., anova constrained ordination).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-5","text":"ordistep improved interpretation scope: lower scope missing, formula starting solution taken lower scope instead using empty model. See Stackoverflow question 46985029. fitspecaccum gained new support functions nobs logLik allow better co-operation packages functions. See GitHub issue #250. “backtracking” null model community simulation faster. However, “backtracking” biased legacy model used except comparative studies.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-4","dir":"Changelog","previous_headings":"","what":"vegan 2.4-4","title":"vegan 2.4-4","text":"CRAN release: 2017-08-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-4","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-4","text":"orditkplot longer give warnings CRAN tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-4","text":"anova(..., = \"axis\") constrained ordination (cca, rda, dbrda) ignored partial terms Condition(). inertcomp summary.cca failed constrained component defined, explained nothing zero rank. See StackOverflow: R - Error message RDA analysis - vegan package. Labels longer cropped meandist plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-4","text":"significance tests axes constrained ordination use now forward testing strategy. extensive analysis indicated previous marginal tests biased. conflict Legendre, Oksanen & ter Braak, Methods Ecol Evol 2, 269–277 (2011) regarded marginal tests unbiased. Canberra distance vegdist can now handle negative input entries similarly latest versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-3","dir":"Changelog","previous_headings":"","what":"vegan 2.4-3","title":"vegan 2.4-3","text":"CRAN release: 2017-04-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-3","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-3","text":"vegan registers native C Fortran routines. avoids warnings model checking, may also give small gain speed. Future versions vegan deprecate remove elements pCCA$Fit, CCA$Xbar, CA$Xbar cca result objects. release provides new function ordiYbar able construct elements current future releases. Scripts functions directly accessing elements switch ordiYbar smooth transition.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-3","text":".mlm methods constrained ordination include zero intercept give correct residual degrees freedom derived statistics. biplot method rda passes correlation argument scaling algorithm. Biplot scores wrongly centred cca caused small error values. Weighting centring corrected intersetcor spenvcor. fix can make small difference analysing cca results. Partial models correctly handled intersetcor. envfit ordisurf functions failed applied species scores. Non-standard variable names can used within Condition() partial ordination. Partial models used internally within several functions, problem reported Albin Meyer (Univ Lorraine, Metz, France) ordiR2step using variable name contained hyphen (wrongly interpreted minus sign partial ordination). ordispider pass graphical arguments used show difference LC WA scores constrained ordination. ordiR2step uses forward selection avoid several problems model evaluation. tolerance function return NaN cases returned 0. Partial models correctly analysed. Misleading (non-zero) tolerances sometimes given species occurred sampling units one species.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-2","dir":"Changelog","previous_headings":"","what":"vegan 2.4-2","title":"vegan 2.4-2","text":"CRAN release: 2017-01-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-2","text":"Permutation tests (permutests, anova) first axis failed constrained distance-based ordination (dbrda, capscale). Now capscale also throw away negative eigenvalues first eigenvalues tested. permutation tests first axis now faster. problem reported Cleo Tebby fixes discussed GitHub issue #198 pull request #199. support functions dbrda capscale gave results components wrong scale. Fixes stressplot, simulate, predict fitted functions. intersetcor use correct weighting cca results slightly . anova permutest failed betadisper fitted argument bias.adjust = TRUE. Fixes Github issue #219 reported Ross Cunning, O’ahu, Hawaii. ordicluster return invisibly coordinates internal points (clusters points joined), last rows contained coordinates external points (ordination scores points). cca method tolerance returning incorrect values second axis sample heterogeneities species tolerances. See issue #216 details.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-2","text":"Biplot scores scaled similarly site scores constrained ordination methods cca, rda, capscale dbrda. Earlier unscaled (technically, equal scaling axes). tabasco adds argument scale colours rows columns addition old equal scale whole plot. New arguments labRow labCex can used change column row labels. Function also takes care -zero observations coloured: earlier tiny observed values merged zeros distinct plots. Sequential null models somewhat faster (10%). Non-sequential null models may marginally faster. null models generated function nullmodel also used oecosimu. vegdist much faster. used clearly slower stats::dist, now nearly equally fast dissimilarity measure. Handling data= formula interface robust, messages user errors improved. fixes points raised Github issue #200. families orders dune.taxon updated APG IV (Bot J Linnean Soc 181, 1–20; 2016) corresponding classification higher levels (Chase & Reveal, Bot J Linnean Soc 161, 122-127; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-1","dir":"Changelog","previous_headings":"","what":"vegan 2.4-1","title":"vegan 2.4-1","text":"CRAN release: 2016-09-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-4-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.4-1","text":"Fortran code modernized avoid warnings latest R. modernization visible effect functions. Please report suspect cases vegan issues.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-1","text":"Several support functions ordination methods failed solution one ordination axis, instance, one constraining variable CCA, RDA friends. concerned goodness constrained ordination, inertcomp, fitted capscale, stressplot RDA, CCA (GitHub issue #189). goodness CCA & friends ignored choices argument (GitHub issue #190). goodness function consider negative eigenvalues db-RDA (function dbrda). Function meandist failed cases one groups one observation. linestack handle expressions labels. regression discussed GitHub issue #195. Nestedness measures nestedbetajac nestedbetasor expecting binary data cope quantitative input evaluating Baselga’s matrix-wide Jaccard Sørensen dissimilarity indices. Function .mcmc cast oecosimu result MCMC object (coda package) failed one chain.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-1","text":"diversity function returns now NA observation NA values instead returning 0. function also checks input refuses handle data negative values. GitHub issue #187. rarefy function work robustly marginal case user asks one individual can one species zero variance. Several functions robust factor arguments contain missing values (NA): betadisper, adipart, multipart, hiersimu, envfit constrained ordination methods cca, rda, capscale dbrda. GitHub issues #192 #193.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-0","dir":"Changelog","previous_headings":"","what":"vegan 2.4-0","title":"vegan 2.4-0","text":"CRAN release: 2016-06-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"distance-based-analysis-2-4-0","dir":"Changelog","previous_headings":"","what":"Distance-based Analysis","title":"vegan 2.4-0","text":"Distance-based methods redesigned made consistent ordination (capscale, new dbrda), permutational ANOVA (adonis, new adonis2), multivariate dispersion (betadisper) variation partitioning (varpart). methods can produce negative eigenvalues several popular semimetric dissimilarity indices, handled similarly functions. Now functions designed McArdle & Anderson (Ecology 82, 290–297; 2001). dbrda new function distance-based Redundancy Analysis following McArdle & Anderson (Ecology 82, 290–297; 2001). metric dissimilarities, function equivalent old capscale, negative eigenvalues semimetric indices handled differently. dbrda dissimilarities decomposed directly conditions, constraints residuals negative eigenvalues, components can imaginary dimensions. Function mostly compatible capscale constrained ordination methods, full compatibility achieved (see issue #140 Github). function based code Pierre Legendre. old capscale function constrained ordination still based real components, total inertia components assessed similarly dbrda. significance tests differ previous version, function oldCapscale cast capscale result similar form previously. adonis2 new function permutational ANOVA dissimilarities. based algorithm dbrda. function can perform overall tests independent variables well sequential marginal tests term. old adonis still available, can perform sequential tests. settings, adonis adonis2 give identical results (see Github issue #156 differences). Function varpart can partition dissimilarities using algorithm dbrda. Argument sqrt.dist takes square roots dissimilarities can change many popular semimetric indices metric distances capscale, dbrda, wcmdscale, adonis2, varpart betadisper (issue #179 Github). Lingoes Cailliez adjustments change dissimilarity metric distance capscale, dbrda, adonis2, varpart, betadisper wcmdscale. Earlier Cailliez adjustment capscale (issue #179 Github). RsquareAdj works capscale dbrda allows using ordiR2step model building.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-0","text":"specaccum: plot failed line type (lty) given. Reported Lila Nath Sharma (Univ Bergen, Norway)","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-4-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.4-0","text":"ordibar new function draw crosses standard deviations standard errors ordination diagrams instead corresponding ellipses. Several permustats results can combined new c() function. New function smbind binds together null models row, column replication. sequential models bound together, can treated parallel chains subsequent analysis (e.g., .mcmc). See issue #164 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-0","text":"Null model analysis upgraded: New \"curveball\" algorithm provides fast null model fixed row column sums binary matrices Strona et al. (Nature Commun. 5: 4114; 2014). \"quasiswap\" algorithm gained argument thin can reduce bias null models. \"backtracking\" now much faster, still slow, provided mainly allow comparison better faster methods. Compiled code can now interrupted null model simulations. designdist can now use beta diversity notation (gamma, alpha) easier definition beta diversity indices. metaMDS new iteration strategy: Argument try gives minimum number random starts, trymax maximum number. Earlier hand try gave maximum number, now run least try times. reduces risk trapped local optimum (issue #154 Github). convergent solutions, metaMDS now tabulate stopping criteria (trace = TRUE). can help deciding criteria made stringent number iterations increased. documentation monoMDS metaMDS give detailed information convergence criteria. summary permustats prints now P-values, test direction (alternative) can changed. qqmath function permustats can now plot standardized statistics. partial solution issue #172 Github. MDSrotate can rotate ordination show maximum separation factor levels (classes) using linear discriminant analysis (lda MASS package). adipart, hiersimu multipart expose argument method specify null model. RsquareAdj works cca allows using ordiR2step model building. code developed Dan McGlinn (issue #161 Github). However, cca still used varpart. ordiellipse ordihull allow setting colours, line types graphical parameters. alpha channel can now given also real number 0 … 1 addition integer 0 … 255. ordiellipse can now draw ellipsoid hulls enclose points group. ordicluster, ordisegments, ordispider lines plot functions isomap spantree can use mixture colours connected points. behaviour similar analogous functions vegan3d package. plot betadisper configurable. See issues #128 #166 Github details. text points methods orditkplot respect stored graphical parameters. Environmental data Barro Colorado Island forest plots gained new variables Harms et al. (J. Ecol. 89, 947–959; 2001). Issue #178 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-4-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.4-0","text":"Function metaMDSrotate removed replaced MDSrotate. density densityplot methods various vegan objects deprecated replaced density densityplot permustats. Function permustats can extract permutation simulation results vegan result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-5","dir":"Changelog","previous_headings":"","what":"vegan 2.3-5","title":"vegan 2.3-5","text":"CRAN release: 2016-04-09","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-5","text":"eigenvals fails prcomp results R-devel. next version prcomp argument limit number eigenvalues shown (rank.), breaks eigenvals vegan. calibrate failed cca friends rank given.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-4","dir":"Changelog","previous_headings":"","what":"vegan 2.3-4","title":"vegan 2.3-4","text":"CRAN release: 2016-02-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-4","text":"betadiver index 19 wrong sign one terms. linestack failed labels given, input scores names. Reported Jeff Wood (ANU, Canberra, ACT).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-3-4","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.3-4","text":"vegandocs deprecated. Current R provides better tools seeing extra documentation (news() browseVignettes()).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vignettes-2-3-4","dir":"Changelog","previous_headings":"","what":"Vignettes","title":"vegan 2.3-4","text":"vignettes built standard R tools can browsed browseVignettes. FAQ-vegan partitioning accessible vegandocs function.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"building-2-3-4","dir":"Changelog","previous_headings":"","what":"Building","title":"vegan 2.3-4","text":"Dependence external software texi2dvi removed. Version 6.1 texi2dvi incompatible R prevented building vegan. FAQ-vegan earlier built texi2dvi uses now knitr. , vegan now dependent R-3.0.0. Fixes issue #158 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-3","dir":"Changelog","previous_headings":"","what":"vegan 2.3-3","title":"vegan 2.3-3","text":"CRAN release: 2016-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-3","text":"metaMDS monoMDS fail input dissimilarities huge: reported case magnitude 1E85. Fixes issue #152 Github. Permutations failed defined permute control structures estaccum, ordiareatest, renyiaccum tsallisaccum. Reported Dan Gafta (Cluj-Napoca) renyiaccum. rarefy gave false warnings input vector single sampling unit. extrapolated richness indices specpool needed number doubletons (= number species occurring two sampling units), failed one sampling unit supplied. extrapolated richness estimated single sampling unit, now cases handled smoothly instead failing: observed non-extrapolated richness zero standard error reported. issue reported StackOverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-3","text":"treedist treedive refuse handle trees reversals, .e, higher levels homogeneous lower levels. Function treeheight estimate total height absolute values branch lengths. Function treedive refuses handle trees negative branch heights indicating negative dissimilarities. Function treedive faster. gdispweight works input data matrix instead data frame. Input dissimilarities supplied symmetric matrices data frames robustly recognized anosim, bioenv mrpp.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-2","dir":"Changelog","previous_headings":"","what":"vegan 2.3-2","title":"vegan 2.3-2","text":"CRAN release: 2015-11-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-2","text":"Printing details gridded permutation design fail grid within-plot level. ordicluster joined branches wrong coordinates cases. ordiellipse ignored weights calculating standard errors (kind = \"se\"). influenced plots cca, also influenced ordiareatest.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-2","text":"adonis capscale functions recognize symmetric square matrices dissimilarities. Formerly dissimilarities given \"dist\" objects produced dist vegdist functions, data frames matrices regarded observations x variables data confuse users (e.g., issue #147). mso accepts \"dist\" objects distances among locations alternative coordinates locations. text, points lines functions procrustes analysis gained new argument truemean allows adding procrustes items plots original analysis. rrarefy returns observed non-rarefied communities (warning) users request subsamples larger observed community instead failing. Function drarefy similar returned sampling probabilities 1, now also issues warning. Fixes issue #144 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-1","dir":"Changelog","previous_headings":"","what":"vegan 2.3-1","title":"vegan 2.3-1","text":"CRAN release: 2015-09-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-1","text":"Permutation tests always correctly recognize ties observed statistic result low P-values. happen particular predictor variables factors (classes). changes concern functions adonis, anosim, anova permutest functions cca, rda capscale, permutest betadisper, envfit, mantel mantel.partial, mrpp, mso, oecosimu, ordiareatest, protest simper. also fixes issues #120 #132 GitHub. Automated model building constrained ordination (cca, rda, capscale) step, ordistep ordiR2step fail aliased candidate variables, constraints completely explained variables already model. regression introduced vegan 2.2-0. Constrained ordination methods cca, rda capscale treat character variables factors analysis, return centroids plotting. Recovery original data metaMDS computing WA scores species fail expression supplied argument comm long & got deparsed multiple strings. metaMDSdist now returns (possibly modified) data frame community data comm attribute \"comm\" returned dist object. metaMDS now uses compute WA species scores NMDS. addition, deparsed expression comm now robust long expressions. Reported Richard Telford. metaMDS monoMDS rejected dissimilarities missing values. Function rarecurve check input cause confusing error messages. Now function checks input data integers can interpreted counts individuals sampling units species. Unchecked bad inputs reason problems reported Stackoverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-1","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-1","text":"Scaling ordination axes cca, rda capscale can now expressed descriptive strings \"none\", \"sites\", \"species\" \"symmetric\" tell kind scores scaled eigenvalues. can modified arguments hill cca correlation rda. old numeric scaling can still used. permutation data can extracted anova results constrained ordination (cca, rda, capscale) analysed permustats function. New data set BCI.env site information Barro Colorado Island tree community data. useful variables UTM coordinates sample plots. variables constant nearly constant little use normal analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-0","dir":"Changelog","previous_headings":"","what":"vegan 2.3-0","title":"vegan 2.3-0","text":"CRAN release: 2015-05-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-0","text":"Constrained ordination functions cca, rda capscale now robust. Scoping data set names variable names much improved. fix numerous long-standing problems, instance reported Benedicte Bachelot (email) Richard Telford (Twitter), well issues #16 #100 GitHub. Ordination functions cca rda silently accepted dissimilarities input although analysis makes sense methods. Dissimilarities analysed distance-based redundancy analysis (capscale). variance conditional component -estimated goodness rda results, results wrong partial RDA. problems reported R-sig-ecology message Christoph von Redwitz.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"windows-2-3-0","dir":"Changelog","previous_headings":"","what":"Windows","title":"vegan 2.3-0","text":"orditkplot add file type identifier saved graphics Windows although required. problem concerned Windows OS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-0","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-0","text":"goodness function constrained ordination (cca, rda, capscale) redesigned. Function gained argument addprevious add variation explained previous ordination components axes statistic = \"explained\". option, model = \"CCA\" include variation explained partialled-conditions, model = \"CA\" include accumulated variation explained conditions constraints. former behaviour addprevious = TRUE model = \"CCA\", addprevious = FALSE model = \"CA\". argument effect statistic = \"distance\", always show residual distance previous components. Formerly displayed residual distance currently analysed model. Functions ordiArrowMul ordiArrowTextXY exported can used normal interactive sessions. functions used scale bunch arrows fit ordination graphics, formerly internal functions used within vegan functions. orditkplot can export graphics SVG format. SVG vector graphics format can edited several external programs, Illustrator Inkscape. Rarefaction curve (rarecurve) species accumulation models (specaccum, fitspecaccum) gained new functions estimate slope curve given location. Originally based response R-SIG-ecology query. rarefaction curves, function rareslope, species accumulation models specslope. functions based analytic equations, can also evaluated interpolated non-integer values. specaccum models functions can evaluated analytic models \"exact\", \"rarefaction\" \"coleman\". \"random\" \"collector\" methods can use finite differences (diff(fitted())). Analytic functions slope used non-linear regression models known fitspecaccum. Species accumulation models (specaccum) non-liner regression models species accumulation (fitspecaccum) work consistently weights. cases, models defined using number sites independent variable, weights means observations can non-integer numbers virtual sites. predict models also use number sites newdata, analytic models can estimate expected values non-integer number sites, non-analytic randomized collector models can interpolate non-integer values. fitspecaccum gained support functions AIC deviance. varpart plots four-component models redesigned following Legendre, Borcard & Roberts Ecology 93, 1234–1240 (2012), use now four ellipses instead three circles two rectangles. components now labelled plots, circles ellipses can easily filled transparent background colour.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-1","dir":"Changelog","previous_headings":"","what":"vegan 2.2-1","title":"vegan 2.2-1","text":"CRAN release: 2015-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-1","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-1","text":"maintenance release avoid warning messages caused changes CRAN repository. namespace usage also stringent avoid warnings notes development versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-1","text":"vegan can installed loaded without tcltk package. tcltk package needed orditkplot function interactive editing ordination graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-1","text":"ordisurf failed gam package loaded due namespace issues: support functions gam used instead mgcv functions. tolerance function failed unconstrained correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-1","text":"estimateR uses exact variance formula bias-corrected Chao estimate extrapolated number species. new formula may unpublished, derived following guidelines Chiu, Wang, Walther & Chao, Biometrics 70, 671–682 (2014), doi:10.1111/biom.12200, online supplementary material. Diversity accumulation functions specaccum, renyiaccum, tsallisaccum, poolaccum estaccumR use now permute package permutations order sampling sites. Normally functions need simple random permutation sites, restricted permutation permute package user-supplied permutation matrices can used. estaccumR function can use parallel processing. linestack accepts now expressions labels. allows using mathematical symbols formula given mathematical expressions.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-0","dir":"Changelog","previous_headings":"","what":"vegan 2.2-0","title":"vegan 2.2-0","text":"CRAN release: 2014-11-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-0","text":"Several vegan functions can now use parallel processing slow repeating calculations. functions argument parallel. argument can integer giving number parallel processes. unix-alikes (Mac OS, Linux) launch \"multicore\" processing Windows set \"snow\" clusters desribed documentation parallel package. option \"mc.cores\" set integer > 1, used automatically start parallel processing. Finally, argument can also previously set \"snow\" cluster used Windows unix-alikes. Vegan vignette Design decision explains implementation (use vegandocs(\"decission\"), parallel package extensive documentation parallel processing R. following function use parallel processing analysing permutation statistics: adonis, anosim, anova.cca (permutest.cca), mantel (mantel.partial), mrpp, ordiareatest, permutest.betadisper simper. addition, bioenv can compare several candidate sets models paralle, metaMDS can launch several random starts parallel, oecosimu can evaluate test statistics several null models parallel. permutation tests based permute package offers strong tools restricted permutation. functions argument permutations. default usage simple non-restricted permutations achieved giving single integer number. Restricted permutations can defined using function permute package. Finally, argument can permutation matrix rows define permutations. possible use external user constructed permutations. See help(permutations) brief introduction permutations vegan, permute package full documention. vignette permute package can read vegan command vegandocs(\"permutations\"). following functions use permute package: CCorA, adonis, anosim, anova.cca (plus associated permutest.cca, add1.cca, drop1.cca, ordistep, ordiR2step), envfit (plus associated factorfit vectorfit), mantel (mantel.partial), mrpp, mso, ordiareatest, permutest.betadisper, protest simper. Community null model generation completely redesigned rewritten. communities constructed new nullmodel function defined low level commsim function. actual null models generated simulate function builds array null models. new null models include wide array quantitative models addition old binary models, users can plug generating functions. basic tool invoking analysing null models oecosimu. null models often used analysis nestedness, implementation oecosimu allows analysing statistic, null models better seen alternative permutation tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-0","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-0","text":"vegan package dependencies namespace imports adapted changes R, trigger warnings notes package tests. Three-dimensional ordination graphics using scatterplot3d static plots rgl dynamic plots removed vegan moved companion package vegan3d. package available CRAN.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-2-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.2-0","text":"Function dispweight implements dispersion weighting Clarke et al. (Marine Ecology Progress Series, 320, 11–27). addition, implemented new method generalized dispersion weighting gdispweight. methods downweight species significantly -dispersed. New hclust support functions reorder, rev scores. Functions reorder rev similar functions dendrogram objects base R. However, reorder can use (defaults ) weighted mean. weighted mean node average always mean member leaves, whereas dendrogram uses always unweighted means joined branches. Function ordiareatest supplements ordihull ordiellipse provides randomization test one-sided alternative hypothesis convex hulls ellipses two-dimensional ordination space smaller areas randomized groups. Function permustats extracts inspects permutation results support functions summary, density, densityplot, qqnorm qqmath. density qqnorm standard R tools work one statistic, densityplot qqmath lattice graphics work univariate multivariate statistics. results following functions can extracted: anosim, adonis, mantel (mantel.partial), mrpp, oecosimu, permustest.cca (corresponding anova methods), permutest.betadisper, protest. stressplot functions display ordination distances given number dimensions original distances. method functins similar stressplot metaMDS, always use inherent distances ordination method. functions available results capscale, cca, princomp, prcomp, rda, wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-0","text":"cascadeKM one group NA instead random value. ordiellipse can handle points exactly line, including two points (warning). plotting radfit results several species failed communities species one species. RsquareAdj capscale negative eigenvalues now report NA instead using biased method rda results. simper failed group single member.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-0","text":"anova.cca functions re-written use permute package. Old results may exactly reproduced, models missing data may fail several cases. new option analysing sequence models . simulate functions cca rda can return several simulations nullmodel compatible object. functions can produce simulations correlated errors (also capscale) parametric simulation Gaussian error. bioenv can use Manhattan, Gower Mahalanobis distances addition default Euclidean. New helper function bioenvdist can extract dissimilarities applied best model model. metaMDS(..., trace = 2) show convergence information default monoMDS engine. Function MDSrotate can rotate k-dimensional ordination k-1 variables. variables correlated (like usually case), vectors can also correlated previously rotated dimensions, uncorrelated later ones. vegan 2.0-10 changed weighted nestednodf weighted analysis binary data equivalent binary analysis. However, broke equivalence original method. Now function argument wbinary select method analysis. problem reported fix submitted Vanderlei Debastiani (Universidade Federal Rio Grande Sul, Brasil). ordiellipse, ordihull ordiellipse can handle missing values groups. ordispider can now use spatial medians instead means. rankindex can use Manhattan, Gower Mahalanobis distance addition default Euclidean. User can set colours line types function rarecurve plotting rarefaction curves. spantree gained support function .hclust change minimum spanning tree hclust tree. fitspecaccum can weighted analysis. Gained lines method. Functions extrapolated number species size species pool using Chao method modified following Chiu et al., Biometrics 70, 671–682 (2014). Incidence based specpool can now use (defaults ) small sample correction number sites sample size. Function uses basic Chao extrapolation based ratio singletons doubletons, switches now bias corrected Chao extrapolation doubletons (species found twice). variance formula bias corrected Chao derived following supporting line material doi:10.1111/biom.12200 differs slightly Chiu et al. (2014). poolaccum function changed similarly, small sample correction used always. abundance based estimateR uses bias corrected Chao extrapolation, earlier estimated variance classic Chao model. Now use widespread approximate estimate EstimateS variance. changes functions similar EstimateS tabasco uses now reorder.hclust hclust object better ordering previously cast trees dendrogram objects. treedive treedist default now match.force = TRUE can silenced verbose = FALSE. vegdist gained Mahalanobis distance. Nomenclature updated plant community data help Taxonstand taxize packages. taxonomy dune data adapted sources APG III. varespec dune use 8-character names (4 genus + 4 species epithet). New data set phylogenetic distances dune extracted Zanne et al. (Nature 506, 89–92; 2014). User configurable plots rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-2-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.2-0","text":"strata deprecated permutations. still accepted phased next releases. Use permute package. cca, rda capscale return scores scaled eigenvalues: use scores function extract scaled results. commsimulator deprecated. Replace commsimulator(x, method) simulate(nullmodel(x, method)). density densityplot permutation results deprecated: use permustats density densityplot method.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-10","dir":"Changelog","previous_headings":"","what":"vegan 2.0-10","title":"vegan 2.0-10","text":"CRAN release: 2013-12-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-10","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-10","text":"version adapted changes permute package version 0.8-0 triggers NOTEs package checks. release may last 2.0 series, next vegan release scheduled major release newly designed oecosimu community pattern simulation, support parallel processing, full support permute package. interested developments, may try development versions vegan GitHub report problems user experience us.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-10","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-10","text":"envfit function assumed external variables either numeric factors, failed , say, character strings. Now numeric variables taken continuous vectors, variables (character strings, logical) coerced factors possible. function also work degenerate data, like one level factor constant value continuous environmental variable. ties wrongly assessing permutation P-values vectorfit. nestednodf quantitative data consistent binary models, fill wrongly calculated quantitative data. oecosimu now correctly adapts displayed quantiles simulated values alternative test direction. renyiaccum plotting failed one level diversity scale used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-10","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-10","text":"Kempton Taylor algorithm found unreliable fisherfit fisher.alpha, now estimation Fisher α based number species number individuals. estimation standard errors profile confidence intervals also scrapped. renyiaccum, specaccum tsallisaccum functions gained subset argument. renyiaccum can now add collector curve analysis. collector curve diversity accumulation order sampling units. interesting ordering sampling units allows comparing actual species accumulations expected randomized accumulation. specaccum can now perform weighted accumulation using sampling effort weights.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-9","dir":"Changelog","previous_headings":"","what":"vegan 2.0-9","title":"vegan 2.0-9","text":"CRAN release: 2013-09-25 version released due changes programming interface testing procedures R 3.0.2. using older version R, need upgrade vegan. new features bug fixes. user-visible changes documentation output messages formatting. R changes, version dependent R version 2.14.0 newer lattice package.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-8","dir":"Changelog","previous_headings":"","what":"vegan 2.0-8","title":"vegan 2.0-8","text":"CRAN release: 2013-07-10","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-8","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-8","text":"maintenance release fixes issues raised changed R toolset processing vignettes. also fix typographic issues vignettes.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-8","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-8","text":"ordisurf gained new arguments flexible definition fitted models better utilize mgcv::gam function. linewidth contours can now set argument lwd. Labels arrows positioned better way plot functions results envfit, cca, rda capscale. labels longer overlap arrow tips. setting test direction clearer oecosimu. ordipointlabel gained plot method can used replot saved result.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-7","dir":"Changelog","previous_headings":"","what":"vegan 2.0-7","title":"vegan 2.0-7","text":"CRAN release: 2013-03-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-7","text":"tabasco() new function graphical display community data matrix. Technically interface R heatmap, use closer vegan function vegemite. function can reorder community data matrix similarly vegemite, instance, ordination results. Unlike heatmap, displays dendrograms supplied user, defaults re-order dendrograms correspondence analysis. Species ordered match site ordering like determined user.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-7","text":"Function fitspecaccum(..., model = \"asymp\") fitted logistic model instead asymptotic model (model = \"logis\"). nestedtemp() failed sparse data (fill < 0.38%).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-7","text":"plot function constrained ordination results (cca, rda, capscale) gained argument axis.bp (defaults TRUE) can used suppress axis scale biplot arrays. Number iterations nonmetric multidimensional scaling (NMDS) can set keyword maxit (defaults 200) metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-0-7","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.0-7","text":"result objects cca, rda capscale longer scores u.eig, v.eig wa.eig future versions vegan. change influence normal usage, vegan functions need items. However, external scripts packages may need changes future versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-6","dir":"Changelog","previous_headings":"","what":"vegan 2.0-6","title":"vegan 2.0-6","text":"CRAN release: 2013-02-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-6","text":"species scores scaled wrongly capscale(). scaled correctly Euclidean distances used, usually capscale() used non-Euclidean distances. graphics change redone. change scaling mainly influences spread species scores respect site scores. Function clamtest() failed set minimum abundance threshold cases. addition, output wrong possible species groups missing. problems reported Richard Telford (Bergen, Norway). Plotting object fitted envfit() fail p.max used unused levels one factors. unused levels result deletion observations missing values simply result supplying subset larger data set envfit(). multipart() printed wrong information analysis type (analysis correctly). Reported Valerie Coudrain. oecosimu() failed nestedfun returned data frame. fundamental fix vegan 2.2-0, structure oecosimu() result change. plot two-dimensional procrustes() solutions often draw original axes wrong angle. problem reported Elizabeth Ottesen (MIT). Function treedive() functional phylogenetic diversity correctly match species names community data species tree tree contained species occur data. Related function treedist() phylogenetic distances try match names .","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-6","text":"output capscale() displays value additive constant argument add = TRUE used. fitted() functions cca(), rda() capscale() can now return conditioned (partial) component response: Argument model gained new alternative model = \"pCCA\". dispindmorisita() output gained new column Chi-squared based probabilities null hypothesis (random distribution) true. metaMDS() monoMDS() new default convergence criteria. importantly, scale factor gradient (sfgrmin) stricter. former limit slack large data sets iterations stopped early without getting close solution. addition, scores() ignore now requests dimensions beyond calculated instead failing, scores() metaMDS() results drop dimensions. msoplot() gained legend argument positioning legend. Nestedness function nestednodf() gained plot method. ordiR2step() gained new argument R2scope (defaults TRUE) can used turn criterion stopping adjusted R2 current model exceeds scope. option allows model building scope overdetermined (number predictors higher number observations). ordiR2step() now handles partial redundancy analysis (pRDA). orditorp() gained argument select select rows columns results display. protest() prints standardized residual statistic squared m12 addition squared Procrustes correlation R2. calculated, latter displayed. Permutation tests much faster protest(). Instead calling repeatedly procrustes(), goodness fit statistic evaluated within function. wcmdscale() gained methods print, plot etc. results. methods used full wcmdscale result returned , e.g., argument eig = TRUE. default still return matrix scores similarly standard R function cmdscale(), case new methods used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-5","dir":"Changelog","previous_headings":"","what":"vegan 2.0-5","title":"vegan 2.0-5","text":"CRAN release: 2012-10-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-5","text":"anova(, ...) failed = \"axis\" = \"term\". bug reported Dr Sven Neulinger (Christian Albrecht University, Kiel, Germany). radlattice honour argument BIC = TRUE, always displayed AIC.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-5","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-5","text":"vegan functions permutation tests now density method can used find empirical probability distributions permutations. new plot method functions displays density observed statistic. density function available adonis, anosim, mantel, mantel.partial, mrpp, permutest.cca procrustes. Function adonis can return several statistics, now densityplot method (based lattice). Function oecosimu already density densityplot, now similar vegan methods, also work adipart, hiersimu multipart. radfit functions got predict method also accepts arguments newdata total new ranks site totals prediction. functions can also interpolate non-integer “ranks”, models also extrapolate.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-5","text":"Labels can now set plot envfit results. labels must given order function uses internally, new support function labels can used display default labels correct order. Mantel tests (functions mantel mantel.partial) gained argument na.rm can used remove missing values. options used care: Permutation tests can biased missing values originally matching fixed positions. radfit results can consistently accessed methods whether single model single site, models single site models sites data. functions now methods AIC, coef, deviance, logLik, fitted, predict residuals.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-0-5","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.0-5","text":"Building vegan vignettes failed latest version LaTeX (TeXLive 2012). R versions later 2.15-1 (including development version) report warnings errors installing checking vegan, must upgrade vegan version. warnings concern functions cIndexKM betadisper, error occurs betadisper. errors warnings triggered internal changes R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-4","dir":"Changelog","previous_headings":"","what":"vegan 2.0-4","title":"vegan 2.0-4","text":"CRAN release: 2012-06-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-4","text":"adipart assumed constant gamma diversity simulations assessing P-value. give biased results null model produces variable gamma diversities option weights = \"prop\" used. default null model (\"r2dtable\") default option (weights = \"unif\") analysed correctly. anova(, = \"axis\") cases failed due ‘NAMESPACE’ issues. clamtest wrongly used frequencies instead counts calculating sample coverage. detectable differences produced rerunning examples Chazdon et al. 2011 vegan help page. envfit failed unused factor levels. predict cca results type = \"response\" type = \"working\" failed newdata number rows match original data. Now newdata ignored wrong number rows. number rows must match results cca must weighted original row totals. problem concern rda capscale results need row weights. Reported Glenn De’ath.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-4","text":"Functions diversity partitioning (adipart, hiersimu multipart) now formula default methods. formula method identical previous functions, default method can take two matrices input. Functions adipart multipart can used fast easy overall partitioning alpha, beta gamma diversities omitting argument describing hierarchy. method betadisper biased small sample sizes. effects bias strongest unequal sample sizes. bias adjusted version developed Adrian Stier Ben Bolker, can invoked argument bias.adjust (defaults FALSE). bioenv accepts dissimilarities (square matrices can interpreted dissimilarities) alternative community data. allows using dissimilarities available vegdist. plot function envfit results gained new argument bg can used set background colour plotted labels. msoplot configurable, allows, instance, setting y-axis limits. Hulls ellipses now filled using semitransparent colours ordihull ordiellipse, user can set degree transparency new argument alpha. filled shapes used functions called argument draw = \"polygon\". Function ordihull puts labels (argument label = TRUE) now real polygon centre. ordiplot3d returns function envfit.convert projected location origin. Together can used add envfit results existing ordiplot3d plots. Equal aspect ratio set exactly ordiplot3d underlying core routines allow . Now ordiplot3d sets equal axis ranges, documents urge users verify aspect ratio reasonably equal graph looks like cube. problems solved future, ordiplot3d may removed next releases vegan. Function ordipointlabel gained argument select items plotting. argument can used one set points.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-3","dir":"Changelog","previous_headings":"","what":"vegan 2.0-3","title":"vegan 2.0-3","text":"CRAN release: 2012-03-03","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-3","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-3","text":"Added new nestedness functions nestedbetasor nestedbetajac implement multiple-site dissimilarity indices decomposition turnover nestedness components following Baselga (Global Ecology Biogeography 19, 134–143; 2010). Added function rarecurve draw rarefaction curves row (sampling unit) input data, optionally lines showing rarefied species richness given sample size curve. Added function simper implements “similarity percentages” Clarke (Australian Journal Ecology 18, 117–143; 1993). method compares two groups decomposes average -group Bray-Curtis dissimilarity index contributions individual species. code developed GitHub Eduard Szöcs (Uni Landau, Germany).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-3","text":"betadisper() failed groups factor empty levels. constrained ordination methods support functions robust border cases (completely aliased effects, saturated models, user requests non-existng scores etc). Concerns capscale, ordistep, varpart, plot function constrained ordination, anova(, = \"margin\"). scores function monoMDS honour choices argument hence dimensions chosen plot. default scores method failed number requested axes higher ordination object . reported error ordiplot R-sig-ecology mailing list.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-3","text":"metaMDS argument noshare = 0 now regarded numeric threshold always triggers extended dissimilarities (stepacross), instead treated synonymous noshare = FALSE always suppresses extended dissimilarities. Nestedness discrepancy index nesteddisc gained new argument allows user set number iterations optimizing index. oecosimu displays mean simulations describes alternative hypothesis clearly printed output. Implemented adjusted R2 partial RDA. partial model rda(Y ~ X1 + Condition(X2)) component [] = X1|X2 variance partition varpart describes marginal (unique) effect constraining term adjusted R2. Added Cao dissimilarity (CYd) new dissimilarity method vegdist following Cao et al., Water Envir Res 69, 95–106 (1997). index good data high beta diversity variable sampling intensity. Thanks consultation Yong Cao (Univ Illinois, USA).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-2","dir":"Changelog","previous_headings":"","what":"vegan 2.0-2","title":"vegan 2.0-2","text":"CRAN release: 2011-11-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-2","text":"Function capscale failed constrained component zero rank. happened likely partial models conditions aliased constraints. problem observed anova(..., =\"margin\") uses partial models analyses marginal effects, reported email message R-News mailing list. stressplot goodness sometimes failed metaMDS based isoMDS (MASS package) metaMDSdist use defaults step-across (extended) dissimilarities metaMDS(..., engine = \"isoMDS\"). change defaults can also influence triggering step-across capscale(..., metaMDSdist = TRUE). adonis contained minor bug resulting incomplete implementation speed-affect results. fixing bug, bug identified transposing hat matrices. second bug active following fixing first bug. fixing bugs, speed-internal f.test() function fully realised. Reported Nicholas Lewin-Koh.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-2","text":"ordiarrows ordisegments gained argument order.gives variable sort points within groups. Earlier points assumed order. Function ordispider invisibly returns coordinates points connected. Typically class centroids point, constrained ordination groups LC scores.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-1","dir":"Changelog","previous_headings":"","what":"vegan 2.0-1","title":"vegan 2.0-1","text":"CRAN release: 2011-10-20","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-1","text":"clamtest: new function classify species generalists specialists two distinct habitats (CLAM test Chazdon et al., Ecology 92, 1332–1343; 2011). test based multinomial distribution individuals two habitat types sampling units, applicable count data -dispersion. .preston gained plot lines methods, .fisher gained plot method (also can add items existing plots). similar plot lines prestonfit fisherfit, display data without fitted lines. raupcrick: new function implement Raup-Crick dissimilarity probability number co-occurring species occurrence probabilities proportional species frequencies. Vegan Raup-Crick index choice vegdist, uses equal sampling probabilities species analytic equations. new raupcrick function uses simulation oecosimu. function follows Chase et al. (2011) Ecosphere 2:art24 [doi:10.1890/ES10-00117.1], developed consultation Brian Inouye.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-1","text":"Function meandist scramble items give wrong results, especially grouping numerical. problem reported Dr Miguel Alvarez (Univ. Bonn). metaMDS reset tries new model started previous.best solution different model. Function permatswap community null models using quantitative swap never swapped items 2x2 submatrix cells filled. result permutest.cca updated ‘NAMESPACE’ issue. R 2.14.0 changed accept using sd() function matrices (behaviour least since R 1.0-0), several vegan functions changed adapt change (rda, capscale, simulate methods rda, cca capscale). change R 2.14.0 influence results probably wish upgrade vegan avoid annoying warnings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-1","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-1","text":"nesteddisc slacker hence faster trying optimize statistic tied column frequencies. Tracing showed cases improved ordering found rather early tries, results equally good cases.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-0","dir":"Changelog","previous_headings":"","what":"vegan 2.0-0","title":"vegan 2.0-0","text":"CRAN release: 2011-09-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-0","text":"Peter Minchin joins vegan team. vegan implements standard R ‘NAMESPACE’. general, S3 methods exported means directly use see contents functions like cca.default, plot.cca anova.ccabyterm. use functions rely R delegation simply use cca result objects use plot anova without suffix .cca. see contents function can use :::, vegan:::cca.default. change may break packages, documents scripts rely non-exported names. vegan depends permute package. package provides powerful tools restricted permutation schemes. vegan permutation gradually move use permute, currently betadisper uses new feature.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-0","text":"monoMDS: new function non-metric multidimensional scaling (NMDS). function replaces MASS::isoMDS default method metaMDS. Major advantages monoMDS ‘weak’ (‘primary’) tie treatment means can split tied observed dissimilarities. ‘Weak’ tie treatment improves ordination heterogeneous data sets, maximum dissimilarities 1 can split. addition global NMDS, monoMDS can perform local hybrid NMDS metric MDS. can also handle missing zero dissimilarities. Moreover, monoMDS faster previous alternatives. function uses Fortran code written Peter Minchin. MDSrotate new function replace metaMDSrotate. function can rotate metaMDS monoMDS results first axis parallel environmental vector. eventstar finds minimum evenness profile Tsallis entropy, uses find corresponding values diversity, evenness numbers equivalent following Mendes et al. (Ecography 31, 450-456; 2008). code contributed Eduardo Ribeira Cunha Heloisa Beatriz Antoniazi Evangelista adapted vegan Peter Solymos. fitspecaccum fits non-linear regression models species accumulation results specaccum. function can use new self-starting species accumulation models vegan self-starting non-linear regression models R. function can fit Arrhenius, Gleason, Gitay, Lomolino (vegan), asymptotic, Gompertz, Michaelis-Menten, logistic Weibull (base R) models. function plot predict methods. Self-starting non-linear species accumulation models SSarrhenius, SSgleason, SSgitay SSlomolino. can used fitspecaccum directly non-linear regression nls. functions implemented found good species-area models Dengler (J. Biogeogr. 36, 728-744; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-0","text":"adonis, anosim, meandist mrpp warn negative dissimilarities, betadisper refuses analyse . functions expect dissimilarities, giving something else (like correlations) probably user error. betadisper uses restricted permutation permute package. metaMDS uses monoMDS default ordination engine. Function gains new argument engine can used alternatively select MASS::isoMDS. default use stepacross monoMDS ‘weak’ tie treatment can cope tied maximum dissimilarities one. However, stepacross default isoMDS handle adequately tied maximum dissimilarities. specaccum gained predict method uses either linear spline interpolation data observed points. Extrapolation possible spline interpolation, may make little sense. specpool can handle missing values empty factor levels grouping factor pool. Now also checks length pool matches number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-0-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.0-0","text":"metaMDSrotate replaced MDSrotate can also handle results monoMDS. permuted.index2 “new” permutation code removed favour permute package. code intended normal use, packages depending code vegan instead depend permute.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-0","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-0","text":"treeheight uses much snappier code. results unchanged.","code":""}]
+[{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"vegan-faq","dir":"Articles","previous_headings":"","what":"vegan FAQ","title":"","text":"document contains answers frequently asked questions R package vegan. work licensed Creative Commons Attribution 3.0 License. view copy license, visit https://creativecommons.org/licenses//3.0/ send letter Creative Commons, 543 Howard Street, 5th Floor, San Francisco, California, 94105, USA. Copyright © 2008-2016 vegan development team","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is vegan?","title":"","text":"Vegan R package community ecologists. contains popular methods multivariate analysis needed analysing ecological communities, tools diversity analysis, potentially useful functions. Vegan self-contained must run R statistical environment, also depends many R packages. Vegan free software distributed GPL2 license.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-is-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What is R?","title":"","text":"R system statistical computation graphics. consists language plus run-time environment graphics, debugger, access certain system functions, ability run programs stored script files. R home page https://www.R-project.org/. free software distributed GNU-style copyleft, official part GNU project (“GNU S”).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-obtain-vegan-and-r","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to obtain vegan and R?","title":"","text":"R latest release version vegan can obtained CRAN. Unstable development version vegan can obtained GitHub. github page gives instructions obtaining installing development versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-r-packages-vegan-depends-on","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What R packages vegan depends on?","title":"","text":"Vegan depends permute package provide advanced flexible permutation routines vegan. permute package developed together vegan GitHub. individual vegan functions depend packages MASS, mgcv, parallel, cluster lattice. Vegan dependence tcltk deprecated removed future releases. base recommended R packages available every R installation. Vegan declares suggested imported packages, can install vegan use functions without packages. Vegan accompanied supporting package vegan3d three-dimensional dynamic plotting. vegan3d package needs tcltk non-standard packages rgl scatterplot3d.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-packages-are-available-for-ecologists","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other packages are available for ecologists?","title":"","text":"CRAN Task Views include entries like Environmetrics, Multivariate Spatial describe several useful packages functions. install R package ctv, can inspect Task Views R session, automatically install sets important packages.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-other-documentation-is-available-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"What other documentation is available for vegan?","title":"","text":"Vegan fully documented R package standard help pages. authoritative sources documentation (last resource can use force read source, vegan open source). Vegan package ships documents can read browseVignettes(\"vegan\") command. documents included vegan package Vegan NEWS can accessed via news() command. document (FAQ-vegan). Short introduction basic ordination methods vegan (intro-vegan). Introduction diversity methods vegan (diversity-vegan). Discussion design decisions vegan (decision-vegan). Description variance partition procedures function varpart (partitioning). Web documents outside package include: https://github.com/vegandevs/vegan: development page. https://vegandevs.github.io/vegan/: vegan homepage.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-a-graphical-user-interface-gui-for-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is there a Graphical User Interface (GUI) for vegan?","title":"","text":"Roeland Kindt made package BiodiversityR provides GUI vegan. package available CRAN. mere GUI vegan, adds new functions complements vegan functions order provide workbench biodiversity analysis. can install BiodiversityR using install.packages(\"BiodiversityR\") graphical package management menu R. GUI works Windows, MacOS X Linux.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-cite-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to cite vegan?","title":"","text":"Use command citation(\"vegan\") R see recommended citation used publications.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-build-vegan-from-sources","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to build vegan from sources?","title":"","text":"general, need build vegan sources, binary builds release versions available CRAN Windows MacOS X. use operating systems, may use source packages. Vegan standard R package, can built like instructed R documentation. Vegan contains source files C FORTRAN, need appropriate compilers (may need work Windows MacOS X).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"are-there-binaries-for-devel-versions","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Are there binaries for devel versions?","title":"","text":"Binaries can available R Universe: see https://github.com/vegandevs/vegan instructions.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-report-a-bug-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"How to report a bug in vegan?","title":"","text":"think found bug vegan, report vegan maintainers developers. preferred forum report bugs GitHub. bug report detailed bug can replicated corrected. Preferably, send example causes bug. needs data set available R, send minimal data set well. also paste output error message message. also specify version vegan used. Bug reports welcome: way make vegan non-buggy. Please note shall send bug reports R mailing lists, since vegan standard R package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-a-bug-or-a-feature","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Is it a bug or a feature?","title":"","text":"necessarily bug function gives different results expect: may deliberate design decision. may useful check documentation function see intended behaviour. may also happen function argument switch behaviour match expectation. instance, function vegdist always calculates quantitative indices (possible). expect calculate binary index, use argument binary = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-contribute-to-vegan","dir":"Articles","previous_headings":"vegan FAQ > Introduction","what":"Can I contribute to vegan?","title":"","text":"Vegan dependent user contribution. feedback welcome. problems vegan, may simple incomplete documentation, shall best improve documents. Feature requests also welcome, necessarily fulfilled. new feature added easy looks useful, submit code. can write code , best forum contribute vegan GitHub.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-only-numeric-and-positive-data-but-vegan-still-complains","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have only numeric and positive data but vegan still complains","title":"","text":"wrong! Computers painfully pedantic, find non-numeric negative data entries, really . Check data! common reasons non-numeric data row names read non-numeric variable instead used row names (check argument row.names reading data), column names interpreted data (check argument header = TRUE reading data). Another common reason empty cells input data, interpreted missing values.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-analyse-binary-or-cover-class-data","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I analyse binary or cover class data?","title":"","text":"Yes. vegan methods can handle binary data cover abundance data. statistical tests based permutation, make distributional assumptions. methods (mainly diversity analysis) need count data. methods check input data integers, may fooled cover class data.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-dissimilarities-in-vegan-differ-from-other-sources","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why dissimilarities in vegan differ from other sources?","title":"","text":"commonly reason software use presence–absence data whereas vegan used quantitative data. Usually vegan indices quantitative, can use argument binary = TRUE make presence–absence. However, index name cases, although different names usually occur literature. instance, Jaccard index actually refers binary index, vegan uses name \"jaccard\" quantitative index, . Another reason may indices indeed defined differently, people use names different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-nmds-stress-is-sometimes-0-1-and-sometimes-10","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Why NMDS stress is sometimes 0.1 and sometimes 10?","title":"","text":"Stress proportional measure badness fit. proportions can expressed either parts one percents. Function isoMDS (MASS package) uses percents, function monoMDS (vegan package) uses proportions, therefore stress 100 times higher isoMDS. results goodness function also depend definition stress, goodness 100 times higher isoMDS monoMDS. conventions equally correct.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-get-zero-stress-but-no-repeated-solutions-in-metamds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I get zero stress but no repeated solutions in metaMDS","title":"","text":"first (try 0) run metaMDS starts metric scaling solution usually good, sofware return solution. However, metaMDS tries see standard solution can repeated, improved improved solution still repeated. cases, return best solution found, burning need anything get message tha solution repeated. keen know solution really global optimum, may follow instructions metaMDS help section “Results Repeated” try . common reason observations NMDS. n observations (points) k dimensions need estimate n*k parameters (ordination scores) using n*(n-1)/2 dissimilarities. k dimensions must n > 2*k + 1, two dimensions least six points. degenerate situations may need even larger number points. lower number points, can find undefined number perfect (stress zero) different solutions. Conventional wisdom due Kruskal n > 4*k + 1 points k dimensions. typical symptom insufficient data (nearly) zero stress two convergent solutions. cases reduce number dimensions (k) small data sets use NMDS, rely metric methods. seems local hybrid scaling monoMDS similar lower limits practice (although theoretically differ). However, higher number dimensions can used metric scaling, monoMDS principal coordinates analysis (cmdscale stats, wcmdscale vegan).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"zero-dissimilarities-in-isomds","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Zero dissimilarities in isoMDS","title":"","text":"Function metaMDS uses function monoMDS default method NMDS, function can handle zero dissimilarities. Alternative function isoMDS handle zero dissimilarities. want use isoMDS, can use argument zerodist = \"add\" metaMDS handle zero dissimilarities. argument, zero dissimilarities replaced small positive value, can handled isoMDS. kluge, people like . principal solution remove duplicate sites using R command unique. However, standardizations dissimilarity indices, originally non-unique sites can zero dissimilarity, resort kluge (work harder data). Usually better use monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-have-heard-that-you-cannot-fit-environmental-vectors-or-surfaces-to-nmds-results-which-only-have-rank-order-scores","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I have heard that you cannot fit environmental vectors or surfaces to NMDS results which only have rank-order scores","title":"","text":"Claims like indeed large Internet, based grave misunderstanding plainly wrong. NMDS ordination results strictly metric, vegan metaMDS monoMDS even strictly Euclidean. method called “non-metric” Euclidean distances ordination space non-metric rank-order relationship community dissimilarities. can inspect non-linear step curve using function stressplot vegan. ordination scores strictly Euclidean, correct use vegan functions envfit ordisurf NMDS results.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"where-can-i-find-numerical-scores-of-ordination-axes","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Where can I find numerical scores of ordination axes?","title":"","text":"Normally can use function scores extract ordination scores ordination method. scores function can also find ordination scores many non-vegan functions prcomp princomp ade4 functions. cases ordination result object stores raw scores, axes also scaled appropriate access scores. instance, cca rda ordination object -called normalized scores, scaled ordination plots use accessed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-the-rda-results-are-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How the RDA results are scaled?","title":"","text":"scaling RDA results indeed differ software packages. scaling RDA complicated issue explained FAQ, explained separate pdf document “Design decision implementation details vegan” can read command browseVignettes(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-cannot-print-and-plot-rda-results-properly","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I cannot print and plot RDA results properly","title":"","text":"RDA ordination results weird format plot properly, probably name clash klaR package also function rda, klaR print, plot predict functions used vegan RDA results. can choose rda functions using vegan::rda() klaR::rda(): get obscure error messages use wrong function. general, vegan able work normally vegan loaded klaR, klaR loaded later, functions take precedence vegan. Sometimes vegan namespace loaded automatically restoring previously stored workspace start-, klaR methods always take precedence vegan. check loaded packages. klaR may also loaded indirectly via packages (reported cases often loaded via agricolae package). Vegan klaR function name (rda), may possible use packages simultaneously, safest choice unload one packages possible. See discussion vegan github issues.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"ordination-fails-with-error-in-la-svd","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Ordination fails with “Error in La.svd”","title":"","text":"Constrained ordination (cca, rda, dbrda, capscale) sometimes fail error message Error La.svd(x, nu, nv): error code 1 Lapack routine 'dgesdd'. seems basic problem svd function LAPACK used numerical analysis R. LAPACK external library beyond control package developers R core team problems may unsolvable. Reducing range constraints (environmental variables) helps sometimes. instance, multiplying constraints constant < 1. rescaling influence numerical results constrained ordination, can complicate analyses values constraints needed, scaling must applied . reports problems getting rare may problem fixed R LAPACK.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"variance-explained-by-ordination-axes-","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Variance explained by ordination axes.","title":"","text":"general, vegan directly give statistics “variance explained” ordination axes constrained axes. design decision: think information normally useless often misleading. community ordination, goal typically explain variance, find “gradients” main trends data. “total variation” often meaningless, proportions meaningless values also meaningless. Often better solution explains smaller part “total variation”. instance, unstandardized principal components analysis variance generated small number abundant species, easy “explain” data really multivariate. standardize data, species equally important. first axes explains much less “total variation”, now explain species equally, results typically much useful whole community. Correspondence analysis uses another measure variation (variance), typically explains “smaller proportion” principal components better result. Detrended correspondence analysis nonmetric multidimensional scaling even try “explain” variation, use criteria. methods incommensurable, impossible compare methods using “explanation variation”. still want get “explanation variation” (deranged editor requests ), possible get information methods: Eigenvector methods: Functions rda, cca, dbrda capscale give variation conditional (partialled), constrained (canonical) residual components. Function eigenvals extracts eigenvalues, summary(eigenvals(ord)) reports proportions explained result object ord, also works decorana wcmdscale. Function RsquareAdj gives R-squared adjusted R-squared (available) constrained components. Function goodness gives statistics individual species sites. addition, special function varpart unbiased partitioning variance four separate components redundancy analysis. Nonmetric multidimensional scaling. NMDS method nonlinear mapping, concept variation explained make sense. However, 1 - stress^2 transforms nonlinear stress quantity analogous squared correlation coefficient. Function stressplot displays nonlinear fit gives statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-have-random-effects-in-constrained-ordination-or-in-adonis","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I have random effects in constrained ordination or in adonis?","title":"","text":". Strictly speaking, impossible. However, can define models respond similar goals random effects models, although strictly speaking use fixed effects. Constrained ordination functions cca, rda dbrda can Condition() terms formula. Condition() define partial terms fitted constraints can used remove effects background variables, contribution decomposing inertia (variance) reported separately. partial terms often regarded similar random effects, still fitted way terms strictly speaking fixed terms. Function adonis2 can evaluate terms sequentially. model right-hand-side ~ + B effects evaluated first, effects B removing effects . Sequential tests also available anova function constrained ordination results setting argument = \"term\". way, first terms can serve similar role random effects, although fitted way terms, strictly speaking fixed terms. permutation tests vegan based permute package allows constructing various restricted permutation schemes. instance, can set levels plots blocks factor regarded random term. major reason real random effects models impossible vegan functions tests based permutation data. data given, fixed, therefore permutation tests basically tests fixed terms fixed data. Random effect terms require permutations data random component instead given, fixed data, tests available vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-it-possible-to-have-passive-points-in-ordination","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Is it possible to have passive points in ordination?","title":"","text":"Vegan concept passive points, point little influence ordination results. However, can add points eigenvector methods using predict functions newdata. can first perform ordination without species sites, can find scores points using complete data newdata. predict functions available basic eigenvector methods vegan (cca, rda, decorana, --date list, use command methods(\"predict\")).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"class-variables-and-dummies","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Class variables and dummies","title":"","text":"define class variable R factor, vegan automatically handle . R (vegan) knows unordered ordered factors. Unordered factors internally coded dummy variables, one redundant level removed aliased. default contrasts, removed level first one. Ordered factors expressed polynomial contrasts. contrasts explained standard R documentation.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-are-environmental-arrows-scaled","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How are environmental arrows scaled?","title":"","text":"printed output envfit gives direction cosines coordinates unit length arrows. plotting, scaled correlation (square roots column r2). can see scaled lengths envfit arrows using command scores. scaled environmental vectors envfit arrows continuous environmental variables constrained ordination (cca, rda, dbrda) adjusted fill current graph. lengths arrows fixed meaning respect points (species, sites), can compared , therefore relative lengths important. want change scaling arrows, can use text (plotting arrows text) points (plotting arrows) functions constrained ordination. functions argument arrow.mul sets multiplier. plot function envfit also arrow.mul argument set arrow multiplier. save invisible result constrained ordination plot command, can see value currently used arrow.mul saved attribute biplot scores. Function ordiArrowMul used find scaling current plot. can use function see arrows scaled:","code":"sol <- cca(varespec) ef <- envfit(sol ~ ., varechem) plot(sol) ordiArrowMul(scores(ef, display=\"vectors\"))"},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"i-want-to-use-helmert-or-sum-contrasts","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"I want to use Helmert or sum contrasts","title":"","text":"vegan uses standard R utilities defining contrasts. default standard installations use treatment contrasts, can change behaviour globally setting options locally using keyword contrasts. Please check R help pages user manuals details.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"what-are-aliased-variables-and-how-to-see-them","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"What are aliased variables and how to see them?","title":"","text":"Aliased variable information can expressed help variables. variables automatically removed constrained ordination vegan. aliased variables can redundant levels factors whole variables. Vegan function alias gives defining equations aliased variables. want see names aliased variables levels solution sol, use alias(sol, names.=TRUE).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"plotting-aliased-variables","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Plotting aliased variables","title":"","text":"can fit vectors class centroids aliased variables using envfit function. envfit function uses weighted fitting, fitted vectors identical vectors correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"restricted-permutations-in-vegan","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Restricted permutations in vegan","title":"","text":"Vegan uses permute package permutation tests. permute package allow restricted permutation designs time series, line transects, spatial grids blocking factors. construction restricted permutation schemes explained manual page permutations vegan documentation permute package.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-use-different-plotting-symbols-in-ordination-graphics","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to use different plotting symbols in ordination graphics?","title":"","text":"default ordination plot function intended fast plotting configurable. use different plotting symbols, first create empty ordination plot plot(..., type=\"n\"), add points text created empty frame (... means arguments want give plot command). points text commands fully configurable, allow different plotting symbols characters.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-to-avoid-cluttered-ordination-graphs","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"How to avoid cluttered ordination graphs?","title":"","text":"really high number species sites, graphs often congested many labels overwritten. may impossible complete readable graphics data sets. give brief overview tricks can use. Gavin Simpson’s blog bottom heap series articles “decluttering ordination plots” detailed discussion examples. Use points, possibly different types need see labels. may need first create empty plot using plot(..., type=\"n\"), satisfied default graph. (... means arguments want give plot command.) Use points add labels desired points using interactive identify command need see labels. Add labels using function ordilabel uses non-transparent background text. labels still shadow , uppermost labels readable. Argument priority help displaying interesting labels (see Decluttering blog, part 1). Use orditorp function uses labels can added graph without overwriting labels, points otherwise, need see labels. must first create empty plot using plot(..., type=\"n\"), add labels points orditorp (see Decluttering blog). Use ordipointlabel uses points text labels points, tries optimize location text minimize overlap (see Decluttering blog). Ordination text points functions argument select can used full control selecting items plotted text points. Use interactive orditkplot function (vegan3d package) lets drag labels points better positions need see labels. one set points can used (see Decluttering blog). plot functions allow zoom part graph using xlim ylim arguments reduce clutter congested areas.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-flip-an-axis-in-ordination-diagram","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I flip an axis in ordination diagram?","title":"","text":"Use xlim ylim flipped limits. model mod <- cca(dune) can flip first axis plot(mod, xlim = c(3, -2)).","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"can-i-zoom-into-an-ordination-plot","dir":"Articles","previous_headings":"vegan FAQ > Ordination","what":"Can I zoom into an ordination plot?","title":"","text":"can use xlim ylim arguments plot ordiplot zoom ordination diagrams. Normally must set xlim ylim ordination plots keep equal aspect ratio axes, fill graph longer axis fit. Dynamic zooming can done function orditkplot CRAN package vegan3d. can directly save edited orditkplot graph various graphic formats, can export graph object back R session use plot display results.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"is-there-twinspan","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Is there TWINSPAN?","title":"","text":"TWINSPAN R available github.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"why-restricted-permutation-does-not-influence-adonis-results","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"Why restricted permutation does not influence adonis results?","title":"","text":"permutation scheme influences permutation distribution statistics probably significance levels, influence calculation statistics.","code":""},{"path":"https://vegandevs.github.io/vegan/articles/FAQ-vegan.html","id":"how-is-deviance-calculated","dir":"Articles","previous_headings":"vegan FAQ > Other analysis methods","what":"How is deviance calculated?","title":"","text":"vegan functions, radfit use base R facility family maximum likelihood estimation. allows use several alternative error distributions, among \"poisson\" \"gaussian\". R family also defines deviance. can see equations deviance commands like poisson()$dev gaussian()$dev. general, deviance 2 times log.likelihood shifted models exact fit zero deviance.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Jari Oksanen. Author, maintainer. Gavin L. Simpson. Author. F. Guillaume Blanchet. Author. Roeland Kindt. Author. Pierre Legendre. Author. Peter R. Minchin. Author. R.B. O'Hara. Author. Peter Solymos. Author. M. Henry H. Stevens. Author. Eduard Szoecs. Author. Helene Wagner. Author. Matt Barbour. Author. Michael Bedward. Author. Ben Bolker. Author. Daniel Borcard. Author. Gustavo Carvalho. Author. Michael Chirico. Author. Miquel De Caceres. Author. Sebastien Durand. Author. Heloisa Beatriz Antoniazi Evangelista. Author. Rich FitzJohn. Author. Michael Friendly. Author. Brendan Furneaux. Author. Geoffrey Hannigan. Author. Mark O. Hill. Author. Leo Lahti. Author. Dan McGlinn. Author. Marie-Helene Ouellette. Author. Eduardo Ribeiro Cunha. Author. Tyler Smith. Author. Adrian Stier. Author. Cajo J.F. Ter Braak. Author. James Weedon. Author.","code":""},{"path":"https://vegandevs.github.io/vegan/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O'Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier , Ter Braak C, Weedon J (2024). vegan: Community Ecology Package. R package version 2.6-7, https://github.com/vegandevs/vegan, https://vegandevs.github.io/vegan/.","code":"@Manual{, title = {vegan: Community Ecology Package}, author = {Jari Oksanen and Gavin L. Simpson and F. Guillaume Blanchet and Roeland Kindt and Pierre Legendre and Peter R. Minchin and R.B. O'Hara and Peter Solymos and M. Henry H. Stevens and Eduard Szoecs and Helene Wagner and Matt Barbour and Michael Bedward and Ben Bolker and Daniel Borcard and Gustavo Carvalho and Michael Chirico and Miquel {De Caceres} and Sebastien Durand and Heloisa Beatriz Antoniazi Evangelista and Rich FitzJohn and Michael Friendly and Brendan Furneaux and Geoffrey Hannigan and Mark O. Hill and Leo Lahti and Dan McGlinn and Marie-Helene Ouellette and Eduardo {Ribeiro Cunha} and Tyler Smith and Adrian Stier and Cajo J.F. {Ter Braak} and James Weedon}, year = {2024}, note = {R package version 2.6-7, https://github.com/vegandevs/vegan}, url = {https://vegandevs.github.io/vegan/}, }"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"vegan-an-r-package-for-community-ecologists","dir":"","previous_headings":"","what":"vegan: an R package for community ecologists","title":"vegan: an R package for community ecologists","text":"Ordination methods, diversity analysis functions community vegetation ecologists. Website development version vegan package. Vignettes available R-universe Introduction ordination vegan Partition Variation Diversity analysis vegan Design decisions implementation","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"vegan: an R package for community ecologists","text":"install development version vegan can use usual git R CMD build -> R CMD INSTALL dance cloned repo (downloaded sources). ’ll need able install packages source work; don’t relevant developer tools, won’t able install vegan way.","code":""},{"path":"https://vegandevs.github.io/vegan/index.html","id":"using-remotes","dir":"","previous_headings":"","what":"Using remotes","title":"vegan: an R package for community ecologists","text":"developer tools installed don’t want hassle keeping local source code tree --date, use remotes package:","code":"install.packages(\"remotes\") remotes::install_github(\"vegandevs/vegan\")"},{"path":"https://vegandevs.github.io/vegan/index.html","id":"installing-binaries-from-r-universe","dir":"","previous_headings":"","what":"Installing binaries from R Universe","title":"vegan: an R package for community ecologists","text":"just want install binary version packages, just CRAN, can install R Universe repository. Run following R session:","code":"install.packages('vegan', repos = c('https://vegandevs.r-universe.dev','https://cloud.r-project.org'))"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":null,"dir":"Reference","previous_headings":"","what":"Barro Colorado Island Tree Counts — BCI","title":"Barro Colorado Island Tree Counts — BCI","text":"Tree counts 1-hectare plots Barro Colorado Island associated site information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI) data(BCI.env)"},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Barro Colorado Island Tree Counts — BCI","text":"data frame 50 plots (rows) 1 hectare counts trees plot total 225 species (columns). Full Latin names used tree species. names updated http://www.theplantlist.org Kress et al. (2009) allows matching 207 species doi:10.5061/dryad.63q27 (Zanne et al., 2014). original species names available attribute original.names BCI. See Examples changed names. BCI.env, data frame 50 plots (rows) nine site variables derived Pyke et al. (2001) Harms et al. (2001): UTM.EW: UTM coordinates (zone 17N) East-West. UTM.NS: UTM coordinates (zone 17N) North-South. Precipitation: Precipitation mm per year. Elevation: Elevation m sea level. Age.cat: Forest age category. Geology: Underlying geological formation. Habitat: Dominant habitat type based map habitat types 25 grid cells plot (Harms et al. 2001, excluding streamside habitat). habitat types Young forests (ca. 100 years), old forests > 7 degree slopes (OldSlope), old forests 152 m elevation (OldLow) higher elevation (OldHigh) Swamp forests. River: \"Yes\" streamside habitat plot. EnvHet: Environmental Heterogeneity assessed Simpson diversity frequencies Habitat types 25 grid cells plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Barro Colorado Island Tree Counts — BCI","text":"Data give numbers trees least 10 cm diameter breast height (DBH) one hectare quadrat 1982 BCI plot. Within plot, individuals tallied recorded table. full survey included smaller trees DBH 1 cm larger, BCI dataset subset larger trees compiled Condit et al. (2002). full data thinner trees densities 4000 stems per hectare, ten times stems data. dataset BCI provided (2003) illustrate analysis methods vegan. scientific research ecological issues strongly recommend access complete modern data (Condit et al. 2019) updated taxonomy (Condit et al. 2020). data frame contains Barro Colorado Island subset full data table Condit et al. (2002). quadrats located regular grid. See BCI.env coordinates. full description site information BCI.env given Pyke et al. (2001) Harms et al. (2001). N.B. Pyke et al. (2001) Harms et al. (2001) give conflicting information forest age categories elevation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Barro Colorado Island Tree Counts — BCI","text":"https://www.science.org/doi/10.1126/science.1066854 community data References environmental data. updated complete data (incl. thinner trees 1 cm), see Condit et al. (2019).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Barro Colorado Island Tree Counts — BCI","text":"Condit, R, Pitman, N, Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., Nuñez, P., Aguilar, S., Valencia, R., Villa, G., Muller-Landau, H.C., Losos, E. & Hubbell, S.P. (2002). Beta-diversity tropical forest trees. Science 295, 666--669. Condit R., Pérez, R., Aguilar, S., Lao, S., Foster, R. & Hubbell, S. (2019). Complete data Barro Colorado 50-ha plot: 423617 trees, 35 years [Dataset]. Dryad. doi:10.15146/5xcp-0d46 Condit, R., Aguilar, S., Lao, S., Foster, R., Hubbell, S. (2020). BCI 50-ha Plot Taxonomy [Dataset]. Dryad. doi:10.15146/R3FH61 Harms K.E., Condit R., Hubbell S.P. & Foster R.B. (2001) Habitat associations trees shrubs 50-ha neotropical forest plot. J. Ecol. 89, 947--959. Kress W.J., Erickson D.L, Jones F.., Swenson N.G, Perez R., Sanjur O. & Bermingham E. (2009) Plant DNA barcodes community phylogeny tropical forest dynamics plot Panama. PNAS 106, 18621--18626. Pyke, C. R., Condit, R., Aguilar, S., & Lao, S. (2001). Floristic composition across climatic gradient neotropical lowland forest. Journal Vegetation Science 12, 553--566. doi:10.2307/3237007 Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506, 89--92. doi:10.1038/nature12872 (published online Dec 22, 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/BCI.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Barro Colorado Island Tree Counts — BCI","text":"","code":"data(BCI, BCI.env) head(BCI.env) #> UTM.EW UTM.NS Precipitation Elevation Age.cat Geology Habitat Stream EnvHet #> 1 625754 1011569 2530 120 c3 Tb OldSlope Yes 0.6272 #> 2 625754 1011669 2530 120 c3 Tb OldLow Yes 0.3936 #> 3 625754 1011769 2530 120 c3 Tb OldLow No 0.0000 #> 4 625754 1011869 2530 120 c3 Tb OldLow No 0.0000 #> 5 625754 1011969 2530 120 c3 Tb OldSlope No 0.4608 #> 6 625854 1011569 2530 120 c3 Tb OldLow No 0.0768 ## see changed species names oldnames <- attr(BCI, \"original.names\") taxa <- cbind(\"Old Names\" = oldnames, \"Current Names\" = names(BCI)) noquote(taxa[taxa[,1] != taxa[,2], ]) #> Old Names Current Names #> [1,] Abarema.macradenium Abarema.macradenia #> [2,] Acacia.melanoceras Vachellia.melanoceras #> [3,] Apeiba.aspera Apeiba.glabra #> [4,] Aspidosperma.cruenta Aspidosperma.desmanthum #> [5,] Cassipourea.elliptica Cassipourea.guianensis #> [6,] Cespedezia.macrophylla Cespedesia.spathulata #> [7,] Chlorophora.tinctoria Maclura.tinctoria #> [8,] Coccoloba.manzanillensis Coccoloba.manzinellensis #> [9,] Coussarea.curvigemmia Coussarea.curvigemma #> [10,] Cupania.sylvatica Cupania.seemannii #> [11,] Dipteryx.panamensis Dipteryx.oleifera #> [12,] Eugenia.coloradensis Eugenia.florida #> [13,] Eugenia.oerstedeana Eugenia.oerstediana #> [14,] Guapira.standleyana Guapira.myrtiflora #> [15,] Hyeronima.alchorneoides Hieronyma.alchorneoides #> [16,] Inga.marginata Inga.semialata #> [17,] Lonchocarpus.latifolius Lonchocarpus.heptaphyllus #> [18,] Maquira.costaricana Maquira.guianensis.costaricana #> [19,] Phoebe.cinnamomifolia Cinnamomum.triplinerve #> [20,] Swartzia.simplex.var.ochnacea Swartzia.simplex.continentalis #> [21,] Tabebuia.guayacan Handroanthus.guayacan"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":null,"dir":"Reference","previous_headings":"","what":"Canonical Correlation Analysis — CCorA","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis, following Brian McArdle's unpublished graduate course notes, plus improvements allow calculations case sparse collinear matrices, permutation test Pillai's trace statistic.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Canonical Correlation Analysis — CCorA","text":"","code":"CCorA(Y, X, stand.Y=FALSE, stand.X=FALSE, permutations = 0, ...) # S3 method for CCorA biplot(x, plot.type=\"ov\", xlabs, plot.axes = 1:2, int=0.5, col.Y=\"red\", col.X=\"blue\", cex=c(0.7,0.9), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Canonical Correlation Analysis — CCorA","text":"Y Left matrix (object class: matrix data.frame). X Right matrix (object class: matrix data.frame). stand.Y Logical; Y standardized? stand.X Logical; X standardized? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x CCoaR result object. plot.type character string indicating following plots produced: \"objects\", \"variables\", \"ov\" (separate graphs objects variables), \"biplots\". unambiguous subset containing first letters names can used instead full names. xlabs Row labels. default use row names, NULL uses row numbers instead, NA suppresses plotting row names completely. plot.axes vector 2 values containing order numbers canonical axes plotted. Default: first two axes. int Radius inner circles plotted visual references plots variables. Default: int=0.5. int=0, inner circle plotted. col.Y Color used objects variables first data table (Y) plots. biplots, objects black. col.X Color used objects variables second data table (X) plots. cex vector 2 values containing size reduction factors object variable names, respectively, plots. Default values: cex=c(0.7,0.9). ... arguments passed functions. function biplot.CCorA passes graphical arguments biplot biplot.default. CCorA currently ignores extra arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Canonical Correlation Analysis — CCorA","text":"Canonical correlation analysis (Hotelling 1936) seeks linear combinations variables Y maximally correlated linear combinations variables X. analysis estimates relationships displays graphs. Pillai's trace statistic computed tested parametrically (F-test); permutation test also available. Algorithmic note -- blunt approach read two matrices, compute covariance matrices, matrix S12 %*% inv(S22) %*% t(S12) %*% inv(S11). trace Pillai's trace statistic. approach may fail, however, heavy multicollinearity sparse data matrices. safe approach replace data matrices PCA object scores. function can produce different types plots depending option chosen: \"objects\" produces two plots objects, one space Y, second space X; \"variables\" produces two plots variables, one variables Y space Y, second variables X space X; \"ov\" produces four plots, two objects two variables; \"biplots\" produces two biplots, one first matrix (Y) one second matrix (X) solutions. biplots, function passes arguments biplot.default; consult help page configuring biplots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Canonical Correlation Analysis — CCorA","text":"Function CCorA returns list containing following elements: Pillai Pillai's trace statistic = sum canonical eigenvalues. Eigenvalues Canonical eigenvalues. squares canonical correlations. CanCorr Canonical correlations. Mat.ranks Ranks matrices Y X. RDA.Rsquares Bimultivariate redundancy coefficients (R-squares) RDAs Y|X X|Y. RDA.adj.Rsq RDA.Rsquares adjusted n number explanatory variables. nperm Number permutations. p.Pillai Parametric probability value associated Pillai's trace. p.perm Permutational probability associated Pillai's trace. Cy Object scores Y biplot. Cx Object scores X biplot. corr.Y.Cy Scores Y variables Y biplot, computed cor(Y,Cy). corr.X.Cx Scores X variables X biplot, computed cor(X,Cx). corr.Y.Cx cor(Y,Cy) available plotting variables Y space X manually. corr.X.Cy cor(X,Cx) available plotting variables X space Y manually. control list control values permutations returned function . call Call CCorA function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Canonical Correlation Analysis — CCorA","text":"Hotelling, H. 1936. Relations two sets variates. Biometrika 28: 321-377. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Canonical Correlation Analysis — CCorA","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal. Implemented vegan help Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/CCorA.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Canonical Correlation Analysis — CCorA","text":"","code":"# Example using two mite groups. The mite data are available in vegan data(mite) # Two mite species associations (Legendre 2005, Fig. 4) group.1 <- c(1,2,4:8,10:15,17,19:22,24,26:30) group.2 <- c(3,9,16,18,23,25,31:35) # Separate Hellinger transformations of the two groups of species mite.hel.1 <- decostand(mite[,group.1], \"hel\") mite.hel.2 <- decostand(mite[,group.2], \"hel\") rownames(mite.hel.1) = paste(\"S\",1:nrow(mite),sep=\"\") rownames(mite.hel.2) = paste(\"S\",1:nrow(mite),sep=\"\") out <- CCorA(mite.hel.1, mite.hel.2) out #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mite.hel.1, X = mite.hel.2) #> #> Y X #> Matrix Ranks 24 11 #> #> Pillai's trace: 4.573009 #> #> Significance of Pillai's trace: #> from F-distribution: 0.0032737 #> CanAxis1 CanAxis2 CanAxis3 CanAxis4 CanAxis5 CanAxis6 #> Canonical Correlations 0.92810 0.82431 0.81209 0.74981 0.70795 0.65950 #> CanAxis7 CanAxis8 CanAxis9 CanAxis10 CanAxis11 #> Canonical Correlations 0.50189 0.48179 0.41089 0.37823 0.28 #> #> Y | X X | Y #> RDA R squares 0.33224 0.5376 #> adj. RDA R squares 0.20560 0.2910 #> biplot(out, \"ob\") # Two plots of objects biplot(out, \"v\", cex=c(0.7,0.6)) # Two plots of variables biplot(out, \"ov\", cex=c(0.7,0.6)) # Four plots (2 for objects, 2 for variables) biplot(out, \"b\", cex=c(0.7,0.6)) # Two biplots biplot(out, xlabs = NA, plot.axes = c(3,5)) # Plot axes 3, 5. No object names biplot(out, plot.type=\"biplots\", xlabs = NULL) # Replace object names by numbers # Example using random numbers. No significant relationship is expected mat1 <- matrix(rnorm(60),20,3) mat2 <- matrix(rnorm(100),20,5) out2 = CCorA(mat1, mat2, permutations=99) out2 #> #> Canonical Correlation Analysis #> #> Call: #> CCorA(Y = mat1, X = mat2, permutations = 99) #> #> Y X #> Matrix Ranks 3 5 #> #> Pillai's trace: 0.480458 #> #> Significance of Pillai's trace: #> from F-distribution: 0.90606 #> based on permutations: 0.94 #> Permutation: free #> Number of permutations: 99 #> #> CanAxis1 CanAxis2 CanAxis3 #> Canonical Correlations 0.64421 0.23458 0.1021 #> #> Y | X X | Y #> RDA R squares 0.214302 0.0839 #> adj. RDA R squares -0.066305 -0.0879 #> biplot(out2, \"b\")"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Add New Points to NMDS ordination — MDSaddpoints","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Add new points existing metaMDS monoMDS ordination.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"MDSaddpoints(nmds, dis, neighbours = 5, maxit = 200) dist2xy(dist, pick, type = c(\"xy\", \"xx\"), invert = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"nmds Result object metaMDS monoMDS. configuration points fixed, new points added. dis Rectangular non-symmetric dissimilarity matrix among new points (rows) old fixed points (columns). matrix can extracted complete dissimilarities old new points dist2xy, calculated designdist2. neighbours Number nearest points used get starting locations new points. maxit Maximum number iterations. dist Input dissimilarities. pick Indices (integers) selected observations logical vector TRUE picked items. output original order reordered argument. type \"xy\" returns rectangular data picked picked observations, \"xx\" subset symmetric dissimilarities. invert Invert pick: drop elements listed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function return list class \"nmds\" (objects type vegan) following elements points Coordinates added new points seeds Starting coordinates new points. deltastress Change stress added points. iters Number iterations. cause Cause termination iterations. Integer convergence criteria monoMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"Function provides interface monoMDS Fortran code add new points existing ordination regarded fixed. function similar role predict functions newdata Euclidean ordination (e.g. predict.cca). Input data must rectangular matrix distances among new added points (rows) fixed old points (columns). matrices can extracted complete dissimilarities helper function dist2xy. Function designdist2 can directly calculate rectangular dissimilarity matrices sampling units (rows) two matries. addition, analogue distance function can calculate dissimilarities among two matrices, including functions specified designdist2. Great care needed preparing dissimilarities input. dissimilarity index must exactly fixed ordination, columns must match old fixed points, rows added new points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSaddpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add New Points to NMDS ordination — MDSaddpoints","text":"","code":"## Cross-validation: remove a point when performing NMDS and add as ## a new points data(dune) d <- vegdist(dune) ## remove point 3 from ordination mod3 <- metaMDS(dist2xy(d, 3, \"xx\", invert = TRUE), trace=0) ## add point 3 to the result MDSaddpoints(mod3, dist2xy(d, 3)) #> $points #> MDS1 MDS2 #> 3 -0.09726881 -0.4560917 #> #> $seed #> [,1] [,2] #> [1,] 0.01222493 -0.4353716 #> #> $deltastress #> [1] 0.001811377 #> #> $iters #> [1] 18 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\" ## Use designdist2 d15 <- designdist(dune[1:15,]) m15 <- metaMDS(d15, trace=0) MDSaddpoints(m15, designdist2(dune[1:15,], dune[16:20,])) #> $points #> MDS1 MDS2 #> 16 0.4296366 -0.3068902 #> 17 -0.4122463 0.4118309 #> 18 -0.1135994 0.3120708 #> 19 0.1421922 0.4869128 #> 20 0.5892538 0.1751219 #> #> $seed #> [,1] [,2] #> [1,] 0.2832566 0.019571976 #> [2,] -0.1920994 0.137270818 #> [3,] -0.1529998 0.094374124 #> [4,] -0.0244597 0.137988191 #> [5,] 0.3212258 0.009941098 #> #> $deltastress #> [1] 0.03394768 #> #> $iters #> [1] 33 #> #> $cause #> [1] 4 #> #> attr(,\"class\") #> [1] \"nmds\""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":null,"dir":"Reference","previous_headings":"","what":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function rotates multidimensional scaling result first dimension parallel external (environmental variable). function can handle results metaMDS monoMDS functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"MDSrotate(object, vec, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"object result object metaMDS monoMDS. vec environmental variable matrix variables. number variables must lower number dimensions, solution rotated variables order appear matrix. Alternatively vec can factor, solution rotated optimal separation factor levels using lda. na.rm Remove missing values continuous variable vec. ... arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"orientation rotation undefined multidimensional scaling. Functions metaMDS metaMDS can rotate solutions principal components dispersion points highest first dimension. Sometimes different rotation intuitive, MDSrotate allows rotation result first axis parallel given external variable two first variables completely two-dimensional plane etc. several external variables supplied, applied order matrix. First axis rotated first supplied variable, second axis second variable. variables usually correlated, second variable usually aligned second axis, uncorrelated later dimensions. must least one free dimension: number external variables must lower number dimensions, used environmental variables uncorrelated free dimension. Alternatively method can rotate discriminate levels factor using linear discriminant analysis (lda). hardly meaningful two-dimensional solutions, since rotations two dimensions separation cluster levels. However, function can useful finding two-dimensional projection clusters two dimensions. last dimension always show residual variation, \\(k\\) dimensions, \\(k-1\\) discrimination vectors used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Function returns original ordination result, rotated scores (site species available), pc attribute scores set FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Rotation factor variable experimental feature may removed. discriminant analysis weights dimensions discriminating power, MDSrotate performs rigid rotation. Therefore solution may optimal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MDSrotate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rotate First MDS Dimension Parallel to an External Variable — MDSrotate","text":"","code":"data(varespec) data(varechem) mod <- monoMDS(vegdist(varespec)) mod <- with(varechem, MDSrotate(mod, pH)) plot(mod) ef <- envfit(mod ~ pH, varechem, permutations = 0) plot(ef) ordisurf(mod ~ pH, varechem, knots = 1, add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ poly(x1, 1) + poly(x2, 1) #> Total model degrees of freedom 3 #> #> REML score: -2.736059"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":null,"dir":"Reference","previous_headings":"","what":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds & Shaw test concerns location highest (hump) lowest (pit) value quadratic curve given points. Typically, used study whether quadratic hump pit located within studied interval. current test generalized applies generalized linear models (glm) link function instead simple quadratic curve. test popularized ecology analysis humped species richness patterns (Mittelbach et al. 2001), general. logarithmic link function, quadratic response defines Gaussian response model ecological gradients (ter Braak & Looman 1986), test can used inspecting location Gaussian optimum within given range gradient. can also used replace Tokeshi's test “bimodal” species frequency distribution.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"MOStest(x, y, interval, ...) # S3 method for MOStest plot(x, which = c(1,2,3,6), ...) fieller.MOStest(object, level = 0.95) # S3 method for MOStest profile(fitted, alpha = 0.01, maxsteps = 10, del = zmax/5, ...) # S3 method for MOStest confint(object, parm = 1, level = 0.95, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"x independent variable plotting object plot. y dependent variable. interval two points test statistic evaluated. missing, extremes x used. Subset plots produced. Values =1 2 define plots specific MOStest (see Details), larger values select graphs plot.lm (minus 2). object, fitted result object MOStest. level confidence level required. alpha Maximum significance level allowed. maxsteps Maximum number steps profile. del step length parameter profile (see code). parm Ignored. ... variables passed functions. Function MOStest passes glm can include family. functions pass underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function fits quadratic curve \\(\\mu = b_0 + b_1 x + b_2 x^2\\) given family link function. \\(b_2 < 0\\), defines unimodal curve highest point \\(u = -b_1/(2 b_2)\\) (ter Braak & Looman 1986). \\(b_2 > 0\\), parabola minimum \\(u\\) response sometimes called “bimodal”. null hypothesis extreme point \\(u\\) located within interval given points \\(p_1\\) \\(p_2\\). extreme point \\(u\\) exactly \\(p_1\\), \\(b_1 = 0\\) shifted axis \\(x - p_1\\). test, origin x shifted values \\(p_1\\) \\(p_2\\), test statistic based differences deviances original model model origin forced given location using standard anova.glm function (Oksanen et al. 2001). Mitchell-Olds & Shaw (1987) used first degree coefficient significance estimated summary.glm function. give identical results Normal error, error distributions preferable use test based differences deviances fitted models. test often presented general test location hump, really dependent quadratic fitted curve. hump different form quadratic, test may insignificant. strong assumptions test, use support functions inspect fit. Function plot(..., =1) displays data points, fitted quadratic model, approximate 95% confidence intervals (2 times SE). Function plot = 2 displays approximate confidence interval polynomial coefficients, together two lines indicating combinations coefficients produce evaluated points x. Moreover, cross-hair shows approximate confidence intervals polynomial coefficients ignoring correlations. Higher values produce corresponding graphs plot.lm. , must add 2 value plot.lm. Function fieller.MOStest approximates confidence limits location extreme point (hump pit) using Fieller's theorem following ter Braak & Looman (1986). test based quasideviance except family poisson binomial. Function profile evaluates profile deviance fitted model, confint finds profile based confidence limits following Oksanen et al. (2001). test typically used assessing significance diversity hump productivity gradient (Mittelbach et al. 2001). also can used location pit (deepest points) instead Tokeshi test. , can used test location Gaussian optimum ecological gradient analysis (ter Braak & Looman 1986, Oksanen et al. 2001).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"function based glm, returns result object glm amended result test. new items MOStest : isHump TRUE response hump. isBracketed TRUE hump pit bracketed evaluated points. hump Sorted vector location hump pit points test evaluated. coefficients Table test statistics significances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Mitchell-Olds, T. & Shaw, R.G. 1987. Regression analysis natural selection: statistical inference biological interpretation. Evolution 41, 1149--1161. Mittelbach, G.C. Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, R.M., Dodson, S.. & Gough, L. 2001. observed relationship species richness productivity? Ecology 82, 2381--2396. Oksanen, J., Läärä, E., Tolonen, K. & Warner, B.G. 2001. Confidence intervals optimum Gaussian response function. Ecology 82, 1191--1197. ter Braak, C.J.F & Looman, C.W.N 1986. Weighted averaging, logistic regression Gaussian response model. Vegetatio 65, 3--11.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"Function fieller.MOStest based package optgrad Ecological Archives (https://figshare.com/articles/dataset/Full_Archive/3521975) accompanying Oksanen et al. (2001). Ecological Archive package optgrad also contains profile deviance method location hump pit, current implementation profile confint rather follow example profile.glm confint.glm MASS package.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/MOStest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mitchell-Olds and Shaw Test for the Location of Quadratic Extreme — MOStest","text":"","code":"## The Al-Mufti data analysed in humpfit(): mass <- c(140,230,310,310,400,510,610,670,860,900,1050,1160,1900,2480) spno <- c(1, 4, 3, 9, 18, 30, 20, 14, 3, 2, 3, 2, 5, 2) mod <- MOStest(mass, spno) ## Insignificant mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: gaussian #> Link function: identity #> #> hump min max #> 46.89749 140.00000 2480.00000 #> ***** Caution: hump/pit not bracketed by the data ****** #> #> min/max F Pr(>F) #> hump at min 140 0.0006 0.9816 #> hump at max 2480 0.3161 0.5852 #> Combined 0.9924 ## ... but inadequate shape of the curve op <- par(mfrow=c(2,2), mar=c(4,4,1,1)+.1) plot(mod) ## Looks rather like log-link with Poisson error and logarithmic biomass mod <- MOStest(log(mass), spno, family=quasipoisson) mod #> #> Mitchell-Olds and Shaw test #> Null: hump of a quadratic linear predictor is at min or max #> #> Family: quasipoisson #> Link function: log #> #> min hump max #> 4.941642 6.243371 7.816014 #> #> min/max F Pr(>F) #> hump at min 4.9416 7.1367 0.02174 * #> hump at max 7.8160 9.0487 0.01191 * #> Combined 0.03338 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod) par(op) ## Confidence Limits fieller.MOStest(mod) #> 2.5 % 97.5 % #> 5.255827 6.782979 confint(mod) #> 2.5 % 97.5 % #> 5.816021 6.574378 plot(profile(mod))"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":null,"dir":"Reference","previous_headings":"","what":"Adjusted R-square — RsquareAdj","title":"Adjusted R-square — RsquareAdj","text":"functions finds adjusted R-square.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Adjusted R-square — RsquareAdj","text":"","code":"# S3 method for default RsquareAdj(x, n, m, ...) # S3 method for rda RsquareAdj(x, ...) # S3 method for cca RsquareAdj(x, permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Adjusted R-square — RsquareAdj","text":"x Unadjusted R-squared object terms evaluation adjusted R-squared can found. n, m Number observations number degrees freedom fitted model. permutations Number permutations use computing adjusted R-squared cca. permutations can calculated parallel specifying number cores passed permutest ... arguments (ignored) except case cca arguments passed permutest.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Adjusted R-square — RsquareAdj","text":"default method finds adjusted \\(R^2\\) unadjusted \\(R^2\\), number observations, number degrees freedom fitted model. specific methods find information fitted result object. specific methods rda (also used distance-based RDA), cca, lm glm. Adjusted, even unadjusted, \\(R^2\\) may available cases, functions return NA. \\(R^2\\) values available gaussian models glm. adjusted, \\(R^2\\) cca computed using permutation approach developed Peres-Neto et al. (2006). default 1000 permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Adjusted R-square — RsquareAdj","text":"functions return list items r.squared adj.r.squared.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Adjusted R-square — RsquareAdj","text":"Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277. Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87, 2614--2625.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/RsquareAdj.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Adjusted R-square — RsquareAdj","text":"","code":"data(mite) data(mite.env) ## rda m <- rda(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.5265047 #> #> $adj.r.squared #> [1] 0.4367038 #> ## cca m <- cca(decostand(mite, \"hell\") ~ ., mite.env) RsquareAdj(m) #> $r.squared #> [1] 0.4471676 #> #> $adj.r.squared #> [1] 0.3446879 #> ## default method RsquareAdj(0.8, 20, 5) #> [1] 0.7285714"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":null,"dir":"Reference","previous_headings":"","what":"Self-Starting nls Species-Area Models — SSarrhenius","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions provide self-starting species-area models non-linear regression (nls). can also used fitting species accumulation models fitspecaccum. models (many ) reviewed Dengler (2009).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"SSarrhenius(area, k, z) SSgleason(area, k, slope) SSgitay(area, k, slope) SSlomolino(area, Asym, xmid, slope)"},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"area Area size sample: independent variable. k, z, slope, Asym, xmid Estimated model parameters: see Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"functions assumed used species richness (number species) independent variable, area sample size independent variable. Basically, define least squares models untransformed data, differ models transformed species richness models non-Gaussian error. Arrhenius model (SSarrhenius) expression k*area^z. classical model can found textbook ecology (also Dengler 2009). Parameter z steepness species-area curve, k expected number species unit area. Gleason model (SSgleason) linear expression k + slope*log(area) (Dengler 200). linear model, starting values give final estimates; provided ease comparison models. Gitay model (SSgitay) quadratic logarithmic expression (k + slope*log(area))^2 (Gitay et al. 1991, Dengler 2009). Parameter slope steepness species-area curve, k square root expected richness unit area. Lomolino model (SSlomolino) Asym/(1 + slope^log(xmid/area)) (Lomolino 2000, Dengler 2009). Parameter Asym asymptotic maximum number species, slope maximum slope increase richness, xmid area half maximum richness achieved. addition models, several models studied Dengler (2009) available standard R self-starting models: Michaelis-Menten (SSmicmen), Gompertz (SSgompertz), logistic (SSlogis), Weibull (SSweibull), others may useful.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Numeric vector length area. value expression model. arguments names objects gradient matrix respect names attached attribute named gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Dengler, J. (2009) function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gitay, H., Roxburgh, S.H. & Wilson, J.B. (1991) Species-area relationship New Zealand tussock grassland, implications nature reserve design community structure. Journal Vegetation Science 2, 113--118. Lomolino, M. V. (2000) Ecology's general, yet protean pattern: species-area relationship. Journal Biogeography 27, 17--26.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/SSarrhenius.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Self-Starting nls Species-Area Models — SSarrhenius","text":"","code":"## Get species area data: sipoo.map gives the areas of islands data(sipoo, sipoo.map) S <- specnumber(sipoo) plot(S ~ area, sipoo.map, xlab = \"Island Area (ha)\", ylab = \"Number of Species\", ylim = c(1, max(S))) ## The Arrhenius model marr <- nls(S ~ SSarrhenius(area, k, z), data=sipoo.map) marr #> Nonlinear regression model #> model: S ~ SSarrhenius(area, k, z) #> data: sipoo.map #> k z #> 3.4062 0.4364 #> residual sum-of-squares: 78.1 #> #> Number of iterations to convergence: 5 #> Achieved convergence tolerance: 1.056e-06 ## confidence limits from profile likelihood confint(marr) #> Waiting for profiling to be done... #> 2.5% 97.5% #> k 2.6220312 4.3033906 #> z 0.3813576 0.4944693 ## draw a line xtmp <- with(sipoo.map, seq(min(area), max(area), len=51)) lines(xtmp, predict(marr, newdata=data.frame(area = xtmp)), lwd=2) ## The normal way is to use linear regression on log-log data, ## but this will be different from the previous: mloglog <- lm(log(S) ~ log(area), data=sipoo.map) mloglog #> #> Call: #> lm(formula = log(S) ~ log(area), data = sipoo.map) #> #> Coefficients: #> (Intercept) log(area) #> 1.0111 0.4925 #> lines(xtmp, exp(predict(mloglog, newdata=data.frame(area=xtmp))), lty=2) ## Gleason: log-linear mgle <- nls(S ~ SSgleason(area, k, slope), sipoo.map) lines(xtmp, predict(mgle, newdata=data.frame(area=xtmp)), lwd=2, col=2) ## Gitay: quadratic of log-linear mgit <- nls(S ~ SSgitay(area, k, slope), sipoo.map) lines(xtmp, predict(mgit, newdata=data.frame(area=xtmp)), lwd=2, col = 3) ## Lomolino: using original names of the parameters (Lomolino 2000): mlom <- nls(S ~ SSlomolino(area, Smax, A50, Hill), sipoo.map) mlom #> Nonlinear regression model #> model: S ~ SSlomolino(area, Smax, A50, Hill) #> data: sipoo.map #> Smax A50 Hill #> 53.493 94.697 2.018 #> residual sum-of-squares: 55.37 #> #> Number of iterations to convergence: 6 #> Achieved convergence tolerance: 9.715e-07 lines(xtmp, predict(mlom, newdata=data.frame(area=xtmp)), lwd=2, col = 4) ## One canned model of standard R: mmic <- nls(S ~ SSmicmen(area, Asym, slope), sipoo.map) lines(xtmp, predict(mmic, newdata = data.frame(area=xtmp)), lwd =2, col = 5) legend(\"bottomright\", c(\"Arrhenius\", \"log-log linear\", \"Gleason\", \"Gitay\", \"Lomolino\", \"Michaelis-Menten\"), col=c(1,1,2,3,4,5), lwd=c(2,1,2,2,2,2), lty=c(1,2,1,1,1,1)) ## compare models (AIC) allmods <- list(Arrhenius = marr, Gleason = mgle, Gitay = mgit, Lomolino = mlom, MicMen= mmic) sapply(allmods, AIC) #> Arrhenius Gleason Gitay Lomolino MicMen #> 83.49847 96.94018 80.54984 79.30718 83.02003"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Compute single terms can added dropped constrained ordination model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"# S3 method for cca add1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...) # S3 method for cca drop1(object, scope, test = c(\"none\", \"permutation\"), permutations = how(nperm=199), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"object constrained ordination object cca, rda, dbrda capscale. scope formula giving terms considered adding dropping; see add1 details. test permutation test added using anova.cca. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ... arguments passed add1.default, drop1.default, anova.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"argument test = \"none\" functions call add1.default drop1.default. argument test = \"permutation\" functions add test results anova.cca. Function drop1.cca call anova.cca argument = \"margin\". Function add1.cca implement test single term additions directly available anova.cca. Functions used implicitly step, ordiR2step ordistep. deviance.cca deviance.rda used step firm basis, setting argument test = \"permutation\" may help getting useful insight validity model building. Function ordistep calls alternately drop1.cca add1.cca argument test = \"permutation\" selects variables permutation \\(P\\)-values. Meticulous use add1.cca drop1.cca allow judicious model building. default number permutations set low value, permutation tests can take long time. sufficient give impression significances terms, higher values permutations used \\(P\\) values really important.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Returns similar object add1 drop1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/add1.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Drop Single Terms to a Constrained Ordination Model — add1.cca","text":"","code":"data(dune) data(dune.env) ## Automatic model building based on AIC but with permutation tests step(cca(dune ~ 1, dune.env), reformulate(names(dune.env)), test=\"perm\") #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 86.608 2.2536 0.005 ** #> + Management 3 86.935 2.1307 0.005 ** #> + A1 1 87.411 2.1400 0.045 * #> 87.657 #> + Manure 4 88.832 1.5251 0.030 * #> + Use 2 89.134 1.1431 0.200 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC F Pr(>F) #> 86.608 #> + Management 3 86.813 1.4565 0.045 * #> + A1 1 86.992 1.2624 0.175 #> + Use 2 87.259 1.2760 0.085 . #> + Manure 4 87.342 1.3143 0.050 * #> - Moisture 3 87.657 2.2536 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #> ## see ?ordistep to do the same, but based on permutation P-values if (FALSE) { ordistep(cca(dune ~ 1, dune.env), reformulate(names(dune.env))) } ## Manual model building ## -- define the maximal model for scope mbig <- rda(dune ~ ., dune.env) ## -- define an empty model to start with m0 <- rda(dune ~ 1, dune.env) ## -- manual selection and updating add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 89.620 #> A1 1 89.591 1.9217 0.015 * #> Moisture 3 87.707 2.5883 0.005 ** #> Management 3 87.082 2.8400 0.005 ** #> Use 2 91.032 1.1741 0.225 #> Manure 4 89.232 1.9539 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Management) add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 87.082 #> A1 1 87.424 1.2965 0.215 #> Moisture 3 85.567 1.9764 0.010 ** #> Use 2 88.284 1.0510 0.400 #> Manure 3 87.517 1.3902 0.105 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 m0 <- update(m0, . ~ . + Moisture) ## -- included variables still significant? drop1(m0, test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> Management 3 87.707 2.1769 0.010 ** #> Moisture 3 87.082 1.9764 0.015 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 add1(m0, scope=formula(mbig), test=\"perm\") #> Df AIC F Pr(>F) #> 85.567 #> A1 1 86.220 0.8359 0.605 #> Use 2 86.842 0.8027 0.765 #> Manure 3 85.762 1.1225 0.390"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"additive diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set (gamma diversity). hierarchical null model testing, statistic returned function evaluated according nested hierarchical sampling design (hiersimu).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"adipart(...) # S3 method for default adipart(y, x, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula adipart(formula, data, index=c(\"richness\", \"shannon\", \"simpson\"), weights=c(\"unif\", \"prop\"), relative = FALSE, nsimul=99, method = \"r2dtable\", ...) hiersimu(...) # S3 method for default hiersimu(y, x, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula hiersimu(formula, data, FUN, location = c(\"mean\", \"median\"), relative = FALSE, drop.highest = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, function performs overall decomposition alpha, beta gamma diversities. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variables referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, diversity index calculated (see Details). weights Character, \"unif\" uniform weights, \"prop\" weighting proportional sample abundances use weighted averaging individual alpha values within strata given level sampling hierarchy. relative Logical, TRUE alpha beta diversity values given relative value gamma function adipart. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. FUN function used hiersimu. must fully specified, currently arguments passed function via .... location Character, identifies function (mean median) used calculate location samples. drop.highest Logical, drop highest level . FUN evaluates arrays least 2 dimensions, highest level dropped, selected . ... arguments passed functions, e.g. base logarithm Shannon diversity, method, thin burnin arguments oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Additive diversity partitioning means mean alpha beta diversities add gamma diversity, thus beta diversity measured dimensions alpha gamma (Lande 1996). additive procedure extended across multiple scales hierarchical sampling design \\(= 1, 2, 3, \\ldots, m\\) levels sampling (Crist et al. 2003). Samples lower hierarchical levels nested within higher level units, thus \\(=1\\) \\(=m\\) grain size increasing constant survey extent. level \\(\\), \\(\\alpha_i\\) denotes average diversity found within samples. highest sampling level, diversity components calculated $$\\beta_m = \\gamma - \\alpha_m$$ lower sampling level $$\\beta_i = \\alpha_{+1} - \\alpha_i$$ , additive partition diversity $$\\gamma = \\alpha_1 + \\sum_{=1}^m \\beta_i$$ Average alpha components can weighted uniformly (weight=\"unif\") calculate simple average, proportionally sample abundances (weight=\"prop\") calculate weighted average follows $$\\alpha_i = \\sum_{j=1}^{n_i} D_{ij} w_{ij}$$ \\(D_{ij}\\) diversity index \\(w_{ij}\\) weight calculated \\(j\\)th sample \\(\\)th sampling level. implementation additive diversity partitioning adipart follows Crist et al. 2003. based species richness (\\(S\\), \\(S-1\\)), Shannon's Simpson's diversity indices stated index argument. expected diversity components calculated nsimul times individual based randomisation community data matrix. done \"r2dtable\" method oecosimu default. hiersimu works almost way adipart, without comparing actual statistic values returned FUN highest possible value (cf. gamma diversity). , cases, difficult ensure additive properties mean statistic values along hierarchy.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"object class \"adipart\" \"hiersimu\" structure oecosimu objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Crist, T.O., Veech, J.., Gering, J.C. Summerville, K.S. (2003). Partitioning species diversity across landscapes regions: hierarchical analysis \\(\\alpha\\), \\(\\beta\\), \\(\\gamma\\)-diversity. . Nat., 162, 734--743. Lande, R. (1996). Statistics partitioning species diversity, similarity among multiple communities. Oikos, 76, 5--13.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Additive Diversity Partitioning and Hierarchical Null Model Testing — adipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l1=1:nrow(mite), l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)), l4=rep(1, nrow(mite)))) ## Let's see in a map par(mfrow=c(1,3)) plot(mite.xy, main=\"l1\", col=as.numeric(levsm$l1)+1, asp = 1) plot(mite.xy, main=\"l2\", col=as.numeric(levsm$l2)+1, asp = 1) plot(mite.xy, main=\"l3\", col=as.numeric(levsm$l3)+1, asp = 1) par(mfrow=c(1,1)) ## Additive diversity partitioning adipart(mite, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -28.979 22.393 21.978 22.429 22.765 0.05 * #> gamma 35.000 0.000 35.000 35.000 35.000 35.000 1.00 #> beta.1 19.886 28.979 12.607 12.235 12.571 13.022 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## the next two define identical models adipart(mite, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(y = mite, x = levsm, index = \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -54.1889 22.35038 22.10429 22.37143 22.584 0.05 * #> alpha.2 29.750 -25.1503 34.81579 34.50000 34.75000 35.000 0.05 * #> alpha.3 33.000 0.0000 35.00000 35.00000 35.00000 35.000 0.05 * #> gamma 35.000 0.0000 35.00000 35.00000 35.00000 35.000 1.00 #> beta.1 14.636 9.9124 12.46541 12.10000 12.45000 12.779 0.05 * #> beta.2 3.250 15.2208 0.18421 0.00000 0.25000 0.500 0.05 * #> beta.3 2.000 0.0000 0.00000 0.00000 0.00000 0.000 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adipart(mite ~ l2 + l3, levsm, index=\"richness\", nsimul=19) #> adipart object #> #> Call: adipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"richness\", nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index richness, weights unif #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 15.114 -51.771 22.378947 22.100714 22.400000 22.587 0.05 * #> alpha.2 29.750 -25.578 34.736842 34.500000 34.750000 35.000 0.05 * #> alpha.3 33.000 -17.206 34.973684 34.725000 35.000000 35.000 0.05 * #> gamma 35.000 0.000 35.000000 35.000000 35.000000 35.000 1.00 #> beta.1 14.636 11.165 12.357895 12.025714 12.428571 12.686 0.05 * #> beta.2 3.250 15.455 0.236842 0.000000 0.250000 0.500 0.05 * #> beta.3 2.000 17.206 0.026316 0.000000 0.000000 0.275 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical null model testing ## diversity analysis (similar to adipart) hiersimu(mite, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(y = mite, FUN = diversity, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> level_1 0.76064 -52.201 0.93892 0.93378 0.93866 0.9444 0.05 * #> leve_2 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 hiersimu(mite ~ l2 + l3, levsm, FUN=diversity, relative=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ l2 + l3, data = levsm, FUN = diversity, #> relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> unit 0.76064 -60.783 0.93927 0.93425 0.93883 0.9435 0.05 * #> l2 0.89736 -116.072 0.99795 0.99617 0.99808 0.9991 0.05 * #> l3 0.92791 -421.205 0.99932 0.99902 0.99935 0.9995 0.05 * #> all 1.00000 0.000 1.00000 1.00000 1.00000 1.0000 1.00 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Hierarchical testing with the Morisita index morfun <- function(x) dispindmorisita(x)$imst hiersimu(mite ~., levsm, morfun, drop.highest=TRUE, nsimul=19) #> hiersimu object #> #> Call: hiersimu(formula = mite ~ ., data = levsm, FUN = morfun, #> drop.highest = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> l1 0.52070 6.4968 0.35648 0.31140 0.35449 0.3978 0.05 * #> l2 0.60234 12.5142 0.15359 0.08900 0.14525 0.2039 0.05 * #> l3 0.67509 17.6125 -0.19189 -0.26300 -0.20619 -0.1149 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Analysis variance using distance matrices --- partitioning distance matrices among sources variation fitting linear models (e.g., factors, polynomial regression) distance matrices; uses permutation test pseudo-\\(F\\) ratios.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"adonis2(formula, data, permutations = 999, method = \"bray\", sqrt.dist = FALSE, add = FALSE, by = \"terms\", parallel = getOption(\"mc.cores\"), na.action = na.fail, strata = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"formula Model formula. left-hand side (LHS) formula must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist used find dissimilarities. right-hand side (RHS) formula defines independent variables. can continuous variables factors, can transformed within formula, can interactions typical formula. data data frame independent variables, rows order community data matrix dissimilarity matrix named LHS formula. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. method name method used vegdist calculate pairwise distances left hand side formula data frame matrix. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" analyse one-degree--freedom contrasts sequentially, = NULL assess overall significance terms together. arguments passed anova.cca. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. na.action Handling missing values right-hand-side formula (see na.fail explanation alternatives). Missing values allowed left-hand-side. NB, argument subset implemented. strata Groups within constrain permutations. traditional non-movable strata set Blocks permute package, flexible alternatives may appropriate. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"adonis2 function analysis partitioning sums squares using dissimilarities. function based principles McArdle & Anderson (2001) can perform sequential, marginal overall tests. function also allows using additive constants squareroot dissimilarities avoid negative eigenvalues, can also handle semimetric indices (Bray-Curtis) produce negative eigenvalues. adonis2 tests identical anova.cca dbrda. Euclidean distances, tests also identical anova.cca rda. function partitions sums squares multivariate data set, directly analogous MANOVA (multivariate analysis variance). McArdle Anderson (2001) Anderson (2001) refer method “permutational MANOVA” (formerly “nonparametric MANOVA”). , inputs linear predictors, response matrix arbitrary number columns, robust alternative parametric MANOVA ordination methods describing variation attributed different experimental treatments uncontrolled covariates. method also analogous distance-based redundancy analysis algorithmically similar dbrda (Legendre Anderson 1999), provides alternative AMOVA (nested analysis molecular variance, Excoffier, Smouse, Quattro, 1992; amova ade4 package) crossed nested factors.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"function returns anova.cca result object new column partial \\(R^2\\): proportion sum squares total, marginal models (= \"margin\") \\(R^2\\) terms add 1.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson (2001, Fig. 4) warns method may confound location dispersion effects: significant differences may caused different within-group variation (dispersion) instead different mean values groups (see Warton et al. 2012 general analysis). However, seems adonis2 less sensitive dispersion effects alternatives (anosim, mrpp). Function betadisper sister function adonis2 study differences dispersion within geometric framework.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Anderson, M.J. 2001. new method non-parametric multivariate analysis variance. Austral Ecology, 26: 32--46. Excoffier, L., P.E. Smouse, J.M. Quattro. 1992. Analysis molecular variance inferred metric distances among DNA haplotypes: Application human mitochondrial DNA restriction data. Genetics, 131:479--491. Legendre, P. M.J. Anderson. 1999. Distance-based redundancy analysis: Testing multispecies responses multifactorial ecological experiments. Ecological Monographs, 69:1--24. McArdle, B.H. M.J. Anderson. 2001. Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology, 82: 290--297. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"Martin Henry H. Stevens Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/adonis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutational Multivariate Analysis of Variance Using Distance Matrices — adonis","text":"","code":"data(dune) data(dune.env) ## default test by terms adonis2(dune ~ Management*A1, data = dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.2629 0.002 ** #> A1 1 0.4409 0.10256 2.9387 0.012 * #> Management:A1 3 0.5892 0.13705 1.3090 0.217 #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## overall tests adonis2(dune ~ Management*A1, data = dune.env, by = NULL) #> Permutation test for adonis under reduced model #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management * A1, data = dune.env, by = NULL) #> Df SumOfSqs R2 F Pr(>F) #> Model 7 2.4987 0.58122 2.3792 0.004 ** #> Residual 12 1.8004 0.41878 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ### Example of use with strata, for nested (e.g., block) designs. dat <- expand.grid(rep=gl(2,1), NO3=factor(c(0,10)),field=gl(3,1) ) dat #> rep NO3 field #> 1 1 0 1 #> 2 2 0 1 #> 3 1 10 1 #> 4 2 10 1 #> 5 1 0 2 #> 6 2 0 2 #> 7 1 10 2 #> 8 2 10 2 #> 9 1 0 3 #> 10 2 0 3 #> 11 1 10 3 #> 12 2 10 3 Agropyron <- with(dat, as.numeric(field) + as.numeric(NO3)+2) +rnorm(12)/2 Schizachyrium <- with(dat, as.numeric(field) - as.numeric(NO3)+2) +rnorm(12)/2 total <- Agropyron + Schizachyrium dotplot(total ~ NO3, dat, jitter.x=TRUE, groups=field, type=c('p','a'), xlab=\"NO3\", auto.key=list(columns=3, lines=TRUE) ) Y <- data.frame(Agropyron, Schizachyrium) mod <- metaMDS(Y, trace = FALSE) plot(mod) ### Ellipsoid hulls show treatment with(dat, ordiellipse(mod, NO3, kind = \"ehull\", label = TRUE)) ### Spider shows fields with(dat, ordispider(mod, field, lty=3, col=\"red\", label = TRUE)) ### Incorrect (no strata) adonis2(Y ~ NO3, data = dat, permutations = 199) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.06 . #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Correct with strata with(dat, adonis2(Y ~ NO3, data = dat, permutations = 199, strata = field)) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Blocks: strata #> Permutation: free #> Number of permutations: 199 #> #> adonis2(formula = Y ~ NO3, data = dat, permutations = 199, strata = field) #> Df SumOfSqs R2 F Pr(>F) #> NO3 1 0.036688 0.24632 3.2682 0.005 ** #> Residual 10 0.112256 0.75368 #> Total 11 0.148944 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":null,"dir":"Reference","previous_headings":"","what":"Analysis of Similarities — anosim","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Analysis of Similarities — anosim","text":"","code":"anosim(x, grouping, permutations = 999, distance = \"bray\", strata = NULL, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Analysis of Similarities — anosim","text":"x Data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. distance Choice distance metric measures dissimilarity two observations. See vegdist options. used x dissimilarity structure symmetric square matrix. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Analysis of Similarities — anosim","text":"Analysis similarities (ANOSIM) provides way test statistically whether significant difference two groups sampling units. Function anosim operates directly dissimilarity matrix. suitable dissimilarity matrix produced functions dist vegdist. method philosophically allied NMDS ordination (monoMDS), uses rank order dissimilarity values. two groups sampling units really different species composition, compositional dissimilarities groups greater within groups. anosim statistic \\(R\\) based difference mean ranks groups (\\(r_B\\)) within groups (\\(r_W\\)): $$R = (r_B - r_W)/(N (N-1) / 4)$$ divisor chosen \\(R\\) interval \\(-1 \\dots +1\\), value \\(0\\) indicating completely random grouping. statistical significance observed \\(R\\) assessed permuting grouping vector obtain empirical distribution \\(R\\) null-model. See permutations additional details permutation tests Vegan. distribution simulated values can inspected permustats function. function summary plot methods. show valuable information assess validity method: function assumes ranked dissimilarities within groups equal median range. plot method uses boxplot options notch=TRUE varwidth=TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Analysis of Similarities — anosim","text":"function returns list class \"anosim\" following items: call Function call. statistic value ANOSIM statistic \\(R\\) signif Significance permutation. perm Permutation values \\(R\\). distribution permutation values can inspected function permustats. class.vec Factor value dissimilarities classes class name corresponding dissimilarity within class. dis.rank Rank dissimilarity entry. dissimilarity name dissimilarity index: \"method\" entry dist object. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Analysis of Similarities — anosim","text":"Clarke, K. R. (1993). Non-parametric multivariate analysis changes community structure. Australian Journal Ecology 18, 117--143. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Analysis of Similarities — anosim","text":"Jari Oksanen, help Peter R. Minchin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Analysis of Similarities — anosim","text":"anosim function can confound differences groups dispersion within groups results can difficult interpret (cf. Warton et al. 2012). function returns lot information ease studying performance. anosim models analysed adonis2 seems robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anosim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Analysis of Similarities — anosim","text":"","code":"data(dune) data(dune.env) dune.dist <- vegdist(dune) dune.ano <- with(dune.env, anosim(dune.dist, Management)) summary(dune.ano) #> #> Call: #> anosim(x = dune.dist, grouping = Management) #> Dissimilarity: bray #> #> ANOSIM statistic R: 0.2579 #> Significance: 0.009 #> #> Permutation: free #> Number of permutations: 999 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.121 0.164 0.205 0.242 #> #> Dissimilarity ranks between and within classes: #> 0% 25% 50% 75% 100% N #> Between 4 58.50 104.00 145.500 188.0 147 #> BF 5 15.25 25.50 41.250 57.0 3 #> HF 1 7.25 46.25 68.125 89.5 10 #> NM 6 64.75 124.50 156.250 181.0 15 #> SF 3 32.75 53.50 99.250 184.0 15 #> plot(dune.ano) #> Warning: some notches went outside hinges ('box'): maybe set notch=FALSE"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function performs ANOVA like permutation test Constrained Correspondence Analysis (cca), Redundancy Analysis (rda) distance-based Redundancy Analysis (dbRDA, dbrda) assess significance constraints.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"# S3 method for cca anova(object, ..., permutations = how(nperm=999), by = NULL, model = c(\"reduced\", \"direct\", \"full\"), parallel = getOption(\"mc.cores\"), strata = NULL, cutoff = 1, scope = NULL) # S3 method for cca permutest(x, permutations = how(nperm = 99), model = c(\"reduced\", \"direct\", \"full\"), by = NULL, first = FALSE, strata = NULL, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"object One several result objects cca, rda, dbrda capscale. several result objects, compared order supplied. single object, test specified overall test given. x single ordination result object. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Setting = \"axis\" assess significance constrained axis, setting = \"terms\" assess significance term (sequentially first last), setting = \"margin\" assess marginal effects terms (marginal term analysed model variables), = \"onedf\" assess sequentially one-degree--freedom contrasts split factors. model Permutation model: model=\"direct\" permutes community data, model=\"reduced\" permutes residuals community data Conditions (partial model), model = \"full\" permutes residuals Conditions Constraints. parallel Use parallel processing given number cores. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. error use permutations matrix, defines blocks. legacy argument deprecated future: use permutations = (..., blocks) instead. cutoff effective =\"axis\" stops permutations axis equals exceeds cutoff \\(p\\)-value. scope effective =\"margin\" can used select marginal terms testing. default test marginal terms drop.scope. first Analyse significance first axis. ... Parameters passed functions. anova.cca passes arguments permutest.cca. anova = \"axis\" can use argument cutoff (defaults 1) stops permutations exceeding given level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Functions anova.cca permutest.cca implement ANOVA like permutation tests joint effect constraints cca, rda, dbrda capscale. Function anova intended user-friendly alternative permutest (real workhorse). Function anova can analyse sequence constrained ordination models. analysis based differences residual deviance permutations nested models. default test sum constrained eigenvalues. Setting first = TRUE perform test first constrained eigenvalue. Argument first can set either anova.cca permutest.cca. also possible perform significance tests axis term (constraining variable) using argument anova.cca. Setting = \"axis\" perform separate significance tests constrained axis. previous constrained axes used conditions (“partialled ”) test first constrained eigenvalues performed (Legendre et al. 2011). can stop permutation tests exceeding given significance level argument cutoff speed calculations large models. Setting = \"terms\" perform separate significance test term (constraining variable). terms assessed sequentially first last, order terms influence significances. Setting = \"onedf\" perform similar sequential test one-degree--freedom effects, multi-level factors split contrasts. Setting = \"margin\" perform separate significance test marginal term model terms. marginal test also accepts scope argument drop.scope can character vector term labels analysed, fitted model lower scope. marginal effects also known “Type III” effects, current function evaluates marginal terms. , instance, ignore main effects included interaction terms. calculating pseudo-\\(F\\), terms compared residual full model. Community data permuted choice model=\"direct\", residuals partial CCA/ RDA/ dbRDA choice model=\"reduced\" (default). partial CCA/ RDA/ dbRDA stage, model=\"reduced\" simply permutes data equivalent model=\"direct\". test statistic “pseudo-\\(F\\)”, ratio constrained unconstrained total Inertia (Chi-squares, variances something similar), divided respective degrees freedom. conditions (“partial” terms), sum eigenvalues remains constant, pseudo-\\(F\\) eigenvalues give equal results. partial CCA/ RDA/ dbRDA, effect conditioning variables (“covariables”) removed permutation, total Chi-square fixed, test based pseudo-\\(F\\) differ test based plain eigenvalues.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"function anova.cca calls permutest.cca fills anova table. Additional attributes Random.seed (random seeds used), control (permutation design, see ) F.perm (permuted test statistics).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Legendre, P. Legendre, L. (2012). Numerical Ecology. 3rd English ed. Elsevier. Legendre, P., Oksanen, J. ter Braak, C.J.F. (2011). Testing significance canonical axes redundancy analysis. Methods Ecology Evolution 2, 269--277.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/anova.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation Test for Constrained Correspondence Analysis, Redundancy Analysis and Constrained Analysis of Principal Coordinates — anova.cca","text":"","code":"data(dune, dune.env) mod <- cca(dune ~ Moisture + Management, dune.env) ## overall test anova(mod) #> Permutation test for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Model 6 1.0024 1.9515 0.003 ** #> Residual 13 1.1129 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## tests for individual terms anova(mod, by=\"term\") #> Permutation test for cca under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.62831 2.4465 0.001 *** #> Management 3 0.37407 1.4565 0.049 * #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 anova(mod, by=\"margin\") #> Permutation test for cca under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture 3 0.39854 1.5518 0.036 * #> Management 3 0.37407 1.4565 0.053 . #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential test for contrasts anova(mod, by = \"onedf\") #> Permutation test for cca under reduced model #> Sequential test for contrasts #> Permutation: free #> Number of permutations: 999 #> #> Model: cca(formula = dune ~ Moisture + Management, data = dune.env) #> Df ChiSquare F Pr(>F) #> Moisture.L 1 0.41081 4.7988 0.001 *** #> Moisture.Q 1 0.11261 1.3154 0.167 #> Moisture.C 1 0.10489 1.2253 0.226 #> ManagementHF 1 0.08849 1.0337 0.359 #> ManagementNM 1 0.20326 2.3744 0.011 * #> ManagementSF 1 0.08231 0.9615 0.455 #> Residual 13 1.11289 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## test for adding all environmental variables anova(mod, cca(dune ~ ., dune.env)) #> Permutation tests for cca under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model 1: dune ~ Moisture + Management #> Model 2: dune ~ A1 + Moisture + Management + Use + Manure #> ResDf ResChiSquare Df ChiSquare F Pr(>F) #> 1 13 1.1129 #> 2 7 0.6121 6 0.50079 0.9545 0.532"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Averaged Subsampled Dissimilarity Matrices — avgdist","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function computes dissimilarity matrix dataset multiple times using vegdist randomly subsampling dataset time. subsampled iterations averaged (mean) provide distance matrix represents average multiple subsampling iterations. emulates behavior distance matrix calculator within Mothur microbial ecology toolkit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"avgdist(x, sample, distfun = vegdist, meanfun = mean, transf = NULL, iterations = 100, dmethod = \"bray\", diag = TRUE, upper = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"x Community data matrix. sample subsampling depth used iteration. Samples meet threshold removed analysis, identity returned user stdout. distfun dissimilarity matrix function used. Default vegan vegdist meanfun calculation use average (mean median). transf Option transforming count data calculating distance matrix. base transformation option can used (e.g. sqrt) iterations number random iterations perform averaging. Default 100 iterations. dmethod Dissimilarity index used specified dissimilarity matrix function. Default Bray-Curtis diag, upper Return dissimilarities diagonal upper triangle. NB. default differs vegdist returns symmetric \"dist\" structure instead lower diagonal. However, object accessed matrix indices unless cast matrix .matrix. ... additional arguments add distance function mean/median function specified.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"Geoffrey Hannigan, minor tweaks Gavin L. Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"function builds function rrarefy additional distance matrix function (e.g. vegdist) add meaningful representations distances among randomly subsampled datasets presenting average multiple random iterations. function runs using vegdist. functionality utilized Mothur standalone microbial ecology toolkit .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/avgdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Averaged Subsampled Dissimilarity Matrices — avgdist","text":"","code":"# Import an example count dataset data(BCI) # Test the base functionality mean.avg.dist <- avgdist(BCI, sample = 50, iterations = 10) # Test the transformation function mean.avg.dist.t <- avgdist(BCI, sample = 50, iterations = 10, transf = sqrt) # Test the median functionality median.avg.dist <- avgdist(BCI, sample = 50, iterations = 10, meanfun = median) # Print the resulting tables head(as.matrix(mean.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 #> 1 0.000 0.556 0.606 0.648 0.620 0.624 0.610 0.606 0.654 0.624 0.638 0.648 0.708 #> 2 0.556 0.000 0.568 0.568 0.632 0.568 0.532 0.554 0.576 0.578 0.588 0.560 0.678 #> 3 0.606 0.568 0.000 0.582 0.578 0.618 0.600 0.568 0.610 0.574 0.610 0.622 0.738 #> 4 0.648 0.568 0.582 0.000 0.624 0.654 0.610 0.572 0.594 0.594 0.596 0.636 0.672 #> 5 0.620 0.632 0.578 0.624 0.000 0.602 0.662 0.626 0.670 0.606 0.678 0.716 0.754 #> 6 0.624 0.568 0.618 0.654 0.602 0.000 0.598 0.566 0.616 0.644 0.576 0.568 0.706 #> 14 15 16 17 18 19 20 21 22 23 24 25 26 #> 1 0.630 0.648 0.600 0.646 0.744 0.682 0.626 0.600 0.646 0.700 0.658 0.620 0.620 #> 2 0.600 0.592 0.554 0.564 0.660 0.584 0.598 0.606 0.544 0.642 0.574 0.624 0.592 #> 3 0.628 0.612 0.616 0.646 0.730 0.660 0.646 0.568 0.636 0.722 0.652 0.672 0.660 #> 4 0.602 0.602 0.618 0.634 0.666 0.622 0.636 0.656 0.568 0.642 0.614 0.654 0.638 #> 5 0.610 0.642 0.630 0.760 0.766 0.698 0.654 0.642 0.738 0.702 0.652 0.688 0.652 #> 6 0.618 0.696 0.626 0.618 0.668 0.634 0.664 0.648 0.630 0.694 0.608 0.682 0.674 #> 27 28 29 30 31 32 33 34 35 36 37 38 39 #> 1 0.702 0.684 0.674 0.634 0.680 0.724 0.692 0.684 0.744 0.666 0.694 0.716 0.700 #> 2 0.632 0.586 0.568 0.582 0.646 0.606 0.618 0.634 0.722 0.638 0.590 0.590 0.616 #> 3 0.668 0.660 0.654 0.676 0.660 0.674 0.694 0.700 0.764 0.684 0.628 0.684 0.708 #> 4 0.644 0.596 0.612 0.608 0.662 0.594 0.610 0.632 0.730 0.648 0.562 0.588 0.636 #> 5 0.712 0.752 0.728 0.734 0.676 0.710 0.698 0.768 0.830 0.678 0.690 0.752 0.760 #> 6 0.628 0.640 0.630 0.668 0.668 0.636 0.662 0.682 0.768 0.678 0.658 0.664 0.700 #> 40 41 42 43 44 45 46 47 48 49 50 #> 1 0.714 0.686 0.624 0.688 0.654 0.668 0.668 0.650 0.676 0.694 0.676 #> 2 0.638 0.670 0.644 0.654 0.626 0.664 0.648 0.638 0.648 0.678 0.672 #> 3 0.716 0.702 0.634 0.672 0.668 0.646 0.720 0.730 0.712 0.698 0.704 #> 4 0.680 0.652 0.618 0.680 0.698 0.696 0.658 0.666 0.658 0.696 0.708 #> 5 0.828 0.688 0.654 0.658 0.690 0.654 0.772 0.720 0.704 0.692 0.656 #> 6 0.724 0.686 0.678 0.670 0.666 0.698 0.734 0.692 0.714 0.734 0.704 head(as.matrix(mean.avg.dist.t)) #> 1 2 3 4 5 6 7 #> 1 0.0000000 0.5073247 0.5451162 0.5511308 0.5863431 0.5693314 0.5561033 #> 2 0.5073247 0.0000000 0.5076177 0.5454488 0.5590474 0.5361407 0.5075526 #> 3 0.5451162 0.5076177 0.0000000 0.5169507 0.5442375 0.5536933 0.5580432 #> 4 0.5511308 0.5454488 0.5169507 0.0000000 0.5556915 0.6029668 0.5894129 #> 5 0.5863431 0.5590474 0.5442375 0.5556915 0.0000000 0.5730661 0.5904097 #> 6 0.5693314 0.5361407 0.5536933 0.6029668 0.5730661 0.0000000 0.5161496 #> 8 9 10 11 12 13 14 #> 1 0.5673559 0.5627535 0.5792425 0.5729159 0.6020539 0.7016723 0.5490011 #> 2 0.5366592 0.5380985 0.5623811 0.5321790 0.5515688 0.6779465 0.5572373 #> 3 0.5073755 0.5536056 0.5389882 0.5801096 0.5868026 0.7260347 0.5816036 #> 4 0.5259962 0.5706750 0.5613498 0.5523447 0.5753112 0.6804780 0.5838035 #> 5 0.5484180 0.5855322 0.5881431 0.6281267 0.6276212 0.7377994 0.5875698 #> 6 0.5529963 0.5552535 0.6215621 0.5618841 0.5336650 0.6716372 0.5641578 #> 15 16 17 18 19 20 21 #> 1 0.6050878 0.5605091 0.6262520 0.6790944 0.6158975 0.5894641 0.6238942 #> 2 0.5492945 0.5571279 0.5683865 0.6719160 0.5864794 0.5817164 0.5785775 #> 3 0.5651808 0.5907658 0.5986803 0.7180656 0.6208023 0.5831755 0.6141896 #> 4 0.5846365 0.5995058 0.6019378 0.6723126 0.6185437 0.5788246 0.6204008 #> 5 0.5886468 0.6313116 0.6908350 0.7310610 0.6738026 0.6056946 0.6248085 #> 6 0.6245763 0.5734551 0.5897524 0.6553604 0.5945132 0.6253476 0.6236726 #> 22 23 24 25 26 27 28 #> 1 0.6087932 0.6864876 0.5674543 0.6124061 0.6000094 0.6287013 0.6246696 #> 2 0.5851407 0.6506247 0.5554757 0.5519726 0.6143138 0.5810440 0.5887911 #> 3 0.6301892 0.7101043 0.5608707 0.5964616 0.6536689 0.6015372 0.6254962 #> 4 0.6116276 0.6570546 0.5806733 0.5942875 0.6756654 0.6170196 0.6088559 #> 5 0.6692645 0.7281868 0.6019467 0.6240673 0.6717579 0.6824310 0.6781349 #> 6 0.5779917 0.6752783 0.5474248 0.6048853 0.6297599 0.6032439 0.6025756 #> 29 30 31 32 33 34 35 #> 1 0.5877126 0.5931557 0.6122546 0.6435203 0.6114740 0.6243358 0.7012874 #> 2 0.5493477 0.5852872 0.5988185 0.6153332 0.5921568 0.6244034 0.6905708 #> 3 0.5678376 0.6210073 0.6158757 0.6377790 0.6371270 0.6555867 0.7323038 #> 4 0.5759931 0.5912841 0.5983615 0.6066556 0.6016788 0.6087876 0.6950808 #> 5 0.6276522 0.6379576 0.6034350 0.6457655 0.6563073 0.6866576 0.7235042 #> 6 0.5816252 0.6242943 0.6152092 0.6137642 0.6116231 0.6246531 0.7312678 #> 36 37 38 39 40 41 42 #> 1 0.5974168 0.6467898 0.6716840 0.6643855 0.6683070 0.6353745 0.6004874 #> 2 0.5903695 0.5752574 0.6187077 0.6192344 0.6573943 0.6571387 0.6027516 #> 3 0.5687077 0.6126409 0.6617885 0.6959650 0.6909293 0.6525653 0.6014759 #> 4 0.6194344 0.5761296 0.6102164 0.6295049 0.6560977 0.5989171 0.6281164 #> 5 0.6056532 0.6441953 0.6593381 0.7110348 0.7252049 0.6568578 0.6156351 #> 6 0.6109414 0.6067715 0.6656711 0.6870593 0.6686455 0.6769139 0.6426746 #> 43 44 45 46 47 48 49 #> 1 0.6310309 0.5900500 0.5774603 0.6474425 0.6488217 0.6157111 0.6070289 #> 2 0.6028820 0.6126007 0.6208170 0.6652427 0.6134982 0.6243098 0.6319083 #> 3 0.6162330 0.6398741 0.6039060 0.7021972 0.6677691 0.6678864 0.6696487 #> 4 0.6197236 0.6379939 0.6156272 0.6582549 0.6062615 0.6139683 0.6466872 #> 5 0.6203780 0.6542059 0.6040162 0.7417980 0.6946591 0.6954255 0.6537119 #> 6 0.6386277 0.6418793 0.6306854 0.6948586 0.6326248 0.6678749 0.6692019 #> 50 #> 1 0.5867459 #> 2 0.6227903 #> 3 0.6365131 #> 4 0.6260672 #> 5 0.6250031 #> 6 0.6423431 head(as.matrix(median.avg.dist)) #> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 #> 1 0.00 0.56 0.60 0.62 0.66 0.63 0.58 0.62 0.65 0.61 0.59 0.65 0.74 0.64 0.64 #> 2 0.56 0.00 0.53 0.59 0.64 0.63 0.60 0.58 0.58 0.60 0.59 0.62 0.68 0.58 0.61 #> 3 0.60 0.53 0.00 0.61 0.63 0.58 0.56 0.57 0.60 0.55 0.60 0.64 0.76 0.66 0.61 #> 4 0.62 0.59 0.61 0.00 0.64 0.66 0.62 0.58 0.57 0.58 0.63 0.62 0.66 0.55 0.57 #> 5 0.66 0.64 0.63 0.64 0.00 0.57 0.65 0.64 0.62 0.62 0.68 0.67 0.76 0.66 0.65 #> 6 0.63 0.63 0.58 0.66 0.57 0.00 0.58 0.62 0.61 0.66 0.60 0.59 0.74 0.69 0.70 #> 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 #> 1 0.63 0.65 0.76 0.67 0.65 0.65 0.66 0.68 0.66 0.66 0.62 0.66 0.65 0.66 0.66 #> 2 0.62 0.59 0.69 0.61 0.56 0.64 0.62 0.67 0.60 0.58 0.62 0.64 0.63 0.56 0.63 #> 3 0.61 0.64 0.72 0.69 0.64 0.62 0.62 0.74 0.65 0.65 0.66 0.64 0.61 0.60 0.64 #> 4 0.63 0.62 0.67 0.61 0.60 0.68 0.59 0.67 0.62 0.60 0.66 0.60 0.62 0.59 0.62 #> 5 0.65 0.73 0.81 0.70 0.65 0.64 0.72 0.69 0.71 0.67 0.71 0.68 0.72 0.70 0.67 #> 6 0.63 0.60 0.68 0.63 0.66 0.68 0.63 0.70 0.61 0.64 0.66 0.63 0.65 0.62 0.69 #> 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 #> 1 0.65 0.69 0.66 0.66 0.74 0.65 0.68 0.69 0.70 0.68 0.67 0.63 0.63 0.64 0.67 #> 2 0.62 0.61 0.61 0.63 0.75 0.66 0.60 0.60 0.67 0.68 0.68 0.62 0.64 0.64 0.69 #> 3 0.63 0.61 0.69 0.68 0.78 0.67 0.61 0.65 0.68 0.72 0.71 0.65 0.66 0.64 0.65 #> 4 0.59 0.65 0.61 0.66 0.79 0.61 0.57 0.61 0.65 0.66 0.68 0.63 0.66 0.68 0.71 #> 5 0.65 0.71 0.70 0.75 0.82 0.70 0.65 0.73 0.79 0.76 0.72 0.64 0.73 0.70 0.68 #> 6 0.68 0.68 0.67 0.68 0.82 0.72 0.71 0.69 0.71 0.75 0.74 0.69 0.67 0.69 0.72 #> 46 47 48 49 50 #> 1 0.69 0.66 0.66 0.66 0.67 #> 2 0.69 0.62 0.67 0.63 0.65 #> 3 0.73 0.70 0.70 0.68 0.72 #> 4 0.68 0.65 0.65 0.69 0.64 #> 5 0.76 0.70 0.75 0.72 0.70 #> 6 0.72 0.71 0.73 0.72 0.71 # Run example to illustrate low variance of mean, median, and stdev results # Mean and median std dev are around 0.05 sdd <- avgdist(BCI, sample = 50, iterations = 100, meanfun = sd) summary(mean.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4640 0.6180 0.6540 0.6543 0.6900 0.8360 summary(median.avg.dist) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.4600 0.6100 0.6500 0.6501 0.6900 0.8600 summary(sdd) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.03849 0.05617 0.05921 0.05919 0.06208 0.07661 # Test for when subsampling depth excludes some samples # Return samples that are removed for not meeting depth filter depth.avg.dist <- avgdist(BCI, sample = 450, iterations = 10) #> Warning: The following sampling units were removed because they were below sampling depth: 1, 2, 6, 7, 8, 9, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 31, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 # Print the result depth.avg.dist #> 3 4 5 10 15 30 32 #> 3 0.0000000 0.3282222 0.3635556 0.2942222 0.3562222 0.4555556 0.4726667 #> 4 0.3282222 0.0000000 0.3775556 0.3173333 0.3480000 0.3908889 0.4273333 #> 5 0.3635556 0.3775556 0.0000000 0.3900000 0.3995556 0.4968889 0.5231111 #> 10 0.2942222 0.3173333 0.3900000 0.0000000 0.3135556 0.4266667 0.4200000 #> 15 0.3562222 0.3480000 0.3995556 0.3135556 0.0000000 0.4535556 0.4657778 #> 30 0.4555556 0.3908889 0.4968889 0.4266667 0.4535556 0.0000000 0.3811111 #> 32 0.4726667 0.4273333 0.5231111 0.4200000 0.4657778 0.3811111 0.0000000 #> 35 0.6591111 0.6266667 0.6966667 0.6817778 0.6573333 0.5157778 0.5893333 #> 40 0.5624444 0.5286667 0.6413333 0.5240000 0.5628889 0.4540000 0.4057778 #> 35 40 #> 3 0.6591111 0.5624444 #> 4 0.6266667 0.5286667 #> 5 0.6966667 0.6413333 #> 10 0.6817778 0.5240000 #> 15 0.6573333 0.5628889 #> 30 0.5157778 0.4540000 #> 32 0.5893333 0.4057778 #> 35 0.0000000 0.4280000 #> 40 0.4280000 0.0000000"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":null,"dir":"Reference","previous_headings":"","what":"Beals Smoothing and Degree of Absence — beals","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing replaces entry community data probability target species occurring particular site, based joint occurrences target species species actually occur site. Swan's (1970) degree absence applies Beals smoothing zero items long zeros replaced smoothed values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"beals(x, species = NA, reference = x, type = 0, include = TRUE) swan(x, maxit = Inf, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Beals Smoothing and Degree of Absence — beals","text":"x Community data frame matrix. species Column index used compute Beals function single species. default (NA) indicates function computed species. reference Community data frame matrix used compute joint occurrences. default, x used reference compute joint occurrences. type Numeric. Specifies abundance values used function beals. See details explanation. include logical flag indicates whether target species included computing mean conditioned probabilities. original Beals (1984) definition equivalent include=TRUE, formulation Münzbergová Herben equal include=FALSE. maxit Maximum number iterations. default Inf means iterations continued zeros number zeros change. Probably maxit = 1 makes sense addition default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals smoothing estimated probability \\(p_{ij}\\) species \\(j\\) occurs site \\(\\). defined \\(p_{ij} = \\frac{1}{S_i} \\sum_k \\frac{N_{jk} I_{ik}}{N_k}\\), \\(S_i\\) number species site \\(\\), \\(N_{jk}\\) number joint occurrences species \\(j\\) \\(k\\), \\(N_k\\) number occurrences species \\(k\\), \\(\\) incidence (0 1) species (last term usually omitted equation, necessary). \\(N_{jk}\\) can interpreted mean conditional probability, beals function can interpreted mean conditioned probabilities (De Cáceres & Legendre 2008). present function generalized abundance values (De Cáceres & Legendre 2008). type argument specifies abundance values used. type = 0 presence/absence mode. type = 1 abundances reference (x) used compute conditioned probabilities. type = 2 abundances x used compute weighted averages conditioned probabilities. type = 3 abundances used compute conditioned probabilities weighted averages. Beals smoothing originally suggested method data transformation remove excessive zeros (Beals 1984, McCune 1994). However, suitable method purpose since maintain information species presences: species may higher probability occurrence site occur sites occurs. Moreover, regularizes data strongly. method may useful identifying species belong species pool (Ewald 2002) identify suitable unoccupied patches metapopulation analysis (Münzbergová & Herben 2004). case, function called include=FALSE cross-validation smoothing species; argument species can used one species studied. Swan (1970) suggested replacing zero values degrees absence species community data matrix. Swan expressed method terms similarity matrix, equivalent applying Beals smoothing zero values, step shifting smallest initially non-zero item value one, repeating many times zeros left data. actually similar extended dissimilarities (implemented function stepacross), rarely used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Beals Smoothing and Degree of Absence — beals","text":"function returns transformed data matrix vector Beals smoothing requested single species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Beals Smoothing and Degree of Absence — beals","text":"Beals, E.W. 1984. Bray-Curtis ordination: effective strategy analysis multivariate ecological data. Pp. 1--55 : MacFadyen, . & E.D. Ford [eds.] Advances Ecological Research, 14. Academic Press, London. De Cáceres, M. & Legendre, P. 2008. Beals smoothing revisited. Oecologia 156: 657--669. Ewald, J. 2002. probabilistic approach estimating species pools large compositional matrices. J. Veg. Sci. 13: 191--198. McCune, B. 1994. Improving community ordination Beals smoothing function. Ecoscience 1: 82--86. Münzbergová, Z. & Herben, T. 2004. Identification suitable unoccupied habitats metapopulation studies using co-occurrence species. Oikos 105: 408--414. Swan, J.M.. 1970. examination ordination problems use simulated vegetational data. Ecology 51: 89--102.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Beals Smoothing and Degree of Absence — beals","text":"Miquel De Cáceres Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/beals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Beals Smoothing and Degree of Absence — beals","text":"","code":"data(dune) ## Default x <- beals(dune) ## Remove target species x <- beals(dune, include = FALSE) ## Smoothed values against presence or absence of species pa <- decostand(dune, \"pa\") boxplot(as.vector(x) ~ unlist(pa), xlab=\"Presence\", ylab=\"Beals\") ## Remove the bias of tarbet species: Yields lower values. beals(dune, type =3, include = FALSE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord #> 1 0.49590853 0.38333415 0.01157407 0.4923280 0.30827883 0.4935662 0.43263047 #> 2 0.47083676 0.39501120 0.03361524 0.4718807 0.34723984 0.4917791 0.42000984 #> 3 0.34063019 0.52738394 0.01520046 0.5309152 0.21609954 0.4033301 0.33010938 #> 4 0.30816435 0.51198853 0.02876960 0.5971801 0.21542662 0.4398775 0.35732610 #> 5 0.59949785 0.27622698 0.06632771 0.3349203 0.48876285 0.4322142 0.44309579 #> 6 0.58819821 0.26299306 0.05967771 0.2700508 0.53154426 0.3696613 0.39760652 #> 7 0.56496165 0.29412293 0.05329633 0.3403047 0.48010987 0.4051777 0.40471531 #> 8 0.21230502 0.66906674 0.02588333 0.5187956 0.16247716 0.2720122 0.21219877 #> 9 0.30323659 0.59744543 0.02213662 0.5792855 0.21896113 0.3292320 0.28613526 #> 10 0.54083871 0.26902092 0.07349127 0.3372958 0.42671693 0.4705094 0.42934344 #> 11 0.40509331 0.31656550 0.10259239 0.3185489 0.38766111 0.3713794 0.31413659 #> 12 0.21008725 0.66278454 0.03625297 0.5753377 0.20078932 0.2802946 0.22974415 #> 13 0.21850759 0.68239707 0.02191119 0.6404427 0.16737280 0.2939740 0.24942466 #> 14 0.13570397 0.76284476 0.02298398 0.4107645 0.12128973 0.1682755 0.13757552 #> 15 0.09168815 0.79412733 0.02538032 0.4505613 0.10117099 0.1420251 0.09794548 #> 16 0.06335463 0.87877202 0.00742115 0.5232448 0.05538377 0.1516354 0.09458531 #> 17 0.55254140 0.07330247 0.29233391 0.1013889 0.69331132 0.3129358 0.34982363 #> 18 0.37751017 0.34451209 0.08535723 0.2838834 0.36918166 0.3676424 0.30478244 #> 19 0.29826049 0.25952255 0.35137675 0.1934048 0.51929869 0.2237843 0.18074796 #> 20 0.05429986 0.76675441 0.06144615 0.4063662 0.10738280 0.1450721 0.06706410 #> Chenalbu Cirsarve Comapalu Eleopalu Elymrepe Empenigr #> 1 0.025132275 0.09504980 0.000000000 0.05592045 0.4667439 0.00000000 #> 2 0.043866562 0.08570299 0.026548839 0.08656209 0.4407282 0.01829337 #> 3 0.065338638 0.08967477 0.031898812 0.16099072 0.4137888 0.01074444 #> 4 0.057970906 0.12920228 0.039859621 0.16112450 0.4399661 0.02527165 #> 5 0.026434737 0.05520104 0.015892090 0.05419613 0.3575948 0.03029752 #> 6 0.021256367 0.03223112 0.030347896 0.08784329 0.3138879 0.03093489 #> 7 0.038467708 0.04706743 0.017083997 0.06694311 0.3586644 0.02304603 #> 8 0.063278453 0.06688407 0.100703044 0.29777644 0.3046956 0.02102222 #> 9 0.069879277 0.07647268 0.045830682 0.19018562 0.3523460 0.01838883 #> 10 0.025686639 0.06037513 0.029746617 0.07787078 0.3736128 0.03425596 #> 11 0.021234732 0.05778318 0.035740922 0.11146095 0.2884798 0.07310076 #> 12 0.103543341 0.07799259 0.045375827 0.19518888 0.3354080 0.03413656 #> 13 0.122547745 0.07905124 0.056084315 0.22437598 0.3511708 0.01840390 #> 14 0.042990591 0.03618335 0.241811837 0.55982776 0.1428372 0.01989756 #> 15 0.035609053 0.04022968 0.198176675 0.53973883 0.1462975 0.02215971 #> 16 0.056246994 0.05184498 0.201352298 0.51523810 0.1832397 0.00742115 #> 17 0.007716049 0.01049383 0.009876543 0.02777778 0.1929470 0.21968254 #> 18 0.014640428 0.04454602 0.042890320 0.17341352 0.2651538 0.06763669 #> 19 0.019591245 0.03668466 0.031845637 0.12592768 0.1422725 0.26011417 #> 20 0.037623741 0.03453783 0.185726965 0.58476297 0.1168700 0.05905666 #> Hyporadi Juncarti Juncbufo Lolipere Planlanc Poaprat Poatriv #> 1 0.07702746 0.14794933 0.1987270 0.9226190 0.40103107 0.9863946 0.8826329 #> 2 0.07454127 0.13017869 0.2070478 0.8272395 0.40700777 0.8972046 0.8288385 #> 3 0.05562332 0.22291082 0.2544828 0.7205525 0.27933493 0.8083020 0.8383185 #> 4 0.06985986 0.21320122 0.2318440 0.7197924 0.25797285 0.7940926 0.8197302 #> 5 0.10245961 0.10406655 0.2164230 0.8380779 0.52628928 0.9035899 0.8094632 #> 6 0.11463153 0.11631772 0.2166255 0.8000021 0.58765018 0.8666677 0.7782619 #> 7 0.10837376 0.11293676 0.2110045 0.8053380 0.51905808 0.8925059 0.8018775 #> 8 0.06550319 0.33219882 0.2323566 0.5403355 0.20596764 0.6160461 0.7101299 #> 9 0.05343787 0.23134366 0.2675624 0.6874068 0.25274756 0.7523318 0.8247374 #> 10 0.13692492 0.09080902 0.1678040 0.8102783 0.52588347 0.8915882 0.7543592 #> 11 0.18108995 0.13478872 0.1656396 0.7180948 0.47012501 0.8062720 0.6404351 #> 12 0.06777311 0.27306206 0.3231724 0.5875943 0.22110550 0.6932541 0.8199960 #> 13 0.04250245 0.28204736 0.3339728 0.5714581 0.18153869 0.7063028 0.7993754 #> 14 0.04665747 0.46685537 0.1206518 0.3356311 0.14342002 0.3817081 0.5090703 #> 15 0.05040404 0.51561767 0.1370235 0.3689922 0.13523214 0.4078219 0.5263520 #> 16 0.01731602 0.54304667 0.1776781 0.3561752 0.07269979 0.4124222 0.6071083 #> 17 0.36492870 0.03333333 0.1038156 0.5858415 0.59641331 0.7434618 0.5036834 #> 18 0.17491099 0.18956922 0.1376386 0.7124388 0.45087176 0.7368632 0.5859071 #> 19 0.39145281 0.13543701 0.1127832 0.4289185 0.40784415 0.5548077 0.3605827 #> 20 0.07795311 0.53056145 0.1192488 0.3262685 0.13059496 0.3662817 0.4523029 #> Ranuflam Rumeacet Sagiproc Salirepe Scorautu Trifprat Trifrepe #> 1 0.08105273 0.3160963 0.3371121 0.02729885 0.8898317 0.21701279 0.8782576 #> 2 0.13042865 0.3031318 0.3302063 0.05983781 0.9349640 0.20673650 0.9125666 #> 3 0.22632936 0.2909068 0.4204104 0.06065155 0.9036443 0.14654749 0.8817430 #> 4 0.21909541 0.2610006 0.4191908 0.07579199 0.9204237 0.12896524 0.8943213 #> 5 0.08063087 0.3979230 0.2612828 0.07589611 0.9576838 0.34808957 0.9142360 #> 6 0.10909966 0.4330705 0.2539380 0.08921540 0.9590466 0.35423465 0.9110822 #> 7 0.10541081 0.4113622 0.2954682 0.07094548 0.9550487 0.32489503 0.9171688 #> 8 0.40134447 0.2331043 0.4009544 0.11569906 0.8755515 0.09897600 0.8002526 #> 9 0.26006489 0.3464870 0.4531178 0.07351827 0.9145996 0.16269563 0.8714833 #> 10 0.10355742 0.3226025 0.2732735 0.09037489 0.9568824 0.26807372 0.9003730 #> 11 0.13269569 0.2753878 0.3673397 0.16465286 0.9442707 0.19982976 0.8979262 #> 12 0.29873222 0.3507140 0.5122033 0.08041977 0.9377963 0.13849854 0.9079979 #> 13 0.33309468 0.3107471 0.5131337 0.06572594 0.9255312 0.11199114 0.8841739 #> 14 0.64674225 0.1241545 0.2528665 0.15917563 0.8477706 0.06176123 0.6485949 #> 15 0.64449081 0.1459458 0.3151199 0.17750323 0.8430677 0.05831084 0.7170446 #> 16 0.66893881 0.1508409 0.3480368 0.15783292 0.8131968 0.03916718 0.6776273 #> 17 0.03549383 0.2913631 0.3292030 0.25651777 0.9839744 0.27593101 0.8141660 #> 18 0.18805395 0.2668100 0.3154533 0.17191937 0.9554011 0.20461193 0.8600701 #> 19 0.14551892 0.1831168 0.4798245 0.36429493 0.9902041 0.11966159 0.8147968 #> 20 0.62796060 0.1098600 0.3105433 0.21674174 0.8457313 0.03708580 0.6350394 #> Vicilath Bracruta Callcusp #> 1 0.17244420 0.7476589 0.003527337 #> 2 0.18494940 0.7415172 0.034597921 #> 3 0.12833142 0.7666969 0.075789630 #> 4 0.12550967 0.7919786 0.081110164 #> 5 0.16693075 0.8079786 0.023129027 #> 6 0.18035860 0.8387650 0.040981168 #> 7 0.19027523 0.8089116 0.024070054 #> 8 0.10213052 0.8109194 0.201958942 #> 9 0.08630413 0.7972178 0.092982775 #> 10 0.23383453 0.7660374 0.033527777 #> 11 0.24317802 0.8182692 0.043950322 #> 12 0.08049055 0.8061715 0.100954127 #> 13 0.06604026 0.7465509 0.122856392 #> 14 0.07857237 0.7238162 0.347514804 #> 15 0.07370069 0.7997141 0.381379395 #> 16 0.03353260 0.8029953 0.364128496 #> 17 0.20728700 0.7635487 0.009876543 #> 18 0.26222869 0.8397471 0.109959916 #> 19 0.18188455 0.8161275 0.082361157 #> 20 0.09111967 0.8397124 0.397924041 ## Uses abundance information. ## Vector with beals smoothing values corresponding to the first species ## in dune. beals(dune, species=1, include=TRUE) #> 1 2 3 4 5 6 7 8 #> 0.5923077 0.5032372 0.3499038 0.3306953 0.5944041 0.5928780 0.5824352 0.2082532 #> 9 10 11 12 13 14 15 16 #> 0.2960799 0.5462492 0.3659392 0.2610043 0.1982372 0.0922619 0.1140625 0.1066506 #> 17 18 19 20 #> 0.6020408 0.3844577 0.2865741 0.0750000"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Multivariate homogeneity of groups dispersions (variances) — betadisper","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Implements Marti Anderson's PERMDISP2 procedure analysis multivariate homogeneity group dispersions (variances). betadisper multivariate analogue Levene's test homogeneity variances. Non-euclidean distances objects group centres (centroids medians) handled reducing original distances principal coordinates. procedure latterly used means assessing beta diversity. anova, scores, plot boxplot methods. TukeyHSD.betadisper creates set confidence intervals differences mean distance--centroid levels grouping factor specified family-wise probability coverage. intervals based Studentized range statistic, Tukey's 'Honest Significant Difference' method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"betadisper(d, group, type = c(\"median\",\"centroid\"), bias.adjust = FALSE, sqrt.dist = FALSE, add = FALSE) # S3 method for betadisper anova(object, ...) # S3 method for betadisper scores(x, display = c(\"sites\", \"centroids\"), choices = c(1,2), ...) # S3 method for betadisper eigenvals(x, ...) # S3 method for betadisper plot(x, axes = c(1,2), cex = 0.7, pch = seq_len(ng), col = NULL, lty = \"solid\", lwd = 1, hull = TRUE, ellipse = FALSE, conf, segments = TRUE, seg.col = \"grey\", seg.lty = lty, seg.lwd = lwd, label = TRUE, label.cex = 1, ylab, xlab, main, sub, ...) # S3 method for betadisper boxplot(x, ylab = \"Distance to centroid\", ...) # S3 method for betadisper TukeyHSD(x, which = \"group\", ordered = FALSE, conf.level = 0.95, ...) # S3 method for betadisper print(x, digits = max(3, getOption(\"digits\") - 3), neigen = 8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"d distance structure returned dist, betadiver vegdist. group vector describing group structure, usually factor object can coerced factor using .factor. Can consist factor single level (.e., one group). type type analysis perform. Use spatial median group centroid? spatial median now default. bias.adjust logical: adjust small sample bias beta diversity estimates? sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. display character; partial match access scores \"sites\" \"species\". object, x object class \"betadisper\", result call betadisper. choices, axes principal coordinate axes wanted. hull logical; convex hull group plotted? ellipse logical; standard deviation data ellipse group plotted? conf Expected fractions data coverage data ellipses, e.g. 0.95. default draw 1 standard deviation data ellipse, supplied, conf multiplied corresponding value found Chi-squared distribution 2df provide requested coverage (probability contour). pch plot symbols groups, vector length equal number groups. col colors plot symbols centroid labels groups, vector length equal number groups. lty, lwd linetype, linewidth convex hulls confidence ellipses. segments logical; segments joining points centroid drawn? seg.col colour draw segments points centroid. Can vector, case one colour per group. seg.lty, seg.lwd linetype line width segments. label logical; centroids labelled respective factor label? label.cex numeric; character expansion centroid labels. cex, ylab, xlab, main, sub graphical parameters. details, see plot.default. character vector listing terms fitted model intervals calculated. Defaults grouping factor. ordered logical; see TukeyHSD. conf.level numeric value zero one giving family-wise confidence level use. digits, neigen numeric; print method, sets number digits use (per print.default) maximum number axes display eigenvalues , repsectively. ... arguments, including graphical parameters ( plot.betadisper boxplot.betadisper), passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"One measure multivariate dispersion (variance) group samples calculate average distance group members group centroid spatial median (referred 'centroid' now unless stated otherwise) multivariate space. test dispersions (variances) one groups different, distances group members group centroid subject ANOVA. multivariate analogue Levene's test homogeneity variances distances group members group centroids Euclidean distance. However, better measures distance Euclidean distance available ecological data. can accommodated reducing distances produced using dissimilarity coefficient principal coordinates, embeds within Euclidean space. analysis proceeds calculating Euclidean distances group members group centroid basis principal coordinate axes rather original distances. Non-metric dissimilarity coefficients can produce principal coordinate axes negative Eigenvalues. correspond imaginary, non-metric part distance objects. negative Eigenvalues produced, must correct imaginary distances. distance centroid point $$z_{ij}^c = \\sqrt{\\Delta^2(u_{ij}^+, c_i^+) - \\Delta^2(u_{ij}^-, c_i^-)},$$ \\(\\Delta^2\\) squared Euclidean distance \\(u_{ij}\\), principal coordinate \\(j\\)th point \\(\\)th group, \\(c_i\\), coordinate centroid \\(\\)th group. super-scripted ‘\\(+\\)’ ‘\\(-\\)’ indicate real imaginary parts respectively. equation (3) Anderson (2006). imaginary part greater magnitude real part, taking square root negative value, resulting NaN, cases changed zero distances (warning). line behaviour Marti Anderson's PERMDISP2 programme. test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance \\(F\\). alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution \\(F\\) Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can also performed using permutest.betadisper. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper. simple wrapper TukeyHSD. user directed read help file TukeyHSD using function. particular, note statement using function unbalanced designs. results analysis can visualised using plot boxplot methods. One additional use functions assessing beta diversity (Anderson et al 2006). Function betadiver provides popular dissimilarity measures purpose. noted passing Anderson (2006) related context O'Neill (2000), estimates dispersion around central location (median centroid) calculated data biased downward. bias matters comparing diversity among treatments small, unequal numbers samples. Setting bias.adjust=TRUE using betadisper imposes \\(\\sqrt{n/(n-1)}\\) correction (Stier et al. 2013).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"anova method returns object class \"anova\" inheriting class \"data.frame\". scores method returns list one components \"sites\" \"centroids\". plot function invisibly returns object class \"ordiplot\", plotting structure can used identify.ordiplot (identify points) functions ordiplot family. boxplot function invisibly returns list whose components documented boxplot. eigenvals.betadisper returns named vector eigenvalues. TukeyHSD.betadisper returns list. See TukeyHSD details. betadisper returns list class \"betadisper\" following components: eig numeric; eigenvalues principal coordinates analysis. vectors matrix; eigenvectors principal coordinates analysis. distances numeric; Euclidean distances principal coordinate space samples respective group centroid median. group factor; vector describing group structure centroids matrix; locations group centroids medians principal coordinates. group.distances numeric; mean distance group centroid median. call matched function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"group consists single level group, anova permutest methods appropriate used data stop error. Missing values either d group removed prior performing analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Stewart Schultz noticed permutation test type=\"centroid\" wrong type error anti-conservative. , default type changed \"median\", uses spatial median group centroid. Tests suggests permutation test type analysis gives correct error rates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62, 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. O'Neill, M.E. (2000) Weighted Least Squares Approach Levene's Test Homogeneity Variance. Australian & New Zealand Journal Statistics 42, 81-–100. Stier, .C., Geange, S.W., Hanson, K.M., & Bolker, B.M. (2013) Predator density timing arrival affect reef fish community assembly. Ecology 94, 1057--1068.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"Gavin L. Simpson; bias correction Adrian Stier Ben Bolker.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multivariate homogeneity of groups dispersions (variances) — betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F permutest(mod, pairwise = TRUE, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 99 0.07 . #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Pairwise comparisons: #> (Observed p-value below diagonal, permuted p-value above diagonal) #> grazed ungrazed #> grazed 0.08 #> ungrazed 0.04295 ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Plot the groups and distances to centroids on the ## first two PCoA axes plot(mod) ## with data ellipses instead of hulls plot(mod, ellipse = TRUE, hull = FALSE) # 1 sd data ellipse plot(mod, ellipse = TRUE, hull = FALSE, conf = 0.90) # 90% data ellipse # plot with manual colour specification my_cols <- c(\"#1b9e77\", \"#7570b3\") plot(mod, col = my_cols, pch = c(16,17), cex = 1.1) ## can also specify which axes to plot, ordering respected plot(mod, axes = c(3,1), seg.col = \"forestgreen\", seg.lty = \"dashed\") ## Draw a boxplot of the distances to centroid for each group boxplot(mod) ## `scores` and `eigenvals` also work scrs <- scores(mod) str(scrs) #> List of 2 #> $ sites : num [1:24, 1:2] 0.0946 -0.3125 -0.3511 -0.3291 -0.1926 ... #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:24] \"18\" \"15\" \"24\" \"27\" ... #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" #> $ centroids: num [1:2, 1:2] -0.1455 0.2786 0.0758 -0.2111 #> ..- attr(*, \"dimnames\")=List of 2 #> .. ..$ : chr [1:2] \"grazed\" \"ungrazed\" #> .. ..$ : chr [1:2] \"PCoA1\" \"PCoA2\" head(scores(mod, 1:4, display = \"sites\")) #> PCoA1 PCoA2 PCoA3 PCoA4 #> 18 0.09459373 0.15914576 0.074400844 -0.202466025 #> 15 -0.31248809 0.10032751 -0.062243360 0.110844864 #> 24 -0.35106507 -0.05954096 -0.038079447 0.095060928 #> 27 -0.32914546 -0.17019348 0.231623720 0.019110623 #> 23 -0.19259443 -0.01459250 -0.005679372 -0.209718312 #> 19 -0.06794575 -0.14501690 -0.085645653 0.002431355 # group centroids/medians scores(mod, 1:4, display = \"centroids\") #> PCoA1 PCoA2 PCoA3 PCoA4 #> grazed -0.1455200 0.07584572 -0.01366220 -0.0178990 #> ungrazed 0.2786095 -0.21114993 -0.03475586 0.0220129 # eigenvalues from the underlying principal coordinates analysis eigenvals(mod) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 #> 1.7552165 1.1334455 0.4429018 0.3698054 0.2453532 0.1960921 0.1751131 #> PCoA8 PCoA9 PCoA10 PCoA11 PCoA12 PCoA13 PCoA14 #> 0.1284467 0.0971594 0.0759601 0.0637178 0.0583225 0.0394934 0.0172699 #> PCoA15 PCoA16 PCoA17 PCoA18 PCoA19 PCoA20 PCoA21 #> 0.0051011 -0.0004131 -0.0064654 -0.0133147 -0.0253944 -0.0375105 -0.0480069 #> PCoA22 PCoA23 #> -0.0537146 -0.0741390 ## try out bias correction; compare with mod3 (mod3B <- betadisper(dis, groups, type = \"median\", bias.adjust=TRUE)) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"median\", bias.adjust #> = TRUE) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.4055 0.2893 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 anova(mod3B) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 0.06468 . #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 permutest(mod3B, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.07193 0.071927 3.7826 99 0.05 * #> Residuals 22 0.41834 0.019015 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## should always work for a single group group <- factor(rep(\"grazed\", NROW(varespec))) (tmp <- betadisper(dis, group, type = \"median\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"median\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed #> 0.4255 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 (tmp <- betadisper(dis, group, type = \"centroid\")) #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = group, type = \"centroid\") #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to centroid: #> grazed #> 0.4261 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## simulate missing values in 'd' and 'group' ## using spatial medians groups[c(2,20)] <- NA dis[c(2, 20)] <- NA mod2 <- betadisper(dis, groups) ## messages #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod2 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to median: #> grazed ungrazed #> 0.3984 0.3008 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod2, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 99 0.16 #> Residuals 18 0.296910 0.016495 anova(mod2) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.039979 0.039979 2.4237 0.1369 #> Residuals 18 0.296910 0.016495 plot(mod2) boxplot(mod2) plot(TukeyHSD(mod2)) ## Using group centroids mod3 <- betadisper(dis, groups, type = \"centroid\") #> missing observations due to 'group' removed #> missing observations due to 'd' removed mod3 #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups, type = \"centroid\") #> #> No. of Positive Eigenvalues: 14 #> No. of Negative Eigenvalues: 5 #> #> Average distance to centroid: #> grazed ungrazed #> 0.4001 0.3108 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 19 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.4755 0.8245 0.4218 0.3456 0.2159 0.1688 0.1150 0.1060 permutest(mod3, permutations = 99) #> #> Permutation test for homogeneity of multivariate dispersions #> Permutation: free #> Number of permutations: 99 #> #> Response: Distances #> Df Sum Sq Mean Sq F N.Perm Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 99 0.12 #> Residuals 18 0.189749 0.010542 anova(mod3) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.033468 0.033468 3.1749 0.09166 . #> Residuals 18 0.189749 0.010542 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod3) boxplot(mod3) plot(TukeyHSD(mod3))"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of beta Diversity — betadiver","title":"Indices of beta Diversity — betadiver","text":"function estimates 24 indices beta diversity reviewed Koleff et al. (2003). Alternatively, finds co-occurrence frequencies triangular plots (Koleff et al. 2003).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of beta Diversity — betadiver","text":"","code":"betadiver(x, method = NA, order = FALSE, help = FALSE, ...) # S3 method for betadiver plot(x, ...) # S3 method for betadiver scores(x, triangular = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of beta Diversity — betadiver","text":"x Community data matrix, betadiver result plot scores functions. method index beta diversity defined Koleff et al. (2003), Table 1. can use either subscript \\(\\beta\\) number index. See argument help . order Order sites increasing number species. influence configuration triangular plot non-symmetric indices. help Show numbers, subscript names defining equations indices exit. triangular Return scores suitable triangular plotting proportions. FALSE, returns 3-column matrix raw counts. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of beta Diversity — betadiver","text":"commonly used index beta diversity \\(\\beta_w = S/\\alpha - 1\\), \\(S\\) total number species, \\(\\alpha\\) average number species per site (Whittaker 1960). drawback model \\(S\\) increases sample size, expectation \\(\\alpha\\) remains constant, beta diversity increases sample size. solution problem study beta diversity pairs sites (Marion et al. 2017). denote number species shared two sites \\(\\) numbers unique species (shared) \\(b\\) \\(c\\), \\(S = + b + c\\) \\(\\alpha = (2 + b + c)/2\\) \\(\\beta_w = (b+c)/(2 + b + c)\\). Sørensen dissimilarity defined vegan function vegdist argument binary = TRUE. Many indices dissimilarity indices well. Function betadiver finds indices reviewed Koleff et al. (2003). indices found function designdist, current function provides conventional shortcut. function finds indices. proper analysis must done functions betadisper, adonis2 mantel. indices directly taken Table 1 Koleff et al. (2003), can selected either index number subscript name used Koleff et al. numbers, names defining equations can seen using betadiver(help = TRUE). cases two alternative forms, one term \\(-1\\) used. several duplicate indices, number distinct alternatives much lower 24 formally provided. formulations used functions differ occasionally Koleff et al. (2003), still mathematically equivalent. method = NA, index calculated, instead object class betadiver returned. list elements , b c. Function plot can used display proportions elements triangular plot suggested Koleff et al. (2003), scores extracts triangular coordinates raw scores. Function plot returns invisibly triangular coordinates \"ordiplot\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of beta Diversity — betadiver","text":"method = NA, function returns object class \"betadisper\" elements , b, c. method specified, function returns \"dist\" object can used function analysing dissimilarities. beta diversity, particularly useful functions betadisper study betadiversity groups, adonis2 model, mantel compare beta diversities dissimilarities distances (including geographical distances). Although betadiver returns \"dist\" object, indices similarities used place dissimilarities, user error. Functions 10 (\"j\"), 11 (\"sor\") 21 (\"rlb\") similarity indices. Function sets argument \"maxdist\" similarly vegdist, using NA fixed upper limit, 0 similarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of beta Diversity — betadiver","text":"Baselga, . (2010) Partitioning turnover nestedness components beta diversity. Global Ecology Biogeography 19, 134--143. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence-absence data. Journal Animal Ecology 72, 367--382. Marion, Z.H., Fordyce, J.. Fitzpatrick, B.M. (2017) Pairwise beta diversity resolves underappreciated source confusion calculating species turnover. Ecology 98, 933--939. Whittaker, R.H. (1960) Vegetation Siskiyou mountains, Oregon California. Ecological Monographs 30, 279--338.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of beta Diversity — betadiver","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Indices of beta Diversity — betadiver","text":"indices return similarities instead dissimilarities.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/betadiver.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of beta Diversity — betadiver","text":"","code":"## Raw data and plotting data(sipoo) m <- betadiver(sipoo) plot(m) ## The indices betadiver(help=TRUE) #> 1 \"w\" = (b+c)/(2*a+b+c) #> 2 \"-1\" = (b+c)/(2*a+b+c) #> 3 \"c\" = (b+c)/2 #> 4 \"wb\" = b+c #> 5 \"r\" = 2*b*c/((a+b+c)^2-2*b*c) #> 6 \"I\" = log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c) #> 7 \"e\" = exp(log(2*a+b+c) - 2*a*log(2)/(2*a+b+c) - ((a+b)*log(a+b) + #> (a+c)*log(a+c)) / (2*a+b+c))-1 #> 8 \"t\" = (b+c)/(2*a+b+c) #> 9 \"me\" = (b+c)/(2*a+b+c) #> 10 \"j\" = a/(a+b+c) #> 11 \"sor\" = 2*a/(2*a+b+c) #> 12 \"m\" = (2*a+b+c)*(b+c)/(a+b+c) #> 13 \"-2\" = pmin.int(b,c)/(pmax.int(b,c)+a) #> 14 \"co\" = (a*c+a*b+2*b*c)/(2*(a+b)*(a+c)) #> 15 \"cc\" = (b+c)/(a+b+c) #> 16 \"g\" = (b+c)/(a+b+c) #> 17 \"-3\" = pmin.int(b,c)/(a+b+c) #> 18 \"l\" = (b+c)/2 #> 19 \"19\" = 2*(b*c+1)/(a+b+c)/(a+b+c-1) #> 20 \"hk\" = (b+c)/(2*a+b+c) #> 21 \"rlb\" = a/(a+c) #> 22 \"sim\" = pmin.int(b,c)/(pmin.int(b,c)+a) #> 23 \"gl\" = 2*abs(b-c)/(2*a+b+c) #> 24 \"z\" = (log(2)-log(2*a+b+c)+log(a+b+c))/log(2) ## The basic Whittaker index d <- betadiver(sipoo, \"w\") ## This should be equal to Sorensen index (binary Bray-Curtis in ## vegan) range(d - vegdist(sipoo, binary=TRUE)) #> [1] 0 0"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":null,"dir":"Reference","previous_headings":"","what":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function computes coefficients dispersal direction geographically connected areas, defined Legendre Legendre (1984), also described Legendre Legendre (2012, section 13.3.4).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"bgdispersal(mat, PAonly = FALSE, abc = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"mat Data frame matrix containing community composition data table (species presence-absence abundance data). PAonly FALSE four types coefficients, DD1 DD4, requested; TRUE DD1 DD2 sought (see Details). abc TRUE, return tables , b c used DD1 DD2.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"signs DD coefficients indicate direction dispersal, provided asymmetry significant. positive sign indicates dispersal first (row DD tables) second region (column); negative sign indicates opposite. McNemar test asymmetry computed presence-absence data test hypothesis significant asymmetry two areas comparison. input data table, rows sites areas, columns taxa. often, taxa species, coefficients can computed genera families well. DD1 DD2 computed presence-absence data. four types coefficients computed quantitative data, converted presence-absence computation DD1 DD2. PAonly = FALSE indicates four types coefficients requested. PAonly = TRUE DD1 DD2 sought.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Function bgdispersal returns list containing following matrices: DD1 \\(DD1_{j,k} = ((b - c))/((+ b + c)^2)\\) DD2 \\(DD2_{j,k} = (2 (b - c))/((2a + b + c) (+ b + c))\\) \\(\\), \\(b\\), \\(c\\) meaning computation binary similarity coefficients. DD3 \\(DD3_{j,k} = {W(-B) / (+B-W)^2} \\) DD4 \\(DD4_{j,k} = 2W(-B) / ((+B)(+B-W))\\) W = sum(pmin(vector1, vector2)), = sum(vector1), B = sum(vector2) McNemar McNemar chi-square statistic asymmetry (Sokal Rohlf 1995): \\(2(b \\log(b) + c \\log(c) - (b+c) \\log((b+c)/2)) / q\\), \\(q = 1 + 1/(2(b+c))\\) (Williams correction continuity) prob.McNemar probabilities associated McNemar statistics, chi-square test. H0: asymmetry \\((b-c)\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Legendre, P. V. Legendre. 1984. Postglacial dispersal freshwater fishes Québec peninsula. Can. J. Fish. Aquat. Sci. 41: 1781-1802. Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Sokal, R. R. F. J. Rohlf. 1995. Biometry. principles practice statistics biological research. 3rd edn. W. H. Freeman, New York.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"function uses powerful alternative McNemar test classical formula. classical formula constructed spirit Pearson's Chi-square, formula function constructed spirit Wilks Chi-square \\(G\\) statistic. Function mcnemar.test uses classical formula. new formula introduced vegan version 1.10-11, older implementations bgdispersal used classical formula.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bgdispersal.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Coefficients of Biogeographical Dispersal Direction — bgdispersal","text":"","code":"mat <- matrix(c(32,15,14,10,70,30,100,4,10,30,25,0,18,0,40, 0,0,20,0,0,0,0,4,0,30,20,0,0,0,0,25,74,42,1,45,89,5,16,16,20), 4, 10, byrow=TRUE) bgdispersal(mat) #> $DD1 #> [,1] [,2] [,3] [,4] #> [1,] 0.00 0.24 0.21 0.00 #> [2,] -0.24 0.00 0.08 -0.24 #> [3,] -0.21 -0.08 0.00 -0.21 #> [4,] 0.00 0.24 0.21 0.00 #> #> $DD2 #> [,1] [,2] [,3] [,4] #> [1,] 0.0000000 0.3428571 0.3230769 0.0000000 #> [2,] -0.3428571 0.0000000 0.1142857 -0.3428571 #> [3,] -0.3230769 -0.1142857 0.0000000 -0.3230769 #> [4,] 0.0000000 0.3428571 0.3230769 0.0000000 #> #> $DD3 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.1567922 0.1420408 -0.01325831 #> [2,] -0.15679216 0.0000000 0.1101196 -0.20049485 #> [3,] -0.14204082 -0.1101196 0.0000000 -0.13586560 #> [4,] 0.01325831 0.2004949 0.1358656 0.00000000 #> #> $DD4 #> [,1] [,2] [,3] [,4] #> [1,] 0.00000000 0.2513176 0.2425087 -0.01960102 #> [2,] -0.25131757 0.0000000 0.1725441 -0.30993929 #> [3,] -0.24250871 -0.1725441 0.0000000 -0.23381521 #> [4,] 0.01960102 0.3099393 0.2338152 0.00000000 #> #> $McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 7.677938 9.0571232 0.000000 #> [2,] NA NA 0.2912555 7.677938 #> [3,] NA NA NA 9.057123 #> [4,] NA NA NA NA #> #> $prob.McNemar #> [,1] [,2] [,3] [,4] #> [1,] NA 0.005590001 0.002616734 1.000000000 #> [2,] NA NA 0.589417103 0.005590001 #> [3,] NA NA NA 0.002616734 #> [4,] NA NA NA NA #>"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":null,"dir":"Reference","previous_headings":"","what":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Function finds best subset environmental variables, Euclidean distances scaled environmental variables maximum (rank) correlation community dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# S3 method for default bioenv(comm, env, method = \"spearman\", index = \"bray\", upto = ncol(env), trace = FALSE, partial = NULL, metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), parallel = getOption(\"mc.cores\"), ...) # S3 method for formula bioenv(formula, data, ...) bioenvdist(x, which = \"best\")"},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"comm Community data frame dissimilarity object square matrix can interpreted dissimilarities. env Data frame continuous environmental variables. method correlation method used cor. index dissimilarity index used community data (comm) vegdist. ignored comm dissimilarities. upto Maximum number parameters studied subsets. formula, data Model formula data. trace Trace calculations partial Dissimilarities partialled inspecting variables env. metric Metric used distances environmental distances. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. x bioenv result object. number model environmental distances evaluated, \"best\" model. ... arguments passed vegdist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function calculates community dissimilarity matrix using vegdist. selects possible subsets environmental variables, scales variables, calculates Euclidean distances subset using dist. function finds correlation community dissimilarities environmental distances, size subsets, saves best result. \\(2^p-1\\) subsets \\(p\\) variables, exhaustive search may take , , long time (parameter upto offers partial relief). argument metric defines distances given set environmental variables. metric = \"euclidean\", variables scaled unit variance Euclidean distances calculated. metric = \"mahalanobis\", Mahalanobis distances calculated: addition scaling unit variance, matrix current set environmental variables also made orthogonal (uncorrelated). metric = \"manhattan\", variables scaled unit range Manhattan distances calculated, distances sums differences environmental variables. metric = \"gower\", Gower distances calculated using function daisy. allows also using factor variables, continuous variables results equal metric = \"manhattan\". function can called model formula LHS data matrix RHS lists environmental variables. formula interface practical selecting transforming environmental variables. argument partial can perform “partial” analysis. partializing item must dissimilarity object class dist. partial item can used correlation method, strictly correct Pearson. Function bioenvdist recalculates environmental distances used within function. default calculate distances best model, number model can given. Clarke & Ainsworth (1993) suggested method used selecting best subset environmental variables interpreting results nonmetric multidimensional scaling (NMDS). recommended parallel display NMDS community dissimilarities NMDS Euclidean distances best subset scaled environmental variables. warned use Procrustes analysis, looks like good way comparing two ordinations. Clarke & Ainsworth wrote computer program BIO-ENV giving name current function. Presumably BIO-ENV later incorporated Clarke's PRIMER software (available Windows). addition, Clarke & Ainsworth suggested novel method rank correlation available current function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"function returns object class bioenv summary method.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Clarke, K. R & Ainsworth, M. 1993. method linking multivariate community structure environmental variables. Marine Ecology Progress Series, 92, 205--219.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"want study ‘significance’ bioenv results, can use function mantel mantel.partial use definition correlation. However, bioenv standardizes environmental variables depending used metric, must mantel comparable results (standardized data returned item x result object). safest use bioenvdist extract environmental distances really used within bioenv. NB., bioenv selects variables maximize Mantel correlation, significance tests based priori selection variables biased.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/bioenv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Best Subset of Environmental Variables with Maximum (Rank) Correlation with Community Dissimilarities — bioenv","text":"","code":"# The method is very slow for large number of possible subsets. # Therefore only 6 variables in this example. data(varespec) data(varechem) sol <- bioenv(wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, varechem) sol #> #> Call: #> bioenv(formula = wisconsin(varespec) ~ log(N) + P + K + Ca + pH + Al, data = varechem) #> #> Subset of environmental variables with best correlation to community data. #> #> Correlations: spearman #> Dissimilarities: bray #> Metric: euclidean #> #> Best model has 3 parameters (max. 6 allowed): #> P Ca Al #> with correlation 0.4004806 #> ## IGNORE_RDIFF_BEGIN summary(sol) #> size correlation #> P 1 0.2513 #> P Al 2 0.4004 #> P Ca Al 3 0.4005 #> P Ca pH Al 4 0.3619 #> log(N) P Ca pH Al 5 0.3216 #> log(N) P K Ca pH Al 6 0.2822 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA biplot — biplot.rda","title":"PCA biplot — biplot.rda","text":"Draws PCA biplot species scores indicated biplot arrows","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA biplot — biplot.rda","text":"","code":"# S3 method for rda biplot(x, choices = c(1, 2), scaling = \"species\", display = c(\"sites\", \"species\"), type, xlim, ylim, col = c(1,2), const, correlation = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA biplot — biplot.rda","text":"x rda result object. choices Axes show. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. negative scaling values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. display Scores shown. must alternatives \"species\" species scores, /\"sites\" site scores. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. Can length 2 (e.g. type = c(\"text\", \"points\")), case first element describes species scores handled, second site scores drawn. xlim, ylim x y limits (min, max) plot. col Colours used sites species (order). one colour given, used . const General scaling constant scores.rda. ... parameters plotting functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA biplot — biplot.rda","text":"Produces plot biplot results call rda. common \"species\" scores PCA drawn biplot arrows point direction increasing values variable. biplot.rda function provides wrapper plot.cca allow easy production plot. biplot.rda suitable unconstrained models. used ordination object constraints, error issued. species scores drawn using \"text\", arrows drawn origin 0.85 * species score, whilst labels drawn species score. type used \"points\", labels drawn therefore arrows drawn origin actual species score.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"PCA biplot — biplot.rda","text":"plot function returns invisibly plotting structure can used identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"PCA biplot — biplot.rda","text":"Gavin Simpson, based plot.cca Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/biplot.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA biplot — biplot.rda","text":"","code":"data(dune) mod <- rda(dune, scale = TRUE) biplot(mod, scaling = \"symmetric\") ## different type for species and site scores biplot(mod, scaling = \"symmetric\", type = c(\"text\", \"points\")) ## We can use ordiplot pipes in R 4.1 to build similar plots with ## flexible control if (FALSE) { if (getRversion() >= \"4.1\") { plot(mod, scaling = \"symmetric\", type=\"n\") |> text(\"sites\", cex=0.8) |> text(\"species\", arrows=TRUE, length=0.02, col=\"red\", cex=0.6) } }"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":null,"dir":"Reference","previous_headings":"","what":"K-means partitioning using a range of values of K — cascadeKM","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function wrapper kmeans function. creates several partitions forming cascade small large number groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"cascadeKM(data, inf.gr, sup.gr, iter = 100, criterion = \"calinski\", parallel = getOption(\"mc.cores\")) cIndexKM(y, x, index = \"all\") # S3 method for cascadeKM plot(x, min.g, max.g, grpmts.plot = TRUE, sortg = FALSE, gridcol = NA, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"K-means partitioning using a range of values of K — cascadeKM","text":"data data matrix. objects (samples) rows. inf.gr number groups partition smallest number groups cascade (min). sup.gr number groups partition largest \t number groups cascade (max). iter number random starting configurations value \\(K\\). criterion criterion used select best partition. default value \"calinski\", refers Calinski-Harabasz (1974) criterion. simple structure index (\"ssi\") also available. indices available package cclust. experience, two indices work best likely return maximum value near optimal number clusters \"calinski\" \"ssi\". y Object class \"kmeans\" returned clustering algorithm kmeans x Data matrix columns correspond variables rows observations, plotting object plot index available indices : \"calinski\" \"ssi\". Type \"\" obtain indices. Abbreviations names also accepted. min.g, max.g minimum maximum numbers groups displayed. grpmts.plot Show plot (TRUE FALSE). sortg Sort objects function group membership produce easily interpretable graph. See Details. original object names kept; used labels output table x, although graph. row names, sequential row numbers used keep track original order objects. gridcol colour grid lines plots. NA, default value, removes grid lines. ... parameters functions (ignored). parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"K-means partitioning using a range of values of K — cascadeKM","text":"function creates several partitions forming cascade small large number groups formed kmeans. work performed function cIndex based clustIndex package cclust). criteria removed version computation errors generated one object found group. default value \"calinski\", refers well-known Calinski-Harabasz (1974) criterion. available index simple structure index \"ssi\" (Dolnicar et al. 1999). case groups equal sizes, \"calinski\" generally good criterion indicate correct number groups. Users take indications literally groups equal size. Type \"\" obtain indices. indices defined : calinski: \\((SSB/(K-1))/(SSW/(n-K))\\), \\(n\\) number data points \\(K\\) number clusters. \\(SSW\\) sum squares within clusters \\(SSB\\) sum squares among clusters. index simply \\(F\\) (ANOVA) statistic. ssi: “Simple Structure Index” multiplicatively combines several elements influence interpretability partitioning solution. best partition indicated highest SSI value. simulation study, Milligan Cooper (1985) found Calinski-Harabasz criterion recovered correct number groups often. recommend criterion , groups equal sizes, maximum value \"calinski\" usually indicates correct number groups. Another available index simple structure index \"ssi\". Users take indications indices literally groups equal size explore groups corresponding values \\(K\\). Function cascadeKM plot method. Two plots produced. graph left objects abscissa number groups ordinate. groups represented colours. graph right shows values criterion (\"calinski\" \"ssi\") determining best partition. highest value criterion marked red. Points marked orange, , indicate partitions producing increase criterion value number groups increases; may represent interesting partitions. sortg=TRUE, objects reordered following procedure: (1) simple matching distance matrix computed among objects, based table K-means assignments groups, \\(K\\) = min.g \\(K\\) = max.g. (2) principal coordinate analysis (PCoA, Gower 1966) computed centred distance matrix. (3) first principal coordinate used new order objects graph. simplified algorithm used compute first principal coordinate , using iterative algorithm described Legendre & Legendre (2012). full distance matrix among objects never computed; avoids problem storing number objects large. Distance values computed needed algorithm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Function cascadeKM returns object class cascadeKM items: partition Table partitions found different numbers groups \\(K\\), \\(K\\) = inf.gr \\(K\\) = sup.gr. results Values criterion select best partition. criterion name criterion used. size number objects found group, partitions (columns). Function cIndex returns vector index values. maximum value indices supposed indicate best partition. indices work best groups equal sizes. groups equal sizes, one put much faith maximum indices, also explore groups corresponding values \\(K\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Calinski, T. J. Harabasz. 1974. dendrite method cluster analysis. Commun. Stat. 3: 1--27. Dolnicar, S., K. Grabler J. . Mazanec. 1999. tale three cities: perceptual charting analyzing destination images. Pp. 39-62 : Woodside, . et al. [eds.] Consumer psychology tourism, hospitality leisure. CAB International, New York. Gower, J. C. 1966. distance properties latent root vector methods used multivariate analysis. Biometrika 53: 325--338. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Milligan, G. W. & M. C. Cooper. 1985. examination procedures determining number clusters data set. Psychometrika 50: 159--179. Weingessel, ., Dimitriadou, . Dolnicar, S. 2002. examination indexes determining number clusters binary data sets. Psychometrika 67: 137--160.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"K-means partitioning using a range of values of K — cascadeKM","text":"Marie-Helene Ouellette Marie-Helene.Ouellette@UMontreal.ca, Sebastien Durand Sebastien.Durand@UMontreal.ca Pierre Legendre Pierre.Legendre@UMontreal.ca. Parallel processing Virgilio Gómez-Rubio. Edited vegan Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cascadeKM.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"K-means partitioning using a range of values of K — cascadeKM","text":"","code":"# Partitioning a (10 x 10) data matrix of random numbers mat <- matrix(runif(100),10,10) res <- cascadeKM(mat, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning an autocorrelated time series vec <- sort(matrix(runif(30),30,1)) res <- cascadeKM(vec, 2, 5, iter = 25, criterion = 'calinski') toto <- plot(res) # Partitioning a large autocorrelated time series # Note that we remove the grid lines vec <- sort(matrix(runif(1000),1000,1)) res <- cascadeKM(vec, 2, 7, iter = 10, criterion = 'calinski') toto <- plot(res, gridcol=NA)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca performs correspondence analysis, optionally constrained correspondence analysis (.k.. canonical correspondence analysis), optionally partial constrained correspondence analysis. Function rda performs redundancy analysis, optionally principal components analysis. popular ordination techniques community ecology.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"# S3 method for formula cca(formula, data, na.action = na.fail, subset = NULL, ...) # S3 method for formula rda(formula, data, scale=FALSE, na.action = na.fail, subset = NULL, ...) # S3 method for default cca(X, Y, Z, ...) # S3 method for default rda(X, Y, Z, scale=FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"formula Model formula, left hand side gives community data matrix, right hand side gives constraining variables, conditioning variables can given within special function Condition. data Data frame containing variables right hand side model formula. X Community data matrix. Y Constraining matrix, typically environmental variables. Can missing. data.frame, expanded model.matrix factors expanded contrasts (“dummy variables”). better use formula instead argument, analyses work formula used. Z Conditioning matrix, effect removed (“partialled ”) next step. Can missing. data.frame, expanded similarly constraining matrix. scale Scale species unit variance (like correlations). na.action Handling missing values constraints conditions. default (na.fail) stop missing value. Choice na.omit removes rows missing values. Choice na.exclude keeps observations gives NA results calculated. WA scores rows may found also missing values constraints. Missing values never allowed dependent community data. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data. ... arguments print plot functions (ignored functions).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Since introduction (ter Braak 1986), constrained, canonical, correspondence analysis spin-, redundancy analysis, popular ordination methods community ecology. Functions cca rda similar popular proprietary software Canoco, although implementation completely different. functions based Legendre & Legendre's (2012) algorithm: cca Chi-square transformed data matrix subjected weighted linear regression constraining variables, fitted values submitted correspondence analysis performed via singular value decomposition (svd). Function rda similar, uses ordinary, unweighted linear regression unweighted SVD. Legendre & Legendre (2012), Table 11.5 (p. 650) give skeleton RDA algorithm vegan. algorithm CCA similar, involves standardization row column weights. functions can called either matrix-like entries community data constraints, formula interface. general, formula interface preferred, allows better control model allows factor constraints. analyses ordination results possible model fitted formula (e.g., cases anova.cca, automatic model building). following sections, X, Y Z, although referred matrices, commonly data frames. matrix interface, community data matrix X must given, data matrices may omitted, corresponding stage analysis skipped. matrix Z supplied, effects removed community matrix, residual matrix submitted next stage. called partial correspondence redundancy analysis. matrix Y supplied, used constrain ordination, resulting constrained canonical correspondence analysis, redundancy analysis. Finally, residual submitted ordinary correspondence analysis (principal components analysis). matrices Z Y missing, data matrix analysed ordinary correspondence analysis ( principal components analysis). Instead separate matrices, model can defined using model formula. left hand side must community data matrix (X). right hand side defines constraining model. constraints can contain ordered unordered factors, interactions among variables functions variables. defined contrasts honoured factor variables. constraints can also matrices (data frames). formula can include special term Condition conditioning variables (“covariables”) partialled analysis. following commands equivalent: cca(X, Y, Z), cca(X ~ Y + Condition(Z)), Y Z refer constraints conditions matrices respectively. Constrained correspondence analysis indeed constrained method: CCA try display variation data, part can explained used constraints. Consequently, results strongly dependent set constraints transformations interactions among constraints. shotgun method use environmental variables constraints. However, exploratory problems better analysed unconstrained methods correspondence analysis (decorana, corresp) non-metric multidimensional scaling (metaMDS) environmental interpretation analysis (envfit, ordisurf). CCA good choice user clear strong priori hypotheses constraints interested major structure data set. CCA able correct curve artefact commonly found correspondence analysis forcing configuration linear constraints. However, curve artefact can avoided low number constraints curvilinear relation . curve can reappear even two badly chosen constraints single factor. Although formula interface makes easy include polynomial interaction terms, terms often produce curved artefacts (difficult interpret), probably avoided. According folklore, rda used ``short gradients'' rather cca. However, based research finds methods based Euclidean metric uniformly weaker based Chi-squared metric. However, standardized Euclidean distance may appropriate measures (see Hellinger standardization decostand particular). Partial CCA (pCCA; alternatively partial RDA) can used remove effect conditioning background random variables covariables CCA proper. fact, pCCA compares models cca(X ~ Z) cca(X ~ Y + Z) attributes difference effect Y cleansed effect Z. people used method extracting “components variance” CCA. However, effect variables together stronger sum separately, can increase total Chi-square partialling variation, give negative “components variance”. general, components “variance” trusted due interactions two sets variables. functions summary plot methods documented separately (see plot.cca, summary.cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"Function cca returns huge object class cca, described separately cca.object. Function rda returns object class rda inherits class cca described cca.object. scaling used rda scores described separate vignette package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"original method ter Braak, current implementation follows Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. McCune, B. (1997) Influence noisy environmental data canonical correspondence analysis. Ecology 78, 2617-2623. Palmer, M. W. (1993) Putting things even better order: advantages canonical correspondence analysis. Ecology 74,2215-2230. Ter Braak, C. J. F. (1986) Canonical Correspondence Analysis: new eigenvector technique multivariate direct gradient analysis. Ecology 67, 1167-1179.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"responsible author Jari Oksanen, code borrows heavily Dave Roberts (Montana State University, USA).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] [Constrained] Correspondence Analysis and Redundancy Analysis — cca","text":"","code":"data(varespec) data(varechem) ## Common but bad way: use all variables you happen to have in your ## environmental data matrix vare.cca <- cca(varespec, varechem) vare.cca #> Call: cca(X = varespec, Y = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Constrained 1.4415 0.6920 14 #> Unconstrained 0.6417 0.3080 9 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4389 0.2918 0.1628 0.1421 0.1180 0.0890 0.0703 0.0584 0.0311 0.0133 0.0084 #> CCA12 CCA13 CCA14 #> 0.0065 0.0062 0.0047 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 #> 0.19776 0.14193 0.10117 0.07079 0.05330 0.03330 0.01887 0.01510 0.00949 #> plot(vare.cca) ## Formula interface and a better model vare.cca <- cca(varespec ~ Al + P*(K + Baresoil), data=varechem) vare.cca #> Call: cca(formula = varespec ~ Al + P * (K + Baresoil), data = #> varechem) #> #> Inertia Proportion Rank #> Total 2.083 1.000 #> Constrained 1.046 0.502 6 #> Unconstrained 1.038 0.498 17 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 0.3756 0.2342 0.1407 0.1323 0.1068 0.0561 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.27577 0.15411 0.13536 0.11803 0.08887 0.05511 0.04919 0.03781 #> (Showing 8 of 17 unconstrained eigenvalues) #> plot(vare.cca) ## Partialling out and negative components of variance cca(varespec ~ Ca, varechem) #> Call: cca(formula = varespec ~ Ca, data = varechem) #> #> Inertia Proportion Rank #> Total 2.08320 1.00000 #> Constrained 0.15722 0.07547 1 #> Unconstrained 1.92598 0.92453 22 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.15722 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.4745 0.2939 0.2140 0.1954 0.1748 0.1171 0.1121 0.0880 #> (Showing 8 of 22 unconstrained eigenvalues) #> cca(varespec ~ Ca + Condition(pH), varechem) #> Call: cca(formula = varespec ~ Ca + Condition(pH), data = varechem) #> #> Inertia Proportion Rank #> Total 2.0832 1.0000 #> Conditional 0.1458 0.0700 1 #> Constrained 0.1827 0.0877 1 #> Unconstrained 1.7547 0.8423 21 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 #> 0.18269 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3834 0.2749 0.2123 0.1760 0.1701 0.1161 0.1089 0.0880 #> (Showing 8 of 21 unconstrained eigenvalues) #> ## RDA data(dune) data(dune.env) dune.Manure <- rda(dune ~ Manure, dune.env) plot(dune.Manure)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":null,"dir":"Reference","previous_headings":"","what":"Result Object from Constrained Ordination — cca.object","title":"Result Object from Constrained Ordination — cca.object","text":"Ordination methods cca, rda, dbrda capscale return similar result objects. methods use internal function ordConstrained. differ (1) initial transformation data defining inertia, (2) weighting, (3) use rectangular rows \\(\\times\\) columns data symmetric rows \\(\\times\\) rows dissimilarities: rda initializes data give variance correlations inertia, cca based double-standardized data give Chi-square inertia uses row column weights, capscale maps real part dissimilarities rectangular data performs RDA, dbrda performs RDA-like analysis directly symmetric dissimilarities. Function ordConstrained returns result components methods, calling function may add components final result. However, access result components directly (using $): internal structure regarded stable application interface (API), can change release. access results components directly, take risk breakage vegan release. vegan provides wide set accessor functions components, functions updated result object changes. documentation gives overview accessor functions cca result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Result Object from Constrained Ordination — cca.object","text":"","code":"ordiYbar(x, model = c(\"CCA\", \"CA\", \"pCCA\", \"partial\", \"initial\")) # S3 method for cca model.frame(formula, ...) # S3 method for cca model.matrix(object, ...) # S3 method for cca weights(object, display = \"sites\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Result Object from Constrained Ordination — cca.object","text":"object, x, formula result object cca, rda, dbrda, capscale. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. ordiYbar value can also \"initial\" internal working input data, \"partial\" internal working input data removing partial effects. display Display either \"sites\" \"species\". ... arguments passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Result Object from Constrained Ordination — cca.object","text":"internal (“working”) form dependent (community) data can accessed function ordiYbar. form depends ordination method: instance, cca data weighted Chi-square transformed, dbrda Gower-centred dissimilarities. input data original (“response”) form can accessed fitted.cca residuals.cca. Function predict.cca can return either working response data, also lower-rank approximations. model matrix independent data (“Constraints” “Conditions”) can extracted model.matrix. partial analysis, function returns list design matrices called Conditions Constraints. either component missing, single matrix returned. redundant (aliased) terms appear model matrix. terms can found alias.cca. Function model.frame tries reconstruct data frame model matrices derived. possible original model fitted formula data arguments, still fails data unavailable. number observations can accessed nobs.cca, residual degrees freedom df.residual.cca. information observations missing values can accessed na.action. terms formula fitted model can accessed formula terms. weights used cca can accessed weights. unweighted methods (rda) weights equal. ordination results saved separate components partial terms, constraints residual unconstrained ordination. guarantee components internal names currently, cautious developing scripts functions directly access components. constrained ordination algorithm based QR decomposition constraints conditions (environmental data), QR component saved separately partial constrained components. QR decomposition constraints can accessed qr.cca. also include residual effects partial terms (Conditions), used together ordiYbar(x, \"partial\"). environmental data first centred rda weighted centred cca. QR decomposition used many functions access cca results, can used find many items directly stored object. examples, see coef.cca, coef.rda, vif.cca, permutest.cca, predict.cca, predict.rda, calibrate.cca. See qr possible uses component. instance, rank constraints can found QR decomposition. eigenvalues solution can accessed eigenvals.cca. Eigenvalues evaluated partial component, available constrained residual components. ordination scores internally stored (weighted) orthonormal scores matrices. results can accessed scores.cca scores.rda functions. ordination scores scaled accessed scores functions, internal (weighted) orthonormal scores can accessed setting scaling = FALSE. Unconstrained residual component species site scores, constrained component also fitted site scores linear combination scores sites biplot scores centroids constraint variables. biplot scores correspond model.matrix, centroids calculated factor variables used. scores can selected defining axes, direct way accessing scores certain component. number dimensions can assessed eigenvals. addition, types can derived results although saved results. instance, regression scores model coefficients can accessed scores coef functions. Partial component scores. Distance-based methods (dbrda, capscale) can negative eigenvalues associated imaginary axis scores. addition, species scores initially missing dbrda accessory found analysis capscale (may misleading). Function sppscores can used add species scores replace meaningful ones.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Result Object from Constrained Ordination — cca.object","text":"latest large change result object made release 2.5-1 2016. can modernize ancient stray results modernobject <- update(ancientobject).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Result Object from Constrained Ordination — cca.object","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/cca.object.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Result Object from Constrained Ordination — cca.object","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":null,"dir":"Reference","previous_headings":"","what":"Multinomial Species Classification Method (CLAM) — clamtest","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"CLAM statistical approach classifying generalists specialists two distinct habitats described Chazdon et al. (2011).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"clamtest(comm, groups, coverage.limit = 10, specialization = 2/3, npoints = 20, alpha = 0.05/20) # S3 method for clamtest summary(object, ...) # S3 method for clamtest plot(x, xlab, ylab, main, pch = 21:24, col.points = 1:4, col.lines = 2:4, lty = 1:3, position = \"bottomright\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"comm Community matrix, consisting counts. groups vector identifying two habitats. Must exactly two unique values levels. Habitat IDs grouping vector must match corresponding rows community matrix comm. coverage.limit Integer, sample coverage based correction applied rare species counts limit. Sample coverage calculated separately two habitats. Sample relative abundances used species higher equal coverage.limit total counts per habitat. specialization Numeric, specialization threshold value 0 1. value \\(2/3\\) represents ‘supermajority’ rule, value \\(1/2\\) represents ‘simple majority’ rule assign shared species habitat specialists. npoints Integer, number points used determine boundary lines plots. alpha Numeric, nominal significance level individual tests. default value reduces conventional limit \\(0.05\\) account overdispersion multiple testing several species simultaneously. However, firm reason exactly limit. x, object Fitted model object class \"clamtest\". xlab, ylab Labels plot axes. main Main title plot. pch, col.points Symbols colors used plotting species groups. lty, col.lines Line types colors boundary lines plot separate species groups. position Position figure legend, see legend specification details. Legend shown position = NULL. ... Additional arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"method uses multinomial model based estimated species relative abundance two habitats (, B). minimizes bias due differences sampling intensities two habitat types well bias due insufficient sampling within habitat. method permits robust statistical classification habitat specialists generalists, without excluding rare species priori (Chazdon et al. 2011). Based user-defined specialization threshold, model classifies species one four groups: (1) generalists; (2) habitat specialists; (3) habitat B specialists; (4) rare classify confidence.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"data frame (class attribute \"clamtest\"), columns: Species: species name (column names comm), Total_**: total count habitat , Total_*B*: total count habitat B, Classes: species classification, factor levels Generalist, Specialist_**, Specialist_*B*, Too_rare. ** *B* placeholders habitat names/labels found data. summary method returns descriptive statistics results. plot method returns values invisibly produces bivariate scatterplot species total abundances two habitats. Symbols boundary lines shown species groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Chazdon, R. L., Chao, ., Colwell, R. K., Lin, S.-Y., Norden, N., Letcher, S. G., Clark, D. B., Finegan, B. Arroyo J. P.(2011). novel statistical method classifying habitat generalists specialists. Ecology 92, 1332--1343.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"Peter Solymos solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"code tested standalone CLAM software provided website Anne Chao (http://chao.stat.nthu.edu.tw/wordpress); minor inconsistencies found, especially finding threshold 'rare' species. inconsistencies probably due numerical differences two implementation. current R implementation uses root finding iso-lines instead iterative search. original method (Chazdon et al. 2011) two major problems: assumes error distribution multinomial. justified choice individuals freely distributed, -dispersion clustering individuals. ecological data, variance much higher multinomial assumption, therefore test statistic optimistic. original authors suggest multiple testing adjustment multiple testing based number points (npoints) used draw critical lines plot, whereas adjustment based number tests (.e., tested species). function uses numerical values original paper, automatic connection npoints alpha arguments, must work adjustment .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/clamtest.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multinomial Species Classification Method (CLAM) — clamtest","text":"","code":"data(mite) data(mite.env) sol <- with(mite.env, clamtest(mite, Shrub==\"None\", alpha=0.005)) summary(sol) #> Two Groups Species Classification Method (CLAM) #> #> Specialization threshold = 0.6666667 #> Alpha level = 0.005 #> #> Estimated sample coverage: #> FALSE TRUE #> 1.0000 0.9996 #> #> Minimum abundance for classification: #> FALSE TRUE #> 27 9 #> #> Species Proportion #> Generalist 10 0.286 #> Specialist_FALSE 14 0.400 #> Specialist_TRUE 4 0.114 #> Too_rare 7 0.200 head(sol) #> Species Total_FALSE Total_TRUE Classes #> 1 Brachy 534 77 Generalist #> 2 PHTH 89 0 Specialist_FALSE #> 3 HPAV 389 207 Generalist #> 4 RARD 85 0 Specialist_FALSE #> 5 SSTR 22 0 Too_rare #> 6 Protopl 26 0 Too_rare plot(sol)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":null,"dir":"Reference","previous_headings":"","what":"Create an Object for Null Model Algorithms — commsim","title":"Create an Object for Null Model Algorithms — commsim","text":"commsim function can used feed Null Model algorithms nullmodel analysis. make.commsim function returns various predefined algorithm types (see Details). functions represent low level interface community null model infrastructure vegan intent extensibility, less emphasis direct use users.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"commsim(method, fun, binary, isSeq, mode) make.commsim(method) # S3 method for commsim print(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create an Object for Null Model Algorithms — commsim","text":"method Character, name algorithm. fun function. possible formal arguments function see Details. binary Logical, algorithm applies presence-absence count matrices. isSeq Logical, algorithm sequential (needs burnin thinning) . mode Character, storage mode community matrix, either \"integer\" \"double\". x object class commsim. ... Additional arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Create an Object for Null Model Algorithms — commsim","text":"function fun must return array dim(nr, nc, n), must take following arguments: x: input matrix, n: number permuted matrices output, nr: number rows, nc: number columns, rs: vector row sums, cs: vector column sums, rf: vector row frequencies (non-zero cells), cf: vector column frequencies (non-zero cells), s: total sum x, fill: matrix fill (non-zero cells), thin: thinning value sequential algorithms, ...: additional arguments. can define null model, several null model algorithm pre-defined can called name. predefined algorithms described detail following chapters. binary null models produce matrices zeros (absences) ones (presences) also input matrix quantitative. two types quantitative data: Counts integers natural unit individuals can shuffled, abundances can real (floating point) values natural subunit shuffling. quantitative models can handle counts, able handle real values. null models sequential next matrix derived current one. makes models dependent previous models, usually must thin matrices study sequences stability: see oecosimu details instructions. See Examples structural constraints imposed algorithm defining null model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"binary-null-models","dir":"Reference","previous_headings":"","what":"Binary null models","title":"Create an Object for Null Model Algorithms — commsim","text":"binary null models preserve fill: number presences conversely number absences. classic models may also preserve column (species) frequencies (c0) row frequencies species richness site (r0) take account commonness rarity species (r1, r2). Algorithms swap, tswap, curveball, quasiswap backtracking preserve row column frequencies. Three first ones sequential two latter non-sequential produce independent matrices. Basic algorithms reviewed Wright et al. (1998). \"r00\": non-sequential algorithm binary matrices preserves number presences (fill). \"r0\": non-sequential algorithm binary matrices preserves site (row) frequencies. \"r1\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses column marginal frequencies probabilities selecting species. \"r2\": non-sequential algorithm binary matrices preserves site (row) frequencies, uses squared column marginal frequencies probabilities selecting species. \"c0\": non-sequential algorithm binary matrices preserves species frequencies (Jonsson 2001). \"swap\": sequential algorithm binary matrices changes matrix structure, influence marginal sums (Gotelli & Entsminger 2003). inspects \\(2 \\times 2\\) submatrices long swap can done. \"tswap\": sequential algorithm binary matrices. \"swap\" algorithm, tries fixed number times performs zero many swaps one step (according thin argument call). approach suggested Miklós & Podani (2004) found ordinary swap may lead biased sequences, since columns rows easily swapped. \"curveball\": sequential method binary matrices implements ‘Curveball’ algorithm Strona et al. (2014). algorithm selects two random rows finds set unique species occur one rows. algorithm distributes set unique species rows preserving original row frequencies. Zero several species swapped one step, usually matrix perturbed strongly sequential methods. \"quasiswap\": non-sequential algorithm binary matrices implements method matrix first filled honouring row column totals, integers may larger one. method inspects random \\(2 \\times 2\\) matrices performs quasiswap . addition ordinary swaps, quasiswap can reduce numbers one ones preserving marginal totals (Miklós & Podani 2004). method non-sequential, accepts thin argument: convergence checked every thin steps. allows performing several ordinary swaps addition fill changing swaps helps reducing removing bias. \"greedyqswap\": greedy variant quasiswap. greedy step, one element \\(2 \\times 2\\) matrix taken \\(> 1\\) elements. greedy steps biased, method can thinned, first thin steps greedy. Even modest thinning (say thin = 20) removes reduces bias, thin = 100 (1% greedy steps) looks completely safe still speeds simulation. code experimental provided scrutiny, tested bias use. \"backtracking\": non-sequential algorithm binary matrices implements filling method constraints row column frequencies (Gotelli & Entsminger 2001). matrix first filled randomly, typically row column sums reached incidences filled . begins \"backtracking\", incidences removed, filling started , backtracking done many times incidences filled matrix. results may biased inspected carefully use.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-models-for-counts-with-fixed-marginal-sums","dir":"Reference","previous_headings":"","what":"Quantitative Models for Counts with Fixed Marginal Sums","title":"Create an Object for Null Model Algorithms — commsim","text":"models shuffle individuals counts keep marginal sums fixed, marginal frequencies preserved. Algorithm r2dtable uses standard R function r2dtable also used simulated \\(P\\)-values chisq.test. Algorithm quasiswap_count uses , preserves original fill. Typically means increasing numbers zero cells result zero-inflated respect r2dtable. \"r2dtable\": non-sequential algorithm count matrices. algorithm keeps matrix sum row/column sums constant. Based r2dtable. \"quasiswap_count\": non-sequential algorithm count matrices. algorithm similar Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function preserving row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping \\(2 \\times 2\\) submatrices (see \"swap_count\" algorithm details) maintain row column totals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative swap models similar binary swap, swap largest permissible value. models section maintain fill perform quantitative swap can done without changing fill. Single step swap often changes matrix little. particular, cell counts variable, high values change slowly. Checking chain stability independence even crucial binary swap, strong thinning often needed. models never used without inspecting properties current data. null models can also defined using permatswap function. \"swap_count\": sequential algorithm count matrices. algorithm find \\(2 \\times 2\\) submatrices can swapped leaving column row totals fill unchanged. algorithm finds largest value submatrix can swapped (\\(d\\)). Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. \"abuswap_r\": sequential algorithm count nonnegative real valued matrices fixed row frequencies (see also permatswap). algorithm similar swap_count, uses different swap value row \\(2 \\times 2\\) submatrix. step changes corresponding column sums, honours matrix fill, row sums, row/column frequencies (Hardy 2008; randomization scheme 2x). \"abuswap_c\": sequential algorithm count nonnegative real valued matrices fixed column frequencies (see also permatswap). algorithm similar previous one, operates columns. step changes corresponding row sums, honours matrix fill, column sums, row/column frequencies (Hardy 2008; randomization scheme 3x).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-swap-and-shuffle-models","dir":"Reference","previous_headings":"","what":"Quantitative Swap and Shuffle Models","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative Swap Shuffle methods (swsh methods) preserve fill column row frequencies, also either row column sums. methods first perform binary quasiswap shuffle original quantitative data non-zero cells. samp methods shuffle original non-zero cell values can used also non-integer data. methods redistribute individuals randomly among non-zero cells can used integer data. shuffling either free whole matrix, within rows (r methods) within columns (c methods). Shuffling within row preserves row sums, shuffling within column preserves column sums. models can also defined permatswap. \"swsh_samp\": non-sequential algorithm quantitative data (either integer counts non-integer values). Original non-zero values values shuffled. \"swsh_both\": non-sequential algorithm count data. Individuals shuffled freely non-zero cells. \"swsh_samp_r\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately row. \"swsh_samp_c\": non-sequential algorithm quantitative data. Non-zero values (samples) shuffled separately column. \"swsh_both_r\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values within row. \"swsh_both_c\": non-sequential algorithm count matrices. Individuals shuffled freely non-zero values column.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"quantitative-shuffle-methods","dir":"Reference","previous_headings":"","what":"Quantitative Shuffle Methods","title":"Create an Object for Null Model Algorithms — commsim","text":"Quantitative shuffle methods generalizations binary models r00, r0 c0. _ind methods shuffle individuals grand sum, row sum column sums preserved. methods similar r2dtable still slacker constraints marginal sums. _samp _both methods first apply corresponding binary model similar restriction marginal frequencies distribute quantitative values non-zero cells. _samp models shuffle original cell values can therefore handle also non-count real values. _both models shuffle individuals among non-zero values. shuffling whole matrix r00_, within row r0_ within column c0_ cases. \"r00_ind\": non-sequential algorithm count matrices. algorithm preserves grand sum individuals shuffled among cells matrix. \"r0_ind\": non-sequential algorithm count matrices. algorithm preserves row sums individuals shuffled among cells row matrix. \"c0_ind\": non-sequential algorithm count matrices. algorithm preserves column sums individuals shuffled among cells column matrix. \"r00_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves grand sum cells matrix shuffled. \"r0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves row sums cells within row shuffled. \"c0_samp\": non-sequential algorithm count nonnegative real valued (mode = \"double\") matrices. algorithm preserves column sums constant cells within column shuffled. \"r00_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells matrix shuffled. \"r0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells row shuffled. \"c0_both\": non-sequential algorithm count matrices. algorithm preserves grand sum cells individuals among cells column shuffled.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create an Object for Null Model Algorithms — commsim","text":"object class commsim elements corresponding arguments (method, binary, isSeq, mode, fun). input make.comsimm commsim object, returned without evaluation. case, character method argument matched predefined algorithm names. error message issued none found. method argument missing, function returns names currently available null model algorithms character vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Create an Object for Null Model Algorithms — commsim","text":"Gotelli, N.J. & Entsminger, N.J. (2001). Swap fill algorithms null model analysis: rethinking knight's tour. Oecologia 129, 281--291. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97. Strona, G., Nappo, D., Boccacci, F., Fattorini, S. & San-Miguel-Ayanz, J. (2014). fast unbiased procedure randomize ecological binary matrices fixed row column totals. Nature Communications 5:4114 doi:10.1038/ncomms5114 . Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Create an Object for Null Model Algorithms — commsim","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/commsim.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Create an Object for Null Model Algorithms — commsim","text":"","code":"## write the r00 algorithm f <- function(x, n, ...) array(replicate(n, sample(x)), c(dim(x), n)) (cs <- commsim(\"r00\", fun=f, binary=TRUE, isSeq=FALSE, mode=\"integer\")) #> An object of class “commsim” #> ‘r00’ method (binary, non-sequential, integer mode) #> ## retrieving the sequential swap algorithm (cs <- make.commsim(\"swap\")) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## feeding a commsim object as argument make.commsim(cs) #> An object of class “commsim” #> ‘swap’ method (binary, sequential, integer mode) #> ## making the missing c1 model using r1 as a template ## non-sequential algorithm for binary matrices ## that preserves the species (column) frequencies, ## but uses row marginal frequencies ## as probabilities of selecting sites f <- function (x, n, nr, nc, rs, cs, ...) { out <- array(0L, c(nr, nc, n)) J <- seq_len(nc) storage.mode(rs) <- \"double\" for (k in seq_len(n)) for (j in J) out[sample.int(nr, cs[j], prob = rs), j, k] <- 1L out } cs <- make.commsim(\"r1\") cs$method <- \"c1\" cs$fun <- f ## structural constraints diagfun <- function(x, y) { c(sum = sum(y) == sum(x), fill = sum(y > 0) == sum(x > 0), rowSums = all(rowSums(y) == rowSums(x)), colSums = all(colSums(y) == colSums(x)), rowFreq = all(rowSums(y > 0) == rowSums(x > 0)), colFreq = all(colSums(y > 0) == colSums(x > 0))) } evalfun <- function(meth, x, n) { m <- nullmodel(x, meth) y <- simulate(m, nsim=n) out <- rowMeans(sapply(1:dim(y)[3], function(i) diagfun(attr(y, \"data\"), y[,,i]))) z <- as.numeric(c(attr(y, \"binary\"), attr(y, \"isSeq\"), attr(y, \"mode\") == \"double\")) names(z) <- c(\"binary\", \"isSeq\", \"double\") c(z, out) } x <- matrix(rbinom(10*12, 1, 0.5)*rpois(10*12, 3), 12, 10) algos <- make.commsim() a <- t(sapply(algos, evalfun, x=x, n=10)) print(as.table(ifelse(a==1,1,0)), zero.print = \".\") #> binary isSeq double sum fill rowSums colSums rowFreq colFreq #> r00 1 . . 1 1 . . . . #> c0 1 . . 1 1 . 1 . 1 #> r0 1 . . 1 1 1 . 1 . #> r1 1 . . 1 1 1 . 1 . #> r2 1 . . 1 1 1 . 1 . #> quasiswap 1 . . 1 1 1 1 1 1 #> greedyqswap 1 . . 1 1 1 1 1 1 #> swap 1 1 . 1 1 1 1 1 1 #> tswap 1 1 . 1 1 1 1 1 1 #> curveball 1 1 . 1 1 1 1 1 1 #> backtrack 1 . . 1 1 1 1 1 1 #> r2dtable . . . 1 . 1 1 . . #> swap_count . 1 . 1 1 1 1 . . #> quasiswap_count . . . 1 1 1 1 . . #> swsh_samp . . 1 1 1 . . 1 1 #> swsh_both . . . 1 1 . . 1 1 #> swsh_samp_r . . 1 1 1 1 . 1 1 #> swsh_samp_c . . 1 1 1 . 1 1 1 #> swsh_both_r . . . 1 1 1 . 1 1 #> swsh_both_c . . . 1 1 . 1 1 1 #> abuswap_r . 1 1 1 1 1 . 1 1 #> abuswap_c . 1 1 1 1 . 1 1 1 #> r00_samp . . 1 1 1 . . . . #> c0_samp . . 1 1 1 . 1 . 1 #> r0_samp . . 1 1 1 1 . 1 . #> r00_ind . . . 1 . . . . . #> c0_ind . . . 1 . . 1 . . #> r0_ind . . . 1 . 1 . . . #> r00_both . . . 1 1 . . . . #> c0_both . . . 1 1 . 1 . 1 #> r0_both . . . 1 1 1 . 1 ."},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":null,"dir":"Reference","previous_headings":"","what":"Contribution Diversity Approach — contribdiv","title":"Contribution Diversity Approach — contribdiv","text":"contribution diversity approach based differentiation within-unit among-unit diversity using additive diversity partitioning unit distinctiveness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Contribution Diversity Approach — contribdiv","text":"","code":"contribdiv(comm, index = c(\"richness\", \"simpson\"), relative = FALSE, scaled = TRUE, drop.zero = FALSE) # S3 method for contribdiv plot(x, sub, xlab, ylab, ylim, col, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Contribution Diversity Approach — contribdiv","text":"comm community data matrix samples rows species column. index Character, diversity index calculated. relative Logical, TRUE contribution diversity values expressed signed deviation mean. See details. scaled Logical, TRUE relative contribution diversity values scaled sum gamma values (index = \"richness\") sum gamma values times number rows comm (index = \"simpson\"). See details. drop.zero Logical, empty rows dropped result? empty rows dropped, corresponding results NAs. x object class \"contribdiv\". sub, xlab, ylab, ylim, col Graphical arguments passed plot. ... arguments passed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Contribution Diversity Approach — contribdiv","text":"approach proposed Lu et al. (2007). Additive diversity partitioning (see adipart references) deals relation mean alpha total (gamma) diversity. Although alpha diversity values often vary considerably. Thus, contributions sites total diversity uneven. site specific contribution measured contribution diversity components. unit e.g. many unique species contribute higher level (gamma) diversity another unit number species, common. Distinctiveness species \\(j\\) can defined number sites occurs (\\(n_j\\)), sum relative frequencies (\\(p_j\\)). Relative frequencies computed sitewise \\(sum_j{p_ij}\\)s site \\(\\) sum \\(1\\). contribution site \\(\\) total diversity given \\(alpha_i = sum_j(1 / n_ij)\\) dealing richness \\(alpha_i = sum(p_{ij} * (1 - p_{ij}))\\) Simpson index. unit distinctiveness site \\(\\) average species distinctiveness, averaging species occur site \\(\\). species richness: \\(alpha_i = mean(n_i)\\) (paper, second equation contains typo, \\(n\\) without index). Simpson index: \\(alpha_i = mean(n_i)\\). Lu et al. (2007) gives -depth description different indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Contribution Diversity Approach — contribdiv","text":"object class \"contribdiv\" inheriting data frame. Returned values alpha, beta gamma components sites (rows) community matrix. \"diff.coef\" attribute gives differentiation coefficient (see Examples).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Contribution Diversity Approach — contribdiv","text":"Lu, H. P., Wagner, H. H. Chen, X. Y. 2007. contribution diversity approach evaluate species diversity. Basic Applied Ecology, 8, 1--12.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Contribution Diversity Approach — contribdiv","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/contribdiv.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Contribution Diversity Approach — contribdiv","text":"","code":"## Artificial example given in ## Table 2 in Lu et al. 2007 x <- matrix(c( 1/3,1/3,1/3,0,0,0, 0,0,1/3,1/3,1/3,0, 0,0,0,1/3,1/3,1/3), 3, 6, byrow = TRUE, dimnames = list(LETTERS[1:3],letters[1:6])) x #> a b c d e f #> A 0.3333333 0.3333333 0.3333333 0.0000000 0.0000000 0.0000000 #> B 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 0.0000000 #> C 0.0000000 0.0000000 0.0000000 0.3333333 0.3333333 0.3333333 ## Compare results with Table 2 contribdiv(x, \"richness\") #> alpha beta gamma #> A 1 1.5 2.5 #> B 1 0.5 1.5 #> C 1 1.0 2.0 contribdiv(x, \"simpson\") #> alpha beta gamma #> A 0.6666667 0.1851852 0.8518519 #> B 0.6666667 0.1111111 0.7777778 #> C 0.6666667 0.1481481 0.8148148 ## Relative contribution (C values), compare with Table 2 (cd1 <- contribdiv(x, \"richness\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.5 0.5 #> B 0 -0.5 -0.5 #> C 0 0.0 0.0 (cd2 <- contribdiv(x, \"simpson\", relative = TRUE, scaled = FALSE)) #> alpha beta gamma #> A 0 0.03703704 0.03703704 #> B 0 -0.03703704 -0.03703704 #> C 0 0.00000000 0.00000000 ## Differentiation coefficients attr(cd1, \"diff.coef\") # D_ST #> [1] 0.5 attr(cd2, \"diff.coef\") # D_DT #> [1] 0.1818182 ## BCI data set data(BCI) opar <- par(mfrow=c(2,2)) plot(contribdiv(BCI, \"richness\"), main = \"Absolute\") plot(contribdiv(BCI, \"richness\", relative = TRUE), main = \"Relative\") plot(contribdiv(BCI, \"simpson\")) plot(contribdiv(BCI, \"simpson\", relative = TRUE)) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":null,"dir":"Reference","previous_headings":"","what":"[Partial] Distance-based Redundancy Analysis — dbrda","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Distance-based redundancy analysis (dbRDA) ordination method similar Redundancy Analysis (rda), allows non-Euclidean dissimilarity indices, Manhattan Bray-Curtis distance. Despite non-Euclidean feature, analysis strictly linear metric. called Euclidean distance, results identical rda, dbRDA less efficient. Functions dbrda constrained versions metric scaling, .k.. principal coordinates analysis, based Euclidean distance can used, useful, dissimilarity measures. Function capscale simplified version based Euclidean approximation dissimilarities. functions can also perform unconstrained principal coordinates analysis, optionally using extended dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"dbrda(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...) capscale(formula, data, distance = \"euclidean\", sqrt.dist = FALSE, comm = NULL, add = FALSE, dfun = vegdist, metaMDSdist = FALSE, na.action = na.fail, subset = NULL, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"formula Model formula. function can called formula interface. usual features formula hold, especially defined cca rda. LHS must either community data matrix dissimilarity matrix, e.g., vegdist dist. LHS data matrix, function vegdist function given dfun used find dissimilarities. RHS defines constraints. constraints can continuous variables factors, can transformed within formula, can interactions typical formula. RHS can special term Condition defines variables “partialled ” constraints, just like rda cca. allows use partial dbRDA. data Data frame containing variables right hand side model formula. distance name dissimilarity (distance) index LHS formula data frame instead dissimilarity matrix. sqrt.dist Take square roots dissimilarities. See section Details . comm Community data frame used finding species scores LHS formula dissimilarity matrix. used LHS data frame. supplied, “species scores” unavailable dissimilarities supplied. N.B., available capscale: dbrda return species scores. Function sppscores can used add species scores missing. add Add constant non-diagonal dissimilarities eigenvalues non-negative underlying Principal Co-ordinates Analysis (see wcmdscale details). \"lingoes\" (TRUE) uses recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. latter one cmdscale. dfun Distance dissimilarity function used. function returning standard \"dist\" taking index name first argument can used. metaMDSdist Use metaMDSdist similarly metaMDS. means automatic data transformation using extended flexible shortest path dissimilarities (function stepacross) many dissimilarities based shared species. na.action Handling missing values constraints conditions. default (na.fail) stop missing values. Choices na.omit na.exclude delete rows missing values, differ representation results. na.omit non-missing site scores shown, na.exclude gives NA scores missing observations. Unlike rda, WA scores available missing constraints conditions. subset Subset data rows. can logical vector TRUE kept observations, logical expression can contain variables working environment, data species names community data (given formula comm argument). ... parameters passed underlying functions (e.g., metaMDSdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Functions dbrda capscale provide two alternative implementations dbRDA. Function dbrda based McArdle & Anderson (2001) directly decomposes dissimilarities. Euclidean distances results identical rda. Non-Euclidean dissimilarities may give negative eigenvalues associated imaginary axes. Function capscale based Legendre & Anderson (1999): dissimilarity data first ordinated using metric scaling, ordination results analysed rda. capscale ignores imaginary component give negative eigenvalues (report magnitude imaginary component). user supplied community data frame instead dissimilarities, functions find dissimilarities using vegdist distance function given dfun specified distance. functions accept distance objects vegdist, dist, method producing compatible objects. constraining variables can continuous factors , can interaction terms, can transformed call. Moreover, can special term Condition just like rda cca “partial” analysis can performed. Function dbrda return species scores, can also missing capscale, can added analysis using function sppscores. Non-Euclidean dissimilarities can produce negative eigenvalues (Legendre & Anderson 1999, McArdle & Anderson 2001). negative eigenvalues, printed output capscale add column sums positive eigenvalues item sum negative eigenvalues, dbrda add column giving number real dimensions positive eigenvalues. negative eigenvalues disturbing, functions let distort dissimilarities non-negative eigenvalues produced argument add = TRUE. Alternatively, sqrt.dist = TRUE, square roots dissimilarities can used may help avoiding negative eigenvalues (Legendre & Anderson 1999). functions can also used perform ordinary metric scaling .k.. principal coordinates analysis using formula constant right hand side, comm ~ 1. metaMDSdist = TRUE, function can automatic data standardization use extended dissimilarities using function stepacross similarly non-metric multidimensional scaling metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"functions return object class dbrda capscale inherit rda. See cca.object description result object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Anderson, M.J. & Willis, T.J. (2003). Canonical analysis principal coordinates: useful method constrained ordination ecology. Ecology 84, 511--525. Gower, J.C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. Legendre, P. & Legendre, L. (2012). Numerical Ecology. 3rd English Edition. Elsevier. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290--297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"Function dbrda implements real distance-based RDA preferred capscale. capscale originally developed variant constrained analysis proximities (Anderson & Willis 2003), developments made similar dbRDA. However, discards imaginary dimensions negative eigenvalues ordination significance tests area based real dimensions positive eigenvalues. capscale may removed vegan future. vegan since 2003 (CRAN release 1.6-0) dbrda first released 2016 (version 2.4-0), removal capscale may disruptive historical examples scripts, modern times dbrda used. inertia named dissimilarity index defined dissimilarity data, unknown distance information missing. largest original dissimilarity larger 4, capscale handles input similarly rda bases analysis variance instead sum squares. Keyword mean added inertia cases, e.g. Euclidean Manhattan distances. Inertia based squared index, keyword squared added name distance, unless data square root transformed (argument sqrt.dist=TRUE). additive constant used argument add, Lingoes Cailliez adjusted added name inertia, value constant printed.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dbrda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"[Partial] Distance-based Redundancy Analysis — dbrda","text":"","code":"data(varespec, varechem) ## dbrda dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\") #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\") #> #> Inertia Proportion Rank RealDims #> Total 4.5444 1.0000 #> Conditional 0.9726 0.2140 1 #> Constrained 0.9731 0.2141 3 3 #> Unconstrained 2.5987 0.5718 19 13 #> Inertia is squared Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5362 0.3198 0.1171 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9054 0.5070 0.3336 0.2581 0.2027 0.1605 0.1221 0.0825 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues with sqrt distances dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"bray\", sqrt.dist = TRUE) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"bray\", sqrt.dist = TRUE) #> #> Inertia Proportion Rank #> Total 6.9500 1.0000 #> Conditional 0.9535 0.1372 1 #> Constrained 1.2267 0.1765 3 #> Unconstrained 4.7698 0.6863 19 #> Inertia is Bray distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5817 0.4086 0.2365 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 0.9680 0.6100 0.4469 0.3837 0.3371 0.3012 0.2558 0.2010 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## avoid negative eigenvalues also with Jaccard distances (m <- dbrda(varespec ~ N + P + K + Condition(Al), varechem, dist=\"jaccard\")) #> Call: dbrda(formula = varespec ~ N + P + K + Condition(Al), data = #> varechem, distance = \"jaccard\") #> #> Inertia Proportion Rank #> Total 6.5044 1.0000 #> Conditional 1.0330 0.1588 1 #> Constrained 1.2068 0.1855 3 #> Unconstrained 4.2646 0.6557 19 #> Inertia is squared Jaccard distance #> #> Eigenvalues for constrained axes: #> dbRDA1 dbRDA2 dbRDA3 #> 0.5992 0.3994 0.2082 #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.0388 0.6441 0.4518 0.3759 0.3239 0.2785 0.2279 0.1644 #> (Showing 8 of 19 unconstrained eigenvalues) #> ## add species scores sppscores(m) <- wisconsin(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":null,"dir":"Reference","previous_headings":"","what":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Performs detrended correspondence analysis basic reciprocal averaging orthogonal correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"decorana(veg, iweigh=0, iresc=4, ira=0, mk=26, short=0, before=NULL, after=NULL) # S3 method for decorana plot(x, choices=c(1,2), origin=TRUE, display=c(\"both\",\"sites\",\"species\",\"none\"), cex = 0.8, cols = c(1,2), type, xlim, ylim, ...) # S3 method for decorana text(x, display = c(\"sites\", \"species\"), labels, choices = 1:2, origin = TRUE, select, ...) # S3 method for decorana points(x, display = c(\"sites\", \"species\"), choices=1:2, origin = TRUE, select, ...) # S3 method for decorana scores(x, display=\"sites\", choices=1:4, origin=TRUE, tidy=FALSE, ...) downweight(veg, fraction = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"veg Community data, matrix-like object. iweigh Downweighting rare species (0: ). iresc Number rescaling cycles (0: rescaling). ira Type analysis (0: detrended, 1: basic reciprocal averaging). mk Number segments rescaling. short Shortest gradient rescaled. Hill's piecewise transformation: values transformation. Hill's piecewise transformation: values transformation -- must correspond values . x decorana result object. choices Axes shown. origin Use true origin even detrended correspondence analysis. display Display sites, species, neither. cex Plot character size. cols Colours used sites species. type Type plots, partial match \"text\", \"points\" \"none\". labels Optional text used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. xlim, ylim x y limits (min,max) plot. fraction Abundance fraction downweighting begins. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score (\"sites\", \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. ... arguments plot function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"late 1970s, correspondence analysis became method choice ordination vegetation science, since seemed better able cope non-linear species responses principal components analysis. However, even correspondence analysis can produce arc-shaped configuration single gradient. Mark Hill developed detrended correspondence analysis correct two assumed ‘faults’ correspondence analysis: curvature straight gradients packing sites ends gradient. curvature removed replacing orthogonalization axes detrending. orthogonalization successive axes made non-correlated, detrending remove systematic dependence axes. Detrending performed using smoothing window mk segments. packing sites ends gradient undone rescaling axes extraction. rescaling, axis supposed scaled ‘SD’ units, average width Gaussian species responses supposed one whole axis. innovations piecewise linear transformation species abundances downweighting rare species regarded unduly high influence ordination axes. seems detrending actually works twisting ordination space, results look non-curved two-dimensional projections (‘lolly paper effect’). result, points usually easily recognized triangular diamond shaped pattern, obviously artefact detrending. Rescaling works differently commonly presented, . decorana use, even evaluate, widths species responses. Instead, tries equalize weighted standard deviation species scores axis segments (parameter mk effect, since decorana finds segments internally). Function tolerance returns internal criterion can used assess success rescaling. plot method plots species site scores. Classical decorana scaled axes smallest site score 0 ( smallest species score negative), summary, plot scores use true origin, unless origin = FALSE. addition proper eigenvalues, function reports ‘decorana values’ detrended analysis. ‘decorana values’ values legacy code decorana returns eigenvalues. estimated iteration, describe joint effects axes detrending. ‘decorana values’ estimated rescaling show effect eigenvalues. proper eigenvalues estimated extraction axes ratio weighted sum squares site species scores even detrended rescaled solutions. eigenvalues estimated axis separately, additive, higher decorana axes can show effects already explained prior axes. ‘Additive eigenvalues’ cleansed effects prior axes, can assumed add total inertia (scaled Chi-square). proportions cumulative proportions explained can use eigenvals.decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana returns object class \"decorana\", print, summary, scores, plot, points text methods, support functions eigenvals, bstick, screeplot, predict tolerance. downweight independent function can also used methods decorana.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Hill, M.O. Gauch, H.G. (1980). Detrended correspondence analysis: improved ordination technique. Vegetatio 42, 47--58. Oksanen, J. Minchin, P.R. (1997). Instability ordination results changes input data order: explanations remedies. Journal Vegetation Science 8, 447--454.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"Mark O. Hill wrote original Fortran code, R port Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"decorana uses central numerical engine original Fortran code (public domain), 1/3 original program. tried implement original behaviour, although great part preparatory steps written R language, may differ somewhat original code. However, well-known bugs corrected strict criteria used (Oksanen & Minchin 1997). Please note really need piecewise transformation even downweighting within decorana, since powerful extensive alternatives R, options included compliance original software. different fraction abundance needed downweighting, function downweight must applied decorana. Function downweight indeed can applied prior correspondence analysis, can used together cca, . Github package natto R implementation decorana allows easier inspection algorithm also easier development function. vegan 2.6-6 earlier summary method, nothing useful now defunct. former information can extracted scores weights.decorana.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/decorana.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Detrended Correspondence Analysis and Basic Reciprocal Averaging — decorana","text":"","code":"data(varespec) vare.dca <- decorana(varespec) vare.dca #> #> Call: #> decorana(veg = varespec) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.0832 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.5235 0.3253 0.20010 0.19176 #> Additive Eigenvalues 0.5235 0.3217 0.17919 0.11922 #> Decorana values 0.5249 0.1572 0.09669 0.06075 #> Axis lengths 2.8161 2.2054 1.54650 1.64864 #> plot(vare.dca) ### the detrending rationale: gaussresp <- function(x,u) exp(-(x-u)^2/2) x <- seq(0,6,length=15) ## The gradient u <- seq(-2,8,len=23) ## The optima pack <- outer(x,u,gaussresp) matplot(x, pack, type=\"l\", main=\"Species packing\") opar <- par(mfrow=c(2,2)) plot(scores(prcomp(pack)), asp=1, type=\"b\", main=\"PCA\") plot(scores(decorana(pack, ira=1)), asp=1, type=\"b\", main=\"CA\") plot(scores(decorana(pack)), asp=1, type=\"b\", main=\"DCA\") plot(scores(cca(pack ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") ### Let's add some noise: noisy <- (0.5 + runif(length(pack)))*pack par(mfrow=c(2,1)) matplot(x, pack, type=\"l\", main=\"Ideal model\") matplot(x, noisy, type=\"l\", main=\"Noisy model\") par(mfrow=c(2,2)) plot(scores(prcomp(noisy)), type=\"b\", main=\"PCA\", asp=1) plot(scores(decorana(noisy, ira=1)), type=\"b\", main=\"CA\", asp=1) plot(scores(decorana(noisy)), type=\"b\", main=\"DCA\", asp=1) plot(scores(cca(noisy ~ x), dis=\"sites\"), asp=1, type=\"b\", main=\"CCA\") par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":null,"dir":"Reference","previous_headings":"","what":"Standardization Methods for Community Ecology — decostand","title":"Standardization Methods for Community Ecology — decostand","text":"function provides popular (effective) standardization methods community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"decostand(x, method, MARGIN, range.global, logbase = 2, na.rm=FALSE, ...) wisconsin(x) decobackstand(x, zap = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Standardization Methods for Community Ecology — decostand","text":"x Community data, matrix-like object. decobackstand standardized data. method Standardization method. See Details available options. MARGIN Margin, default acceptable. 1 = rows, 2 = columns x. range.global Matrix range found method = \"range\". allows using ranges across subsets data. dimensions MARGIN must match x. logbase logarithm base used method = \"log\". na.rm Ignore missing values row column standardizations. zap Make near-zero values exact zeros avoid negative values exaggerated estimates species richness. ... arguments function (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Standardization Methods for Community Ecology — decostand","text":"function offers following standardization methods community data: total: divide margin total (default MARGIN = 1). max: divide margin maximum (default MARGIN = 2). frequency: divide margin total multiply number non-zero items, average non-zero entries one (Oksanen 1983; default MARGIN = 2). normalize: make margin sum squares equal one (default MARGIN = 1). range: standardize values range 0 ... 1 (default MARGIN = 2). values constant, transformed 0. rank, rrank: rank replaces abundance values increasing ranks leaving zeros unchanged, rrank similar uses relative ranks maximum 1 (default MARGIN = 1). Average ranks used tied values. standardize: scale x zero mean unit variance (default MARGIN = 2). pa: scale x presence/absence scale (0/1). chi.square: divide row sums square root column sums, adjust square root matrix total (Legendre & Gallagher 2001). used Euclidean distance, distances similar Chi-square distance used correspondence analysis. However, results cmdscale still differ, since CA weighted ordination method (default MARGIN = 1). hellinger: square root method = \"total\" (Legendre & Gallagher 2001). log: logarithmic transformation suggested Anderson et al. (2006): \\(\\log_b (x) + 1\\) \\(x > 0\\), \\(b\\) base logarithm; zeros left zeros. Higher bases give less weight quantities presences, logbase = Inf gives presence/absence scaling. Please note \\(\\log(x+1)\\). Anderson et al. (2006) suggested (strongly) modified Gower distance (implemented method = \"altGower\" vegdist), standardization can used independently distance indices. alr: Additive log ratio (\"alr\") transformation (Aitchison 1986) reduces data skewness compositionality bias. transformation assumes positive values, pseudocounts can added argument pseudocount. One rows/columns reference can given reference (name index). first row/column used default (reference = 1). Note transformation drops one row column transformed output data. alr transformation defined formally follows: $$alr = [log\\frac{x_1}{x_D}, ..., log\\frac{x_{D-1}}{x_D}]$$ denominator sample \\(x_D\\) can chosen arbitrarily. transformation often used pH chemistry measurenments. also commonly used multinomial logistic regression. Default MARGIN = 1 uses row reference. clr: centered log ratio (\"clr\") transformation proposed Aitchison (1986) used reduce data skewness compositionality bias. transformation frequent applications microbial ecology (see e.g. Gloor et al., 2017). clr transformation defined : $$clr = log\\frac{x}{g(x)} = log x - log g(x)$$ \\(x\\) single value, g(x) geometric mean \\(x\\). method can operate positive data; common way deal zeroes add pseudocount (e.g. smallest positive value data), either adding manually input data, using argument pseudocount decostand(x, method = \"clr\", pseudocount = 1). Adding pseudocount inevitably introduce bias; see rclr method one available solution. rclr: robust clr (\"rclr\") similar regular clr (see ) allows data contains zeroes. method use pseudocounts, unlike standard clr. robust clr (rclr) divides values geometric mean observed features; zero values kept zeroes, taken account. high dimensional data, geometric mean rclr approximates true geometric mean; see e.g. Martino et al. (2019) rclr transformation defined formally follows: $$rclr = log\\frac{x}{g(x > 0)}$$ \\(x\\) single value, \\(g(x > 0)\\) geometric mean sample-wide values \\(x\\) positive (> 0). Standardization, contrasted transformation, means entries transformed relative entries. methods default margin. MARGIN=1 means rows (sites normal data set) MARGIN=2 means columns (species normal data set). Command wisconsin shortcut common Wisconsin double standardization species (MARGIN=2) first standardized maxima (max) sites (MARGIN=1) site totals (tot). standardization methods give nonsense results negative data entries normally occur community data. empty sites species (constant method = \"range\"), many standardization change NaN. Function decobackstand can used transform standardized data back original. possible standardization may implemented cases possible. round-errors back-transformation exact, wise overwrite original data. zap=TRUE original zeros exact.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Standardization Methods for Community Ecology — decostand","text":"Returns standardized data frame, adds attribute \"decostand\" giving name applied standardization \"method\" attribute \"parameters\" appropriate transformation parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Standardization Methods for Community Ecology — decostand","text":"Jari Oksanen, Etienne Laliberté (method = \"log\"), Leo Lahti (alr, \"clr\" \"rclr\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Standardization Methods for Community Ecology — decostand","text":"Common transformations can made standard R functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Standardization Methods for Community Ecology — decostand","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Egozcue, J.J., Pawlowsky-Glahn, V., Mateu-Figueras, G., Barcel'o-Vidal, C. (2003) Isometric logratio transformations compositional data analysis. Mathematical Geology 35, 279--300. Gloor, G.B., Macklaim, J.M., Pawlowsky-Glahn, V. & Egozcue, J.J. (2017) Microbiome Datasets Compositional: Optional. Frontiers Microbiology 8, 2224. Legendre, P. & Gallagher, E.D. (2001) Ecologically meaningful transformations ordination species data. Oecologia 129, 271--280. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Oksanen, J. (1983) Ordination boreal heath-like vegetation principal component analysis, correspondence analysis multidimensional scaling. Vegetatio 52, 181--189.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/decostand.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Standardization Methods for Community Ecology — decostand","text":"","code":"data(varespec) sptrans <- decostand(varespec, \"max\") apply(sptrans, 2, max) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv Descflex Betupube #> 1 1 1 1 1 1 1 1 #> Vacculig Diphcomp Dicrsp Dicrfusc Dicrpoly Hylosple Pleuschr Polypili #> 1 1 1 1 1 1 1 1 #> Polyjuni Polycomm Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 1 1 1 1 1 1 1 1 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris Cladchlo Cladbotr #> 1 1 1 1 1 1 1 1 #> Cladamau Cladsp Cetreric Cetrisla Flavniva Nepharct Stersp Peltapht #> 1 1 1 1 1 1 1 1 #> Icmaeric Cladcerv Claddefo Cladphyl #> 1 1 1 1 sptrans <- wisconsin(varespec) # CLR transformation for rows, with pseudocount varespec.clr <- decostand(varespec, \"clr\", pseudocount=1) # ALR transformation for rows, with pseudocount and reference sample varespec.alr <- decostand(varespec, \"alr\", pseudocount=1, reference=1) ## Chi-square: PCA similar but not identical to CA. ## Use wcmdscale for weighted analysis and identical results. sptrans <- decostand(varespec, \"chi.square\") plot(procrustes(rda(sptrans), cca(varespec)))"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Design your own Dissimilarities — designdist","title":"Design your own Dissimilarities — designdist","text":"Function designdist lets define dissimilarities using terms shared total quantities, number rows number columns. shared total quantities can binary, quadratic minimum terms. binary terms, shared component number shared species, totals numbers species sites. quadratic terms cross-products sums squares, minimum terms sums parallel minima row totals. Function designdist2 similar, finds dissimilarities among two data sets. Function chaodist lets define dissimilarities using terms supposed take account “unseen species” (see Chao et al., 2005 Details vegdist).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Design your own Dissimilarities — designdist","text":"","code":"designdist(x, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) designdist2(x, y, method = \"(A+B-2*J)/(A+B)\", terms = c(\"binary\", \"quadratic\", \"minimum\"), abcd = FALSE, alphagamma = FALSE, name, maxdist) chaodist(x, method = \"1 - 2*U*V/(U+V)\", name)"},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Design your own Dissimilarities — designdist","text":"x Input data. y Another input data set: dissimilarities calculated among rows x rows y. method Equation dissimilarities. can use terms J shared quantity, B totals, N number rows (sites) P number columns (species) chaodist can use terms U V. equation can also contain R functions accepts vector arguments returns vectors length. terms shared total components found. vectors x y \"quadratic\" terms J = sum(x*y), = sum(x^2), B = sum(y^2), \"minimum\" terms J = sum(pmin(x,y)), = sum(x) B = sum(y), \"binary\" terms either transforming data binary form (shared number species, number species row). abcd Use 2x2 contingency table notation binary data: \\(\\) number shared species, \\(b\\) \\(c\\) numbers species occurring one sites , \\(d\\) number species occur neither sites. alphagamma Use beta diversity notation terms alpha average alpha diversity compared sites, gamma diversity pooled sites, delta absolute value difference average alpha alpha diversities compared sites. Terms B refer alpha diversities compared sites. name name want use index. default combine method equation terms argument. maxdist Theoretical maximum dissimilarity, NA index open absolute maximum. necessary argument, used vegan functions, certain maximum, better supply value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Design your own Dissimilarities — designdist","text":"popular dissimilarity measures ecology can expressed help terms J, B, also involve matrix dimensions N P. examples can define designdist : function designdist can implement dissimilarity indices vegdist elsewhere, can also used implement many indices, amongst , described Legendre & Legendre (2012). can also used implement indices beta diversity described Koleff et al. (2003), also specific function betadiver purpose. want implement binary dissimilarities based 2x2 contingency table notation, can set abcd = TRUE. notation = J, b = -J, c = B-J, d = P--B+J. notation often used instead tangible default notation reasons opaque . alphagamma = TRUE possible use beta diversity notation terms alpha average alpha diversity gamma gamma diversity two compared sites. terms calculated alpha = (+B)/2, gamma = +B-J delta = abs(-B)/2. Terms B also available give alpha diversities individual compared sites. beta diversity terms may make sense binary terms (diversities expressed numbers species), calculated quadratic minimum terms well ( warning). Function chaodist similar designgist, uses terms U V Chao et al. (2005). terms supposed take account effects unseen species. U V scaled range \\(0 \\dots 1\\). take place B product U*V used place J designdist. Function chaodist can implement commonly used Chao et al. (2005) style dissimilarity: Function vegdist implements Jaccard-type Chao distance, documentation contains complete discussion calculation terms.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Design your own Dissimilarities — designdist","text":"designdist returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Design your own Dissimilarities — designdist","text":"Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005) new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Koleff, P., Gaston, K.J. Lennon, J.J. (2003) Measuring beta diversity presence--absence data. J. Animal Ecol. 72, 367--382. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Design your own Dissimilarities — designdist","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Design your own Dissimilarities — designdist","text":"designdist use compiled code, based vectorized R code. designdist function can much faster vegdist, although latter uses compiled code. However, designdist skip missing values uses much memory calculations. use sum terms can numerically unstable. particularly, terms large, precision may lost. risk large number columns high, particularly large quadratic terms. precise calculations better use functions like dist vegdist robust numerical problems.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/designdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Design your own Dissimilarities — designdist","text":"","code":"data(BCI) ## Four ways of calculating the same Sørensen dissimilarity d0 <- vegdist(BCI, \"bray\", binary = TRUE) d1 <- designdist(BCI, \"(A+B-2*J)/(A+B)\") d2 <- designdist(BCI, \"(b+c)/(2*a+b+c)\", abcd = TRUE) d3 <- designdist(BCI, \"gamma/alpha - 1\", alphagamma = TRUE) ## Arrhenius dissimilarity: the value of z in the species-area model ## S = c*A^z when combining two sites of equal areas, where S is the ## number of species, A is the area, and c and z are model parameters. ## The A below is not the area (which cancels out), but number of ## species in one of the sites, as defined in designdist(). dis <- designdist(BCI, \"(log(A+B-J)-log(A+B)+log(2))/log(2)\") ## This can be used in clustering or ordination... ordiplot(cmdscale(dis)) #> species scores not available ## ... or in analysing beta diversity (without gradients) summary(dis) #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2733 0.3895 0.4192 0.4213 0.4537 0.5906"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions extract statistics resemble deviance AIC result constrained correspondence analysis cca redundancy analysis rda. functions rarely needed directly, called step automatic model building. Actually, cca rda AIC functions certainly wrong.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# S3 method for cca deviance(object, ...) # S3 method for cca extractAIC(fit, scale = 0, k = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"object result constrained ordination (cca rda). fit fitted model constrained ordination. scale optional numeric specifying scale parameter model, see scale step. k numeric specifying \"weight\" equivalent degrees freedom (=:edf) part AIC formula. ... arguments.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions find statistics resemble deviance AIC constrained ordination. Actually, constrained ordination methods log-Likelihood, means AIC deviance. Therefore use functions, use , trust . use functions, remains responsibility check adequacy result. deviance cca equal Chi-square residual data matrix fitting constraints. deviance rda defined residual sum squares. deviance function rda also used distance-based RDA dbrda. Function extractAIC mimics extractAIC.lm translating deviance AIC. little need call functions directly. However, called implicitly step function used automatic selection constraining variables. check resulting model criteria, statistics used unfounded. particular, penalty k properly defined, default k = 2 justified theoretically. continuous covariates, step function base model building magnitude eigenvalues, value k influences stopping point ( variables highest eigenvalues necessarily significant permutation tests anova.cca). also multi-class factors, value k capricious effect model building. step function pass arguments add1.cca drop1.cca, setting test = \"permutation\" provide permutation tests deletion addition can help judging validity model building.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"deviance functions return “deviance”, extractAIC returns effective degrees freedom “AIC”.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Godínez-Domínguez, E. & Freire, J. (2003) Information-theoretic approach selection spatial temporal models community organization. Marine Ecology Progress Series 253, 17--24.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"functions unfounded untested used directly implicitly. Moreover, usual caveats using step valid.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/deviance.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Statistics Resembling Deviance and AIC for Constrained Ordination — deviance.cca","text":"","code":"# The deviance of correspondence analysis equals Chi-square data(dune) data(dune.env) chisq.test(dune) #> Warning: Chi-squared approximation may be incorrect #> #> \tPearson's Chi-squared test #> #> data: dune #> X-squared = 1449, df = 551, p-value < 2.2e-16 #> deviance(cca(dune)) #> [1] 1448.956 # Stepwise selection (forward from an empty model \"dune ~ 1\") ord <- cca(dune ~ ., dune.env) step(cca(dune ~ 1, dune.env), scope = formula(ord)) #> Start: AIC=87.66 #> dune ~ 1 #> #> Df AIC #> + Moisture 3 86.608 #> + Management 3 86.935 #> + A1 1 87.411 #> 87.657 #> + Manure 4 88.832 #> + Use 2 89.134 #> #> Step: AIC=86.61 #> dune ~ Moisture #> #> Df AIC #> 86.608 #> + Management 3 86.813 #> + A1 1 86.992 #> + Use 2 87.259 #> + Manure 4 87.342 #> - Moisture 3 87.657 #> Call: cca(formula = dune ~ Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 0.6283 0.2970 3 #> Unconstrained 1.4870 0.7030 16 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 #> 0.4187 0.1330 0.0766 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> 0.4098 0.2259 0.1761 0.1234 0.1082 0.0908 0.0859 0.0609 0.0566 0.0467 0.0419 #> CA12 CA13 CA14 CA15 CA16 #> 0.0201 0.0143 0.0099 0.0085 0.0080 #>"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":null,"dir":"Reference","previous_headings":"","what":"Morisita index of intraspecific aggregation — dispindmorisita","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Calculates Morisita index dispersion, standardized index values, called clumpedness uniform indices.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"dispindmorisita(x, unique.rm = FALSE, crit = 0.05, na.rm = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"x community data matrix, sites (samples) rows species columns. unique.rm logical, TRUE, unique species (occurring one sample) removed result. crit two-sided p-value used calculate critical Chi-squared values. na.rm logical. missing values (including NaN) omitted calculations?","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita index dispersion defined (Morisita 1959, 1962): Imor = n * (sum(xi^2) - sum(xi)) / (sum(xi)^2 - sum(xi)) \\(xi\\) count individuals sample \\(\\), \\(n\\) number samples (\\(= 1, 2, \\ldots, n\\)). \\(Imor\\) values 0 \\(n\\). uniform (hyperdispersed) patterns value falls 0 1, clumped patterns falls 1 \\(n\\). increasing sample sizes (.e. joining neighbouring quadrats), \\(Imor\\) goes \\(n\\) quadrat size approaches clump size. random patterns, \\(Imor = 1\\) counts samples follow Poisson frequency distribution. deviation random expectation (null hypothesis) can tested using critical values Chi-squared distribution \\(n-1\\) degrees freedom. Confidence intervals around 1 can calculated clumped \\(Mclu\\) uniform \\(Muni\\) indices (Hairston et al. 1971, Krebs 1999) (Chi2Lower Chi2Upper refers e.g. 0.025 0.975 quantile values Chi-squared distribution \\(n-1\\) degrees freedom, respectively, crit = 0.05): Mclu = (Chi2Lower - n + sum(xi)) / (sum(xi) - 1) Muni = (Chi2Upper - n + sum(xi)) / (sum(xi) - 1) Smith-Gill (1975) proposed scaling Morisita index [0, n] interval [-1, 1], setting -0.5 0.5 values confidence limits around random distribution rescaled value 0. rescale Morisita index, one following four equations apply calculate standardized index \\(Imst\\): () Imor >= Mclu > 1: Imst = 0.5 + 0.5 (Imor - Mclu) / (n - Mclu), (b) Mclu > Imor >= 1: Imst = 0.5 (Imor - 1) / (Mclu - 1), (c) 1 > Imor > Muni: Imst = -0.5 (Imor - 1) / (Muni - 1), (d) 1 > Muni > Imor: Imst = -0.5 + 0.5 (Imor - Muni) / Muni.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Returns data frame many rows number columns input data, four columns. Columns : imor unstandardized Morisita index, mclu clumpedness index, muni uniform index, imst standardized Morisita index, pchisq Chi-squared based probability null hypothesis random expectation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Morisita, M. 1959. Measuring dispersion individuals analysis distributional patterns. Mem. Fac. Sci. Kyushu Univ. Ser. E 2, 215--235. Morisita, M. 1962. Id-index, measure dispersion individuals. Res. Popul. Ecol. 4, 1--7. Smith-Gill, S. J. 1975. Cytophysiological basis disruptive pigmentary patterns leopard frog, Rana pipiens. II. Wild type mutant cell specific patterns. J. Morphol. 146, 35--54. Hairston, N. G., Hill, R. Ritte, U. 1971. interpretation aggregation patterns. : Patil, G. P., Pileou, E. C. Waters, W. E. eds. Statistical Ecology 1: Spatial Patterns Statistical Distributions. Penn. State Univ. Press, University Park. Krebs, C. J. 1999. Ecological Methodology. 2nd ed. Benjamin Cummings Publishers.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"common error found several papers standardizing case (b), denominator given Muni - 1. results hiatus [0, 0.5] interval standardized index. root typo book Krebs (1999), see Errata book (Page 217, https://www.zoology.ubc.ca/~krebs/downloads/errors_2nd_printing.pdf).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispindmorisita.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Morisita index of intraspecific aggregation — dispindmorisita","text":"","code":"data(dune) x <- dispindmorisita(dune) x #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Chenalbu NaN Inf -Inf NaN NaN #> Cirsarve 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Empenigr 20.0000000 14.852327 -9.093483518 1.00000000 5.934709e-03 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 y <- dispindmorisita(dune, unique.rm = TRUE) y #> imor mclu muni imst pchisq #> Achimill 2.1666667 1.923488 0.327101099 0.50672636 9.157890e-03 #> Agrostol 1.8085106 1.294730 0.785245032 0.51373357 1.142619e-05 #> Airaprae 8.0000000 4.463082 -1.523370880 0.61382303 3.571702e-04 #> Alopgeni 2.5396825 1.395781 0.711614757 0.53074307 3.024441e-08 #> Anthodor 2.6666667 1.692616 0.495325824 0.52660266 5.897217e-05 #> Bellpere 2.0512821 2.154361 0.158876373 0.45535255 3.451547e-02 #> Bromhord 3.2380952 1.989452 0.279036892 0.53466422 1.170437e-04 #> Comapalu 6.6666667 5.617442 -2.364494506 0.53647558 1.055552e-02 #> Eleopalu 3.7333333 1.577180 0.579438187 0.55851854 2.958285e-10 #> Elymrepe 2.7692308 1.554093 0.596260659 0.53293787 1.180195e-06 #> Hyporadi 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Juncarti 3.1372549 1.814843 0.406265675 0.53635966 2.066336e-05 #> Juncbufo 4.1025641 2.154361 0.158876373 0.55458486 1.503205e-05 #> Lolipere 1.5849970 1.243023 0.822921342 0.50911591 5.873839e-05 #> Planlanc 2.4615385 1.554093 0.596260659 0.52459747 1.921730e-05 #> Poaprat 1.1702128 1.294730 0.785245032 0.28876015 1.046531e-01 #> Poatriv 1.4644137 1.223425 0.837201879 0.50641728 2.747301e-04 #> Ranuflam 2.4175824 2.065564 0.223578191 0.50981405 7.010483e-03 #> Rumeacet 3.9215686 1.814843 0.406265675 0.55792432 1.530085e-07 #> Sagiproc 2.4210526 1.729070 0.468764025 0.51893672 4.956394e-04 #> Salirepe 5.8181818 2.385233 -0.009348352 0.59744520 2.687397e-07 #> Scorautu 0.9643606 1.261365 0.809556915 -0.09356972 5.823404e-01 #> Trifprat 6.6666667 2.731541 -0.261685440 0.61393969 7.832274e-07 #> Trifrepe 1.2210916 1.301138 0.780576445 0.36709402 6.335449e-02 #> Vicilath 3.3333333 5.617442 -2.364494506 0.25266513 1.301890e-01 #> Bracruta 1.1904762 1.288590 0.789719093 0.33001160 8.071762e-02 #> Callcusp 5.3333333 2.539147 -0.121498169 0.58001287 7.982634e-06 dim(x) ## with unique species #> [1] 30 5 dim(y) ## unique species removed #> [1] 27 5"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":null,"dir":"Reference","previous_headings":"","what":"Dispersion-based weighting of species counts — dispweight","title":"Dispersion-based weighting of species counts — dispweight","text":"Transform abundance data downweighting species overdispersed Poisson error.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"dispweight(comm, groups, nsimul = 999, nullmodel = \"c0_ind\", plimit = 0.05) gdispweight(formula, data, plimit = 0.05) # S3 method for dispweight summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dispersion-based weighting of species counts — dispweight","text":"comm Community data matrix. groups Factor describing group structure. missing, sites regarded belonging one group. NA values allowed. nsimul Number simulations. nullmodel nullmodel used commsim within groups. default follows Clarke et al. (2006). plimit Downweight species \\(p\\)-value limit. formula, data Formula left-hand side community data frame right-hand side gives explanatory variables. explanatory variables found data frame given data parent frame. object Result object dispweight gdispweight. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dispersion-based weighting of species counts — dispweight","text":"dispersion index (\\(D\\)) calculated ratio variance expected value species. species abundances follow Poisson distribution, expected dispersion \\(E(D) = 1\\), \\(D > 1\\), species overdispersed. inverse \\(1/D\\) can used downweight species abundances. Species downweighted overdispersion judged statistically significant (Clarke et al. 2006). Function dispweight implements original procedure Clarke et al. (2006). one factor can used group sites find species means. significance overdispersion assessed freely distributing individuals species within factor levels. achieved using nullmodel \"c0_ind\" (accords Clarke et al. 2006), nullmodels can used, though may meaningful (see commsim alternatives). species absent factor level, whole level ignored calculation overdispersion, number degrees freedom can vary among species. reduced number degrees freedom used divisor overdispersion \\(D\\), species higher dispersion hence lower weights transformation. Function gdispweight generalized parametric version dispweight. function based glm quasipoisson error family. glm model can used, including several factors continuous covariates. Function gdispweight uses test statistic dispweight (Pearson Chi-square), ignore factor levels species absent, number degrees freedom equal species. Therefore transformation weights can higher dispweight. gdispweight function evaluates significance overdispersion parametrically Chi-square distribution (pchisq). Functions dispweight gdispweight transform data, add information overdispersion weights attributes result. summary can used extract print information.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dispersion-based weighting of species counts — dispweight","text":"Function returns transformed data following new attributes: D Dispersion statistic. df Degrees freedom species. p \\(p\\)-value Dispersion statistic \\(D\\). weights weights applied community data. nsimul Number simulations used assess \\(p\\)-value, NA simulations performed. nullmodel name commsim null model, NA simulations performed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dispersion-based weighting of species counts — dispweight","text":"Clarke, K. R., M. G. Chapman, P. J. Somerfield, H. R. Needham. 2006. Dispersion-based weighting species counts assemblage analyses. Marine Ecology Progress Series, 320, 11–27.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dispersion-based weighting of species counts — dispweight","text":"Eduard Szöcs eduardszoesc@gmail.com wrote original dispweight, Jari Oksanen significantly modified code, provided support functions developed gdispweight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dispweight.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dispersion-based weighting of species counts — dispweight","text":"","code":"data(mite, mite.env) ## dispweight and its summary mite.dw <- with(mite.env, dispweight(mite, Shrub, nsimul = 99)) ## IGNORE_RDIFF_BEGIN summary(mite.dw) #> Dispersion Weight Df Pr(Disp.) #> Brachy 9.6908 0.1031909 67 0.01 ** #> PHTH 3.2809 0.3047900 49 0.01 ** #> HPAV 6.5263 0.1532264 67 0.01 ** #> RARD 6.0477 0.1653525 49 0.01 ** #> SSTR 2.2619 0.4421053 49 0.01 ** #> Protopl 5.4229 0.1844031 49 0.01 ** #> MEGR 4.5354 0.2204860 67 0.01 ** #> MPRO 1.2687 0.7882353 67 0.04 * #> TVIE 2.5956 0.3852706 67 0.01 ** #> HMIN 10.0714 0.0992906 67 0.01 ** #> HMIN2 7.5674 0.1321466 49 0.01 ** #> NPRA 2.6743 0.3739344 67 0.01 ** #> TVEL 9.6295 0.1038474 49 0.01 ** #> ONOV 11.3628 0.0880064 67 0.01 ** #> SUCT 8.7372 0.1144533 67 0.01 ** #> LCIL 129.4436 0.0077254 67 0.01 ** #> Oribatl1 4.1250 0.2424248 67 0.01 ** #> Ceratoz1 1.7150 0.5830768 67 0.01 ** #> PWIL 2.2943 0.4358538 67 0.01 ** #> Galumna1 2.8777 0.3474943 49 0.01 ** #> Stgncrs2 3.8242 0.2614953 49 0.01 ** #> HRUF 1.7575 0.5690021 67 0.01 ** #> Trhypch1 14.9225 0.0670128 67 0.01 ** #> PPEL 1.3628 1.0000000 49 0.07 . #> NCOR 2.5875 0.3864771 67 0.01 ** #> SLAT 2.7857 0.3589744 49 0.01 ** #> FSET 4.8901 0.2044944 49 0.01 ** #> Lepidzts 1.6577 0.6032360 49 0.02 * #> Eupelops 1.4611 0.6844033 67 0.04 * #> Miniglmn 1.6505 0.6058733 49 0.01 ** #> LRUG 12.0658 0.0828790 67 0.01 ** #> PLAG2 3.2403 0.3086090 67 0.01 ** #> Ceratoz3 3.5125 0.2846947 67 0.01 ** #> Oppiminu 3.1680 0.3156525 67 0.01 ** #> Trimalc2 10.5927 0.0944046 67 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Based on 99 simulations on 'c0_ind' nullmodel ## IGNORE_RDIFF_END ## generalized dispersion weighting mite.dw <- gdispweight(mite ~ Shrub + WatrCont, data = mite.env) rda(mite.dw ~ Shrub + WatrCont, data = mite.env) #> Call: rda(formula = mite.dw ~ Shrub + WatrCont, data = mite.env) #> #> Inertia Proportion Rank #> Total 38.1640 1.0000 #> Constrained 9.2129 0.2414 3 #> Unconstrained 28.9511 0.7586 35 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 7.986 0.748 0.480 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 5.886 3.634 2.791 2.592 1.932 1.573 1.210 1.078 #> (Showing 8 of 35 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":null,"dir":"Reference","previous_headings":"","what":"Connectedness of Dissimilarities — distconnected","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected finds groups connected disregarding dissimilarities threshold NA. function can used find groups can ordinated together transformed stepacross. Function .shared returns logical dissimilarity object, TRUE means sites species common. minimal structure distconnected can used set missing values dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"distconnected(dis, toolong = 1, trace = TRUE) no.shared(x)"},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Connectedness of Dissimilarities — distconnected","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. trace Summarize results distconnected x Community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Connectedness of Dissimilarities — distconnected","text":"Data sets disconnected sample plots groups sample plots share species sites groups sites. data sets sensibly ordinated unconstrained method subsets related . instance, correspondence analysis polarize subsets eigenvalue 1. Neither can dissimilarities transformed stepacross, path points, result contain NAs. Function distconnected find subsets dissimilarity matrices. function return grouping vector can used sub-setting data. data connected, result vector \\(1\\)s. connectedness two points can defined either threshold toolong using input dissimilarities NAs. Function .shared returns dist structure value TRUE two sites nothing common, value FALSE least one shared species. minimal structure can analysed distconnected. function can used select dissimilarities shared species indices fixed upper limit. Function distconnected uses depth-first search (Sedgewick 1990).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Connectedness of Dissimilarities — distconnected","text":"Function distconnected returns vector observations using integers identify connected groups. data connected, values 1. Function .shared returns object class dist.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Connectedness of Dissimilarities — distconnected","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Connectedness of Dissimilarities — distconnected","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/distconnected.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Connectedness of Dissimilarities — distconnected","text":"","code":"## There are no disconnected data in vegan, and the following uses an ## extremely low threshold limit for connectedness. This is for ## illustration only, and not a recommended practice. data(dune) dis <- vegdist(dune) gr <- distconnected(dis, toolong=0.4) #> Connectivity of distance matrix with threshold dissimilarity 0.4 #> Data are disconnected: 6 groups #> Groups sizes #> 1 2 3 4 5 6 #> 1 11 2 4 1 1 # Make sites with no shared species as NA in Manhattan dissimilarities dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":null,"dir":"Reference","previous_headings":"","what":"Ecological Diversity Indices — diversity","title":"Ecological Diversity Indices — diversity","text":"Shannon, Simpson, Fisher diversity indices species richness.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ecological Diversity Indices — diversity","text":"","code":"diversity(x, index = \"shannon\", groups, equalize.groups = FALSE, MARGIN = 1, base = exp(1)) simpson.unb(x, inverse = FALSE) fisher.alpha(x, MARGIN = 1, ...) specnumber(x, groups, MARGIN = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ecological Diversity Indices — diversity","text":"x Community data, matrix-like object vector. index Diversity index, one \"shannon\", \"simpson\" \"invsimpson\". MARGIN Margin index computed. base logarithm base used shannon. inverse Use inverse Simpson similarly diversity(x, \"invsimpson\"). groups grouping factor: given, finds diversity communities pooled groups. equalize.groups Instead observed abundances, standardize communities unit total. ... Parameters passed function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ecological Diversity Indices — diversity","text":"Shannon Shannon--Weaver (Shannon--Wiener) index defined \\(H' = -\\sum_i p_i \\log_{b} p_i\\), \\(p_i\\) proportional abundance species \\(\\) \\(b\\) base logarithm. popular use natural logarithms, argue base \\(b = 2\\) (makes sense, real difference). variants Simpson's index based \\(D = \\sum p_i^2\\). Choice simpson returns \\(1-D\\) invsimpson returns \\(1/D\\). simpson.unb finds unbiased Simpson indices discrete samples (Hurlbert 1971, eq. 5). less sensitive sample size basic Simpson indices. unbiased indices can calculated data integer counts. diversity function can find total (gamma) diversity pooled communities argument groups. average alpha diversity can found mean diversities groups, difference ratio estimate beta diversity (see Examples). pooling can based either observed abundancies, communities can equalized unit total pooling; see Jost (2007) discussion. Functions adipart multipart provide canned alternatives estimating alpha, beta gamma diversities hierarchical settings. fisher.alpha estimates \\(\\alpha\\) parameter Fisher's logarithmic series (see fisherfit). estimation possible genuine counts individuals. None diversity indices usable empty sampling units without species, indices can give numeric value. Filtering cases left user. Function specnumber finds number species. MARGIN = 2, finds frequencies species. groups given, finds total number species group (see example finding one kind beta diversity option). Better stories can told Simpson's index Shannon's index, still grander narratives rarefaction (Hurlbert 1971). However, indices closely related (Hill 1973), reason despise one others (graduate student, drag , obey Professor's orders). particular, exponent Shannon index linearly related inverse Simpson (Hill 1973) although former may sensitive rare species. Moreover, inverse Simpson asymptotically equal rarefied species richness sample two individuals, Fisher's \\(\\alpha\\) similar inverse Simpson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ecological Diversity Indices — diversity","text":"vector diversity indices numbers species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Ecological Diversity Indices — diversity","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12, 42--58. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ecological Diversity Indices — diversity","text":"Jari Oksanen Bob O'Hara (fisher.alpha).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/diversity.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ecological Diversity Indices — diversity","text":"","code":"data(BCI, BCI.env) H <- diversity(BCI) simp <- diversity(BCI, \"simpson\") invsimp <- diversity(BCI, \"inv\") ## Unbiased Simpson unbias.simp <- simpson.unb(BCI) ## Fisher alpha alpha <- fisher.alpha(BCI) ## Plot all pairs(cbind(H, simp, invsimp, unbias.simp, alpha), pch=\"+\", col=\"blue\") ## Species richness (S) and Pielou's evenness (J): S <- specnumber(BCI) ## rowSums(BCI > 0) does the same... J <- H/log(S) ## beta diversity defined as gamma/alpha - 1: ## alpha is the average no. of species in a group, and gamma is the ## total number of species in the group (alpha <- with(BCI.env, tapply(specnumber(BCI), Habitat, mean))) #> OldHigh OldLow OldSlope Swamp Young #> 85.75000 91.76923 91.58333 94.00000 90.00000 (gamma <- with(BCI.env, specnumber(BCI, Habitat))) #> OldHigh OldLow OldSlope Swamp Young #> 158 210 183 128 117 gamma/alpha - 1 #> OldHigh OldLow OldSlope Swamp Young #> 0.8425656 1.2883487 0.9981802 0.3617021 0.3000000 ## similar calculations with Shannon diversity (alpha <- with(BCI.env, tapply(diversity(BCI), Habitat, mean))) # average #> OldHigh OldLow OldSlope Swamp Young #> 3.638598 3.876413 3.887122 4.003780 3.246729 (gamma <- with(BCI.env, diversity(BCI, groups=Habitat))) # pooled #> OldHigh OldLow OldSlope Swamp Young #> 3.873186 4.284972 4.212098 4.164335 3.387536 ## additive beta diversity based on Shannon index gamma-alpha #> OldHigh OldLow OldSlope Swamp Young #> 0.2345878 0.4085595 0.3249760 0.1605548 0.1408068"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and Environment in Dutch Dune Meadows. — dune","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune meadow vegetation data, dune, cover class values 30 species 20 sites. corresponding environmental data frame dune.env following entries:","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"dune data frame observations 30 species 20 sites. species names abbreviated 4+4 letters (see make.cepnames). following names changed original source (Jongman et al. 1987): Leontodon autumnalis Scorzoneroides, Potentilla palustris Comarum. dune.env data frame 20 observations following 5 variables: A1: numeric vector thickness soil A1 horizon. Moisture: ordered factor levels: 1 < 2 < 4 < 5. Management: factor levels: BF (Biological farming), HF (Hobby farming), NM (Nature Conservation Management), SF (Standard Farming). Use: ordered factor land-use levels: Hayfield < Haypastu < Pasture. Manure: ordered factor levels: 0 < 1 < 2 < 3 < 4.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"Jongman, R.H.G, ter Braak, C.J.F & van Tongeren, O.F.R. (1987). Data Analysis Community Landscape Ecology. Pudoc, Wageningen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and Environment in Dutch Dune Meadows. — dune","text":"","code":"data(dune) data(dune.env)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":null,"dir":"Reference","previous_headings":"","what":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"Classification table species dune data set.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"dune.taxon data frame 30 species (rows) classified five taxonomic levels (columns). dune.phylodis dist object estimated coalescence ages extracted doi:10.5061/dryad.63q27 (Zanne et al. 2014) using tools packages ape phylobase.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"families orders based APG IV (2016) vascular plants Hill et al. (2006) mosses. higher levels (superorder subclass) based Chase & Reveal (2009). Chase & Reveal (2009) treat Angiosperms mosses subclasses class Equisetopsida (land plants), brylogists traditionally used much inflated levels adjusted match Angiosperm classification.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"APG IV [Angiosperm Phylogeny Group] (2016) update Angiosperm Phylogeny Group classification orders families flowering plants: APG IV. Bot. J. Linnean Soc. 181: 1--20. Chase, M.W. & Reveal, J. L. (2009) phylogenetic classification land plants accompany APG III. Bot. J. Linnean Soc. 161: 122--127. Hill, M.O et al. (2006) annotated checklist mosses Europe Macaronesia. J. Bryology 28: 198--267. Zanne .E., Tank D.C., Cornwell, W.K., Eastman J.M., Smith, S.., FitzJohn, R.G., McGlinn, D.J., O’Meara, B.C., Moles, .T., Reich, P.B., Royer, D.L., Soltis, D.E., Stevens, P.F., Westoby, M., Wright, .J., Aarssen, L., Bertin, R.., Calaminus, ., Govaerts, R., Hemmings, F., Leishman, M.R., Oleksyn, J., Soltis, P.S., Swenson, N.G., Warman, L. & Beaulieu, J.M. (2014) Three keys radiation angiosperms freezing environments. Nature 506: 89--92.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/dune.taxon.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Taxonomic Classification and Phylogeny of Dune Meadow Species — dune.taxon","text":"","code":"data(dune.taxon) data(dune.phylodis)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract Eigenvalues from an Ordination Object — eigenvals","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Function extracts eigenvalues object . Many multivariate methods return objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"eigenvals(x, ...) # S3 method for cca eigenvals(x, model = c(\"all\", \"unconstrained\", \"constrained\"), constrained = NULL, ...) # S3 method for decorana eigenvals(x, kind = c(\"additive\", \"axiswise\", \"decorana\"), ...) # S3 method for eigenvals summary(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"x object extract eigenvalues. object eigenvals result object. model eigenvalues return objects inherit class \"cca\" . constrained Return constrained eigenvalues. Deprecated vegan 2.5-0. Use model instead. kind Kind eigenvalues returned decorana. \"additive\" eigenvalues can used reporting importances components summary. \"axiswise\" gives non-additive eigenvalues, \"decorana\" decorana values (see decorana details). ... arguments functions (usually ignored)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"generic function methods cca, wcmdscale, pcnm, prcomp, princomp, dudi (ade4), pca pco ( labdsv) result objects. default method also extracts eigenvalues result looks like eigen svd. Functions prcomp princomp contain square roots eigenvalues called standard deviations, eigenvals function returns squares. Function svd contains singular values, function eigenvals returns squares. constrained ordination methods cca, rda capscale function returns constrained unconstrained eigenvalues concatenated one vector, partial component ignored. However, argument constrained = TRUE constrained eigenvalues returned. summary eigenvals result returns eigenvalues, proportion explained cumulative proportion explained. result object can negative eigenvalues (wcmdscale, dbrda, pcnm) correspond imaginary axes Euclidean mapping non-Euclidean distances (Gower 1985). case real axes (corresponding positive eigenvalues) \"explain\" proportion >1 total variation, negative eigenvalues bring cumulative proportion 1. capscale find positive eigenvalues used finding proportions. decorana importances cumulative proportions reported kind = \"additive\", alternatives add total inertia input data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"object class \"eigenvals\", vector eigenvalues. summary method returns object class \"summary.eigenvals\", matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"Gower, J. C. (1985). Properties Euclidean non-Euclidean distance matrices. Linear Algebra Applications 67, 81--97.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eigenvals.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract Eigenvalues from an Ordination Object — eigenvals","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ev <- eigenvals(mod) ev #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 CA5 #> 0.3615566 0.1699600 0.1126167 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 #> CA6 CA7 CA8 CA9 CA10 CA11 CA12 CA13 #> 0.1002670 0.0772991 0.0536938 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 #> CA14 CA15 CA16 CA17 CA18 CA19 CA20 #> 0.0101910 0.0094701 0.0055090 0.0030529 0.0025118 0.0019485 0.0005178 summary(ev) #> Importance of components: #> CCA1 CCA2 CCA3 CA1 CA2 CA3 CA4 #> Eigenvalue 0.3616 0.16996 0.11262 0.3500 0.2201 0.18507 0.15512 #> Proportion Explained 0.1736 0.08159 0.05406 0.1680 0.1056 0.08884 0.07446 #> Cumulative Proportion 0.1736 0.25514 0.30920 0.4772 0.5829 0.67172 0.74618 #> CA5 CA6 CA7 CA8 CA9 CA10 CA11 #> Eigenvalue 0.13511 0.10027 0.07730 0.05369 0.03656 0.03509 0.02823 #> Proportion Explained 0.06485 0.04813 0.03711 0.02577 0.01755 0.01684 0.01355 #> Cumulative Proportion 0.81104 0.85917 0.89627 0.92205 0.93960 0.95644 0.96999 #> CA12 CA13 CA14 CA15 CA16 CA17 #> Eigenvalue 0.017065 0.012247 0.010191 0.009470 0.005509 0.003053 #> Proportion Explained 0.008192 0.005879 0.004892 0.004546 0.002644 0.001465 #> Cumulative Proportion 0.978183 0.984062 0.988954 0.993500 0.996145 0.997610 #> CA18 CA19 CA20 #> Eigenvalue 0.002512 0.0019485 0.0005178 #> Proportion Explained 0.001206 0.0009353 0.0002486 #> Cumulative Proportion 0.998816 0.9997514 1.0000000 ## choose which eignevalues to return eigenvals(mod, model = \"unconstrained\") #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3500372 0.2200788 0.1850741 0.1551179 0.1351054 0.1002670 0.0772991 0.0536938 #> CA9 CA10 CA11 CA12 CA13 CA14 CA15 CA16 #> 0.0365603 0.0350887 0.0282291 0.0170651 0.0122474 0.0101910 0.0094701 0.0055090 #> CA17 CA18 CA19 CA20 #> 0.0030529 0.0025118 0.0019485 0.0005178"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fits an Environmental Vector or Factor onto an Ordination — envfit","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"function fits environmental vectors factors onto ordination. projections points onto vectors maximum correlation corresponding environmental variables, factors show averages factor levels. continuous varaibles equal fitting linear trend surface (plane 2D) variable (see ordisurf); trend surface can presented showing gradient (direction steepest increase) using arrow. environmental variables dependent variables explained ordination scores, dependent variable analysed separately.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"# S3 method for default envfit(ord, env, permutations = 999, strata = NULL, choices=c(1,2), display = \"sites\", w = weights(ord, display), na.rm = FALSE, ...) # S3 method for formula envfit(formula, data, ...) # S3 method for envfit plot(x, choices = c(1,2), labels, arrow.mul, at = c(0,0), axis = FALSE, p.max = NULL, col = \"blue\", bg, add = TRUE, ...) # S3 method for envfit scores(x, display, choices, arrow.mul=1, tidy = FALSE, ...) vectorfit(X, P, permutations = 0, strata = NULL, w, ...) factorfit(X, P, permutations = 0, strata = NULL, w, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"ord ordination object structure ordination scores can extracted (including data frame matrix scores). env Data frame, matrix vector environmental variables. variables can mixed type (factors, continuous variables) data frames. X Matrix data frame ordination scores. P Data frame, matrix vector environmental variable(s). must continuous vectorfit factors characters factorfit. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. Set permutations = 0 skip permutations. formula, data Model formula data. na.rm Remove points missing values ordination scores environmental variables. operation casewise: whole row data removed missing value na.rm = TRUE. x result object envfit. ordiArrowMul ordiArrowTextXY must two-column matrix ( matrix-like object) containing coordinates arrow heads two plot axes, methods extract structure envfit results. choices Axes plotted. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable scores (\"vectors\" \"factors\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Change plotting labels. argument list elements vectors factors give new plotting labels. either elements omitted, default labels used. one type elements (vectors factors), labels can given vector. default labels can displayed labels command. arrow.mul Multiplier vector lengths. arrows automatically scaled similarly plot.cca given plot add = TRUE. However, scores can used adjust arrow lengths plot function used. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. axis Plot axis showing scaling fitted arrows. p.max Maximum estimated \\(P\\) value displayed variables. must calculate \\(P\\) values setting permutations use option. col Colour plotting. bg Background colour labels. bg set, labels displayed ordilabel instead text. See Examples using semitransparent background. add Results added existing ordination plot. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. display fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). scores function either \"vectors\" \"factors\" (synonyms \"bp\" \"cn\", resp.). w Weights used fitting (concerns mainly cca decorana results nonconstant weights). ... Parameters passed scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Function envfit finds vectors factor averages environmental variables. Function plot.envfit adds ordination diagram. X data.frame, envfit uses factorfit factor variables vectorfit variables. X matrix vector, envfit uses vectorfit. Alternatively, model can defined simplified model formula, left hand side must ordination result object matrix ordination scores, right hand side lists environmental variables. formula interface can used easier selection /transformation environmental variables. main effects analysed even interaction terms defined formula. ordination results extracted scores extra arguments passed scores. fitted models apply results defined extracting scores using envfit. instance, scaling constrained ordination (see scores.rda, scores.cca) must set way envfit plot ordination results (see Examples). printed output continuous variables (vectors) gives direction cosines coordinates heads unit length vectors. plot scaled correlation (square root column r2) “weak” predictors shorter arrows “strong” predictors. can see scaled relative lengths using command scores. plotted (scaled) arrows adjusted current graph using constant multiplier: keep relative r2-scaled lengths arrows tries fill current plot. can see multiplier using ordiArrowMul(result_of_envfit), set argument arrow.mul. Functions vectorfit factorfit can called directly. Function vectorfit finds directions ordination space towards environmental vectors change rapidly maximal correlations ordination configuration. Function factorfit finds averages ordination scores factor levels. Function factorfit treats ordered unordered factors similarly. permutations \\(> 0\\), significance fitted vectors factors assessed using permutation environmental variables. goodness fit statistic squared correlation coefficient (\\(r^2\\)). factors defined \\(r^2 = 1 - ss_w/ss_t\\), \\(ss_w\\) \\(ss_t\\) within-group total sums squares. See permutations additional details permutation tests Vegan. User can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. fitted vectors similar biplot arrows constrained ordination fitted LC scores (display = \"lc\") set scaling = \"species\" (see scores.cca). weighted fitting gives similar results biplot arrows class centroids cca. lengths arrows fitted vectors automatically adjusted physical size plot, arrow lengths compared across plots. similar scaling arrows, must explicitly set arrow.mul argument plot command; see ordiArrowMul ordiArrowTextXY. results can accessed scores.envfit function returns either fitted vectors scaled correlation coefficient centroids fitted environmental variables, named list .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Functions vectorfit factorfit return lists classes vectorfit factorfit print method. result object following items: arrows Arrow endpoints vectorfit. arrows scaled unit length. centroids Class centroids factorfit. r Goodness fit statistic: Squared correlation coefficient permutations Number permutations. control list control values permutations returned function . pvals Empirical P-values variable. Function envfit returns list class envfit results vectorfit envfit items. Function plot.envfit scales vectors correlation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Jari Oksanen. permutation test derives code suggested Michael Scroggie.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"Fitted vectors become method choice displaying environmental variables ordination. Indeed, optimal way presenting environmental variables Constrained Correspondence Analysis cca, since linear constraints. unconstrained ordination relation external variables ordination configuration may less linear, therefore methods arrows may useful. simplest adjust plotting symbol sizes (cex, symbols) environmental variables. Fancier methods involve smoothing regression methods abound R, ordisurf provides wrapper .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/envfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fits an Environmental Vector or Factor onto an Ordination — envfit","text":"","code":"data(varespec, varechem) library(MASS) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2276827 #> Run 2 stress 0.2204286 #> Run 3 stress 0.1967393 #> Run 4 stress 0.2066032 #> Run 5 stress 0.2234314 #> Run 6 stress 0.1869637 #> Run 7 stress 0.18584 #> Run 8 stress 0.18458 #> ... Procrustes: rmse 0.04935318 max resid 0.1575045 #> Run 9 stress 0.2178486 #> Run 10 stress 0.195049 #> Run 11 stress 0.2300388 #> Run 12 stress 0.1948413 #> Run 13 stress 0.1955836 #> Run 14 stress 0.2390981 #> Run 15 stress 0.2229031 #> Run 16 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04162605 max resid 0.1518104 #> Run 17 stress 0.2390411 #> Run 18 stress 0.1825658 #> ... Procrustes: rmse 1.358437e-05 max resid 2.641802e-05 #> ... Similar to previous best #> Run 19 stress 0.2092456 #> Run 20 stress 0.1982376 #> *** Best solution repeated 1 times (fit <- envfit(ord, varechem, perm = 999)) #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05732 -0.99836 0.2536 0.046 * #> P 0.61973 0.78481 0.1938 0.093 . #> K 0.76646 0.64229 0.1809 0.121 #> Ca 0.68520 0.72835 0.4119 0.008 ** #> Mg 0.63253 0.77454 0.4270 0.004 ** #> S 0.19140 0.98151 0.1752 0.123 #> Al -0.87159 0.49023 0.5269 0.001 *** #> Fe -0.93602 0.35195 0.4451 0.002 ** #> Mn 0.79871 -0.60172 0.5231 0.002 ** #> Zn 0.61756 0.78652 0.1879 0.107 #> Mo -0.90306 0.42951 0.0609 0.539 #> Baresoil 0.92488 -0.38025 0.2508 0.041 * #> Humdepth 0.93283 -0.36032 0.5201 0.002 ** #> pH -0.64798 0.76165 0.2308 0.070 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> scores(fit, \"vectors\") #> NMDS1 NMDS2 #> N -0.02886657 -0.5027884 #> P 0.27284738 0.3455254 #> K 0.32602389 0.2732093 #> Ca 0.43973450 0.4674260 #> Mg 0.41334179 0.5061403 #> S 0.08011287 0.4108193 #> Al -0.63269483 0.3558584 #> Fe -0.62444176 0.2347947 #> Mn 0.57765881 -0.4351900 #> Zn 0.26771371 0.3409563 #> Mo -0.22293623 0.1060315 #> Baresoil 0.46316234 -0.1904201 #> Humdepth 0.67271099 -0.2598473 #> pH -0.31130599 0.3659163 plot(ord) plot(fit) plot(fit, p.max = 0.05, col = \"red\") ## Adding fitted arrows to CCA. We use \"lc\" scores, and hope ## that arrows are scaled similarly in cca and envfit plots ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, type=\"p\") fit <- envfit(ord, varechem, perm = 999, display = \"lc\") plot(fit, p.max = 0.05, col = \"red\") ## 'scaling' must be set similarly in envfit and in ordination plot plot(ord, type = \"p\", scaling = \"sites\") fit <- envfit(ord, varechem, perm = 0, display = \"lc\", scaling = \"sites\") plot(fit, col = \"red\") ## Class variables, formula interface, and displaying the ## inter-class variability with ordispider, and semitransparent ## white background for labels (semitransparent colours are not ## supported by all graphics devices) data(dune) data(dune.env) ord <- cca(dune) fit <- envfit(ord ~ Moisture + A1, dune.env, perm = 0) plot(ord, type = \"n\") with(dune.env, ordispider(ord, Moisture, col=\"skyblue\")) with(dune.env, points(ord, display = \"sites\", col = as.numeric(Moisture), pch=16)) plot(fit, cex=1.2, axis=TRUE, bg = rgb(1, 1, 1, 0.5)) ## Use shorter labels for factor centroids labels(fit) #> $vectors #> [1] \"A1\" #> #> $factors #> [1] \"Moisture1\" \"Moisture2\" \"Moisture4\" \"Moisture5\" #> plot(ord) plot(fit, labels=list(factors = paste(\"M\", c(1,2,4,5), sep = \"\")), bg = rgb(1,1,0,0.5))"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds minimum (\\(q^*\\)) evenness profile based Tsallis entropy. scale factor entropy represents specific weighting species relative frequencies leads minimum evenness community (Mendes et al. 2008).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"eventstar(x, qmax = 5)"},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"x community matrix numeric vector. qmax Maximum scale parameter Tsallis entropy used finding minimum Tsallis based evenness range c(0, qmax).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"function eventstar finds characteristic value scale parameter \\(q\\) Tsallis entropy corresponding minimum evenness (equitability) profile based Tsallis entropy. value proposed Mendes et al. (2008) \\(q^*\\). \\(q^\\ast\\) index represents scale parameter one parameter Tsallis diversity family leads greatest deviation maximum equitability given relative abundance vector community. value \\(q^\\ast\\) found identifying minimum evenness profile scaling factor \\(q\\) one-dimensional minimization. evenness profile known convex function, guaranteed underlying optimize function find unique solution range c(0, qmax). scale parameter value \\(q^\\ast\\) used find corresponding values diversity (\\(H_{q^\\ast}\\)), evenness (\\(H_{q^\\ast}(\\max)\\)), numbers equivalent (\\(D_{q^\\ast}\\)). calculation details, see tsallis Examples . Mendes et al. (2008) advocated use \\(q^\\ast\\) corresponding diversity, evenness, Hill numbers, unique value representing diversity profile, positively associated rare species community, thus potentially useful indicator certain relative abundance distributions communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"data frame columns: qstar scale parameter value \\(q\\ast\\) corresponding minimum value Tsallis based evenness profile. Estar Value evenness based normalized Tsallis entropy \\(q^\\ast\\). Hstar Value Tsallis entropy \\(q^\\ast\\). Dstar Value Tsallis entropy \\(q^\\ast\\) converted numbers equivalents (also called Hill numbers, effective number species, ‘true’ diversity; cf. Jost 2007). See tsallis calculation details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Mendes, R.S., Evangelista, L.R., Thomaz, S.M., Agostinho, .. Gomes, L.C. (2008) unified index measure ecological diversity species rarity. Ecography 31, 450--456. Jost, L. (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439. Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Values \\(q^\\ast\\) found Mendes et al. (2008) ranged 0.56 1.12 presenting low variability, interval 0 5 safely encompass possibly expected \\(q^\\ast\\) values practice, profiling evenness changing value qmax argument advised output values near range limits found.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"Eduardo Ribeiro Cunha edurcunha@gmail.com Heloisa Beatriz Antoniazi Evangelista helobeatriz@gmail.com, technical input Péter Sólymos.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/eventstar.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Scale Parameter at the Minimum of the Tsallis Evenness Profile — eventstar","text":"","code":"data(BCI) (x <- eventstar(BCI[1:5,])) #> qstar Estar Hstar Dstar #> 1 0.6146389 0.4263687 10.524584 67.03551 #> 2 0.6249249 0.4080263 9.534034 57.66840 #> 3 0.6380858 0.4062032 9.225458 57.69174 #> 4 0.6245808 0.4062213 10.140189 65.50247 #> 5 0.6404825 0.4219957 9.828138 66.96440 ## profiling y <- as.numeric(BCI[10,]) (z <- eventstar(y)) #> qstar Estar Hstar Dstar #> 1 0.6372529 0.4117557 9.546332 61.77715 q <- seq(0, 2, 0.05) Eprof <- tsallis(y, scales=q, norm=TRUE) Hprof <- tsallis(y, scales=q) Dprof <- tsallis(y, scales=q, hill=TRUE) opar <- par(mfrow=c(3,1)) plot(q, Eprof, type=\"l\", main=\"Evenness\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, norm=TRUE), col=2) plot(q, Hprof, type=\"l\", main=\"Diversity\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar), col=2) plot(q, Dprof, type=\"l\", main=\"Effective number of species\") abline(v=z$qstar, h=tsallis(y, scales=z$qstar, hill=TRUE), col=2) par(opar)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Function fisherfit fits Fisher's logseries abundance data. Function prestonfit groups species frequencies doubling octave classes fits Preston's lognormal model, function prestondistr fits truncated lognormal model without pooling data octaves.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"fisherfit(x, ...) prestonfit(x, tiesplit = TRUE, ...) prestondistr(x, truncate = -1, ...) # S3 method for prestonfit plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", line.col = \"red\", lwd = 2, ...) # S3 method for prestonfit lines(x, line.col = \"red\", lwd = 2, ...) veiledspec(x, ...) as.fisher(x, ...) # S3 method for fisher plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", kind = c(\"bar\", \"hiplot\", \"points\", \"lines\"), add = FALSE, ...) as.preston(x, tiesplit = TRUE, ...) # S3 method for preston plot(x, xlab = \"Frequency\", ylab = \"Species\", bar.col = \"skyblue\", ...) # S3 method for preston lines(x, xadjust = 0.5, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"x Community data vector fitting functions result object plot functions. tiesplit Split frequencies \\(1, 2, 4, 8\\) etc adjacent octaves. truncate Truncation point log-Normal model, log2 units. Default value \\(-1\\) corresponds left border zero Octave. choice strongly influences fitting results. xlab, ylab Labels x y axes. bar.col Colour data bars. line.col Colour fitted line. lwd Width fitted line. kind Kind plot drawn: \"bar\" similar bar plot plot.fisherfit, \"hiplot\" draws vertical lines plot(..., type=\"h\"), \"points\" \"lines\" obvious. add Add existing plot. xadjust Adjustment horizontal positions octaves. ... parameters passed functions. Ignored prestonfit tiesplit passed .preston prestondistr.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher's logarithmic series expected number species \\(f\\) \\(n\\) observed individuals \\(f_n = \\alpha x^n / n\\) (Fisher et al. 1943). estimation possible genuine counts individuals. parameter \\(\\alpha\\) used diversity index can estimated separate function fisher.alpha. parameter \\(x\\) taken nuisance parameter estimated separately taken \\(n/(n+\\alpha)\\). Helper function .fisher transforms abundance data Fisher frequency table. Diversity given NA communities one (zero) species: reliable way estimating diversity, even equations return bogus numeric value cases. Preston (1948) satisfied Fisher's model seemed imply infinite species richness, postulated rare species diminishing class species middle frequency scale. achieved collapsing higher frequency classes wider wider “octaves” doubling class limits: 1, 2, 3--4, 5--8, 9--16 etc. occurrences. seems Preston regarded frequencies 1, 2, 4, etc.. “tied” octaves (Williamson & Gaston 2005). means half species frequency 1 shown lowest octave, rest transferred second octave. Half species second octave transferred higher one well, usually large number species. practise makes data look lognormal reducing usually high lowest octaves. can achieved setting argument tiesplit = TRUE. tiesplit = FALSE frequencies split, ones lowest octave, twos second, etc. Williamson & Gaston (2005) discuss alternative definitions detail, consulted critical review log-Normal model. logseries data look like lognormal plotted Preston's way. expected frequency \\(f\\) abundance octave \\(o\\) defined \\(f_o = S_0 \\exp(-(\\log_2(o) - \\mu)^2/2/\\sigma^2)\\), \\(\\mu\\) location mode \\(\\sigma\\) width, \\(\\log_2\\) scale, \\(S_0\\) expected number species mode. lognormal model usually truncated left rare species observed. Function prestonfit fits truncated lognormal model second degree log-polynomial octave pooled data using Poisson ( tiesplit = FALSE) quasi-Poisson (tiesplit = TRUE) error. Function prestondistr fits left-truncated Normal distribution \\(\\log_2\\) transformed non-pooled observations direct maximization log-likelihood. Function prestondistr modelled function fitdistr can used alternative distribution models. functions common print, plot lines methods. lines function adds fitted curve octave range line segments showing location mode width (sd) response. Function .preston transforms abundance data octaves. Argument tiesplit influence fit prestondistr, influence barplot octaves. total extrapolated richness fitted Preston model can found function veiledspec. function accepts results prestonfit prestondistr. veiledspec called species count vector, internally use prestonfit. Function specpool provides alternative ways estimating number unseen species. fact, Preston's lognormal model seems truncated ends, may main reason result differ lognormal models fitted Rank--Abundance diagrams functions rad.lognormal.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"function prestonfit returns object fitted coefficients, observed (freq) fitted (fitted) frequencies, string describing fitting method. Function prestondistr omits entry fitted. function fisherfit returns result nlm, item estimate \\(\\alpha\\). result object amended nuisance parameter item fisher observed data .fisher","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Fisher, R.., Corbet, .S. & Williams, C.B. (1943). relation number species number individuals random sample animal population. Journal Animal Ecology 12: 42--58. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Williamson, M. & Gaston, K.J. (2005). lognormal distribution appropriate null hypothesis species--abundance distribution. Journal Animal Ecology 74, 409--422.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"Bob O'Hara Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/fisherfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit Fisher's Logseries and Preston's Lognormal Model to Abundance Data — fisherfit","text":"","code":"data(BCI) mod <- fisherfit(BCI[5,]) mod #> #> Fisher log series model #> No. of species: 101 #> Fisher alpha: 37.96423 #> # prestonfit seems to need large samples mod.oct <- prestonfit(colSums(BCI)) mod.ll <- prestondistr(colSums(BCI)) mod.oct #> #> Preston lognormal model #> Method: Quasi-Poisson fit to octaves #> No. of species: 225 #> #> mode width S0 #> 4.885798 2.932690 32.022923 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.500000 16.00000 18.00000 19.000 30.00000 35.00000 31.00000 26.50000 #> Fitted 7.994154 13.31175 19.73342 26.042 30.59502 31.99865 29.79321 24.69491 #> 8 9 10 11 #> Observed 18.00000 13.00000 7.000000 2.0000 #> Fitted 18.22226 11.97021 7.000122 3.6443 #> mod.ll #> #> Preston lognormal model #> Method: maximized likelihood to log2 abundances #> No. of species: 225 #> #> mode width S0 #> 4.365002 2.753531 33.458185 #> #> Frequencies by Octave #> 0 1 2 3 4 5 6 7 #> Observed 9.50000 16.00000 18.00000 19.00000 30.00000 35.00000 31.00000 26.50000 #> Fitted 9.52392 15.85637 23.13724 29.58961 33.16552 32.58022 28.05054 21.16645 #> 8 9 10 11 #> Observed 18.00000 13.000000 7.000000 2.00000 #> Fitted 13.99829 8.113746 4.121808 1.83516 #> plot(mod.oct) lines(mod.ll, line.col=\"blue3\") # Different ## Smoothed density den <- density(log2(colSums(BCI))) lines(den$x, ncol(BCI)*den$y, lwd=2) # Fairly similar to mod.oct ## Extrapolated richness veiledspec(mod.oct) #> Extrapolated Observed Veiled #> 235.40577 225.00000 10.40577 veiledspec(mod.ll) #> Extrapolated Observed Veiled #> 230.931018 225.000000 5.931018"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Functions goodness inertcomp can used assess goodness fit individual sites species. Function vif.cca alias.cca can used analyse linear dependencies among constraints conditions. addition, diagnostic tools (see 'Details').","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"# S3 method for cca goodness(object, choices, display = c(\"species\", \"sites\"), model = c(\"CCA\", \"CA\"), summarize = FALSE, addprevious = FALSE, ...) inertcomp(object, display = c(\"species\", \"sites\"), unity = FALSE, proportional = FALSE) spenvcor(object) intersetcor(object) vif.cca(object) # S3 method for cca alias(object, names.only = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"object result object cca, rda, dbrda capscale. display Display \"species\" \"sites\". Species available dbrda capscale. choices Axes shown. Default show axes \"model\". model Show constrained (\"CCA\") unconstrained (\"CA\") results. summarize Show accumulated total. addprevious Add variation explained previous components statistic=\"explained\". model = \"CCA\" add conditioned (partialled ) variation, model = \"CA\" add conditioned constrained variation. give cumulative explanation previous components. unity Scale inertia components unit sum (sum items 1). proportional Give inertia components proportional corresponding total item (sum row 1). option takes precedence unity. names.Return names aliased variable(s) instead defining equations. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Function goodness gives cumulative proportion inertia accounted species chosen axes. proportions can assessed either species sites depending argument display, species available distance-based dbrda. function implemented capscale. Function inertcomp decomposes inertia partial, constrained unconstrained components site species. Legendre & De Cáceres (2012) called inertia components local contributions beta-diversity (LCBD) species contributions beta-diversity (SCBD), give relative contributions summing unity (argument unity = TRUE). interpretation, appropriate dissimilarity measures used dbrda appropriate standardization rda (Legendre & De Cáceres 2012). function implemented capscale. Function spenvcor finds -called “species -- environment correlation” (weighted) correlation weighted average scores linear combination scores. bad measure goodness ordination, sensitive extreme scores (like correlations ), sensitive overfitting using many constraints. Better models often poorer correlations. Function ordispider can show graphically. Function intersetcor finds -called “interset correlation” (weighted) correlation weighted averages scores constraints. defined contrasts used factor variables. bad measure since correlation. , focuses correlations single contrasts single axes instead looking multivariate relationship. Fitted vectors (envfit) provide better alternative. Biplot scores (see scores.cca) multivariate alternative (weighted) correlation linear combination scores constraints. Function vif.cca gives variance inflation factors constraint contrast factor constraints. partial ordination, conditioning variables analysed together constraints. Variance inflation diagnostic tool identify useless constraints. common rule values 10 indicate redundant constraints. later constraints complete linear combinations conditions previous constraints, completely removed estimation, biplot scores centroids calculated aliased constraints. note printed default output aliased constraints. Function alias give linear coefficients defining aliased constraints, names argument names.= TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"functions return matrices vectors appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London. Gross, J. (2003). Variance inflation factors. R News 3(1), 13--15. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"Jari Oksanen. vif.cca relies heavily code W. N. Venables. alias.cca simplified version alias.lm.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Diagnostic Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — goodness.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) goodness(mod, addprevious = TRUE) #> CCA1 CCA2 CCA3 CCA4 #> Achimill 0.36630013 0.3822685 0.3838616 0.4934158 #> Agrostol 0.67247051 0.6724758 0.6779597 0.7773267 #> Airaprae 0.36213737 0.3698100 0.3816619 0.3908018 #> Alopgeni 0.61547145 0.6966105 0.7042650 0.7212918 #> Anthodor 0.24619147 0.2795001 0.3509172 0.3609709 #> Bellpere 0.41185412 0.4179432 0.4847618 0.4849622 #> Bromhord 0.33487622 0.3397416 0.3870032 0.5505037 #> Chenalbu 0.23594716 0.2684323 0.2828928 0.2885321 #> Cirsarve 0.29041563 0.3013655 0.3080671 0.3591280 #> Comapalu 0.16338257 0.6836790 0.7390659 0.7963425 #> Eleopalu 0.55132024 0.6099415 0.6193301 0.6259818 #> Elymrepe 0.25239595 0.2710266 0.2761491 0.2882666 #> Empenigr 0.27089495 0.3132399 0.3153052 0.3154203 #> Hyporadi 0.31349648 0.3371809 0.3387669 0.3388716 #> Juncarti 0.43923609 0.4492937 0.4871043 0.5224072 #> Juncbufo 0.70439967 0.7226263 0.7228786 0.7257471 #> Lolipere 0.48141171 0.5720410 0.5727299 0.6034007 #> Planlanc 0.54969676 0.6084389 0.6802195 0.6826265 #> Poaprat 0.40267189 0.4944813 0.5014516 0.5326546 #> Poatriv 0.49694972 0.5409439 0.5468830 0.5594817 #> Ranuflam 0.68677962 0.6983001 0.7020461 0.7064850 #> Rumeacet 0.44788204 0.5211145 0.7673956 0.7691199 #> Sagiproc 0.27039747 0.3497634 0.3553109 0.3613746 #> Salirepe 0.64788354 0.7264891 0.7276110 0.7639711 #> Scorautu 0.54312496 0.5510319 0.6078931 0.6140593 #> Trifprat 0.37328840 0.4101104 0.6624199 0.6625703 #> Trifrepe 0.03048149 0.2115857 0.3300132 0.4207437 #> Vicilath 0.17824132 0.1784611 0.3762406 0.4279428 #> Bracruta 0.15585567 0.1641095 0.1672797 0.2449864 #> Callcusp 0.30771429 0.3143582 0.3308502 0.3518027 goodness(mod, addprevious = TRUE, summ = TRUE) #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 0.4934158 0.7773267 0.3908018 0.7212918 0.3609709 0.4849622 0.5505037 0.2885321 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 0.3591280 0.7963425 0.6259818 0.2882666 0.3154203 0.3388716 0.5224072 0.7257471 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 0.6034007 0.6826265 0.5326546 0.5594817 0.7064850 0.7691199 0.3613746 0.7639711 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 0.6140593 0.6625703 0.4207437 0.4279428 0.2449864 0.3518027 # Inertia components inertcomp(mod, prop = TRUE) #> pCCA CCA CA #> Achimill 0.34271900 0.15069678 0.5065842 #> Agrostol 0.55602406 0.22130269 0.2226733 #> Airaprae 0.06404726 0.32675457 0.6091982 #> Alopgeni 0.34238968 0.37890210 0.2787082 #> Anthodor 0.10259139 0.25837947 0.6390291 #> Bellpere 0.40972447 0.07523776 0.5150378 #> Bromhord 0.33046684 0.22003683 0.4494963 #> Chenalbu 0.11064346 0.17788865 0.7114679 #> Cirsarve 0.26649913 0.09262886 0.6408720 #> Comapalu 0.16096277 0.63537969 0.2036575 #> Eleopalu 0.53954819 0.08643366 0.3740182 #> Elymrepe 0.22234322 0.06592337 0.7117334 #> Empenigr 0.10361994 0.21180040 0.6845797 #> Hyporadi 0.03889627 0.29997533 0.6611284 #> Juncarti 0.43439190 0.08801527 0.4775928 #> Juncbufo 0.66622672 0.05952038 0.2742529 #> Lolipere 0.46273045 0.14067027 0.3965993 #> Planlanc 0.51993753 0.16268893 0.3173735 #> Poaprat 0.39408053 0.13857406 0.4673454 #> Poatriv 0.05598349 0.50349824 0.4405183 #> Ranuflam 0.68509904 0.02138594 0.2935150 #> Rumeacet 0.40125987 0.36786003 0.2308801 #> Sagiproc 0.26050435 0.10087025 0.6386254 #> Salirepe 0.12527838 0.63869277 0.2360289 #> Scorautu 0.10895437 0.50510492 0.3859407 #> Trifprat 0.34544815 0.31712212 0.3374297 #> Trifrepe 0.02132183 0.39942191 0.5792563 #> Vicilath 0.12125433 0.30668844 0.5720572 #> Bracruta 0.07222706 0.17275938 0.7550136 #> Callcusp 0.29447422 0.05732850 0.6481973 inertcomp(mod) #> pCCA CCA CA #> Achimill 0.0173766015 0.007640656 0.02568493 #> Agrostol 0.0456558521 0.018171449 0.01828399 #> Airaprae 0.0066672285 0.034014687 0.06341666 #> Alopgeni 0.0325977567 0.036073980 0.02653486 #> Anthodor 0.0096274015 0.024246897 0.05996790 #> Bellpere 0.0154640710 0.002839669 0.01943887 #> Bromhord 0.0180126793 0.011993496 0.02450059 #> Chenalbu 0.0031913088 0.005130874 0.02052099 #> Cirsarve 0.0110663060 0.003846389 0.02661204 #> Comapalu 0.0127652351 0.050389111 0.01615116 #> Eleopalu 0.0797827194 0.012780901 0.05530588 #> Elymrepe 0.0193932154 0.005749967 0.06207879 #> Empenigr 0.0063826176 0.013046147 0.04216766 #> Hyporadi 0.0046669914 0.035992710 0.07932587 #> Juncarti 0.0359126341 0.007276518 0.03948420 #> Juncbufo 0.0494087668 0.004414156 0.02033917 #> Lolipere 0.0368344271 0.011197683 0.03157023 #> Planlanc 0.0366139947 0.011456552 0.02234944 #> Poaprat 0.0142991623 0.005028142 0.01695757 #> Poatriv 0.0028845344 0.025942611 0.02269759 #> Ranuflam 0.0446783229 0.001394671 0.01914141 #> Rumeacet 0.0288221948 0.026423110 0.01658394 #> Sagiproc 0.0151161507 0.005853146 0.03705718 #> Salirepe 0.0142756439 0.072779924 0.02689581 #> Scorautu 0.0030643984 0.014206339 0.01085478 #> Trifprat 0.0228613139 0.020986733 0.02233067 #> Trifrepe 0.0008339368 0.015622139 0.02265580 #> Vicilath 0.0049088357 0.012415912 0.02315905 #> Bracruta 0.0032317812 0.007730074 0.03378289 #> Callcusp 0.0319130878 0.006212868 0.07024716 # vif.cca vif.cca(mod) #> Moisture.L Moisture.Q Moisture.C A1 ManagementHF ManagementNM #> 1.504327 1.284489 1.347660 1.367328 2.238653 2.570972 #> ManagementSF #> 2.424444 # Aliased constraints mod <- cca(dune ~ ., dune.env) mod #> Call: cca(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Constrained 1.5032 0.7106 12 #> Unconstrained 0.6121 0.2894 7 #> Inertia is scaled Chi-square #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 CCA11 #> 0.4671 0.3410 0.1761 0.1532 0.0953 0.0703 0.0589 0.0499 0.0318 0.0260 0.0228 #> CCA12 #> 0.0108 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 #> 0.27237 0.10876 0.08975 0.06305 0.03489 0.02529 0.01798 #> vif.cca(mod) #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> 2.208249 2.858927 3.072715 3.587087 6.608315 142.359372 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> 12.862713 2.642718 3.007238 80.828330 49.294455 21.433337 #> Manure^4 #> NA alias(mod) #> Model : #> dune ~ A1 + Moisture + Management + Use + Manure #> #> Complete : #> A1 Moisture.L Moisture.Q Moisture.C ManagementHF ManagementNM #> Manure^4 8.366600 #> ManagementSF Use.L Use.Q Manure.L Manure.Q Manure.C #> Manure^4 5.291503 -4.472136 2.645751 #> with(dune.env, table(Management, Manure)) #> Manure #> Management 0 1 2 3 4 #> BF 0 2 1 0 0 #> HF 0 1 2 2 0 #> NM 6 0 0 0 0 #> SF 0 0 1 2 3 # The standard correlations (not recommended) ## IGNORE_RDIFF_BEGIN spenvcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 #> 0.9636709 0.9487249 0.9330741 0.8734876 0.9373716 0.8362687 0.9748793 0.8392720 #> CCA9 CCA10 CCA11 CCA12 #> 0.8748741 0.6087512 0.6633248 0.7581210 intersetcor(mod) #> CCA1 CCA2 CCA3 CCA4 CCA5 #> A1 -0.5332506 0.13691202 -0.47996401 -0.259859587 -0.09894964 #> Moisture.L -0.8785505 0.17867589 0.03714134 0.181952935 -0.09826534 #> Moisture.Q -0.1956664 -0.33044917 -0.27321286 -0.180333890 0.26609291 #> Moisture.C -0.2023782 -0.09698397 0.28596824 -0.261712720 -0.49103002 #> ManagementHF 0.3473460 0.01680324 -0.51205769 0.194144965 0.30752664 #> ManagementNM -0.5699549 -0.61111645 0.14751127 -0.013777789 0.04571982 #> ManagementSF -0.1197499 0.64084416 0.19780650 0.134892908 -0.09679992 #> Use.L -0.1871999 0.32990444 -0.30941161 -0.372747011 0.09586963 #> Use.Q -0.1820298 -0.48874152 -0.01997442 -0.009812946 0.04812588 #> Manure.L 0.3175126 0.65945634 0.03724864 -0.025383543 -0.04077470 #> Manure.Q -0.4075615 -0.21149073 0.49297244 -0.176686201 0.11973190 #> Manure.C 0.4676279 0.11376054 0.29132473 -0.173382982 0.14219924 #> Manure^4 0.2222349 -0.12789494 -0.12921227 0.108367170 -0.02559567 #> CCA6 CCA7 CCA8 CCA9 CCA10 #> A1 -0.15225816 0.25788462 0.19247720 -0.27694466 -0.1158449480 #> Moisture.L -0.02923342 0.07858647 -0.10772510 0.07101300 0.0952517164 #> Moisture.Q -0.11211675 0.05062810 -0.48302647 0.06138704 -0.2053304965 #> Moisture.C -0.23581275 -0.38693407 -0.10144580 -0.21907160 0.1875632770 #> ManagementHF -0.24278705 0.16364055 -0.14053438 0.31066725 0.1310215145 #> ManagementNM -0.06430101 0.23917584 0.14375754 -0.27103732 0.0002768613 #> ManagementSF -0.01611984 -0.49726250 0.08073472 -0.30235728 -0.1381281272 #> Use.L 0.19127262 -0.44624831 -0.18450714 0.12950951 0.0452826749 #> Use.Q 0.13485545 0.10367354 -0.11020112 0.41245485 -0.0766932005 #> Manure.L -0.22265819 -0.49627772 -0.16971786 -0.03943343 -0.0045229147 #> Manure.Q -0.19402211 -0.11937394 0.17611673 -0.44002593 0.0903998202 #> Manure.C 0.14760330 0.07842345 0.37774417 0.10181374 0.1055057288 #> Manure^4 -0.36683782 0.05953330 0.40927409 -0.06054381 -0.1500198368 #> CCA11 CCA12 #> A1 -0.03550223 -0.08881387 #> Moisture.L 0.06404776 -0.08587882 #> Moisture.Q -0.21810558 0.16917878 #> Moisture.C 0.13701079 -0.14260914 #> ManagementHF 0.17283125 0.13296499 #> ManagementNM -0.01358436 0.09533598 #> ManagementSF -0.01468592 -0.06614834 #> Use.L -0.08584883 0.32559307 #> Use.Q 0.41893616 0.04881247 #> Manure.L 0.02396993 0.13049087 #> Manure.Q 0.12987366 0.07137031 #> Manure.C 0.05176927 -0.41550238 #> Manure^4 -0.41603287 0.01661279 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS find goodness fit measure points nonmetric multidimensional scaling, function stressplot makes Shepard diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"# S3 method for metaMDS goodness(object, dis, ...) # S3 method for default stressplot(object, dis, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"object result object metaMDS, monoMDS isoMDS. dis Dissimilarities. used metaMDS monoMDS, must used isoMDS. pch Plotting character points. Default dependent number points. p.col, l.col Point line colours. lwd Line width. monoMDS default lwd = 1 two lines drawn, lwd = 2 otherwise. ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness.metaMDS finds goodness fit statistic observations (points). defined sum squared values equal squared stress. Large values indicate poor fit. absolute values goodness statistic depend definition stress: isoMDS expresses stress percents, therefore goodness values 100 times higher monoMDS expresses stress proportion. Function stressplot draws Shepard diagram plot ordination distances monotone linear fit line original dissimilarities. addition, displays two correlation-like statistics goodness fit graph. nonmetric fit based stress \\(S\\) defined \\(R^2 = 1-S^2\\). “linear fit” squared correlation fitted values ordination distances. monoMDS, “linear fit” \\(R^2\\) “stress type 2” equal. functions can used metaMDS, monoMDS isoMDS. original dissimilarities given monoMDS metaMDS results (latter tries reconstruct dissimilarities using metaMDSredist isoMDS used engine). isoMDS dissimilarities must given. either case, functions inspect dissimilarities consistent current ordination, refuse analyse inconsistent dissimilarities. Function goodness.metaMDS generic vegan, must spell name completely isoMDS class.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Function goodness returns vector values. Function stressplot returns invisibly object items original dissimilarities, ordination distances fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/goodness.metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Goodness of Fit and Shepard Plot for Nonmetric Multidimensional Scaling — goodness.metaMDS","text":"","code":"data(varespec) mod <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.2410643 #> Run 2 stress 0.1948413 #> Run 3 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.0416189 max resid 0.1517616 #> Run 4 stress 0.2383056 #> Run 5 stress 0.2096851 #> Run 6 stress 0.2329653 #> Run 7 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 2.732084e-05 max resid 8.096006e-05 #> ... Similar to previous best #> Run 8 stress 0.1825658 #> ... Procrustes: rmse 1.167373e-05 max resid 3.019892e-05 #> ... Similar to previous best #> Run 9 stress 0.2092456 #> Run 10 stress 0.1976154 #> Run 11 stress 0.2331577 #> Run 12 stress 0.2330486 #> Run 13 stress 0.1948413 #> Run 14 stress 0.2356726 #> Run 15 stress 0.1955838 #> Run 16 stress 0.2339637 #> Run 17 stress 0.2088293 #> Run 18 stress 0.2138946 #> Run 19 stress 0.2136761 #> Run 20 stress 0.2352729 #> *** Best solution repeated 2 times stressplot(mod) gof <- goodness(mod) gof #> [1] 0.02984516 0.03513715 0.04189194 0.04598243 0.04003134 0.03441442 #> [7] 0.03294898 0.03050145 0.03060754 0.02994058 0.03526332 0.02621426 #> [13] 0.03830984 0.02980913 0.03369611 0.02225897 0.03561569 0.03505254 #> [19] 0.06577496 0.03268352 0.03503121 0.02956628 0.05168013 0.04601957 plot(mod, display = \"sites\", type = \"n\") points(mod, display = \"sites\", cex = 2*gof/mean(gof))"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":null,"dir":"Reference","previous_headings":"","what":"Indicator Power of Species — indpower","title":"Indicator Power of Species — indpower","text":"Indicator power calculation Halme et al. (2009) congruence indicator target species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indicator Power of Species — indpower","text":"","code":"indpower(x, type = 0)"},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indicator Power of Species — indpower","text":"x Community data frame matrix. type type statistic returned. See Details explanation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indicator Power of Species — indpower","text":"Halme et al. (2009) described index indicator power defined \\(IP_I = \\sqrt{\\times b}\\), \\(= S / O_I\\) \\(b = 1 - (O_T - S) / (N - O_I)\\). \\(N\\) number sites, \\(S\\) number shared occurrences indicator (\\(\\)) target (\\(T\\)) species. \\(O_I\\) \\(O_T\\) number occurrences indicator target species. type argument function call enables choose statistic return. type = 0 returns \\(IP_I\\), type = 1 returns \\(\\), type = 2 returns \\(b\\). Total indicator power (TIP) indicator species column mean (without value, see examples). Halme et al. (2009) explain calculate confidence intervals statistics, see Examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indicator Power of Species — indpower","text":"matrix indicator species rows target species columns (indicated first letters row/column names).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indicator Power of Species — indpower","text":"Halme, P., Mönkkönen, M., Kotiaho, J. S, Ylisirniö, -L. 2009. Quantifying indicator power indicator species. Conservation Biology 23: 1008--1016.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indicator Power of Species — indpower","text":"Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/indpower.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indicator Power of Species — indpower","text":"","code":"data(dune) ## IP values ip <- indpower(dune) ## and TIP values diag(ip) <- NA (TIP <- rowMeans(ip, na.rm=TRUE)) #> i.Achimill i.Agrostol i.Airaprae i.Alopgeni i.Anthodor i.Bellpere i.Bromhord #> 0.3186250 0.3342800 0.2168133 0.3416198 0.3567884 0.3432281 0.3665632 #> i.Chenalbu i.Cirsarve i.Comapalu i.Eleopalu i.Elymrepe i.Empenigr i.Hyporadi #> 0.2095044 0.2781640 0.1713273 0.2414787 0.3263516 0.2016196 0.2378197 #> i.Juncarti i.Juncbufo i.Lolipere i.Planlanc i.Poaprat i.Poatriv i.Ranuflam #> 0.2915850 0.3331330 0.3998442 0.3426064 0.4094319 0.3929520 0.2663080 #> i.Rumeacet i.Sagiproc i.Salirepe i.Scorautu i.Trifprat i.Trifrepe i.Vicilath #> 0.3484684 0.3788905 0.2898512 0.4362493 0.3145854 0.4503764 0.2605349 #> i.Bracruta i.Callcusp #> 0.4252676 0.2070766 ## p value calculation for a species ## from Halme et al. 2009 ## i is ID for the species i <- 1 fun <- function(x, i) indpower(x)[i,-i] ## 'c0' randomizes species occurrences os <- oecosimu(dune, fun, \"c0\", i=i, nsimul=99) #> Warning: nullmodel transformed 'comm' to binary data ## get z values from oecosimu output z <- os$oecosimu$z ## p-value (p <- sum(z) / sqrt(length(z))) #> [1] -1.471628 ## 'heterogeneity' measure (chi2 <- sum((z - mean(z))^2)) #> [1] 86.33867 pchisq(chi2, df=length(z)-1) #> [1] 0.9999999 ## Halme et al.'s suggested output out <- c(TIP=TIP[i], significance=p, heterogeneity=chi2, minIP=min(fun(dune, i=i)), varIP=sd(fun(dune, i=i)^2)) out #> TIP.i.Achimill significance heterogeneity minIP varIP #> 0.3186250 -1.4716280 86.3386705 0.0000000 0.2142097"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Linear Model Diagnostics for Constrained Ordination — influence.cca","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"set function extracts influence statistics linear model statistics directly constrained ordination result object cca, rda, capscale dbrda. constraints linear model functions support functions return identical results corresponding linear models (lm), can use documentation. main functions normal usage leverage values (hatvalues), standardized residuals (rstandard), studentized leave-one-residuals (rstudent), Cook's distance (cooks.distance). addition, vcov returns variance-covariance matrix coefficients, diagonal values variances coefficients. functions mainly support functions , can used directly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"# S3 method for cca hatvalues(model, ...) # S3 method for cca rstandard(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca rstudent(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca cooks.distance(model, type = c(\"response\", \"canoco\"), ...) # S3 method for cca sigma(object, type = c(\"response\", \"canoco\"), ...) # S3 method for cca vcov(object, type = \"canoco\", ...) # S3 method for cca SSD(object, type = \"canoco\", ...) # S3 method for cca qr(x, ...) # S3 method for cca df.residual(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"model, object, x constrained ordination result object. type Type statistics used extracting raw residuals residual standard deviation (sigma). Either \"response\" species data difference WA LC scores \"canoco\". ... arguments functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"vegan algorithm constrained ordination uses linear model (weighted linear model cca) find fitted values dependent community data, constrained ordination based fitted response (Legendre & Legendre 2012). hatvalues give leverage values constraints, leverage independent response data. influence statistics (rstandard, rstudent, cooks.distance) based leverage, raw residuals residual standard deviation (sigma). type = \"response\" raw residuals given unconstrained component constrained ordination, influence statistics matrix dimensions . observations times . species. cca statistics obtained lm model using Chi-square standardized species data (see decostand) dependent variable, row sums community data weights, rda lm model uses non-modified community data weights. algorithm CANOCO software constraints results iteration performing linear regression weighted averages (WA) scores constraints taking fitted values regression linear combination (LC) scores (ter Braak 1984). WA scores directly found species scores, LC scores linear combinations constraints regression. type = \"canoco\" raw residuals differences WA LC scores, residual standard deviation (sigma) taken axis sum squared WA scores minus one. quantities relationship residual component ordination, rather methodological artefacts algorithm used vegan. result matrix dimensions . observations times . constrained axes. Function vcov returns matrix variances covariances regression coefficients. diagonal values matrix variances, square roots give standard errors regression coefficients. function based SSD extracts sum squares crossproducts residuals. residuals defined similarly influence measures type similar properties limitations, define dimensions result matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Function .mlm casts ordination object multiple linear model class \"mlm\" (see lm), similar statistics can derived modified object set functions. However, problems R implementation analysis multiple linear model objects. results differ, current set functions probable correct. use .mlm objects avoided.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/influence.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Linear Model Diagnostics for Constrained Ordination — influence.cca","text":"","code":"data(varespec, varechem) mod <- cca(varespec ~ Al + P + K, varechem) ## leverage hatvalues(mod) #> 18 15 24 27 23 19 22 #> 0.06904416 0.06666628 0.15245083 0.18944882 0.09291510 0.05122338 0.15309307 #> 16 28 13 14 20 25 7 #> 0.09605909 0.27139695 0.75889765 0.04958141 0.06582891 0.10590183 0.20630888 #> 5 6 3 4 2 9 12 #> 0.19797654 0.16280522 0.22738889 0.30915530 0.15557066 0.14855598 0.09046701 #> 10 11 21 #> 0.12745850 0.10984996 0.14195559 plot(hatvalues(mod), type = \"h\") ## ordination plot with leverages: points with high leverage have ## similar LC and WA scores plot(mod, type = \"n\") ordispider(mod) # segment from LC to WA scores points(mod, dis=\"si\", cex=5*hatvalues(mod), pch=21, bg=2) # WA scores text(mod, dis=\"bp\", col=4) ## deviation and influence head(rstandard(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.4084518 0.9442480 -0.68178124 -0.798241724 0.9883838 -0.3086868 #> 15 -1.3902462 -1.5717947 -0.70784872 -0.645563228 0.2353736 -0.1679226 #> 24 0.9622453 -0.9520875 -0.08884556 -0.654099911 0.2420416 0.4832198 #> 27 -1.1080099 1.0938951 1.70146427 -0.196668562 -0.3937467 -0.7424140 #> 23 0.3979939 1.3218254 -0.63872221 -1.003315524 1.8996365 -0.4495408 #> 19 -1.5874575 0.7894087 -0.59609083 -0.006142973 0.1334143 -0.1060450 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 -0.5785258 -0.4585683 0.7640788 4.3748349 -0.39301720 -0.656213958 #> 15 -0.5416812 -0.4594716 -0.3410155 -0.2704388 -0.05769657 0.406022095 #> 24 -0.6409619 0.1908003 0.0198320 -0.2175720 3.75416938 -0.009140093 #> 27 4.2976822 -0.2704153 0.9211453 -0.2037075 -0.84410200 -0.712968237 #> 23 -0.8356637 -0.2779318 -0.0206861 -0.2899895 -0.67773316 -0.294674408 #> 19 -0.4453843 -0.3739569 -0.3174207 -0.2429740 -0.14281646 -0.819230368 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 -0.4933634 -0.6030042 -1.3812122 -0.07430854 -0.5332731 -0.7140513 #> 15 -0.3630434 -0.4055353 1.9930791 0.02126658 -0.2274864 -0.6730241 #> 24 2.2312025 -1.3907968 0.5792314 -0.45358405 -0.4628096 -0.2153846 #> 27 -0.5715008 1.6310289 0.8124329 -0.34628172 -0.8630615 0.9423113 #> 23 -0.4684368 -1.1995321 -0.8241477 -0.04953929 0.7890327 -0.6774675 #> 19 -0.3186606 -0.3948224 0.6180114 0.05035990 0.8260214 2.3626483 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang Cladstel #> 18 0.2918311 -0.42027512 -0.40191333 1.5303054 0.85056948 -0.2218833 #> 15 -0.6168627 -0.47096984 -0.42715787 0.1758993 -0.45409579 -0.3739920 #> 24 1.9274063 0.09984384 0.21066100 0.1668363 -0.08508535 -1.2835676 #> 27 -1.5729552 -0.31944598 -0.09167396 -0.1864165 0.51080623 -0.1083797 #> 23 0.4875575 0.61798970 -0.18848086 0.5574309 0.28079468 -0.4026464 #> 19 -0.1141521 -0.25344373 -0.31179839 -0.2391669 -0.65577452 0.6584127 #> Cladunci Cladcocc Cladcorn Cladgrac Cladfimb Cladcris #> 18 -0.39796095 0.93836573 -0.2564543 0.33864378 1.1572584 -0.2083305 #> 15 0.06761526 0.24340663 -0.1767166 0.27491203 1.0721632 1.9724410 #> 24 1.24902375 -0.98471253 -0.4801382 2.51184311 -1.4063518 -0.3084304 #> 27 -0.59021669 -1.25354423 -0.2460447 -1.09351514 -1.1681499 -1.0337232 #> 23 -0.34604539 -0.10730202 3.9477300 2.51924664 0.3536280 3.3882402 #> 19 -0.33866721 0.02698153 0.1776632 0.03968833 -0.7512944 -0.4763562 #> Cladchlo Cladbotr Cladamau Cladsp Cetreric Cetrisla #> 18 -0.5711604 -0.4914716 4.0852019 0.2489284 -0.4428064 -0.5834462 #> 15 -0.4347061 -0.6517740 -0.3155708 -0.2568784 0.2369559 -0.3468005 #> 24 0.6013607 0.4603779 -0.1576900 -0.5995616 2.7826114 0.3931826 #> 27 -0.5436659 -0.2788962 -0.1606997 0.1557879 -0.7298364 -0.5867612 #> 23 0.1029999 0.6494142 -0.3415411 -0.2330698 -0.6309632 -0.5301977 #> 19 0.2793712 -0.1315438 -0.2705486 -0.2865837 -0.4664078 -0.5056846 #> Flavniva Nepharct Stersp Peltapht Icmaeric Cladcerv #> 18 0.30428187 -0.3624631 -0.23665431 -0.1571633 -0.630333375 0.12058739 #> 15 0.17345018 -0.1919943 0.05469573 -0.3233311 -0.561177494 0.08518455 #> 24 -0.74154401 -0.4146848 -0.05500461 -0.7609417 0.255058737 -0.92761801 #> 27 0.11301489 -0.5064006 -0.08681568 -0.1247151 -0.001277338 -0.04038189 #> 23 0.09411988 -0.4627811 0.47668055 3.5826478 -0.274664798 -0.05517988 #> 19 0.07211309 -0.1693122 -0.17244475 -0.3155345 -0.461532920 -0.02709075 #> Claddefo Cladphyl #> 18 -0.43581630 -0.2098378 #> 15 0.94176661 -0.1028102 #> 24 -0.07508682 -1.0479632 #> 27 -1.06110299 -0.4908554 #> 23 2.66430575 -0.4244333 #> 19 -0.08427954 -0.1692474 head(cooks.distance(mod)) #> Callvulg Empenigr Rhodtome Vaccmyrt Vaccviti Pinusylv #> 18 0.003093283 0.01653142 0.0086184263 1.181427e-02 0.0181129462 0.0017667454 #> 15 0.034513793 0.04411649 0.0089472619 7.441951e-03 0.0009892926 0.0005035324 #> 24 0.041636714 0.04076229 0.0003549575 1.923947e-02 0.0026344196 0.0105001237 #> 27 0.071736260 0.06992022 0.1691597848 2.260067e-03 0.0090591037 0.0322065174 #> 23 0.004056312 0.04474315 0.0104472601 2.577825e-02 0.0924100906 0.0051750754 #> 19 0.034013281 0.00841101 0.0047958896 5.093326e-07 0.0002402422 0.0001517834 #> Descflex Betupube Vacculig Diphcomp Dicrsp Dicrfusc #> 18 0.006205594 0.003898934 1.082466e-02 0.3548634058 2.863921e-03 7.984152e-03 #> 15 0.005239584 0.003769873 2.076622e-03 0.0013060122 5.944416e-05 2.943802e-03 #> 24 0.018474359 0.001637053 1.768633e-05 0.0021286828 6.337714e-01 3.756697e-06 #> 27 1.079245024 0.004272814 4.958014e-02 0.0024247421 4.163335e-02 2.970242e-02 #> 23 0.017883042 0.001978130 1.095811e-05 0.0021534900 1.176240e-02 2.223634e-03 #> 19 0.002677405 0.001887502 1.359924e-03 0.0007968265 2.752966e-04 9.058502e-03 #> Dicrpoly Hylosple Pleuschr Polypili Polyjuni Polycomm #> 18 0.004513066 0.006741841 0.035371924 1.023801e-04 0.0052727512 0.009453588 #> 15 0.002353565 0.002936747 0.070934636 8.076156e-06 0.0009241023 0.008088546 #> 24 0.223863289 0.086982577 0.015087217 9.251676e-03 0.0096318471 0.002086096 #> 27 0.019084693 0.155444288 0.038567940 7.006653e-03 0.0435246182 0.051884812 #> 23 0.005619275 0.036846999 0.017393559 6.284604e-05 0.0159429407 0.011753178 #> 19 0.001370570 0.002104010 0.005155103 3.423056e-05 0.0092093061 0.075342921 #> Pohlnuta Ptilcili Barbhatc Cladarbu Cladrang #> 18 0.0015790689 0.0032749544 0.0029950405 0.0434204290 0.0134139742 #> 15 0.0067949512 0.0039609143 0.0032582625 0.0005525065 0.0036821731 #> 24 0.1670519112 0.0004482780 0.0019955986 0.0012516596 0.0003255476 #> 27 0.1445719869 0.0059627459 0.0004910701 0.0020305802 0.0152462776 #> 23 0.0060873754 0.0097800450 0.0009097308 0.0079572035 0.0020190921 #> 19 0.0001758783 0.0008669766 0.0013121758 0.0007720518 0.0058043470 #> Cladstel Cladunci Cladcocc Cladcorn Cladgrac Cladfimb #> 18 0.0009128242 2.936424e-03 1.632609e-02 0.0012194323 2.126298e-03 0.024831249 #> 15 0.0024976637 8.163919e-05 1.057972e-03 0.0005576530 1.349574e-03 0.020527262 #> 24 0.0740870642 7.015301e-02 4.360375e-02 0.0103666250 2.837200e-01 0.088939118 #> 27 0.0006863532 2.035516e-02 9.181862e-02 0.0035373629 6.987166e-02 0.079734929 #> 23 0.0041517010 3.066511e-03 2.948453e-04 0.3990922140 1.625248e-01 0.003202372 #> 19 0.0058511426 1.548070e-03 9.826017e-06 0.0004260291 2.126033e-05 0.007618415 #> Cladcris Cladchlo Cladbotr Cladamau Cladsp Cetreric #> 18 0.0008047178 0.0060485887 0.0044785217 0.3094317912 0.001148912 0.003635513 #> 15 0.0694732020 0.0033744318 0.0075858327 0.0017782910 0.001178323 0.001002639 #> 24 0.0042777886 0.0162620357 0.0095308986 0.0011181833 0.016164882 0.348184969 #> 27 0.0624395810 0.0172709295 0.0045450274 0.0015089732 0.001418140 0.031124511 #> 23 0.2939860714 0.0002716767 0.0107999564 0.0029872002 0.001391075 0.010194981 #> 19 0.0030627243 0.0010534342 0.0002335528 0.0009879496 0.001108529 0.002936134 #> Cetrisla Flavniva Nepharct Stersp Peltapht Icmaeric #> 18 0.006311601 1.716683e-03 0.0024359336 1.038405e-03 0.0004579733 7.366793e-03 #> 15 0.002147676 5.372282e-04 0.0006582422 5.342149e-05 0.0018668272 5.623539e-03 #> 24 0.006951741 2.472742e-02 0.0077328775 1.360514e-04 0.0260380002 2.925400e-03 #> 27 0.020117510 7.463161e-04 0.0149844191 4.404005e-04 0.0009088445 9.533741e-08 #> 23 0.007198701 2.268512e-04 0.0054844061 5.818798e-03 0.3286900860 1.931899e-03 #> 19 0.003451467 7.018956e-05 0.0003869197 4.013694e-04 0.0013438101 2.875078e-03 #> Cladcerv Claddefo Cladphyl #> 18 2.696135e-04 3.521639e-03 0.0008164040 #> 15 1.295779e-04 1.583784e-02 0.0001887477 #> 24 3.869397e-02 2.535317e-04 0.0493852158 #> 27 9.528504e-05 6.579101e-02 0.0140785723 #> 23 7.797221e-05 1.817802e-01 0.0046131462 #> 19 9.905729e-06 9.587131e-05 0.0003866237 ## Influence measures from lm y <- decostand(varespec, \"chi.square\") # needed in cca y1 <- with(y, Cladstel) # take one species for lm lmod1 <- lm(y1 ~ Al + P + K, varechem, weights = rowSums(varespec)) ## numerically identical within 2e-15 all(abs(cooks.distance(lmod1) - cooks.distance(mod)[, \"Cladstel\"]) < 1e-8) #> [1] TRUE ## t-values of regression coefficients based on type = \"canoco\" ## residuals coef(mod) #> CCA1 CCA2 CCA3 #> Al 0.007478556 -0.001883637 0.003380774 #> P -0.006491081 -0.102189737 -0.022306682 #> K -0.006755568 0.015343662 0.017067351 coef(mod)/sqrt(diag(vcov(mod, type = \"canoco\"))) #> CCA1 CCA2 CCA3 #> Al 6.5615451 -1.397643 3.313629 #> P -0.4576132 -6.092557 -1.756774 #> K -2.0862129 4.007159 5.887926"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":null,"dir":"Reference","previous_headings":"","what":"Isometric Feature Mapping Ordination — isomap","title":"Isometric Feature Mapping Ordination — isomap","text":"function performs isometric feature mapping consists three simple steps: (1) retain shortest dissimilarities among objects, (2) estimate dissimilarities shortest path distances, (3) perform metric scaling (Tenenbaum et al. 2000).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"isomap(dist, ndim=10, ...) isomapdist(dist, epsilon, k, path = \"shortest\", fragmentedOK =FALSE, ...) # S3 method for isomap summary(object, ...) # S3 method for isomap plot(x, net = TRUE, n.col = \"gray\", type = \"points\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Isometric Feature Mapping Ordination — isomap","text":"dist Dissimilarities. ndim Number axes metric scaling (argument k cmdscale). epsilon Shortest dissimilarity retained. k Number shortest dissimilarities retained point. epsilon k given, epsilon used. path Method used stepacross estimate shortest path, alternatives \"shortest\" \"extended\". fragmentedOK dissimilarity matrix fragmented. TRUE, analyse largest connected group, otherwise stop error. x, object isomap result object. net Draw net retained dissimilarities. n.col Colour drawn net segments. can also vector recycled points, colour net segment mixture joined points. type Plot observations either \"points\", \"text\" use \"none\" plot observations. \"text\" use ordilabel net = TRUE ordiplot net = FALSE, pass extra arguments functions. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Isometric Feature Mapping Ordination — isomap","text":"function isomap first calls function isomapdist dissimilarity transformation, performs metric scaling result. arguments isomap passed isomapdist. functions separate isompadist transformation easily used functions simple linear mapping cmdscale. Function isomapdist retains either dissimilarities equal shorter epsilon, epsilon given, least k shortest dissimilarities point. complete dissimilarity matrix reconstructed using stepacross using either flexible shortest paths extended dissimilarities (details, see stepacross). De'ath (1999) actually published essentially method Tenenbaum et al. (2000), De'ath's function available function xdiss non-CRAN package mvpart. differences isomap introduced k criterion, whereas De'ath used epsilon criterion. practice, De'ath also retains higher proportion dissimilarities typical isomap. plot function uses internally ordiplot, except adds text net using ordilabel. plot function passes extra arguments functions. addition, vegan3d package function rgl.isomap make dynamic 3D plots can rotated screen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Isometric Feature Mapping Ordination — isomap","text":"Function isomapdist returns dissimilarity object similar dist. Function isomap returns object class isomap plot summary methods. plot function returns invisibly object class ordiplot. Function scores can extract ordination scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Isometric Feature Mapping Ordination — isomap","text":"De'ath, G. (1999) Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecology 144, 191--199 Tenenbaum, J.B., de Silva, V. & Langford, J.C. (2000) global network framework nonlinear dimensionality reduction. Science 290, 2319--2323.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Isometric Feature Mapping Ordination — isomap","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Isometric Feature Mapping Ordination — isomap","text":"Tenenbaum et al. (2000) justify isomap tool unfolding manifold (e.g. 'Swiss Roll'). Even manifold structure, sampling must even dense dissimilarities along manifold shorter across folds. data manifold structure, results sensitive parameter values.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/isomap.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Isometric Feature Mapping Ordination — isomap","text":"","code":"## The following examples also overlay minimum spanning tree to ## the graphics in red. op <- par(mar=c(4,4,1,1)+0.2, mfrow=c(2,2)) data(BCI) dis <- vegdist(BCI) tr <- spantree(dis) pl <- ordiplot(cmdscale(dis), main=\"cmdscale\") #> species scores not available lines(tr, pl, col=\"red\") ord <- isomap(dis, k=3) ord #> #> Isometric Feature Mapping (isomap) #> #> Call: #> isomap(dist = dis, k = 3) #> #> Distance method: bray shortest isomap #> Criterion: k = 3 pl <- plot(ord, main=\"isomap k=3\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, k=5), main=\"isomap k=5\") lines(tr, pl, col=\"red\") pl <- plot(isomap(dis, epsilon=0.45), main=\"isomap epsilon=0.45\") lines(tr, pl, col=\"red\") par(op) ## colour points and web by the dominant species dom <- apply(BCI, 1, which.max) ## need nine colours, but default palette has only eight op <- palette(c(palette(\"default\"), \"sienna\")) plot(ord, pch = 16, col = dom, n.col = dom) palette(op)"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":null,"dir":"Reference","previous_headings":"","what":"Kendall coefficient of concordance — kendall.global","title":"Kendall coefficient of concordance — kendall.global","text":"Function kendall.global computes tests coefficient concordance among several judges (variables, species) permutation test. Function kendall.post carries posteriori tests contributions individual judges (variables, species) overall concordance group permutation tests. several groups judges identified data table, coefficients concordance (kendall.global) posteriori tests (kendall.post) computed group separately. Use ecology: identify significant species associations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"kendall.global(Y, group, nperm = 999, mult = \"holm\") kendall.post(Y, group, nperm = 999, mult = \"holm\")"},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kendall coefficient of concordance — kendall.global","text":"Y Data file (data frame matrix) containing quantitative semiquantitative data. Rows objects columns judges (variables). community ecology, table often site--species table. group vector defining judges divided groups. See example . groups explicitly defined, judges data file considered forming single group. nperm Number permutations performed. Default 999. mult Correct P-values multiple testing using alternatives described p.adjust addition \"sidak\" (see Details). Bonferroni correction overly conservative; recommended. included allow comparisons methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Kendall coefficient of concordance — kendall.global","text":"Y must contain quantitative data. transformed ranks within column computation coefficient concordance. search species associations described Legendre (2005) proceeds 3 steps: (1) Correlation analysis species. possible method compute Ward's agglomerative clustering matrix correlations among species. detail: (1.1) compute Pearson Spearman correlation matrix (correl.matrix) among species; (1.2) turn distance matrix: mat.D = .dist(1-correl.matrix); (1.3) carry Ward's hierarchical clustering matrix using hclust: clust.ward = hclust(mat.D, \"ward\"); (1.4) plot dendrogram: plot(clust.ward, hang=-1); (1.5) cut dendrogram two groups, retrieve vector species membership: group.2 = cutree(clust.ward, k=2). (1.6) steps 2 3 , may come back try divisions species k = \\(3, 4, 5, \\dots\\) groups. (2) Compute global tests significance 2 () groups using function kendall.global vector defining groups. Groups globally significant must refined abandoned. (3) Compute posteriori tests contribution individual species concordance group using function kendall.post vector defining groups. species negative values \"Spearman.mean\", means species clearly belong group, hence group inclusive. Go back (1.5) cut dendrogram finely. left right groups can cut separately, independently levels along dendrogram; write vector group membership cutree produce desired groups. corrections used multiple testing applied list P-values (P); take account number tests (k) carried simultaneously (number groups kendall.global, number species kendall.post). corrections performed using function p.adjust; see function description correction methods. addition, Šidák correction defined \\(P_{corr} = 1 -(1 - P)^k\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Kendall coefficient of concordance — kendall.global","text":"table containing following information rows. columns correspond groups \"judges\" defined vector \"group\". function Kendall.post used, many tables number predefined groups. W Kendall's coefficient concordance, W. F F statistic. F = W*(m-1)/(1-W) m number judges. Prob.F Probability associated F statistic, computed F distribution nu1 = n-1-(2/m) nu2 = nu1*(m-1); n number objects. Corrected prob.F Probabilities associated F, corrected using method selected parameter mult. Shown one group. Chi2 Friedman's chi-square statistic (Friedman 1937) used permutation test W. Prob.perm Permutational probabilities, uncorrected. Corrected prob.perm Permutational probabilities corrected using method selected parameter mult. Shown one group. Spearman.mean Mean Spearman correlations judge test judges group. W.per.species Contribution judge test overall concordance statistic group.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Kendall coefficient of concordance — kendall.global","text":"Friedman, M. 1937. use ranks avoid assumption normality implicit analysis variance. Journal American Statistical Association 32: 675-701. Kendall, M. G. B. Babington Smith. 1939. problem m rankings. Annals Mathematical Statistics 10: 275-287. Legendre, P. 2005. Species associations: Kendall coefficient concordance revisited. Journal Agricultural, Biological, Environmental Statistics 10: 226-245. Legendre, P. 2009. Coefficient concordance. : Encyclopedia Research Design. SAGE Publications (press). Siegel, S. N. J. Castellan, Jr. 1988. Nonparametric statistics behavioral sciences. 2nd edition. McGraw-Hill, New York.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Kendall coefficient of concordance — kendall.global","text":"F. Guillaume Blanchet, University Alberta, Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/kendall.global.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kendall coefficient of concordance — kendall.global","text":"","code":"data(mite) mite.hel <- decostand(mite, \"hel\") # Reproduce the results shown in Table 2 of Legendre (2005), a single group mite.small <- mite.hel[c(4,9,14,22,31,34,45,53,61,69),c(13:15,23)] kendall.global(mite.small, nperm=49) #> $Concordance_analysis #> Group.1 #> W 0.44160305 #> F 2.37252221 #> Prob.F 0.04403791 #> Chi2 15.89770992 #> Prob.perm 0.12000000 #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.small, mult=\"holm\", nperm=49) #> $A_posteriori_tests #> TVEL ONOV SUCT Trhypch1 #> Spearman.mean 0.3265678 0.3965503 0.4570402 -0.1681251 #> W.per.species 0.4949258 0.5474127 0.5927802 0.1239061 #> Prob 0.0800000 0.0200000 0.0200000 0.7200000 #> Corrected prob 0.1600000 0.0800000 0.0800000 0.7200000 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # Reproduce the results shown in Tables 3 and 4 of Legendre (2005), 2 groups group <-c(1,1,2,1,1,1,1,1,2,1,1,1,1,1,1,2,1,2,1,1,1,1,2,1,2,1,1,1,1,1,2,2,2,2,2) kendall.global(mite.hel, group=group, nperm=49) #> $Concordance_analysis #> Group.1 Group.2 #> W 3.097870e-01 2.911888e-01 #> F 1.032305e+01 4.108130e+00 #> Prob.F 1.177138e-85 4.676566e-22 #> Corrected prob.F 2.354275e-85 4.676566e-22 #> Chi2 5.130073e+02 2.210123e+02 #> Prob.perm 2.000000e-02 2.000000e-02 #> Corrected prob.perm 4.000000e-02 4.000000e-02 #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.global\" kendall.post(mite.hel, group=group, mult=\"holm\", nperm=49) #> $A_posteriori_tests_Group #> $A_posteriori_tests_Group[[1]] #> Brachy PHTH RARD SSTR Protopl MEGR #> Spearman.mean 0.1851177 0.4258111 0.359058 0.2505486 0.1802160 0.2833298 #> W.per.species 0.2190711 0.4497357 0.385764 0.2817757 0.2143736 0.3131911 #> Prob 0.0200000 0.0200000 0.020000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 #> MPRO HMIN HMIN2 NPRA TVEL ONOV #> Spearman.mean 0.09248024 0.2444656 0.4138494 0.1263751 0.4177343 0.3301159 #> W.per.species 0.13029357 0.2759462 0.4382723 0.1627761 0.4419954 0.3580278 #> Prob 0.14000000 0.0200000 0.0200000 0.0400000 0.0200000 0.0200000 #> Corrected prob 0.70000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> SUCT Oribatl1 PWIL Galumna1 Stgncrs2 HRUF #> Spearman.mean 0.2185421 0.421216 0.2574779 0.4180699 0.3623428 0.1250230 #> W.per.species 0.2511028 0.445332 0.2884163 0.4423170 0.3889118 0.1614804 #> Prob 0.0200000 0.020000 0.0200000 0.0200000 0.0200000 0.0600000 #> Corrected prob 0.7000000 0.700000 0.7000000 0.7000000 0.7000000 0.7000000 #> PPEL SLAT FSET Lepidzts Eupelops Miniglmn #> Spearman.mean 0.2188216 0.3016159 0.4217606 0.2577037 0.1108022 0.2301430 #> W.per.species 0.2513707 0.3307153 0.4458539 0.2886327 0.1478521 0.2622203 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0600000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> $A_posteriori_tests_Group[[2]] #> HPAV TVIE LCIL Ceratoz1 Trhypch1 NCOR #> Spearman.mean 0.1222579 0.2712078 0.1906408 0.1375601 0.1342409 0.3342345 #> W.per.species 0.2020527 0.3374616 0.2642189 0.2159637 0.2129463 0.3947586 #> Prob 0.0400000 0.0200000 0.0200000 0.0200000 0.0400000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> LRUG PLAG2 Ceratoz3 Oppiminu Trimalc2 #> Spearman.mean 0.3446561 0.1833099 0.3188922 0.1764232 0.2498877 #> W.per.species 0.4042328 0.2575544 0.3808111 0.2512938 0.3180797 #> Prob 0.0200000 0.0200000 0.0200000 0.0200000 0.0200000 #> Corrected prob 0.7000000 0.7000000 0.7000000 0.7000000 0.7000000 #> #> #> $Correction.type #> [1] \"holm\" #> #> attr(,\"class\") #> [1] \"kendall.post\" # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Function linestack plots vertical one-dimensional plots numeric vectors. plots always labelled, labels moved vertically avoid overwriting.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"linestack(x, labels, cex = 0.8, side = \"right\", hoff = 2, air = 1.1, at = 0, add = FALSE, axis = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"x Numeric vector plotted. labels Labels used instead default (names x). May expressions drawn plotmath. cex Size labels. side Put labels \"right\" \"left\" axis. hoff Distance vertical axis label units width letter “m”. air Multiplier string height leave empty space labels. Position plot horizontal axis. add Add existing plot. axis Add axis plot. ... graphical parameters labels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function returns invisibly shifted positions labels user coordinates.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"Jari Oksanen modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"function always draws labelled diagrams. want unlabelled diagrams, can use, e.g., plot, stripchart rug.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/linestack.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots One-dimensional Diagrams without Overwriting Labels — linestack","text":"","code":"## First DCA axis data(dune) ord <- decorana(dune) linestack(scores(ord, choices=1, display=\"sp\")) linestack(scores(ord, choices=1, display=\"si\"), side=\"left\", add=TRUE) title(main=\"DCA axis 1\") ## Expressions as labels N <- 10 # Number of sites df <- data.frame(Ca = rlnorm(N, 2), NO3 = rlnorm(N, 4), SO4 = rlnorm(N, 10), K = rlnorm(N, 3)) ord <- rda(df, scale = TRUE) ### vector of expressions for labels labs <- expression(Ca^{2+phantom()}, NO[3]^{-phantom()}, SO[4]^{2-phantom()}, K^{+phantom()}) scl <- \"sites\" linestack(scores(ord, choices = 1, display = \"species\", scaling = scl), labels = labs, air = 2) linestack(scores(ord, choices = 1, display = \"site\", scaling = scl), side = \"left\", add = TRUE) title(main = \"PCA axis 1\")"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":null,"dir":"Reference","previous_headings":"","what":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"standard CEP name four first letters generic name four first letters specific epithet Latin name. last epithet, may subspecific name, used current function. name one component, abbreviated eight characters (see abbreviate). returned names made unique function make.unique adds numbers end CEP names needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(names, seconditem = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"names names formatted CEP names. seconditem Take always second item original name abbreviated name instead last original item (default).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Cornell Ecology Programs (CEP) used eight-letter abbreviations species site names. species, names formed taking four first letters generic name four first letters specific subspecific epithet. current function first makes valid R names using make.names, splits elements. CEP name made taking four first letters first element, four first letters last (default) second element ( seconditem = TRUE). one name element, abbreviated eight letters. Finally, names made unique may add numbers duplicated names. CEP names originally used, old FORTRAN IV CHARACTER data type, text stored numerical variables, popular computers hold four characters. modern times, reason limitation, ecologists used names, may practical avoid congestion ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Function returns CEP names.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"function simpleminded rigid. must write better one need.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/make.cepnames.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Abbreviates a Botanical or Zoological Latin Name into an Eight-character Name — make.cepnames","text":"","code":"make.cepnames(c(\"Aa maderoi\", \"Poa sp.\", \"Cladina rangiferina\", \"Cladonia cornuta\", \"Cladonia cornuta var. groenlandica\", \"Cladonia rangiformis\", \"Bryoerythrophyllum\")) #> [1] \"Aamade\" \"Poasp\" \"Cladrang\" \"Cladcorn\" \"Cladgroe\" #> [6] \"Cladrang.1\" \"Bryrythr\" data(BCI) colnames(BCI) <- make.cepnames(colnames(BCI))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel Correlogram — mantel.correlog","title":"Mantel Correlogram — mantel.correlog","text":"Function mantel.correlog computes multivariate Mantel correlogram. Proposed Sokal (1986) Oden Sokal (1986), method also described Legendre Legendre (2012, pp. 819--821).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel Correlogram — mantel.correlog","text":"","code":"mantel.correlog(D.eco, D.geo=NULL, XY=NULL, n.class=0, break.pts=NULL, cutoff=TRUE, r.type=\"pearson\", nperm=999, mult=\"holm\", progressive=TRUE) # S3 method for mantel.correlog plot(x, alpha=0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel Correlogram — mantel.correlog","text":"D.eco ecological distance matrix, class either dist matrix. D.geo geographic distance matrix, class either dist matrix. Provide either D.geo XY. Default: D.geo=NULL. XY file Cartesian geographic coordinates points. Default: XY=NULL. n.class Number classes. n.class=0, Sturges equation used unless break points provided. break.pts Vector containing break points distance distribution. Provide (n.class+1) breakpoints, , list beginning ending point. Default: break.pts=NULL. cutoff second half distance classes, cutoff = TRUE limits correlogram distance classes include points. cutoff = FALSE, correlogram includes distance classes. r.type Type correlation calculation Mantel statistic. Default: r.type=\"pearson\". choices r.type=\"spearman\" r.type=\"kendall\", functions cor mantel. nperm Number permutations tests significance. Default: nperm=999. large data files, permutation tests rather slow. mult Correct P-values multiple testing. correction methods \"holm\" (default), \"hochberg\", \"sidak\", methods available p.adjust function: \"bonferroni\" (best known, recommended overly conservative), \"hommel\", \"BH\", \"\", \"fdr\", \"none\". progressive Default: progressive=TRUE progressive correction multiple-testing, described Legendre Legendre (1998, p. 721). Test first distance class: correction; second distance class: correct 2 simultaneous tests; distance class k: correct k simultaneous tests. progressive=FALSE: correct tests n.class simultaneous tests. x Output mantel.correlog. alpha Significance level points drawn black symbols correlogram. Default: alpha=0.05. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel Correlogram — mantel.correlog","text":"correlogram graph spatial correlation values plotted, ordinate, function geographic distance classes among study sites along abscissa. Mantel correlogram, Mantel correlation (Mantel 1967) computed multivariate (e.g. multi-species) distance matrix user's choice design matrix representing geographic distance classes turn. Mantel statistic tested permutational Mantel test performed vegan's mantel function. correction multiple testing applied, permutations necessary -correction case, obtain significant p-values higher correlogram classes. print.mantel.correlog function prints correlogram. See examples.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel Correlogram — mantel.correlog","text":"mantel.res table distance classes rows class indices, number distances per class, Mantel statistics (computed using Pearson's r, Spearman's r, Kendall's tau), p-values columns. positive Mantel statistic indicates positive spatial correlation. additional column p-values corrected multiple testing added unless mult=\"none\". n.class n umber distance classes. break.pts break points provided user computed program. mult name correction multiple testing. correction: mult=\"none\". progressive logical (TRUE, FALSE) value indicating whether progressive correction multiple testing requested. n.tests number distance classes Mantel tests computed tested significance. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel Correlogram — mantel.correlog","text":"Pierre Legendre, Université de Montréal","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel Correlogram — mantel.correlog","text":"Legendre, P. L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. Mantel, N. 1967. detection disease clustering generalized regression approach. Cancer Res. 27: 209-220. Oden, N. L. R. R. Sokal. 1986. Directional autocorrelation: extension spatial correlograms two dimensions. Syst. Zool. 35: 608-617. Sokal, R. R. 1986. Spatial data analysis historical processes. 29-43 : E. Diday et al. [eds.] Data analysis informatics, IV. North-Holland, Amsterdam. Sturges, H. . 1926. choice class interval. Journal American Statistical Association 21: 65–66.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.correlog.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel Correlogram — mantel.correlog","text":"","code":"# Mite data available in \"vegan\" data(mite) data(mite.xy) mite.hel <- decostand(mite, \"hellinger\") # Detrend the species data by regression on the site coordinates mite.hel.resid <- resid(lm(as.matrix(mite.hel) ~ ., data=mite.xy)) # Compute the detrended species distance matrix mite.hel.D <- dist(mite.hel.resid) # Compute Mantel correlogram with cutoff, Pearson statistic mite.correlog <- mantel.correlog(mite.hel.D, XY=mite.xy, nperm=49) summary(mite.correlog) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 4 -none- call mite.correlog #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.135713 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.118174 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.037820 0.04 0.06 . #> D.cl.4 2.699274 696.000000 -0.098605 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.112682 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.107603 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.022264 0.12 0.14 #> D.cl.8 5.612730 326.000000 NA NA NA #> D.cl.9 6.341094 260.000000 NA NA NA #> D.cl.10 7.069458 184.000000 NA NA NA #> D.cl.11 7.797822 130.000000 NA NA NA #> D.cl.12 8.526186 66.000000 NA NA NA #> D.cl.13 9.254550 32.000000 NA NA NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 # or: print(mite.correlog) # or: print.mantel.correlog(mite.correlog) plot(mite.correlog) # Compute Mantel correlogram without cutoff, Spearman statistic mite.correlog2 <- mantel.correlog(mite.hel.D, XY=mite.xy, cutoff=FALSE, r.type=\"spearman\", nperm=49) summary(mite.correlog2) #> Length Class Mode #> mantel.res 65 -none- numeric #> n.class 1 -none- numeric #> break.pts 14 -none- numeric #> mult 1 -none- character #> n.tests 1 -none- numeric #> call 6 -none- call mite.correlog2 #> #> Mantel Correlogram Analysis #> #> Call: #> #> mantel.correlog(D.eco = mite.hel.D, XY = mite.xy, cutoff = FALSE, r.type = \"spearman\", nperm = 49) #> #> class.index n.dist Mantel.cor Pr(Mantel) Pr(corrected) #> D.cl.1 0.514182 358.000000 0.134229 0.02 0.02 * #> D.cl.2 1.242546 650.000000 0.121270 0.02 0.04 * #> D.cl.3 1.970910 796.000000 0.035413 0.02 0.06 . #> D.cl.4 2.699274 696.000000 -0.095899 0.02 0.08 . #> D.cl.5 3.427638 500.000000 -0.118692 0.02 0.10 . #> D.cl.6 4.156002 468.000000 -0.117148 0.02 0.12 #> D.cl.7 4.884366 364.000000 -0.031123 0.08 0.14 #> D.cl.8 5.612730 326.000000 0.026064 0.14 0.16 #> D.cl.9 6.341094 260.000000 0.050573 0.08 0.24 #> D.cl.10 7.069458 184.000000 0.057017 0.04 0.20 #> D.cl.11 7.797822 130.000000 0.036195 0.10 0.32 #> D.cl.12 8.526186 66.000000 -0.054242 0.02 0.24 #> D.cl.13 9.254550 32.000000 -0.066677 0.04 0.26 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mite.correlog2) # NOTE: 'nperm' argument usually needs to be larger than 49. # It was set to this low value for demonstration purposes."},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":null,"dir":"Reference","previous_headings":"","what":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Function mantel finds Mantel statistic matrix correlation two dissimilarity matrices, function mantel.partial finds partial Mantel statistic partial matrix correlation three dissimilarity matrices. significance statistic evaluated permuting rows columns first dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"mantel(xdis, ydis, method=\"pearson\", permutations=999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\")) mantel.partial(xdis, ydis, zdis, method = \"pearson\", permutations = 999, strata = NULL, na.rm = FALSE, parallel = getOption(\"mc.cores\"))"},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"xdis, ydis, zdis Dissimilarity matrices ordist objects. first object xdis permuted permutation tests. method Correlation method, accepted cor: \"pearson\", \"spearman\" \"kendall\". permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. na.rm Remove missing values calculation Mantel correlation. Use option care: Permutation tests can biased, particular two matrices missing values matching positions. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Mantel statistic simply correlation entries two dissimilarity matrices (use cross products, linearly related). However, significance directly assessed, \\(N(N-1)/2\\) entries just \\(N\\) observations. Mantel developed asymptotic test, use permutations \\(N\\) rows columns dissimilarity matrix. first matrix (xdist) permuted, second kept constant. See permutations additional details permutation tests Vegan. Partial Mantel statistic uses partial correlation conditioned third matrix. first matrix permuted correlation structure second first matrices kept constant. Although mantel.partial silently accepts methods \"pearson\", partial correlations probably wrong methods. function uses cor, accept alternatives pearson product moment correlations spearman kendall rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"function returns list class mantel following components: Call Function call. method Correlation method used, returned cor.test. statistic Mantel statistic. signif Empirical significance level permutations. perm vector permuted values. distribution permuted values can inspected permustats function. permutations Number permutations. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"test due Mantel, course, current implementation based Legendre Legendre. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English Edition. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Legendre & Legendre (2012, Box 10.4) warn using partial Mantel correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mantel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Mantel and Partial Mantel Tests for Dissimilarity Matrices — mantel","text":"","code":"## Is vegetation related to environment? data(varespec) data(varechem) veg.dist <- vegdist(varespec) # Bray-Curtis env.dist <- vegdist(scale(varechem), \"euclid\") mantel(veg.dist, env.dist) #> #> Mantel statistic based on Pearson's product-moment correlation #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist) #> #> Mantel statistic r: 0.3047 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.116 0.147 0.176 0.210 #> Permutation: free #> Number of permutations: 999 #> mantel(veg.dist, env.dist, method=\"spear\") #> #> Mantel statistic based on Spearman's rank correlation rho #> #> Call: #> mantel(xdis = veg.dist, ydis = env.dist, method = \"spear\") #> #> Mantel statistic r: 0.2838 #> Significance: 0.001 #> #> Upper quantiles of permutations (null model): #> 90% 95% 97.5% 99% #> 0.123 0.162 0.182 0.208 #> Permutation: free #> Number of permutations: 999 #>"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS performs Nonmetric Multidimensional Scaling (NMDS), tries find stable solution using several random starts. addition, standardizes scaling result, configurations easier interpret, adds species scores site ordination. metaMDS function provide actual NMDS, calls another function purpose. Currently monoMDS default choice, also possible call isoMDS (MASS package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"metaMDS(comm, distance = \"bray\", k = 2, try = 20, trymax = 20, engine = c(\"monoMDS\", \"isoMDS\"), autotransform =TRUE, noshare = (engine == \"isoMDS\"), wascores = TRUE, expand = TRUE, trace = 1, plot = FALSE, previous.best, ...) # S3 method for metaMDS plot(x, display = c(\"sites\", \"species\"), choices = c(1, 2), type = \"p\", shrink = FALSE, ...) # S3 method for metaMDS points(x, display = c(\"sites\", \"species\"), choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS text(x, display = c(\"sites\", \"species\"), labels, choices = c(1,2), shrink = FALSE, select, ...) # S3 method for metaMDS scores(x, display = c(\"sites\", \"species\"), shrink = FALSE, choices, tidy = FALSE, ...) metaMDSdist(comm, distance = \"bray\", autotransform = TRUE, noshare = TRUE, trace = 1, commname, zerodist = \"ignore\", distfun = vegdist, ...) metaMDSiter(dist, k = 2, try = 20, trymax = 20, trace = 1, plot = FALSE, previous.best, engine = \"monoMDS\", maxit = 200, parallel = getOption(\"mc.cores\"), ...) initMDS(x, k=2) postMDS(X, dist, pc=TRUE, center=TRUE, halfchange, threshold=0.8, nthreshold=10, plot=FALSE, ...) metaMDSredist(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"comm Community data. Alternatively, dissimilarities either dist structure symmetric square matrix. latter case stages skipped except random starts centring pc rotation axes. distance Dissimilarity index used vegdist. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher global non-metric MDS, still higher local NMDS. try, trymax Minimum maximum numbers random starts search stable solution. try reached, iteration stop similar solutions repeated trymax reached. engine function used MDS. default use monoMDS function vegan, backward compatibility also possible use isoMDS MASS. autotransform Use simple heuristics possible data transformation typical community data (see ). community data, probably set autotransform = FALSE. noshare Triggering calculation step-across extended dissimilarities function stepacross. argument can logical numerical value greater zero less one. TRUE, extended dissimilarities used always shared species sites, FALSE, never used. noshare numerical value, stepacross used proportion site pairs shared species exceeds noshare. number pairs shared species found .shared function, noshare effect input data dissimilarities instead community data. wascores Calculate species scores using function wascores. expand Expand weighted averages species wascores. trace Trace function; trace = 2 higher voluminous. plot Graphical tracing: plot interim results. may want set par(ask = TRUE) option. previous.best Start searches previous solution. x metaMDS result (dissimilarity structure initMDS). choices Axes shown. type Plot type: \"p\" points, \"t\" text, \"n\" axes . display Display \"sites\" \"species\". shrink Shrink back species scores expanded originally. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable code (\"sites\" \"species\"), names variable label. scores incompatible conventional plot functions, can used ggplot2. labels Optional test used instead row names. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. X Configuration multidimensional scaling. commname name comm: given function called directly. zerodist Handling zero dissimilarities: either \"fail\" \"add\" small positive value, \"ignore\". monoMDS accepts zero dissimilarities default zerodist = \"ignore\", isoMDS may need set zerodist = \"add\". distfun Dissimilarity function. function returning dist object accepting argument method can used (extra arguments may cause name conflicts). maxit Maximum number iterations single NMDS run; passed engine function monoMDS isoMDS. parallel Number parallel processes predefined socket cluster. use pre-defined socket clusters (say, clus), must issue clusterEvalQ(clus, library(vegan)) make available internal vegan functions. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist Dissimilarity matrix used multidimensional scaling. pc Rotate principal components. center Centre configuration. halfchange Scale axes half-change units. defaults TRUE dissimilarities known theoretical maximum value (ceiling). Function vegdist information attribute maxdist, distfun interpreted simple test (can fail), information may available input data distances. FALSE, ordination dissimilarities scaled range input dissimilarities. threshold Largest dissimilarity used half-change scaling. dissimilarities known (inferred) ceiling, threshold relative ceiling (see halfchange). nthreshold Minimum number points half-change scaling. object result object metaMDS. ... parameters passed functions. Function metaMDS passes arguments component functions metaMDSdist, metaMDSiter, postMDS, distfun engine.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-metric Multidimensional Scaling (NMDS) commonly regarded robust unconstrained ordination method community ecology (Minchin 1987). Function metaMDS wrapper function calls several functions combine Minchin's (1987) recommendations one command. complete steps metaMDS : Transformation: data values larger common abundance class scales, function performs Wisconsin double standardization (wisconsin). values look large, function also performs sqrt transformation. standardizations generally found improve results. However, limits completely arbitrary (present, data maximum 50 triggers sqrt \\(>9\\) triggers wisconsin). want full control analysis, set autotransform = FALSE standardize transform data independently. autotransform intended community data, data types, set autotransform = FALSE. step perfomed using metaMDSdist, step skipped input dissimilarities. Choice dissimilarity: good result, use dissimilarity indices good rank order relation ordering sites along gradients (Faith et al. 1987). default Bray-Curtis dissimilarity, often test winner. However, dissimilarity index vegdist can used. Function rankindex can used finding test winner data gradients. default choice may bad analyse community data, probably select appropriate index using argument distance. step performed using metaMDSdist, step skipped input dissimilarities. Step-across dissimilarities: Ordination may difficult large proportion sites shared species. case, results may improved stepacross dissimilarities, flexible shortest paths among sites. default NMDS engine monoMDS able break tied values maximum dissimilarity, often sufficient handle cases shared species, therefore default use stepacross monoMDS. Function isoMDS handle tied values adequately, therefore default use stepacross always sites shared species engine = \"isoMDS\". stepacross triggered option noshare. like manipulation original distances, set noshare = FALSE. step skipped input data dissimilarities instead community data. step performed using metaMDSdist, step skipped always input dissimilarities. NMDS random starts: NMDS easily gets trapped local optima, must start NMDS several times random starts confident found global solution. strategy metaMDS first run NMDS starting metric scaling (cmdscale usually finds good solution often close local optimum), use previous.best solution supplied, take solution standard (Run 0). metaMDS starts NMDS several random starts (minimum number given try maximum number trymax). random starts generated initMDS. solution better (lower stress) previous standard, taken new standard. solution better close standard, metaMDS compares two solutions using Procrustes analysis (function procrustes option symmetric = TRUE). solutions similar Procrustes rmse largest residual small, solutions regarded repeated better one taken new standard. conditions stringent, may found good relatively similar solutions although function yet satisfied. Setting trace = TRUE monitor final stresses, plot = TRUE display Procrustes overlay plots comparison. step performed using metaMDSiter. first step performed input data (comm) dissimilarities. Random starts can run parallel processing (argument parallel). Scaling results: metaMDS run postMDS final result. Function postMDS provides following ways “fixing” indeterminacy scaling orientation axes NMDS: Centring moves origin average axes; Principal components rotate configuration variance points maximized first dimension (function MDSrotate can alternatively rotate configuration first axis parallel environmental variable); Half-change scaling scales configuration one unit means halving community similarity replicate similarity. Half-change scaling based closer dissimilarities relation ordination distance community dissimilarity rather linear (limit set argument threshold). enough points threshold (controlled parameter nthreshold), dissimilarities regressed distances. intercept regression taken replicate dissimilarity, half-change distance similarity halves according linear regression. Obviously method applicable dissimilarity indices scaled \\(0 \\ldots 1\\), Kulczynski, Bray-Curtis Canberra indices. half-change scaling used, ordination scaled range original dissimilarities. Half-change scaling skipped default input dissimilarities, can turned argument halfchange = TRUE. NB., PC rotation changes directions reference axes, influence configuration solution general. Species scores: Function adds species scores final solution weighted averages using function wascores given value parameter expand. expansion weighted averages can undone shrink = TRUE plot scores functions, calculation species scores can suppressed wascores = FALSE. step skipped input dissimilarities community data unavailable. However, species scores can added replaced sppscores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"results-could-not-be-repeated","dir":"Reference","previous_headings":"","what":"Results Could Not Be Repeated","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Non-linear optimization hard task, best possible solution (“global optimum”) may found random starting configuration. software solve starting result metric scaling (cmdscale). probably give good result, necessarily “global optimum”. Vegan , metaMDS tries verify improve first solution (“try 0”) using several random starts seeing result can repeated improved improved solution repeated. succeed, get message result repeated. However, result least good usual standard strategy starting metric scaling may improved. may need anything message, can satisfied result. want sure probably “global optimum” may try following instructions. default engine = \"monoMDS\" function tabulate stopping criteria used, can see criterion made stringent. criteria can given arguments metaMDS current values described monoMDS. particular, reach maximum number iterations, increase value maxit. may ask larger number random starts without losing old ones giving previous solution argument previous.best. addition slack convergence criteria low number random starts, wrong number dimensions (argument k) common reason able repeat similar solutions. NMDS usually run low number dimensions (k=2 k=3), complex data increasing k one may help. run NMDS much higher number dimensions (say, k=10 ), reconsider drastically reduce k. heterogeneous data sets partial disjunctions, may help set stepacross, data sets default weakties = TRUE sufficient. Please note can give arguments metaMDS* functions NMDS engine (default monoMDS) metaMDS command, check documentation functions details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"common-wrong-claims","dir":"Reference","previous_headings":"","what":"Common Wrong Claims","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"NMDS often misunderstood wrong claims properties common Web even publications. often claimed NMDS configuration non-metric means fit environmental variables species onto space. false statement. fact, result configuration NMDS metric, can used like ordination result. NMDS rank orders Euclidean distances among points ordination non-metric monotone relationship observed dissimilarities. transfer function observed dissimilarities ordination distances non-metric (Kruskal 1964a, 1964b), ordination result configuration metric observed dissimilarities can kind (metric non-metric). ordination configuration usually rotated principal components metaMDS. rotation performed finding result, changes direction reference axes. important feature NMDS solution ordination distances, change rotation. Similarly, rank order distances change uniform scaling centring configuration points. can also rotate NMDS solution external environmental variables MDSrotate. rotation also change orientation axes, change configuration points distances points ordination space. Function stressplot displays method graphically: plots observed dissimilarities distances ordination space, also shows non-metric monotone regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS returns object class metaMDS. final site ordination stored item points, species ordination item species, stress item stress (NB, scaling stress depends engine: isoMDS uses percents, monoMDS proportions range \\(0 \\ldots 1\\)). items store information steps taken items returned engine function. object print, plot, points text methods. Functions metaMDSdist metaMDSredist return vegdist objects. Function initMDS returns random configuration intended used within isoMDS . Functions metaMDSiter postMDS returns result NMDS updated configuration.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Kruskal, J.B. (1964a). Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. (1964b). Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. (1987). evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"Function metaMDS simple wrapper NMDS engine (either monoMDS isoMDS) support functions (metaMDSdist, stepacross, metaMDSiter, initMDS, postMDS, wascores). can call support functions separately better control results. Data transformation, dissimilarities possible stepacross made function metaMDSdist returns dissimilarity result. Iterative search ( starting values initMDS monoMDS) made metaMDSiter. Processing result configuration done postMDS, species scores added wascores. want certain reaching global solution, can compare results several independent runs. can also continue analysis previous results configuration. Function may save used dissimilarity matrix (monoMDS ), metaMDSredist tries reconstruct used dissimilarities original data transformation possible stepacross. metaMDS function designed used community data. type data, probably set arguments non-default values: probably least wascores, autotransform noshare FALSE. negative data entries, metaMDS set previous FALSE warning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"metaMDS uses monoMDS NMDS engine vegan version 2.0-0, replaced isoMDS function. can set argument engine select old engine.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/metaMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nonmetric Multidimensional Scaling with Stable Solution from Random Starts, Axis Scaling and Species Scores — metaMDS","text":"","code":"## The recommended way of running NMDS (Minchin 1987) ## data(dune) ## IGNORE_RDIFF_BEGIN ## Global NMDS using monoMDS sol <- metaMDS(dune) #> Run 0 stress 0.1192678 #> Run 1 stress 0.3016052 #> Run 2 stress 0.1192678 #> ... New best solution #> ... Procrustes: rmse 9.804396e-06 max resid 2.993113e-05 #> ... Similar to previous best #> Run 3 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 0.02027462 max resid 0.06499128 #> Run 4 stress 0.1192679 #> Run 5 stress 0.1183186 #> ... Procrustes: rmse 3.964007e-06 max resid 1.2573e-05 #> ... Similar to previous best #> Run 6 stress 0.1192679 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 6.675204e-06 max resid 1.892118e-05 #> ... Similar to previous best #> Run 8 stress 0.1192678 #> Run 9 stress 0.1183186 #> ... Procrustes: rmse 2.31839e-06 max resid 6.973206e-06 #> ... Similar to previous best #> Run 10 stress 0.1889645 #> Run 11 stress 0.1183186 #> ... Procrustes: rmse 4.550156e-06 max resid 1.4618e-05 #> ... Similar to previous best #> Run 12 stress 0.1192678 #> Run 13 stress 0.1192678 #> Run 14 stress 0.1192678 #> Run 15 stress 0.1192678 #> Run 16 stress 0.1192679 #> Run 17 stress 0.1808912 #> Run 18 stress 0.1183186 #> ... Procrustes: rmse 1.954831e-05 max resid 6.115607e-05 #> ... Similar to previous best #> Run 19 stress 0.1808911 #> Run 20 stress 0.1183186 #> ... New best solution #> ... Procrustes: rmse 1.3197e-06 max resid 3.321941e-06 #> ... Similar to previous best #> *** Best solution repeated 1 times sol #> #> Call: #> metaMDS(comm = dune) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1183186 #> Stress type 1, weak ties #> Best solution was repeated 1 time in 20 tries #> The best solution was from try 20 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> plot(sol, type=\"t\") ## Start from previous best solution sol <- metaMDS(dune, previous.best = sol) #> Starting from 2-dimensional configuration #> Run 0 stress 0.1183186 #> Run 1 stress 0.1183186 #> ... Procrustes: rmse 5.279158e-06 max resid 1.797511e-05 #> ... Similar to previous best #> Run 2 stress 0.1183186 #> ... Procrustes: rmse 2.460953e-06 max resid 5.877957e-06 #> ... Similar to previous best #> Run 3 stress 0.1192678 #> Run 4 stress 0.1886532 #> Run 5 stress 0.1192678 #> Run 6 stress 0.1808911 #> Run 7 stress 0.1183186 #> ... Procrustes: rmse 7.828309e-06 max resid 2.405585e-05 #> ... Similar to previous best #> Run 8 stress 0.1183186 #> ... Procrustes: rmse 6.32253e-06 max resid 2.037397e-05 #> ... Similar to previous best #> Run 9 stress 0.1808911 #> Run 10 stress 0.1183186 #> ... Procrustes: rmse 1.809694e-06 max resid 5.978548e-06 #> ... Similar to previous best #> Run 11 stress 0.1901491 #> Run 12 stress 0.1183186 #> ... Procrustes: rmse 1.546268e-05 max resid 4.957148e-05 #> ... Similar to previous best #> Run 13 stress 0.1812932 #> Run 14 stress 0.1183186 #> ... Procrustes: rmse 5.112711e-06 max resid 1.098817e-05 #> ... Similar to previous best #> Run 15 stress 0.1183186 #> ... Procrustes: rmse 6.246131e-06 max resid 2.000824e-05 #> ... Similar to previous best #> Run 16 stress 0.1192679 #> Run 17 stress 0.1192678 #> Run 18 stress 0.1886532 #> Run 19 stress 0.1192679 #> Run 20 stress 0.1192678 #> *** Best solution repeated 9 times ## Local NMDS and stress 2 of monoMDS sol2 <- metaMDS(dune, model = \"local\", stress=2) #> Run 0 stress 0.1928478 #> Run 1 stress 0.1928482 #> ... Procrustes: rmse 0.0006537695 max resid 0.001899655 #> ... Similar to previous best #> Run 2 stress 0.1928476 #> ... New best solution #> ... Procrustes: rmse 0.0004471422 max resid 0.001294366 #> ... Similar to previous best #> Run 3 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 0.0002885977 max resid 0.0008320155 #> ... Similar to previous best #> Run 4 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 5.59458e-05 max resid 0.0001616763 #> ... Similar to previous best #> Run 5 stress 0.1928475 #> ... Procrustes: rmse 4.864069e-05 max resid 0.0001571056 #> ... Similar to previous best #> Run 6 stress 0.1928477 #> ... Procrustes: rmse 0.0002420844 max resid 0.0006909829 #> ... Similar to previous best #> Run 7 stress 0.1928475 #> ... Procrustes: rmse 0.0001346077 max resid 0.0003887498 #> ... Similar to previous best #> Run 8 stress 0.1928475 #> ... Procrustes: rmse 0.0001625818 max resid 0.0004729866 #> ... Similar to previous best #> Run 9 stress 0.1928475 #> ... Procrustes: rmse 8.000053e-05 max resid 0.0002277773 #> ... Similar to previous best #> Run 10 stress 0.192849 #> ... Procrustes: rmse 0.0006301494 max resid 0.001813666 #> ... Similar to previous best #> Run 11 stress 0.1928475 #> ... Procrustes: rmse 0.0001014964 max resid 0.0002751379 #> ... Similar to previous best #> Run 12 stress 0.192848 #> ... Procrustes: rmse 0.0002964116 max resid 0.0008527077 #> ... Similar to previous best #> Run 13 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 6.461832e-05 max resid 0.0001901839 #> ... Similar to previous best #> Run 14 stress 0.1928482 #> ... Procrustes: rmse 0.0003439501 max resid 0.001017732 #> ... Similar to previous best #> Run 15 stress 0.1928479 #> ... Procrustes: rmse 0.0002821687 max resid 0.0008172892 #> ... Similar to previous best #> Run 16 stress 0.1928475 #> ... New best solution #> ... Procrustes: rmse 3.13617e-05 max resid 8.909983e-05 #> ... Similar to previous best #> Run 17 stress 0.1928475 #> ... Procrustes: rmse 4.709197e-05 max resid 0.0001417594 #> ... Similar to previous best #> Run 18 stress 0.1928477 #> ... Procrustes: rmse 0.0002414341 max resid 0.0007100825 #> ... Similar to previous best #> Run 19 stress 0.1928479 #> ... Procrustes: rmse 0.0003023992 max resid 0.0008897382 #> ... Similar to previous best #> Run 20 stress 0.1928476 #> ... Procrustes: rmse 0.0002063278 max resid 0.0006048887 #> ... Similar to previous best #> *** Best solution repeated 5 times sol2 #> #> Call: #> metaMDS(comm = dune, model = \"local\", stress = 2) #> #> local Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: bray #> #> Dimensions: 2 #> Stress: 0.1928475 #> Stress type 2, weak ties #> Best solution was repeated 5 times in 20 tries #> The best solution was from try 16 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## Use Arrhenius exponent 'z' as a binary dissimilarity measure sol <- metaMDS(dune, distfun = betadiver, distance = \"z\") #> Run 0 stress 0.1067169 #> Run 1 stress 0.1073148 #> Run 2 stress 0.1073148 #> Run 3 stress 0.1073148 #> Run 4 stress 0.1073148 #> Run 5 stress 0.1073148 #> Run 6 stress 0.1067169 #> ... New best solution #> ... Procrustes: rmse 2.925047e-06 max resid 7.049929e-06 #> ... Similar to previous best #> Run 7 stress 0.1069781 #> ... Procrustes: rmse 0.006690411 max resid 0.02349817 #> Run 8 stress 0.1067169 #> ... Procrustes: rmse 6.432458e-06 max resid 1.886945e-05 #> ... Similar to previous best #> Run 9 stress 0.1069786 #> ... Procrustes: rmse 0.006774522 max resid 0.02386629 #> Run 10 stress 0.1067169 #> ... Procrustes: rmse 2.346391e-06 max resid 5.306599e-06 #> ... Similar to previous best #> Run 11 stress 0.1067169 #> ... Procrustes: rmse 2.816635e-06 max resid 7.563776e-06 #> ... Similar to previous best #> Run 12 stress 0.1069784 #> ... Procrustes: rmse 0.006735205 max resid 0.02369698 #> Run 13 stress 0.1067169 #> ... Procrustes: rmse 3.36442e-06 max resid 1.127286e-05 #> ... Similar to previous best #> Run 14 stress 0.2355185 #> Run 15 stress 0.1644741 #> Run 16 stress 0.1067169 #> ... Procrustes: rmse 8.389288e-06 max resid 2.174836e-05 #> ... Similar to previous best #> Run 17 stress 0.1073148 #> Run 18 stress 0.1067169 #> ... Procrustes: rmse 5.169725e-06 max resid 1.633663e-05 #> ... Similar to previous best #> Run 19 stress 0.1067169 #> ... Procrustes: rmse 3.667758e-06 max resid 9.091241e-06 #> ... Similar to previous best #> Run 20 stress 0.1073148 #> *** Best solution repeated 8 times sol #> #> Call: #> metaMDS(comm = dune, distance = \"z\", distfun = betadiver) #> #> global Multidimensional Scaling using monoMDS #> #> Data: dune #> Distance: beta.z #> #> Dimensions: 2 #> Stress: 0.1067169 #> Stress type 1, weak ties #> Best solution was repeated 8 times in 20 tries #> The best solution was from try 6 (random start) #> Scaling: centring, PC rotation, halfchange scaling #> Species: expanded scores based on ‘dune’ #> ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":null,"dir":"Reference","previous_headings":"","what":"Oribatid Mite Data with Explanatory Variables — mite","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Oribatid mite data. 70 soil cores collected Daniel Borcard 1989. See Borcard et al. (1992, 1994) details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) data(mite.xy)"},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"three linked data sets: mite contains data 35 species Oribatid mites, mite.env contains environmental data sampling sites, mite.xy contains geographic coordinates, mite.pcnm contains 22 PCNM base functions (columns) computed geographic coordinates 70 sampling sites (Borcard & Legendre 2002). whole sampling area 2.5 m x 10 m size. fields environmental data : SubsDens Substrate density (g/L) WatrCont Water content substrate (g/L) Substrate Substrate type, factor levels Sphagn1, \tSphagn2 Sphagn3 Sphagn Litter Barepeat Interface Shrub Shrub density, ordered factor levels 1 < 2 < 3 Topo Microtopography, factor levels Blanket Hummock","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Pierre Legendre","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"Borcard, D., P. Legendre P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045-1055. Borcard, D. P. Legendre. 1994. Environmental control spatial structure ecological communities: example using Oribatid mites (Acari, Oribatei). Environmental Ecological Statistics 1: 37-61. Borcard, D. P. Legendre. 2002. -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153: 51-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Oribatid Mite Data with Explanatory Variables — mite","text":"","code":"data(mite)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":null,"dir":"Reference","previous_headings":"","what":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Function implements Kruskal's (1964a,b) non-metric multidimensional scaling (NMDS) using monotone regression primary (“weak”) treatment ties. addition traditional global NMDS, function implements local NMDS, linear hybrid multidimensional scaling.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"monoMDS(dist, y, k = 2, model = c(\"global\", \"local\", \"linear\", \"hybrid\"), threshold = 0.8, maxit = 200, weakties = TRUE, stress = 1, scaling = TRUE, pc = TRUE, smin = 1e-4, sfgrmin = 1e-7, sratmax=0.999999, ...) # S3 method for monoMDS scores(x, choices = NA, ...) # S3 method for monoMDS plot(x, choices = c(1,2), type = \"t\", ...) # S3 method for monoMDS points(x, choices = c(1,2), select, ...) # S3 method for monoMDS text(x, labels, choices = c(1,2), select, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"dist Input dissimilarities. y Starting configuration. random configuration generated missing. k Number dimensions. NB., number points \\(n\\) \\(n > 2k + 1\\), preferably higher non-metric MDS. model MDS model: \"global\" normal non-metric MDS monotone regression, \"local\" non-metric MDS separate regressions point, \"linear\" uses linear regression, \"hybrid\" uses linear regression dissimilarities threshold addition monotone regression. See Details. threshold Dissimilarity linear regression used alternately monotone regression. maxit Maximum number iterations. weakties Use primary weak tie treatment, equal observed dissimilarities allowed different fitted values. FALSE, secondary (strong) tie treatment used, tied values broken. stress Use stress type 1 2 (see Details). scaling Scale final scores unit root mean squares. pc Rotate final scores principal components. smin, sfgrmin, sratmax Convergence criteria: iterations stop stress drops smin, scale factor gradient drops sfgrmin, stress ratio two iterations goes sratmax (still \\(< 1\\)). x monoMDS result. choices Dimensions returned plotted. default NA returns dimensions. type type plot: \"t\" text, \"p\" points, \"n\" none. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. labels Labels use used instead row names. ... parameters functions (ignored monoMDS, passed graphical functions plot.).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"several versions non-metric multidimensional scaling R, monoMDS offers following unique combination features: “Weak” treatment ties (Kruskal 1964a,b), tied dissimilarities can broken monotone regression. especially important cases compared sites share species dissimilarities tied maximum value one. Breaking ties allows points different distances can help recovering long coenoclines (gradients). Functions smacof package also hav adequate tie treatment. Handles missing values meaningful way. Offers “local” “hybrid” scaling addition usual “global” NMDS (see ). Uses fast compiled code (isoMDS MASS package also uses compiled code). Function monoMDS uses Kruskal's (1964b) original monotone regression minimize stress. two alternatives stress: Kruskal's (1964a,b) original “stress 1” alternative version “stress 2” (Sibson 1972). stresses can expressed general formula $$s^2 = \\frac{\\sum (d - \\hat d)^2}{\\sum(d - d_0)^2}$$ \\(d\\) distances among points ordination configuration, \\(\\hat d\\) fitted ordination distances, \\(d_0\\) ordination distances null model. “stress 1” \\(d_0 = 0\\), “stress 2” \\(d_0 = \\bar{d}\\) mean distances. “Stress 2” can expressed \\(s^2 = 1 - R^2\\), \\(R^2\\) squared correlation fitted values ordination distances, related “linear fit” stressplot. Function monoMDS can fit several alternative NMDS variants can selected argument model. default model = \"global\" fits global NMDS, Kruskal's (1964a,b) original NMDS similar isoMDS (MASS). Alternative model = \"local\" fits local NMDS independent monotone regression used point (Sibson 1972). Alternative model = \"linear\" fits linear MDS. fits linear regression instead monotone, identical metric scaling principal coordinates analysis (cmdscale) performs eigenvector decomposition dissimilarities (Gower 1966). Alternative model = \"hybrid\" implements hybrid MDS uses monotone regression points linear regression dissimilarities threshold dissimilarity alternating steps (Faith et al. 1987). Function stressplot can used display kind regression model. Scaling, orientation direction axes arbitrary. However, function always centres axes, default scaling scale configuration unit root mean square rotate axes (argument pc) principal components first dimension shows major variation. possible rotate solution first axis parallel given environmental variable using function MDSrotate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"convergence-criteria","dir":"Reference","previous_headings":"","what":"Convergence Criteria","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"NMDS iterative, function stops convergence criteria met. actually criterion assured convergence, solution can local optimum. compare several random starts (use monoMDS via metaMDS) assess solutions likely global optimum. stopping criteria : maxit: Maximum number iterations. Reaching criterion means solutions almost certainly found, maxit increased. smin: Minimum stress. stress nearly zero, fit almost perfect. Usually means data set small requested analysis, may several different solutions almost perfect. reduce number dimensions (k), get data ( observations) use method, metric scaling (cmdscale, wcmdscale). sratmax: Change stress. Values close one mean almost unchanged stress. may mean solution, can also signal stranding suboptimal solution flat stress surface. sfgrmin: Minimum scale factor. Values close zero mean almost unchanged configuration. may mean solution, also happen local optima.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Returns object class \"monoMDS\". final scores returned item points (function scores extracts results), stress item stress. addition, large number items (may change without notice future releases).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Faith, D.P., Minchin, P.R Belbin, L. 1987. Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J.C. (1966). distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Kruskal, J.B. 1964a. Multidimensional scaling optimizing goodness--fit nonmetric hypothesis. Psychometrika 29, 1--28. Kruskal, J.B. 1964b. Nonmetric multidimensional scaling: numerical method. Psychometrika 29, 115--129. Minchin, P.R. 1987. evaluation relative robustness techniques ecological ordinations. Vegetatio 69, 89--107. Sibson, R. 1972. Order invariant methods data analysis. Journal Royal Statistical Society B 34, 311--349.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"Peter R. Michin (Fortran core) Jari Oksanen (R interface).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"default NMDS function used metaMDS. Function metaMDS adds support functions NMDS can run like recommended Minchin (1987).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/monoMDS.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Global and Local Non-metric Multidimensional Scaling and Linear and Hybrid Scaling — monoMDS","text":"","code":"data(dune) dis <- vegdist(dune) m <- monoMDS(dis, model = \"loc\") m #> #> Call: #> monoMDS(dist = dis, model = \"loc\") #> #> Local non-metric Multidimensional Scaling #> #> 20 points, dissimilarity ‘bray’, call ‘vegdist(x = dune)’ #> #> Dimensions: 2 #> Stress: 0.07596745 #> Stress type 1, weak ties #> Scores scaled to unit root mean square, rotated to principal components #> Stopped after 62 iterations: Stress nearly unchanged (ratio > sratmax) plot(m)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":null,"dir":"Reference","previous_headings":"","what":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. Function meandist finds mean within block dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"mrpp(dat, grouping, permutations = 999, distance = \"euclidean\", weight.type = 1, strata = NULL, parallel = getOption(\"mc.cores\")) meandist(dist, grouping, ...) # S3 method for meandist summary(object, ...) # S3 method for meandist plot(x, kind = c(\"dendrogram\", \"histogram\"), cluster = \"average\", ylim, axes = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"dat data matrix data frame rows samples columns response variable(s), dissimilarity object symmetric square matrix dissimilarities. grouping Factor numeric index grouping observations. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. used assess significance MRPP statistic, \\(delta\\). distance Choice distance metric measures dissimilarity two observations . See vegdist options. used dat dissimilarity structure symmetric square matrix. weight.type choice group weights. See Details options. strata integer vector factor specifying strata permutation. supplied, observations permuted within specified strata. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. dist dist object dissimilarities, produced functions dist, vegdist designdist. . object, x meandist result object. kind Draw dendrogram histogram; see Details. cluster clustering method hclust function kind = \"dendrogram\". hclust method can used, perhaps \"average\" \"single\" make sense. ylim Limits vertical axes (optional). axes Draw scale vertical axis. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"Multiple Response Permutation Procedure (MRPP) provides test whether significant difference two groups sampling units. difference may one location (differences mean) one spread (differences within-group distance; cf. Warton et al. 2012). Function mrpp operates data.frame matrix rows observations responses data matrix. response(s) may uni- multivariate. method philosophically mathematically allied analysis variance, compares dissimilarities within among groups. two groups sampling units really different (e.g. species composition), average within-group compositional dissimilarities less average dissimilarities two random collection sampling units drawn entire population. mrpp statistic \\(\\delta\\) overall weighted mean within-group means pairwise dissimilarities among sampling units. choice group weights currently clear. mrpp function offers three choices: (1) group size (\\(n\\)), (2) degrees--freedom analogue (\\(n-1\\)), (3) weight number unique distances calculated among \\(n\\) sampling units (\\(n(n-1)/2\\)). mrpp algorithm first calculates pairwise distances entire dataset, calculates \\(\\delta\\). permutes sampling units associated pairwise distances, recalculates \\(\\delta\\) based permuted data. repeats permutation step permutations times. significance test fraction permuted deltas less observed delta, small sample correction. function also calculates change-corrected within-group agreement \\(= 1 -\\delta/E(\\delta)\\), \\(E(\\delta)\\) expected \\(\\delta\\) assessed average dissimilarities. first argument dat can interpreted dissimilarities, used directly. cases function treats dat observations, uses vegdist find dissimilarities. default distance Euclidean traditional use method, dissimilarities vegdist also available. Function meandist calculates matrix mean within-cluster dissimilarities (diagonal) -cluster dissimilarities (-diagonal elements), attribute n grouping counts. Function summary finds within-class, -class overall means dissimilarities, MRPP statistics weight.type options Classification Strength, CS (Van Sickle Hughes, 2000). CS defined dissimilarities \\(\\bar{B} - \\bar{W}\\), \\(\\bar{B}\\) mean cluster dissimilarity \\(\\bar{W}\\) mean within cluster dissimilarity weight.type = 1. function perform significance tests statistics, must use mrpp appropriate weight.type. currently significance test CS, mrpp weight.type = 1 gives correct test \\(\\bar{W}\\) good approximation CS. Function plot draws dendrogram histogram result matrix based within-group group dissimilarities. dendrogram found method given cluster argument using function hclust. terminal segments hang within-cluster dissimilarity. clusters heterogeneous combined class, leaf segment reversed. histograms based dissimilarities, ore otherwise similar Van Sickle Hughes (2000): horizontal line drawn level mean -cluster dissimilarity vertical lines connect within-cluster dissimilarities line.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"function returns list class mrpp following items: call Function call. delta overall weighted mean group mean distances. E.delta expected delta, null hypothesis group structure. mean original dissimilarities. CS Classification strength (Van Sickle Hughes, 2000). Currently implemented always NA. n Number observations class. classdelta Mean dissimilarities within classes. overall \\(\\delta\\) weighted average values given weight.type . Pvalue Significance test. chance-corrected estimate proportion distances explained group identity; value analogous coefficient determination linear model. distance Choice distance metric used; \"method\" entry dist object. weight.type choice group weights used. boot.deltas vector \"permuted deltas,\" deltas calculated permuted datasets. distribution item can inspected permustats function. permutations number permutations used. control list control values permutations returned function .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"B. McCune J. B. Grace. 2002. Analysis Ecological Communities. MjM Software Design, Gleneden Beach, Oregon, USA. P. W. Mielke K. J. Berry. 2001. Permutation Methods: Distance Function Approach. Springer Series Statistics. Springer. J. Van Sickle R. M. Hughes 2000. Classification strengths ecoregions, catchments, geographic clusters aquatic vertebrates Oregon. J. N. . Benthol. Soc. 19:370--384. Warton, D.., Wright, T.W., Wang, Y. 2012. Distance-based multivariate analyses confound location dispersion effects. Methods Ecology Evolution, 3, 89--101","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"M. Henry H. Stevens HStevens@muohio.edu Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"difference may one location (differences mean) one spread (differences within-group distance). , may find significant difference two groups simply one groups greater dissimilarities among sampling units. mrpp models can analysed adonis2 seems suffer problems mrpp robust alternative.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mrpp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multi Response Permutation Procedure and Mean Dissimilarity Matrix — mrpp","text":"","code":"data(dune) data(dune.env) dune.mrpp <- with(dune.env, mrpp(dune, Management)) dune.mrpp #> #> Call: #> mrpp(dat = dune, grouping = Management) #> #> Dissimilarity index: euclidean #> Weights for groups: n #> #> Class means and counts: #> #> BF HF NM SF #> delta 10.03 11.08 10.66 12.27 #> n 3 5 6 6 #> #> Chance corrected within-group agreement A: 0.1246 #> Based on observed delta 11.15 and expected delta 12.74 #> #> Significance of delta: 0.001 #> Permutation: free #> Number of permutations: 999 #> # Save and change plotting parameters def.par <- par(no.readonly = TRUE) layout(matrix(1:2,nr=1)) plot(dune.ord <- metaMDS(dune, trace=0), type=\"text\", display=\"sites\" ) with(dune.env, ordihull(dune.ord, Management)) with(dune.mrpp, { fig.dist <- hist(boot.deltas, xlim=range(c(delta,boot.deltas)), main=\"Test of Differences Among Groups\") abline(v=delta); text(delta, 2*mean(fig.dist$counts), adj = -0.5, expression(bold(delta)), cex=1.5 ) } ) par(def.par) ## meandist dune.md <- with(dune.env, meandist(vegdist(dune), Management)) dune.md #> BF HF NM SF #> BF 0.4159972 0.4736637 0.7296979 0.6247169 #> HF 0.4736637 0.4418115 0.7217933 0.5673664 #> NM 0.7296979 0.7217933 0.6882438 0.7723367 #> SF 0.6247169 0.5673664 0.7723367 0.5813015 #> attr(,\"class\") #> [1] \"meandist\" \"matrix\" #> attr(,\"n\") #> grouping #> BF HF NM SF #> 3 5 6 6 summary(dune.md) #> #> Mean distances: #> Average #> within groups 0.5746346 #> between groups 0.6664172 #> overall 0.6456454 #> #> Summary statistics: #> Statistic #> MRPP A weights n 0.1423836 #> MRPP A weights n-1 0.1339124 #> MRPP A weights n(n-1) 0.1099842 #> Classification strength 0.1127012 plot(dune.md) plot(dune.md, kind=\"histogram\")"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":null,"dir":"Reference","previous_headings":"","what":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso adds attribute vario object class \"cca\" describes spatial partitioning cca object performs optional permutation test spatial independence residuals. function plot.mso creates diagnostic plot spatial partitioning \"cca\" object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"mso(object.cca, object.xy, grain = 1, round.up = FALSE, permutations = 0) msoplot(x, alpha = 0.05, explained = FALSE, ylim = NULL, legend = \"topleft\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"object.cca object class cca, created cca rda function. object.xy vector, matrix data frame spatial coordinates data represented object.cca. number rows must match number observations (given nobs) cca.object. Alternatively, interpoint distances can supplied dist object. grain Interval size distance classes. round.Determines choice breaks. false, distances rounded nearest multiple grain. true, distances rounded upper multiple grain. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. x result object mso. alpha Significance level two-sided permutation test Mantel statistic spatial independence residual inertia point-wise envelope variogram total variance. Bonferroni-type correction can achieved dividing overall significance value (e.g. 0.05) number distance classes. explained false, suppresses plotting variogram explained variance. ylim Limits y-axis. legend x y co-ordinates used position legend. can specified keyword way accepted legend. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Mantel test adaptation function mantel vegan package parallel testing several distance classes. compares mean inertia distance class pooled mean inertia distance classes. explanatory variables (RDA, CCA, pRDA, pCCA) significance test residual autocorrelation performed running function mso, function plot.mso print estimate much autocorrelation (based significant distance classes) causes global error variance regression analysis underestimated","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function mso returns amended cca rda object additional attributes grain, H, H.test vario. grain grain attribute defines interval size distance classes . H H object class 'dist' contains geographic distances observations. H.test H.test contains set dummy variables describe pairs observations (rows = elements object$H) fall distance class (columns). vario vario attribute data frame contains following components rda case (cca case brackets): H Distance class multiples grain. Dist Average distance pairs observations distance class H. n Number unique pairs observations distance class \tH. Empirical (chi-square) variogram total variance \t(inertia). Sum Sum empirical (chi-square) variograms explained \tresidual variance (inertia). CA Empirical (chi-square) variogram residual variance \t(inertia). CCA Empirical (chi-square) variogram explained variance \t(inertia). pCCA Empirical (chi-square) variogram conditioned \tvariance (inertia). se Standard error empirical (chi-square) variogram \ttotal variance (inertia). CA.signif P-value permutation test spatial \tindependence residual variance (inertia).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"Wagner, H.H. 2004. Direct multi-scale ordination canonical correspondence analysis. Ecology 85: 342--351.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"responsible author Helene Wagner.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"function based code published Ecological Archives E085-006 (doi:10.1890/02-0738 ).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/mso.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functions for performing and displaying a spatial partitioning of cca or rda results — mso","text":"","code":"## Reconstruct worked example of Wagner (submitted): X <- matrix(c(1, 2, 3, 2, 1, 0), 3, 2) Y <- c(3, -1, -2) tmat <- c(1:3) ## Canonical correspondence analysis (cca): Example.cca <- cca(X, Y) Example.cca <- mso(Example.cca, tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.cca) Example.cca$vario #> H Dist n All Sum CA CCA se #> 1 1 1 2 0.25 0.3456633 0.07461735 0.2710459 0 #> 2 2 2 1 1.00 0.8086735 0.01147959 0.7971939 NA ## Correspondence analysis (ca): Example.ca <- mso(cca(X), tmat) #> Set of permutations < 'minperm'. Generating entire set. msoplot(Example.ca) ## Unconstrained ordination with test for autocorrelation ## using oribatid mite data set as in Wagner (2004) data(mite) data(mite.env) data(mite.xy) mite.cca <- cca(log(mite + 1)) mite.cca <- mso(mite.cca, mite.xy, grain = 1, permutations = 99) msoplot(mite.cca) mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, grain = 1, #> permutations = 99) #> #> Inertia Rank #> Total 1.164 #> Unconstrained 1.164 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.3662 0.1328 0.0723 0.0658 0.0559 0.0481 0.0418 0.0391 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All CA CA.signif #> 0 0 0.3555 63 0.6250 0.6250 0.01 #> 1 1 1.0659 393 0.7556 0.7556 0.01 #> 2 2 2.0089 534 0.8931 0.8931 0.01 #> 3 3 2.9786 417 1.0988 1.0988 0.01 #> 4 4 3.9817 322 1.3321 1.3321 0.01 #> 5 5 5.0204 245 1.5109 1.5109 0.01 #> 10 10 6.8069 441 1.7466 1.7466 0.01 #> #> Permutation: free #> Number of permutations: 99 #> ## Constrained ordination with test for residual autocorrelation ## and scale-invariance of species-environment relationships mite.cca <- cca(log(mite + 1) ~ SubsDens + WatrCont + Substrate + Shrub + Topo, mite.env) mite.cca <- mso(mite.cca, mite.xy, permutations = 99) msoplot(mite.cca) #> Error variance of regression model underestimated by 0.4 percent mite.cca #> Call: mso(object.cca = mite.cca, object.xy = mite.xy, permutations = #> 99) #> #> Inertia Proportion Rank #> Total 1.1638 1.0000 #> Constrained 0.5211 0.4478 11 #> Unconstrained 0.6427 0.5522 34 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 CCA7 CCA8 CCA9 CCA10 #> 0.31207 0.06601 0.04117 0.02938 0.02438 0.01591 0.01201 0.00752 0.00612 0.00373 #> CCA11 #> 0.00284 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 #> 0.07888 0.06752 0.05457 0.04023 0.03855 0.03491 0.03233 0.02692 #> (Showing 8 of 34 unconstrained eigenvalues) #> #> mso variogram: #> #> H Dist n All Sum CA CCA se CA.signif #> 0 0 0.3555 63 0.6250 0.7479 0.5512 0.1967 0.03506 0.01 #> 1 1 1.0659 393 0.7556 0.8820 0.6339 0.2482 0.01573 0.18 #> 2 2 2.0089 534 0.8931 0.9573 0.6473 0.3100 0.01487 0.68 #> 3 3 2.9786 417 1.0988 1.1010 0.6403 0.4607 0.01858 0.46 #> 4 4 3.9817 322 1.3321 1.2548 0.6521 0.6027 0.02439 0.98 #> 5 5 5.0204 245 1.5109 1.4564 0.6636 0.7928 0.02801 0.41 #> 10 10 6.8069 441 1.7466 1.6266 0.6914 0.9351 0.02052 0.19 #> #> Permutation: free #> Number of permutations: 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":null,"dir":"Reference","previous_headings":"","what":"Multiplicative Diversity Partitioning — multipart","title":"Multiplicative Diversity Partitioning — multipart","text":"multiplicative diversity partitioning, mean values alpha diversity lower levels sampling hierarchy compared total diversity entire data set pooled samples (gamma diversity).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"multipart(...) # S3 method for default multipart(y, x, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...) # S3 method for formula multipart(formula, data, index=c(\"renyi\", \"tsallis\"), scales = 1, global = FALSE, relative = FALSE, nsimul=99, method = \"r2dtable\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Multiplicative Diversity Partitioning — multipart","text":"y community matrix. x matrix number rows y, columns coding levels sampling hierarchy. number groups within hierarchy must decrease left right. x missing, two levels assumed: row group first level, rows group second level. formula two sided model formula form y ~ x, y community data matrix samples rows species column. Right hand side (x) must grouping variable(s) referring levels sampling hierarchy, terms right left treated nested (first column lowest, last highest level). formula add unique indentifier rows constant rows always produce estimates row-level alpha overall gamma diversities. must use non-formula interface avoid behaviour. Interaction terms allowed. data data frame look variables defined right hand side formula. missing, variables looked global environment. index Character, entropy index calculated (see Details). relative Logical, TRUE beta diversity standardized maximum (see Details). scales Numeric, length 1, order generalized diversity index used. global Logical, indicates calculation beta diversity values, see Details. nsimul Number permutations use. nsimul = 0, FUN argument evaluated. thus possible reuse statistic values without null model. method Null model method: either name (character string) method defined make.commsim commsim function. default \"r2dtable\" keeps row sums column sums fixed. See oecosimu Details Examples. ... arguments passed oecosimu, .e. method, thin burnin.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Multiplicative Diversity Partitioning — multipart","text":"Multiplicative diversity partitioning based Whittaker's (1972) ideas, recently generalised one parametric diversity families (.e. Rényi Tsallis) Jost (2006, 2007). Jost recommends use numbers equivalents (Hill numbers), instead pure diversities, proofs, satisfies multiplicative partitioning requirements. current implementation multipart calculates Hill numbers based functions renyi tsallis (provided index argument). values one scales desired, done separate runs, adds extra dimensionality implementation, resolved efficiently. Alpha diversities averages Hill numbers hierarchy levels, global gamma diversity alpha value calculated highest hierarchy level. global = TRUE, beta calculated relative global gamma value: $$\\beta_i = \\gamma / \\alpha_{}$$ global = FALSE, beta calculated relative local gamma values (local gamma means diversity calculated particular cluster based pooled abundance vector): $$\\beta_ij = \\alpha_{(+1)j} / mean(\\alpha_{ij})$$ \\(j\\) particular cluster hierarchy level \\(\\). beta diversity value level \\(\\) mean beta values clusters level, \\(\\beta_{} = mean(\\beta_{ij})\\). relative = TRUE, respective beta diversity values standardized maximum possible values (\\(mean(\\beta_{ij}) / \\beta_{max,ij}\\)) given \\(\\beta_{max,ij} = n_{j}\\) (number lower level units given cluster \\(j\\)). expected diversity components calculated nsimul times individual based randomization community data matrix. done \"r2dtable\" method oecosimu default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Multiplicative Diversity Partitioning — multipart","text":"object class \"multipart\" structure \"oecosimu\" objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Multiplicative Diversity Partitioning — multipart","text":"Jost, L. (2006). Entropy diversity. Oikos, 113, 363--375. Jost, L. (2007). Partitioning diversity independent alpha beta components. Ecology, 88, 2427--2439. Whittaker, R. (1972). Evolution measurement species diversity. Taxon, 21, 213--251.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Multiplicative Diversity Partitioning — multipart","text":"Péter Sólymos, solymos@ualberta.ca","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/multipart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Multiplicative Diversity Partitioning — multipart","text":"","code":"## NOTE: 'nsimul' argument usually needs to be >= 99 ## here much lower value is used for demonstration data(mite) data(mite.xy) data(mite.env) ## Function to get equal area partitions of the mite data cutter <- function (x, cut = seq(0, 10, by = 2.5)) { out <- rep(1, length(x)) for (i in 2:(length(cut) - 1)) out[which(x > cut[i] & x <= cut[(i + 1)])] <- i return(out)} ## The hierarchy of sample aggregation levsm <- with(mite.xy, data.frame( l2=cutter(y, cut = seq(0, 10, by = 2.5)), l3=cutter(y, cut = seq(0, 10, by = 5)))) ## Multiplicative diversity partitioning multipart(mite, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(y = mite, x = levsm, index = \"renyi\", scales = 1, #> nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 11.235 -81.754 14.07974 14.02539 14.07261 14.148 0.05 * #> gamma 12.006 -256.950 14.13219 14.11654 14.13205 14.146 0.05 * #> beta.1 1.071 29.363 1.00373 0.99935 1.00399 1.008 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ l2 + l3, levsm, index=\"renyi\", scales=1, nsimul=19) #> multipart object #> #> Call: multipart(formula = mite ~ l2 + l3, data = levsm, index = #> \"renyi\", scales = 1, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -46.004 12.17819 12.02119 12.15824 12.3415 0.05 * #> alpha.2 11.2353 -87.290 14.08463 14.03637 14.07817 14.1383 0.05 * #> alpha.3 12.0064 -490.481 14.13774 14.13229 14.13634 14.1450 0.05 * #> gamma 14.1603 0.000 14.16027 14.16027 14.16027 14.1603 1.00 #> beta.1 1.3568 20.034 1.16000 1.14399 1.16150 1.1779 0.05 * #> beta.2 1.0710 29.394 1.00379 0.99989 1.00410 1.0073 0.05 * #> beta.3 1.1794 577.616 1.00159 1.00108 1.00169 1.0020 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, relative=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, relative = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global FALSE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.055481 -46.949 12.221716 12.073934 12.244586 12.3654 0.05 * #> alpha.2 11.235261 -123.496 14.087872 14.043345 14.090654 14.1218 0.05 * #> alpha.3 12.006443 -323.402 14.135613 14.124838 14.135306 14.1443 0.05 * #> gamma 14.160271 0.000 14.160271 14.160271 14.160271 14.1603 1.00 #> beta.1 0.078594 17.624 0.068099 0.067172 0.068222 0.0691 0.05 * #> beta.2 0.535514 42.418 0.501697 0.500646 0.501814 0.5031 0.05 * #> beta.3 0.589695 380.700 0.500872 0.500566 0.500883 0.5013 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 multipart(mite ~ ., levsm, index=\"renyi\", scales=1, nsimul=19, global=TRUE) #> multipart object #> #> Call: multipart(formula = mite ~ ., data = levsm, index = \"renyi\", #> scales = 1, global = TRUE, nsimul = 19) #> #> nullmodel method ‘r2dtable’ with 19 simulations #> options: index renyi, scales 1, global TRUE #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> alpha.1 8.0555 -74.528 12.1813 12.0860 12.1748 12.2676 0.05 * #> alpha.2 11.2353 -92.796 14.0859 14.0459 14.0812 14.1484 0.05 * #> alpha.3 12.0064 -331.084 14.1370 14.1256 14.1378 14.1455 0.05 * #> gamma 14.1603 0.000 14.1603 14.1603 14.1603 14.1603 1.00 #> beta.1 1.7578 112.658 1.1625 1.1543 1.1631 1.1716 0.05 * #> beta.2 1.2603 116.533 1.0053 1.0008 1.0056 1.0081 0.05 * #> beta.3 1.1794 389.764 1.0016 1.0010 1.0016 1.0025 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":null,"dir":"Reference","previous_headings":"","what":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Patches local communities regarded nested subsets community. general, species poor communities subsets species rich communities, rare species occur species rich communities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"nestedchecker(comm) nestedn0(comm) nesteddisc(comm, niter = 200) nestedtemp(comm, ...) nestednodf(comm, order = TRUE, weighted = FALSE, wbinary = FALSE) nestedbetasor(comm) nestedbetajac(comm) # S3 method for nestedtemp plot(x, kind = c(\"temperature\", \"incidence\"), col=rev(heat.colors(100)), names = FALSE, ...) # S3 method for nestednodf plot(x, col = \"red\", names = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"comm Community data. niter Number iterations reorder tied columns. x Result object plot. col Colour scheme matrix temperatures. kind kind plot produced. names Label columns rows plot using names comm. logical vector length 2, row column labels returned accordingly. order Order rows columns frequencies. weighted Use species abundances weights interactions. wbinary Modify original method binary data give result weighted unweighted analysis. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"nestedness functions evaluate alternative indices nestedness. functions intended used together Null model communities used argument oecosimu analyse non-randomness results. Function nestedchecker gives number checkerboard units, 2x2 submatrices species occur different sites (Stone & Roberts 1990). Function nestedn0 implements nestedness measure N0 number absences sites richer pauperate site species occurs (Patterson & Atmar 1986). Function nesteddisc implements discrepancy index number ones shifted fill row ones table arranged species frequencies (Brualdi & Sanderson 1999). original definition arranges species (columns) frequencies, method handling tied frequencies. nesteddisc function tries order tied columns minimize discrepancy statistic rather slow, large number tied columns guarantee best ordering found (argument niter gives maximum number tried orders). case warning tied columns issued. Function nestedtemp finds matrix temperature defined sum “surprises” arranged matrix. arranged unsurprising matrix species within proportion given matrix fill upper left corner matrix, surprise absence presences diagonal distance fill line (Atmar & Patterson 1993). Function tries pack species sites low temperature (Rodríguez-Gironés & Santamaria 2006), iterative procedure, temperatures usually vary among runs. Function nestedtemp also plot method can display either incidences temperatures surprises. Matrix temperature rather vaguely described (Atmar & Patterson 1993), Rodríguez-Gironés & Santamaria (2006) explicit description used . However, results probably differ implementations, users cautious interpreting results. details calculations explained vignette Design decisions implementation can read using functions browseVignettes. Function nestedness bipartite package direct port BINMATNEST programme Rodríguez-Gironés & Santamaria (2006). Function nestednodf implements nestedness metric based overlap decreasing fill (Almeida-Neto et al., 2008). Two basic properties required matrix maximum degree nestedness according metric: (1) complete overlap 1's right left columns rows, (2) decreasing marginal totals pairs columns pairs rows. nestedness statistic evaluated separately columns (N columns) rows (N rows) combined whole matrix (NODF). set order = FALSE, statistic evaluated current matrix ordering allowing tests meaningful hypothesis matrix structure default ordering row column totals (breaking ties total abundances weighted = TRUE) (see Almeida-Neto et al. 2008). weighted = TRUE, function finds weighted version index (Almeida-Neto & Ulrich, 2011). However, requires quantitative null models adequate testing. Almeida-Neto & Ulrich (2011) say positive nestedness values first row/column higher second. condition, weighted analysis binary data always give zero nestedness. argument wbinary = TRUE, equality rows/columns also indicates nestedness, binary data give identical results weighted unweighted analysis. However, can also influence results weighted analysis results may differ Almeida-Neto & Ulrich (2011). Functions nestedbetasor nestedbetajac find multiple-site dissimilarities decompose components turnover nestedness following Baselga (2012); pairwise dissimilarities can found designdist. can seen decomposition beta diversity (see betadiver). Function nestedbetasor uses Sørensen dissimilarity turnover component Simpson dissimilarity (Baselga 2012), nestedbetajac uses analogous methods Jaccard index. functions return vector three items: turnover, nestedness sum multiple Sørensen Jaccard dissimilarity. last one total beta diversity (Baselga 2012). functions treat data presence/absence (binary) can used binary nullmodel. overall dissimilarity constant nullmodels fix species (column) frequencies (\"c0\"), components constant row columns also fixed (e.g., model \"quasiswap\"), functions meaningful null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"result returned nestedness function contains item called statistic, components differ among functions. functions constructed can handled oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Almeida-Neto, M., Guimarães, P., Guimarães, P.R., Loyola, R.D. & Ulrich, W. (2008). consistent metric nestedness analysis ecological systems: reconciling concept measurement. Oikos 117, 1227--1239. Almeida-Neto, M. & Ulrich, W. (2011). straightforward computational approach measuring nestedness using quantitative matrices. Env. Mod. Software 26, 173--178. Atmar, W. & Patterson, B.D. (1993). measurement order disorder distribution species fragmented habitat. Oecologia 96, 373--382. Baselga, . (2012). relationship species replacement, dissimilarity derived nestedness, nestedness. Global Ecol. Biogeogr. 21, 1223--1232. Brualdi, R.. & Sanderson, J.G. (1999). Nested species subsets, gaps, discrepancy. Oecologia 119, 256--264. Patterson, B.D. & Atmar, W. (1986). Nested subsets structure insular mammalian faunas archipelagos. Biol. J. Linnean Soc. 28, 65--82. Rodríguez-Gironés, M.. & Santamaria, L. (2006). new algorithm calculate nestedness temperature presence-absence matrices. J. Biogeogr. 33, 924--935. Stone, L. & Roberts, . (1990). checkerboard score species distributions. Oecologia 85, 74--79. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"Jari Oksanen Gustavo Carvalho (nestednodf).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nestedtemp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Nestedness Indices for Communities of Islands or Patches — nestedtemp","text":"","code":"data(sipoo) ## Matrix temperature out <- nestedtemp(sipoo) out #> nestedness temperature: 10.23506 #> with matrix fill 0.2233333 plot(out) plot(out, kind=\"incid\") ## Use oecosimu to assess the non-randomness of checker board units nestedchecker(sipoo) #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 oecosimu(sipoo, nestedchecker, \"quasiswap\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = #> \"quasiswap\") #> #> nullmodel method ‘quasiswap’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.55254 2715.3 2564.9 2713.0 2911.2 0.57 ## Another Null model and standardized checkerboard score oecosimu(sipoo, nestedchecker, \"r00\", statistic = \"C.score\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r00\", #> statistic = \"C.score\") #> #> nullmodel method ‘r00’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> C.score 2.2588 -28.215 9.2285 8.7036 9.2294 9.6735 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract the Number of Observations from a vegan Fit. — nobs.cca","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Extract number ‘observations’ vegan model fit.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"","code":"# S3 method for cca nobs(object, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"object fitted model object. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Function nobs generic R, vegan provides methods objects betadisper, cca related methods, CCorA, decorana, isomap, metaMDS, pcnm, procrustes, radfit, varpart wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"single number, normally integer, giving number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nobs.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract the Number of Observations from a vegan Fit. — nobs.cca","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Null Model and Simulation — nullmodel","title":"Null Model and Simulation — nullmodel","text":"nullmodel function creates object, can serve basis Null Model simulation via simulate method. update method updates nullmodel object without sampling (effective sequential algorithms). smbind binds together multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Null Model and Simulation — nullmodel","text":"","code":"nullmodel(x, method) # S3 method for nullmodel print(x, ...) # S3 method for nullmodel simulate(object, nsim = 1, seed = NULL, burnin = 0, thin = 1, ...) # S3 method for nullmodel update(object, nsim = 1, seed = NULL, ...) # S3 method for simmat print(x, ...) smbind(object, ..., MARGIN, strict = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Null Model and Simulation — nullmodel","text":"x community matrix. print method, object printed. method Character, specifying one null model algorithms listed help page commsim. can user supplied object class commsim. object object class nullmodel returned function nullmodel. case smbind simmat object returned update simulate methods. nsim Positive integer, number simulated matrices return. update method, number burnin steps made sequential algorithms update status input model object. seed object specifying random number generator initialized (\"seeded\"). Either NULL integer used call set.seed simulating matrices. set, value saved \"seed\" attribute returned value. default, NULL change random generator state, return .Random.seed \"seed\" attribute, see Value. burnin Nonnegative integer, specifying number steps discarded starting simulation. Active sequential null model algorithms. Ignored non-sequential null model algorithms. thin Positive integer, number simulation steps made returned matrix. Active sequential null model algorithms. Ignored non-sequential null model algorithms. MARGIN Integer, indicating dimension multiple simmat objects bound together smbind. 1: matrices stacked (row bound), 2: matrices column bound, 3: iterations combined. Needs length 1. dimensions expected match across objects. strict Logical, consistency time series attributes (\"start\", \"end\", \"thin\", number simulated matrices) simmat objects strictly enforced binding multiple objects together using smbind. Applies input objects based sequential null model algorithms. ... Additional arguments supplied algorithms. case smbind can contain multiple simmat objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Null Model and Simulation — nullmodel","text":"purpose nullmodel function create object, necessary statistics input matrix calculated . information reused, recalculated step simulation process done simulate method. simulate method carries simulation, simulated matrices stored array. sequential algorithms, method updates state input nullmodel object. Therefore, possible diagnostic tests returned simmat object, make simulations, use increased thinning value desired. update method makes burnin steps case sequential algorithms update status input model without attempt return matrices. non-sequential algorithms method nothing. update preferred way making burnin iterations without sampling. Alternatively, burnin can done via simulate method. convergence diagnostics, recommended use simulate method without burnin. input nullmodel object updated, samples can simulated desired without start process . See Examples. smbind function can used combine multiple simmat objects. comes handy null model simulations stratified sites (MARGIN = 1) species (MARGIN = 2), case multiple objects returned identical/consistent settings e.g. parallel computations (MARGIN = 3). Sanity checks made ensure combining multiple objects sensible, user's responsibility check independence simulated matrices null distribution converged case sequential null model algorithms. strict = FALSE setting can relax checks regarding start, end, thinning values sequential null models.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Null Model and Simulation — nullmodel","text":"function nullmodel returns object class nullmodel. set objects sharing environment: data: original matrix integer mode. nrow: number rows. ncol: number columns. rowSums: row sums. colSums: column sums. rowFreq: row frequencies (number nonzero cells). colFreq: column frequencies (number nonzero cells). totalSum: total sum. fill: number nonzero cells matrix. commsim: commsim object result method argument. state: current state permutations, matrix similar original. NULL non-sequential algorithms. iter: current number iterations sequential algorithms. NULL non-sequential algorithms. simulate method returns object class simmat. array simulated matrices (third dimension corresponding nsim argument). update method returns current state (last updated matrix) invisibly, update input object sequential algorithms. non sequential algorithms, returns NULL. smbind function returns object class simmat.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Null Model and Simulation — nullmodel","text":"Jari Oksanen Peter Solymos","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Null Model and Simulation — nullmodel","text":"Care must taken input matrix contains single row column. input invalid swapping hypergeometric distribution (calling r2dtable) based algorithms. also applies cases input stratified subsets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/nullmodel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Null Model and Simulation — nullmodel","text":"","code":"data(mite) x <- as.matrix(mite)[1:12, 21:30] ## non-sequential nullmodel (nm <- nullmodel(x, \"r00\")) #> An object of class “nullmodel” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> (sm <- simulate(nm, nsim=10)) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## sequential nullmodel (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 55, End = 100, Thin = 5 #> ## sequential nullmodel with burnin and extra updating (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> (sm1 <- simulate(nm, burnin=10, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 5, End = 50, Thin = 5 #> ## sequential nullmodel with separate initial burnin (nm <- nullmodel(x, \"swap\")) #> An object of class “nullmodel” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Iterations = 0 #> nm <- update(nm, nsim=10) (sm2 <- simulate(nm, nsim=10, thin=5)) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> Start = 15, End = 60, Thin = 5 #> ## combining multiple simmat objects ## stratification nm1 <- nullmodel(x[1:6,], \"r00\") sm1 <- simulate(nm1, nsim=10) nm2 <- nullmodel(x[7:12,], \"r00\") sm2 <- simulate(nm2, nsim=10) smbind(sm1, sm2, MARGIN=1) #> An object of class “simmat” #> ‘r00’ method (binary, non-sequential) #> 12 x 10 matrix #> Number of permuted matrices = 10 #> ## binding subsequent samples from sequential algorithms ## start, end, thin retained nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=10) sm1 <- simulate(nm, nsim=10, thin=5) sm2 <- simulate(nm, nsim=20, thin=5) sm3 <- simulate(nm, nsim=10, thin=5) smbind(sm3, sm2, sm1, MARGIN=3) #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 40 #> Start = 15, End = 210, Thin = 5 #> ## 'replicate' based usage which is similar to the output ## of 'parLapply' or 'mclapply' in the 'parallel' package ## start, end, thin are set, also noting number of chains smfun <- function(x, burnin, nsim, thin) { nm <- nullmodel(x, \"swap\") nm <- update(nm, nsim=burnin) simulate(nm, nsim=nsim, thin=thin) } smlist <- replicate(3, smfun(x, burnin=50, nsim=10, thin=5), simplify=FALSE) smbind(smlist, MARGIN=3) # Number of permuted matrices = 30 #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 12 x 10 matrix #> Number of permuted matrices = 30 #> Start = 55, End = 100, Thin = 5 (3 chains) #> if (FALSE) { ## parallel null model calculations library(parallel) if (.Platform$OS.type == \"unix\") { ## forking on Unix systems smlist <- mclapply(1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) smbind(smlist, MARGIN=3) } ## socket type cluster, works on all platforms cl <- makeCluster(3) clusterEvalQ(cl, library(vegan)) clusterExport(cl, c(\"smfun\", \"x\")) smlist <- parLapply(cl, 1:3, function(i) smfun(x, burnin=50, nsim=10, thin=5)) stopCluster(cl) smbind(smlist, MARGIN=3) }"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":null,"dir":"Reference","previous_headings":"","what":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function evaluates statistic vector statistics community evaluates significance series simulated random communities. approach used traditionally analysis nestedness, function general can used statistics evaluated simulated communities. Function oecosimu collects evaluates statistics. Null model communities described make.commsim permatfull/ permatswap, definition Null models nullmodel, nestedness statistics nestednodf (describes several alternative statistics, including nestedness temperature, \\(N0\\), checker board units, nestedness discrepancy NODF).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"oecosimu(comm, nestfun, method, nsimul = 99, burnin = 0, thin = 1, statistic = \"statistic\", alternative = c(\"two.sided\", \"less\", \"greater\"), batchsize = NA, parallel = getOption(\"mc.cores\"), ...) # S3 method for oecosimu as.ts(x, ...) # S3 method for oecosimu toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"comm Community data, Null model object generated nullmodel object class simmat (array permuted matrices simulate.nullmodel). comm community data, null model simulation method must specified. comm nullmodel, simulation method ignored, comm simmat object, arguments ignored except nestfun, statistic alternative. nestfun Function analysed. nestedness functions provided vegan (see nestedtemp), function can used accepts community first argument, returns either plain number vector result list item name defined argument statistic. See Examples defining functions. method Null model method: either name (character string) method defined make.commsim commsim function. argument ignored comm nullmodel simmat object. See Details Examples. nsimul Number simulated null communities (ignored comm simmat object). burnin Number null communities discarded proper analysis sequential methods (\"tswap\") (ignored non-sequential methods comm simmat object). thin Number discarded null communities two evaluations nestedness statistic sequential methods (ignored non-sequential methods comm simmat object). statistic name statistic returned nestfun. alternative character string specifying alternative hypothesis, must one \"two.sided\" (default), \"greater\" \"less\". Please note \\(p\\)-value two-sided test approximately two times higher corresponding one-sided test (\"greater\" \"less\" depending sign difference). batchsize Size Megabytes largest simulation object. larger structure produced, analysis broken internally batches. default NA analysis broken batches. See Details. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. define nestfun Windows needs R packages vegan permute, must set socket cluster call. x oecosimu result object. ... arguments functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu wrapper evaluates statistic using function given nestfun, simulates series null models based nullmodel, evaluates statistic null models. vegan packages contains nestedness functions described separately (nestedchecker, nesteddisc, nestedn0, nestedtemp, nestednodf), many functions can used long meaningful simulated communities. applicable function must return either statistic plain number vector, list element \"statistic\" (like chisq.test), item whose name given argument statistic. statistic can single number (like typical nestedness index), can vector. vector indices can used analyse site (row) species (column) properties, see treedive example. Raup-Crick index (raupcrick) gives example using dissimilarities. Null model type can given name (quoted character string) used define Null model make.commsim. include binary models described Wright et al. (1998), Jonsson (2001), Gotelli & Entsminger (2003), Miklós & Podani (2004), others. several quantitative Null models, discussed Hardy (2008), several unpublished (see make.commsim, permatfull, permatswap discussion). user can also define commsim function (see Examples). Function works first defining nullmodel given commsim, generating series simulated communities simulate.nullmodel. shortcut can used stages input can Community data (comm), Null model function (nestfun) number simulations (nsimul). nullmodel object number simulations, argument method ignored. three-dimensional array simulated communities generated simulate.nullmodel, arguments method nsimul ignored. last case allows analysing several statistics simulations. function first generates simulations given nullmodel analyses using nestfun. large data sets /large number simulations, generated objects can large, memory exhausted, analysis can become slow system can become unresponsive. simulation broken several smaller batches simulated nullmodel objective set batchsize avoid memory problems (see object.size estimating size current data set). parallel processing still increases memory needs. parallel processing used evaluating nestfun. main load may simulation nullmodel, parallel argument help . Function .ts transforms simulated results sequential methods time series ts object. allows using analytic tools time series studying sequences (see examples). Function toCoda transforms simulated results sequential methods \"mcmc\" object coda package. coda package provides functions analysis stationarity, adequacy sample size, autocorrelation, need burn-much sequential methods, summary results. Please consult documentation coda package. Function permustats provides support standard density, densityplot, qqnorm qqmath functions simulated values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Function oecosimu returns object class \"oecosimu\". result object items statistic oecosimu. statistic contains complete object returned nestfun original data. oecosimu component contains following items: statistic Observed values statistic. simulated Simulated values statistic. means Mean values statistic simulations. z Standardized effect sizes (SES, .k.. \\(z\\)-values) observed statistic based simulations. pval \\(P\\)-values statistic based simulations. alternative type testing given argument alternative. method method used nullmodel. isSeq TRUE method sequential.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Gotelli, N.J. & Entsminger, N.J. (2003). Swap algorithms null model analysis. Ecology 84, 532--535. Jonsson, B.G. (2001) null model randomization tests nestedness species assemblages. Oecologia 127, 309--313. Miklós, . & Podani, J. (2004). Randomization presence-absence matrices: comments new algorithms. Ecology 85, 86--92. Wright, D.H., Patterson, B.D., Mikkelson, G.M., Cutler, . & Atmar, W. (1998). comparative analysis nested subset patterns species composition. Oecologia 113, 1--20.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"Jari Oksanen Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"wonder name oecosimu, look journal names References (nestedtemp). internal structure function radically changed vegan 2.2-0 introduction commsim nullmodel deprecation commsimulator.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/oecosimu.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Evaluate Statistics with Null Models of Biological Communities — oecosimu","text":"","code":"## Use the first eigenvalue of correspondence analysis as an index ## of structure: a model for making your own functions. data(sipoo) ## Traditional nestedness statistics (number of checkerboard units) oecosimu(sipoo, nestedchecker, \"r0\") #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = nestedchecker, method = \"r0\") #> #> nullmodel method ‘r0’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 -18.742 8017.0 7469.0 8021.0 8450.5 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## sequential model, one-sided test, a vector statistic out <- oecosimu(sipoo, decorana, \"swap\", burnin=100, thin=10, statistic=\"evals\", alt = \"greater\") out #> oecosimu object #> #> Call: oecosimu(comm = sipoo, nestfun = decorana, method = \"swap\", #> burnin = 100, thin = 10, statistic = \"evals\", alternative = \"greater\") #> #> nullmodel method ‘swap’ with 99 simulations #> options: thin 10, burnin 100 #> alternative hypothesis: statistic is greater than simulated values #> #> #> Call: #> nestfun(veg = comm) #> #> Detrended correspondence analysis with 26 segments. #> Rescaling of axes with 4 iterations. #> Total inertia (scaled Chi-square): 2.4436 #> #> DCA1 DCA2 DCA3 DCA4 #> Eigenvalues 0.3822 0.2612 0.1668 0.08723 #> Additive Eigenvalues 0.3822 0.2609 0.1631 0.07650 #> Decorana values 0.4154 0.2465 0.1391 0.04992 #> Axis lengths 2.9197 2.5442 2.7546 1.78074 #> #> #> statistic SES mean 50% 95% Pr(sim.) #> DCA1 0.382249 2.04544 0.32960 0.33249 0.3677 0.01 ** #> DCA2 0.261208 1.77368 0.21549 0.21404 0.2587 0.05 * #> DCA3 0.166788 0.63257 0.15363 0.15405 0.1907 0.22 #> DCA4 0.087226 -1.69622 0.12533 0.12788 0.1636 0.96 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Inspect the swap sequence as a time series object plot(as.ts(out)) lag.plot(as.ts(out)) acf(as.ts(out)) ## Density plot densityplot(permustats(out), as.table = TRUE, layout = c(1,4)) ## Use quantitative null models to compare ## mean Bray-Curtis dissimilarities data(dune) meandist <- function(x) mean(vegdist(x, \"bray\")) mbc1 <- oecosimu(dune, meandist, \"r2dtable\") mbc1 #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = \"r2dtable\") #> #> nullmodel method ‘r2dtable’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 13.84 0.46601 0.44155 0.46756 0.4928 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Define your own null model as a 'commsim' function: shuffle cells ## in each row foo <- function(x, n, nr, nc, ...) { out <- array(0, c(nr, nc, n)) for (k in seq_len(n)) out[,,k] <- apply(x, 2, function(z) sample(z, length(z))) out } cf <- commsim(\"myshuffle\", foo, isSeq = FALSE, binary = FALSE, mode = \"double\") oecosimu(dune, meandist, cf) #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = meandist, method = cf) #> #> nullmodel method ‘myshuffle’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> statistic 0.64565 3.8862 0.63505 0.62995 0.63508 0.64 0.01 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Use pre-built null model nm <- simulate(nullmodel(sipoo, \"curveball\"), 99) oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘curveball’ with 99 simulations #> options: thin 1, burnin 0 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 1.4459 2710.6 2635.0 2723.0 2762.7 0.05 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Several chains of a sequential model -- this can be generalized ## for parallel processing (see ?smbind) nm <- replicate(5, simulate(nullmodel(sipoo, \"swap\"), 99, thin=10, burnin=100), simplify = FALSE) ## nm is now a list of nullmodels: use smbind to combine these into one ## nullmodel with several chains ## IGNORE_RDIFF_BEGIN nm <- smbind(nm, MARGIN = 3) nm #> An object of class “simmat” #> ‘swap’ method (binary, sequential) #> 18 x 50 matrix #> Number of permuted matrices = 495 #> Start = 110, End = 1090, Thin = 10 (5 chains) #> oecosimu(nm, nestedchecker) #> oecosimu object #> #> Call: oecosimu(comm = nm, nestfun = nestedchecker) #> #> nullmodel method ‘swap’ with 495 simulations #> options: thin 10, burnin 100, chains 5 #> alternative hypothesis: statistic is less or greater than simulated values #> #> Checkerboard Units : 2767 #> C-score (species mean): 2.258776 #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> checkerboards 2767 0.71842 2698.1 2572.0 2676.0 2904.7 0.4577 ## IGNORE_RDIFF_END ## After this you can use toCoda() and tools in the coda package to ## analyse the chains (these will show that thin, burnin and nsimul are ## all too low for real analysis)."},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":null,"dir":"Reference","previous_headings":"","what":"Support Functions for Drawing Vectors — ordiArrowTextXY","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Support functions assist drawing vectors (arrows) ordination plots. ordiArrowMul finds multiplier coordinates head vector occupy fill proportion plot region. ordiArrowTextXY finds coordinates locations labels drawn just beyond head vector.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"ordiArrowTextXY(x, labels, display, choices = c(1,2), rescale = TRUE, fill = 0.75, at = c(0,0), cex = NULL, ...) ordiArrowMul(x, at = c(0,0), fill = 0.75, display, choices = c(1,2), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"x R object, scores can determine suitable ordination scores object created envfit, two-column matrix coordinates arrow heads two plot axes. labels Change plotting labels. character vector labels label coordinates sought. supplied, determined row names x, scores(x, ...) required. either defined, suitable labels generated. display character string known scores one methods indicates type scores extract. fitting functions ordinary site scores linear combination scores (\"lc\") constrained ordination (cca, rda, dbrda). x created envfit display can set user takes value \"vectors\". Ignored x matrix. choices Axes plotted. rescale logical; coordinates extracted x rescaled fill fill proportion plot region? default always rescale coordinates usually desired objects x coordinates retrieved. supplying x 2-column matrix already rescaled, set FALSE. fill numeric; proportion plot fill span arrows. origin fitted arrows plot. plot arrows places origin, probably specify arrrow.mul. cex Character expansion text. ... Parameters passed scores, strwidth strheight.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowMul finds multiplier scale bunch arrows fill ordination plot, ordiArrowTextXY finds coordinates labels arrows. NB., ordiArrowTextXY draw labels; simply returns coordinates labels drawn use another function, text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"ordiArrowTextXY, 2-column matrix coordinates label centres coordinate system currently active plotting device. ordiArrowMul, length-1 vector containing scaling factor.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"Jari Oksanen, modifications Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiArrowTextXY.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Support Functions for Drawing Vectors — ordiArrowTextXY","text":"","code":"## Scale arrows by hand to fill 80% of the plot ## Biplot arrows by hand data(varespec, varechem) ord <- cca(varespec ~ Al + P + K, varechem) plot(ord, display = c(\"species\",\"sites\")) ## biplot scores bip <- scores(ord, choices = 1:2, display = \"bp\") ## scaling factor for arrows to fill 80% of plot (mul <- ordiArrowMul(bip, fill = 0.8)) #> [1] 2.092173 bip.scl <- bip * mul # Scale the biplot scores labs <- rownames(bip) # Arrow labels ## calculate coordinate of labels for arrows (bip.lab <- ordiArrowTextXY(bip.scl, rescale = FALSE, labels = labs)) #> [,1] [,2] #> Al 1.9098765 -0.3562415 #> P -0.9298005 -1.6652122 #> K -1.0069931 -0.3764923 ## draw arrows and text labels arrows(0, 0, bip.scl[,1], bip.scl[,2], length = 0.1) text(bip.lab, labels = labs) ## Handling of ordination objects directly mul2 <- ordiArrowMul(ord, display = \"bp\", fill = 0.8) stopifnot(all.equal(mul, mul2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Functions add arrows, line segments, regular grids points. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"ordiarrows(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, startmark, label = FALSE, length = 0.1, ...) ordisegments(ord, groups, levels, replicates, order.by, display = \"sites\", col = 1, show.groups, label = FALSE, ...) ordigrid(ord, levels, replicates, display = \"sites\", lty = c(1,1), col = c(1,1), lwd = c(1,1), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. levels, replicates Alternatively, regular groups can defined arguments levels replicates, levels gives number groups, replicates number successive items group. order.Order points increasing order variable within groups. Reverse sign variable decreasing ordering. display Item displayed. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names. ordiellipse, ordihull ordispider group name centroid object, ordiarrows start arrow, ordisegments ends. ordiellipse ordihull use standard text, others use ordilabel. startmark plotting character used mark first item. default use mark, instance, startmark = 1 draw circle. plotting characters, see pch points. col Colour lines, label borders startmark ordiarrows ordisegments. can vector recycled groups. ordigrid can vector length 2 used levels replicates. length Length edges arrow head (inches). lty, lwd Line type, line width used levels replicates ordigrid. ... Parameters passed graphical functions lines, segments, arrows, scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Function ordiarrows draws arrows ordisegments draws line segments successive items groups. Function ordigrid draws line segments within groups corresponding items among groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"functions add graphical items ordination graph: must draw graph first.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiarrows.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Arrows and Line Segments to Ordination Diagrams — ordiarrows","text":"","code":"example(pyrifos) #> #> pyrifs> data(pyrifos) #> #> pyrifs> ditch <- gl(12, 1, length=132) #> #> pyrifs> week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) #> #> pyrifs> dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) mod <- rda(pyrifos) plot(mod, type = \"n\") ## Annual succession by ditches, colour by dose ordiarrows(mod, ditch, label = TRUE, col = as.numeric(dose)) legend(\"topright\", levels(dose), lty=1, col=1:5, title=\"Dose\") ## Show only control and highest Pyrifos treatment plot(mod, type = \"n\") ordiarrows(mod, ditch, label = TRUE, show.groups = c(\"2\", \"3\", \"5\", \"11\")) ordiarrows(mod, ditch, label = TRUE, show = c(\"6\", \"9\"), col = 2) legend(\"topright\", c(\"Control\", \"Pyrifos 44\"), lty = 1, col = c(1,2))"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions add convex hulls, “spider” graphs, ellipses cluster dendrogram ordination diagrams. ordination diagrams can produced vegan plot.cca, plot.decorana ordiplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"ordihull(ord, groups, display = \"sites\", draw = c(\"lines\",\"polygon\", \"none\"), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd = NULL, ...) ordiellipse(ord, groups, display=\"sites\", kind = c(\"sd\",\"se\", \"ehull\"), conf, draw = c(\"lines\",\"polygon\", \"none\"), w = weights(ord, display), col = NULL, alpha = 127, show.groups, label = FALSE, border = NULL, lty = NULL, lwd=NULL, ...) ordibar(ord, groups, display = \"sites\", kind = c(\"sd\", \"se\"), conf, w = weights(ord, display), col = 1, show.groups, label = FALSE, lwd = NULL, length = 0, ...) ordispider(ord, groups, display=\"sites\", w = weights(ord, display), spiders = c(\"centroid\", \"median\"), show.groups, label = FALSE, col = NULL, lty = NULL, lwd = NULL, ...) ordicluster(ord, cluster, prune = 0, display = \"sites\", w = weights(ord, display), col = 1, draw = c(\"segments\", \"none\"), ...) # S3 method for ordihull summary(object, ...) # S3 method for ordiellipse summary(object, ...) ordiareatest(ord, groups, area = c(\"hull\", \"ellipse\"), kind = \"sd\", permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"ord ordination object ordiplot object. groups Factor giving groups graphical item drawn. display Item displayed. draw character; objects represented plot? ordihull ordiellipse use either lines polygon draw lines. ordicluster, line segments drawn using segments. suppress plotting, use \"none\". Graphical parameters passed . main difference polygons may filled non-transparent. none nothing drawn, function returns invisible plotting. col Colour hull ellipse lines (draw = \"lines\") fills (draw = \"polygon\") ordihull ordiellipse. draw = \"polygon\", colour bordering lines can set argument border polygon function. functions effect depends underlining functions argument passed . multiple values col specified used element names(table(groups)) (order), shorter vectors recycled. Function ordicluster groups, argument recycled points, colour connecting lines mixture point s cluster. alpha Transparency fill colour draw = \"polygon\" ordihull ordiellipse. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. show.groups Show given groups. can vector, TRUE want show items condition TRUE. argument makes possible use different colours line types groups. default show groups. label Label groups names centroid object. ordiellipse ordihull use standard text, others use ordilabel. w Weights used find average within group. Weights used automatically cca decorana results, unless undone user. w=NULL sets equal weights points. kind Draw standard deviations points (sd), standard errors (se) ellipsoid hulls enclose points group (ehull). conf Confidence limit ellipses, e.g. 0.95. given, corresponding sd se multiplied corresponding value found Chi-squared distribution 2df. spiders centres spider bodies calculated either centroids (averages) spatial medians. cluster Result hierarchic cluster analysis, hclust agnes. prune Number upper level hierarchies removed dendrogram. prune \\(>0\\), dendrogram disconnected. object result object ordihull ordiellipse. result invisible, can saved, used summaries (areas etc. hulls ellipses). area Evaluate area convex hulls ordihull, ellipses ordiellipse. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. lty, lwd, border Vectors parameters can supplied applied (appropriate) element names(table(groups)) (order). Shorter vectors recycled. length Width (inches) small (“caps”) ends bar segment (passed arrows). ... Parameters passed graphical functions scores select axes scaling etc.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Function ordihull draws lines polygons convex hulls found function chull encircling items groups. Function ordiellipse draws lines polygons ellipses groups. function can either draw standard deviation points (kind=\"sd\") standard error (weighted) centroids (kind=\"se\"), (weighted) correlation defines direction principal axis ellipse. kind = \"se\" used together argument conf, ellipses show confidence regions locations group centroids. kind=\"ehull\" function draws ellipse encloses points group using ellipsoidhull (cluster package). Function ordibar draws crossed “error bars” using either either standard deviation point scores standard error (weighted) average scores. principal axes corresponding ordiellipse, found principal component analysis (weighted) covariance matrix. Functions ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups. Argument kind can used select kind ellipse, effect convex hulls. Function ordispider draws ‘spider’ diagram point connected group centroid segments. Weighted centroids used correspondence analysis methods cca decorana user gives weights call. ordispider called cca rda result without groups argument, function connects ‘WA’ scores corresponding ‘LC’ score. argument (invisible) ordihull object, function connect points hull centroid. Function ordicluster overlays cluster dendrogram onto ordination. needs result hierarchic clustering hclust agnes, similar structure. Function ordicluster connects cluster centroids line segments. Function uses centroids points clusters, therefore similar average linkage methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"functions add graphical items ordination graph: must draw graph first. draw line segments, grids arrows, see ordisegments, ordigrid andordiarrows.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Functions ordihull, ordiellipse ordispider return invisible plotting structure. Function ordispider return coordinates point connected (centroids ‘LC’ scores). Function ordihull ordiellipse return invisibly object summary method returns coordinates centroids areas hulls ellipses. Function ordiareatest studies one-sided hypothesis areas smaller randomized groups.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordihull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Groups or Factor Levels in Ordination Diagrams — ordihull","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ Management, dune.env) plot(mod, type=\"n\", scaling = \"symmetric\") ## Catch the invisible result of ordihull... pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", label = TRUE)) ## ... and find centres and areas of the hulls summary(pl) #> BF HF NM SF #> CCA1 0.2917476 0.36826105 -1.3505642 0.2762936 #> CCA2 0.8632208 0.09419919 0.2681515 -0.8139398 #> Area 0.1951715 0.59943363 1.7398193 1.0144372 ## use more colours and add ellipsoid hulls plot(mod, type = \"n\") pl <- with(dune.env, ordihull(mod, Management, scaling = \"symmetric\", col = 1:4, draw=\"polygon\", label =TRUE)) with(dune.env, ordiellipse(mod, Management, scaling = \"symmetric\", kind = \"ehull\", col = 1:4, lwd=3)) ## ordispider to connect WA and LC scores plot(mod, dis=c(\"wa\",\"lc\"), type=\"p\") ordispider(mod) ## Other types of plots plot(mod, type = \"p\", display=\"sites\") cl <- hclust(vegdist(dune)) ordicluster(mod, cl, prune=3, col = cutree(cl, 4)) ## confidence ellipse: location of the class centroids plot(mod, type=\"n\", display = \"sites\") with(dune.env, text(mod, display=\"sites\", labels = as.character(Management), col=as.numeric(Management))) pl <- with(dune.env, ordiellipse(mod, Management, kind=\"se\", conf=0.95, lwd=2, draw = \"polygon\", col=1:4, border=1:4, alpha=63)) summary(pl) #> BF HF NM SF #> CCA1 0.4312652 0.5583211 -1.87848340 0.5601499 #> CCA2 1.3273917 0.6373120 -0.05503211 -1.3859924 #> Area 1.4559842 1.3806668 2.73667419 1.5559135 ## add confidence bars with(dune.env, ordibar(mod, Management, kind=\"se\", conf=0.95, lwd=2, col=1:4, label=TRUE))"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Function ordilabel similar text, text opaque label. can help crowded ordination plots: still see text labels, least uppermost readable. Argument priority helps make important labels visible.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"ordilabel(x, display, labels, choices = c(1, 2), priority, select, cex = 0.8, fill = \"white\", border = NULL, col = NULL, xpd = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"x ordination object object known scores. display Kind scores displayed (passed scores). labels Optional text used plots. given, text found ordination object. choices Axes shown (passed scores). priority Vector length number labels. items high priority plotted uppermost. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. cex Character expansion text (passed text). fill Background colour labels (col argument polygon). border colour visibility border label defined polygon. col Text colour. Default NULL give value border par(\"fg\") border NULL. xpd Draw labels also outside plot region (see par). ... arguments (passed text).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"function may useful crowded ordination plots, particular together argument priority. see text labels, least readable. alternatives crowded plots identify.ordiplot, orditorp orditkplot ( vegan3d package).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordilabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text on Non-transparent Label to an Ordination Plot. — ordilabel","text":"","code":"data(dune) ord <- cca(dune) plot(ord, type = \"n\") ordilabel(ord, dis=\"sites\", cex=1.2, font=3, fill=\"hotpink\", col=\"blue\") ## You may prefer separate plots, but here species as well ordilabel(ord, dis=\"sp\", font=2, priority=colSums(dune))"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Alternative plot and identify Functions for Ordination — ordiplot","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot alternative plotting function can worked vegan ordination result many non-vegan results. addition, plot functions vegan ordinations return invisibly \"ordiplot\" result object, allows using ordiplot support functions result: identify can used add labels selected site, species constraint points, points text can add elements plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"ordiplot(ord, choices = c(1, 2), type=\"points\", display, xlim, ylim, cex = 0.7, ...) # S3 method for ordiplot identify(x, what, labels, ...) # S3 method for ordiplot points(x, what, select, arrows = FALSE, ...) # S3 method for ordiplot text(x, what, labels, select, arrows = FALSE, length = 0.05, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"ord result ordination. choices Axes shown. type type graph may \"points\", \"text\" \"none\" ordination method. display Display \"sites\" \"species\". default methods display , cca, rda, dbrda capscale plot.cca. xlim, ylim x y limits (min,max) plot. cex Character expansion factor points text. ... graphical parameters. x result object ordiplot. Items identified ordination plot. types depend kind plot used. methods know sites species, functions cca rda know addition constraints (LC scores), centroids biplot, plot.procrustes ordination plot heads points. labels Optional text used labels. Row names used missing. arrows Draw arrows origin. always TRUE biplot scores value ignored. Setting TRUE draw arrows type scores. allows, e.g, using biplot arrows species. arrow head value scores, possible text moved outwards. length Length arrow heads (see arrows). select Items displayed. can either logical vector TRUE displayed items vector indices displayed items.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot draws ordination diagram using black circles sites red crosses species. returns invisibly object class ordiplot can used identify.ordiplot label selected sites species, constraints cca rda. function can handle output several alternative ordination methods. cca, rda decorana uses plot method option type = \"points\". addition, plot functions methods return invisibly ordiplot object can used identify.ordiplot label points. ordinations relies scores extract scores. full user control plots, best call ordiplot type = \"none\" save result, add sites species using points.ordiplot text.ordiplot pass arguments corresponding default graphical functions. functions can chained pipes allows alternative intuitive way building plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Function ordiplot returns invisibly object class ordiplot used scores. general, vegan plot functions ordination results also return invisible ordiplot object. plot(..., type = \"n\") used originally, plot empty, items can added invisible object. Functions points text return input object without modification, allows chaining commands pipes. Function identify.ordiplot uses object label point.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"purpose functions provide similar functionality plot, plotid specid methods library labdsv. functions somewhat limited parametrization, can call directly standard identify plot functions better user control.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Alternative plot and identify Functions for Ordination — ordiplot","text":"","code":"## Draw a plot for a non-vegan ordination (cmdscale). data(dune) dune.dis <- vegdist(wisconsin(dune)) dune.mds <- cmdscale(dune.dis, eig = TRUE) dune.mds$species <- wascores(dune.mds$points, dune, expand = TRUE) pl <- ordiplot(dune.mds, type = \"none\") points(pl, \"sites\", pch=21, col=\"red\", bg=\"yellow\") text(pl, \"species\", col=\"blue\", cex=0.9) if (FALSE) { ## same plot using pipes (|>) ordiplot(dune.mds, type=\"n\") |> points(\"sites\", pch=21, col=\"red\", bg=\"yellow\") |> text(\"species\", col=\"blue\", cex=0.9) ## Some people think that species should be shown with arrows in PCA. ## Other ordination methods also return an invisible ordiplot object and ## we can use pipes to draw those arrows. mod <- rda(dune) plot(mod, type=\"n\") |> points(\"sites\", pch=16, col=\"red\") |> text(\"species\", arrows = TRUE, length=0.05, col=\"blue\") } ## Default plot of the previous using identify to label selected points if (FALSE) { pl <- ordiplot(dune.mds) identify(pl, \"spec\")}"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":null,"dir":"Reference","previous_headings":"","what":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function ordipointlabel produces ordination plots points text label points. points exact location given ordination, function tries optimize location text labels minimize overplotting text. function may useful moderately crowded ordination plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"ordipointlabel(x, display = c(\"sites\", \"species\"), choices = c(1, 2), col = c(1, 2), pch = c(\"o\", \"+\"), font = c(1, 1), cex = c(0.8, 0.8), add = FALSE, select, ...) # S3 method for ordipointlabel plot(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"x ordipointlabel() result object ordination function. plot.ordipointlabel object resulting call ordipointlabel(). display Scores displayed plot. choices Axes shown. col, pch, font, cex Colours, point types, font style character expansion kind scores displayed plot. vectors length number items display. add Add existing plot. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. select used single set scores plotted (.e. length(display) == 1), otherwise ignored warning issued. logical vector used, must length scores plotted. ... arguments passed points text.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function uses simulated annealing (optim, method = \"SANN\") optimize location text labels points. eight possible locations: , , sides corners. weak preference text right point, weak avoidance corner positions. exact locations goodness solution varies runs, guarantee finding global optimum. optimization can take long time difficult cases high number potential overlaps. Several sets scores can displayed one plot. function modelled pointLabel maptools package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function returns invisibly object class ordipointlabel items xy coordinates points, labels coordinates labels, items pch, cex font graphical parameters point label. addition, returns result optim attribute \"optim\". unit overlap area character \"m\", variable cex smallest alternative. plot method based orditkplot alter reset graphical parameters via par. result object ordipointlabel inherits orditkplot vegan3d package, may possible edit result object orditkplot, good results necessary points span whole horizontal axis without empty margins.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"function designed ordination graphics, optimization works properly plots isometric aspect ratio.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordipointlabel.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Ordination Plots with Points and Optimized Locations for Text — ordipointlabel","text":"","code":"data(dune) ord <- cca(dune) plt <- ordipointlabel(ord) ## set scaling - should be no warnings! ordipointlabel(ord, scaling = \"sites\") ## plot then add plot(ord, scaling = \"symmetric\", type = \"n\") ordipointlabel(ord, display = \"species\", scaling = \"symm\", add = TRUE) ordipointlabel(ord, display = \"sites\", scaling = \"symm\", add = TRUE) ## redraw plot without rerunning SANN optimisation plot(plt)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":null,"dir":"Reference","previous_headings":"","what":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function provides plot.lm style diagnostic plots results constrained ordination cca, rda, dbrda capscale. Normally need plots, ordination descriptive make assumptions distribution residuals. However, permute residuals significance tests (anova.cca), may interested inspecting residuals really exchangeable independent fitted values.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"ordiresids(x, kind = c(\"residuals\", \"scale\", \"qqmath\"), residuals = \"working\", type = c(\"p\", \"smooth\", \"g\"), formula, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"x Ordination result cca, rda, dbrda, capscale. kind type plot: \"residuals\" plot residuals fitted values, \"scale\" square root absolute residuals fitted values, \"qqmath\" residuals expected distribution (defaults qnorm), unless defined differently formula argument. residuals kind residuals fitted values, alternatives \"working\", \"response\", \"standardized\" \"studentized\" (see Details). type type plot. argument passed lattice functions. formula Formula override default plot. formula can contain items Fitted, Residuals, Species Sites (provided names species sites available ordination result). ... arguments passed lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"default plots similar plot.lm, use Lattice functions xyplot qqmath. alternatives default formulae can replaced user. elements available formula groups argument Fitted, Residuals, Species Sites. residuals = \"response\" residuals = \"working\" fitted values residuals found functions fitted.cca residuals.cca. residuals = \"standardized\" residuals found rstandard.cca, residuals = \"studentized\" found rstudent.cca, cases fitted values standardized sigma.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"function returns Lattice object can displayed plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordiresids.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plots of Residuals and Fitted Values for Constrained Ordination — ordiresids","text":"","code":"data(varespec) data(varechem) mod <- cca(varespec ~ Al + P + K, varechem) ordiresids(mod) ordiresids(mod, formula = Residuals ~ Fitted | Species, residuals=\"standard\", cex = 0.5)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":null,"dir":"Reference","previous_headings":"","what":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Automatic stepwise model building constrained ordination methods (cca, rda, dbrda, capscale). function ordistep modelled step can forward, backward stepwise model selection using permutation tests. Function ordiR2step performs forward model choice solely adjusted \\(R^2\\) \\(P\\)-value.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"ordistep(object, scope, direction = c(\"both\", \"backward\", \"forward\"), Pin = 0.05, Pout = 0.1, permutations = how(nperm = 199), steps = 50, trace = TRUE, ...) ordiR2step(object, scope, Pin = 0.05, R2scope = TRUE, permutations = how(nperm = 499), trace = TRUE, R2permutations = 1000, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"object ordistep, ordination object inheriting cca rda. scope Defines range models examined stepwise search. can list containing components upper lower, formulae. single item, interpreted target scope, depending direction. direction \"forward\", single item interpreted upper scope formula input object lower scope. See step details. ordiR2step, defines upper scope; can also ordination object model extracted. direction mode stepwise search, can one \"\", \"backward\", \"forward\", default \"\". scope argument missing, default direction \"backward\" ordistep ( ordiR2step argument, works forward). Pin, Pout Limits permutation \\(P\\)-values adding (Pin) term model, dropping (Pout) model. Term added \\(P \\le\\) Pin, removed \\(P >\\) Pout. R2scope Use adjusted \\(R^2\\) stopping criterion: models lower adjusted \\(R^2\\) scope accepted. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. passed anova.cca: see details. steps Maximum number iteration steps dropping adding terms. trace positive, information printed model building. Larger values may give information. R2permutations Number permutations used estimation adjusted \\(R^2\\) cca using RsquareAdj. ... additional arguments add1.cca drop1.cca.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"basic functions model choice constrained ordination add1.cca drop1.cca. functions, ordination models can chosen standard R function step bases term choice AIC. AIC-like statistics ordination provided functions deviance.cca extractAIC.cca ( similar functions rda). Actually, constrained ordination methods AIC, therefore step may trusted. function provides alternative using permutation \\(P\\)-values. Function ordistep defines model, scope models considered, direction procedure similarly step. function alternates drop add steps stops model changed one step. - + signs summary table indicate stage performed. often sensible Pout \\(>\\) Pin stepwise models avoid cyclic adds drops single terms. Function ordiR2step builds model forward maximizes adjusted \\(R^2\\) (function RsquareAdj) every step, stopping adjusted \\(R^2\\) starts decrease, adjusted \\(R^2\\) scope exceeded, selected permutation \\(P\\)-value exceeded (Blanchet et al. 2008). second criterion ignored option R2scope = FALSE, third criterion can ignored setting Pin = 1 (higher). function used adjusted \\(R^2\\) calculated. number predictors higher number observations, adjusted \\(R^2\\) also unavailable. models can analysed R2scope = FALSE, variable selection stop models become overfitted adjusted \\(R^2\\) calculated, adjusted \\(R^2\\) reported zero. \\(R^2\\) cca based simulations (see RsquareAdj) different runs ordiR2step can give different results. Functions ordistep (based \\(P\\) values) ordiR2step (based adjusted \\(R^2\\) hence eigenvalues) can select variables different order.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Functions return selected model one additional component, anova, contains brief information steps taken. can suppress voluminous output model building setting trace = FALSE, find summary model history anova item.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Blanchet, F. G., Legendre, P. & Borcard, D. (2008) Forward selection explanatory variables. Ecology 89, 2623--2632.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordistep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Choose a Model by Permutation Tests in Constrained Ordination — ordistep","text":"","code":"## See add1.cca for another example ### Dune data data(dune) data(dune.env) mod0 <- rda(dune ~ 1, dune.env) # Model with intercept only mod1 <- rda(dune ~ ., dune.env) # Model with all explanatory variables ## With scope present, the default direction is \"both\" mod <- ordistep(mod0, scope = formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.035 * #> + Use 2 91.032 1.1741 0.300 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.085 . #> + A1 1 87.424 1.2965 0.170 #> + Use 2 88.284 1.0510 0.355 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.010 ** #> - Management 3 87.707 2.1769 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.535 #> + Use 2 86.842 0.8027 0.785 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## summary table of steps mod$anova #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 85.567 1.9764 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Example of ordistep, forward ordistep(mod0, scope = formula(mod1), direction=\"forward\") #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.010 ** #> + A1 1 89.591 1.9217 0.020 * #> + Use 2 91.032 1.1741 0.290 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.055 . #> + A1 1 87.424 1.2965 0.225 #> + Use 2 88.284 1.0510 0.410 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.300 #> + A1 1 86.220 0.8359 0.595 #> + Use 2 86.842 0.8027 0.730 #> #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> ## Example of ordiR2step (always forward) ## stops because R2 of 'mod1' exceeded ordiR2step(mod0, mod1) #> Step: R2.adj= 0 #> Call: dune ~ 1 #> #> R2.adjusted #> 0.32508817 #> + Management 0.22512409 #> + Moisture 0.20050225 #> + Manure 0.16723149 #> + A1 0.04626579 #> + Use 0.01799755 #> 0.00000000 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.84 0.002 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: R2.adj= 0.2251241 #> Call: dune ~ Management #> #> R2.adjusted #> + Moisture 0.3450334 #> 0.3250882 #> + Manure 0.2779515 #> + A1 0.2392216 #> + Use 0.2300349 #> 0.2251241 #> #> Call: rda(formula = dune ~ Management, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 29.2307 0.3475 3 #> Unconstrained 54.8930 0.6525 16 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 #> 14.865 10.690 3.675 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 #> 15.270 8.428 6.899 5.675 3.988 3.121 2.588 2.380 1.818 1.376 0.995 #> PC12 PC13 PC14 PC15 PC16 #> 0.785 0.661 0.467 0.283 0.159 #>"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":null,"dir":"Reference","previous_headings":"","what":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface given variable plots result ordination diagram.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"# S3 method for default ordisurf(x, y, choices = c(1, 2), knots = 10, family = \"gaussian\", col = \"red\", isotropic = TRUE, thinplate = TRUE, bs = \"tp\", fx = FALSE, add = FALSE, display = \"sites\", w = weights(x, display), main, nlevels = 10, levels, npoints = 31, labcex = 0.6, bubble = FALSE, cex = 1, select = TRUE, method = \"REML\", gamma = 1, plot = TRUE, lwd.cl = par(\"lwd\"), ...) # S3 method for formula ordisurf(formula, data, ...) # S3 method for ordisurf calibrate(object, newdata, ...) # S3 method for ordisurf plot(x, what = c(\"contour\",\"persp\",\"gam\"), add = FALSE, bubble = FALSE, col = \"red\", cex = 1, nlevels = 10, levels, labcex = 0.6, lwd.cl = par(\"lwd\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"x ordisurf ordination configuration, either matrix result known scores. plot.ordisurf object class \"ordisurf\" returned ordisurf. y Variable plotted / modelled function ordination scores. choices Ordination axes. knots Number initial knots gam (one degrees freedom). knots = 0 knots = 1 function fit linear trend surface, knots = 2 function fit quadratic trend surface instead smooth surface. vector length 2 allowed isotropic = FALSE, first second elements knots referring first second ordination dimensions (indicated choices) respectively. family Error distribution gam. col Colour contours. isotropic, thinplate Fit isotropic smooth surface (.e. smoothness ordination dimensions) via gam. Use thinplate deprecated removed future version package. bs two letter character string indicating smoothing basis use. (e.g. \"tp\" thin plate regression spline, \"cr\" cubic regression spline). One c(\"tp\", \"ts\", \"cr\", \"cs\", \"ds\", \"ps\", \"ad\"). See smooth.terms view refer . default use thin plate splines: bs = \"tp\". fx indicates whether smoothers fixed degree freedom regression splines (fx = FALSE) penalised regression splines (fx = TRUE). Can vector length 2 anisotropic surfaces (isotropic = FALSE). make sense use fx = TRUE select = TRUE error . warning issued specify fx = TRUE forget use select = FALSE though fitting continues using select = FALSE. add Add contours existing diagram draw new plot? display Type scores known scores: typically \"sites\" ordinary site scores \"lc\" linear combination scores. w Prior weights data. Concerns mainly cca decorana results nonconstant weights. main main title plot, default name plotted variable new plot. nlevels, levels Either vector levels contours drawn, suggested number contours nlevels levels supplied. npoints numeric; number locations evaluate fitted surface. represents number locations dimension. labcex Label size contours. Setting zero suppress labels. bubble Use “bubble plot” points, vary point diameter value plotted variable. bubble numeric, value used maximum symbol size ( cex), bubble = TRUE, value cex gives maximum. minimum size always cex = 0.4. option effect add = FALSE. cex Character expansion plotting symbols. select Logical; specify gam argument \"select\". TRUE gam can add extra penalty term can penalized zero. means smoothing parameter estimation part fitting can completely remove terms model. corresponding smoothing parameter estimated zero extra penalty effect. method character; smoothing parameter estimation method. Options allowed : \"GCV.Cp\" uses GCV models unknown scale parameter Mallows' Cp/UBRE/AIC models known scale; \"GACV.Cp\" \"GCV.Cp\" uses GACV (Generalised Approximate CV) instead GCV; \"REML\" \"ML\" use restricted maximum likelihood maximum likelihood estimation known unknown scale; \"P-REML\" \"P-ML\" use REML ML estimation use Pearson estimate scale. gamma Multiplier inflate model degrees freedom GCV UBRE/AIC score . effectively places extra penalty complex models. oft-used value gamma = 1.4. plot logical; plotting done ordisurf? Useful want fitted response surface model. lwd.cl numeric; lwd (line width) parameter use drawing contour lines. formula, data Alternative definition fitted model x ~ y, left-hand side ordination x right-hand side single fitted continuous variable y. variable y must working environment data frame environment given data. arguments passed default method. object ordisurf result object. newdata Coordinates two-dimensional ordination new points. character; type plot produce. \"contour\" produces contour plot response surface, see contour details. \"persp\" produces perspective plot , see persp details. \"gam\" plots fitted GAM model, object inherits class \"gam\" returned ordisurf, see plot.gam. ... parameters passed scores, graphical functions. See Note exceptions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Function ordisurf fits smooth surface using penalised splines (Wood 2003) gam, uses predict.gam find fitted values regular grid. smooth surface can fitted extra penalty allows entire smoother penalized back 0 degrees freedom, effectively removing term model (see Marra & Wood, 2011). addition extra penalty invoked setting argument select TRUE. alternative use spline basis includes shrinkage (bs = \"ts\" bs = \"cs\"). ordisurf() exposes large number options gam specifying basis functions used surface. stray defaults, read Notes section relevant documentation s smooth.terms. function plots fitted contours convex hull data points either existing ordination diagram draws new plot. select = TRUE smooth effectively penalised model, contours plotted. gam determines degree smoothness fitted response surface model fitting, unless fx = TRUE. Argument method controls gam performs smoothness selection. See gam details available options. Using \"REML\" \"ML\" yields p-values smooths best coverage properties things matter . function uses scores extract ordination scores, x can result object known function. user can supply vector prior weights w. ordination object weights, used. practise means row totals used weights cca decorana results. like , want give equal weights sites, set w = NULL. behaviour consistent envfit. complete accordance constrained cca, set display = \"lc\". Function calibrate returns fitted values response variable. newdata must coordinates points fitted values desired. function based predict.gam pass extra arguments function.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"ordisurf usually called side effect drawing contour plot. function returns result object class \"ordisurf\" inherits gam used internally fit surface, adds item grid contains data grid surface. item grid elements x y vectors axis coordinates, element z matrix fitted values contour. values outside convex hull observed points indicated NA z. gam component result can used analysis like predicting new values (see predict.gam).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Dave Roberts, Jari Oksanen Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"default use isotropic smoother via s employing thin plate regression splines (bs = \"tp\"). make sense ordination equal smoothing directions rotation invariant. However, different degrees smoothness along dimensions required, anisotropic smooth surface may applicable. can achieved use isotropic = FALSE, wherein surface fitted via tensor product smoother via te (unless bs = \"ad\", case separate splines dimension fitted using s). Cubic regression splines P splines can used isotropic = FALSE. Adaptive smooths (bs = \"ad\"), especially two dimensions, require large number observations; without many hundreds observations, default complexities smoother exceed number observations fitting fail. get old behaviour ordisurf use select = FALSE, method = \"GCV.Cp\", fx = FALSE, bs = \"tp\". latter two options current defaults. Graphical arguments supplied plot.ordisurf passed underlying plotting functions, contour, persp, plot.gam. exception arguments col cex can currently passed plot.gam bug way function evaluates arguments arranging plot. work-around call plot.gam directly result call ordisurf. See Examples illustration .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"warning","dir":"Reference","previous_headings":"","what":"Warning","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"fitted GAM regression model usual assumptions models. particular note assumption independence residuals. observations independent (e.g. repeat measures set objects, experimental design, inter alia) trust p-values GAM output. need control (.e. add additional fixed effects model, use complex smoothers), extract ordination scores using scores function generate gam call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"Marra, G.P & Wood, S.N. (2011) Practical variable selection generalized additive models. Comput. Stat. Data Analysis 55, 2372--2387. Wood, S.N. (2003) Thin plate regression splines. J. R. Statist. Soc. B 65, 95--114.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordisurf.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Fit and Plot Smooth Surfaces of Variables on Ordination. — ordisurf","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(varespec) vare.mds <- monoMDS(vare.dist) ## IGNORE_RDIFF_BEGIN ordisurf(vare.mds ~ Baresoil, varechem, bubble = 5) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## as above but without the extra penalties on smooth terms, ## and using GCV smoothness selection (old behaviour of `ordisurf()`): ordisurf(vare.mds ~ Baresoil, varechem, col = \"blue\", add = TRUE, select = FALSE, method = \"GCV.Cp\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.34 total = 8.34 #> #> GCV score: 154.2897 ## Cover of Cladina arbuscula fit <- ordisurf(vare.mds ~ Cladarbu, varespec, family=quasipoisson) ## Get fitted values calibrate(fit) #> 18 15 24 27 23 19 22 #> 21.6245333 8.0344376 3.8614798 2.4595115 6.3958275 5.4260377 6.7158154 #> 16 28 13 14 20 25 7 #> 11.6759768 0.8041457 31.1726268 16.1942820 9.5897933 5.4219853 29.8214589 #> 5 6 3 4 2 9 12 #> 22.8500283 29.9936019 7.1653707 15.5309226 2.8467735 0.9051372 3.5541639 #> 10 11 21 #> 1.3099294 10.7783687 0.9177922 ## Variable selection via additional shrinkage penalties ## This allows non-significant smooths to be selected out ## of the model not just to a linear surface. There are 2 ## options available: ## - option 1: `select = TRUE` --- the *default* ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", select = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94112 ## - option 2: use a basis with shrinkage ordisurf(vare.mds ~ Baresoil, varechem, method = \"REML\", bs = \"ts\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ts\", fx = FALSE) #> #> Estimated degrees of freedom: #> 6.26 total = 7.26 #> #> REML score: 98.83943 ## or bs = \"cs\" with `isotropic = FALSE` ## IGNORE_RDIFF_END ## Plot method plot(fit, what = \"contour\") ## Plotting the \"gam\" object plot(fit, what = \"gam\") ## 'col' and 'cex' not passed on ## or via plot.gam directly library(mgcv) #> Loading required package: nlme #> This is mgcv 1.9-1. For overview type 'help(\"mgcv-package\")'. plot.gam(fit, cex = 2, pch = 1, col = \"blue\") ## 'col' effects all objects drawn... ### controlling the basis functions used ## Use Duchon splines ordisurf(vare.mds ~ Baresoil, varechem, bs = \"ds\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"ds\", fx = FALSE) #> #> Estimated degrees of freedom: #> 5.33 total = 6.33 #> #> REML score: 93.94117 ## A fixed degrees of freedom smooth, must use 'select = FALSE' ordisurf(vare.mds ~ Baresoil, varechem, knots = 4, fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 4, bs = \"tp\", fx = TRUE) #> #> Estimated degrees of freedom: #> 3 total = 4 #> #> REML score: 85.55126 ## An anisotropic smoother with cubic regression spline bases ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cr\", knots = 4) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(4, 4), bs = c(\"cr\", \"cr\"), fx = c(FALSE, #> FALSE)) #> #> Estimated degrees of freedom: #> 3.08 total = 4.08 #> #> REML score: 93.02782 ## An anisotropic smoother with cubic regression spline with ## shrinkage bases & different degrees of freedom in each dimension ordisurf(vare.mds ~ Baresoil, varechem, isotropic = FALSE, bs = \"cs\", knots = c(3,4), fx = TRUE, select = FALSE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ te(x1, x2, k = c(3, 4), bs = c(\"cs\", \"cs\"), fx = c(TRUE, #> TRUE)) #> #> Estimated degrees of freedom: #> 11 total = 12 #> #> REML score: 42.9834"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Text or Points to Ordination Plots — orditorp","title":"Add Text or Points to Ordination Plots — orditorp","text":"function adds text points ordination plots. Text used can done without overwriting text labels, points used otherwise. function can help reducing clutter ordination graphics, manual editing may still necessary.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"orditorp(x, display, labels, choices = c(1, 2), priority, select, cex = 0.7, pcex, col = par(\"col\"), pcol, pch = par(\"pch\"), air = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Text or Points to Ordination Plots — orditorp","text":"x result object ordination ordiplot result. display Items displayed plot. one alternative allowed. Typically \"sites\" \"species\". labels Optional text used labels. Row names used missing. choices Axes shown. priority Text used items higher priority labels overlap. vector length number items plotted. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. logical vector used, must length scores plotted. cex, pcex Text point sizes, see plot.default.. col, pcol Text point colours, see plot.default. pch Plotting character, see points. air Amount empty space text labels. Values <1 allow overlapping text. ... arguments scores (various methods), text points.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Text or Points to Ordination Plots — orditorp","text":"Function orditorp add either text points existing plot. items high priority added first text used can done without overwriting previous labels,points used otherwise. priority missing, labels added outskirts centre. Function orditorp can used ordination results, plotting results ordiplot ordination plot functions (plot.cca, plot.decorana, plot.metaMDS). Arguments can passed relevant scores method ordination object (x) drawn. See relevant scores help page arguments can used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add Text or Points to Ordination Plots — orditorp","text":"function returns invisibly logical vector TRUE means item labelled text FALSE means marked point. returned vector can used select argument ordination text points functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add Text or Points to Ordination Plots — orditorp","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/orditorp.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Text or Points to Ordination Plots — orditorp","text":"","code":"## A cluttered ordination plot : data(BCI) mod <- cca(BCI) plot(mod, dis=\"sp\", type=\"t\") # Now with orditorp and abbreviated species names cnam <- make.cepnames(names(BCI)) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, pch=\"+\", pcol=\"grey\") ## show select in action set.seed(1) take <- sample(ncol(BCI), 50) plot(mod, dis=\"sp\", type=\"n\") stems <- colSums(BCI) orditorp(mod, \"sp\", label = cnam, priority=stems, select = take, pch=\"+\", pcol=\"grey\") # \\dontshow{ ## example(orditorp) should not set random seed in the user session rm(.Random.seed) #> Warning: object '.Random.seed' not found # }"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":null,"dir":"Reference","previous_headings":"","what":"Trellis (Lattice) Plots for Ordination — ordixyplot","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Functions ordicloud, ordisplom ordixyplot provide interface plot ordination results using Trellis functions cloud, splom xyplot package lattice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"ordixyplot(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", aspect = \"iso\", envfit, type = c(\"p\", \"biplot\"), ...) ordisplom(x, data=NULL, formula = NULL, display = \"sites\", choices = 1:3, panel = \"panel.ordi\", type = \"p\", ...) ordicloud(x, data = NULL, formula, display = \"sites\", choices = 1:3, panel = \"panel.ordi3d\", prepanel = \"prepanel.ordi3d\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"x ordination result scores knows: ordination result vegan many others. data Optional data amend ordination results. ordination results found x, may give data variables needed plots. Typically environmental data. formula Formula define plots. default formula used omitted. ordination axes must called names ordination results (names vary among methods). ordisplom, special character . refers ordination result. display kind scores: argument passed scores. choices axes selected: argument passed scores. panel, prepanel names panel prepanel functions. aspect aspect plot (passed lattice function). envfit Result envfit function displayed ordixyplot. Please note needs choices ordixyplot. type type plot. knows alternatives panel.xyplot. addition ordixyplot alternatives \"biplot\", \"arrows\" \"polygon\". first displays fitted vectors factor centroids envfit, constrained ordination, biplot arrows factor centroids envfit given. second (type = \"arrows\") trellis variant ordiarrows draws arrows groups. line parameters controlled trellis.par.set superpose.line, user can set length, angle ends parameters panel.arrows. last one (type = \"polygon\") draws polygon enclosing points panel polygon enclosing points data. overall polygon controlled trellis.par.set plot.polygon, panel polygon controlled superpose.polygon. ... Arguments passed scores methods lattice functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"functions provide interface corresponding lattice functions. graphical parameters passed lattice function graphs extremely configurable. See Lattice xyplot, splom cloud details, usage possibilities. argument x must always ordination result. scores extracted vegan function scores functions work vegan ordinations many others. formula used define models. functions simple default formulae used formula missing. formula omitted ordisplom produces pairs plot ordination axes variables data. formula given, ordination results must referred . variables names. functions, formula must use names ordination scores names data. ordination scores found x, data optional. data contain variables ordination scores used plots. Typically, environmental variables (typically factors) define panels plot symbols. proper work done panel function. layout can changed defining panel functions. See panel.xyplot, panel.splom panel.cloud details survey possibilities. Ordination graphics always isometric: scale used axes. controlled (can changed) argument aspect ordixyplot. ordicloud isometric scaling defined panel prepanel functions. must replace functions want non-isometric scaling graphs. select isometric scaling ordisplom.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"function return Lattice objects class \"trellis\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/ordixyplot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Trellis (Lattice) Plots for Ordination — ordixyplot","text":"","code":"data(dune, dune.env) ord <- cca(dune) ## Pairs plots ordisplom(ord) ordisplom(ord, data=dune.env, choices=1:2) ordisplom(ord, data=dune.env, form = ~ . | Management, groups=Manure) ## Scatter plot with polygons ordixyplot(ord, data=dune.env, form = CA1 ~ CA2 | Management, groups=Manure, type = c(\"p\",\"polygon\")) ## Choose a different scaling ordixyplot(ord, scaling = \"symmetric\") ## ... Slices of third axis ordixyplot(ord, form = CA1 ~ CA2 | equal.count(CA3, 4), type = c(\"g\",\"p\", \"polygon\")) ## Display environmental variables ordixyplot(ord, envfit = envfit(ord ~ Management + A1, dune.env, choices=1:3)) ## 3D Scatter plots ordicloud(ord, form = CA2 ~ CA3*CA1, groups = Manure, data = dune.env) ordicloud(ord, form = CA2 ~ CA3*CA1 | Management, groups = Manure, data = dune.env, auto.key = TRUE, type = c(\"p\",\"h\"))"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Coordinates of Neighbourhood Matrix — pcnm","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"function computed classical PCNM principal coordinate analysis truncated distance matrix. commonly used transform (spatial) distances rectangular data suitable constrained ordination regression.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"pcnm(dis, threshold, w, dist.ret = FALSE)"},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"dis distance matrix. threshold threshold value truncation distance. missing, minimum distance giving connected network used. found longest distance minimum spanning tree dis. w Prior weights rows. dist.ret Return distances used calculate PCNMs.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Principal Coordinates Neighbourhood Matrix (PCNM) map distances rows onto rectangular matrix rows using truncation threshold long distances (Borcard & Legendre 2002). original distances Euclidean distances two dimensions (like normal spatial distances), mapped onto two dimensions truncation distances. truncation, higher number principal coordinates. selection truncation distance huge influence PCNM vectors. default use longest distance keep data connected. distances truncation threshold given arbitrary value 4 times threshold. regular data, first PCNM vectors show wide scale variation later PCNM vectors show smaller scale variation (Borcard & Legendre 2002), irregular data interpretation clear. PCNM functions used express distances rectangular form similar normal explanatory variables used , e.g., constrained ordination (rda, cca dbrda) univariate regression (lm) together environmental variables (row weights supplied cca; see Examples). regarded powerful method forcing rectangular environmental data distances using partial mantel analysis (mantel.partial) together geographic distances (Legendre et al. 2008, see Tuomisto & Ruokolainen 2008). function based pcnm function Dray's unreleased spacemakeR package. differences current function uses spantree internal support function. current function also can use prior weights rows using weighted metric scaling wcmdscale. use row weights allows finding orthonormal PCNMs also correspondence analysis (e.g., cca).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"list following elements: values Eigenvalues obtained principal coordinates analysis. vectors Eigenvectors obtained principal coordinates analysis. scaled unit norm. vectors can extracted scores function. default return PCNM vectors, argument choices selects given vectors. threshold Truncation distance. dist distance matrix values threshold replaced arbitrary value four times threshold. String \"pcnm\" added method attribute, new attribute threshold added distances. returned dist.ret = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Borcard D. Legendre P. (2002) -scale spatial analysis ecological data means principal coordinates neighbour matrices. Ecological Modelling 153, 51--68. Legendre, P., Borcard, D Peres-Neto, P. (2008) Analyzing explaining beta diversity? Comment. Ecology 89, 3238--3244. Tuomisto, H. & Ruokolainen, K. (2008) Analyzing explaining beta diversity? reply. Ecology 89, 3244--3256.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"Jari Oksanen, based code Stephane Dray.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/pcnm.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Coordinates of Neighbourhood Matrix — pcnm","text":"","code":"## Example from Borcard & Legendre (2002) data(mite.xy) pcnm1 <- pcnm(dist(mite.xy)) op <- par(mfrow=c(1,3)) ## Map of PCNMs in the sample plot ordisurf(mite.xy, scores(pcnm1, choi=1), bubble = 4, main = \"PCNM 1\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.71 total = 9.71 #> #> REML score: -120.7705 ordisurf(mite.xy, scores(pcnm1, choi=2), bubble = 4, main = \"PCNM 2\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 7.18 total = 8.18 #> #> REML score: -103.4662 ordisurf(mite.xy, scores(pcnm1, choi=3), bubble = 4, main = \"PCNM 3\") #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 8.32 total = 9.32 #> #> REML score: -94.19053 par(op) ## Plot first PCNMs against each other ordisplom(pcnm1, choices=1:4) ## Weighted PCNM for CCA data(mite) rs <- rowSums(mite)/sum(mite) pcnmw <- pcnm(dist(mite.xy), w = rs) ord <- cca(mite ~ scores(pcnmw)) ## Multiscale ordination: residual variance should have no distance ## trend msoplot(mso(ord, mite.xy))"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":null,"dir":"Reference","previous_headings":"","what":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Individual (count data) incidence (presence-absence data) based null models can generated community level simulations. Options preserving characteristics original matrix (rows/columns sums, matrix fill) restricted permutations (based strata) discussed Details section.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"permatfull(m, fixedmar = \"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, ...) permatswap(m, method = \"quasiswap\", fixedmar=\"both\", shuffle = \"both\", strata = NULL, mtype = \"count\", times = 99, burnin = 0, thin = 1, ...) # S3 method for permat print(x, digits = 3, ...) # S3 method for permat summary(object, ...) # S3 method for summary.permat print(x, digits = 2, ...) # S3 method for permat plot(x, type = \"bray\", ylab, xlab, col, lty, lowess = TRUE, plot = TRUE, text = TRUE, ...) # S3 method for permat lines(x, type = \"bray\", ...) # S3 method for permat as.ts(x, type = \"bray\", ...) # S3 method for permat toCoda(x)"},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"m community data matrix plots (samples) rows species (taxa) columns. fixedmar character, stating row/column sums preserved (\"none\", \"rows\", \"columns\", \"\"). strata Numeric vector factor length nrow(m) grouping rows within strata restricted permutations. Unique values levels used. mtype Matrix data type, either \"count\" count data, \"prab\" presence-absence type incidence data. times Number permuted matrices. method Character method used swap algorithm (\"swap\", \"tswap\", \"quasiswap\", \"backtrack\") described function make.commsim. mtype=\"count\" \"quasiswap\", \"swap\", \"swsh\" \"abuswap\" methods available (see details). shuffle Character, indicating whether individuals (\"ind\"), samples (\"samp\") (\"\") shuffled, see details. burnin Number null communities discarded proper analysis sequential (\"swap\", \"tswap\") methods. thin Number discarded permuted matrices two evaluations sequential (\"swap\", \"tswap\") methods. x, object Object class \"permat\" digits Number digits used rounding. ylab, xlab, col, lty graphical parameters plot method. type Character, type plot displayed: \"bray\" Bray-Curtis dissimilarities, \"chisq\" Chi-squared values. lowess, plot, text Logical arguments plot method, whether locally weighted regression curve drawn, plot drawn, statistic values printed plot. ... arguments passed simulate.nullmodel methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"function permatfull useful matrix fill allowed vary, matrix type count. fixedmar argument used set constraints permutation. none margins fixed, cells randomised within matrix. rows columns fixed, cells within rows columns randomised, respectively. margins fixed, r2dtable function used based Patefield's (1981) algorithm. presence absence data, matrix fill necessarily fixed, permatfull wrapper function make.commsim. r00, r0, c0, quasiswap algorithms make.commsim used \"none\", \"rows\", \"columns\", \"\" values fixedmar argument, respectively shuffle argument effect mtype = \"count\" permatfull function used \"none\", \"rows\", \"columns\" values fixedmar. cases count data individual based randomisations. \"samp\" \"\" options result fixed matrix fill. \"\" option means individuals shuffled among non zero cells ensuring cell zeros result, cell (zero new valued cells) shuffled. function permatswap useful matrix fill (.e. proportion empty cells) row/columns sums kept constant. permatswap uses different kinds swap algorithms, row columns sums fixed cases. presence-absence data, swap tswap methods make.commsim can used. count data, special swap algorithm ('swapcount') implemented results permuted matrices fixed marginals matrix fill time. 'quasiswapcount' algorithm (method=\"quasiswap\" mtype=\"count\") uses trick Carsten Dormann's swap.web function package bipartite. First, random matrix generated r2dtable function retaining row column sums. original matrix fill reconstructed sequential steps increase decrease matrix fill random matrix. steps based swapping 2x2 submatrices (see 'swapcount' algorithm details) maintain row column totals. algorithm generates independent matrices step, burnin thin arguments considered. default method, sequential (swapcount ) independence subsequent matrices checked. swapcount algorithm (method=\"swap\" mtype=\"count\") tries find 2x2 submatrices (identified 2 random row 2 random column indices), can swapped order leave column row totals fill unchanged. First, algorithm finds largest value submatrix can swapped (\\(d\\)) whether diagonal antidiagonal way. Submatrices contain values larger zero either diagonal antidiagonal position can swapped. Swap means values diagonal antidiagonal positions decreased \\(d\\), remaining cells increased \\(d\\). swap made fill change. algorithm sequential, subsequent matrices independent, swaps modify little matrix large. cases many burnin steps thinning needed get independent random matrices. Although algorithm implemented C, large burnin thin values can slow considerably. WARNING: according simulations, algorithm seems biased non random, thus use avoided! algorithm \"swsh\" function permatswap hybrid algorithm. First, makes binary quasiswaps keep row column incidences constant, non-zero values modified according shuffle argument (\"samp\" \"\" available case, applied non-zero values). also recognizes fixedmar argument \"\" (vegan versions <= 2.0 algorithm fixedmar = \"none\"). algorithm \"abuswap\" produces two kinds null models (based fixedmar=\"columns\" fixedmar=\"rows\") described Hardy (2008; randomization scheme 2x 3x, respectively). preserve column row occurrences, column row sums time. (Note similar constraints can achieved non sequential \"swsh\" algorithm fixedmar argument set \"columns\" \"rows\", respectively.) Constraints row/column sums, matrix fill, total sum sums within strata can checked summary method. plot method visually testing randomness permuted matrices, especially sequential swap algorithms. tendency graph, higher burnin thin values can help sequential methods. New lines can added existing plot lines method. Unrestricted restricted permutations: strata NULL, functions perform unrestricted permutations. Otherwise, used restricted permutations. strata contain least 2 rows order perform randomization (case low row numbers, swap algorithms can rather slow). design well balanced (.e. number observations within stratum), permuted matrices may biased constraints forced submatrices different dimensions. often means, number potential permutations decrease dimensions. constraints put, less randomness can expected. plot method useful graphically testing trend independence permuted matrices. especially important using sequential algorithms (\"swap\", \"tswap\", \"abuswap\"). .ts method can used extract Bray-Curtis dissimilarities Chi-squared values time series. can used testing independence (see Examples). method toCoda useful accessing diagnostic tools available coda package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Functions permatfull permatswap return object class \"permat\" containing function call (call), original data matrix used permutations (orig) list permuted matrices length times (perm). summary method returns various statistics list (including mean Bray-Curtis dissimilarities calculated pairwise among original permuted matrices, Chi-square statistics, check results constraints; see Examples). Note strata used original call, summary calculation may take longer. plot creates plot side effect. .ts method returns object class \"ts\".","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Original references presence-absence algorithms given help page make.commsim. Hardy, O. J. (2008) Testing spatial phylogenetic structure local communities: statistical performances different null models test statistics locally neutral community. Journal Ecology 96, 914--926. Patefield, W. M. (1981) Algorithm AS159. efficient method generating r x c tables given row column totals. Applied Statistics 30, 91--97.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"Péter Sólymos, solymos@ualberta.ca Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permatfull.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Matrix Permutation Algorithms for Presence-Absence and Count Data — permat","text":"","code":"## A simple artificial community data matrix. m <- matrix(c( 1,3,2,0,3,1, 0,2,1,0,2,1, 0,0,1,2,0,3, 0,0,0,1,4,3 ), 4, 6, byrow=TRUE) ## Using the quasiswap algorithm to create a ## list of permuted matrices, where ## row/columns sums and matrix fill are preserved: x1 <- permatswap(m, \"quasiswap\") summary(x1) #> Summary of object of class 'permat' #> #> Call: permatswap(m = m, method = \"quasiswap\") #> #> Matrix type: count #> Permutation type: swap #> Method: quasiswap_count, burnin: 0, thin: 1 #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 13.13 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3667 0.4333 0.4145 0.4667 0.6000 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 16.27 19.59 21.10 21.51 23.36 31.69 ## Unrestricted permutation retaining ## row/columns sums but not matrix fill: x2 <- permatfull(m) summary(x2) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: FALSE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 16.16 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 0 % #> Column incidences retained: 1.01 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.3667 0.3865 0.4333 0.6333 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 7.824 12.046 15.660 16.015 19.413 28.132 ## Unrestricted permutation of presence-absence type ## not retaining row/columns sums: x3 <- permatfull(m, \"none\", mtype=\"prab\") x3$orig ## note: original matrix is binarized! #> [,1] [,2] [,3] [,4] [,5] [,6] #> [1,] 1 1 1 0 1 1 #> [2,] 0 1 1 0 1 1 #> [3,] 0 0 1 1 0 1 #> [4,] 0 0 0 1 1 1 summary(x3) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, fixedmar = \"none\", mtype = \"prab\") #> #> Matrix type: prab #> Permutation type: full #> Method: r00 #> Restricted: FALSE #> Fixed margins: none #> Individuals and samples are shuffled #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 15 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 100 % #> Row sums retained: 4.04 % #> Column sums retained: 0 % #> Row incidences retained: 4.04 % #> Column incidences retained: 0 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.2000 0.3333 0.4000 0.3852 0.4000 0.5333 #> #> Chi-squared for original matrix: 8.4 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 8.812 13.583 15.208 15.295 17.083 21.896 ## Restricted permutation, ## check sums within strata: x4 <- permatfull(m, strata=c(1,1,2,2)) summary(x4) #> Summary of object of class 'permat' #> #> Call: permatfull(m = m, strata = c(1, 1, 2, 2)) #> #> Matrix type: count #> Permutation type: full #> Method: r2dtable #> Restricted: TRUE #> Fixed margins: both #> #> Matrix dimensions: 4 rows, 6 columns #> Sum of original matrix: 30 #> Fill of original matrix: 0.62 #> Number of permuted matrices: 99 #> #> Matrix sums retained: 100 % #> Matrix fill retained: 38.38 % #> Row sums retained: 100 % #> Column sums retained: 100 % #> Row incidences retained: 1.01 % #> Column incidences retained: 2.02 % #> Sums within strata retained: 100 % #> #> Bray-Curtis dissimilarities among original and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 0.06667 0.20000 0.26667 0.23502 0.26667 0.46667 #> #> Chi-squared for original matrix: 18.55 #> Chi-squared values among expected and permuted matrices: #> Min. 1st Qu. Median Mean 3rd Qu. Max. #> 14.21 18.68 21.05 22.09 25.26 36.50 ## NOTE: 'times' argument usually needs to be >= 99 ## here much lower value is used for demonstration ## Not sequential algorithm data(BCI) a <- permatswap(BCI, \"quasiswap\", times=19) ## Sequential algorithm b <- permatswap(BCI, \"abuswap\", fixedmar=\"col\", burnin=0, thin=100, times=19) opar <- par(mfrow=c(2,2)) plot(a, main=\"Not sequential\") plot(b, main=\"Sequential\") plot(a, \"chisq\") plot(b, \"chisq\") par(opar) ## Extract Bray-Curtis dissimilarities ## as time series bc <- as.ts(b) ## Lag plot lag.plot(bc) ## First order autoregressive model mar <- arima(bc, c(1,0,0)) mar #> #> Call: #> arima(x = bc, order = c(1, 0, 0)) #> #> Coefficients: #> ar1 intercept #> 0.9915 0.1850 #> s.e. 0.0120 0.1374 #> #> sigma^2 estimated as 0.000346: log likelihood = 46.71, aic = -87.42 ## Ljung-Box test of residuals Box.test(residuals(mar)) #> #> \tBox-Pierce test #> #> data: residuals(mar) #> X-squared = 0.35725, df = 1, p-value = 0.55 #> ## Graphical diagnostics tsdiag(mar)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":null,"dir":"Reference","previous_headings":"","what":"Extract, Analyse and Display Permutation Results — permustats","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results vegan functions. support functions can find quantiles standardized effect sizes, plot densities Q-Q plots.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"permustats(x, ...) # S3 method for permustats summary(object, interval = 0.95, alternative, ...) # S3 method for permustats densityplot(x, data, xlab = \"Permutations\", ...) # S3 method for permustats density(x, observed = TRUE, ...) # S3 method for permustats qqnorm(y, observed = TRUE, ...) # S3 method for permustats qqmath(x, data, observed = TRUE, sd.scale = FALSE, ylab = \"Permutations\", ...) # S3 method for permustats boxplot(x, scale = FALSE, names, ...) # S3 method for permustats pairs(x, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extract, Analyse and Display Permutation Results — permustats","text":"object, x, y object handled. interval numeric; coverage interval reported. alternative character string specifying limits used interval direction test evaluating \\(p\\)-values. Must one \"two.sided\" (upper lower limit), \"greater\" (upper limit), \"less\" (lower limit). Usually alternative given result object, can specified argument. xlab, ylab Arguments densityplot qqmath functions. observed Add observed statistic among permutations. sd.scale Scale permutations unit standard deviation observed statistic standardized effect size. data Ignored. scale Use standardized effect size (SES). names Names boxes (default: names statistics). ... arguments passed function. density passed density.default, boxplot boxplot.default.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function extracts permutation results observed statistics several vegan functions perform permutations simulations. summary method permustats estimates standardized effect sizes (SES) difference observed statistic mean permutations divided standard deviation permutations (also known \\(z\\)-values). also prints mean, median, limits contain interval percent permuted values. default (interval = 0.95), two-sided test (2.5%, 97.5%) one-sided tests either 5% 95% quantile \\(p\\)-value depending test direction. mean, quantiles \\(z\\) values evaluated permuted values without observed statistic, \\(p\\)-value evaluated observed statistic. intervals \\(p\\)-value evaluated test direction original test, can changed argument alternative. Several permustats objects can combined c function. c function checks statistics equal, performs sanity tests. density densityplot methods display kernel density estimates permuted values. observed value statistic included permuted values, densityplot method marks observed statistic vertical line. However density method uses standard plot method mark observed value. qqnorm qqmath display Q-Q plots permutations, optionally together observed value (default) shown horizontal line plots. qqnorm plots permutation values standard Normal variate. qqmath defaults standard Normal well, can accept alternatives (see standard qqmath). qqmath function can also plot observed statistic standardized effect size (SES) standandized permutations (argument sd.scale). permutations standardized without observed statistic, similarly summary. Functions density qqnorm based standard R methods accept arguments. handle one statistic, used several test statistic evaluated. densityplot qqmath lattice graphics, can used either one several statistics. functions pass arguments underlying functions; see documentation. Functions qqmath densityplot default use axis scaling subplots lattice. can use argument scales set independent scaling subplots appropriate (see xyplot exhaustive list arguments). Function boxplot draws box--whiskers plots effect size, difference permutations observed statistic. scale = TRUE, permutations standardized unit standard deviation, plot show standardized effect sizes. Function pairs plots permutation values statistics . function passes extra arguments pairs. permustats can extract permutation statistics results adonis2, anosim, anova.cca, mantel, mantel.partial, mrpp, oecosimu, ordiareatest, permutest.cca, protest, permutest.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extract, Analyse and Display Permutation Results — permustats","text":"permustats function returns object class \"permustats\". list items \"statistic\" observed statistics, permutations contains permuted values, alternative contains text defining character test (\"two.sided\", \"less\" \"greater\"). qqnorm density methods return standard result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extract, Analyse and Display Permutation Results — permustats","text":"Jari Oksanen contributions Gavin L. Simpson (permustats.permutest.betadisper method related modifications summary.permustats print method) Eduard Szöcs (permustats.anova.cca).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permustats.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extract, Analyse and Display Permutation Results — permustats","text":"","code":"data(dune, dune.env) mod <- adonis2(dune ~ Management + A1, data = dune.env) ## use permustats perm <- permustats(mod) summary(perm) #> #> statistic SES mean lower median upper Pr(perm) #> Management 3.0730 4.6870 1.0387 0.9565 1.8217 0.004 ** #> A1 2.7676 2.7175 1.0022 0.8484 2.2436 0.028 * #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.95) densityplot(perm) qqmath(perm) boxplot(perm, scale=TRUE, lty=1, pch=16, cex=0.6, col=\"hotpink\", ylab=\"SES\") abline(h=0, col=\"skyblue\") ## example of multiple types of statistic mod <- with(dune.env, betadisper(vegdist(dune), Management)) pmod <- permutest(mod, nperm = 99, pairwise = TRUE) perm <- permustats(pmod) summary(perm, interval = 0.90) #> #> statistic SES mean lower median upper Pr(perm) #> Overall (F) 1.9506 0.7173 1.1427 0.8211 2.4909 0.154 #> BF-HF (t) -0.5634 -0.4124 -0.0443 -2.0202 -0.0293 1.8851 0.591 #> BF-NM (t) -2.2387 -1.8672 -0.0045 -1.8423 0.0074 2.0628 0.067 . #> BF-SF (t) -1.1675 -0.9341 -0.0086 -1.9337 -0.0450 1.9486 0.283 #> HF-NM (t) -2.1017 -1.9328 0.0277 -1.6716 0.0346 1.7582 0.067 . #> HF-SF (t) -0.8789 -0.7872 0.0321 -1.8598 0.0284 1.8394 0.379 #> NM-SF (t) 0.9485 0.8265 0.0121 -1.9118 0.0690 1.7827 0.379 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> (Interval (Upper - Lower) = 0.9)"},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation tests in Vegan — permutations","title":"Permutation tests in Vegan — permutations","text":"version 2.2-0, vegan significantly improved access restricted permutations brings line offered Canoco. permutation designs modelled permutation schemes Canoco 3.1 (ter Braak, 1990). vegan currently provides following features within permutation tests: Free permutation DATA, also known randomisation, Free permutation DATA within levels grouping variable, Restricted permutations line transects time series, Permutation groups samples whilst retaining within-group ordering, Restricted permutations spatial grids, Blocking, samples never permuted blocks, Split-plot designs, permutation whole plots, split plots, . , use DATA mean either observed data function data, example residuals ordination model presence covariables. capabilities provided functions permute package. user can request particular type permutation supplying permutations argument function object returned , defines samples permuted. Alternatively, user can simply specify required number permutations simple randomisation procedure performed. Finally, user can supply matrix permutations ( number rows equal number permutations number columns equal number observations data) vegan use permutations instead generating new permutations. majority functions vegan allow full range possibilities outlined . Exceptions include kendall.post kendall.global. Null hypothesis first two types permutation test listed assumes free exchangeability DATA (within levels grouping variable, specified). Dependence observations, arises due spatial temporal autocorrelation, -complicated experimental designs, split-plot designs, violates fundamental assumption test requires complex restricted permutation test designs. designs available via permute package vegan provides access version 2.2-0 onwards. Unless otherwise stated help pages specific functions, permutation tests vegan follow format/structure: appropriate test statistic chosen. statistic chosen described help pages individual functions. value test statistic evaluate observed data analysis/model recorded. Denote value \\(x_0\\). DATA randomly permuted according one schemes, value test statistic permutation evaluated recorded. Step 3 repeated total \\(n\\) times, \\(n\\) number permutations requested. Denote values \\(x_i\\), \\(= 1, ..., n\\) Count number values test statistic, \\(x_i\\), Null distribution extreme test statistic observed data \\(x_0\\). Denote count \\(N\\). use phrase extreme include cases two-sided test performed large negative values test statistic considered. permutation p-value computed $$p = \\frac{N + 1}{n + 1}$$ description illustrates default number permutations specified vegan functions takes values 199 999 example. Pretty p values achieved \\(+ 1\\) denominator results division 200 1000, 199 999 random permutations used test. simple intuition behind presence \\(+ 1\\) numerator denominator represent inclusion observed value statistic Null distribution (e.g. Manly 2006). Phipson & Smyth (2010) present compelling explanation inclusion \\(+ 1\\) numerator denominator p value calculation. Fisher (1935) mind permutation test involve enumeration possible permutations data yielding exact test. However, complete enumeration may feasible practice owing potentially vast number arrangements data, even modestly-sized data sets free permutation samples. result evaluate p value tail probability Null distribution test statistic directly random sample possible permutations. Phipson & Smyth (2010) show naive calculation permutation p value $$p = \\frac{N}{n}$$ leads invalid test incorrect type error rate. go show replacing unknown tail probability ( p value) Null distribution biased estimator $$p = \\frac{N + 1}{n + 1}$$ positive bias induced just right size account uncertainty estimation tail probability set randomly sampled permutations yield test correct type error rate. estimator described correct situation permutations data samples randomly without replacement. strictly happens vegan permutations drawn pseudo-randomly independent one another. Note actual chance happening practice small functions permute guarantee generate unique set permutations unless complete enumeration permutations requested. feasible smallest data sets restrictive permutation designs, cases chance drawing set permutations repeats lessened sample size, thence size set possible permutations, increases. situation sampling permutations replacement , tail probability \\(p\\) calculated biased estimator described somewhat conservative, large amount depends number possible values test statistic can take permutation data (Phipson & Smyth, 2010). represents slight loss statistical power conservative p value calculation used . However, unless sample sizes small permutation design set values test statistic can take also small, loss power unlikely critical. minimum achievable p-value $$p_{\\mathrm{min}} = \\frac{1}{n + 1}$$ hence depends number permutations evaluated. However, one simply increase number permutations (\\(n\\)) achieve potentially lower p-value unless number observations available permits number permutations. unlikely problem smallest data sets free permutation (randomisation) valid, restricted permutation designs low number observations, may many unique permutations data might desire reach required level significance. currently responsibility user determine total number possible permutations DATA. number possible permutations allowed specified design can calculated using numPerms permute package. Heuristics employed within shuffleSet function used vegan can triggered generate entire set permutations instead random set. settings controlling triggering complete enumeration step contained within permutation design created using link[permute]{} can set user. See details. Limits total number permutations DATA severe temporally spatially ordered data experimental designs low replication. example, time series \\(n = 100\\) observations just 100 possible permutations including observed ordering. situations low number permutations possible due nature DATA experimental design, enumeration permutations becomes important achievable computationally. , provided brief overview capabilities vegan permute. get best new functionality details set permutation designs using , consult vignette Restricted permutations; using permute package supplied permute accessible via vignette(\"permutations\", package = \"permute\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"random-number-generation","dir":"Reference","previous_headings":"","what":"Random Number Generation","title":"Permutation tests in Vegan — permutations","text":"permutations based random number generator provided R. may change R releases change permutations vegan test results. One change R release 3.6.0. new version clearly better permutation tests use . However, need reproduce old results, can set R random number generator previous version RNGversion.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation tests in Vegan — permutations","text":"Manly, B. F. J. (2006). Randomization, Bootstrap Monte Carlo Methods Biology, Third Edition. Chapman Hall/CRC. Phipson, B., & Smyth, G. K. (2010). Permutation P-values never zero: calculating exact P-values permutations randomly drawn. Statistical Applications Genetics Molecular Biology, 9, Article 39. DOI: 10.2202/1544-6115.1585 ter Braak, C. J. F. (1990). Update notes: CANOCO version 3.1. Wageningen: Agricultural Mathematics Group. (UR). See also: Davison, . C., & Hinkley, D. V. (1997). Bootstrap Methods Application. Cambridge University Press.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutations.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation tests in Vegan — permutations","text":"Gavin L. Simpson","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":null,"dir":"Reference","previous_headings":"","what":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Implements permutation-based test multivariate homogeneity group dispersions (variances) results call betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"# S3 method for betadisper permutest(x, pairwise = FALSE, permutations = 999, parallel = getOption(\"mc.cores\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"x object class \"betadisper\", result call betadisper. pairwise logical; perform pairwise comparisons group means? permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. ... Arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"test one groups variable others, ANOVA distances group centroids can performed parametric theory used interpret significance F. alternative use permutation test. permutest.betadisper permutes model residuals generate permutation distribution F Null hypothesis difference dispersion groups. Pairwise comparisons group mean dispersions can performed setting argument pairwise TRUE. classical t test performed pairwise group dispersions. combined permutation test based t statistic calculated pairwise group dispersions. alternative classical comparison group dispersions, calculate Tukey's Honest Significant Differences groups, via TukeyHSD.betadisper.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"permutest.betadisper returns list class \"permutest.betadisper\" following components: tab ANOVA table object inheriting class \"data.frame\". pairwise list components observed permuted containing observed permuted p-values pairwise comparisons group mean distances (dispersions variances). groups character; levels grouping factor. control list, result call .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Anderson, M.J. (2006) Distance-based tests homogeneity multivariate dispersions. Biometrics 62(1), 245--253. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006) Multivariate dispersion measure beta diversity. Ecology Letters 9(6), 683--693.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/permutest.betadisper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Permutation test of multivariate homogeneity of groups dispersions (variances) — permutest.betadisper","text":"","code":"data(varespec) ## Bray-Curtis distances between samples dis <- vegdist(varespec) ## First 16 sites grazed, remaining 8 sites ungrazed groups <- factor(c(rep(1,16), rep(2,8)), labels = c(\"grazed\",\"ungrazed\")) ## Calculate multivariate dispersions mod <- betadisper(dis, groups) mod #> #> \tHomogeneity of multivariate dispersions #> #> Call: betadisper(d = dis, group = groups) #> #> No. of Positive Eigenvalues: 15 #> No. of Negative Eigenvalues: 8 #> #> Average distance to median: #> grazed ungrazed #> 0.3926 0.2706 #> #> Eigenvalues for PCoA axes: #> (Showing 8 of 23 eigenvalues) #> PCoA1 PCoA2 PCoA3 PCoA4 PCoA5 PCoA6 PCoA7 PCoA8 #> 1.7552 1.1334 0.4429 0.3698 0.2454 0.1961 0.1751 0.1284 ## Perform test anova(mod) #> Analysis of Variance Table #> #> Response: Distances #> Df Sum Sq Mean Sq F value Pr(>F) #> Groups 1 0.07931 0.079306 4.6156 0.04295 * #> Residuals 22 0.37801 0.017182 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test for F pmod <- permutest(mod, permutations = 99, pairwise = TRUE) ## Tukey's Honest Significant Differences (mod.HSD <- TukeyHSD(mod)) #> Tukey multiple comparisons of means #> 95% family-wise confidence level #> #> Fit: aov(formula = distances ~ group, data = df) #> #> $group #> diff lwr upr p adj #> ungrazed-grazed -0.1219422 -0.2396552 -0.004229243 0.0429502 #> plot(mod.HSD) ## Has permustats() method pstat <- permustats(pmod) densityplot(pstat, scales = list(x = list(relation = \"free\"))) qqmath(pstat, scales = list(relation = \"free\"))"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Functions plot extract results constrained correspondence analysis (cca), redundancy analysis (rda), distance-based redundancy analysis (dbrda) constrained analysis principal coordinates (capscale).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"# S3 method for cca plot(x, choices = c(1, 2), display = c(\"sp\", \"wa\", \"cn\"), scaling = \"species\", type, xlim, ylim, const, correlation = FALSE, hill = FALSE, cex = 0.7, ...) # S3 method for cca text(x, display = \"sites\", labels, choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca points(x, display = \"sites\", choices = c(1, 2), scaling = \"species\", arrow.mul, head.arrow = 0.05, select, const, axis.bp = FALSE, correlation = FALSE, hill = FALSE, ...) # S3 method for cca scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", hill = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for rda scores(x, choices = c(1,2), display = \"all\", scaling = \"species\", const, correlation = FALSE, tidy = FALSE, droplist = TRUE, ...) # S3 method for cca summary(object, digits = max(3, getOption(\"digits\") - 3), ...) # S3 method for cca labels(object, display, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"x, object cca result object. choices Axes shown. display Scores shown. must include alternatives \"species\" \"sp\" species scores, sites \"wa\" site scores, \"lc\" linear constraints LC scores, \"bp\" biplot arrows \"cn\" centroids factor constraints instead arrow, \"reg\" regression coefficients (.k.. canonical coefficients). alternative \"\" selects available scores. scaling Scaling species site scores. Either species (2) site (1) scores scaled eigenvalues, set scores left unscaled, 3 scaled symmetrically square root eigenvalues. Corresponding negative values can used cca additionally multiply results \\(\\sqrt(1/(1-\\lambda))\\). scaling know Hill scaling (although nothing Hill's rescaling decorana). corresponding negative values rda, species scores divided standard deviation species multiplied equalizing constant. Unscaled raw scores stored result can accessed scaling = 0. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Arguments correlation hill scores.rda scores.cca respectively can used combination character descriptions get corresponding negative value. correlation, hill logical; scaling character description scaling type, correlation hill used select corresponding negative scaling type; either correlation-like scores Hill's scaling PCA/RDA CA/CCA respectively. See argument scaling details. tidy Return scores compatible ggplot2: scores single data.frame, score type identified factor variable score, names variable label, weights (CCA) variable weight. possible values score species, sites (WA scores), constraints (LC scores sites calculated directly constraining variables), biplot (biplot arrows), centroids (levels factor variables), factorbiplot (biplot arrows model centroids), regression (regression coefficients find LC scores constraints). scores used conventional plot, directly suitable used ggplot2 package. type Type plot: partial match text text labels, points points, none setting frames . omitted, text selected smaller data sets, points larger. xlim, ylim x y limits (min,max) plot. labels Optional text used instead row names. use , good check default labels order using labels command. arrow.mul Factor expand arrows graph. Arrows scaled automatically fit graph missing. head.arrow Default length arrow heads. select Items displayed. can either logical vector TRUE displayed items vector indices displayed items. const General scaling constant rda scores. default use constant gives biplot scores, , scores approximate original data (see vignette ‘Design Decisions’ browseVignettes(\"vegan\") details discussion). const vector two items, first used species, second item site scores. droplist Return matrix instead named list one kind scores requested. axis.bp Draw axis biplot arrows. digits Number digits output. cex Character expansion. ... Parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function used cca rda. produces quick, standard plot current scaling. plot function sets colours (col), plotting characters (pch) character sizes (cex) certain standard values. fuller control produced plot, best call plot type=\"none\" first, add plotting item separately using text.cca points.cca functions. use default settings standard text points functions accept parameters, allowing full user control produced plots. Environmental variables receive special treatment. display=\"bp\", arrows drawn. labelled text unlabelled points. arrows basically unit scaling, sites scaled (scaling \"sites\" \"symmetric\"), scores requested axes adjusted relative axis highest eigenvalue. scaling = \"species\" scaling = \"none\", arrows consistent vectors fitted linear combination scores (display = \"lc\" function envfit), scaling alternatives differ. basic plot function uses simple heuristics adjusting unit-length arrows current plot area, user can give expansion factor mul.arrow. display=\"cn\" centroids levels factor variables displayed (available factors formula interface used cca rda). option continuous variables still presented arrows ordered factors arrows centroids. display = \"reg\" arrows drawn regression coefficients (.k.. canonical coefficients) constraints conditions. Biplot arrows can interpreted individually, regression coefficients must interpreted together: LC score site sum regressions displayed arrows. partialled conditions zero shown biplot arrows, shown regressions, show effect must partialled get LC scores. biplot arrows standard easily interpreted, regression arrows used know need . want better control plots, best construct plot text points commands accept graphical parameters. important remember use scaling, correlation hill arguments calls. plot.cca command returns invisibly ordiplot result object, consistent scaling elements. also possible use R pipes (|>) maintain consistent scaling setting. easiest way full control graphics first set plot frame using plot type = \"n\" needed scores display save result use first command pipe. points text commands ordiplot allow full graphical control (see section Examples). Utility function labels returns default labels order applied text. Palmer (1993) suggested using linear constraints (“LC scores”) ordination diagrams, gave better results simulations site scores (“WA scores”) step constrained unconstrained analysis. However, McCune (1997) showed noisy environmental variables (environmental measurements noisy) destroy “LC scores” whereas “WA scores” little affected. Therefore plot function uses site scores (“WA scores”) default. consistent usage statistics functions R (lda, cancor).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"plot function returns invisibly plotting structure can used function identify.ordiplot identify points functions ordiplot family.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/plot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plot or Extract Results of Constrained Correspondence Analysis or Redundancy Analysis — plot.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Moisture + Management, dune.env) ## better control -- remember to set scaling etc identically plot(mod, type=\"n\", scaling=\"sites\") text(mod, dis=\"cn\", scaling=\"sites\") points(mod, pch=21, col=\"red\", bg=\"yellow\", cex=1.2, scaling=\"sites\") text(mod, \"species\", col=\"blue\", cex=0.8, scaling=\"sites\") ## catch the invisible result and use ordiplot support - the example ## will make a biplot with arrows for species and correlation scaling pca <- rda(dune) pl <- plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) with(dune.env, points(pl, \"site\", pch=21, col=1, bg=Management)) text(pl, \"sp\", arrow=TRUE, length=0.05, col=4, cex=0.6, xpd=TRUE) with(dune.env, legend(\"bottomleft\", levels(Management), pch=21, pt.bg=1:4, bty=\"n\")) ## Configuration with pipes: make an arrow biplot plot(pca, type=\"n\", scaling=\"sites\", correlation=TRUE) |> points(\"sites\", pch=21, col = 1, cex=1.5, bg = dune.env$Management) |> text(\"species\", col = \"blue\", arrow = TRUE, xpd = TRUE) ## Scaling can be numeric or more user-friendly names ## e.g. Hill's scaling for (C)CA scrs <- scores(mod, scaling = \"sites\", hill = TRUE) ## or correlation-based scores in PCA/RDA scrs <- scores(rda(dune ~ A1 + Moisture + Management, dune.env), scaling = \"sites\", correlation = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Response Curves for Treatments with Repeated Observations — prc","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Principal Response Curves (PRC) special case Redundancy Analysis (rda) multivariate responses repeated observation design. originally suggested ecological communities. easier interpret traditional constrained ordination. can also used study effects factor depend levels factor B, + :B, multivariate response experiment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"prc(response, treatment, time, ...) # S3 method for prc summary(object, axis = 1, scaling = \"sites\", const, digits = 4, correlation = FALSE, ...) # S3 method for prc plot(x, species = TRUE, select, scaling = \"symmetric\", axis = 1, correlation = FALSE, const, type = \"l\", xlab, ylab, ylim, lty = 1:5, col = 1:6, pch, legpos, cex = 0.8, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"response Multivariate response data. Typically community (species) data. data counts, probably log transformed prior analysis. treatment factor treatments. time unordered factor defining observations times repeated design. object, x prc result object. axis Axis shown (one axis can selected). scaling Scaling species scores, identical scaling scores.rda. type scores can also specified one \"none\", \"sites\", \"species\", \"symmetric\", correspond values 0, 1, 2, 3 respectively. Argument correlation can used combination character descriptions get corresponding negative value. const General scaling constant species scores (see scores.rda details). Lower values reduce range species scores, influence regression coefficients. digits Number significant digits displayed. correlation logical; scaling character description scaling type, correlation can used select correlation-like scores PCA. See argument scaling details. species Display species scores. select Vector select displayed species. can vector indices logical vector TRUE selected species type Type plot: \"l\" lines, \"p\" points \"b\" . xlab, ylab Text replace default axis labels. ylim Limits vertical axis. lty, col, pch Line type, colour plotting characters (defaults supplied). legpos position legend. guess made supplied, NA suppress legend. cex Character expansion symbols species labels. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"PRC special case rda single factor treatment single factor time points repeated observations. vegan, corresponding rda model defined rda(response ~ treatment * time + Condition(time)). Since time appears twice model formula, main effects aliased, main effect treatment interaction terms available, used PRC. Instead usual multivariate ordination diagrams, PRC uses canonical (regression) coefficients species scores single axis. current functions provide special summary plot methods display rda results PRC fashion. current version works default contrasts (contr.treatment) coefficients contrasts first level, levels must arranged first level control ( baseline). necessary, must change baseline level function relevel. Function summary prints species scores coefficients. Function plot plots coefficients time using matplot, similar defaults. graph (PRC) meaningful first treatment level control, results contrasts first level unordered factors used. plot also displays species scores right vertical axis using function linestack. Typically number species high can displayed default settings, users can reduce character size padding (air) linestack, select subset species. legend displayed unless suppressed legpos = NA, functions tries guess put legend legpos supplied.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"function special case rda returns result object (see cca.object). However, special summary plot methods display returns differently rda.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"Jari Oksanen Cajo ter Braak","code":""},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"warning-","dir":"Reference","previous_headings":"","what":"Warning","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"first level treatment must control: use function relevel guarantee correct reference level. current version ignore user setting contrasts always use treatment contrasts (contr.treatment). time must unordered factor.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/prc.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Response Curves for Treatments with Repeated Observations — prc","text":"","code":"## Chlorpyrifos experiment and experimental design: Pesticide ## treatment in ditches (replicated) and followed over from 4 weeks ## before to 24 weeks after exposure data(pyrifos) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11)) ditch <- gl(12, 1, length=132) ## IGNORE_RDIFF_BEGIN ## PRC mod <- prc(pyrifos, dose, week) mod # RDA #> Call: prc(response = pyrifos, treatment = dose, time = week) #> #> Inertia Proportion Rank #> Total 288.9920 1.0000 #> Conditional 63.3493 0.2192 10 #> Constrained 96.6837 0.3346 44 #> Unconstrained 128.9589 0.4462 77 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 25.282 8.297 6.044 4.766 4.148 3.857 3.587 3.334 3.087 2.551 2.466 #> RDA12 RDA13 RDA14 RDA15 RDA16 RDA17 RDA18 RDA19 RDA20 RDA21 RDA22 #> 2.209 2.129 1.941 1.799 1.622 1.579 1.440 1.398 1.284 1.211 1.133 #> RDA23 RDA24 RDA25 RDA26 RDA27 RDA28 RDA29 RDA30 RDA31 RDA32 RDA33 #> 1.001 0.923 0.862 0.788 0.750 0.712 0.685 0.611 0.584 0.537 0.516 #> RDA34 RDA35 RDA36 RDA37 RDA38 RDA39 RDA40 RDA41 RDA42 RDA43 RDA44 #> 0.442 0.417 0.404 0.368 0.340 0.339 0.306 0.279 0.271 0.205 0.179 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 17.156 9.189 7.585 6.064 5.730 4.843 4.518 4.105 #> (Showing 8 of 77 unconstrained eigenvalues) #> summary(mod) # PRC #> #> Call: #> prc(response = pyrifos, treatment = dose, time = week) #> Species scores: #> Simve Daplo Cerpu Alogu Aloco Alore Aloaf Copsp #> 2.688099 1.464566 0.542739 0.280040 0.177019 0.315038 0.426524 1.169368 #> Ostsp Slyla Acrha Aloex Chysp Alona Plead Oxyte #> 2.312186 -0.556899 0.105535 0.228092 0.095042 0.063689 0.138397 0.025401 #> Grate Copdi NauLa CilHa Strvi amosp Ascmo Synsp #> 0.096840 1.428854 4.847070 0.895241 3.069709 -1.357663 0.069736 -0.026494 #> Squro Squmu Polar Kerqu Anufi Mytve Mytvi Mytmu #> 0.264390 -0.452667 0.461989 0.495348 0.432767 0.074372 0.090928 0.105891 #> Lepsp Leppa Colob Colbi Colun Lecsp Lecqu Lecco #> 0.998286 0.084809 -0.723051 -0.139569 -0.828338 0.472866 -0.088860 -0.290531 #> Leclu Lecfl Tripo Cepsp Monlo Monae Scalo Trilo #> 0.049788 -0.408035 0.215234 -0.809597 -0.527913 -0.089948 -0.077192 -0.039086 #> Tripo.1 Tricy Trisp Tepat Rotne Notla Filsp Lopox #> 0.246435 0.335400 0.078423 0.007719 0.143730 -0.114301 -0.168356 -0.030990 #> hydrspec bothrosp olchaeta erpoocto glsicomp alglhete hebdstag sphidae #> -0.048698 0.398665 -1.165154 -0.901030 -0.144389 -0.073049 0.902223 1.463655 #> ansuvote armicris bathcont binitent gyraalbu hippcomp lymnstag lymnaes7 #> 0.140685 1.680010 0.073282 -1.950500 0.033051 0.404473 -0.263679 0.135150 #> physfont plbacorn popyanti radiovat radipere valvcris valvpisc hycarina #> -0.026383 0.084761 1.272223 -0.019815 -0.625468 0.010579 -0.267577 1.044034 #> gammpule aselaqua proameri collembo caenhora caenluct caenrobu cloedipt #> 1.526450 1.578743 0.116577 0.029906 5.767844 2.376188 0.126181 4.734035 #> cloesimi aeshniae libellae conagrae corident coripanz coripunc cymabons #> 1.242207 0.212699 -0.081867 1.630574 0.013761 0.120439 -0.176746 0.046360 #> hesplinn hespsahl notoglau notomacu notoobli notoviri pacoconc pleaminu #> -0.069465 0.033240 0.555201 0.050060 0.082356 0.215863 0.016709 0.071168 #> sigadist sigafall sigastri sigarasp colyfusc donacis6 gyrimari haliconf #> 0.076479 -0.018364 0.060206 -0.277312 0.036493 -0.078608 0.010579 0.446911 #> haliflav haligruf haliobli herubrev hya_herm hyglpusi hyhyovat hypoplan #> 0.044538 0.450146 0.131472 -0.128486 -0.322379 -0.011775 0.024196 0.014976 #> hyporusp hytuinae hytuvers laphminu noteclav rhantusp sialluta ablalong #> 0.232042 2.316179 1.772351 0.632897 0.007912 0.067618 1.109341 0.014976 #> ablaphmo cltanerv malopisp mopetenu prdiussp pstavari chironsp crchirsp #> 2.992695 0.075631 0.047658 0.008716 0.554911 0.082842 1.889916 0.016709 #> crclglat ditendsp mitegchl pachgarc pachgvit popegnub popedisp acriluce #> 0.028953 0.083481 0.230629 -0.012187 -0.029907 0.224272 -0.069649 -0.007950 #> chclpige conescut cricotsp liesspec psclbarb psclgsli psclobvi psclplat #> -0.008744 0.821036 0.121530 0.107387 -0.028639 0.601568 -0.362378 -0.052054 #> psclpsil pscladsp cladotsp laa_spec patanysp tatarssp zaa_spec anopmacu #> 0.007339 0.005674 0.539894 0.034105 0.146807 0.669430 0.049949 -0.163731 #> cepogoae chaoobsc cucidae4 tabanusp agdasphr athrater cyrncren holodubi #> 2.555403 2.442310 0.033240 -0.011601 0.271815 0.067618 0.071168 0.094754 #> holopici leceriae lilurhom monaangu mystazur mystloni oecefurv oecelacu #> 0.611618 0.298633 0.009063 0.644295 0.033240 2.998460 0.536628 0.259064 #> triabico paponysp #> 0.088915 0.097788 #> #> Coefficients for dose + week:dose interaction #> which are contrasts to dose 0 #> rows are dose, columns are week #> -4 -1 0.1 1 2 4 8 12 15 #> 0.1 -0.07218 -0.1375 -0.1020 -0.04068 -0.2101 -0.1364 -0.08077 -0.1536 -0.1123 #> 0.9 -0.08106 -0.1935 -0.1936 -0.47699 -0.4977 -0.4306 -0.13532 -0.3548 -0.2408 #> 6 -0.16616 -0.1232 -0.4539 -1.15638 -1.0835 -1.1511 -0.56112 -0.4698 -0.3078 #> 44 -0.13979 -0.1958 -0.7308 -1.26088 -1.2978 -1.4627 -1.29139 -1.0081 -0.7819 #> 19 24 #> 0.1 -0.2163 -0.07835 #> 0.9 -0.1756 -0.15442 #> 6 -0.3293 -0.18227 #> 44 -0.5768 -0.31022 logabu <- colSums(pyrifos) plot(mod, select = logabu > 100) ## IGNORE_RDIFF_END ## Ditches are randomized, we have a time series, and are only ## interested in the first axis ctrl <- how(plots = Plots(strata = ditch,type = \"free\"), within = Within(type = \"series\"), nperm = 99) anova(mod, permutations = ctrl, first=TRUE) #> Permutation test for rda under reduced model #> Plots: ditch, plot permutation: free #> Permutation: series #> Number of permutations: 99 #> #> Model: prc(response = pyrifos, treatment = dose, time = week) #> Df Variance F Pr(>F) #> RDA1 1 25.282 15.096 0.01 ** #> Residual 77 128.959 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function predict can used find site species scores estimates response data new data sets, Function calibrate estimates values constraints new data set. Functions fitted residuals return estimates response data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"# S3 method for cca fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\"), type = c(\"response\", \"working\"), ...) # S3 method for capscale fitted(object, model = c(\"CCA\", \"CA\", \"pCCA\", \"Imaginary\"), type = c(\"response\", \"working\"), ...) # S3 method for cca residuals(object, ...) # S3 method for cca predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", hill = FALSE, ...) # S3 method for rda predict(object, newdata, type = c(\"response\", \"wa\", \"sp\", \"lc\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", correlation = FALSE, const, ...) # S3 method for dbrda predict(object, newdata, type = c(\"response\", \"lc\", \"wa\", \"working\"), rank = \"full\", model = c(\"CCA\", \"CA\"), scaling = \"none\", const, ...) # S3 method for cca calibrate(object, newdata, rank = \"full\", ...) # S3 method for cca coef(object, norm = FALSE, ...) # S3 method for decorana predict(object, newdata, type = c(\"response\", \"sites\", \"species\"), rank = 4, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"object result object cca, rda, dbrda, capscale decorana. model Show constrained (\"CCA\"), unconstrained (\"CA\") conditioned “partial” (\"pCCA\") results. fitted method capscale can also \"Imaginary\" imaginary components negative eigenvalues newdata New data frame used prediction calibration. Usually new community data frame, type = \"lc\" constrained component type = \"response\" type = \"working\" must data frame constraints. newdata must number rows original community data cca result type = \"response\" type = \"working\". original model row column names, new data must contain rows columns names (row names species scores, column names \"wa\" scores constraint names \"lc\" scores). cases rows columns must match directly. argument implemented \"wa\" scores dbrda. type type prediction, fitted values residuals: \"response\" scales results ordination gives results, \"working\" gives values used internally, Chi-square standardization cca scaling centring rda. capscale dbrda \"response\" gives dissimilarities, \"working\" internal data structure analysed ordination. Alternative \"wa\" gives site scores weighted averages community data, \"lc\" site scores linear combinations environmental data, \"sp\" species scores. predict.decorana alternatives scores \"sites\" \"species\". rank rank number axes used approximation. default use axes (full rank) \"model\" available four axes predict.decorana. scaling logical, character, numeric; Scaling predicted scores meaning cca, rda, dbrda, capscale. See scores.cca details acceptable values. correlation, hill logical; correlation-like scores Hill's scaling appropriate RDA CCA respectively. See scores.cca additional details. const Constant multiplier RDA scores. used scaling FALSE, default value give similar scaling scores.rda. norm Coefficients variables centred scaled unit norm. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Function fitted gives approximation original data matrix dissimilarities ordination result either scale response scaled internally function. Function residuals gives approximation original data unconstrained ordination. argument type = \"response\" fitted.cca residuals.cca function give marginal totals original data matrix, fitted residuals add original data. Functions fitted residuals dbrda capscale give dissimilarities type = \"response\", additive. However, \"working\" scores additive capscale ( dbrda). fitted residuals capscale dbrda include additive constant requested function call. variants fitted residuals defined model mod <- cca(y ~ x), cca(fitted(mod)) equal constrained ordination, cca(residuals(mod)) equal unconstrained part ordination. Function predict can find estimate original data matrix dissimilarities (type = \"response\") rank. rank = \"full\" identical fitted. addition, function can find species scores site scores community data matrix cca rda. function can used new data, can used add new species site scores existing ordinations. function returns (weighted) orthonormal scores default, must specify explicit scaling add scores ordination diagrams. type = \"wa\" function finds site scores species scores. case, new data can contain new sites, species must match original new data. type=\"sp\" function finds species scores site constraints (linear combination scores). case new data can contain new species, sites must match original new data. type = \"lc\" function finds linear combination scores sites environmental data. case new data frame must contain constraining conditioning environmental variables model formula. type = \"response\" type = \"working\" new data must contain environmental variables constrained component desired, community data matrix residual unconstrained component desired. types, function uses newdata find new \"lc\" (constrained) \"wa\" scores (unconstrained) finds response working data new row scores species scores. original site (row) species (column) weights used type = \"response\" type = \"working\" correspondence analysis (cca) therefore number rows must match original data newdata. completely new data frame created, extreme care needed defining variables similarly original model, particular (ordered) factors. ordination performed formula interface, newdata can data frame matrix, extreme care needed columns match original newdata. Function calibrate.cca finds estimates constraints community ordination \"wa\" scores cca, rda capscale. often known calibration, bioindication environmental reconstruction. Basically, method similar projecting site scores onto biplot arrows, uses regression coefficients. function can called newdata cross-validation possible. newdata may contain new sites, species must match original new data. function work ‘partial’ models Condition term, used newdata capscale dbrda results. results may interpretable continuous variables. Function coef give regression coefficients centred environmental variables (constraints conditions) linear combination scores. coefficients unstandardized environmental variables. coefficients NA aliased effects. Function predict.decorana similar predict.cca. However, type = \"species\" available detrended correspondence analysis (DCA), detrending destroys mutual reciprocal averaging (except first axis rescaling used). Detrended CA attempt approximate original data matrix, type = \"response\" meaning detrended analysis (except rank = 1).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"functions return matrices, vectors dissimilarities appropriate.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Greenacre, M. J. (1984). Theory applications correspondence analysis. Academic Press, London.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/predict.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Prediction Tools for [Constrained] Ordination (CCA, RDA, DCA, CA, PCA) — predict.cca","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ A1 + Management + Condition(Moisture), data=dune.env) # Definition of the concepts 'fitted' and 'residuals' mod #> Call: cca(formula = dune ~ A1 + Management + Condition(Moisture), data #> = dune.env) #> #> Inertia Proportion Rank #> Total 2.1153 1.0000 #> Conditional 0.6283 0.2970 3 #> Constrained 0.5109 0.2415 4 #> Unconstrained 0.9761 0.4615 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for constrained axes: #> CCA1 CCA2 CCA3 CCA4 #> 0.24932 0.12090 0.08160 0.05904 #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> cca(fitted(mod)) #> Call: cca(X = fitted(mod)) #> #> Inertia Rank #> Total 0.5109 #> Unconstrained 0.5109 4 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 #> 0.24932 0.12090 0.08160 0.05904 #> cca(residuals(mod)) #> Call: cca(X = residuals(mod)) #> #> Inertia Rank #> Total 0.9761 #> Unconstrained 0.9761 12 #> Inertia is scaled Chi-square #> #> Eigenvalues for unconstrained axes: #> CA1 CA2 CA3 CA4 CA5 CA6 CA7 CA8 CA9 CA10 #> 0.30637 0.13191 0.11516 0.10947 0.07724 0.07575 0.04871 0.03758 0.03106 0.02102 #> CA11 CA12 #> 0.01254 0.00928 #> # Remove rare species (freq==1) from 'cca' and find their scores # 'passively'. freq <- specnumber(dune, MARGIN=2) freq #> Achimill Agrostol Airaprae Alopgeni Anthodor Bellpere Bromhord Chenalbu #> 7 10 2 8 6 6 5 1 #> Cirsarve Comapalu Eleopalu Elymrepe Empenigr Hyporadi Juncarti Juncbufo #> 1 2 5 6 1 3 5 4 #> Lolipere Planlanc Poaprat Poatriv Ranuflam Rumeacet Sagiproc Salirepe #> 12 7 14 13 6 5 7 3 #> Scorautu Trifprat Trifrepe Vicilath Bracruta Callcusp #> 18 3 16 3 15 3 mod <- cca(dune[, freq>1] ~ A1 + Management + Condition(Moisture), dune.env) ## IGNORE_RDIFF_BEGIN predict(mod, type=\"sp\", newdata=dune[, freq==1], scaling=\"species\") #> CCA1 CCA2 CCA3 CCA4 #> Chenalbu 1.5737337 0.7842538 0.5503660 -0.35108333 #> Cirsarve 0.5945146 0.3714228 -0.2862647 -0.88373727 #> Empenigr -1.8771953 0.9904299 -0.2446222 -0.04858656 # New sites predict(mod, type=\"lc\", new=data.frame(A1 = 3, Management=\"NM\", Moisture=\"2\"), scal=2) #> CCA1 CCA2 CCA3 CCA4 #> 1 -2.38829 1.230652 -0.2363485 0.3338258 # Calibration and residual plot mod <- cca(dune ~ A1 + Moisture, dune.env) pred <- calibrate(mod) pred #> A1 Moisture.L Moisture.Q Moisture.C #> 1 2.2630533 -0.62633470 -0.20456759 0.220761764 #> 2 4.0510042 -0.47341146 -0.36986691 0.474939409 #> 3 4.2752294 -0.07214500 -0.60797514 0.303213289 #> 4 4.5398659 0.03192745 -1.12417368 0.932223234 #> 5 5.0409406 -0.84235946 0.43000738 -0.291599200 #> 6 5.1962100 -0.91316862 1.11354235 -0.804453944 #> 7 4.2452549 -0.76452556 0.60464291 -0.484842066 #> 8 5.0208369 0.43886340 0.08169514 0.132995916 #> 9 4.2663219 0.10720486 -0.34067849 -0.675151598 #> 10 4.0411356 -0.65472729 0.02832164 0.558402684 #> 11 2.8280051 -0.45762457 0.63079135 -0.089977975 #> 12 5.1204137 0.36328912 -0.69118581 -0.665622948 #> 13 4.9034218 0.47069541 -0.54378271 -0.118643453 #> 14 11.6455841 0.60920550 0.78341426 0.532852308 #> 15 10.7829689 0.69208513 0.82190786 0.237311062 #> 16 7.9892176 0.96421599 0.46793089 0.373647014 #> 17 0.9218684 -0.15822891 0.14593271 1.189161582 #> 18 3.1680733 -0.41737900 1.03352732 -0.236938282 #> 19 -1.2003506 0.57033354 0.72777285 0.509955590 #> 20 4.7876770 1.00324330 1.49898460 0.009202396 ## IGNORE_RDIFF_END with(dune.env, plot(A1, pred[,\"A1\"] - A1, ylab=\"Prediction Error\")) abline(h=0)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":null,"dir":"Reference","previous_headings":"","what":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes rotates configuration maximum similarity another configuration. Function protest tests non-randomness (significance) two configurations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"procrustes(X, Y, scale = TRUE, symmetric = FALSE, scores = \"sites\", ...) # S3 method for procrustes summary(object, digits = getOption(\"digits\"), ...) # S3 method for procrustes plot(x, kind=1, choices=c(1,2), to.target = TRUE, type = \"p\", xlab, ylab, main, ar.col = \"blue\", length=0.05, cex = 0.7, ...) # S3 method for procrustes points(x, display = c(\"target\", \"rotated\"), choices = c(1,2), truemean = FALSE, ...) # S3 method for procrustes text(x, display = c(\"target\", \"rotated\"), choices = c(1,2), labels, truemean = FALSE, ...) # S3 method for procrustes lines(x, type = c(\"segments\", \"arrows\"), choices = c(1, 2), truemean = FALSE, ...) # S3 method for procrustes residuals(object, ...) # S3 method for procrustes fitted(object, truemean = TRUE, ...) # S3 method for procrustes predict(object, newdata, truemean = TRUE, ...) protest(X, Y, scores = \"sites\", permutations = how(nperm = 999), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"X Target matrix Y Matrix rotated. scale Allow scaling axes Y. symmetric Use symmetric Procrustes statistic (rotation still non-symmetric). scores Kind scores used. display argument used corresponding scores function: see scores, scores.cca scores.cca alternatives. x, object object class procrustes. digits Number digits output. kind plot function, kind plot produced: kind = 1 plots shifts two configurations, kind = 0 draws corresponding empty plot, kind = 2 plots impulse diagram residuals. choices Axes (dimensions) plotted. xlab, ylab Axis labels, defaults unacceptable. main Plot title, default unacceptable. display Show \"target\" \"rotated\" matrix points. .target Draw arrows point target. type type plot drawn. plot, type can \"points\" \"text\" select marker tail arrow, \"none\" drawing empty plot. lines type selects either arrows line segments connect target rotated configuration. truemean Use original range target matrix instead centring fitted values. Function plot.procrustes needs truemean = FALSE, adding graphical items plots original results may need truemean = TRUE. newdata Matrix coordinates rotated translated target. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. ar.col Arrow colour. length Width arrow head. labels Character vector text labels. Rownames result object used default. cex Character expansion points text. ... parameters passed functions. procrustes protest parameters passed scores, graphical functions underlying graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Procrustes rotation rotates matrix maximum similarity target matrix minimizing sum squared differences. Procrustes rotation typically used comparison ordination results. particularly useful comparing alternative solutions multidimensional scaling. scale=FALSE, function rotates matrix Y. scale=TRUE, scales linearly configuration Y maximum similarity. Since Y scaled fit X, scaling non-symmetric. However, symmetric=TRUE, configurations scaled equal dispersions symmetric version Procrustes statistic computed. Instead matrix, X Y can results ordination scores can extract results. Function procrustes passes extra arguments scores, scores.cca etc. can specify arguments scaling. Function plot plots procrustes object returns invisibly ordiplot object function identify.ordiplot can used identifying points. items ordiplot object called heads points kind=1 (ordination diagram) sites kind=2 (residuals). ordination diagrams, arrow heads point target configuration .target = TRUE, rotated configuration .target = FALSE. Target original rotated axes shown cross hairs two-dimensional Procrustes analysis, higher number dimensions, rotated axes projected onto plot scaled centred range. Function plot passes parameters underlying plotting functions. full control plots, can draw axes using plot kind = 0, add items points lines. functions pass parameters underlying functions can select plotting characters, size, colours etc., can select width, colour type line segments arrows, can select orientation head width arrows. Function residuals returns pointwise residuals, fitted fitted values, either centred zero mean (truemean=FALSE) original scale ( hardly make sense symmetric = TRUE). addition, summary print methods. matrix X lower number columns matrix Y, matrix X filled zero columns match dimensions. means function can used rotate ordination configuration environmental variable ( practically extracting result fitted function). Function predict can used add new rotated coordinates target. predict function always translate coordinates original non-centred matrix. function used newdata symmetric analysis. Function protest performs symmetric Procrustes analysis repeatedly estimate significance Procrustes statistic. Function protest uses correlation-like statistic derived symmetric Procrustes sum squares \\(ss\\) \\(r =\\sqrt{1-ss}\\), also prints sum squares symmetric analysis, sometimes called \\(m_{12}^2\\). Function protest print method, otherwise uses procrustes methods. Thus plot protest object yields Procrustean superimposition plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Function procrustes returns object class procrustes items. Function protest inherits procrustes, amends new items: Yrot Rotated matrix Y. X Target matrix. ss Sum squared differences X Yrot. rotation Orthogonal rotation matrix. translation Translation origin. scale Scaling factor. xmean centroid target. symmetric Type ss statistic. call Function call. t0 following items class protest: Procrustes correlation non-permuted solution. t Procrustes correlations permutations. distribution correlations can inspected permustats function. signif Significance t permutations Number permutations. control list control values permutations returned function . control list passed argument control describing permutation design.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate Analysis. Academic Press. Peres-Neto, P.R. Jackson, D.. (2001). well multivariate data sets match? advantages Procrustean superimposition approach Mantel test. Oecologia 129: 169-178.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"function protest follows Peres-Neto & Jackson (2001), implementation still Mardia et al. (1979).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/procrustes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Procrustes Rotation of Two Configurations and PROTEST — procrustes","text":"","code":"## IGNORE_RDIFF_BEGIN data(varespec) vare.dist <- vegdist(wisconsin(varespec)) mds.null <- monoMDS(vare.dist, y = cmdscale(vare.dist)) mds.alt <- monoMDS(vare.dist) vare.proc <- procrustes(mds.alt, mds.null) vare.proc #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Procrustes sum of squares: #> 11.17 #> summary(vare.proc) #> #> Call: #> procrustes(X = mds.alt, Y = mds.null) #> #> Number of objects: 24 Number of dimensions: 2 #> #> Procrustes sum of squares: #> 11.17448 #> Procrustes root mean squared error: #> 0.6823512 #> Quantiles of Procrustes errors: #> Min 1Q Median 3Q Max #> 0.1642438 0.2425785 0.2783603 0.4983976 2.4447632 #> #> Rotation matrix: #> [,1] [,2] #> [1,] 0.99937107 -0.03546065 #> [2,] 0.03546065 0.99937107 #> #> Translation of averages: #> [,1] [,2] #> [1,] -1.713781e-17 1.56769e-17 #> #> Scaling of target: #> [1] 0.7310245 #> plot(vare.proc) plot(vare.proc, kind=2) residuals(vare.proc) #> 18 15 24 27 23 19 22 16 #> 0.2734040 0.2032392 0.4708118 0.4420710 0.3547337 0.1642438 0.2515286 0.2611623 #> 28 13 14 20 25 7 5 6 #> 0.7773604 0.3075051 0.2833167 0.1749943 0.2684784 0.5167965 0.9747233 0.2437827 #> 3 4 2 9 12 10 11 21 #> 0.2252071 0.7586954 2.4447632 0.2389659 0.2093104 0.2597721 0.4922646 1.0884414 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":null,"dir":"Reference","previous_headings":"","what":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data log transformed abundances aquatic invertebrate twelve ditches studied eleven times insecticide treatment.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos)"},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data frame 132 observations log-transformed (log(10*x + 1)) abundances 178 species. twelve sites (ditches, mesocosms), studied repeatedly eleven occasions. treatment levels, treatment times, ditch ID's data frame, data regular, example shows obtain external variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"data set obtained experiment outdoor experimental ditches. Twelve mesocosms allocated random treatments; four served controls, remaining eight treated insecticide chlorpyrifos, nominal dose levels 0.1, 0.9, 6, 44 \\(\\mu\\)g/ L two mesocosms . example data set invertebrates. Sampling done 11 times, week -4 pre-treatment week 24 post-treatment, giving total 132 samples (12 mesocosms times 11 sampling dates), see van den Brink & ter Braak (1999) details. data set contains species data, example shows obtain treatment, time ditch ID variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"CANOCO 4 example data, permission Cajo J. F. ter Braak.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"van den Brink, P.J. & ter Braak, C.J.F. (1999). Principal response curves: Analysis time-dependent multivariate responses biological community stress. Environmental Toxicology Chemistry, 18, 138--148.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/pyrifos.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Response of Aquatic Invertebrates to Insecticide Treatment — pyrifos","text":"","code":"data(pyrifos) ditch <- gl(12, 1, length=132) week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24)) dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank -- Abundance or Dominance / Diversity Models — radfit","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions construct rank -- abundance dominance / diversity Whittaker plots fit brokenstick, preemption, log-Normal, Zipf Zipf-Mandelbrot models species abundance.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"# S3 method for default radfit(x, ...) rad.null(x, family=poisson, ...) rad.preempt(x, family = poisson, ...) rad.lognormal(x, family = poisson, ...) rad.zipf(x, family = poisson, ...) rad.zipfbrot(x, family = poisson, ...) # S3 method for radline predict(object, newdata, total, ...) # S3 method for radfit plot(x, BIC = FALSE, legend = TRUE, ...) # S3 method for radfit.frame plot(x, order.by, BIC = FALSE, model, legend = TRUE, as.table = TRUE, ...) # S3 method for radline plot(x, xlab = \"Rank\", ylab = \"Abundance\", type = \"b\", ...) radlattice(x, BIC = FALSE, ...) # S3 method for radfit lines(x, ...) # S3 method for radfit points(x, ...) as.rad(x) # S3 method for rad plot(x, xlab = \"Rank\", ylab = \"Abundance\", log = \"y\", ...)"},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"x Data frame, matrix vector giving species abundances, object plotted. family Error distribution (passed glm). alternatives accepting link = \"log\" family can used, although make sense. object fitted result object. newdata Ranks used ordinations. models can interpolate non-integer “ranks” (although may approximate), extrapolation may fail total new total used predicting abundance. Observed total count used omitted. order.vector used ordering sites plots. BIC Use Bayesian Information Criterion, BIC, instead Akaike's AIC. penalty BIC \\(k = \\log(S)\\) \\(S\\) number species, whereas AIC uses \\(k = 2\\). model Show specified model. missing, AIC used select model. model names (can abbreviated) Null, Preemption, Lognormal, Zipf, Mandelbrot. legend Add legend line colours. .table Arrange panels starting upper left corner (passed xyplot). xlab,ylab Labels x y axes. type Type plot, \"b\" plotting observed points fitted lines, \"p\" points, \"l\" fitted lines, \"n\" setting frame. log Use logarithmic scale given axis. default log = \"y\" gives traditional plot community ecology preemption model straight line, log = \"xy\" Zipf model straight line. log = \"\" axes original arithmetic scale. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Rank--Abundance Dominance (RAD) Dominance/Diversity plots (Whittaker 1965) display logarithmic species abundances species rank order. plots supposed effective analysing types abundance distributions communities. functions fit popular models mainly following Wilson (1991). Functions rad.null, rad.preempt, rad.lognormal, rad.zipf zipfbrot fit individual models (described ) single vector (row data frame), function radfit fits models. argument function radfit can either vector single community data frame row represents distinct community. Function rad.null fits brokenstick model expected abundance species rank \\(r\\) \\(a_r = (J/S) \\sum_{x=r}^S (1/x)\\) (Pielou 1975), \\(J\\) total number individuals (site total) \\(S\\) total number species community. gives Null model individuals randomly distributed among observed species, fitted parameters. Function rad.preempt fits niche preemption model, .k.. geometric series Motomura model, expected abundance \\(\\) species rank \\(r\\) \\(a_r = J \\alpha (1 - \\alpha)^{r-1}\\). estimated parameter preemption coefficient \\(\\alpha\\) gives decay rate abundance per rank. niche preemption model straight line RAD plot. Function rad.lognormal fits log-Normal model assumes logarithmic abundances distributed Normally, \\(a_r = \\exp( \\log \\mu + \\log \\sigma N)\\), \\(N\\) Normal deviate. Function rad.zipf fits Zipf model \\(a_r = J p_1 r^\\gamma\\) \\(p_1\\) fitted proportion abundant species, \\(\\gamma\\) decay coefficient. Zipf--Mandelbrot model (rad.zipfbrot) adds one parameter: \\(a_r = J c (r + \\beta)^\\gamma\\) \\(p_1\\) Zipf model changes meaningless scaling constant \\(c\\). Log-Normal Zipf models generalized linear models (glm) logarithmic link function. Zipf--Mandelbrot adds one nonlinear parameter Zipf model, fitted using nlm nonlinear parameter estimating parameters log-Likelihood glm. Preemption model fitted purely nonlinear model. estimated parameters Null model. default family poisson appropriate genuine counts (integers), families accept link = \"log\" can used. Families Gamma gaussian may appropriate abundance data, cover. best model selected AIC. Therefore ‘quasi’ families quasipoisson used: AIC log-Likelihood needed non-linear models. functions plot functions. radfit applied data frame, plot uses Lattice graphics, plot functions use ordinary graphics. ordinary graphics functions return invisibly ordiplot object observed points, function identify.ordiplot can used label selected species. Alternatively, radlattice uses Lattice graphics display radfit model single site separate panel together AIC BIC values. Function .rad base function construct ordered RAD data. plot used RAD plot functions pass extra arguments (xlab log) function. function returns ordered vector taxa occurring site, corresponding attribute \"index\" included taxa.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Functions rad.null, rad.preempt, rad.lognormal, zipf zipfbrot fit single RAD model single site. result object class \"radline\" inherits glm, can handled ( ) glm methods. Function radfit fits models either single site rows data frame matrix. fitted single site, function returns object class \"radfit\" items y (observed values), family, models list fitted \"radline\" models. applied data frame matrix, radfit function returns object class \"radfit.frame\" list \"radfit\" objects, item names corresponding row name. result objects (\"radline\", \"radfit\", \"radfit.frame\") can accessed method functions. following methods available: AIC, coef, deviance, logLik. addition fit results can accessed fitted, predict residuals (inheriting residuals.glm). graphical functions discussed Details.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Pielou, E.C. (1975) Ecological Diversity. Wiley & Sons. Preston, F.W. (1948) commonness rarity species. Ecology 29, 254--283. Whittaker, R. H. (1965) Dominance diversity plant communities. Science 147, 250--260. Wilson, J. B. (1991) Methods fitting dominance/diversity curves. Journal Vegetation Science 2, 35--46.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"RAD models usually fitted proportions instead original abundances. However, nothing models seems require division abundances site totals, original observations used functions. wish use proportions, must standardize data site totals, e.g. decostand use appropriate family Gamma. lognormal model fitted standard way, think quite correct -- least equivalent fitting Normal density log abundances like originally suggested (Preston 1948). models may fail. particular, estimation Zipf-Mandelbrot model difficult. fitting fails, NA returned. Wilson (1991) defined preemption model \\(a_r = J p_1 (1 - \\alpha)^{r-1}\\), \\(p_1\\) fitted proportion first species. However, parameter \\(p_1\\) completely defined \\(\\alpha\\) since fitted proportions must add one, therefore handle preemption one-parameter model. Veiled log-Normal model included earlier releases function, removed flawed: implicit veil line also appears ordinary log-Normal. latest release version rad.veil 1.6-10.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/radfit.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rank -- Abundance or Dominance / Diversity Models — radfit","text":"","code":"data(BCI) mod <- rad.lognormal(BCI[5,]) mod #> #> RAD model: Log-Normal #> Family: poisson #> No. of species: 101 #> Total abundance: 505 #> #> log.mu log.sigma Deviance AIC BIC #> 0.951926 1.165929 17.077549 317.656487 322.886728 plot(mod) mod <- radfit(BCI[1,]) ## Standard plot overlaid for all models ## Preemption model is a line plot(mod) ## log for both axes: Zipf model is a line plot(mod, log = \"xy\") ## Lattice graphics separately for each model radlattice(mod) # Take a subset of BCI to save time and nerves mod <- radfit(BCI[3:5,]) mod #> #> Deviance for RAD models: #> #> 3 4 5 #> Null 86.1127 49.8111 80.855 #> Preemption 58.9295 39.7817 76.311 #> Lognormal 29.2719 16.6588 17.078 #> Zipf 50.1262 47.9108 30.936 #> Mandelbrot 5.7342 5.5665 10.573 plot(mod, pch=\".\")"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":null,"dir":"Reference","previous_headings":"","what":"Compares Dissimilarity Indices for Gradient Detection — rankindex","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Rank correlations dissimilarity indices gradient separation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"rankindex(grad, veg, indices = c(\"euc\", \"man\", \"gow\", \"bra\", \"kul\"), stepacross = FALSE, method = \"spearman\", metric = c(\"euclidean\", \"mahalanobis\", \"manhattan\", \"gower\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"grad gradient variable matrix. veg community data matrix. indices Dissimilarity indices compared, partial matches alternatives vegdist. Alternatively, can (named) list functions returning objects class 'dist'. stepacross Use stepacross find shorter path dissimilarity. dissimilarities site pairs shared species set NA using .shared indices fixed upper limit can also analysed. method Correlation method used. metric Metric evaluate gradient separation. See Details. ... parameters stepacross.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"good dissimilarity index multidimensional scaling high rank-order similarity gradient separation. function compares indices vegdist gradient separation using rank correlation coefficients cor. gradient separation point assessed using given metric. default use Euclidean distance continuous variables scaled unit variance, use Gower metric mixed data using function daisy grad factors. alternatives Mahalanabis distances based grad matrix scaled columns orthogonal (uncorrelated) unit variance, Manhattan distances grad variables scaled unit range. indices argument can accept dissimilarity indices besides ones calculated vegdist function. , argument value (possibly named) list functions. function must return valid 'dist' object dissimilarities, similarities accepted converted dissimilarities beforehand.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Returns named vector rank correlations.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Faith, F.P., Minchin, P.R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57-68.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"Jari Oksanen, additions Peter Solymos","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"several problems using rank correlation coefficients. Typically many ties \\(n(n-1)/2\\) gradient separation values derived just \\(n\\) observations. Due floating point arithmetics, many tied values differ machine epsilon arbitrarily ranked differently rank used cor.test. Two indices identical certain transformation standardization may differ slightly (magnitude \\(10^{-15}\\)) may lead third fourth decimal instability rank correlations. Small differences rank correlations taken seriously. Probably method replaced sounder method, yet know ... may experiment mantel, anosim even protest. Earlier version function used method = \"kendall\", far slow large data sets. functions returning dissimilarity objects self contained, ... argument passes additional parameters stepacross functions supplied via indices argument.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rankindex.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Compares Dissimilarity Indices for Gradient Detection — rankindex","text":"","code":"data(varespec) data(varechem) ## The variables are automatically scaled rankindex(varechem, varespec) #> euc man gow bra kul #> 0.2396330 0.2735087 0.2288358 0.2837910 0.2839834 rankindex(varechem, wisconsin(varespec)) #> euc man gow bra kul #> 0.4200990 0.4215642 0.3708606 0.4215642 0.4215642 ## Using non vegdist indices as functions funs <- list(Manhattan=function(x) dist(x, \"manhattan\"), Gower=function(x) cluster:::daisy(x, \"gower\"), Ochiai=function(x) designdist(x, \"1-J/sqrt(A*B)\")) rankindex(scale(varechem), varespec, funs) #> Manhattan Gower Ochiai #> 0.2735087 0.2288358 0.1696862"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":null,"dir":"Reference","previous_headings":"","what":"Rarefaction Species Richness — rarefy","title":"Rarefaction Species Richness — rarefy","text":"Rarefied species richness community ecologists.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rarefaction Species Richness — rarefy","text":"","code":"rarefy(x, sample, se = FALSE, MARGIN = 1) rrarefy(x, sample) drarefy(x, sample) rarecurve(x, step = 1, sample, xlab = \"Sample Size\", ylab = \"Species\", label = TRUE, col, lty, tidy = FALSE, ...) rareslope(x, sample)"},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rarefaction Species Richness — rarefy","text":"x Community data, matrix-like object vector. MARGIN Margin index computed. sample Subsample size rarefying community, either single value vector. se Estimate standard errors. step Step size sample sizes rarefaction curves. xlab, ylab Axis labels plots rarefaction curves. label Label rarefaction curves rownames x (logical). col, lty plotting colour line type, see par. Can vector length nrow(x), one per sample, extended length internally. tidy Instead drawing plot, return “tidy” data frame can used ggplot2 graphics. data frame variables Site (factor), Sample Species. ... Parameters passed nlm, plot, lines ordilabel rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Rarefaction Species Richness — rarefy","text":"Function rarefy gives expected species richness random subsamples size sample community. size sample smaller total community size, function work larger sample well (warning) return non-rarefied species richness (standard error = 0). sample vector, rarefaction observations performed sample size separately. Rarefaction can performed genuine counts individuals. function rarefy based Hurlbert's (1971) formulation, standard errors Heck et al. (1975). Function rrarefy generates one randomly rarefied community data frame vector given sample size. sample can vector giving sample sizes row. sample size equal larger observed number individuals, non-rarefied community returned. random rarefaction made without replacement variance rarefied communities rather related rarefaction proportion size sample. Random rarefaction sometimes used remove effects different sample sizes. usually bad idea: random rarefaction discards valid data, introduces random error reduces quality data (McMurdie & Holmes 2014). better use normalizing transformations (decostand vegan) possible variance stabilization (decostand dispweight vegan) methods sensitive sample sizes. Function drarefy returns probabilities species occur rarefied community size sample. sample can vector giving sample sizes row. sample equal larger observed number individuals, observed species sampling probability 1. Function rarecurve draws rarefaction curve row input data. rarefaction curves evaluated using interval step sample sizes, always including 1 total sample size. sample specified, vertical line drawn sample horizontal lines rarefied species richnesses. Function rareslope calculates slope rarecurve (derivative rarefy) given sample size; sample need integer. Rarefaction functions used observed counts. think necessary use multiplier data, rarefy first multiply. Removing rare species rarefaction can also give biased results. Observed count data normally include singletons (species count 1), missing, functions issue warnings. may false positives, recommended check observed counts multiplied rare taxa removed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rarefaction Species Richness — rarefy","text":"vector rarefied species richness values. single sample se = TRUE, function rarefy returns 2-row matrix rarefied richness (S) standard error (se). sample vector rarefy, function returns matrix column sample size, se = TRUE, rarefied richness standard error consecutive lines. Function rarecurve returns invisible list rarefy results corresponding drawn curve. Alternatively, tidy = TRUE returns data frame can used ggplot2 graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Rarefaction Species Richness — rarefy","text":"Heck, K.L., van Belle, G. & Simberloff, D. (1975). Explicit calculation rarefaction diversity measurement determination sufficient sample size. Ecology 56, 1459--1461. Hurlbert, S.H. (1971). nonconcept species diversity: critique alternative parameters. Ecology 52, 577--586. McMurdie, P.J. & Holmes, S. (2014). Waste , want : rarefying microbiome data inadmissible. PLoS Comput Biol 10(4): e1003531. doi:10.1371/journal.pcbi.1003531","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Rarefaction Species Richness — rarefy","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/rarefy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Rarefaction Species Richness — rarefy","text":"","code":"data(BCI) S <- specnumber(BCI) # observed number of species (raremax <- min(rowSums(BCI))) #> [1] 340 Srare <- rarefy(BCI, raremax) plot(S, Srare, xlab = \"Observed No. of Species\", ylab = \"Rarefied No. of Species\") abline(0, 1) rarecurve(BCI, step = 20, sample = raremax, col = \"blue\", cex = 0.6)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":null,"dir":"Reference","previous_headings":"","what":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Function finds Raup-Crick dissimilarity probability number co-occurring species species occurrence probabilities proportional species frequencies.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"raupcrick(comm, null = \"r1\", nsimul = 999, chase = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"comm Community data treated presence/absence data. null Null model used method oecosimu. nsimul Number null communities assessing dissimilarity index. chase Use Chase et al. (2011) method tie handling ( recommended except comparing results Chase script). ... parameters passed oecosimu.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Raup-Crick index probability compared sampling units non-identical species composition. probability can regarded dissimilarity, although metric: identical sampling units can dissimilarity slightly \\(0\\), dissimilarity can nearly zero range shared species, sampling units shared species can dissimilarity slightly \\(1\\). Moreover, communities sharing rare species appear similar (lower probability finding rare species together), communities sharing number common species. function always treat data binary (presence/ absence). probability assessed using simulation oecosimu test statistic observed number shared species sampling units evaluated community null model (see Examples). default null model \"r1\" probability selecting species proportional species frequencies. vegdist function implements variant Raup-Crick index equal sampling probabilities species using exact analytic equations without simulation. corresponds null model \"r0\" also can used current function. null model methods oecosimu can used current function, new unpublished methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function returns object inheriting dist can interpreted dissimilarity matrix.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"function developed Brian Inouye contacted us informed us method Chase et al. (2011), function takes idea code published paper. current function written Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"test statistic number shared species, typically tied large number simulation results. tied values handled differently current function function published Chase et al. (2011). vegan, index number simulated values smaller equal observed value, smaller observed value used Chase et al. (2011) option split = FALSE script; can achieved chase = TRUE vegan. Chase et al. (2011) script split = TRUE uses half tied simulation values calculate distance measure, choice directly reproduced vegan ( average vegan raupcrick results chase = TRUE chase = FALSE).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/raupcrick.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Raup-Crick Dissimilarity with Unequal Sampling Densities of Species — raupcrick","text":"","code":"## data set with variable species richness data(sipoo) ## default raupcrick dr1 <- raupcrick(sipoo) ## use null model \"r0\" of oecosimu dr0 <- raupcrick(sipoo, null = \"r0\") ## vegdist(..., method = \"raup\") corresponds to 'null = \"r0\"' d <- vegdist(sipoo, \"raup\") op <- par(mfrow=c(2,1), mar=c(4,4,1,1)+.1) plot(dr1 ~ d, xlab = \"Raup-Crick with Null R1\", ylab=\"vegdist\") plot(dr0 ~ d, xlab = \"Raup-Crick with Null R0\", ylab=\"vegdist\") par(op) ## The calculation is essentially as in the following oecosimu() call, ## except that designdist() is replaced with faster code if (FALSE) oecosimu(sipoo, function(x) designdist(x, \"J\", \"binary\"), method = \"r1\")"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":null,"dir":"Reference","previous_headings":"","what":"Reads a CEP (Canoco) data file — read.cep","title":"Reads a CEP (Canoco) data file — read.cep","text":"read.cep reads file formatted relaxed strict CEP format used Canoco software, among others.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"read.cep(file, positive=TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reads a CEP (Canoco) data file — read.cep","text":"file File name (character variable). positive positive entries, like community data.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reads a CEP (Canoco) data file — read.cep","text":"Cornell Ecology Programs (CEP) introduced several data formats designed punched cards. One ‘condensed strict’ format adopted popular software DECORANA TWINSPAN. relaxed variant format later adopted Canoco software (ter Braak 1984). Function read.cep reads legacy files written format. condensed CEP CANOCO formats : Two three title cards, importantly specifying format number items per record. Data condensed format: First number line site identifier (integer), followed pairs (‘couplets’) numbers identifying species abundance (integer floating point number). Species site names, given Fortran format (10A8): Ten names per line, eight columns . option positive = TRUE function removes rows columns zero negative marginal sums. community data positive entries, removes empty sites species. data entries can negative, ruins data, data sets read option positive = FALSE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reads a CEP (Canoco) data file — read.cep","text":"Returns data frame, columns species rows sites. Column row names taken CEP file, changed unique R names make.names stripping blanks.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Reads a CEP (Canoco) data file — read.cep","text":"ter Braak, C.J.F. (1984--): CANOCO -- FORTRAN program canonical community ordination [partial] [detrended] [canonical] correspondence analysis, principal components analysis redundancy analysis. TNO Inst. Applied Computer Sci., Stat. Dept. Wageningen, Netherlands.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reads a CEP (Canoco) data file — read.cep","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reads a CEP (Canoco) data file — read.cep","text":"Function read.cep used Fortran read data vegan 2.4-5 earlier, Fortran /O longer allowed CRAN packages, function re-written R. original Fortran code robust, several legacy data sets may fail current version, read previous Fortran version. CRAN package cepreader makes available original Fortran-based code run separate subprocess. cepreader package can also read ‘free’ ‘open’ Canoco formats handled function. function based read.fortran. REAL format defines decimal part species abundances (F5.1), read.fortran divides input corresponding power 10 even input data explicit decimal separator. F5.1, 100 become 10, 0.1 become 0.01. Function read.cep tries undo division, check scaling results reading data, necessary, multiply results original scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/read.cep.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reads a CEP (Canoco) data file — read.cep","text":"","code":"## Provided that you have the file \"dune.spe\" if (FALSE) { theclassic <- read.cep(\"dune.spe\")}"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":null,"dir":"Reference","previous_headings":"","what":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi find Rényi diversities scale corresponding Hill number (Hill 1973). Function renyiaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"renyi(x, scales = c(0, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64, Inf), hill = FALSE) # S3 method for renyi plot(x, ...) renyiaccum(x, scales = c(0, 0.5, 1, 2, 4, Inf), permutations = 100, raw = FALSE, collector = FALSE, subset, ...) # S3 method for renyiaccum plot(x, what = c(\"Collector\", \"mean\", \"Qnt 0.025\", \"Qnt 0.975\"), type = \"l\", ...) # S3 method for renyiaccum persp(x, theta = 220, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"x Community data matrix plotting object. scales Scales Rényi diversity. hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. collector Accumulate diversities order sites data set, collector curve can plotted summary permutations. argument ignored raw = TRUE. subset logical expression indicating sites (rows) keep: missing values taken FALSE. Items plotted. type Type plot, type = \"l\" means lines. theta Angle defining viewing direction (azimuthal) persp. col Colours used surface. Single colour passed , vector colours selected midpoint rectangle persp. zlim Limits vertical axis. ... arguments passed renyi graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Common diversity indices special cases Rényi diversity $$H_a = \\frac{1}{1-} \\log \\sum p_i^$$ \\(\\) scale parameter, Hill (1975) suggested use -called ‘Hill numbers’ defined \\(N_a = \\exp(H_a)\\). Hill numbers number species \\(= 0\\), \\(\\exp(H')\\) exponent Shannon diversity \\(= 1\\), inverse Simpson \\(= 2\\) \\(1/ \\max(p_i)\\) \\(= \\infty\\). According theory diversity ordering, one community can regarded diverse another Rényi diversities higher (Tóthmérész 1995). plot method renyi uses lattice graphics, displays diversity values scale separate panel site together minimum, maximum median values complete data. Function renyiaccum similar specaccum finds Rényi Hill diversities given scales random permutations accumulated sites. plot function uses lattice function xyplot display accumulation curves value scales separate panel. addition, persp method plot diversity surface scale number sites. Similar dynamic graphics can made rgl.renyiaccum vegan3d package.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Function renyi returns data frame selected indices. Function renyiaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Hill, M.O. (1973). Diversity evenness: unifying notation consequences. Ecology 54, 427--473. Kindt, R., Van Damme, P., Simons, .J. (2006). Tree diversity western Kenya: using profiles characterise richness evenness. Biodiversity Conservation 15, 1253--1270. Tóthmérész, B. (1995). Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"Roeland Kindt Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/renyi.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Renyi and Hill Diversities and Corresponding Accumulation Curves — renyi","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) mod <- renyi(BCI[i,]) plot(mod) mod <- renyiaccum(BCI[i,]) plot(mod, as.table=TRUE, col = c(1, 2, 2)) persp(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":null,"dir":"Reference","previous_headings":"","what":"Reorder a Hierarchical Clustering Tree — reorder.hclust","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Function takes hierarchical clustering tree hclust vector values reorders clustering tree order supplied vector, maintaining constraints tree. method generic function reorder alternative reordering \"dendrogram\" object reorder.dendrogram","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"# S3 method for hclust reorder(x, wts, agglo.FUN = c(\"mean\", \"min\", \"max\", \"sum\", \"uwmean\"), ...) # S3 method for hclust rev(x) # S3 method for hclust scores(x, display = \"internal\", ...) cutreeord(tree, k = NULL, h = NULL)"},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"x, tree hierarchical clustering hclust. wts numeric vector reordering. agglo.FUN function weights agglomeration, see . display return \"internal\" nodes \"terminal\" nodes (also called \"leaves\"). k, h scalars vectors giving numbers desired groups heights tree cut (passed function cutree). ... additional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Dendrograms can ordered many ways. reorder function reorders hclust tree provides alternative reorder.dendrogram can reorder dendrogram. current function also work differently agglo.FUN \"mean\": reorder.dendrogram always take direct mean member groups ignoring sizes, function used weighted.mean weighted group sizes, group mean always mean member leaves (terminal nodes). want ignore group sizes, can use unweighted mean \"uwmean\". function accepts limited list agglo.FUN functions assessing value wts groups. ordering always ascending, order leaves can reversed rev. Function scores finds coordinates nodes two-column matrix. terminal nodes (leaves) value item merged tree, labels can still hang level (see plot.hclust). Function cutreeord cuts tree groups numbered left right tree. based standard function cutree numbers groups order appear input data instead order tree.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Reordered hclust result object added item value gives value statistic merge level.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"functions really base R.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/reorder.hclust.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Reorder a Hierarchical Clustering Tree — reorder.hclust","text":"","code":"## reorder by water content of soil data(mite, mite.env) hc <- hclust(vegdist(wisconsin(sqrt(mite)))) ohc <- with(mite.env, reorder(hc, WatrCont)) plot(hc) plot(ohc) ## label leaves by the observed value, and each branching point ## (internal node) by the cluster mean with(mite.env, plot(ohc, labels=round(WatrCont), cex=0.7)) ordilabel(scores(ohc), label=round(ohc$value), cex=0.7) ## Slightly different from reordered 'dendrogram' which ignores group ## sizes in assessing means. den <- as.dendrogram(hc) den <- with(mite.env, reorder(den, WatrCont, agglo.FUN = mean)) plot(den)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Species or Site Scores from an Ordination — scores","title":"Get Species or Site Scores from an Ordination — scores","text":"Function access either species site scores specified axes ordination methods. scores function generic vegan, vegan ordination functions scores functions documented separately method (see e.g. scores.cca, scores.metaMDS, scores.decorana). help file documents default scores method used non-vegan ordination objects.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"# S3 method for default scores(x, choices, display=c(\"sites\", \"species\", \"both\"), tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Species or Site Scores from an Ordination — scores","text":"x ordination result. choices Ordination axes. missing, default method returns axes. display Partial match access scores \"sites\" \"species\" \"\". tidy Return \"\" scores data frame compatible ggplot2, variable score labelling scores \"sites\" \"species\". ... parameters (unused).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Species or Site Scores from an Ordination — scores","text":"Function scores generic method vegan. Several vegan functions scores methods defaults new arguments. help page describes default method. methods, see, e.g., scores.cca, scores.rda, scores.decorana. vegan ordination functions scores method used extract scores instead directly accessing . Scaling transformation scores also happen scores function. scores function available, results can plotted using ordiplot, ordixyplot etc., ordination results can compared procrustes analysis. scores.default function used extract scores non-vegan ordination results. Many standard ordination methods libraries specific class, specific method can written . However, scores.default guesses commonly used functions keep site scores possible species scores. x matrix, scores.default returns chosen columns matrix, ignoring whether species sites requested (regard bug feature, please). Currently function seems work least isoMDS, prcomp, princomp ade4 objects. may work cases fail mysteriously.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Species or Site Scores from an Ordination — scores","text":"function returns matrix scores one type requested, named list matrices display = \"\", ggplot2 compatible data frame tidy = TRUE.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Get Species or Site Scores from an Ordination — scores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/scores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Species or Site Scores from an Ordination — scores","text":"","code":"data(varespec) vare.pca <- prcomp(varespec) scores(vare.pca, choices=c(1,2)) #> PC1 PC2 #> 18 -10.7847878 18.7094315 #> 15 -27.8036826 -11.7414745 #> 24 -25.6919559 -14.5399684 #> 27 -31.7820166 -31.2216800 #> 23 -19.6315869 -2.5541193 #> 19 -0.2413294 -11.4974077 #> 22 -26.6771373 -12.3140897 #> 16 -21.9230366 0.4449159 #> 28 -39.6083051 -41.8877392 #> 13 -4.0664328 20.4191153 #> 14 -18.4416245 5.4406988 #> 20 -17.3999191 2.3653380 #> 25 -25.1673547 -13.2508067 #> 7 -11.4065430 41.7356300 #> 5 -8.4243752 45.3805255 #> 6 -2.0759474 36.9311222 #> 3 39.8617580 8.0590041 #> 4 13.1065901 12.8377217 #> 2 57.6827011 -4.8983565 #> 9 63.3138332 -22.4481549 #> 12 44.1073111 -10.1653935 #> 10 64.9418975 -16.7633564 #> 11 11.5313633 3.9720890 #> 21 -3.4194194 -3.0130455"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":null,"dir":"Reference","previous_headings":"","what":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Screeplot methods plotting variances ordination axes/components overlaying broken stick distributions. Also, provides alternative screeplot methods princomp prcomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"# S3 method for cca screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, if (is.null(x$CCA) || x$CCA$rank == 0) x$CA$rank else x$CCA$rank), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for decorana screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = 4, ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for prcomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) # S3 method for princomp screeplot(x, bstick = FALSE, type = c(\"barplot\", \"lines\"), npcs = min(10, length(x$sdev)), ptype = \"o\", bst.col = \"red\", bst.lty = \"solid\", xlab = \"Component\", ylab = \"Inertia\", main = deparse(substitute(x)), legend = bstick, ...) bstick(n, ...) # S3 method for default bstick(n, tot.var = 1, ...) # S3 method for cca bstick(n, ...) # S3 method for prcomp bstick(n, ...) # S3 method for princomp bstick(n, ...) # S3 method for decorana bstick(n, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"x object component variances can determined. bstick logical; broken stick distribution drawn? npcs number components plotted. type type plot. ptype type == \"lines\" bstick = TRUE, character indicating type plotting used lines; actually types plot.default. bst.col, bst.lty colour line type used draw broken stick distribution. xlab, ylab, main graphics parameters. legend logical; draw legend? n object variances can extracted number variances (components) case bstick.default. tot.var total variance split. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"functions provide screeplots ordination methods vegan enhanced versions broken stick prcomp princomp. Function bstick gives brokenstick values ordered random proportions, defined \\(p_i = (tot/n) \\sum_{x=}^n (1/x)\\) (Legendre & Legendre 2012), \\(tot\\) total \\(n\\) number brokenstick components (cf. radfit). Broken stick recommended stopping rule principal component analysis (Jackson 1993): principal components retained long observed eigenvalues higher corresponding random broken stick components. bstick function generic. default needs number components total, specific methods extract information ordination results. also bstick method cca. However, broken stick model strictly valid correspondence analysis (CA), eigenvalues CA defined \\(\\leq 1\\), whereas brokenstick components restrictions. brokenstick components detrended correspondence analysis (DCA) assume input data full rank, additive eigenvalues used screeplot (see decorana).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Function screeplot draws plot currently active device, returns invisibly xy.coords points bars eigenvalues. Function bstick returns numeric vector broken stick components.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Jackson, D. . (1993). Stopping rules principal components analysis: comparison heuristical statistical approaches. Ecology 74, 2204--2214. Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"Gavin L. Simpson","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/screeplot.cca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Screeplots for Ordination Results and Broken Stick Distributions — screeplot.cca","text":"","code":"data(varespec) vare.pca <- rda(varespec, scale = TRUE) bstick(vare.pca) #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 7.1438620 5.2308185 4.2742968 3.6366156 3.1583548 2.7757461 2.4569055 2.1836136 #> PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 #> 1.9444831 1.7319228 1.5406184 1.3667054 1.2072851 1.0601279 0.9234819 0.7959457 #> PC17 PC18 PC19 PC20 PC21 PC22 PC23 #> 0.6763805 0.5638485 0.4575683 0.3568818 0.2612296 0.1701323 0.0831758 screeplot(vare.pca, bstick = TRUE, type = \"lines\")"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":null,"dir":"Reference","previous_headings":"","what":"Similarity Percentages — simper","title":"Similarity Percentages — simper","text":"Discriminating species two groups using Bray-Curtis dissimilarities","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Similarity Percentages — simper","text":"","code":"simper(comm, group, permutations = 999, parallel = 1, ...) # S3 method for simper summary(object, ordered = TRUE, digits = max(3,getOption(\"digits\") - 3), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Similarity Percentages — simper","text":"comm Community data. group Factor describing group structure. missing one level, contributions estimated non-grouped data dissimilarities show overall heterogeneity species abundances. permutations list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. object object returned simper. ordered Logical; species ordered average contribution? digits Number digits output. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. (yet implemented). ... Parameters passed functions. simper extra parameters passed shuffleSet permutations used.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Similarity Percentages — simper","text":"Similarity percentage, simper (Clarke 1993) based decomposition Bray-Curtis dissimilarity index (see vegdist, designdist). contribution individual species \\(\\) overall Bray-Curtis dissimilarity \\(d_{jk}\\) given $$d_{ijk} = \\frac{|x_{ij}-x_{ik}|}{\\sum_{=1}^S (x_{ij}+x_{ik})}$$ \\(x\\) abundance species \\(\\) sampling units \\(j\\) \\(k\\). overall index sum individual contributions \\(S\\) species \\(d_{jk}=\\sum_{=1}^S d_{ijk}\\). simper functions performs pairwise comparisons groups sampling units finds contribution species average -group Bray-Curtis dissimilarity. Although method called “Similarity Percentages”, really studied dissimilarities instead similarities (Clarke 1993). function displays important species pair groups. species contribute least 70 % differences groups. function returns much extensive results (including species) can accessed directly result object (see section Value). Function summary transforms result list data frames. argument ordered = TRUE data frames also include cumulative contributions ordered species contribution. results simper can difficult interpret often misunderstood even publications. method gives contribution species overall dissimilarities, caused variation species abundances, partly differences among groups. Even make groups copies , method single species high contribution, contributions non-existing -group differences random noise variation species abundances. abundant species usually highest variances, high contributions even differ among groups. Permutation tests study differences among groups, can used find species differences among groups important component contribution dissimilarities. Analysis without group argument find species contributions average overall dissimilarity among sampling units. non-grouped contributions can compared grouped contributions see much added value grouping species.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Similarity Percentages — simper","text":"list class \"simper\" following items: species species names. average Species contribution average -group dissimilarity. overall average -group dissimilarity. sum item average. sd Standard deviation contribution. ratio Average sd ratio. ava, avb Average abundances per group. ord index vector order vectors contribution order cusum back original data order. cusum Ordered cumulative contribution. based item average, sum total 1. p Permutation \\(p\\)-value. Probability getting larger equal average contribution random permutation group factor. area available permutations used (default: calculated).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Similarity Percentages — simper","text":"Eduard Szöcs Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Similarity Percentages — simper","text":"Clarke, K.R. 1993. Non-parametric multivariate analyses changes community structure. Australian Journal Ecology, 18, 117–143.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simper.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Similarity Percentages — simper","text":"","code":"data(dune) data(dune.env) (sim <- with(dune.env, simper(dune, Management, permutations = 99))) #> cumulative contributions of most influential species: #> #> $SF_BF #> Agrostol Alopgeni Lolipere Trifrepe Poatriv Scorautu Bromhord #> 0.09824271 0.18254830 0.25956958 0.33367870 0.40734444 0.47729205 0.53120026 #> Achimill Planlanc Elymrepe Bracruta #> 0.57946526 0.62522255 0.67016196 0.71098133 #> #> $SF_HF #> Agrostol Alopgeni Lolipere Planlanc Rumeacet Elymrepe Poatriv #> 0.08350879 0.16534834 0.23934930 0.30843624 0.37716139 0.43334492 0.48351753 #> Bracruta Eleopalu Poaprat Anthodor Sagiproc Trifprat #> 0.52804045 0.57205850 0.61423981 0.65549838 0.69628951 0.73696831 #> #> $SF_NM #> Poatriv Alopgeni Agrostol Lolipere Eleopalu Poaprat Bracruta Elymrepe #> 0.1013601 0.1935731 0.2667383 0.3377578 0.3999419 0.4526707 0.5044725 0.5505643 #> Scorautu Trifrepe Sagiproc Salirepe #> 0.5926117 0.6320111 0.6712478 0.7091528 #> #> $BF_HF #> Rumeacet Poatriv Planlanc Bromhord Lolipere Elymrepe Trifrepe #> 0.08163219 0.15193797 0.21918333 0.27967181 0.33969561 0.39843338 0.45298204 #> Anthodor Achimill Bracruta Alopgeni Trifprat Juncarti #> 0.50276849 0.55222648 0.60021994 0.64584333 0.69126471 0.73366621 #> #> $BF_NM #> Lolipere Poatriv Poaprat Trifrepe Bromhord Bracruta Eleopalu Agrostol #> 0.1242718 0.1992126 0.2711756 0.3414609 0.3958520 0.4448077 0.4910724 0.5369083 #> Achimill Scorautu Anthodor Planlanc #> 0.5823926 0.6253645 0.6638182 0.7012577 #> #> $HF_NM #> Poatriv Lolipere Rumeacet Poaprat Planlanc Bracruta Eleopalu #> 0.09913221 0.17468460 0.23917190 0.29701331 0.35469313 0.40365488 0.44804851 #> Agrostol Trifrepe Elymrepe Anthodor Juncarti Trifprat Salirepe #> 0.49226546 0.53434466 0.57564661 0.61543243 0.65341300 0.68921695 0.72432408 #> ## IGNORE_RDIFF_BEGIN summary(sim) #> #> Contrast: SF_BF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.06137 0.03419 1.79490 4.66700 0.00000 0.098 0.05 * #> Alopgeni 0.05267 0.03648 1.44390 4.33300 0.66700 0.182 0.14 #> Lolipere 0.04812 0.03945 1.21980 3.00000 6.00000 0.260 0.39 #> Trifrepe 0.04630 0.02553 1.81380 1.33300 4.66700 0.334 0.09 . #> Poatriv 0.04602 0.03380 1.36150 4.66700 3.66700 0.407 0.46 #> Scorautu 0.04370 0.02492 1.75340 1.33300 4.33300 0.477 0.04 * #> Bromhord 0.03368 0.02586 1.30230 0.50000 2.66700 0.531 0.02 * #> Achimill 0.03015 0.02082 1.44820 0.16700 2.33300 0.580 0.04 * #> Planlanc 0.02859 0.02155 1.32650 0.00000 2.00000 0.625 0.49 #> Elymrepe 0.02807 0.02978 0.94280 2.00000 1.33300 0.670 0.52 #> Bracruta 0.02550 0.02390 1.06690 2.00000 2.00000 0.711 0.83 #> Poaprat 0.02513 0.02397 1.04850 2.50000 4.00000 0.751 0.82 #> Sagiproc 0.02433 0.02215 1.09830 1.83300 0.66700 0.790 0.39 #> Bellpere 0.01986 0.01709 1.16220 0.66700 1.66700 0.822 0.10 . #> Eleopalu 0.01861 0.04296 0.43330 1.33300 0.00000 0.852 0.82 #> Anthodor 0.01754 0.02580 0.67980 0.00000 1.33300 0.880 0.75 #> Juncbufo 0.01603 0.02371 0.67620 1.16700 0.00000 0.905 0.57 #> Vicilath 0.01467 0.01331 1.10260 0.00000 1.00000 0.929 0.04 * #> Hyporadi 0.01029 0.01520 0.67680 0.00000 0.66700 0.945 0.62 #> Ranuflam 0.00931 0.01360 0.68450 0.66700 0.00000 0.960 0.93 #> Juncarti 0.00698 0.01611 0.43330 0.50000 0.00000 0.972 0.95 #> Callcusp 0.00698 0.01611 0.43330 0.50000 0.00000 0.983 0.79 #> Rumeacet 0.00453 0.01044 0.43330 0.33300 0.00000 0.990 0.95 #> Cirsarve 0.00398 0.00918 0.43360 0.33300 0.00000 0.996 0.37 #> Chenalbu 0.00233 0.00537 0.43330 0.16700 0.00000 1.000 0.41 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_HF #> #> average sd ratio ava avb cumsum p #> Agrostol 0.04738 0.03127 1.51510 4.66700 1.40000 0.084 0.35 #> Alopgeni 0.04643 0.03290 1.41150 4.33300 1.60000 0.165 0.20 #> Lolipere 0.04199 0.02701 1.55460 3.00000 4.00000 0.239 0.75 #> Planlanc 0.03920 0.03321 1.18040 0.00000 3.00000 0.308 0.02 * #> Rumeacet 0.03899 0.02737 1.42470 0.33300 3.20000 0.377 0.01 ** #> Elymrepe 0.03188 0.02955 1.07870 2.00000 2.00000 0.433 0.30 #> Poatriv 0.02847 0.02152 1.32270 4.66700 4.80000 0.484 1.00 #> Bracruta 0.02526 0.02104 1.20040 2.00000 2.80000 0.528 0.92 #> Eleopalu 0.02497 0.03888 0.64240 1.33300 0.80000 0.572 0.77 #> Poaprat 0.02393 0.01918 1.24780 2.50000 3.40000 0.614 0.98 #> Anthodor 0.02341 0.02143 1.09230 0.00000 1.80000 0.655 0.67 #> Sagiproc 0.02314 0.02048 1.13010 1.83300 0.80000 0.696 0.40 #> Trifprat 0.02308 0.02343 0.98500 0.00000 1.80000 0.737 0.01 ** #> Juncarti 0.02285 0.02568 0.88990 0.50000 1.60000 0.777 0.51 #> Trifrepe 0.02238 0.01949 1.14860 1.33300 2.80000 0.817 0.94 #> Juncbufo 0.02164 0.02224 0.97330 1.16700 1.20000 0.855 0.24 #> Scorautu 0.02051 0.01642 1.24890 1.33300 2.80000 0.891 0.79 #> Achimill 0.01518 0.01139 1.33260 0.16700 1.20000 0.918 0.75 #> Bromhord 0.01338 0.01450 0.92220 0.50000 0.80000 0.941 0.79 #> Ranuflam 0.01066 0.01339 0.79640 0.66700 0.40000 0.960 0.86 #> Bellpere 0.00999 0.01257 0.79480 0.66700 0.40000 0.978 0.84 #> Callcusp 0.00662 0.01508 0.43930 0.50000 0.00000 0.989 0.92 #> Cirsarve 0.00381 0.00867 0.43940 0.33300 0.00000 0.996 0.54 #> Chenalbu 0.00221 0.00503 0.43930 0.16700 0.00000 1.000 0.51 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Hyporadi 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Vicilath 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: SF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07828 0.04095 1.91180 4.66700 0.00000 0.101 0.01 ** #> Alopgeni 0.07122 0.04696 1.51670 4.33300 0.00000 0.194 0.01 ** #> Agrostol 0.05651 0.04418 1.27920 4.66700 2.16700 0.267 0.05 * #> Lolipere 0.05485 0.05991 0.91550 3.00000 0.33300 0.338 0.15 #> Eleopalu 0.04803 0.04717 1.01820 1.33300 2.16700 0.400 0.04 * #> Poaprat 0.04072 0.03179 1.28100 2.50000 0.66700 0.453 0.06 . #> Bracruta 0.04001 0.03440 1.16310 2.00000 2.83300 0.504 0.08 . #> Elymrepe 0.03560 0.03852 0.92430 2.00000 0.00000 0.551 0.12 #> Scorautu 0.03247 0.03481 0.93280 1.33300 3.16700 0.593 0.13 #> Trifrepe 0.03043 0.03163 0.96190 1.33300 1.83300 0.632 0.59 #> Sagiproc 0.03030 0.03048 0.99430 1.83300 0.50000 0.671 0.02 * #> Salirepe 0.02928 0.03201 0.91440 0.00000 1.83300 0.709 0.02 * #> Anthodor 0.02454 0.03669 0.66880 0.00000 1.33300 0.741 0.56 #> Callcusp 0.02276 0.02944 0.77310 0.50000 1.16700 0.770 0.08 . #> Ranuflam 0.02257 0.02282 0.98890 0.66700 1.33300 0.800 0.08 . #> Juncarti 0.02254 0.02860 0.78830 0.50000 1.16700 0.829 0.53 #> Hyporadi 0.02011 0.03129 0.64260 0.00000 1.16700 0.855 0.21 #> Juncbufo 0.01986 0.02903 0.68400 1.16700 0.00000 0.881 0.21 #> Planlanc 0.01542 0.02277 0.67720 0.00000 0.83300 0.900 0.98 #> Airaprae 0.01488 0.02188 0.68020 0.00000 0.83300 0.920 0.06 . #> Bellpere 0.01232 0.01592 0.77370 0.66700 0.33300 0.936 0.72 #> Comapalu 0.01188 0.01741 0.68260 0.00000 0.66700 0.951 0.05 * #> Achimill 0.00929 0.01493 0.62240 0.16700 0.33300 0.963 0.98 #> Bromhord 0.00717 0.01633 0.43910 0.50000 0.00000 0.972 0.98 #> Rumeacet 0.00559 0.01275 0.43840 0.33300 0.00000 0.980 0.98 #> Empenigr 0.00523 0.01200 0.43540 0.00000 0.33300 0.986 0.29 #> Cirsarve 0.00478 0.01089 0.43910 0.33300 0.00000 0.993 0.02 * #> Chenalbu 0.00289 0.00660 0.43820 0.16700 0.00000 0.996 0.02 * #> Vicilath 0.00279 0.00642 0.43450 0.00000 0.16700 1.000 0.81 #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_HF #> #> average sd ratio ava avb cumsum p #> Rumeacet 0.03867 0.02606 1.48380 0.00000 3.20000 0.082 0.03 * #> Poatriv 0.03330 0.02579 1.29110 3.66700 4.80000 0.152 0.96 #> Planlanc 0.03185 0.01830 1.74010 2.00000 3.00000 0.219 0.34 #> Bromhord 0.02865 0.01799 1.59260 2.66700 0.80000 0.280 0.07 . #> Lolipere 0.02843 0.02215 1.28340 6.00000 4.00000 0.340 1.00 #> Elymrepe 0.02782 0.02959 0.94040 1.33300 2.00000 0.398 0.56 #> Trifrepe 0.02584 0.01656 1.56030 4.66700 2.80000 0.453 0.78 #> Anthodor 0.02358 0.02042 1.15470 1.33300 1.80000 0.503 0.55 #> Achimill 0.02343 0.01474 1.58930 2.33300 1.20000 0.552 0.24 #> Bracruta 0.02273 0.01802 1.26170 2.00000 2.80000 0.600 0.89 #> Alopgeni 0.02161 0.02308 0.93630 0.66700 1.60000 0.646 0.91 #> Trifprat 0.02151 0.02207 0.97470 0.00000 1.80000 0.691 0.12 #> Juncarti 0.02008 0.02555 0.78600 0.00000 1.60000 0.734 0.59 #> Scorautu 0.01932 0.01357 1.42410 4.33300 2.80000 0.774 0.78 #> Bellpere 0.01829 0.01486 1.23050 1.66700 0.40000 0.813 0.22 #> Agrostol 0.01761 0.02284 0.77080 0.00000 1.40000 0.850 1.00 #> Juncbufo 0.01500 0.02066 0.72600 0.00000 1.20000 0.882 0.67 #> Vicilath 0.01285 0.01140 1.12740 1.00000 0.00000 0.909 0.04 * #> Sagiproc 0.01168 0.01297 0.90080 0.66700 0.80000 0.934 0.92 #> Eleopalu 0.01017 0.02111 0.48170 0.00000 0.80000 0.955 0.93 #> Hyporadi 0.00895 0.01312 0.68240 0.66700 0.00000 0.974 0.64 #> Poaprat 0.00720 0.01010 0.71330 4.00000 3.40000 0.989 1.00 #> Ranuflam 0.00508 0.01055 0.48170 0.00000 0.40000 1.000 0.97 #> Airaprae 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Comapalu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Empenigr 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Salirepe 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Callcusp 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: BF_NM #> #> average sd ratio ava avb cumsum p #> Lolipere 0.09068 0.02644 3.42900 6.00000 0.33300 0.124 0.04 * #> Poatriv 0.05468 0.04465 1.22500 3.66700 0.00000 0.199 0.28 #> Poaprat 0.05251 0.01813 2.89700 4.00000 0.66700 0.271 0.02 * #> Trifrepe 0.05129 0.02756 1.86100 4.66700 1.83300 0.342 0.04 * #> Bromhord 0.03969 0.02920 1.35900 2.66700 0.00000 0.396 0.01 ** #> Bracruta 0.03572 0.02869 1.24500 2.00000 2.83300 0.445 0.40 #> Eleopalu 0.03376 0.03573 0.94500 0.00000 2.16700 0.491 0.43 #> Agrostol 0.03345 0.03473 0.96300 0.00000 2.16700 0.537 0.87 #> Achimill 0.03319 0.02338 1.42000 2.33300 0.33300 0.582 0.02 * #> Scorautu 0.03136 0.02026 1.54800 4.33300 3.16700 0.625 0.29 #> Anthodor 0.02806 0.03295 0.85200 1.33300 1.33300 0.664 0.37 #> Planlanc 0.02732 0.02193 1.24600 2.00000 0.83300 0.701 0.62 #> Salirepe 0.02677 0.02927 0.91400 0.00000 1.83300 0.738 0.11 #> Bellpere 0.02353 0.01909 1.23200 1.66700 0.33300 0.770 0.04 * #> Hyporadi 0.02172 0.02450 0.88600 0.66700 1.16700 0.800 0.26 #> Ranuflam 0.02031 0.02275 0.89300 0.00000 1.33300 0.828 0.27 #> Elymrepe 0.01999 0.02926 0.68300 1.33300 0.00000 0.855 0.79 #> Callcusp 0.01783 0.02681 0.66500 0.00000 1.16700 0.880 0.44 #> Juncarti 0.01769 0.02600 0.68100 0.00000 1.16700 0.904 0.73 #> Vicilath 0.01577 0.01447 1.09000 1.00000 0.16700 0.925 0.01 ** #> Sagiproc 0.01543 0.01857 0.83100 0.66700 0.50000 0.947 0.79 #> Airaprae 0.01341 0.01969 0.68100 0.00000 0.83300 0.965 0.30 #> Comapalu 0.01074 0.01571 0.68400 0.00000 0.66700 0.980 0.42 #> Alopgeni 0.01000 0.01463 0.68300 0.66700 0.00000 0.993 0.99 #> Empenigr 0.00479 0.01105 0.43300 0.00000 0.33300 1.000 0.41 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Juncbufo 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Rumeacet 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Trifprat 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Contrast: HF_NM #> #> average sd ratio ava avb cumsum p #> Poatriv 0.07155 0.01368 5.23000 4.80000 0.00000 0.099 0.01 ** #> Lolipere 0.05453 0.02962 1.84100 4.00000 0.33300 0.175 0.22 #> Rumeacet 0.04655 0.03081 1.51100 3.20000 0.00000 0.239 0.01 ** #> Poaprat 0.04175 0.01885 2.21500 3.40000 0.66700 0.297 0.06 . #> Planlanc 0.04163 0.02956 1.40800 3.00000 0.83300 0.355 0.03 * #> Bracruta 0.03534 0.02010 1.75800 2.80000 2.83300 0.404 0.40 #> Eleopalu 0.03204 0.03231 0.99200 0.80000 2.16700 0.448 0.51 #> Agrostol 0.03192 0.02889 1.10500 1.40000 2.16700 0.492 0.99 #> Trifrepe 0.03037 0.02287 1.32800 2.80000 1.83300 0.534 0.71 #> Elymrepe 0.02981 0.03868 0.77100 2.00000 0.00000 0.576 0.53 #> Anthodor 0.02872 0.02480 1.15800 1.80000 1.33300 0.615 0.24 #> Juncarti 0.02741 0.02854 0.96100 1.60000 1.16700 0.653 0.19 #> Trifprat 0.02584 0.02597 0.99500 1.80000 0.00000 0.689 0.01 ** #> Salirepe 0.02534 0.02729 0.92900 0.00000 1.83300 0.724 0.18 #> Alopgeni 0.02446 0.03240 0.75500 1.60000 0.00000 0.758 0.92 #> Scorautu 0.02070 0.01412 1.46600 2.80000 3.16700 0.787 0.85 #> Ranuflam 0.01928 0.01994 0.96700 0.40000 1.33300 0.814 0.36 #> Juncbufo 0.01818 0.02465 0.73800 1.20000 0.00000 0.839 0.43 #> Hyporadi 0.01714 0.02655 0.64600 0.00000 1.16700 0.863 0.42 #> Callcusp 0.01683 0.02490 0.67600 0.00000 1.16700 0.886 0.42 #> Achimill 0.01656 0.01490 1.11100 1.20000 0.33300 0.909 0.59 #> Sagiproc 0.01528 0.01653 0.92400 0.80000 0.50000 0.930 0.89 #> Airaprae 0.01261 0.01824 0.69100 0.00000 0.83300 0.947 0.30 #> Bromhord 0.01209 0.01517 0.79700 0.80000 0.00000 0.964 0.84 #> Comapalu 0.01011 0.01456 0.69400 0.00000 0.66700 0.978 0.25 #> Bellpere 0.00880 0.01373 0.64100 0.40000 0.33300 0.990 0.94 #> Empenigr 0.00454 0.01033 0.43900 0.00000 0.33300 0.997 0.60 #> Vicilath 0.00240 0.00546 0.43900 0.00000 0.16700 1.000 0.86 #> Chenalbu 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> Cirsarve 0.00000 0.00000 NaN 0.00000 0.00000 1.000 NA #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 99 ## IGNORE_RDIFF_END"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":null,"dir":"Reference","previous_headings":"","what":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Function simulates response data frame adds Gaussian error fitted responses Redundancy Analysis (rda), Constrained Correspondence Analysis (cca) distance-based RDA (capscale). function special case generic simulate, works similarly simulate.lm.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"# S3 method for rda simulate(object, nsim = 1, seed = NULL, indx = NULL, rank = \"full\", correlated = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"object object representing fitted rda, cca capscale model. nsim number response matrices simulated. one dissimilarity matrix returned capscale, larger nsim error. seed object specifying random number generator initialized (‘seeded’). See simulate details. indx Index residuals added fitted values, produced shuffleSet sample. index can duplicate entries bootstrapping allowed. nsim \\(>1\\), output compliant shuffleSet one line simulation. nsim missing, number rows indx used define number simulations, nsim given, match number rows indx. null, parametric simulation used Gaussian error added fitted values. rank rank constrained component: passed predict.rda predict.cca. correlated species regarded correlated parametric simulation indx given? correlated = TRUE, multivariate Gaussian random error generated, FALSE, Gaussian random error generated separately species. argument effect capscale information species. ... additional optional arguments (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"implementation follows \"lm\" method simulate, adds Gaussian (Normal) error fitted values (fitted.rda) using function rnorm correlated = FALSE mvrnorm correlated = TRUE. standard deviations (rnorm) covariance matrices species (mvrnorm) estimated residuals fitting constraints. Alternatively, function can take permutation index used add permuted residuals (unconstrained component) fitted values. Raw data used rda. Internal Chi-square transformed data used cca within function, returned matrix similar original input data. simulation performed internal metric scaling data capscale, function returns Euclidean distances calculated simulated data. simulation uses real components, imaginary dimensions ignored.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"nsim = 1, returns matrix dissimilarities ( capscale) similar additional arguments random number seed simulate. nsim > 1, returns similar array returned simulate.nullmodel similar attributes.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/simulate.rda.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Simulate Responses with Gaussian Error or Permuted Residuals for Constrained Ordination — simulate.rda","text":"","code":"data(dune) data(dune.env) mod <- rda(dune ~ Moisture + Management, dune.env) ## One simulation update(mod, simulate(mod) ~ .) #> Call: rda(formula = simulate(mod) ~ Moisture + Management, data = #> dune.env) #> #> Inertia Proportion Rank #> Total 79.3906 1.0000 #> Constrained 52.2955 0.6587 6 #> Unconstrained 27.0951 0.3413 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 24.007 14.238 5.712 3.314 2.879 2.145 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 6.128 4.391 3.331 3.143 2.121 1.757 1.488 1.152 1.115 1.044 0.682 0.539 0.205 #> ## An impression of confidence regions of site scores plot(mod, display=\"sites\") for (i in 1:5) lines(procrustes(mod, update(mod, simulate(mod) ~ .)), col=\"blue\") ## Simulate a set of null communities with permutation of residuals simulate(mod, indx = shuffleSet(nrow(dune), 99)) #> An object of class “simulate.rda” #> ‘simulate index’ method (abundance, non-sequential) #> 20 x 30 matrix #> Number of permuted matrices = 99 #>"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":null,"dir":"Reference","previous_headings":"","what":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Land birds islands covered coniferous forest Sipoo Archipelago, southern Finland.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map)"},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"sipoo data frame contains data occurrences 50 land bird species 18 islands Sipoo Archipelago (Simberloff & Martin, 1991, Appendix 3). species referred 4+4 letter abbreviation Latin names (using five letters two species names make unique). sipoo.map data contains geographic coordinates islands ETRS89-TM35FIN coordinate system (EPSG:3067) areas islands hectares.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"Simberloff, D. & Martin, J.-L. (1991). Nestedness insular avifaunas: simple summary statistics masking complex species patterns. Ornis Fennica 68:178--192.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sipoo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Birds in the Archipelago of Sipoo (Sibbo and Borgå) — sipoo","text":"","code":"data(sipoo) data(sipoo.map) plot(N ~ E, data=sipoo.map, asp = 1)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":null,"dir":"Reference","previous_headings":"","what":"Minimum Spanning Tree — spantree","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree connecting points, disregarding dissimilarities threshold NA.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Minimum Spanning Tree — spantree","text":"","code":"spantree(d, toolong = 0) # S3 method for spantree as.hclust(x, ...) # S3 method for spantree cophenetic(x) spandepth(x) # S3 method for spantree plot(x, ord, cex = 0.7, type = \"p\", labels, dlim, FUN = sammon, ...) # S3 method for spantree lines(x, ord, display=\"sites\", col = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Minimum Spanning Tree — spantree","text":"d Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . toolong = 0 (negative), dissimilarity regarded long. x spantree result object. ord ordination configuration, ordination result known scores. cex Character expansion factor. type Observations plotted points type=\"p\" type=\"b\", text label type=\"t\". tree (lines) always plotted. labels Text used type=\"t\" node names missing. dlim ceiling value used highest cophenetic dissimilarity. FUN Ordination function find configuration cophenetic dissimilarities. supplied FUN work, supply ordination result argument ord. display Type scores used ord. col Colour line segments. can vector recycled points, line colour mixture two joined points. ... parameters passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Minimum Spanning Tree — spantree","text":"Function spantree finds minimum spanning tree dissimilarities (may several minimum spanning trees, function finds one). Dissimilarities threshold toolong NAs disregarded, spanning tree found dissimilarities. data disconnected, function return disconnected tree (forest), corresponding link NA. Connected subtrees can identified using distconnected. Minimum spanning tree closely related single linkage clustering, .k.. nearest neighbour clustering, genetics neighbour joining tree available hclust agnes functions. important practical difference minimum spanning tree concept cluster membership, always joins individual points . Function .hclust can change spantree result corresponding hclust object. Function cophenetic finds distances points along tree segments. Function spandepth returns depth node. nodes tree either leaves (one link) internal nodes (one link). leaves recursively removed tree, depth layer leaf removed. disconnected spantree object (forest) tree analysed separately disconnected nodes tree depth zero. Function plot displays tree supplied ordination configuration, lines adds spanning tree ordination graph. configuration supplied plot, function ordinates cophenetic dissimilarities spanning tree overlays tree result. default ordination function sammon (package MASS), Sammon scaling emphasizes structure neighbourhood nodes may able beautifully represent tree (may need set dlim, sometimes results remain twisted). ordination methods work disconnected trees, must supply ordination configuration. Function lines overlay tree existing plot. Function spantree uses Prim's method implemented priority-first search dense graphs (Sedgewick 1990). Function cophenetic uses function stepacross option path = \"extended\". spantree fast, cophenetic slow large data sets.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Minimum Spanning Tree — spantree","text":"Function spantree returns object class spantree list two vectors, length \\(n-1\\). number links tree one less number observations, first item omitted. items kid child node parent, starting parent number two. link parent, value NA tree disconnected node. dist Corresponding distance. kid = NA, dist = 0. labels Names nodes found input dissimilarities. call function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Minimum Spanning Tree — spantree","text":"Sedgewick, R. (1990). Algorithms C. Addison Wesley.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Minimum Spanning Tree — spantree","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Minimum Spanning Tree — spantree","text":"principle, minimum spanning tree equivalent single linkage clustering can performed using hclust agnes. However, functions combine clusters information actually connected points (“single link”) recovered result. graphical output single linkage clustering plotted ordicluster look different equivalent spanning tree plotted lines.spantree.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/spantree.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Minimum Spanning Tree — spantree","text":"","code":"data(dune) dis <- vegdist(dune) tr <- spantree(dis) ## Add tree to a metric scaling plot(tr, cmdscale(dis), type = \"t\") ## Find a configuration to display the tree neatly plot(tr, type = \"t\") #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Depths of nodes depths <- spandepth(tr) plot(tr, type = \"t\", label = depths) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500 ## Plot as a dendrogram cl <- as.hclust(tr) plot(cl) ## cut hclust tree to classes and show in colours in spantree plot(tr, col = cutree(cl, 5), pch=16) #> Initial stress : 0.03111 #> stress after 10 iters: 0.01302, magic = 0.500 #> stress after 20 iters: 0.01139, magic = 0.500 #> stress after 30 iters: 0.01118, magic = 0.500 #> stress after 40 iters: 0.01114, magic = 0.500"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":null,"dir":"Reference","previous_headings":"","what":"Species Accumulation Curves — specaccum","title":"Species Accumulation Curves — specaccum","text":"Function specaccum finds species accumulation curves number species certain number sampled sites individuals.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species Accumulation Curves — specaccum","text":"","code":"specaccum(comm, method = \"exact\", permutations = 100, conditioned =TRUE, gamma = \"jack1\", w = NULL, subset, ...) # S3 method for specaccum plot(x, add = FALSE, random = FALSE, ci = 2, ci.type = c(\"bar\", \"line\", \"polygon\"), col = par(\"fg\"), lty = 1, ci.col = col, ci.lty = 1, ci.length = 0, xlab, ylab = x$method, ylim, xvar = c(\"sites\", \"individuals\", \"effort\"), ...) # S3 method for specaccum boxplot(x, add = FALSE, ...) fitspecaccum(object, model, method = \"random\", ...) # S3 method for fitspecaccum plot(x, col = par(\"fg\"), lty = 1, xlab = \"Sites\", ylab = x$method, ...) # S3 method for specaccum predict(object, newdata, interpolation = c(\"linear\", \"spline\"), ...) # S3 method for fitspecaccum predict(object, newdata, ...) specslope(object, at)"},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species Accumulation Curves — specaccum","text":"comm Community data set. method Species accumulation method (partial match). Method \"collector\" adds sites order happen data, \"random\" adds sites random order, \"exact\" finds expected (mean) species richness, \"coleman\" finds expected richness following Coleman et al. 1982, \"rarefaction\" finds mean accumulating individuals instead sites. permutations Number permutations method = \"random\". Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. conditioned Estimation standard deviation conditional empirical dataset exact SAC gamma Method estimating total extrapolated number species survey area function specpool w Weights giving sampling effort. subset logical expression indicating sites (rows) keep: missing values taken FALSE. x specaccum result object add Add existing graph. random Draw random simulation separately instead drawing average confidence intervals. ci Multiplier used get confidence intervals standard deviation (standard error estimate). Value ci = 0 suppresses drawing confidence intervals. ci.type Type confidence intervals graph: \"bar\" draws vertical bars, \"line\" draws lines, \"polygon\" draws shaded area. col Colour drawing lines. lty line type (see par). ci.col Colour drawing lines filling \"polygon\". ci.lty Line type confidence intervals border \"polygon\". ci.length Length horizontal bars (inches) end vertical bars ci.type = \"bar\". xlab,ylab Labels x (defaults xvar) y axis. ylim y limits plot. xvar Variable used horizontal axis: \"individuals\" can used method = \"rarefaction\". object Either community data set fitted specaccum model. model Nonlinear regression model (nls). See Details. newdata Optional data used prediction interpreted number sampling units (sites). missing, fitted values returned. interpolation Interpolation method used newdata. Number plots slope evaluated. Can real number. ... parameters functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species Accumulation Curves — specaccum","text":"Species accumulation curves (SAC) used compare diversity properties community data sets using different accumulator functions. classic method \"random\" finds mean SAC standard deviation random permutations data, subsampling without replacement (Gotelli & Colwell 2001). \"exact\" method finds expected SAC using sample-based rarefaction method independently developed numerous times (Chiarucci et al. 2008) often known Mao Tau estimate (Colwell et al. 2012). unconditional standard deviation exact SAC represents moment-based estimation conditioned empirical data set (sd samples > 0). unconditional standard deviation based estimation extrapolated number species survey area (.k.. gamma diversity), estimated function specpool. conditional standard deviation developed Jari Oksanen ( published, sd=0 samples). Method \"coleman\" finds expected SAC standard deviation following Coleman et al. (1982). methods based sampling sites without replacement. contrast, method = \"rarefaction\" finds expected species richness standard deviation sampling individuals instead sites. achieves applying function rarefy number individuals corresponding average number individuals per site. Methods \"random\" \"collector\" can take weights (w) give sampling effort site. weights w influence order sites accumulated, value sampling effort sites equal. summary results expressed sites even accumulation uses weights (methods \"random\", \"collector\"), based individuals (\"rarefaction\"). actual sampling effort given item Effort Individuals printed result. weighted \"random\" method effort refers average effort per site, sum weights per number sites. weighted method = \"random\", averaged species richness found linear interpolation single random permutations. Therefore least first value (often several first) NA richness, values interpolated cases extrapolated. plot function defaults display results scaled sites, can changed selecting xvar = \"effort\" (weighted methods) xvar = \"individuals\" (method = \"rarefaction\"). summary boxplot methods available method = \"random\". Function predict specaccum can return values corresponding newdata. method \"exact\", \"rarefaction\" \"coleman\" function uses analytic equations interpolated non-integer values, methods linear (approx) spline (spline) interpolation. newdata given, function returns values corresponding data. NB., fitted values method=\"rarefaction\" based rounded integer counts, predict can use fractional non-integer counts newdata give slightly different results. Function fitspecaccum fits nonlinear (nls) self-starting species accumulation model. input object can result specaccum community data frame. latter case function first fits specaccum model proceeds fitting nonlinear model. function can apply limited set nonlinear regression models suggested species-area relationship (Dengler 2009). selfStart models. permissible alternatives \"arrhenius\" (SSarrhenius), \"gleason\" (SSgleason), \"gitay\" (SSgitay), \"lomolino\" (SSlomolino) vegan package. addition following standard R models available: \"asymp\" (SSasymp), \"gompertz\" (SSgompertz), \"michaelis-menten\" (SSmicmen), \"logis\" (SSlogis), \"weibull\" (SSweibull). See functions model specification details. weights w used fit based accumulated effort model = \"rarefaction\" accumulated number individuals. plot still based sites, unless alternative selected xvar. Function predict fitspecaccum uses predict.nls, can pass arguments function. addition, fitted, residuals, nobs, coef, AIC, logLik deviance work result object. Function specslope evaluates derivative species accumulation curve given number sample plots, gives rate increase number species. function works specaccum result object based analytic models \"exact\", \"rarefaction\" \"coleman\", non-linear regression results fitspecaccum. Nonlinear regression may fail reason, fitspecaccum models fragile may succeed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species Accumulation Curves — specaccum","text":"Function specaccum returns object class \"specaccum\", fitspecaccum model class \"fitspecaccum\" adds items \"specaccum\" (see end list ): call Function call. method Accumulator method. sites Number sites. method = \"rarefaction\" number sites corresponding certain number individuals generally integer, average number individuals also returned item individuals. effort Average sum weights corresponding number sites model fitted argument w richness number species corresponding number sites. method = \"collector\" observed richness, methods average expected richness. sd standard deviation SAC (standard error). NULL method = \"collector\", estimated permutations method = \"random\", analytic equations methods. perm Permutation results method = \"random\" NULL cases. column perm holds one permutation. weights Matrix accumulated weights corresponding columns perm matrix model fitted argument w. fitted, residuals, coefficients fitspecacum: fitted values, residuals nonlinear model coefficients. method = \"random\" matrices column random accumulation. models fitspecaccum: list fitted nls models (see Examples accessing models).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Species Accumulation Curves — specaccum","text":"Chiarucci, ., Bacaro, G., Rocchini, D. & Fattorini, L. (2008). Discovering rediscovering sample-based rarefaction formula ecological literature. Commun. Ecol. 9: 121--123. Coleman, B.D, Mares, M.., Willis, M.R. & Hsieh, Y. (1982). Randomness, area species richness. Ecology 63: 1121--1133. Colwell, R.K., Chao, ., Gotelli, N.J., Lin, S.Y., Mao, C.X., Chazdon, R.L. & Longino, J.T. (2012). Models estimators linking individual-based sample-based rarefaction, extrapolation comparison assemblages. J. Plant Ecol. 5: 3--21. Dengler, J. (2009). function describes species-area relationship best? review empirical evaluation. Journal Biogeography 36, 728--744. Gotelli, N.J. & Colwell, R.K. (2001). Quantifying biodiversity: procedures pitfalls measurement comparison species richness. Ecol. Lett. 4, 379--391.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species Accumulation Curves — specaccum","text":"Roeland Kindt r.kindt@cgiar.org Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Species Accumulation Curves — specaccum","text":"SAC method = \"exact\" developed Roeland Kindt, standard deviation Jari Oksanen (unpublished). method = \"coleman\" underestimates SAC handle properly sampling without replacement. , standard deviation take account species correlations, generally low.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specaccum.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species Accumulation Curves — specaccum","text":"","code":"data(BCI) sp1 <- specaccum(BCI) #> Warning: the standard deviation is zero sp2 <- specaccum(BCI, \"random\") sp2 #> Species Accumulation Curve #> Accumulation method: random, with 100 permutations #> Call: specaccum(comm = BCI, method = \"random\") #> #> #> Sites 1.00000 2.00000 3.00000 4.00000 5.00000 6.00000 7.00000 #> Richness 90.42000 121.29000 138.34000 150.46000 158.46000 165.38000 170.34000 #> sd 7.23931 7.43361 7.38305 6.80377 5.92959 5.07475 4.58196 #> #> Sites 8.00000 9.00000 10.00000 11.0000 12.00000 13.00000 14.0000 #> Richness 175.02000 178.93000 182.15000 184.8500 187.47000 189.88000 191.9000 #> sd 4.43808 4.74662 4.58671 4.4093 4.23419 4.08565 4.1451 #> #> Sites 15.00000 16.00000 17.00000 18.00000 19.00000 20.00000 21.00000 #> Richness 193.75000 195.50000 197.32000 198.94000 200.32000 201.68000 202.95000 #> sd 4.13503 4.01889 3.54447 3.53859 3.62617 3.62895 3.41528 #> #> Sites 22.00000 23.00000 24.00000 25.00000 26.00000 27.000 28.00000 #> Richness 204.17000 205.39000 206.65000 207.73000 208.91000 209.880 210.90000 #> sd 3.47009 3.28417 3.22357 3.06448 3.06856 3.217 3.14787 #> #> Sites 29.00000 30.00000 31.00000 32.00000 33.00000 34.00000 35.00000 #> Richness 211.64000 212.74000 213.57000 214.43000 215.15000 215.93000 216.75000 #> sd 3.12862 3.01719 3.01931 2.92069 2.76111 2.77163 2.69446 #> #> Sites 36.00000 37.00000 38.00000 39.00000 40.00000 41.00000 42.00000 #> Richness 217.54000 218.20000 219.00000 219.61000 220.14000 220.77000 221.28000 #> sd 2.62629 2.64384 2.45361 2.22427 2.11307 1.93769 1.82618 #> #> Sites 43.00000 44.00000 45.00000 46.00000 47.00000 48.0000 49.00000 #> Richness 221.86000 222.33000 222.85000 223.28000 223.69000 224.1400 224.60000 #> sd 1.79235 1.49784 1.45904 1.36389 1.07961 0.8764 0.58603 #> #> Sites 50 #> Richness 225 #> sd 0 summary(sp2) #> 1 sites 2 sites 3 sites 4 sites #> Min. : 77.00 Min. :105.0 Min. :119.0 Min. :132.0 #> 1st Qu.: 85.00 1st Qu.:115.8 1st Qu.:134.0 1st Qu.:146.0 #> Median : 88.00 Median :121.5 Median :139.0 Median :151.0 #> Mean : 90.42 Mean :121.3 Mean :138.3 Mean :150.5 #> 3rd Qu.: 94.25 3rd Qu.:126.0 3rd Qu.:143.2 3rd Qu.:155.0 #> Max. :109.00 Max. :144.0 Max. :155.0 Max. :166.0 #> 5 sites 6 sites 7 sites 8 sites 9 sites #> Min. :140.0 Min. :149.0 Min. :157.0 Min. :161 Min. :166.0 #> 1st Qu.:155.0 1st Qu.:162.0 1st Qu.:168.0 1st Qu.:173 1st Qu.:176.0 #> Median :159.0 Median :165.0 Median :171.0 Median :175 Median :179.0 #> Mean :158.5 Mean :165.4 Mean :170.3 Mean :175 Mean :178.9 #> 3rd Qu.:163.0 3rd Qu.:170.0 3rd Qu.:174.0 3rd Qu.:178 3rd Qu.:182.0 #> Max. :172.0 Max. :176.0 Max. :181.0 Max. :186 Max. :190.0 #> 10 sites 11 sites 12 sites 13 sites #> Min. :170.0 Min. :174.0 Min. :174.0 Min. :179.0 #> 1st Qu.:179.0 1st Qu.:182.0 1st Qu.:184.0 1st Qu.:187.0 #> Median :183.0 Median :185.0 Median :188.0 Median :190.5 #> Mean :182.2 Mean :184.8 Mean :187.5 Mean :189.9 #> 3rd Qu.:185.0 3rd Qu.:188.0 3rd Qu.:191.0 3rd Qu.:193.0 #> Max. :191.0 Max. :193.0 Max. :200.0 Max. :202.0 #> 14 sites 15 sites 16 sites 17 sites #> Min. :181.0 Min. :183.0 Min. :185.0 Min. :188.0 #> 1st Qu.:189.0 1st Qu.:191.0 1st Qu.:193.0 1st Qu.:195.0 #> Median :193.0 Median :194.0 Median :196.0 Median :197.5 #> Mean :191.9 Mean :193.8 Mean :195.5 Mean :197.3 #> 3rd Qu.:194.0 3rd Qu.:196.0 3rd Qu.:198.0 3rd Qu.:200.0 #> Max. :203.0 Max. :204.0 Max. :205.0 Max. :205.0 #> 18 sites 19 sites 20 sites 21 sites #> Min. :188.0 Min. :188.0 Min. :190.0 Min. :195.0 #> 1st Qu.:196.8 1st Qu.:198.0 1st Qu.:199.0 1st Qu.:201.0 #> Median :199.0 Median :200.0 Median :202.0 Median :203.0 #> Mean :198.9 Mean :200.3 Mean :201.7 Mean :202.9 #> 3rd Qu.:201.0 3rd Qu.:203.0 3rd Qu.:204.0 3rd Qu.:205.0 #> Max. :208.0 Max. :209.0 Max. :210.0 Max. :211.0 #> 22 sites 23 sites 24 sites 25 sites #> Min. :195.0 Min. :199.0 Min. :199.0 Min. :200.0 #> 1st Qu.:202.0 1st Qu.:203.0 1st Qu.:204.0 1st Qu.:206.0 #> Median :204.0 Median :205.0 Median :207.0 Median :207.5 #> Mean :204.2 Mean :205.4 Mean :206.7 Mean :207.7 #> 3rd Qu.:206.2 3rd Qu.:207.0 3rd Qu.:209.0 3rd Qu.:210.0 #> Max. :214.0 Max. :214.0 Max. :214.0 Max. :215.0 #> 26 sites 27 sites 28 sites 29 sites #> Min. :203.0 Min. :203.0 Min. :204.0 Min. :204.0 #> 1st Qu.:207.0 1st Qu.:208.0 1st Qu.:209.0 1st Qu.:210.0 #> Median :209.0 Median :210.0 Median :211.0 Median :212.0 #> Mean :208.9 Mean :209.9 Mean :210.9 Mean :211.6 #> 3rd Qu.:211.0 3rd Qu.:212.0 3rd Qu.:213.0 3rd Qu.:214.0 #> Max. :217.0 Max. :218.0 Max. :220.0 Max. :220.0 #> 30 sites 31 sites 32 sites 33 sites #> Min. :206.0 Min. :206.0 Min. :206.0 Min. :207.0 #> 1st Qu.:211.0 1st Qu.:212.0 1st Qu.:213.0 1st Qu.:214.0 #> Median :212.5 Median :213.0 Median :214.0 Median :215.0 #> Mean :212.7 Mean :213.6 Mean :214.4 Mean :215.2 #> 3rd Qu.:215.0 3rd Qu.:215.0 3rd Qu.:216.0 3rd Qu.:216.0 #> Max. :221.0 Max. :221.0 Max. :222.0 Max. :222.0 #> 34 sites 35 sites 36 sites 37 sites #> Min. :208.0 Min. :208.0 Min. :208.0 Min. :208.0 #> 1st Qu.:214.0 1st Qu.:215.0 1st Qu.:216.0 1st Qu.:217.0 #> Median :216.0 Median :217.0 Median :218.0 Median :218.0 #> Mean :215.9 Mean :216.8 Mean :217.5 Mean :218.2 #> 3rd Qu.:218.0 3rd Qu.:219.0 3rd Qu.:219.0 3rd Qu.:220.0 #> Max. :222.0 Max. :222.0 Max. :223.0 Max. :224.0 #> 38 sites 39 sites 40 sites 41 sites #> Min. :212.0 Min. :212.0 Min. :212.0 Min. :212.0 #> 1st Qu.:217.0 1st Qu.:218.0 1st Qu.:219.0 1st Qu.:220.0 #> Median :219.0 Median :220.0 Median :220.0 Median :221.0 #> Mean :219.0 Mean :219.6 Mean :220.1 Mean :220.8 #> 3rd Qu.:220.2 3rd Qu.:221.0 3rd Qu.:221.0 3rd Qu.:222.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 42 sites 43 sites 44 sites 45 sites #> Min. :215.0 Min. :215.0 Min. :218.0 Min. :219.0 #> 1st Qu.:220.0 1st Qu.:221.0 1st Qu.:221.0 1st Qu.:222.0 #> Median :221.0 Median :222.0 Median :222.0 Median :223.0 #> Mean :221.3 Mean :221.9 Mean :222.3 Mean :222.8 #> 3rd Qu.:222.2 3rd Qu.:223.0 3rd Qu.:223.2 3rd Qu.:224.0 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 #> 46 sites 47 sites 48 sites 49 sites 50 sites #> Min. :220.0 Min. :221.0 Min. :222.0 Min. :223.0 Min. :225 #> 1st Qu.:222.8 1st Qu.:223.0 1st Qu.:224.0 1st Qu.:224.0 1st Qu.:225 #> Median :223.0 Median :224.0 Median :224.0 Median :225.0 Median :225 #> Mean :223.3 Mean :223.7 Mean :224.1 Mean :224.6 Mean :225 #> 3rd Qu.:224.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225.0 3rd Qu.:225 #> Max. :225.0 Max. :225.0 Max. :225.0 Max. :225.0 Max. :225 plot(sp1, ci.type=\"poly\", col=\"blue\", lwd=2, ci.lty=0, ci.col=\"lightblue\") boxplot(sp2, col=\"yellow\", add=TRUE, pch=\"+\") ## Fit Lomolino model to the exact accumulation mod1 <- fitspecaccum(sp1, \"lomolino\") coef(mod1) #> Asym xmid slope #> 258.440682 2.442061 1.858694 fitted(mod1) #> [1] 94.34749 121.23271 137.45031 148.83053 157.45735 164.31866 169.95946 #> [8] 174.71115 178.78954 182.34254 185.47566 188.26658 190.77402 193.04337 #> [15] 195.11033 197.00350 198.74606 200.35705 201.85227 203.24499 204.54643 #> [22] 205.76612 206.91229 207.99203 209.01150 209.97609 210.89054 211.75903 #> [29] 212.58527 213.37256 214.12386 214.84180 215.52877 216.18692 216.81820 #> [36] 217.42437 218.00703 218.56767 219.10762 219.62811 220.13027 220.61514 #> [43] 221.08369 221.53679 221.97528 222.39991 222.81138 223.21037 223.59747 #> [50] 223.97327 plot(sp1) ## Add Lomolino model using argument 'add' plot(mod1, add = TRUE, col=2, lwd=2) ## Fit Arrhenius models to all random accumulations mods <- fitspecaccum(sp2, \"arrh\") plot(mods, col=\"hotpink\") boxplot(sp2, col = \"yellow\", border = \"blue\", lty=1, cex=0.3, add= TRUE) ## Use nls() methods to the list of models sapply(mods$models, AIC) #> [1] 329.2456 321.8614 331.1568 352.1309 354.8606 323.3767 322.0897 347.7218 #> [9] 327.6283 327.3510 335.4843 308.5300 373.4959 334.3691 320.6975 338.1084 #> [17] 330.2018 343.5560 348.5827 300.2513 315.7909 310.7837 305.5217 295.1005 #> [25] 359.2639 342.7111 339.7456 328.4175 338.4335 368.4988 354.7097 328.8704 #> [33] 342.1528 292.9875 328.3752 299.7201 336.5796 364.3153 336.4822 339.6666 #> [41] 354.7445 320.7841 304.1628 314.9640 349.6753 350.7169 307.6840 296.7971 #> [49] 359.6018 302.9863 319.6832 339.5021 328.8450 329.0553 298.0970 334.0182 #> [57] 291.9027 349.7161 325.1376 347.7797 340.6111 338.6148 352.0292 308.8538 #> [65] 315.1890 289.7947 351.6732 339.3587 320.3515 300.9377 323.3054 290.9419 #> [73] 350.8587 322.6970 340.7948 334.1965 358.2890 300.7034 338.5079 338.8941 #> [81] 341.7687 331.4553 337.5975 349.7867 331.8581 270.3068 354.5083 326.0026 #> [89] 337.8555 322.7217 302.5019 324.7469 322.1415 324.6922 324.5427 347.2050 #> [97] 309.1766 316.1194 343.9958 350.5935"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":null,"dir":"Reference","previous_headings":"","what":"Extrapolated Species Richness in a Species Pool — specpool","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions estimate extrapolated species richness species pool, number unobserved species. Function specpool based incidences sample sites, gives single estimate collection sample sites (matrix). Function estimateR based abundances (counts) single sample site.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"specpool(x, pool, smallsample = TRUE) estimateR(x, ...) specpool2vect(X, index = c(\"jack1\",\"jack2\", \"chao\", \"boot\",\"Species\")) poolaccum(x, permutations = 100, minsize = 3) estaccumR(x, permutations = 100, parallel = getOption(\"mc.cores\")) # S3 method for poolaccum summary(object, display, alpha = 0.05, ...) # S3 method for poolaccum plot(x, alpha = 0.05, type = c(\"l\",\"g\"), ...)"},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"x Data frame matrix species data analysis result plot function. pool vector giving classification pooling sites species data. missing, sites pooled together. smallsample Use small sample correction \\((N-1)/N\\), \\(N\\) number sites within pool. X, object specpool result object. index selected index extrapolated richness. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. minsize Smallest number sampling units reported. parallel Number parallel processes predefined socket cluster. parallel = 1 uses ordinary, non-parallel processing. parallel processing done parallel package. display Indices displayed. alpha Level quantiles shown. proportion left outside symmetric limits. type Type graph produced xyplot. ... parameters (used).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Many species always remain unseen undetected collection sample plots. function uses popular ways estimating number unseen species adding observed species richness (Palmer 1990, Colwell & Coddington 1994). incidence-based estimates specpool use frequencies species collection sites. following, \\(S_P\\) extrapolated richness pool, \\(S_0\\) observed number species collection, \\(a_1\\) \\(a_2\\) number species occurring one two sites collection, \\(p_i\\) frequency species \\(\\), \\(N\\) number sites collection. variants extrapolated richness specpool : specpool normally uses basic Chao equation, doubletons (\\(a2=0\\)) switches bias-corrected version. case Chao equation simplifies \\(S_0 + \\frac{1}{2} a_1 (a_1-1) \\frac{N-1}{N}\\). abundance-based estimates estimateR use counts (numbers individuals) species single site. called matrix data frame, function give separate estimates site. two variants extrapolated richness estimateR bias-corrected Chao ACE (O'Hara 2005, Chiu et al. 2014). Chao estimate similar bias corrected one , \\(a_i\\) refers number species abundance \\(\\) instead number sites, small-sample correction used. ACE estimate defined : \\(a_i\\) refers number species abundance \\(\\) \\(S_{rare}\\) number rare species, \\(S_{abund}\\) number abundant species, arbitrary threshold abundance 10 rare species, \\(N_{rare}\\) number individuals rare species. Functions estimate standard errors estimates. concern number added species, assume variance observed richness. equations standard errors complicated reproduced help page, can studied R source code function discussed vignette can read browseVignettes(\"vegan\"). standard error based following sources: Chiu et al. (2014) Chao estimates Smith van Belle (1984) first-order Jackknife bootstrap (second-order jackknife still missing). variance estimator \\(S_{ace}\\) see O'Hara (2005). Functions poolaccum estaccumR similar specaccum, estimate extrapolated richness indices specpool estimateR addition number species random ordering sampling units. Function specpool uses presence data estaccumR count data. functions share summary plot methods. summary returns quantile envelopes permutations corresponding given level alpha standard deviation permutations sample size. NB., based standard deviations estimated within specpool estimateR, based permutations. plot function shows mean envelope permutations given alpha models. selection models can restricted order changes using display argument summary plot. configuration plot command, see xyplot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Function specpool returns data frame entries observed richness indices class pool vector. utility function specpool2vect maps pooled values vector giving value selected index original site. Function estimateR returns estimates standard errors site. Functions poolaccum estimateR return matrices permutation results richness estimator, vector sample sizes table means permutations estimator.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Chao, . (1987). Estimating population size capture-recapture data unequal catchability. Biometrics 43, 783--791. Chiu, C.H., Wang, Y.T., Walther, B.. & Chao, . (2014). Improved nonparametric lower bound species richness via modified Good-Turing frequency formula. Biometrics 70, 671--682. Colwell, R.K. & Coddington, J.. (1994). Estimating terrestrial biodiversity extrapolation. Phil. Trans. Roy. Soc. London B 345, 101--118. O'Hara, R.B. (2005). Species richness estimators: many species can dance head pin? J. Anim. Ecol. 74, 375--386. Palmer, M.W. (1990). estimation species richness extrapolation. Ecology 71, 1195--1198. Smith, E.P & van Belle, G. (1984). Nonparametric estimation species richness. Biometrics 40, 119--129.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"Bob O'Hara (estimateR) Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"functions based assumption species pool: community closed fixed pool size \\(S_P\\). general, functions give lower limit species richness: real richness \\(S >= S_P\\), consistent bias estimates. Even bias-correction Chao reduces bias, remove completely (Chiu et al. 2014). Optional small sample correction added specpool vegan 2.2-0. used older literature (Chao 1987), recommended recently (Chiu et al. 2014).","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/specpool.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Extrapolated Species Richness in a Species Pool — specpool","text":"","code":"data(dune) data(dune.env) pool <- with(dune.env, specpool(dune, Management)) pool #> Species chao chao.se jack1 jack1.se jack2 boot boot.se n #> BF 16 17.19048 1.5895675 19.33333 2.211083 19.83333 17.74074 1.646379 3 #> HF 21 21.51429 0.9511693 23.40000 1.876166 22.05000 22.56864 1.821518 5 #> NM 21 22.87500 2.1582871 26.00000 3.291403 25.73333 23.77696 2.300982 6 #> SF 21 29.88889 8.6447967 27.66667 3.496029 31.40000 23.99496 1.850288 6 op <- par(mfrow=c(1,2)) boxplot(specnumber(dune) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") boxplot(specnumber(dune)/specpool2vect(pool) ~ Management, data = dune.env, col = \"hotpink\", border = \"cyan3\") par(op) data(BCI) ## Accumulation model pool <- poolaccum(BCI) summary(pool, display = \"chao\") #> $chao #> N Chao 2.5% 97.5% Std.Dev #> [1,] 3 162.3374 142.5272 186.2227 11.217082 #> [2,] 4 176.1243 156.4704 206.3243 12.158304 #> [3,] 5 183.8507 162.9868 209.2821 12.488675 #> [4,] 6 188.9018 165.6732 214.2779 12.888148 #> [5,] 7 193.8679 175.8712 216.1384 11.602711 #> [6,] 8 199.0315 180.2949 227.8918 12.604562 #> [7,] 9 202.0946 183.3317 228.8381 11.958432 #> [8,] 10 204.9843 185.3556 227.5898 12.637762 #> [9,] 11 206.6160 186.7205 234.5167 11.679759 #> [10,] 12 209.1286 189.2824 232.6359 12.351073 #> [11,] 13 211.3655 191.7918 232.0423 11.991967 #> [12,] 14 212.8128 195.8592 239.3318 11.537817 #> [13,] 15 215.5275 198.2739 239.7978 11.422201 #> [14,] 16 218.7094 198.3891 243.3826 12.165145 #> [15,] 17 221.6814 197.6029 256.5328 14.464302 #> [16,] 18 222.7418 201.4964 257.9494 13.227892 #> [17,] 19 224.8619 206.6236 251.9979 13.165360 #> [18,] 20 226.3324 210.2832 249.5616 12.858116 #> [19,] 21 228.4284 210.9920 257.9636 13.784663 #> [20,] 22 228.8425 211.1903 254.3500 11.629250 #> [21,] 23 230.6429 214.1737 251.1681 10.510649 #> [22,] 24 231.9109 213.0874 252.7553 10.925851 #> [23,] 25 232.9972 214.3610 258.4917 11.207306 #> [24,] 26 233.8067 216.9697 263.2733 11.546339 #> [25,] 27 235.6347 217.3831 266.9230 12.800245 #> [26,] 28 235.8942 218.6913 261.4115 11.761873 #> [27,] 29 235.9612 218.6591 258.5682 11.133700 #> [28,] 30 236.6706 219.4315 259.9315 10.291499 #> [29,] 31 236.8530 220.2338 259.2903 9.810390 #> [30,] 32 237.6018 221.4868 262.0864 10.195825 #> [31,] 33 237.4291 222.2990 256.2778 9.227891 #> [32,] 34 237.6712 222.7844 258.2268 8.979536 #> [33,] 35 237.7856 222.2750 257.2540 8.537867 #> [34,] 36 238.4325 223.6473 255.5610 8.566487 #> [35,] 37 238.1834 224.9892 253.5811 7.776992 #> [36,] 38 238.4685 225.2802 254.0155 7.626532 #> [37,] 39 238.2082 226.0390 252.7975 7.642443 #> [38,] 40 238.7991 227.1971 254.1619 7.155802 #> [39,] 41 239.2246 227.3368 258.5178 7.926574 #> [40,] 42 238.4935 228.5195 256.2137 7.287824 #> [41,] 43 238.3856 227.8327 255.8470 7.300153 #> [42,] 44 238.0092 228.5279 252.9488 6.520961 #> [43,] 45 238.1537 228.7980 253.3757 6.443894 #> [44,] 46 238.0532 230.5990 250.9539 5.466587 #> [45,] 47 237.7542 231.3089 251.0192 4.821564 #> [46,] 48 237.2842 231.9015 248.6399 4.038862 #> [47,] 49 236.8407 233.3115 245.4082 2.803888 #> [48,] 50 236.3732 236.3732 236.3732 0.000000 #> #> attr(,\"class\") #> [1] \"summary.poolaccum\" plot(pool) ## Quantitative model estimateR(BCI[1:5,]) #> 1 2 3 4 5 #> S.obs 93.000000 84.000000 90.000000 94.000000 101.000000 #> S.chao1 117.473684 117.214286 141.230769 111.550000 136.000000 #> se.chao1 11.583785 15.918953 23.001405 8.919663 15.467344 #> S.ACE 122.848959 117.317307 134.669844 118.729941 137.114088 #> se.ACE 5.736054 5.571998 6.191618 5.367571 5.848474"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":null,"dir":"Reference","previous_headings":"","what":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distance-based ordination (dbrda, capscale, metaMDS) information species, methods may add species scores community data available. However, species scores may missing ( always dbrda), may close relation used dissimilarity index. function add species scores replace existing species scores distance-based methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"sppscores(object) <- value"},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"object Ordination result. value Community data find species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Distances information species (columns, variables), hence distance-based ordination information species scores. However, species scores can added supplementary information analysis help interpretation results. ordination methods (capscale, metaMDS) can supplement species scores analysis community data available analysis. capscale species scores found projecting community data site ordination (linear combination scores), scores accurate analysis used Euclidean distances. dissimilarity index can expressed Euclidean distances transformed data (instance, Chord Hellinger Distances), species scores based transformed data accurate, function still finds dissimilarities untransformed data. Usually community dissimilarities differ two significant ways Euclidean distances: bound maximum 1, use absolute differences instead squared differences. cases, may better use species scores transformed Euclidean distances good linear relation used dissimilarities. often useful standardize data row unit total, perform squareroot transformation damp effect squared differences (see Examples). Function dbrda never finds species scores, mathematically similar capscale, similar rules followed supplementing species scores. Function metaMDS uses weighted averages (wascores) find species scores. better relationship dissimilarities projection scores used metric ordination, similar transformation community data used dissimilarities species scores.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Replacement function adds species scores replaces old scores ordination object.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/sppscores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add or Replace Species Scores in Distance-Based Ordination — sppscores","text":"","code":"data(BCI, BCI.env) mod <- dbrda(vegdist(BCI) ~ Habitat, BCI.env) ## add species scores sppscores(mod) <- BCI ## Euclidean distances of BCI differ from used dissimilarity plot(vegdist(BCI), dist(BCI)) ## more linear relationship plot(vegdist(BCI), dist(sqrt(decostand(BCI, \"total\")))) ## better species scores sppscores(mod) <- sqrt(decostand(BCI, \"total\"))"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":null,"dir":"Reference","previous_headings":"","what":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function stepacross tries replace dissimilarities shortest paths stepping across intermediate sites regarding dissimilarities threshold missing data (NA). path = \"shortest\" flexible shortest path (Williamson 1978, Bradfield & Kenkel 1987), path = \"extended\" approximation known extended dissimilarities (De'ath 1999). use stepacross improve ordination high beta diversity, many sites species common.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"stepacross(dis, path = \"shortest\", toolong = 1, trace = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"dis Dissimilarity data inheriting class dist object, matrix, can converted dissimilarity matrix. Functions vegdist dist functions producing suitable dissimilarity data. path method stepping across (partial match) Alternative \"shortest\" finds shortest paths, \"extended\" approximation known extended dissimilarities. toolong Shortest dissimilarity regarded NA. function uses fuzz factor, dissimilarities close limit made NA, . trace Trace calculations. ... parameters (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Williamson (1978) suggested using flexible shortest paths estimate dissimilarities sites nothing common, shared species. path = \"shortest\" function stepacross replaces dissimilarities toolong longer NA, tries find shortest paths sites using remaining dissimilarities. Several dissimilarity indices semi-metric means obey triangle inequality \\(d_{ij} \\leq d_{ik} + d_{kj}\\), shortest path algorithm can replace dissimilarities well, even shorter toolong. De'ath (1999) suggested simplified method known extended dissimilarities, calculated path = \"extended\". method, dissimilarities toolong longer first made NA, function tries replace NA dissimilarities path single stepping stone points. NA replaced one pass, function make new passes updated dissimilarities long NA replaced extended dissimilarities. mean second passes, remaining NA dissimilarities allowed one stepping stone site, previously replaced dissimilarities updated. , function consider dissimilarities shorter toolong, although replaced shorter path semi-metric indices, used part paths. optimal cases, extended dissimilarities equal shortest paths, may longer. alternative defining long dissimilarities parameter toolong, input dissimilarities can contain NAs. toolong zero negative, function make dissimilarities NA. NAs input toolong = 0, path = \"shortest\" find shorter paths semi-metric indices, path = \"extended\" nothing. Function .shared can used set dissimilarities NA. data disconnected path points, result contain NAs warning issued. Several methods handle NA dissimilarities, warning taken seriously. Function distconnected can used find connected groups remove rare outlier observations groups observations. Alternative path = \"shortest\" uses Dijkstra's method finding flexible shortest paths, implemented priority-first search dense graphs (Sedgewick 1990). Alternative path = \"extended\" follows De'ath (1999), implementation simpler code.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Function returns object class dist extended dissimilarities (see functions vegdist dist). value path appended method attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Bradfield, G.E. & Kenkel, N.C. (1987). Nonlinear ordination using flexible shortest path adjustment ecological distances. Ecology 68, 750--753. De'ath, G. (1999). Extended dissimilarity: method robust estimation ecological distances high beta diversity data. Plant Ecol. 144, 191--199. Sedgewick, R. (1990). Algorithms C. Addison Wesley. Williamson, M.H. (1978). ordination incidence data. J. Ecol. 66, 911-920.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"function changes original dissimilarities, like . may best use function really must: extremely high beta diversity large proportion dissimilarities upper limit (species common). Semi-metric indices vary degree violating triangle inequality. Morisita Horn--Morisita indices vegdist may strongly semi-metric, shortest paths can change indices much. Mountford index violates basic rules dissimilarities: non-identical sites zero dissimilarity species composition poorer site subset richer. Mountford index, can find three sites \\(, j, k\\) \\(d_{ik} = 0\\) \\(d_{jk} = 0\\), \\(d_{ij} > 0\\). results stepacross Mountford index can weird. stepacross needed, best try use metric indices .","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stepacross.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Stepacross as Flexible Shortest Paths or Extended Dissimilarities — stepacross","text":"","code":"# There are no data sets with high beta diversity in vegan, but this # should give an idea. data(dune) dis <- vegdist(dune) edis <- stepacross(dis) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities... plot(edis, dis, xlab = \"Shortest path\", ylab = \"Original\") ## Manhattan distance have no fixed upper limit. dis <- vegdist(dune, \"manhattan\") is.na(dis) <- no.shared(dune) dis <- stepacross(dis, toolong=0) #> Too long or NA distances: 5 out of 190 (2.6%) #> Stepping across 190 dissimilarities..."},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions plot ordination distances given number dimensions observed distances distances full space eigenvector methods. display similar Shepard diagram (stressplot non-metric multidimensional scaling metaMDS monoMDS), shows linear relationship eigenvector ordinations. stressplot methods available wcmdscale, rda, cca, capscale, dbrda, prcomp princomp.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"# S3 method for wcmdscale stressplot(object, k = 2, pch, p.col = \"blue\", l.col = \"red\", lwd = 2, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"object Result object eigenvector ordination (wcmdscale, rda, cca, dbrda, capscale) k Number dimensions ordination distances displayed. pch, p.col, l.col, lwd Plotting character, point colour line colour like default stressplot ... parameters functions, e.g. graphical parameters.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"functions offer similar display eigenvector ordinations standard Shepard diagram (stressplot) non-metric multidimensional scaling. ordination distances given number dimensions plotted observed distances. metric distances, ordination distances full space (ordination axes) equal observed distances, fit line shows equality. general, fit line go points, points observed distances approach fit line . However, non-Euclidean distances ( wcmdscale, dbrda capscale) negative eigenvalues ordination distances can exceed observed distances real dimensions; imaginary dimensions negative eigenvalues correct excess distances. used dbrda, capscale wcmdscale argument add avoid negative eigenvalues, ordination distances exceed observed dissimilarities. partial ordination (cca, rda, capscale Condition formula), distances partial component included observed distances ordination distances. k=0, ordination distances refer partial ordination. exception dbrda distances partial, constrained residual components additive, first components can shown, partial models shown .","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Functions draw graph return invisibly ordination distances ordination distances.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"Jari Oksanen.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/stressplot.wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Ordination Distances Against Observed Distances in Eigenvector Ordinations — stressplot.wcmdscale","text":"","code":"data(dune, dune.env) mod <- rda(dune) stressplot(mod) mod <- rda(dune ~ Management, dune.env) stressplot(mod, k=3)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":null,"dir":"Reference","previous_headings":"","what":"Indices of Taxonomic Diversity and Distinctness — taxondive","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function finds indices taxonomic diversity distinctness, averaged taxonomic distances among species individuals community (Clarke & Warwick 1998, 2001)","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"taxondive(comm, dis, match.force = FALSE) taxa2dist(x, varstep = FALSE, check = TRUE, labels)"},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"comm Community data. dis Taxonomic distances among taxa comm. dist object symmetric square matrix. match.force Force matching column names comm labels dis. FALSE, matching happens dimensions differ, case species must identical order . x Classification table row species basic taxon, columns identifiers classification higher levels. varstep Vary step lengths successive levels relative proportional loss number distinct classes. check TRUE, remove redundant levels different rows constant rows regard row different basal taxon (species). FALSE levels retained basal taxa (species) also must coded variables (columns). get warning species coded, can ignore intention. labels labels attribute taxonomic distances. Row names used given. Species matched labels comm dis taxondive different dimensions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke & Warwick (1998, 2001) suggested several alternative indices taxonomic diversity distinctness. Two basic indices called taxonomic diversity (\\(\\Delta\\)) distinctness (\\(\\Delta^*\\)): equations give index value single site, summation goes species \\(\\) \\(j\\). \\(\\omega\\) taxonomic distances among taxa, \\(x\\) species abundances, \\(n\\) total abundance site. presence/absence data indices reduce index \\(\\Delta^+\\), index Clarke & Warwick (1998) also estimate standard deviation. Clarke & Warwick (2001) presented two new indices: \\(s\\Delta^+\\) product species richness \\(\\Delta^+\\), index variation taxonomic distinctness (\\(\\Lambda^+\\)) defined dis argument must species dissimilarities. must similar dissimilarities produced dist. customary integer steps taxonomic hierarchies, kind dissimilarities can used, phylogenetic trees genetic differences. , dis need taxonomic, species classifications can used. Function taxa2dist can produce suitable dist object classification table. species (basic taxon) corresponds row classification table, columns give classification different levels. varstep = FALSE successive levels separated equal steps, varstep = TRUE step length relative proportional decrease number classes (Clarke & Warwick 1999). check = TRUE, function removes classes distinct species combine species one class, assumes row presents distinct basic taxon. function scales distances longest path length taxa 100 (necessarily check = FALSE). Function plot.taxondive plots \\(\\Delta^+\\) Number species, together expectation approximate 2*sd limits. Function summary.taxondive finds \\(z\\) values significances Normal distribution \\(\\Delta^+\\).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Function returns object class taxondive following items: Species Number species site. D, Dstar, Dplus, SDplus, Lambda \\(\\Delta\\), \\(\\Delta^*\\), \\(\\Delta^+\\), \\(s\\Delta^+\\) \\(\\Lambda^+\\) site. sd.Dplus Standard deviation \\(\\Delta^+\\). ED, EDstar, EDplus Expected values corresponding statistics. Function taxa2dist returns object class \"dist\", attribute \"steps\" step lengths successive levels.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Clarke, K.R & Warwick, R.M. (1998) taxonomic distinctness index statistical properties. Journal Applied Ecology 35, 523--531. Clarke, K.R. & Warwick, R.M. (1999) taxonomic distinctness measure biodiversity: weighting step lengths hierarchical levels. Marine Ecology Progress Series 184: 21--29. Clarke, K.R. & Warwick, R.M. (2001) biodiversity index applicable species lists: variation taxonomic distinctness. Marine Ecology Progress Series 216, 265--278.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"Jari Oksanen","code":""},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"function still preliminary may change. scaling taxonomic dissimilarities influences results. multiply taxonomic distances (step lengths) constant, values Deltas multiplied constant, value \\(\\Lambda^+\\) square constant.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/taxondive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Indices of Taxonomic Diversity and Distinctness — taxondive","text":"","code":"## Preliminary: needs better data and some support functions data(dune) data(dune.taxon) # Taxonomic distances from a classification table with variable step lengths. taxdis <- taxa2dist(dune.taxon, varstep=TRUE) plot(hclust(taxdis), hang = -1) # Indices mod <- taxondive(dune, taxdis) mod #> Species Delta Delta* Lambda+ Delta+ S Delta+ #> 1 5.000 22.736 29.232 900.298 43.364 216.82 #> 2 10.000 51.046 55.988 822.191 56.232 562.32 #> 3 10.000 41.633 46.194 1025.471 62.869 628.69 #> 4 13.000 50.795 55.140 888.244 64.837 842.88 #> 5 14.000 63.498 67.856 715.393 69.211 968.95 #> 6 11.000 70.201 76.361 628.743 73.281 806.09 #> 7 13.000 61.605 66.187 679.337 69.918 908.94 #> 8 12.000 52.544 56.374 756.375 66.729 800.74 #> 9 13.000 50.526 54.108 849.448 63.205 821.67 #> 10 12.000 60.068 64.960 730.736 69.291 831.49 #> 11 9.000 69.589 77.740 404.609 77.803 700.23 #> 12 9.000 62.405 69.795 552.129 74.470 670.23 #> 13 10.000 47.316 53.842 536.429 66.657 666.57 #> 14 7.000 71.383 82.091 239.543 82.013 574.09 #> 15 8.000 68.564 77.097 334.889 79.010 632.08 #> 16 8.000 55.984 64.400 978.014 69.708 557.66 #> 17 7.000 53.913 60.222 632.990 59.286 415.00 #> 18 9.000 73.235 81.865 438.355 76.288 686.59 #> 19 9.000 68.727 76.091 336.364 78.636 707.73 #> 20 8.000 72.343 80.670 444.915 82.078 656.62 #> Expected 65.330 62.560 71.031 summary(mod) #> Delta Delta* Delta+ sd(Delta+) z(Delta+) Pr(>|z|) #> 1 22.7362 29.2322 43.3636 10.0499 -2.7530 0.005905 ** #> 2 51.0458 55.9878 56.2323 5.7727 -2.5636 0.010359 * #> 3 41.6334 46.1936 62.8687 5.7727 -1.4140 0.157360 #> 4 50.7952 55.1396 64.8368 4.5677 -1.3562 0.175047 #> 5 63.4979 67.8564 69.2108 4.2482 -0.4285 0.668251 #> 6 70.2011 76.3614 73.2810 5.3189 0.4229 0.672332 #> 7 61.6049 66.1871 69.9184 4.5677 -0.2437 0.807499 #> 8 52.5437 56.3743 66.7287 4.9215 -0.8743 0.381975 #> 9 50.5258 54.1079 63.2051 4.5677 -1.7134 0.086640 . #> 10 60.0680 64.9597 69.2906 4.9215 -0.3537 0.723567 #> 11 69.5894 77.7396 77.8030 6.3011 1.0747 0.282517 #> 12 62.4049 69.7949 74.4697 6.3011 0.5457 0.585290 #> 13 47.3158 53.8421 66.6566 5.7727 -0.7578 0.448548 #> 14 71.3834 82.0909 82.0130 7.7069 1.4249 0.154186 #> 15 68.5645 77.0970 79.0097 6.9314 1.1510 0.249714 #> 16 55.9840 64.3999 69.7078 6.9314 -0.1909 0.848565 #> 17 53.9134 60.2224 59.2857 7.7069 -1.5240 0.127501 #> 18 73.2349 81.8645 76.2879 6.3011 0.8342 0.404155 #> 19 68.7273 76.0909 78.6364 6.3011 1.2069 0.227457 #> 20 72.3431 80.6704 82.0779 6.9314 1.5937 0.111005 #> Expected 65.3302 62.5603 71.0313 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 plot(mod)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":null,"dir":"Reference","previous_headings":"","what":"Species tolerances and sample heterogeneities — tolerance","title":"Species tolerances and sample heterogeneities — tolerance","text":"Species tolerances sample heterogeneities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"tolerance(x, ...) # S3 method for cca tolerance(x, choices = 1:2, which = c(\"species\",\"sites\"), scaling = \"species\", useN2 = TRUE, hill = FALSE, ...) # S3 method for decorana tolerance(x, data, choices = 1:4, which = c(\"sites\", \"species\"), useN2 = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Species tolerances and sample heterogeneities — tolerance","text":"Function compute species tolerances site heterogeneity measures unimodal ordinations (CCA & CA). Implements Eq 6.47 6.48 Canoco 4.5 Reference Manual (pages 178--179).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Species tolerances and sample heterogeneities — tolerance","text":"Matrix tolerances/heterogeneities additional attributes: , scaling, N2, latter NA useN2 = FALSE N2 estimated.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Species tolerances and sample heterogeneities — tolerance","text":"Gavin L. Simpson Jari Oksanen (decorana method).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Species tolerances and sample heterogeneities — tolerance","text":"x object class \"cca\". choices numeric; ordination axes compute tolerances heterogeneities . Defaults axes 1 2. character; one \"species\" \"sites\", indicating whether species tolerances sample heterogeneities respectively computed. scaling character numeric; ordination scaling use. See scores.cca details. hill logical; scaling character, control whether Hill's scaling used (C)CA respectively. See scores.cca details. useN2 logical; bias tolerances / heterogeneities reduced via scaling Hill's N2? data Original input data used decorana. missing, function tries get data used decorana call. ... arguments passed methods.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tolerance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Species tolerances and sample heterogeneities — tolerance","text":"","code":"data(dune) data(dune.env) mod <- cca(dune ~ ., data = dune.env) ## defaults to species tolerances tolerance(mod) #> #> Species Tolerance #> #> Scaling: 2 #> #> CCA1 CCA2 #> Achimill 0.32968099 0.9241988 #> Agrostol 0.93670069 0.9238455 #> Airaprae 1.04694096 0.5889849 #> Alopgeni 0.72227472 0.3760138 #> Anthodor 1.00596787 0.8338212 #> Bellpere 0.32891011 0.9962790 #> Bromhord 0.27740999 0.6236199 #> Chenalbu 0.00000000 0.0000000 #> Cirsarve 0.00000000 0.0000000 #> Comapalu 0.47185632 0.8029414 #> Eleopalu 0.50344134 0.9384960 #> Elymrepe 0.35119963 0.5642491 #> Empenigr 0.00000000 0.0000000 #> Hyporadi 1.05840696 0.7523003 #> Juncarti 0.78397702 1.0686743 #> Juncbufo 0.69275956 0.6180830 #> Lolipere 0.51006235 0.8278177 #> Planlanc 0.36040676 0.6962294 #> Poaprat 0.58184277 0.9547104 #> Poatriv 0.78695928 0.7433503 #> Ranuflam 0.56576326 1.1725628 #> Rumeacet 0.58715663 0.8751491 #> Sagiproc 0.70922180 1.1153129 #> Salirepe 0.98530179 0.1077917 #> Scorautu 1.04355761 1.0724439 #> Trifprat 0.03045846 0.3651949 #> Trifrepe 1.21543364 0.9115613 #> Vicilath 0.24853962 0.6194084 #> Bracruta 1.03787313 1.0958331 #> Callcusp 0.57882025 1.0418623 #> ## sample heterogeneities for CCA axes 1:6 tolerance(mod, which = \"sites\", choices = 1:6) #> #> Sample Heterogeneity #> #> Scaling: 2 #> #> CCA1 CCA2 CCA3 CCA4 CCA5 CCA6 #> 1 0.2350112 0.8611530 1.7964571 0.4445499 2.4235732 0.5496289 #> 2 0.7100754 0.4136311 0.8151643 0.6311751 1.0467901 0.2514646 #> 3 0.5076492 0.7279717 0.8306874 0.5590739 0.3904998 0.9162012 #> 4 0.5955037 0.6901907 0.7931255 0.4873638 0.3966068 0.8700581 #> 5 0.6001048 0.5614830 1.1481560 0.3569604 0.4423909 1.9420043 #> 6 0.7272637 0.6867342 1.6068628 0.7778498 0.9187843 0.4938865 #> 7 0.6478967 0.4993262 0.7207318 0.3817131 0.4130713 0.7228173 #> 8 0.8563491 0.5498552 0.4217718 0.3370226 0.3013276 0.9535190 #> 9 0.5599722 0.7399384 0.4170304 1.0535541 1.4612437 0.7626183 #> 10 0.5210280 0.5806978 0.5856634 0.4174860 1.8559344 0.8890262 #> 11 0.4489323 0.6016877 0.3317371 1.8780211 1.2965939 2.1953737 #> 12 0.4948094 1.1084494 0.5226746 1.5064446 0.5703077 1.1561020 #> 13 0.6998985 0.8859365 0.4215474 0.8582272 0.5673698 0.5186678 #> 14 1.5925779 0.6747926 0.8927360 1.6798300 0.3480218 0.1575892 #> 15 1.0107648 0.5294221 1.0975629 1.7632888 0.2240900 0.3727240 #> 16 0.8031479 0.6058313 0.4871527 0.4227451 0.5341256 0.6990815 #> 17 0.5936276 1.5142792 0.5137979 1.0224938 1.7931775 0.6261853 #> 18 0.5689409 1.4067575 0.6398557 0.4983399 0.4364791 0.6590394 #> 19 1.1330387 0.9816332 1.1242398 0.7238920 0.5577662 0.7036044 #> 20 0.6737757 1.4458326 1.4380928 1.0959027 0.4142423 0.5332460 #> ## average should be 1 with scaling = \"sites\", hill = TRUE tol <- tolerance(mod, which = \"sites\", scaling = \"sites\", hill = TRUE, choices = 1:4) colMeans(tol) #> CCA1 CCA2 CCA3 CCA4 #> 1.059199 1.048823 1.000551 1.077612 apply(tol, 2, sd) #> CCA1 CCA2 CCA3 CCA4 #> 0.3174462 0.2793521 0.3714540 0.2681931 ## Rescaling tries to set all tolerances to 1 tol <- tolerance(decorana(dune)) colMeans(tol) #> DCA1 DCA2 DCA3 DCA4 #> 0.9817657 0.9249544 0.9444811 0.9821666 apply(tol, 2, sd) #> DCA1 DCA2 DCA3 DCA4 #> 0.1977777 0.3204058 0.2646872 0.1210543"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":null,"dir":"Reference","previous_headings":"","what":"Functional Diversity and Community Distances from Species Trees — treedive","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Functional diversity defined total branch length trait dendrogram connecting species, excluding unnecessary root segments tree (Petchey Gaston 2006). Tree distance increase total branch length combining two sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"treedive(comm, tree, match.force = TRUE, verbose = TRUE) treeheight(tree) treedist(x, tree, relative = TRUE, match.force = TRUE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"comm, x Community data frame matrix. tree dendrogram treedive must species (columns). match.force Force matching column names data (comm, x) labels tree. FALSE, matching happens dimensions differ (warning message). order data must match order tree matching names done. verbose Print diagnostic messages warnings. relative Use distances relative height combined tree. ... arguments passed functions (ignored).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Function treeheight finds sum lengths connecting segments dendrogram produced hclust, dendrogram can coerced correct type using .hclust. applied clustering species traits, measure functional diversity (Petchey Gaston 2002, 2006), applied phylogenetic trees phylogenetic diversity. Function treedive finds treeheight site (row) community matrix. function uses subset dendrogram species occur site, excludes tree root needed connect species (Petchey Gaston 2006). subset dendrogram found first calculating cophenetic distances input dendrogram, reconstructing dendrogram subset cophenetic distance matrix species occurring site. Diversity 0 one species, NA empty communities. Function treedist finds dissimilarities among trees. Pairwise dissimilarity two trees found combining species common tree seeing much tree height shared much unique. relative = FALSE dissimilarity defined \\(2 (\\cup B) - - B\\), \\(\\) \\(B\\) heights component trees \\(\\cup B\\) height combined tree. relative = TRUE dissimilarity \\((2(\\cup B)--B)/(\\cup B)\\). Although latter formula similar Jaccard dissimilarity (see vegdist, designdist), range \\(0 \\ldots 1\\), since combined tree can add new root. two zero-height trees combined tree zero height, relative index attains maximum value \\(2\\). dissimilarity zero combined zero-height tree. functions need dendrogram species traits phylogenies input. species traits contain factor ordered factor variables, recommended use Gower distances mixed data (function daisy package cluster), usually recommended clustering method UPGMA (method = \"average\" function hclust) (Podani Schmera 2006). Phylogenetic trees can changed dendrograms using function .hclust.phylo ape package. possible analyse non-randomness tree diversity using oecosimu. needs specifying adequate Null model, results change choice.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"vector diversity values single tree height, dissimilarity structure inherits dist can used similarly.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Lozupone, C. Knight, R. 2005. UniFrac: new phylogenetic method comparing microbial communities. Applied Environmental Microbiology 71, 8228--8235. Petchey, O.L. Gaston, K.J. 2002. Functional diversity (FD), species richness community composition. Ecology Letters 5, 402--411. Petchey, O.L. Gaston, K.J. 2006. Functional diversity: back basics looking forward. Ecology Letters 9, 741--758. Podani J. Schmera, D. 2006. dendrogram-based methods functional diversity. Oikos 115, 179--185.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/treedive.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Functional Diversity and Community Distances from Species Trees — treedive","text":"","code":"## There is no data set on species properties yet, and we demonstrate ## the methods using phylogenetic trees data(dune) data(dune.phylodis) cl <- hclust(dune.phylodis) treedive(dune, cl) #> forced matching of 'tree' labels and 'comm' names #> 1 2 3 4 5 6 7 8 #> 384.0913 568.8791 1172.9455 1327.9317 1426.9067 1391.1628 1479.5062 1523.0792 #> 9 10 11 12 13 14 15 16 #> 1460.0423 1316.4832 1366.9960 1423.5582 895.1120 1457.2705 1505.9501 1187.5165 #> 17 18 19 20 #> 517.6920 1394.5162 1470.4671 1439.5571 ## Significance test using Null model communities. ## The current choice fixes numbers of species and picks species ## proportionally to their overall frequency oecosimu(dune, treedive, \"r1\", tree = cl, verbose = FALSE) #> Warning: nullmodel transformed 'comm' to binary data #> oecosimu object #> #> Call: oecosimu(comm = dune, nestfun = treedive, method = \"r1\", tree = #> cl, verbose = FALSE) #> #> nullmodel method ‘r1’ with 99 simulations #> #> alternative hypothesis: statistic is less or greater than simulated values #> #> statistic SES mean 2.5% 50% 97.5% Pr(sim.) #> 1 384.09 -1.238698 773.72 383.47 628.63 1237.7 0.11 #> 2 568.88 -2.353877 1222.61 663.93 1337.50 1561.4 0.01 ** #> 3 1172.95 -0.136353 1210.76 679.53 1327.94 1581.5 0.63 #> 4 1327.93 -0.418624 1427.25 886.63 1505.15 1733.1 0.45 #> 5 1426.91 -0.322462 1496.97 926.98 1559.25 1734.4 0.43 #> 6 1391.16 0.254066 1326.70 753.19 1408.30 1634.7 0.87 #> 7 1479.51 0.280433 1407.90 880.79 1476.23 1735.1 0.99 #> 8 1523.08 0.567886 1389.63 792.74 1428.01 1655.3 0.65 #> 9 1460.04 0.065245 1448.21 899.67 1472.92 1654.4 0.87 #> 10 1316.48 -0.192836 1362.43 708.60 1441.63 1608.0 0.47 #> 11 1367.00 0.768160 1150.05 621.65 1266.73 1486.4 0.43 #> 12 1423.56 1.101039 1088.13 622.24 1208.95 1474.2 0.25 #> 13 895.11 -0.972053 1188.10 656.34 1325.53 1558.5 0.61 #> 14 1457.27 1.477151 980.91 491.11 1101.65 1459.3 0.07 . #> 15 1505.95 1.532058 1052.35 575.91 1171.88 1460.7 0.03 * #> 16 1187.52 0.306608 1089.58 543.75 1238.26 1490.7 0.89 #> 17 517.69 -1.446211 959.50 504.26 1101.16 1367.8 0.09 . #> 18 1394.52 0.961778 1099.50 618.22 1249.49 1493.8 0.33 #> 19 1470.47 1.134795 1121.18 564.70 1240.26 1562.8 0.15 #> 20 1439.56 1.160566 1104.59 629.08 1244.20 1467.4 0.17 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Phylogenetically ordered community table dtree <- treedist(dune, cl) tabasco(dune, hclust(dtree), cl) ## Use tree distances in distance-based RDA dbrda(dtree ~ 1) #> Call: dbrda(formula = dtree ~ 1) #> #> Inertia Rank RealDims #> Total 2.183 #> Unconstrained 2.183 19 10 #> Inertia is squared Treedist distance #> #> Eigenvalues for unconstrained axes: #> MDS1 MDS2 MDS3 MDS4 MDS5 MDS6 MDS7 MDS8 #> 1.1971 0.4546 0.2967 0.1346 0.1067 0.0912 0.0391 0.0190 #> (Showing 8 of 19 unconstrained eigenvalues) #>"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":null,"dir":"Reference","previous_headings":"","what":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis find Tsallis diversities scale corresponding evenness measures. Function tsallisaccum finds statistics accumulating sites.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"tsallis(x, scales = seq(0, 2, 0.2), norm = FALSE, hill = FALSE) tsallisaccum(x, scales = seq(0, 2, 0.2), permutations = 100, raw = FALSE, subset, ...) # S3 method for tsallisaccum persp(x, theta = 220, phi = 15, col = heat.colors(100), zlim, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"x Community data matrix plotting object. scales Scales Tsallis diversity. norm Logical, TRUE diversity values normalized maximum (diversity value equiprobability conditions). hill Calculate Hill numbers. permutations Usually integer giving number permutations, can also list control values permutations returned function , permutation matrix row gives permuted indices. raw FALSE return summary statistics permutations, TRUE returns individual permutations. subset logical expression indicating sites (rows) keep: missing values taken FALSE. theta, phi angles defining viewing direction. theta gives azimuthal direction phi colatitude. col Colours used surface. zlim Limits vertical axis. ... arguments passed tsallis graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis diversity (also equivalent Patil Taillie diversity) one-parametric generalised entropy function, defined : $$H_q = \\frac{1}{q-1} (1-\\sum_{=1}^S p_i^q)$$ \\(q\\) scale parameter, \\(S\\) number species sample (Tsallis 1988, Tothmeresz 1995). diversity concave \\(q>0\\), non-additive (Keylock 2005). \\(q=0\\) gives number species minus one, \\(q\\) tends 1 gives Shannon diversity, \\(q=2\\) gives Simpson index (see function diversity). norm = TRUE, tsallis gives values normalized maximum: $$H_q(max) = \\frac{S^{1-q}-1}{1-q}$$ \\(S\\) number species. \\(q\\) tends 1, maximum defined \\(ln(S)\\). hill = TRUE, tsallis gives Hill numbers (numbers equivalents, see Jost 2007): $$D_q = (1-(q-1) H)^{1/(1-q)}$$ Details plotting methods accumulating values can found help pages functions renyi renyiaccum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Function tsallis returns data frame selected indices. Function tsallisaccum argument raw = FALSE returns three-dimensional array, first dimension accumulated sites, second dimension diversity scales, third dimension summary statistics mean, stdev, min, max, Qnt 0.025 Qnt 0.975. argument raw = TRUE statistics third dimension replaced individual permutation results.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Tsallis, C. (1988) Possible generalization Boltzmann-Gibbs statistics. J. Stat. Phis. 52, 479--487. Tothmeresz, B. (1995) Comparison different methods diversity ordering. Journal Vegetation Science 6, 283--290. Patil, G. P. Taillie, C. (1982) Diversity concept measurement. J. . Stat. Ass. 77, 548--567. Keylock, C. J. (2005) Simpson diversity Shannon-Wiener index special cases generalized entropy. Oikos 109, 203--207. Jost, L (2007) Partitioning diversity independent alpha beta components. Ecology 88, 2427--2439.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"Péter Sólymos, solymos@ualberta.ca, based code Roeland Kindt Jari Oksanen written renyi","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/tsallis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Tsallis Diversity and Corresponding Accumulation Curves — tsallis","text":"","code":"data(BCI) i <- sample(nrow(BCI), 12) x1 <- tsallis(BCI[i,]) x1 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 #> 8 87 39.98200 19.77567 10.60391 6.181934 3.908381 2.659907 1.928610 1.473408 #> 10 93 41.50369 20.10059 10.63988 6.164157 3.889803 2.648176 1.922207 1.470121 #> 5 100 44.20598 21.17686 11.08126 6.350040 3.969940 2.683412 1.937940 1.477217 #> 17 92 40.88788 19.62643 10.29183 5.926929 3.736897 2.553271 1.864739 1.435868 #> 19 108 47.27018 22.32602 11.49839 6.492802 4.013094 2.692466 1.936696 1.473828 #> 13 92 41.98171 20.60707 10.96297 6.342501 3.982373 2.694847 1.945425 1.481616 #> 46 85 38.71044 19.06049 10.22455 5.987209 3.810489 2.611431 1.904913 1.461976 #> 16 92 41.45601 20.19256 10.71573 6.210870 3.916821 2.663870 1.931567 1.475861 #> 36 91 40.76595 19.77916 10.48193 6.081803 3.846109 2.625076 1.910150 1.463936 #> 7 81 37.47840 18.73414 10.17269 6.004908 3.836811 2.631757 1.918091 1.469869 #> 24 94 42.55654 20.75558 10.99284 6.343926 3.979427 2.692702 1.944478 1.481404 #> 22 90 40.16368 19.41416 10.25630 5.939677 3.755413 2.566788 1.872586 1.439722 #> 1.8 2 #> 8 1.173986 0.9671998 #> 10 1.172341 0.9663808 #> 5 1.175553 0.9678267 #> 17 1.152122 0.9545126 #> 19 1.172481 0.9655820 #> 13 1.178033 0.9692075 #> 46 1.168556 0.9646728 #> 16 1.175935 0.9686598 #> 36 1.169232 0.9648567 #> 7 1.173083 0.9672014 #> 24 1.178152 0.9694268 #> 22 1.153640 0.9548316 diversity(BCI[i,],\"simpson\") == x1[[\"2\"]] #> 8 10 5 17 19 13 46 16 36 7 24 22 #> TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE plot(x1) x2 <- tsallis(BCI[i,],norm=TRUE) x2 #> 0 0.2 0.4 0.6 0.8 1 1.2 1.4 #> 8 1 0.9154260 0.8674322 0.8491531 0.8535754 0.8729254 0.8992501 0.9258854 #> 10 1 0.9001004 0.8450759 0.8255446 0.8324355 0.8561634 0.8872568 0.9180261 #> 5 1 0.9038026 0.8502581 0.8308757 0.8372445 0.8602030 0.8904867 0.9204832 #> 17 1 0.8945733 0.8308281 0.8026305 0.8032747 0.8244489 0.8566981 0.8913214 #> 19 1 0.9079031 0.8537683 0.8315799 0.8347890 0.8554245 0.8846675 0.9147428 #> 13 1 0.9185048 0.8723405 0.8549707 0.8595970 0.8786069 0.9042012 0.9298883 #> 46 1 0.9032439 0.8485357 0.8278693 0.8331492 0.8554539 0.8856795 0.9162035 #> 16 1 0.9070033 0.8547937 0.8356888 0.8417570 0.8641446 0.8938075 0.9232641 #> 36 1 0.8998669 0.8431295 0.8216927 0.8272622 0.8505725 0.8820846 0.9137994 #> 7 1 0.9094899 0.8600339 0.8427936 0.8492653 0.8706729 0.8985570 0.9261475 #> 24 1 0.9149445 0.8667059 0.8486326 0.8536824 0.8738548 0.9008891 0.9279024 #> 22 1 0.8945746 0.8334045 0.8082215 0.8109045 0.8325271 0.8637842 0.8965990 #> 1.6 1.8 2 #> 8 0.9486740 0.9660687 0.9783170 #> 10 0.9438798 0.9632975 0.9767719 #> 5 0.9456409 0.9644767 0.9775050 #> 17 0.9223038 0.9469046 0.9648877 #> 19 0.9406572 0.9605041 0.9745226 #> 13 0.9516896 0.9681999 0.9797424 #> 46 0.9422698 0.9621112 0.9760219 #> 16 0.9479928 0.9664763 0.9791887 #> 36 0.9407648 0.9611954 0.9754595 #> 7 0.9493990 0.9669335 0.9791421 #> 24 0.9507028 0.9678509 0.9797399 #> 22 0.9256369 0.9486075 0.9654409 plot(x2) mod1 <- tsallisaccum(BCI[i,]) plot(mod1, as.table=TRUE, col = c(1, 2, 2)) persp(mod1) mod2 <- tsallisaccum(BCI[i,], norm=TRUE) persp(mod2,theta=100,phi=30)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":null,"dir":"Reference","previous_headings":"","what":"Vegetation and environment in lichen pastures — varespec","title":"Vegetation and environment in lichen pastures — varespec","text":"varespec data frame 24 rows 44 columns. Columns estimated cover values 44 species. variable names formed scientific names, self explanatory anybody familiar vegetation type. varechem data frame 24 rows 14 columns, giving soil characteristics sites varespec data frame. chemical measurements obvious names. Baresoil gives estimated cover bare soil, Humdepth thickness humus layer.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varechem) data(varespec)"},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Vegetation and environment in lichen pastures — varespec","text":"Väre, H., Ohtonen, R. Oksanen, J. (1995) Effects reindeer grazing understorey vegetation dry Pinus sylvestris forests. Journal Vegetation Science 6, 523--530.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varechem.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vegetation and environment in lichen pastures — varespec","text":"","code":"data(varespec) data(varechem)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":null,"dir":"Reference","previous_headings":"","what":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"function partitions variation community data community dissimilarities respect two, three, four explanatory tables, using adjusted \\(R^2\\) redundancy analysis ordination (RDA) distance-based redundancy analysis. response single vector, partitioning partial regression. Collinear variables explanatory tables removed prior partitioning.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"varpart(Y, X, ..., data, chisquare = FALSE, transfo, scale = FALSE, add = FALSE, sqrt.dist = FALSE, permutations) # S3 method for varpart summary(object, ...) showvarparts(parts, labels, bg = NULL, alpha = 63, Xnames, id.size = 1.2, ...) # S3 method for varpart234 plot(x, cutoff = 0, digits = 1, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Y Data frame matrix containing response data table dissimilarity structure inheriting dist. community ecology, table often site--species table dissimilarity object. X Two four explanatory models, variables tables. can defined three alternative ways: (1) one-sided model formulae beginning ~ defining model, (2) name single numeric factor variable, (3) name matrix numeric data frame numeric factor variables. model formulae can factors, interaction terms transformations variables. names variables model formula found data frame given data argument, found , user environment. Single variables, data frames matrices found user environment. entries till next argument (data transfo) interpreted explanatory models, names extra arguments abbreviated omitted. ... parameters passed functions. NB, arguments dots abbreviated must spelt completely. data data frame variables used formulae X. chisquare Partition Chi-square inertia Correspondence Analysis (cca). transfo Transformation Y (community data) using decostand. alternatives decostand can used, preserving Euclidean metric include \"hellinger\", \"chi.square\", \"total\", \"norm\". Ignored Y dissimilarities. scale columns Y standardized unit variance. Ignored Y dissimilarities. add Add constant non-diagonal values euclidify dissimilarities (see wcmdscale details). Choice \"lingoes\" (TRUE) use recommended method Legendre & Anderson (1999: “method 1”) \"cailliez\" uses “method 2”. argument effect Y dissimilarities. sqrt.dist Take square root dissimilarities. often euclidifies dissimilarities. NB., argument name abbreviated. argument effect Y dissimilarities. permutations chisquare = TRUE, adjusted \\(R^2\\) estimated permutations, paramater can list control values permutations returned function , number permutations required, permutation matrix row gives permuted indices. parts Number explanatory tables (circles) displayed. labels Labels used displayed fractions. Default use letters printed output. bg Fill colours circles ellipses. alpha Transparency fill colour. argument takes precedence possible transparency definitions colour. value must range \\(0...255\\), low values transparent. Transparency available graphics devices file formats. Xnames Names sources variation. Default names X1, X2, X3 X4. Xnames=NA, Xnames=NULL Xnames=\"\" produce names. names can changed names. often best use short names. id.size numerical value giving character expansion factor names circles ellipses. x, object varpart result. cutoff values cutoff displayed. digits number significant digits; number decimal places least one higher.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"functions partition variation Y components accounted two four explanatory tables combined effects. Y multicolumn data frame matrix, partitioning based redundancy analysis (RDA, see rda) constrained correspondence analysis chisquare = TRUE (CCA, see cca). Y single variable, partitioning based linear regression. Y dissimilarities, decomposition based distance-based redundancy analysis (db-RDA, see dbrda) following McArdle & Anderson (2001). input dissimilarities must compatible results dist. Vegan functions vegdist, designdist, raupcrick betadiver produce objects, many dissimilarity functions R packages. Partitioning made squared dissimilarities analogously using variance rectangular data -- unless sqrt.dist = TRUE specified. function primarily uses adjusted \\(R^2\\) assess partitions explained explanatory tables combinations (see RsquareAdj), unbiased method (Peres-Neto et al., 2006). raw \\(R^2\\) basic fractions also displayed, biased estimates variation explained explanatory table. correspondence analysis (chisquare = TRUE), adjusted \\(R^2\\) found permutation vary repeated analyses. identifiable fractions designated lower case alphabets. meaning symbols can found separate document (use browseVignettes(\"vegan\")), can displayed graphically using function showvarparts. fraction testable can directly expressed RDA db-RDA model. cases printed output also displays corresponding RDA model using notation explanatory tables | conditions (partialled ; see rda details). Although single fractions can testable, mean fractions simultaneously can tested, since number testable fractions higher number estimated models. non-testable components found differences testable components. testable components permutation variance correspondence analysis (chisquare = TRUE), non-testable components even higher variance. abridged explanation alphabetic symbols individual fractions follows, computational details checked vignette (readable browseVignettes(\"vegan\")) source code. two explanatory tables, fractions explained uniquely two tables [] [b], joint effect [c]. three explanatory tables, fractions explained uniquely three tables [] [c], joint fractions two tables [d] [f], joint fraction three tables [g]. four explanatory tables, fractions explained uniquely four tables [] [d], joint fractions two tables [e] [j], joint fractions three variables [k] [n], joint fraction four tables [o]. summary give overview unique overall contribution group variables. overall contribution (labelled “Contributed”) consists unique contribution variable equal shares fraction variable contributes. summary tabulates fraction divided variables, contributed component sum divided fractions. summary based idea Lai et al. (2022), similar output rdacca.hp package. plot function displays Venn diagram labels intersection (individual fraction) adjusted R squared higher cutoff. helper function showvarpart displays fraction labels. circles ellipses labelled short default names names defined user argument Xnames. Longer explanatory file names can written varpart output plot follows: use option Xnames=NA, add new names using text function. bit fiddling coordinates (see locator) character size allow users place names reasonably short lengths varpart plot.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Function varpart returns object class \"varpart\" items scale transfo (can missing) hold information standardizations, tables contains names explanatory tables, call function call. function varpart calls function varpart2, varpart3 varpart4 return object class \"varpart234\" saves result item part. items object : SS.Y Sum squares matrix Y. n Number observations (rows). nsets Number explanatory tables bigwarning Warnings collinearity. fract Basic fractions estimated constrained models. indfract Individual fractions possible subsections Venn diagram (see showvarparts). contr1 Fractions can found conditioning single explanatory table models three four explanatory tables. contr2 Fractions can found conditioning two explanatory tables models four explanatory tables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"fraction-data-frames","dir":"Reference","previous_headings":"","what":"Fraction Data Frames","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Items fract, indfract, contr1 contr2 data frames items: Df: Degrees freedom numerator \\(F\\)-statistic fraction. R.square: Raw \\(R^2\\). calculated fract NA items. Adj.R.square: Adjusted \\(R^2\\). Testable: fraction can expressed (partial) RDA model, directly Testable, field TRUE. case fraction label also gives specification testable RDA model.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"() References variation partitioning Borcard, D., P. Legendre & P. Drapeau. 1992. Partialling spatial component ecological variation. Ecology 73: 1045--1055. Lai J., Y. Zou, J. Zhang & P. Peres-Neto. 2022. Generalizing hierarchical variation partitioning multiple regression canonical analysis using rdacca.hp R package. Methods Ecology Evolution, 13: 782--788. Legendre, P. & L. Legendre. 2012. Numerical ecology, 3rd English edition. Elsevier Science BV, Amsterdam. (b) Reference transformations species data Legendre, P. E. D. Gallagher. 2001. Ecologically meaningful transformations ordination species data. Oecologia 129: 271--280. (c) Reference adjustment bimultivariate redundancy statistic Peres-Neto, P., P. Legendre, S. Dray D. Borcard. 2006. Variation partitioning species data matrices: estimation comparison fractions. Ecology 87: 2614--2625. (d) References partitioning dissimilarities Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecological Monographs 69, 1--24. McArdle, B.H. & Anderson, M.J. (2001). Fitting multivariate models community data: comment distance-based redundancy analysis. Ecology 82, 290-297.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"Pierre Legendre, Departement de Sciences Biologiques, Universite de Montreal, Canada. developed Jari Oksanen.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"can use command browseVignettes(\"vegan\") display document presents Venn diagrams showing fraction names partitioning variation Y respect 2, 3, 4 tables explanatory variables, well equations used variation partitioning. functions frequently give negative estimates variation. Adjusted \\(R^2\\) can negative fraction; unadjusted \\(R^2\\) testable fractions variances non-negative. Non-testable fractions found directly, subtracting different models, subtraction results can negative. fractions orthogonal, linearly independent, complicated nonlinear dependencies can cause negative non-testable fractions. fraction can negative non-Euclidean dissimilarities underlying db-RDA model can yield negative eigenvalues (see dbrda). negative eigenvalues underlying analysis can avoided arguments sqrt.dist add similar effect dbrda: square roots several dissimilarities negative eigenvalues, negative eigenvalues produced Lingoes Cailliez adjustment, effect add random variation dissimilarities. simplified, fast version RDA, CCA adn dbRDA used (functions simpleRDA2, simpleCCA simpleDBRDA). actual calculations done functions varpart2 varpart4, intended called directly user.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/varpart.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Partition the Variation of Community Matrix by 2, 3, or 4 Explanatory Matrices — varpart","text":"","code":"data(mite) data(mite.env) data(mite.pcnm) # Two explanatory data frames -- Hellinger-transform Y mod <- varpart(mite, mite.env, mite.pcnm, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = mite.env, mite.pcnm, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.52650 0.43670 TRUE #> [b+c] = X2 22 0.62300 0.44653 TRUE #> [a+b+c] = X1+X2 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09141 TRUE #> [b] = X2|X1 22 0.10124 TRUE #> [c] 0 0.34530 FALSE #> [d] = Residuals 0.46206 FALSE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0914 0.264 mite.env #> X2 0.1012 0.274 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 #> [a] 0.0914 #> [b] 0.1012 #> [c] 0.1726 0.1726 ## Use fill colours showvarparts(2, bg = c(\"hotpink\",\"skyblue\")) plot(mod, bg = c(\"hotpink\",\"skyblue\")) ## Test fraction [a] using partial RDA, '~ .' in formula tells to use ## all variables of data mite.env. aFrac <- rda(decostand(mite, \"hel\"), mite.env, mite.pcnm) anova(aFrac) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(X = decostand(mite, \"hel\"), Y = mite.env, Z = mite.pcnm) #> Df Variance F Pr(>F) #> Model 11 0.053592 1.8453 0.001 *** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## RsquareAdj gives the same result as component [a] of varpart RsquareAdj(aFrac) #> $r.squared #> [1] 0.1359251 #> #> $adj.r.squared #> [1] 0.09140797 #> ## Partition Bray-Curtis dissimilarities varpart(vegdist(mite), mite.env, mite.pcnm) #> #> Partition of squared Bray distance in dbRDA #> #> Call: varpart(Y = vegdist(mite), X = mite.env, mite.pcnm) #> #> Explanatory tables: #> X1: mite.env #> X2: mite.pcnm #> #> No. of explanatory tables: 2 #> Total variation (SS): 14.696 #> No. of observations: 70 #> #> Partition table: #> Df R.squared Adj.R.squared Testable #> [a+c] = X1 11 0.50512 0.41127 TRUE #> [b+c] = X2 22 0.60144 0.41489 TRUE #> [a+b+c] = X1+X2 33 0.74631 0.51375 TRUE #> Individual fractions #> [a] = X1|X2 11 0.09887 TRUE #> [b] = X2|X1 22 0.10249 TRUE #> [c] 0 0.31240 FALSE #> [d] = Residuals 0.48625 FALSE #> --- #> Use function ‘dbrda’ to test significance of fractions of interest ## Three explanatory tables with formula interface mod <- varpart(mite, ~ SubsDens + WatrCont, ~ Substrate + Shrub + Topo, mite.pcnm, data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm, data = mite.env, transfo = \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm #> #> No. of explanatory tables: 3 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [a+d+f+g] = X1 2 0.32677 0.30667 TRUE #> [b+d+e+g] = X2 9 0.40395 0.31454 TRUE #> [c+e+f+g] = X3 22 0.62300 0.44653 TRUE #> [a+b+d+e+f+g] = X1+X2 11 0.52650 0.43670 TRUE #> [a+c+d+e+f+g] = X1+X3 24 0.67372 0.49970 TRUE #> [b+c+d+e+f+g] = X2+X3 31 0.72400 0.49884 TRUE #> [a+b+c+d+e+f+g] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3 2 0.03910 TRUE #> [b] = X2 | X1+X3 9 0.03824 TRUE #> [c] = X3 | X1+X2 22 0.10124 TRUE #> [d] 0 0.01407 FALSE #> [e] 0 0.09179 FALSE #> [f] 0 0.08306 FALSE #> [g] 0 0.17045 FALSE #> [h] = Residuals 0.46206 FALSE #> Controlling 1 table X #> [a+d] = X1 | X3 2 0.05317 TRUE #> [a+f] = X1 | X2 2 0.12216 TRUE #> [b+d] = X2 | X3 9 0.05231 TRUE #> [b+e] = X2 | X1 9 0.13003 TRUE #> [c+e] = X3 | X1 22 0.19303 TRUE #> [c+f] = X3 | X2 22 0.18429 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.144 ~SubsDens + WatrCont #> X2 0.0382 0.148 ~Substrate + Shrub + Topo #> X3 0.1012 0.245 mite.pcnm #> #> Contributions of fractions to sets: #> #> X1 X2 X3 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10124 #> [d] 0.00703 0.00703 #> [e] 0.04590 0.04590 #> [f] 0.04153 0.04153 #> [g] 0.05682 0.05682 0.05682 showvarparts(3, bg=2:4) plot(mod, bg=2:4) ## Use RDA to test fraction [a] ## Matrix can be an argument in formula rda.result <- rda(decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) anova(rda.result) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = decostand(mite, \"hell\") ~ SubsDens + WatrCont + Condition(Substrate + Shrub + Topo) + Condition(as.matrix(mite.pcnm)), data = mite.env) #> Df Variance F Pr(>F) #> Model 2 0.013771 2.6079 0.005 ** #> Residual 36 0.095050 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Four explanatory tables mod <- varpart(mite, ~ SubsDens + WatrCont, ~Substrate + Shrub + Topo, mite.pcnm[,1:11], mite.pcnm[,12:22], data=mite.env, transfo=\"hel\") mod #> #> Partition of variance in RDA #> #> Call: varpart(Y = mite, X = ~SubsDens + WatrCont, ~Substrate + Shrub + #> Topo, mite.pcnm[, 1:11], mite.pcnm[, 12:22], data = mite.env, transfo = #> \"hel\") #> Species transformation: hellinger #> Explanatory tables: #> X1: ~SubsDens + WatrCont #> X2: ~Substrate + Shrub + Topo #> X3: mite.pcnm[, 1:11] #> X4: mite.pcnm[, 12:22] #> #> No. of explanatory tables: 4 #> Total variation (SS): 27.205 #> Variance: 0.39428 #> No. of observations: 70 #> #> Partition table: #> Df R.square Adj.R.square Testable #> [aeghklno] = X1 2 0.32677 0.30667 TRUE #> [befiklmo] = X2 9 0.40395 0.31454 TRUE #> [cfgjlmno] = X3 11 0.53231 0.44361 TRUE #> [dhijkmno] = X4 11 0.09069 -0.08176 TRUE #> [abefghiklmno] = X1+X2 11 0.52650 0.43670 TRUE #> [acefghjklmno] = X1+X3 13 0.59150 0.49667 TRUE #> [adeghijklmno] = X1+X4 13 0.40374 0.26533 TRUE #> [bcefgijklmno] = X2+X3 20 0.63650 0.48813 TRUE #> [bdefhijklmno] = X2+X4 20 0.53338 0.34292 TRUE #> [cdfghijklmno] = X3+X4 22 0.62300 0.44653 TRUE #> [abcefghijklmno] = X1+X2+X3 22 0.67947 0.52944 TRUE #> [abdefghijklmno] = X1+X2+X4 22 0.61553 0.43557 TRUE #> [acdefghijklmno] = X1+X3+X4 24 0.67372 0.49970 TRUE #> [bcdefghijklmno] = X2+X3+X4 31 0.72400 0.49884 TRUE #> [abcdefghijklmno] = All 33 0.75893 0.53794 TRUE #> Individual fractions #> [a] = X1 | X2+X3+X4 2 0.03910 TRUE #> [b] = X2 | X1+X3+X4 9 0.03824 TRUE #> [c] = X3 | X1+X2+X4 11 0.10237 TRUE #> [d] = X4 | X1+X2+X3 11 0.00850 TRUE #> [e] 0 0.01407 FALSE #> [f] 0 0.13200 FALSE #> [g] 0 0.05355 FALSE #> [h] 0 0.00220 FALSE #> [i] 0 -0.00547 FALSE #> [j] 0 -0.00963 FALSE #> [k] 0 -0.00231 FALSE #> [l] 0 0.24037 FALSE #> [m] 0 -0.03474 FALSE #> [n] 0 0.02730 FALSE #> [o] 0 -0.06761 FALSE #> [p] = Residuals 0 0.46206 FALSE #> Controlling 2 tables X #> [ae] = X1 | X3+X4 2 0.05317 TRUE #> [ag] = X1 | X2+X4 2 0.09265 TRUE #> [ah] = X1 | X2+X3 2 0.04131 TRUE #> [be] = X2 | X3+X4 9 0.05231 TRUE #> [bf] = X2 | X1+X4 9 0.17024 TRUE #> [bi] = X2 | X1+X3 9 0.03277 TRUE #> [cf] = X3 | X1+X4 11 0.23437 TRUE #> [cg] = X3 | X2+X4 11 0.15592 TRUE #> [cj] = X3 | X1+X2 11 0.09274 TRUE #> [dh] = X4 | X2+X3 11 0.01071 TRUE #> [di] = X4 | X1+X3 11 0.00303 TRUE #> [dj] = X4 | X1+X2 11 -0.00113 TRUE #> Controlling 1 table X #> [aghn] = X1 | X2 2 0.12216 TRUE #> [aehk] = X1 | X3 2 0.05306 TRUE #> [aegl] = X1 | X4 2 0.34709 TRUE #> [bfim] = X2 | X1 9 0.13003 TRUE #> [beik] = X2 | X3 9 0.04452 TRUE #> [befl] = X2 | X4 9 0.42468 TRUE #> [cfjm] = X3 | X1 11 0.19000 TRUE #> [cgjn] = X3 | X2 11 0.17359 TRUE #> [cfgl] = X3 | X4 11 0.52830 TRUE #> [dijm] = X4 | X1 11 -0.04134 TRUE #> [dhjn] = X4 | X2 11 0.02837 TRUE #> [dhik] = X4 | X3 11 0.00292 TRUE #> --- #> Use function ‘rda’ to test significance of fractions of interest summary(mod) #> #> Unique fractions and total with shared fractions equally allocated: #> #> Unique Contributed Component #> X1 0.0391 0.1456 ~SubsDens + WatrCont #> X2 0.0382 0.1594 ~Substrate + Shrub + Topo #> X3 0.1024 0.2511 mite.pcnm[, 1:11] #> X4 0.0085 -0.0181 mite.pcnm[, 12:22] #> #> Contributions of fractions to sets: #> #> X1 X2 X3 X4 #> [a] 0.03910 #> [b] 0.03824 #> [c] 0.10237 #> [d] 0.00850 #> [e] 0.00703 0.00703 #> [f] 0.06600 0.06600 #> [g] 0.02678 0.02678 #> [h] 0.00110 0.00110 #> [i] -0.00274 -0.00274 #> [j] -0.00482 -0.00482 #> [k] -0.00077 -0.00077 -0.00077 #> [l] 0.08012 0.08012 0.08012 #> [m] -0.01158 -0.01158 -0.01158 #> [n] 0.00910 0.00910 0.00910 #> [o] -0.01690 -0.01690 -0.01690 -0.01690 plot(mod, bg=2:5) ## Show values for all partitions by putting 'cutoff' low enough: plot(mod, cutoff = -Inf, cex = 0.7, bg=2:5)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":null,"dir":"Reference","previous_headings":"","what":"Defunct Functions in Package vegan — vegan-defunct","title":"Defunct Functions in Package vegan — vegan-defunct","text":"functions variables listed longer part vegan longer needed.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Defunct Functions in Package vegan — vegan-defunct","text":"","code":"## defunct in vegan 2.7-0 (development 2.6-7) adonis(formula, data, permutations = 999, method = \"bray\", strata = NULL, contr.unordered = \"contr.sum\", contr.ordered = \"contr.poly\", parallel = getOption(\"mc.cores\"), ...) ## defunct in vegan 2.6-0 as.mlm(x) humpfit(mass, spno, family = poisson, start) vegandocs(doc = c(\"NEWS\", \"ONEWS\", \"FAQ-vegan\", \"intro-vegan\", \"diversity-vegan\", \"decision-vegan\", \"partitioning\", \"permutations\")) ## defunct in vegan 2.5-0 commsimulator(x, method, thin=1) ## defunct in vegan 2.4-0 # S3 method for adonis density(x, ...) # S3 method for vegandensity plot(x, main = NULL, xlab = NULL, ylab = \"Density\", type = \"l\", zero.line = TRUE, obs.line = TRUE, ...) # S3 method for adonis densityplot(x, data, xlab = \"Null\", ...) ## defunct in vegan 2.2-0 metaMDSrotate(object, vec, na.rm = FALSE, ...) ## defunct in vegan 2.0-0 getNumObs(object, ...) permuted.index2(n, control = permControl())"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-defunct.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Defunct Functions in Package vegan — vegan-defunct","text":"adonis2 replaces adonis extended functionality completely new internal design. shared arguments adonis similar adonis2, arguments contr.unordered contr.ordered can set contrasts within adonis. .mlm function replaced set functions can find statistics directly ordination result object: see hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca, vcov.cca. Function humpfit transferred natto package still available https://github.com/jarioksa/natto/. R functions news used read vegan NEWS (news(package = \"vegan\")), browseVignettes better tool reading vignettes vegandocs. Function commsimulator replaced make.commsim defines Null models, functions nullmodel simulate.nullmodel check input data generate Null model communities. deprecated density densityplot methods replaced similar methods permustats. permustats offers powerful analysis tools permutations, including summary.permustats giving \\(z\\) values (.k.. standardized effect sizes, SES), Q-Q plots (qqnorm.permustats, qqmath.permustats). Function metaMDSrotate replaced MDSrotate can handle monoMDS results addition metaMDS. permutation functions moved permute package, documented . permute package replaces permuted.index permuted.index2 shuffle getNumObs specific nobs-methods.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions in vegan package — vegan-deprecated","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"functions provided compatibility older versions vegan , may defunct soon next release.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"","code":"## moved to vegan3d package: install from CRAN orditkplot(...) ## use toCoda instead as.mcmc.oecosimu(x) as.mcmc.permat(x)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"x object transformed. ... arguments passed functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-deprecated.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Deprecated Functions in vegan package — vegan-deprecated","text":"orditkplot moved vegan3d (version 1.3-0). Install package CRAN use old way. .mcmc functions replaced toCoda.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":null,"dir":"Reference","previous_headings":"","what":"Internal vegan functions — vegan-internal","title":"Internal vegan functions — vegan-internal","text":"Internal vegan functions intended called directly, within functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Internal vegan functions — vegan-internal","text":"","code":"ordiParseFormula(formula, data, xlev = NULL, na.action = na.fail, subset = NULL, X) ordiTerminfo(d, data) ordiNAexclude(x, excluded) ordiNApredict(omit, x) ordiArgAbsorber(..., shrink, origin, scaling, triangular, display, choices, const, truemean, FUN) centroids.cca(x, mf, wt) getPermuteMatrix(perm, N, strata = NULL) howHead(x, ...) pasteCall(call, prefix = \"Call:\") veganCovEllipse(cov, center = c(0, 0), scale = 1, npoints = 100) veganMahatrans(x, s2, tol = sqrt(.Machine$double.eps)) hierParseFormula(formula, data) GowerDblcen(x, na.rm = TRUE) addLingoes(d) addCailliez(d)"},{"path":"https://vegandevs.github.io/vegan/reference/vegan-internal.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Internal vegan functions — vegan-internal","text":"description intended vegan developers: functions intended users, used within functions. general, functions exported namespace, must use get ::: directly call functions. ordiParseFormula returns list three matrices (dependent variables, model.matrix constraints conditions, possibly NULL) needed constrained ordination. Argument xlev passed model.frame. left-hand-side already evaluated calling code, can given argument X re-evaluated. ordiTermInfo finds term information constrained ordination described cca.object. ordiNAexclude implements na.action = na.exclude constrained ordination finding WA scores CCA components site scores unconstrained component excluded rows observations. Function ordiNApredict pads result object WA scores similarly napredict. ordiArgAbsorber absorbs arguments scores function vegan cause superfluous warnings graphical function FUN. implement scores functions new arguments, update ordiArgAbsorber. centroids.cca finds weighted centroids variables. getPermuteMatrix interprets user input returns permutation matrix row gives indices observations permutation. input perm can single number number simple permutations, result defining permutation scheme permutation matrix. howHead formats permutation scheme display. formatting compact one used print permute package, shows non-default choices. output normally used printing results vegan permutations. pasteCall prints function call nicely wrapped Sweave output. veganCovEllipse finds coordinates drawing covariance ellipse. veganMahatrans transforms data matrix Euclidean distances Mahalanobis distances. input data x must matrix centred columns, s2 covariance matrix. s2 given, covariance matrix found x within function. hierParseFormula returns list one matrix (left hand side) model frame factors representing hierarchy levels (right hand side) used adipart, multipart hiersimu. GowerDblcen performs Gower double centring matrix dissimilarities. Similar function earlier available compiled code stats, part official API, therefore poorer replacement. addLingoes addCailliez find constant added non-diagonal (squared) dissimilarities make eigenvalues non-negative Principal Co-ordinates Analysis (wcmdscale, dbrda, capscale). Function cmdscale implements Cailliez method. argument matrix dissimilarities.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":null,"dir":"Reference","previous_headings":"","what":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan package provides tools descriptive community ecology. basic functions diversity analysis, community ordination dissimilarity analysis. multivariate tools can used data types well.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"functions vegan package contain tools diversity analysis, ordination methods tools analysis dissimilarities. Together labdsv package, vegan package provides standard tools descriptive community analysis. Package ade4 provides alternative comprehensive package, several packages complement vegan provide tools deeper analysis specific fields. Package BiodiversityR provides GUI large subset vegan functionality. vegan package developed GitHub (https://github.com/vegandevs/vegan/). GitHub provides --date information forums bug reports. important changes vegan documents can read news(package=\"vegan\") vignettes can browsed browseVignettes(\"vegan\"). vignettes include vegan FAQ, discussion design decisions, short introduction ordination discussion diversity methods. see preferable citation package, type citation(\"vegan\").","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"vegan development team Jari Oksanen, F. Guillaume Blanchet, Roeland Kindt, Pierre Legendre, Peter R. Minchin, R. B. O'Hara, Gavin L. Simpson, Peter Solymos, M. Henry H. Stevens, Helene Wagner. Many people contributed individual functions: see credits function help pages.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegan-package.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Community Ecology Package: Ordination, Diversity and Dissimilarities — vegan-package","text":"","code":"### Example 1: Unconstrained ordination ## NMDS data(varespec) data(varechem) ord <- metaMDS(varespec) #> Square root transformation #> Wisconsin double standardization #> Run 0 stress 0.1843196 #> Run 1 stress 0.184583 #> ... Procrustes: rmse 0.04902295 max resid 0.1551746 #> Run 2 stress 0.1948413 #> Run 3 stress 0.2136761 #> Run 4 stress 0.2079056 #> Run 5 stress 0.2087937 #> Run 6 stress 0.196245 #> Run 7 stress 0.2069725 #> Run 8 stress 0.1948413 #> Run 9 stress 0.1825658 #> ... New best solution #> ... Procrustes: rmse 0.04165046 max resid 0.1519318 #> Run 10 stress 0.1843196 #> Run 11 stress 0.2109853 #> Run 12 stress 0.2177543 #> Run 13 stress 0.2224267 #> Run 14 stress 0.2044974 #> Run 15 stress 0.195049 #> Run 16 stress 0.1969805 #> Run 17 stress 0.22911 #> Run 18 stress 0.2397062 #> Run 19 stress 0.18584 #> Run 20 stress 0.2295494 #> *** Best solution was not repeated -- monoMDS stopping criteria: #> 1: no. of iterations >= maxit #> 19: stress ratio > sratmax plot(ord, type = \"t\") ## Fit environmental variables ef <- envfit(ord, varechem) ef #> #> ***VECTORS #> #> NMDS1 NMDS2 r2 Pr(>r) #> N -0.05721 -0.99836 0.2537 0.038 * #> P 0.61964 0.78489 0.1938 0.102 #> K 0.76634 0.64244 0.1809 0.127 #> Ca 0.68509 0.72846 0.4119 0.005 ** #> Mg 0.63243 0.77462 0.4271 0.003 ** #> S 0.19127 0.98154 0.1752 0.110 #> Al -0.87166 0.49011 0.5269 0.001 *** #> Fe -0.93612 0.35169 0.4451 0.001 *** #> Mn 0.79870 -0.60173 0.5231 0.001 *** #> Zn 0.61752 0.78655 0.1879 0.102 #> Mo -0.90302 0.42960 0.0609 0.521 #> Baresoil 0.92500 -0.37996 0.2508 0.050 * #> Humdepth 0.93288 -0.36018 0.5200 0.002 ** #> pH -0.64804 0.76161 0.2308 0.074 . #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> Permutation: free #> Number of permutations: 999 #> #> plot(ef, p.max = 0.05) ### Example 2: Constrained ordination (RDA) ## The example uses formula interface to define the model data(dune) data(dune.env) ## No constraints: PCA mod0 <- rda(dune ~ 1, dune.env) mod0 #> Call: rda(formula = dune ~ 1, data = dune.env) #> #> Inertia Rank #> Total 84.12 #> Unconstrained 84.12 19 #> Inertia is variance #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 #> 24.795 18.147 7.629 7.153 5.695 4.333 3.199 2.782 #> (Showing 8 of 19 unconstrained eigenvalues) #> plot(mod0) ## All environmental variables: Full model mod1 <- rda(dune ~ ., dune.env) mod1 #> Call: rda(formula = dune ~ A1 + Moisture + Management + Use + Manure, #> data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 63.2062 0.7513 12 #> Unconstrained 20.9175 0.2487 7 #> Inertia is variance #> Some constraints or conditions were aliased because they were redundant #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 RDA7 RDA8 RDA9 RDA10 RDA11 #> 22.396 16.208 7.039 4.038 3.760 2.609 2.167 1.803 1.404 0.917 0.582 #> RDA12 #> 0.284 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 #> 6.627 4.309 3.549 2.546 2.340 0.934 0.612 #> plot(mod1) ## Automatic selection of variables by permutation P-values mod <- ordistep(mod0, scope=formula(mod1)) #> #> Start: dune ~ 1 #> #> Df AIC F Pr(>F) #> + Management 3 87.082 2.8400 0.005 ** #> + Moisture 3 87.707 2.5883 0.005 ** #> + Manure 4 89.232 1.9539 0.015 * #> + A1 1 89.591 1.9217 0.065 . #> + Use 2 91.032 1.1741 0.240 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management #> #> Df AIC F Pr(>F) #> - Management 3 89.62 2.84 0.005 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Moisture 3 85.567 1.9764 0.005 ** #> + Manure 3 87.517 1.3902 0.095 . #> + A1 1 87.424 1.2965 0.200 #> + Use 2 88.284 1.0510 0.360 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Step: dune ~ Management + Moisture #> #> Df AIC F Pr(>F) #> - Moisture 3 87.082 1.9764 0.015 * #> - Management 3 87.707 2.1769 0.010 ** #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 #> #> Df AIC F Pr(>F) #> + Manure 3 85.762 1.1225 0.325 #> + A1 1 86.220 0.8359 0.585 #> + Use 2 86.842 0.8027 0.770 #> mod #> Call: rda(formula = dune ~ Management + Moisture, data = dune.env) #> #> Inertia Proportion Rank #> Total 84.1237 1.0000 #> Constrained 46.4249 0.5519 6 #> Unconstrained 37.6988 0.4481 13 #> Inertia is variance #> #> Eigenvalues for constrained axes: #> RDA1 RDA2 RDA3 RDA4 RDA5 RDA6 #> 21.588 14.075 4.123 3.163 2.369 1.107 #> #> Eigenvalues for unconstrained axes: #> PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 #> 8.241 7.138 5.355 4.409 3.143 2.770 1.878 1.741 0.952 0.909 0.627 0.311 0.227 #> plot(mod) ## Permutation test for all variables anova(mod) #> Permutation test for rda under reduced model #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Model 6 46.425 2.6682 0.001 *** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Permutation test of \"type III\" effects, or significance when a term ## is added to the model after all other terms anova(mod, by = \"margin\") #> Permutation test for rda under reduced model #> Marginal effects of terms #> Permutation: free #> Number of permutations: 999 #> #> Model: rda(formula = dune ~ Management + Moisture, data = dune.env) #> Df Variance F Pr(>F) #> Management 3 18.938 2.1769 0.005 ** #> Moisture 3 17.194 1.9764 0.008 ** #> Residual 13 37.699 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 ## Plot only sample plots, use different symbols and draw SD ellipses ## for Managemenet classes plot(mod, display = \"sites\", type = \"n\") with(dune.env, points(mod, disp = \"si\", pch = as.numeric(Management))) with(dune.env, legend(\"topleft\", levels(Management), pch = 1:4, title = \"Management\")) with(dune.env, ordiellipse(mod, Management, label = TRUE)) ## add fitted surface of diversity to the model ordisurf(mod, diversity(dune), add = TRUE) #> #> Family: gaussian #> Link function: identity #> #> Formula: #> y ~ s(x1, x2, k = 10, bs = \"tp\", fx = FALSE) #> #> Estimated degrees of freedom: #> 1.28 total = 2.28 #> #> REML score: 3.00623 ### Example 3: analysis of dissimilarites a.k.a. non-parametric ### permutational anova adonis2(dune ~ ., dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ ., data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> A1 1 0.7230 0.16817 5.2038 0.002 ** #> Moisture 3 1.1871 0.27613 2.8482 0.006 ** #> Management 3 0.9036 0.21019 2.1681 0.028 * #> Use 2 0.0921 0.02143 0.3315 0.983 #> Manure 3 0.4208 0.09787 1.0096 0.439 #> Residual 7 0.9725 0.22621 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 adonis2(dune ~ Management + Moisture, dune.env) #> Permutation test for adonis under reduced model #> Terms added sequentially (first to last) #> Permutation: free #> Number of permutations: 999 #> #> adonis2(formula = dune ~ Management + Moisture, data = dune.env) #> Df SumOfSqs R2 F Pr(>F) #> Management 3 1.4686 0.34161 3.7907 0.001 *** #> Moisture 3 1.1516 0.26788 2.9726 0.002 ** #> Residual 13 1.6788 0.39051 #> Total 19 4.2990 1.00000 #> --- #> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":null,"dir":"Reference","previous_headings":"","what":"Dissimilarity Indices for Community Ecologists — vegdist","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function computes dissimilarity indices useful popular community ecologists. indices use quantitative data, although named corresponding binary index, can calculate binary index using appropriate argument. find favourite index , can see can implemented using designdist. Gower, Bray--Curtis, Jaccard Kulczynski indices good detecting underlying ecological gradients (Faith et al. 1987). Morisita, Horn--Morisita, Binomial, Cao Chao indices able handle different sample sizes (Wolda 1981, Krebs 1999, Anderson & Millar 2004), Mountford (1962) Raup-Crick indices presence--absence data able handle unknown (variable) sample sizes. indices discussed Krebs (1999) Legendre & Legendre (2012), properties compared Wolda (1981) Legendre & De Cáceres (2012). Aitchison (1986) distance equivalent Euclidean distance CLR-transformed samples (\"clr\") deals positive compositional data. Robust Aitchison distance Martino et al. (2019) uses robust CLR (\"rlcr\"), making applicable non-negative data including zeroes (unlike standard Aitchison).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"vegdist(x, method=\"bray\", binary=FALSE, diag=FALSE, upper=FALSE, na.rm = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"x Community data matrix. method Dissimilarity index, partial match \"manhattan\", \"euclidean\", \"canberra\", \"clark\", \"bray\", \"kulczynski\", \"jaccard\", \"gower\", \"altGower\", \"morisita\", \"horn\", \"mountford\", \"raup\", \"binomial\", \"chao\", \"cao\", \"mahalanobis\", \"chisq\", \"chord\", \"hellinger\", \"aitchison\", \"robust.aitchison\". binary Perform presence/absence standardization analysis using decostand. diag Compute diagonals. upper Return upper diagonal. na.rm Pairwise deletion missing observations computing dissimilarities. ... parameters. ignored, except method =\"gower\" accepts range.global parameter decostand, method=\"aitchison\", accepts pseudocount parameter decostand used clr transformation.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jaccard (\"jaccard\"), Mountford (\"mountford\"), Raup--Crick (\"raup\"), Binomial Chao indices discussed later section. function also finds indices presence/ absence data setting binary = TRUE. following overview gives first quantitative version, \\(x_{ij}\\) \\(x_{ik}\\) refer quantity species (column) \\(\\) sites (rows) \\(j\\) \\(k\\). binary versions \\(\\) \\(B\\) numbers species compared sites, \\(J\\) number species occur compared sites similarly designdist (many indices produce identical binary versions): Jaccard index computed \\(2B/(1+B)\\), \\(B\\) Bray--Curtis dissimilarity. Binomial index derived Binomial deviance null hypothesis two compared communities equal. able handle variable sample sizes. index fixed upper limit, can vary among sites shared species. discussion, see Anderson & Millar (2004). Cao index CYd index (Cao et al. 1997) suggested minimally biased index high beta diversity variable sampling intensity. Cao index fixed upper limit, can vary among sites shared species. index intended count (integer) data, undefined zero abundances; replaced arbitrary value \\(0.1\\) following Cao et al. (1997). Cao et al. (1997) used \\(\\log_{10}\\), current function uses natural logarithms values approximately \\(2.30\\) times higher 10-based logarithms. Anderson & Thompson (2004) give alternative formulation Cao index highlight relationship Binomial index (). Mountford index defined \\(M = 1/\\alpha\\) \\(\\alpha\\) parameter Fisher's logseries assuming compared communities samples community (cf. fisherfit, fisher.alpha). index \\(M\\) found positive root equation \\(\\exp() + \\exp(bM) = 1 + \\exp[(+b-j)M]\\), \\(j\\) number species occurring communities, \\(\\) \\(b\\) number species separate community (index uses presence--absence information). Mountford index usually misrepresented literature: indeed Mountford (1962) suggested approximation used starting value iterations, proper index defined root equation . function vegdist solves \\(M\\) Newton method. Please note either \\(\\) \\(b\\) equal \\(j\\), one communities subset , dissimilarity \\(0\\) meaning non-identical objects may regarded similar index non-metric. Mountford index range \\(0 \\dots \\log(2)\\). Raup--Crick dissimilarity (method = \"raup\") probabilistic index based presence/absence data. defined \\(1 - prob(j)\\), based probability observing least \\(j\\) species shared compared communities. current function uses analytic result hypergeometric distribution (phyper) find probabilities. probability (index) dependent number species missing sites, adding -zero species data removing missing species data influence index. probability ( index) may almost zero almost one wide range parameter values. index nonmetric: two communities shared species may dissimilarity slightly one, two identical communities may dissimilarity slightly zero. index uses equal occurrence probabilities species, Raup Crick originally suggested sampling probabilities proportional species frequencies (Chase et al. 2011). simulation approach unequal species sampling probabilities implemented raupcrick function following Chase et al. (2011). index can also used transposed data give probabilistic dissimilarity index species co-occurrence (identical Veech 2013). Chao index tries take account number unseen species pairs, similarly method = \"chao\" specpool. Function vegdist implements Jaccard, index defined \\(1-\\frac{U \\times V}{U + V - U \\times V}\\); types can defined function chaodist. Chao equation, \\(U = C_j/N_j + (N_k - 1)/N_k \\times a_1/(2 a_2) \\times S_j/N_j\\), \\(V\\) similar except site index \\(k\\). \\(C_j\\) total number individuals species site \\(j\\) shared site \\(k\\), \\(N_j\\) total number individuals site \\(j\\), \\(a_1\\) (\\(a_2\\)) number species occurring site \\(j\\) one (two) individuals site \\(k\\), \\(S_j\\) total number individuals species present site \\(j\\) occur one individual site \\(k\\) (Chao et al. 2005). Morisita index can used genuine count data (integers) . Horn--Morisita variant able handle abundance data. Mahalanobis distances Euclidean distances matrix columns centred, unit variance, uncorrelated. index commonly used community data, sometimes used environmental variables. calculation based transforming data matrix using Euclidean distances following Mardia et al. (1979). Mahalanobis transformation usually fails number columns larger number rows (sampling units). transformation fails, distances nearly constant except small numeric noise. Users must check returned Mahalanobis distances meaningful. Euclidean Manhattan dissimilarities good gradient separation without proper standardization still included comparison special needs. Chi-square distances (\"chisq\") Euclidean distances Chi-square transformed data (see decostand). internal standardization used correspondence analysis (cca, decorana). Weighted principal coordinates analysis distances row sums weights equal correspondence analysis (see Example wcmdscale). Chi-square distance intended non-negative data, typical community data. However, can calculated long margin sums positive, warning issued negative data entries. Chord distances (\"chord\") Euclidean distance matrix rows standardized unit norm (sums squares 1) using decostand. Geometrically standardization moves row points surface multidimensional unit sphere, distances chords across hypersphere. Hellinger distances (\"hellinger\") related Chord distances, data standardized unit total (row sums 1) using decostand, square root transformed. distances upper limit \\(\\sqrt{2}\\). Bray--Curtis Jaccard indices rank-order similar, indices become identical rank-order similar standardizations, especially presence/absence transformation equalizing site totals decostand. Jaccard index metric, probably preferred instead default Bray-Curtis semimetric. Aitchison distance (1986) robust Aitchison distance (Martino et al. 2019) metrics deal compositional data. Aitchison distance said outperform Jensen-Shannon divergence Bray-Curtis dissimilarity, due better stability subsetting aggregation, proper distance (Aitchison et al., 2000). naming conventions vary. one adopted traditional rather truthful priority. function finds either quantitative binary variants indices name, correctly may refer one alternatives instance, Bray index known also Steinhaus, Czekanowski Sørensen index. quantitative version Jaccard probably called Ružička index. abbreviation \"horn\" Horn--Morisita index misleading, since separate Horn index. abbreviation changed index implemented vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Function drop-replacement dist function returns distance object type. result object adds attribute maxdist gives theoretical maximum index sampling units share species, NA maximum.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Aitchison, J. Statistical Analysis Compositional Data (1986). London, UK: Chapman & Hall. Aitchison, J., Barceló-Vidal, C., Martín-Fernández, J.., Pawlowsky-Glahn, V. (2000). Logratio analysis compositional distance. Math. Geol. 32, 271–275. Anderson, M.J. Millar, R.B. (2004). Spatial variation effects habitat temperate reef fish assemblages northeastern New Zealand. Journal Experimental Marine Biology Ecology 305, 191--221. Anderson, M.J., Ellingsen, K.E. & McArdle, B.H. (2006). Multivariate dispersion measure beta diversity. Ecology Letters 9, 683--693. Anderson, M.J & Thompson, .. (2004). Multivariate control charts ecological environmental monitoring. Ecological Applications 14, 1921--1935. Cao, Y., Williams, W.P. & Bark, .W. (1997). Similarity measure bias river benthic Auswuchs community analysis. Water Environment Research 69, 95--106. Chao, ., Chazdon, R. L., Colwell, R. K. Shen, T. (2005). new statistical approach assessing similarity species composition incidence abundance data. Ecology Letters 8, 148--159. Chase, J.M., Kraft, N.J.B., Smith, K.G., Vellend, M. Inouye, B.D. (2011). Using null models disentangle variation community dissimilarity variation \\(\\alpha\\)-diversity. Ecosphere 2:art24 doi:10.1890/ES10-00117.1 Faith, D. P, Minchin, P. R. Belbin, L. (1987). Compositional dissimilarity robust measure ecological distance. Vegetatio 69, 57--68. Gower, J. C. (1971). general coefficient similarity properties. Biometrics 27, 623--637. Krebs, C. J. (1999). Ecological Methodology. Addison Wesley Longman. Legendre, P. & De Cáceres, M. (2012). Beta diversity variance community data: dissimilarity coefficients partitioning. Ecology Letters 16, 951--963. doi:10.1111/ele.12141 Legendre, P. Legendre, L. (2012) Numerical Ecology. 3rd English ed. Elsevier. Mardia, K.V., Kent, J.T. Bibby, J.M. (1979). Multivariate analysis. Academic Press. Martino, C., Morton, J.T., Marotz, C.., Thompson, L.R., Tripathi, ., Knight, R. & Zengler, K. (2019) novel sparse compositional technique reveals microbial perturbations. mSystems 4, 1. Mountford, M. D. (1962). index similarity application classification problems. : P.W.Murphy (ed.), Progress Soil Zoology, 43--50. Butterworths. Veech, J. . (2013). probabilistic model analysing species co-occurrence. Global Ecology Biogeography 22, 252--260. Wolda, H. (1981). Similarity indices, sample size diversity. Oecologia 50, 296--302.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"Jari Oksanen, contributions Tyler Smith (Gower index), Michael Bedward (Raup--Crick index), Leo Lahti (Aitchison robust Aitchison distance).","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"function alternative dist adding ecologically meaningful indices. methods produce similar types objects can interchanged method accepting either. Manhattan Euclidean dissimilarities identical methods. Canberra index divided number variables vegdist, dist. differ constant multiplier, alternative vegdist range (0,1). Function daisy (package cluster) provides alternative implementation Gower index also can handle mixed data numeric class variables. two versions Gower distance (\"gower\", \"altGower\") differ scaling: \"gower\" divides distances number observations (rows) scales column unit range, \"altGower\" omits double-zeros divides number pairs least one -zero value, scale columns (Anderson et al. 2006). can use decostand add range standardization \"altGower\" (see Examples). Gower (1971) suggested omitting double zeros presences, often taken general feature Gower distances. See Examples implementing Anderson et al. (2006) variant Gower index. dissimilarity indices vegdist designed community data, give misleading values negative data entries. results may also misleading NA NaN empty sites. principle, study species composition without species remove empty sites community data.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegdist.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Dissimilarity Indices for Community Ecologists — vegdist","text":"","code":"data(varespec) vare.dist <- vegdist(varespec) # Orlóci's Chord distance: range 0 .. sqrt(2) vare.dist <- vegdist(decostand(varespec, \"norm\"), \"euclidean\") # Anderson et al. (2006) version of Gower vare.dist <- vegdist(decostand(varespec, \"log\"), \"altGower\") #> Warning: non-integer data: divided by smallest positive value # Range standardization with \"altGower\" (that excludes double-zeros) vare.dist <- vegdist(decostand(varespec, \"range\"), \"altGower\")"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":null,"dir":"Reference","previous_headings":"","what":"Display Compact Ordered Community Tables — vegemite","title":"Display Compact Ordered Community Tables — vegemite","text":"Functions vegemite tabasco display compact community tables. Function vegemite prints text tables species rows, site takes one column without spaces. Function tabasco provides interface heatmap colour image data. community table can ordered explicit indexing, environmental variables results ordination cluster analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"vegemite(x, use, scale, sp.ind, site.ind, zero=\".\", select ,...) tabasco(x, use, sp.ind = NULL, site.ind = NULL, select, Rowv = TRUE, Colv = TRUE, labRow = NULL, labCol = NULL, scale, col = heat.colors(12), ...) coverscale(x, scale=c(\"Braun.Blanquet\", \"Domin\", \"Hult\", \"Hill\", \"fix\",\"log\"), maxabund, character = TRUE)"},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Display Compact Ordered Community Tables — vegemite","text":"x Community data. use Either vector, object cca, decorana etc. hclust dendrogram ordering sites species. sp.ind, site.ind Species site indices. tabasco, can also hclust tree, agnes clusterings dendrograms. zero Character used zeros. select Select subset sites. can logical vector (TRUE selected sites), vector indices selected sites. order indices influence results, must specify use site.ind reorder sites. Rowv, Colv Re-order dendrograms rows (sites) columns (species) x. Rowv = TRUE, row dendrograms ordered first axis correspondence analysis, Colv = TRUE column dendrograms weighted average (wascores) row order. Alternatively, arguments can vectors used reorder dendrogram. labRow, labCol character vectors row column labels used heatmap instead default. NB., input matrix transposed row labels used data columns. scale vegemite coverscale: cover scale used (can abbreviated). tabasco: scaling colours heatmap. alternatives coverscale can used tabasco, addition \"column\" \"row\" scale columns rows equal maxima (NB., refer transposed data heatmap), \"none\" uses original values. col vector colours used -zero abundance values. maxabund Maximum abundance used scale = \"log\". Data maximum selected subset used missing. character Return character codes suitable vegemite. FALSE, returns corresponding integers. ... Arguments passed coverscale (.e., maxabund) vegemite heatmap tabasco.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Display Compact Ordered Community Tables — vegemite","text":"function vegemite prints traditional community table. display transposed, species rows sites columns. table printed compact form: one character can used abundance, spaces columns. Species occurrences dropped table. Function tabasco produces similar table vegemite using heatmap, abundances coded colours. function scales abundances equal intervals colour palette, either rows columns can scaled equal maxima, coverscale class systems can used. function can also display dendrograms sites (columns) species given argument (use sites, sp.ind species). parameter use used re-order output. use can vector object hclust agnes, dendrogram ordination result recognized scores (ordination methods vegan vegan). hclust, agnes dendrogram must sites. dendrogram displayed sites tabasco, shown vegemite. dendrogram species displayed, except given sp.ind. use vector, used ordering sites. use object ordination, sites species arranged first axis (provided results available also species). use object hclust, agnes dendrogram, sites ordered similarly cluster dendrogram. Function tabasco re-orders dendrogram Rowv = TRUE Rowv vector. re-ordering available vegemite, can done hand using reorder.dendrogram reorder.hclust. Please note dendrogram hclust reordering can differ: unweighted means merged branches used dendrogram, weighted means (= means leaves cluster) used reorder.hclust. cases species scores missing, species ordered weighted averages (wascores) site order. Species sites can ordered explicitly giving indices names parameters sp.ind site.ind. given, take precedence use. subset sites can displayed using argument select, used order sites, still must give use site.ind. However, tabasco makes two exceptions: site.ind select used use dendrogram (clustering result). addition, sp.ind can hclust tree, agnes clustering dendrogram, case dendrogram plotted left side heatmap. Phylogenetic trees directly used, package ape tools transform hclust trees. scale given, vegemite calls coverscale transform percent cover scale scales traditional class scales used vegetation science (coverscale can called directly, ). Function tabasco can also use traditional class scales, treats transformed values corresponding integers. Braun-Blanquet Domin scales actually strict cover scales, limits used codes r + arbitrary. Scale Hill may inappropriately named, since Mark O. Hill probably never intended cover scale. However, used default “cut levels” TWINSPAN, surprisingly many users stick default, de facto standard publications. traditional scales assume values cover percentages maximum 100. However, non-traditional alternative log can used scale range. class limits integer powers 1/2 maximum (argument maxabund), + used non-zero entries less 1/512 maximum (log stands alternatively logarithmic logical). Scale fix intended “fixing” 10-point scales: truncates scale values integers, replaces 10 X positive values 1 +.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Display Compact Ordered Community Tables — vegemite","text":"functions used mainly display table, return (invisibly) list items species ordered species index, sites ordered site index, table final ordered community table. items can used arguments sp.ind site.ind reproduce table, table can edited. addition table, vegemite prints numbers species sites name used cover scale.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Display Compact Ordered Community Tables — vegemite","text":"cover scales presented many textbooks vegetation science; used: Shimwell, D.W. (1971) Description Classification Vegetation. Sidgwick & Jackson.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Display Compact Ordered Community Tables — vegemite","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Display Compact Ordered Community Tables — vegemite","text":"name vegemite chosen output compact, tabasco just compact, uses heat colours.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/vegemite.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Display Compact Ordered Community Tables — vegemite","text":"","code":"data(varespec) ## Print only more common species freq <- apply(varespec > 0, 2, sum) vegemite(varespec, scale=\"Hult\", sp.ind = freq > 10) #> #> 1122212121122 1112 #> 854739268340575634292011 #> Callvulg 111...11.311...1111.111. #> Empenigr 211332121112211111212213 #> Vaccviti 323332212113221211233234 #> Pinusylv 111.111111111.11.1111111 #> Dicrfusc 12.111441121211111111111 #> Dicrpoly .11.11.11..1.11....1.111 #> Pleuschr 144533435123411111111131 #> Polyjuni 111.11.111112111.111.111 #> Pohlnuta 111111111111.1..11.11111 #> Ptilcili 1111111111..1..11.11..12 #> Cladarbu 321122121332143423111121 #> Cladrang 321221131312145443313241 #> Cladstel 11111311.211.11254555542 #> Cladunci 112111111131111111111111 #> Cladcocc 11..11111111111111.1.11. #> Cladcorn 111111111111111111111111 #> Cladgrac 111111111111111111111.11 #> Cladfimb 11.1111111111111111111.1 #> Cladcris 111111111111111111111111 #> Cladchlo ..1.111..1...1.11.11.1.1 #> Cetreric 111...11.111111111.1.1.. #> Cetrisla .11....1111.1....1.11111 #> Stersp 111.11.1111.112111...11. #> Claddefo 1111111111111111111111.1 #> 24 sites, 24 species #> scale: Hult ## Order by correspondence analysis, use Hill scaling and layout: dca <- decorana(varespec) vegemite(varespec, dca, \"Hill\", zero=\"-\") #> #> 1 1 1 11122211122222 #> 203942561738913046572458 #> Flavniva -1114-11-1-11-1--------- #> Cladstel 5555551451425411111211-- #> Cladphyl -1-1----1-------1------- #> Cladcerv 1---1-----------------1- #> Cladsp ---11--1--11---1-1-11-1- #> Cladamau --1---1----1------------ #> Cladchlo 1111---1-11-111-----11-- #> Cladrang 535254555555223414332321 #> Diphcomp --11-----112----------1- #> Stersp -11-1-4111111-1-111--111 #> Pinusylv 11-111111-111111111-1111 #> Polypili ------111--11-11--1----- #> Cetrisla -1-111--1-1--1--111--111 #> Cladcocc -1111-1111111-11111-1-11 #> Cladarbu 113142453555313343413231 #> Vacculig --1-1----3-1---1---12-11 #> Pohlnuta -11111--11111111111111-1 #> Cladfimb 11111111-111111111111-11 #> Callvulg 111-21-12-51---1221-21-- #> Icmaeric ------1--1------11------ #> Empenigr 342214131314344333143131 #> Vaccviti 342514244334455432444443 #> Cladgrac 1-1111111111111111111111 #> Cetreric -1111-11-111---1111-111- #> Cladcorn 111111111111111111111111 #> Cladcris 111111111111111111111111 #> Peltapht --------1-11--1----1--1- #> Ptilcili 1-11---11-11141--1111111 #> Barbhatc ----------1-121----1---- #> Claddefo 11111111-111111111111111 #> Cladbotr ----------1-1111---1-1-1 #> Betupube -------------1------1-1- #> Dicrpoly -1-1--1-11--1211-11--1-1 #> Cladunci 111111122111111251212311 #> Polycomm ------------11--1--1--1- #> Polyjuni 11-11-1111111121111--131 #> Rhodtome ----------1--2-----21--1 #> Dicrfusc 111111111121112145425-41 #> Pleuschr 111213114132524434555555 #> Vaccmyrt ---1------1-24--12133--4 #> Descflex ----1----1--11-----21-11 #> Nepharct --1------1-1---1------2- #> Dicrsp -----1----1--1-1-11-1541 #> Hylosple ---------------1---3--13 #> 24 sites, 44 species #> scale: Hill ## Show one class from cluster analysis, but retain the ordering above clus <- hclust(vegdist(varespec)) cl <- cutree(clus, 3) sel <- vegemite(varespec, use=dca, select = cl == 3, scale=\"Br\") #> #> 1 12 #> 20921 #> Flavniva .++.. #> Cladstel 55542 #> Cladphyl .++.. #> Cladcerv r.... #> Cladsp ..+.. #> Cladchlo r++.+ #> Cladrang 22121 #> Diphcomp ..+.. #> Stersp .+... #> Pinusylv r++1+ #> Cetrisla .++++ #> Cladcocc .++.. #> Cladarbu +1+1+ #> Pohlnuta .++++ #> Cladfimb r++++ #> Callvulg r+.+. #> Empenigr 22122 #> Vaccviti 22223 #> Cladgrac r.+++ #> Cetreric .++.. #> Cladcorn r+++r #> Cladcris r++++ #> Ptilcili r.+.2 #> Barbhatc ....1 #> Claddefo r++++ #> Cladbotr ....+ #> Betupube ....+ #> Dicrpoly .++.1 #> Cladunci +++1+ #> Polycomm ....+ #> Polyjuni r++.+ #> Rhodtome ....1 #> Dicrfusc r++++ #> Pleuschr ++121 #> Vaccmyrt ..+.2 #> Descflex ....+ #> Dicrsp ...++ #> 5 sites, 37 species #> scale: Braun.Blanquet ## Re-create previous vegemite(varespec, sp=sel$sp, site=sel$site, scale=\"Hult\") #> #> 1 12 #> 20921 #> Flavniva .11.. #> Cladstel 55552 #> Cladphyl .11.. #> Cladcerv 1.... #> Cladsp ..1.. #> Cladchlo 111.1 #> Cladrang 32131 #> Diphcomp ..1.. #> Stersp .1... #> Pinusylv 11111 #> Cetrisla .1111 #> Cladcocc .11.. #> Cladarbu 11111 #> Pohlnuta .1111 #> Cladfimb 11111 #> Callvulg 11.1. #> Empenigr 22123 #> Vaccviti 22334 #> Cladgrac 1.111 #> Cetreric .11.. #> Cladcorn 11111 #> Cladcris 11111 #> Ptilcili 1.1.2 #> Barbhatc ....1 #> Claddefo 11111 #> Cladbotr ....1 #> Betupube ....1 #> Dicrpoly .11.1 #> Cladunci 11111 #> Polycomm ....1 #> Polyjuni 111.1 #> Rhodtome ....1 #> Dicrfusc 11111 #> Pleuschr 11111 #> Vaccmyrt ..1.3 #> Descflex ....1 #> Dicrsp ...11 #> 5 sites, 37 species #> scale: Hult ## Re-order clusters by ordination clus <- as.dendrogram(clus) clus <- reorder(clus, scores(dca, choices=1, display=\"sites\"), agglo.FUN = mean) vegemite(varespec, clus, scale = \"Hult\") #> #> 1 111 1211221212222 #> 431567380922149306254578 #> Flavniva 21.111.111....11........ #> Cladamau .1.1...1................ #> Stersp 111211111....111.1.111.1 #> Polypili ..111..1......111..1.... #> Diphcomp .1...111.1...........1.. #> Cladphyl ..1.....11...1.......... #> Cladrang 344544332133111223121121 #> Cladcerv 1.........1..........1.. #> Cladstel 454121215555213111111.1. #> Cladarbu 322344331111132222121111 #> Vacculig 11...2.1........1.1..111 #> Callvulg 111.1.311.11.1..11111... #> Icmaeric ...1.1.......1...1...... #> Cladsp 1...1.11.1......111..11. #> Cladcocc 1111111111...1111111.1.1 #> Pinusylv 1.111.1111111111111111.1 #> Cladchlo .1..111.111.1.11..1.1... #> Cetrisla 1.1...1.11.111...1.111.1 #> Cladfimb 11.11111111111111111.111 #> Peltapht ..1...11.......1.....11. #> Cetreric 11.1111111...1..111111.. #> Cladgrac 11111111.111111111111111 #> Pohlnuta 111..11111.1111111111.11 #> Ptilcili .11.1.11.11.2.11.1111111 #> Barbhatc ......1.....1.11......1. #> Cladcorn 111111111111111111111111 #> Vaccviti 113122132323412331223232 #> Cladcris 111111111111111111111111 #> Empenigr 111111122122312322111231 #> Cladbotr ......1.....1.111...1.11 #> Betupube ............1.....1..1.. #> Cladunci 111111111111131111112111 #> Claddefo 11.111111111111111111111 #> Dicrpoly ..11.1..11..1.1111.11..1 #> Polycomm ............111......11. #> Rhodtome ......1.....1.....1...11 #> Polyjuni 1.111111111.111111.112.1 #> Dicrfusc 11111111111112111442.211 #> Pleuschr 113111111111123333444455 #> Vaccmyrt ......1..1..311..111..13 #> Nepharct .1...1.1........1....1.. #> Dicrsp ......1....11...111133.1 #> Descflex 1....1......1.1...1..111 #> Hylosple ................1....122 #> 24 sites, 44 species #> scale: Hult ## Abundance values have such a wide range that they must be rescaled tabasco(varespec, dca, scale=\"Braun\") ## Classification trees for species data(dune, dune.taxon) taxontree <- hclust(taxa2dist(dune.taxon)) plotree <- hclust(vegdist(dune), \"average\") ## Automatic reordering of clusters tabasco(dune, plotree, sp.ind = taxontree) ## No reordering of taxonomy tabasco(dune, plotree, sp.ind = taxontree, Colv = FALSE) ## Species cluster: most dissimilarity indices do a bad job when ## comparing rare and common species, but Raup-Crick makes sense sptree <- hclust(vegdist(t(dune), \"raup\"), \"average\") tabasco(dune, plotree, sptree)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Averages Scores for Species — wascores","title":"Weighted Averages Scores for Species — wascores","text":"Computes Weighted Averages scores species ordination configuration environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Averages Scores for Species — wascores","text":"","code":"wascores(x, w, expand=FALSE) eigengrad(x, w)"},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Averages Scores for Species — wascores","text":"x Environmental variables ordination scores. w Weights: species abundances. expand Expand weighted averages weighted variance corresponding environmental variables.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores computes weighted averages. Weighted averages “shrink”: extreme values used calculating averages. expand = TRUE, function “deshrinks” weighted averages making biased weighted variance equal biased weighted variance corresponding environmental variable. Function eigengrad returns inverses squared expansion factors attribute shrinkage wascores result environmental gradient. equal constrained eigenvalue cca one gradient used constraint, describes strength gradient.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Averages Scores for Species — wascores","text":"Function wascores returns matrix species define rows ordination axes environmental variables define columns. expand = TRUE, attribute shrinkage inverses squared expansion factors cca eigenvalues variable. Function eigengrad returns shrinkage attribute.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Weighted Averages Scores for Species — wascores","text":"Jari Oksanen","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wascores.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Averages Scores for Species — wascores","text":"","code":"data(varespec) data(varechem) vare.dist <- vegdist(wisconsin(varespec)) vare.mds <- monoMDS(vare.dist) vare.points <- postMDS(vare.mds$points, vare.dist) vare.wa <- wascores(vare.points, varespec) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa, rownames(vare.wa), cex=0.8, col=\"blue\") ## Omit rare species (frequency <= 4) freq <- apply(varespec>0, 2, sum) plot(scores(vare.points), pch=\"+\", asp=1) text(vare.wa[freq > 4,], rownames(vare.wa)[freq > 4],cex=0.8,col=\"blue\") ## Works for environmental variables, too. wascores(varechem, varespec) #> N P K Ca Mg S Al #> Callvulg 25.12401 41.66188 246.92383 572.3431 99.40828 49.10552 245.15751 #> Empenigr 21.61371 44.17350 158.92517 580.8403 89.20628 35.75327 107.05670 #> Rhodtome 22.34553 40.35530 162.31108 643.8819 95.07712 30.84303 28.83159 #> Vaccmyrt 24.96352 49.80649 189.70177 656.4179 96.75529 34.55190 32.75705 #> Vaccviti 21.14028 45.86984 162.12372 613.2126 94.30084 37.26975 116.14656 #> Pinusylv 18.37299 44.24818 163.60195 670.9387 93.52214 37.02238 150.45523 #> Descflex 22.24089 54.74804 212.20357 771.0159 114.15179 38.00000 24.64143 #> Betupube 21.51034 28.86552 112.28276 513.5586 75.15172 23.55172 33.12414 #> Vacculig 28.00729 33.48758 114.90230 372.9653 70.89954 29.32378 202.89008 #> Diphcomp 22.34228 39.71049 127.44259 446.1565 80.16173 32.32963 122.31574 #> Dicrsp 21.33007 60.03758 185.04563 828.0544 148.67509 46.75427 90.42294 #> Dicrfusc 23.45681 39.14575 162.91954 578.2309 77.81648 33.48086 60.66890 #> Dicrpoly 20.65446 43.87409 150.51485 665.5845 115.22112 36.17079 90.16733 #> Hylosple 26.10599 67.88980 245.78681 779.6520 111.96685 42.27433 24.92738 #> Pleuschr 22.60476 54.22534 199.96241 712.6278 109.23425 40.01132 70.43900 #> Polypili 23.17377 43.75902 144.82623 724.9738 85.42623 30.58525 145.73115 #> Polyjuni 22.89480 47.98022 154.81906 643.3864 87.27819 33.63863 53.24888 #> Polycomm 21.73521 41.17042 154.91549 631.8704 101.84789 32.17324 46.80986 #> Pohlnuta 19.99885 48.59198 169.68855 678.3813 104.31641 39.94427 132.05458 #> Ptilcili 21.27880 33.44211 127.08522 564.5652 85.96417 27.11720 56.60692 #> Barbhatc 21.17461 27.93323 113.13542 497.9138 77.50564 23.72288 42.09749 #> Cladarbu 23.56127 38.04952 142.03073 454.9019 74.00779 33.81002 173.12698 #> Cladrang 24.28421 38.60534 135.31177 463.2750 70.54209 32.53349 183.79979 #> Cladstel 19.28049 46.71060 158.00576 540.4904 80.19153 40.29106 225.89526 #> Cladunci 21.41240 45.49844 163.40402 621.9100 98.35538 40.00734 119.59481 #> Cladcocc 21.72473 42.80681 156.32330 557.9007 80.95448 36.25161 149.82616 #> Cladcorn 22.11640 47.06656 160.36881 623.5185 95.17781 36.75273 104.71463 #> Cladgrac 22.51887 44.06576 156.50214 583.1558 94.10292 36.93930 134.13424 #> Cladfimb 21.77980 41.82652 153.29444 512.4646 78.28232 35.62323 128.96061 #> Cladcris 20.88795 44.12262 171.04016 574.5672 92.52169 38.24003 116.03507 #> Cladchlo 19.51207 45.39655 150.93190 571.0233 95.77586 39.50862 156.81983 #> Cladbotr 22.97660 38.89574 167.20000 590.8021 99.57234 34.79362 87.75957 #> Cladamau 25.07143 35.84286 105.07857 395.2214 68.18571 27.11429 95.91429 #> Cladsp 19.21923 47.37308 168.49231 526.7654 79.54423 45.15385 215.33846 #> Cetreric 21.00944 47.76972 165.07972 579.6322 99.14944 42.25472 163.46000 #> Cetrisla 18.36552 42.73695 151.78374 626.2813 89.77833 35.33498 132.68227 #> Flavniva 18.56110 61.18194 207.67705 502.9203 60.91755 50.22532 396.82405 #> Nepharct 23.33099 49.10019 146.84715 618.1601 64.27319 29.95760 31.72300 #> Stersp 28.19743 32.84800 94.33459 389.5143 53.25377 24.22175 95.39326 #> Peltapht 21.08553 54.45395 193.38816 886.5487 119.35132 37.92500 106.16447 #> Icmaeric 28.88636 27.00000 87.86818 307.0500 40.48182 22.17273 89.94091 #> Cladcerv 20.25000 56.79000 192.36000 519.2300 62.10000 45.18000 314.92000 #> Claddefo 22.19198 45.22981 167.73069 583.7983 92.01320 38.51369 100.46139 #> Cladphyl 15.73750 54.56875 180.39375 775.4500 99.65625 43.35000 208.55000 #> Fe Mn Zn Mo Baresoil Humdepth pH #> Callvulg 75.457843 52.38247 8.281074 0.4734635 27.241036 2.187819 2.845108 #> Empenigr 38.146102 53.49357 7.159938 0.3289657 27.324317 2.367439 2.888078 #> Rhodtome 5.560906 70.48260 7.444100 0.2251490 37.325030 2.689154 2.895352 #> Vaccmyrt 5.589213 75.17221 7.838533 0.2666732 31.404171 2.798935 2.855216 #> Vaccviti 37.586067 51.81515 7.617213 0.3680289 26.307701 2.307879 2.923128 #> Pinusylv 39.121898 35.22311 7.733333 0.3485401 17.762968 1.996350 3.049148 #> Descflex 6.066429 110.87232 9.526607 0.2316071 22.740179 2.834821 2.822857 #> Betupube 5.417241 37.53448 5.637931 0.2068966 51.496552 2.527586 2.979310 #> Vacculig 93.963929 37.73062 4.593824 0.3780552 21.410710 2.041196 3.006965 #> Diphcomp 73.281173 46.88025 4.593827 0.3725309 31.836574 2.103704 2.856790 #> Dicrsp 22.504617 65.35020 13.060765 0.5610370 23.182889 2.232247 2.954272 #> Dicrfusc 13.922252 61.40958 6.922859 0.3218816 26.918674 2.484399 2.806431 #> Dicrpoly 20.973927 33.28779 9.110561 0.3892739 37.304043 2.228713 3.015842 #> Hylosple 4.729157 115.14606 9.885976 0.2851996 20.956264 2.925000 2.807594 #> Pleuschr 19.113811 77.12277 9.007860 0.3405945 24.584979 2.596881 2.858446 #> Polypili 51.993443 36.74754 8.045902 0.2204918 17.368852 1.493443 3.227869 #> Polyjuni 12.885704 82.56274 7.945126 0.2760289 28.116303 2.615523 2.874729 #> Polycomm 7.895775 68.22535 7.843662 0.2591549 38.687324 2.926761 2.860563 #> Pohlnuta 33.089313 42.02290 8.452290 0.3935115 24.709351 2.147328 2.985496 #> Ptilcili 14.036188 33.93547 5.906924 0.2303712 48.941884 2.502498 2.973376 #> Barbhatc 8.199687 31.11379 5.550784 0.2084639 54.331975 2.514734 2.986834 #> Cladarbu 65.470394 38.28429 6.387500 0.4464046 22.592997 2.048540 2.937879 #> Cladrang 76.612752 34.86010 6.616452 0.3903501 17.270158 1.799128 3.022346 #> Cladstel 84.639467 33.27903 7.287216 0.4054879 9.854042 1.851973 3.052167 #> Cladunci 27.463504 40.10322 9.108102 0.5120114 28.312376 2.362687 2.858564 #> Cladcocc 46.653763 43.53584 7.269176 0.3698925 19.972222 2.025090 2.974194 #> Cladcorn 32.916238 53.38441 7.518489 0.3639871 26.620868 2.399518 2.890997 #> Cladgrac 45.758366 44.63911 7.702724 0.4244163 25.546654 2.230350 2.933658 #> Cladfimb 41.710354 47.83687 6.815152 0.3887626 24.763889 2.239394 2.902778 #> Cladcris 34.729585 44.75984 6.933735 0.3746988 29.711352 2.400000 2.841633 #> Cladchlo 42.089655 35.13276 7.908621 0.3814655 22.319655 2.075000 3.022414 #> Cladbotr 23.374468 46.54255 7.263830 0.3106383 45.725532 2.580851 2.904255 #> Cladamau 68.971429 41.98571 4.928571 0.3214286 27.592857 1.857143 2.914286 #> Cladsp 47.913462 49.61923 8.421154 0.6173077 16.864423 2.213462 2.921154 #> Cetreric 42.614167 36.96694 9.516389 0.5687500 21.452639 2.058056 2.923889 #> Cetrisla 29.617734 33.24138 7.518227 0.3192118 26.417980 2.027586 3.065025 #> Flavniva 94.339916 37.03232 9.116371 0.9987764 19.692312 1.799241 2.923629 #> Nepharct 12.910837 115.15684 7.743536 0.2180608 23.135932 2.541065 2.918251 #> Stersp 30.226998 31.97061 7.635502 0.2828767 15.844007 1.477740 3.038756 #> Peltapht 37.598684 56.64079 7.652632 0.2046053 28.321053 2.286842 3.026316 #> Icmaeric 24.236364 23.95909 6.618182 0.2863636 18.727273 1.568182 2.968182 #> Cladcerv 111.090000 52.04000 8.530000 0.6800000 15.393000 1.870000 2.900000 #> Claddefo 25.116325 48.81105 7.599609 0.4096285 33.814545 2.468133 2.823656 #> Cladphyl 50.475000 35.28125 8.568750 0.2812500 7.728125 1.575000 3.231250 ## And the strengths of these variables are: eigengrad(varechem, varespec) #> N P K Ca Mg S Al #> 0.13000842 0.18880078 0.16246365 0.15722067 0.16359171 0.13391967 0.29817406 #> Fe Mn Zn Mo Baresoil Humdepth pH #> 0.20766831 0.27254480 0.16783834 0.09542514 0.20931501 0.25051326 0.14583161"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":null,"dir":"Reference","previous_headings":"","what":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Weighted classical multidimensional scaling, also known weighted principal coordinates analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"wcmdscale(d, k, eig = FALSE, add = FALSE, x.ret = FALSE, w) # S3 method for wcmdscale plot(x, choices = c(1, 2), type = \"t\", ...) # S3 method for wcmdscale scores(x, choices = NA, tidy = FALSE, ...)"},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"d distance structure returned dist full symmetric matrix containing dissimilarities. k dimension space data represented ; must \\(\\{1,2,\\ldots,n-1\\}\\). missing, dimensions zero eigenvalue. eig indicates whether eigenvalues returned. add additive constant \\(c\\) added non-diagonal dissimilarities \\(n-1\\) eigenvalues non-negative. Alternatives \"lingoes\" (default, also used TRUE) \"cailliez\" ( alternative cmdscale). See Legendre & Anderson (1999). x.ret indicates whether doubly centred symmetric distance matrix returned. w Weights points. x wcmdscale result object function called options eig = TRUE x.ret = TRUE (See Details). choices Axes returned; NA returns real axes. type Type graph may \"t\"ext, \"p\"oints \"n\"one. tidy Return scores compatible ggplot2: scores data.frame, score type variable score labelled \"sites\", weights variable weigth, names variable label. ... arguments passed graphical functions.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Function wcmdscale based function cmdscale (package stats base R), uses point weights. Points high weights stronger influence result low weights. Setting equal weights w = 1 give ordinary multidimensional scaling. default options, function returns matrix scores scaled eigenvalues real axes. function called eig = TRUE x.ret = TRUE, function returns object class \"wcmdscale\" print, plot, scores, eigenvals stressplot methods, described section Value. method Euclidean, non-Euclidean dissimilarities eigenvalues can negative. disturbs , can avoided adding constant non-diagonal dissimilarities making eigenvalues non-negative. function implements methods discussed Legendre & Anderson (1999): method Lingoes (add=\"lingoes\") adds constant \\(c\\) squared dissimilarities \\(d\\) using \\(\\sqrt{d^2 + 2 c}\\) method Cailliez (add=\"cailliez\") dissimilarities using \\(d + c\\). Legendre & Anderson (1999) recommend method Lingoes, base R function cmdscale implements method Cailliez.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"eig = FALSE x.ret = FALSE (default), matrix k columns whose rows give coordinates points corresponding positive eigenvalues. Otherwise, object class wcmdscale containing components mostly similar cmdscale: points matrix k columns whose rows give coordinates points chosen represent dissimilarities. eig \\(n-1\\) eigenvalues computed scaling process eig true. x doubly centred weighted distance matrix x.ret true. ac, add additive constant adjustment method used avoid negative eigenvalues. NA FALSE adjustment done. GOF Goodness fit statistics k axes. first value based sum absolute values eigenvalues, second value based sum positive eigenvalues weights Weights. negaxes matrix scores axes negative eigenvalues scaled absolute eigenvalues similarly points. NULL negative eigenvalues k specified, include negative eigenvalues. call Function call.","code":""},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"Gower, J. C. (1966) distance properties latent root vector methods used multivariate analysis. Biometrika 53, 325--328. Legendre, P. & Anderson, M. J. (1999). Distance-based redundancy analysis: testing multispecies responses multifactorial ecological experiments. Ecology 69, 1--24. Mardia, K. V., Kent, J. T. Bibby, J. M. (1979). Chapter 14 Multivariate Analysis, London: Academic Press.","code":""},{"path":[]},{"path":"https://vegandevs.github.io/vegan/reference/wcmdscale.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Weighted Classical (Metric) Multidimensional Scaling — wcmdscale","text":"","code":"## Correspondence analysis as a weighted principal coordinates ## analysis of Euclidean distances of Chi-square transformed data data(dune) rs <- rowSums(dune)/sum(dune) d <- dist(decostand(dune, \"chi\")) ord <- wcmdscale(d, w = rs, eig = TRUE) ## Ordinary CA ca <- cca(dune) ## IGNORE_RDIFF_BEGIN ## Eigevalues are numerically similar ca$CA$eig - ord$eig #> CA1 CA2 CA3 CA4 CA5 #> 6.661338e-16 4.440892e-16 0.000000e+00 1.942890e-16 4.163336e-16 #> CA6 CA7 CA8 CA9 CA10 #> -2.359224e-16 -2.081668e-16 1.110223e-16 -1.387779e-17 -2.775558e-17 #> CA11 CA12 CA13 CA14 CA15 #> 7.632783e-17 1.387779e-17 1.387779e-17 -3.122502e-17 6.938894e-18 #> CA16 CA17 CA18 CA19 #> -3.295975e-17 -1.214306e-17 -1.647987e-17 6.028164e-17 ## Configurations are similar when site scores are scaled by ## eigenvalues in CA procrustes(ord, ca, choices=1:19, scaling = \"sites\") #> #> Call: #> procrustes(X = ord, Y = ca, choices = 1:19, scaling = \"sites\") #> #> Procrustes sum of squares: #> -4.263e-14 #> ## IGNORE_RDIFF_END plot(procrustes(ord, ca, choices=1:2, scaling=\"sites\")) ## Reconstruction of non-Euclidean distances with negative eigenvalues d <- vegdist(dune) ord <- wcmdscale(d, eig = TRUE) ## Only positive eigenvalues: cor(d, dist(ord$points)) #> [1] 0.9975185 ## Correction with negative eigenvalues: cor(d, sqrt(dist(ord$points)^2 - dist(ord$negaxes)^2)) #> [1] 1"},{"path":[]},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-7","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-7 (in development)","text":"vegan longer suggests tcltk, suggests vegan3d (version 1.3-0). See orditkplot section Deprecated Defunct.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-6-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.6-7 (in development)","text":"set functions add new points existing NMDS ordination metaMDS monoMDS. serves purpose adding new points existing eigenvector ordination (instance, predict.rda). main function MDSaddpoints. needs input rectangular matrix dissimilarities new points (rows) old points (columns). Support function dist2xy can extract needed matrices dissimilarities (old new) points, function designdist2 can directly find needed dissimilarities two data matrices. addition, analogue package can calculate rectangular dissimilarities, including many indices defined designdist2. function still experimental. particular user interface may need development. Comments welcome.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-7","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-7 (in development)","text":"adonis defunct: use adonis2 improved functionality. Disabled use summary get ordination scores: use scores! summary.cca see #644. summary.decorana defunct. nothing useful, can extract information scores weights. orditkplot code removed, try launch vegan3d::orditkplot loudly (deprecation message go vegan3d). experimental, may go direct error, depending user experience: please comment!","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-61","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6.1","title":"vegan 2.6-6.1","text":"CRAN release: 2024-05-21","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6.1","text":"C function do_wcentre (weighted centring) can segfault due protection error. problem found automatic CRAN checks. do_wcentre internal function called envfit (vectorfit), wcmdscale varpart (simpleCCA) Fixes bug #653.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-6","dir":"Changelog","previous_headings":"","what":"vegan 2.6-6","title":"vegan 2.6-6","text":"CRAN release: 2024-05-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-6","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-6","text":"vegan depends R version 4.1.0. possible build vegan webR/wasm Fortran compiler. Issue #623.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-6","text":"Permutation tests CCA completely redesigned follow C.J.F ter Braak & D.E. te Beest: Environ Ecol Stat 29, 849–868 (2022) (https://doi.org/10.1007/s10651-022-00545-4). constraints now re-weighted permuted response data, partial model also residualized conditions (partial terms). vegan (release 2.4-6) tests identical Canoco, ter Braak & te Beest demonstrated results biased. old vegan (release 2.4-2 earlier) predictors re-weighted residualized. Re-weighting sufficient remove bias moderate variation weights, residualizing predictors necessary strongly varying weights. See discussion issue #542. new scheme concerns CCA weighted method, RDA dbRDA permutation unchanged. summary ordination results longer prints ordination scores often voluminous hide real summary; see issue #203. Ordination scores extracted scores function. breaks CRAN packages use summary.cca extract scores. switch use scores. maintainers contacted patch files suggested adapt change. See instructions fix packages. scores function constrained ordination (CCA, RDA,dbRDA) default return types scores (display = \"\"). Function can optionally return single type scores list one matrix instead returning matrix (new argument droplist). Constrained ordination objects (cca, rda, dbrda) fitted without formula interface can permutation tests (anova) \"axis\" \"onedf\". Models \"terms\" \"margin\" possible formula interface. Permutation tests constrained ordination objects (cca, rda, dbrda) = \"axis\" stop permutations later axis cutoff limit reached. Earlier cutoff exceeded. default stop permutations P-value 1 reached. analysis takes care P-values axes non-decreasing similarly Canoco. Coefficients effects prc models scaled similarly scaled vegan pre 2.5-1. change suggested Cajo ter Braak. Handling negative eigenvalues changed summary eigenvals. Negative eigenvalues given negative “explanation”, accumulated proportions add 1 last non-negative eigenvalue, 1 last negative eigenvalue. printed output capscale shows proportions real components ignores imaginary dimensions. consistent summary support methods. Issue #636. RsquareAdj capscale based positive eigenvalues, imaginary components ignored. stressplot.dbrda refuses handle partial models. first component variation can displayed dbrda internal (“working”) data structures additive. unconstrained model \"CA\", constrained \"CCA\" partial none. predict dbrda return actual type = \"working\". Earlier returned \"lc\" scores weighted eigenvalues. generated distances eigenvalues, though.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-6","text":"Parallel processing inefficiently implemented slower non-parallel permutation tests constrained ordination adonis2. plot scores cca rda family methods gave error non-existing axes requested. Now ignores requests axes numbers higher result object. summary prc ignored extra parameters (const). -fitted models high number aliased variables caused rare failure adonis2 permutation tests constrained ordination methods (cca, rda, dbrda, capscale) arguments = \"margin\" = \"axis\". also concerned vif.cca intersetcor. Typically occurred high-order interactions factor variables. See issues #452 #622 methods accept rectangular raw data input alternative distances, pass arguments distance functions. arguments vegdist binary = TRUE pseudocount Aitchison distance. concerns dbrda, capscale bioenv. See issue #631 simper gave arbitrary p-values species occur subset. Now given NA. See https://stackoverflow.com/questions/77881877/ Rsquare.adj gave arbitrary p-values -fitted models residual variation. Now returns NA R2 adjusted. Automatic model building proceed cases, fixed ordiR2step returns R2 = 0 overfitted cases. constrained ordination methods issue warning model residual component. See issue #610 inertcomp(..., display = \"sites\", proportional = TRUE) gave wrong values.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"data-sets-2-6-6","dir":"Changelog","previous_headings":"","what":"Data Sets","title":"vegan 2.6-6","text":"Extended description BCI data sets avoid confusion. complete BCI survey includes stems 1 cm DBH, BCI data set vegan subset stems DBH 10 cm published Science 295, 666—669, 2002. data set intended demonstrate methods vegan ecological research suggest contacting BCI team using complete surveys made available Dryad.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-6","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-6","text":"adonis deprecated: use adonis2. several CRAN packages still use adonis although contacted authors June 2022 April 2024, printed message forthcoming deprecation since vegan 2.6-2. See issue #523. See instructions adapt packages functions use adonis2. orditkplot moved CRAN package vegan3d deprecated vegan. See issue #585 announcement #632 use summary extract ordination scores deprecated: use scores extract scores. version still allows extracting scores summary, fail next versions. summary.cca see instructions change package. Support removed ancient cca objects (results cca, rda, dbrda capscale) generated CRAN release 2.5 (2016). still stray relics, use newobject <- update(ancientobject) modernize result. .mcmc.oecosimu .mcmc.permat defunct: use toCoda. Code defunct functions completely removed.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-4","dir":"Changelog","previous_headings":"","what":"vegan 2.6-4","title":"vegan 2.6-4","text":"CRAN release: 2022-10-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-4","text":"Support scores ggplot2 graphics improved extended ordination functions. Suitable scores can requested argument tidy = TRUE, general available types scores returned data frame variable score labelling type. option implemented default method scores structured wcmdscale objects, glitches fixed rda family decorana. Previously tidy scores implemented cca, rda, dbrda family methods, metaMDS, envfit rarecurve. adonis2 anova constrained ordination results can perform sequential test one-degree--freedom effects multi-level factors split contrasts. Previously test available permutest. New summary function varpart brief overview. summary shows unique overall contributed variation set variables. fractions shared several sets variables divided equally contributing sets following Lai J, Zou Y, Zhang J, Peres-Neto P (2022) Methods Ecology Evolution, 13: 782–788. decorana estimates orthogonalized eigenvalues total inertia (scaled Chi-square). Orthogonalized eigenvalues can add total inertia. Together enabled implementing eigenvals, bstick screeplot methods decorana. Axis lengths reported decorana methods. Implemented tolerance method decorana. returns criterion used rescaling DCA, can used inspect success rescaling: constant 1 whole axis. New toCoda function transform sequential null model results oecosimu object can analysed coda convergence independence MCMC model. Function replaces .mcmc.oecosimu .mcmc.permat. metaMDS informative finding similar repeated results random starts uses less confusing language reporting results. Hellinger distance directly available vegdist. vegdist, betadiver raupcrick set attribute maxdist giving numeric value theoretical maximum dissimilarity index. many dissimilarities 1, √2 Chord Hellinger distances, instance. attribute NA open indices ceiling. betadiver three similarity indices set maxdist 0. metaMDS defaults halfchange scaling dissimilarities numeric maxdist attribute, adapt threshold ceiling value. open indices without ceiling, threshold scale dissimilarities. metaMDS used simple test detect index ceiling 1, test now robust can also find maximum values. inference made, function broadcast message assumed value ceiling. Mountford index vegdist now scaled maximum value log(2). Earlier Mountford distances scaled maximum 1. hatvalues constrained ordination objects can sometimes practically 1 1, now cases exactly 1. cases rstandard, rstudent cooks.distance NaN. behaviour similar stats::lm.influence functions. .rad can handle multi-row data frames matrices return list Rank-Abundance data row. Earlier one site handled. decostand returns attribute parameters settings variables used standardization. New function decobackstand can use parameters reconstruct original non-standardized data. Back-transformation exact round-errors, although attempt keep original zeros exact. Back-transformation possible methods pa, rank rrank implemented alr. Back-transformation queried https://stackoverflow.com/questions/73263526/ Rarefaction rarefaction-based methods make sense original observed counts give misleading results data multiplied rare species removed. Observed counts usually singletons (species count one), method issue warning minimum count higher one (may false positive, inspect data). Concerns functions rarefy, drarefy, rrarefy, rarecurve, specaccum(..., method=\"rarefy\"), rareslope avgdist. See github discussion #537. avgdist exposes .dist arguments can return \"dist\"ance objects appear lower triangles instead appearing symmetric matrices. betadisper plots accept col argument (PR #300).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-4","text":"decorana returned wrong results Hill’s piecewise transformation (arguments /) used, unless downweighting also used. scores failed metaMDS result species scores. Bug introduced release 2.6-2. Issue raised https://stackoverflow.com/questions/72483924/ tolerance.cca failed one axis (choice) requested. decostand(..., method=\"alr\") accept name reference, fail cases. CRAN package proxy interfered simper caused obscure error (github issue #528).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-4","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-4","text":"adonis way deprecation. Use adonis2 instead. .mcmc.oecosimu .mcmc.permat deprecated: used S3 methods without depending coda package. Use toCoda instead.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-26-2","dir":"Changelog","previous_headings":"","what":"vegan 2.6-2","title":"vegan 2.6-2","text":"CRAN release: 2022-04-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-6-2","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.6-2","text":"Compiled code adapted changes R 4.2.0. See issues #447, #507. Cross-references function packages adapted stringent tests CRAN","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-6-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.6-2","text":"Aitchison robust Aitchison distances added vegdist. Similar data transformations added decostand. Several functions can return “tidy” data structures can used ggplot2 graphics: rarecurve, scores functions constrained ordination (cca etc.), decorana, envfit, metaMDS. scores.envfit gained argument arrow.mul. vegan plot functions used automatically, now easier use envfit non-vegan plotting. Added function simpson.unb unbiased Simpson diversity robust variation sample sizes. diversity gained argument group calculate indices pooled data. Discussed issue #393. simper much faster even though parallel processing implemented new code. pairs function added plot permustats variables . varpart accepts dissimilarities given symmetric square matrix instead \"dist\" object per wish issue #497. metaMDS adopted user-friendly policy, trymax always maximum number tries. See dicussion https://stackoverflow.com/questions/66748605/. adonis2 accepts strata. adonis2 new main function replaces old adonis. See issue #427. Fisher alpha (fisherfit) badly suited extreme communities follow Fisher’s model. Now fisherfit returns NA communities 0 1 species, issues warning communities consisting singletons extreme Fisher alpha. adipart multipart formulae automatically add unique id constant. always sandwich requested grouping alpha gamma diversities, change results requested groupings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-6-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.6-2","text":"anova function failed marginal tests constrained partial ordination model (cca, rda etc.) interaction terms. Issue #463. Constrained ordination (cca etc.) gave misleading results external variables (constraints, condition) constant explained nothing. decorana fail axes zero eigenvalues. Issue #401. Species accumulation (specaccum) failed one species, several “communities”. Issue #501. Parallel processing failed Windows socket clusters permutest betadisper. Issue #369. orditorp failed numeric labels supplied. Reported https://stackoverflow.com/questions/69272366/. Argument summarize accidentally dropped goodness.cca 2017. taxa2dist failed one usable taxonomic level. See https://stackoverflow.com/questions/67231431/.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-6-2","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.6-2","text":"Function adonis2 replace adonis. humpfit functions defunct removed. available non-CRAN package natto https://github.com/jarioksa/natto. commisimulator defunct. Use simulate nullmodel objects. permuted.index finally defunct (deprecated vegan 2.2-0). .mlm defunct. Use functions documented influence.cca, hatvalues.cca, rstandard.cca, rstudent.cca, cooks.distance.cca others.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-7","dir":"Changelog","previous_headings":"","what":"vegan 2.5-7","title":"vegan 2.5-7","text":"CRAN release: 2020-11-28","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-7","text":"Several distance-based functions failed distances zero (betadisper, capscale, isomap, monoMDS, pcnm, wcmdscale). Reported github issue #372. Non-linear self-starting regression models SSarrhenius, SSgitay, SSgleason SSlomolino failed future R. failure caused internal changes R-devel. Github issue #382. Arrow labels wrong position plot.envfit(..., add = FALSE). rarecurve added unnecessary names results. Github issue #352. permutest betadisper failed parallel processing Windows systems socket clusters used. Github issue #369.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-7","text":"Chi-square Chord distances added vegdist. distances can calculated Euclidean distances transformed data, actually available earlier, many users notice . monoMDS (hence metaMDS) uses stricter convergence criteria. improves possibilities find stable solutions. However, users may still need tweak convergence criteria data. See discussion Github issue #354. text functions constrained ordination plots (cca, rda, dbrda, capscale) accept now expression labels. allows using subscripts, superscripts mathematical expressions. New support function labels.cca returns current text labels authors can change desired ones. See github issue #374. vegemite returns invisibly final formatted table allowing processing. ordiplot passes cex argument linestack decorana plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-6","dir":"Changelog","previous_headings":"","what":"vegan 2.5-6","title":"vegan 2.5-6","text":"CRAN release: 2019-09-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-6","text":"vegdist silently accepted missing values (NA) removed analysis also option na.rm = FALSE. behaviour introduced vegan version 2.5-1. See GitHub issue #319. labels displaced bunch arrows drawn origin ordination graph envfit. See GitHub issue #315. Hill scale coverscale open-ended limited percent data, unlike traditional cover class scales undefined 100% cover.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-6","text":"results .rad longer print index attribute: attribute still object, printing made output messy.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-5","dir":"Changelog","previous_headings":"","what":"vegan 2.5-5","title":"vegan 2.5-5","text":"CRAN release: 2019-05-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-5","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-5","text":"vegan depends R 3.4.0 higher. next vegan release may increase dependence R 3.6.0. R 3.6.0 improved method find random indices permuting sampling data. Vegan relies now R functions ecological null models (functions nullmodel, oecosimu, commsim, permatfull, permatswap others). Technically change compatible R 3.4.0 later, can gain benefits improved code current release R. null models may change due change, certainly change R 3.6.0. See NEWS R 3.6.0 release discussion github issue #312. vegan permutation routines rely permute, gain similar benefits improved randomness upgrade R. Thanks new R dependence, sigma constrained ordination results works without workarounds vegan 2.5-2. fixes completely issue discussed #274. Vegan test results reproduced older versions R 3.6.0. worried , upgrade R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-5","text":"metaMDS failed scaling results engine monoMDS used. However, recommend use monoMDS. See github issue #310.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-5","text":"betadisper changed interpretation negative squared distances give complex-valued distances. Now regarded zero-distances whereas earlier used modulus. change results cases negative squared distances. discussion, see github issue #306.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-4","dir":"Changelog","previous_headings":"","what":"vegan 2.5-4","title":"vegan 2.5-4","text":"CRAN release: 2019-02-04","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-testing-2-5-4","dir":"Changelog","previous_headings":"","what":"Installation and Testing","title":"vegan 2.5-4","text":"code interpreting formula change R 3.6.0, makes constrained ordination methods (cca, rda, dbrda, capscale) fail. See github issue #299. R 3.6.0 introduces new environment variable _R_CHECK_LENGTH_1_LOGIC2_, several functions fail variable set. Changes concern ordiplot, plot summary constrained ordination objects, ordixyplot. See github issue #305.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-4","text":"decorana gave incorrect results downweighting used (argument iweigh = 1). bug introduced vegan 2.5-1 reported github issue #303. goodness constrained ordination methods failed constraints rank = 1 (one constraining variable). Reported Pierre Legendre.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-4","text":"Adjusted R2 enabled partial RDA models (functions rda dbrda) partial CCA models (function cca) function RsquareAdj. feature disabled vegan 2.5-1 . RDA, calculation similar vegan 2.4-6 earlier. Partial CCA now consistent RDA differs earlier implementation. methods, partial models consistent varpart. See github issue #295.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-3","dir":"Changelog","previous_headings":"","what":"vegan 2.5-3","title":"vegan 2.5-3","text":"CRAN release: 2018-10-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-5-3","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.5-3","text":"Tests numerical analysis written robustly give similar results alternative platforms versions R BLAS/Lapack libraries. See github issue #282.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-3","text":"Constrained ordination gave misleading results constraints conditions data NULL variables. rarely happens normal usage, happen marginal anova reported github issue #291. Several functions numerical analysis wrongly accepted non-numeric data (instance, factors) gave either meaningless results confusing error messages. Fixed functions include beals, designdist, diversity, gdispweight, indpower, spantree, specpool, tsallis, tsallisaccum vegdist. See github issue #292. envfit vectors fail missing data. original data scaled centred similarly simulations simulate.rda several simulations returned simmat object (compatible nullmodel simulations can used oecosimu).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-5-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.5-3","text":"anosim checks input avoid confusing error messages like reported StackOverflow question 52082743. Broken-stick distribution (function bstick) longer calculated distance-based Redundancy Analysis (dbrda) negative eigenvalues, clear done. Now dbrda capscale similar respect. print function betadisper results gained new argument neigen select number eigenvalues shown. print robust number eigenvalues lower requested neigen.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-5-3","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.5-3","text":"Function humpfit moved natto package still available https://github.com/jarioksa/natto. scheduled complete removal vegan 2.6-0.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-2","dir":"Changelog","previous_headings":"","what":"vegan 2.5-2","title":"vegan 2.5-2","text":"CRAN release: 2018-05-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-compatibility-2-5-2","dir":"Changelog","previous_headings":"","what":"Installation and Compatibility","title":"vegan 2.5-2","text":"Vegan declares dependence R version 3.2.0. dependence yet noticed previous vegan release. However, generic sigma function defined R-3.3.0, therefore sigma.cca vegan must spelt completely using R-3.2.x. See discussion issue #274. CRAN package klaR function rda, loaded together vegan clashes vegan rda Redundancy Analysis. Vegan tries mitigate problem. cases vegan functions used vegan loaded klaR, error message issued klaR objects handled vegan functions. klaR also tricked print informative message handles vegan objects. However, vegan namespace can attached automatically start-klaR functions take precedence. reported issue #277. Bioconductor package phyloseq problem vegdist function dissimilarities. problem can fixed re-installing phyloseq source package. , must either downgrade vegan version 2.4-6 wait till Bioconductor binary packages upgraded. reported Stackoverflow, vegan issue #272, phyloseq issues #918 #921.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-2","text":"Plotting betadisper failed groups one member. Reported Stackoverflow “Error: Incorrect .dimensions” plotting multivariate data Vegan. Permutation tests constrained ordination (anova.cca, permutest.cca) fail parallel processing socket clusters. Socket clusters always used Windows can also used operating systems created makeCluster. See issue #276.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-25-1","dir":"Changelog","previous_headings":"","what":"vegan 2.5-1","title":"vegan 2.5-1","text":"CRAN release: 2018-04-14","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-5-1","dir":"Changelog","previous_headings":"","what":"GENERAL","title":"vegan 2.5-1","text":"major new release changes package: Nearly 40% program files changed previous release. Please report regressions issues https://github.com/vegandevs/vegan/issues/. Compiled code used much extensively, compiled functions use .Call interface. gives smaller memory footprint also faster. wall clock time, greatest gains permutation tests constrained ordination methods (anova.cca) binary null models (nullmodel). Constrained ordination functions (cca, rda, dbrda, capscale) completely rewritten share code. makes consistent robust. internal structure changed constrained ordination objects, scripts may fail try access result object directly. never guarantee unchanged internal structure, scripts changed use provided support functions access result object (see documentation cca.object github issue #262). support analysis functions may longer work result objects created previous vegan versions. use update(old.result.object) fix old result objects. See github issues #218, #227. vegan includes tests run checking package installation. See github issues #181, #271. informative messages (warnings, notes error messages) cleaned unified also makes possible provide translations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.5-1","text":"avgdist: new function find averaged dissimilarities several random rarefactions communities. Code Geoffrey Hannigan. See github issues #242, #243, #246. chaodist: new function similar designdist, uses Chao terms supposed take account effects unseen species (Chao et al., Ecology Letters 8, 148-159; 2005). Earlier Jaccard-type Chao dissimilarity vegdist, new code allows defining kind Chao dissimilarity. New functions find influence statistics constrained ordination objects: hatvalues, sigma, rstandard, rstudent, cooks.distance, SSD, vcov, df.residual. earlier found via .mlm function deprecated. See github issue #234. boxplot added permustats results display (standardized) effect sizes. sppscores: new function add replace species scores distance-based ordination dbrda, capscale metaMDS. Earlier dbrda species scores, species scores capscale metaMDS based raw input data may consistent used dissimilarity measure. See github issue #254. cutreeord: new function similar stats::cutree, numbers cluster order appear dendrogram (left right) instead labelling order appeared data. sipoo.map: new data set locations sizes islands Sipoo archipelago bird data set sipoo.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-constrained-ordination-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Constrained Ordination","title":"vegan 2.5-1","text":"inertia Correspondence Analysis (cca) called “scaled Chi-square” instead using name little known statistic. elements Constraints Conditions data frames non-formula call rda cca, automatically expanded model matrices can contain factor variables. Earlier numerical model matrices factors used formula interface. Regression scores constraints can extracted plotted constrained ordination methods. See github issue #226. Full model (model = \"full\") enabled permutations tests constrained ordination results anova.cca permutest.cca. permutest.cca gained new option = \"onedf\" perform tests sequential one degree--freedom contrasts factors. option (yet) enabled anova.cca. permutation tests robust, scoping issues fixed. Permutation tests use compiled C code much faster. See github issue #211. permutest printed layout similar anova.cca. eigenvals gained new argument (model) select either constrained unconstrained scores. old argument (constrained) deprecated. See github issue #207. summary.eigenvals returns matrix instead list containing matrix. Adjusted R2 calculated partial ordination, unclear done (function RsquareAdj). ordiresids can display standardized studentized residuals. Function construct model.frame model.matrix constrained ordination robust fail fewer cases. goodness inertcomp constrained ordination result object longer option find distances: explained variation available. inertcomp gained argument unity. give “local contributions beta-diversity” (LCBD) “species contribution beta-diversity” (SCBD) Legendre & De Cáceres (Ecology Letters 16, 951-963; 2012). goodness disabled capscale. prc gained argument const general scaling results similarly rda. prc uses regression scores Canoco-compatibility.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-null-model-communities-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Null Model Communities","title":"vegan 2.5-1","text":"C code swap-based binary null models made efficients, models faster. Many models selected 2 times 2 submatrix, generated four random numbers (two rows, two columns). Now skip selecting third fourth random number obvious matrix swapped. Since time used generating random numbers functions, candidates rejected, speeds functions. However, also means random number sequences change previous vegan versions, old binary model results replicated exactly. See github issues #197, #255 details timing. Ecological null models (nullmodel, simulate, make.commsim, oecosimu) gained new null model \"greedyqswap\" can radically speed quasi-swap models minimal risk introducing bias. Backtracking written C much faster. However, backtracking models biased, provided classic legacy models.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-in-other-functions-2-5-1","dir":"Changelog","previous_headings":"","what":"New Features in Other Functions","title":"vegan 2.5-1","text":"adonis2 gained column R2 similarly old adonis. Great part R code decorana written C makes faster reduces memory footprint. metaMDS results gained new points text methods. ordiplot ordination plot functions can chained points text functions allowing use magrittr pipes. points text functions gained argument draw arrows allowing use drawing biplots adding vectors environmental variables ordiplot. Since many ordination plot methods return invisible \"ordiplot\" object, points text methods also work . See github issue #257. Lattice graphics (ordixyplot) ordination can add polygons enclose points panel complete data. ordicluster gained option suppress drawing plots can easily embedded functions calculations. .rad returns index included taxa attribute. Random rarefaction (function rrarefy) uses compiled C code much faster. plot specaccum can draw short horizontal bars vertical error bars. See StackOverflow question 45378751. decostand gained new standardization methods rank rrank replace abundance values ranks relative ranks. See github issue #225. Clark dissimilarity added vegdist (calculated designdist). designdist evaluates minimum terms compiled code, function faster vegdist also dissimilarities using minimum terms. Although designdist usually faster vegdist, numerically less stable, particular large data sets. swan passes type argument beals. tabasco can use traditional cover scale values function coverscale. Function coverscale can return scaled values integers numerical analysis instead returning characters printing. varpart can partition Chi-squared inertia correspondence analysis new argument chisquare. adjusted R2 based permutation tests, replicate analysis random variation. explanatory tables can data frames factors single factors varpart automatically expanded model matrices. Earlier factors used one-sided model formulae. Based code suggested Daniel Borcard, Univ. Montréal.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-5-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.5-1","text":"long Condition() statements (> 500 characters) failed partial constrained ordination models (cca, rda, dbrda, capscale). problem detected StackOverflow question 49249816. Labels adjusted arrows rescaled envfit plots. See StackOverflow question 49259747. ordiArrowMul failed one arrow plotted envfit.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-5-1","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.5-1","text":".mlm function constrained correspondence analysis deprecated favour new functions directly give influence statistics. See github issue #234. commsimulator now defunct: use simulate nullmodel objects. ade4 cca objects longer handled vegan: ade4 cca since version 1.7-8 (August 9, 2017).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-6","dir":"Changelog","previous_headings":"","what":"vegan 2.4-6","title":"vegan 2.4-6","text":"CRAN release: 2018-01-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-6","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-6","text":"CRAN packages longer allowed use FORTRAN input, read.cep function used FORTRAN format read legacy CEP Canoco files. avoid NOTEs WARNINGs, function re-written R. new read.cep less powerful fragile, can read data “condensed” format, can fail several cases successful old code. old FORTRAN-based function still available cepreader. See github issue #263. cepreader package developed https://github.com/vegandevs/cepreader.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-6","text":"functions rarefaction (rrarefy), species abundance distribution (preston) species pool (estimateR) need exact integer data, test allowed small fuzz. functions worked correctly original data, data transformed back-transformed, pass integer test fuzz give wrong results. instance, sqrt(3)^2 pass test 3, interpreted strictly integer 2. See github issue #259.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-6","text":"ordiresids uses now weighted residuals cca results.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-5","dir":"Changelog","previous_headings":"","what":"vegan 2.4-5","title":"vegan 2.4-5","text":"CRAN release: 2017-12-01","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-5","text":"Several “Swap & Shuffle” null models generated wrong number initial matrices. Usually generated many, dangerous, slow. However, random sequences change fix. Lattice graphics ordination (ordixyplot friends) colour arrows groups instead randomly mixed colours. Information constant mirrored permutations omitted reporting permutation tests (e.g., anova constrained ordination).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-5","text":"ordistep improved interpretation scope: lower scope missing, formula starting solution taken lower scope instead using empty model. See Stackoverflow question 46985029. fitspecaccum gained new support functions nobs logLik allow better co-operation packages functions. See GitHub issue #250. “backtracking” null model community simulation faster. However, “backtracking” biased legacy model used except comparative studies.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-4","dir":"Changelog","previous_headings":"","what":"vegan 2.4-4","title":"vegan 2.4-4","text":"CRAN release: 2017-08-24","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-4","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-4","text":"orditkplot longer give warnings CRAN tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-4","text":"anova(..., = \"axis\") constrained ordination (cca, rda, dbrda) ignored partial terms Condition(). inertcomp summary.cca failed constrained component defined, explained nothing zero rank. See StackOverflow: R - Error message RDA analysis - vegan package. Labels longer cropped meandist plots.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-4","text":"significance tests axes constrained ordination use now forward testing strategy. extensive analysis indicated previous marginal tests biased. conflict Legendre, Oksanen & ter Braak, Methods Ecol Evol 2, 269–277 (2011) regarded marginal tests unbiased. Canberra distance vegdist can now handle negative input entries similarly latest versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-3","dir":"Changelog","previous_headings":"","what":"vegan 2.4-3","title":"vegan 2.4-3","text":"CRAN release: 2017-04-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-4-3","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.4-3","text":"vegan registers native C Fortran routines. avoids warnings model checking, may also give small gain speed. Future versions vegan deprecate remove elements pCCA$Fit, CCA$Xbar, CA$Xbar cca result objects. release provides new function ordiYbar able construct elements current future releases. Scripts functions directly accessing elements switch ordiYbar smooth transition.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-3","text":".mlm methods constrained ordination include zero intercept give correct residual degrees freedom derived statistics. biplot method rda passes correlation argument scaling algorithm. Biplot scores wrongly centred cca caused small error values. Weighting centring corrected intersetcor spenvcor. fix can make small difference analysing cca results. Partial models correctly handled intersetcor. envfit ordisurf functions failed applied species scores. Non-standard variable names can used within Condition() partial ordination. Partial models used internally within several functions, problem reported Albin Meyer (Univ Lorraine, Metz, France) ordiR2step using variable name contained hyphen (wrongly interpreted minus sign partial ordination). ordispider pass graphical arguments used show difference LC WA scores constrained ordination. ordiR2step uses forward selection avoid several problems model evaluation. tolerance function return NaN cases returned 0. Partial models correctly analysed. Misleading (non-zero) tolerances sometimes given species occurred sampling units one species.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-2","dir":"Changelog","previous_headings":"","what":"vegan 2.4-2","title":"vegan 2.4-2","text":"CRAN release: 2017-01-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-2","text":"Permutation tests (permutests, anova) first axis failed constrained distance-based ordination (dbrda, capscale). Now capscale also throw away negative eigenvalues first eigenvalues tested. permutation tests first axis now faster. problem reported Cleo Tebby fixes discussed GitHub issue #198 pull request #199. support functions dbrda capscale gave results components wrong scale. Fixes stressplot, simulate, predict fitted functions. intersetcor use correct weighting cca results slightly . anova permutest failed betadisper fitted argument bias.adjust = TRUE. Fixes Github issue #219 reported Ross Cunning, O’ahu, Hawaii. ordicluster return invisibly coordinates internal points (clusters points joined), last rows contained coordinates external points (ordination scores points). cca method tolerance returning incorrect values second axis sample heterogeneities species tolerances. See issue #216 details.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-2","text":"Biplot scores scaled similarly site scores constrained ordination methods cca, rda, capscale dbrda. Earlier unscaled (technically, equal scaling axes). tabasco adds argument scale colours rows columns addition old equal scale whole plot. New arguments labRow labCex can used change column row labels. Function also takes care -zero observations coloured: earlier tiny observed values merged zeros distinct plots. Sequential null models somewhat faster (10%). Non-sequential null models may marginally faster. null models generated function nullmodel also used oecosimu. vegdist much faster. used clearly slower stats::dist, now nearly equally fast dissimilarity measure. Handling data= formula interface robust, messages user errors improved. fixes points raised Github issue #200. families orders dune.taxon updated APG IV (Bot J Linnean Soc 181, 1–20; 2016) corresponding classification higher levels (Chase & Reveal, Bot J Linnean Soc 161, 122-127; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-1","dir":"Changelog","previous_headings":"","what":"vegan 2.4-1","title":"vegan 2.4-1","text":"CRAN release: 2016-09-07","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-4-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.4-1","text":"Fortran code modernized avoid warnings latest R. modernization visible effect functions. Please report suspect cases vegan issues.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-1","text":"Several support functions ordination methods failed solution one ordination axis, instance, one constraining variable CCA, RDA friends. concerned goodness constrained ordination, inertcomp, fitted capscale, stressplot RDA, CCA (GitHub issue #189). goodness CCA & friends ignored choices argument (GitHub issue #190). goodness function consider negative eigenvalues db-RDA (function dbrda). Function meandist failed cases one groups one observation. linestack handle expressions labels. regression discussed GitHub issue #195. Nestedness measures nestedbetajac nestedbetasor expecting binary data cope quantitative input evaluating Baselga’s matrix-wide Jaccard Sørensen dissimilarity indices. Function .mcmc cast oecosimu result MCMC object (coda package) failed one chain.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-1","text":"diversity function returns now NA observation NA values instead returning 0. function also checks input refuses handle data negative values. GitHub issue #187. rarefy function work robustly marginal case user asks one individual can one species zero variance. Several functions robust factor arguments contain missing values (NA): betadisper, adipart, multipart, hiersimu, envfit constrained ordination methods cca, rda, capscale dbrda. GitHub issues #192 #193.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-24-0","dir":"Changelog","previous_headings":"","what":"vegan 2.4-0","title":"vegan 2.4-0","text":"CRAN release: 2016-06-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"distance-based-analysis-2-4-0","dir":"Changelog","previous_headings":"","what":"Distance-based Analysis","title":"vegan 2.4-0","text":"Distance-based methods redesigned made consistent ordination (capscale, new dbrda), permutational ANOVA (adonis, new adonis2), multivariate dispersion (betadisper) variation partitioning (varpart). methods can produce negative eigenvalues several popular semimetric dissimilarity indices, handled similarly functions. Now functions designed McArdle & Anderson (Ecology 82, 290–297; 2001). dbrda new function distance-based Redundancy Analysis following McArdle & Anderson (Ecology 82, 290–297; 2001). metric dissimilarities, function equivalent old capscale, negative eigenvalues semimetric indices handled differently. dbrda dissimilarities decomposed directly conditions, constraints residuals negative eigenvalues, components can imaginary dimensions. Function mostly compatible capscale constrained ordination methods, full compatibility achieved (see issue #140 Github). function based code Pierre Legendre. old capscale function constrained ordination still based real components, total inertia components assessed similarly dbrda. significance tests differ previous version, function oldCapscale cast capscale result similar form previously. adonis2 new function permutational ANOVA dissimilarities. based algorithm dbrda. function can perform overall tests independent variables well sequential marginal tests term. old adonis still available, can perform sequential tests. settings, adonis adonis2 give identical results (see Github issue #156 differences). Function varpart can partition dissimilarities using algorithm dbrda. Argument sqrt.dist takes square roots dissimilarities can change many popular semimetric indices metric distances capscale, dbrda, wcmdscale, adonis2, varpart betadisper (issue #179 Github). Lingoes Cailliez adjustments change dissimilarity metric distance capscale, dbrda, adonis2, varpart, betadisper wcmdscale. Earlier Cailliez adjustment capscale (issue #179 Github). RsquareAdj works capscale dbrda allows using ordiR2step model building.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-4-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.4-0","text":"specaccum: plot failed line type (lty) given. Reported Lila Nath Sharma (Univ Bergen, Norway)","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-4-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.4-0","text":"ordibar new function draw crosses standard deviations standard errors ordination diagrams instead corresponding ellipses. Several permustats results can combined new c() function. New function smbind binds together null models row, column replication. sequential models bound together, can treated parallel chains subsequent analysis (e.g., .mcmc). See issue #164 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-4-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.4-0","text":"Null model analysis upgraded: New \"curveball\" algorithm provides fast null model fixed row column sums binary matrices Strona et al. (Nature Commun. 5: 4114; 2014). \"quasiswap\" algorithm gained argument thin can reduce bias null models. \"backtracking\" now much faster, still slow, provided mainly allow comparison better faster methods. Compiled code can now interrupted null model simulations. designdist can now use beta diversity notation (gamma, alpha) easier definition beta diversity indices. metaMDS new iteration strategy: Argument try gives minimum number random starts, trymax maximum number. Earlier hand try gave maximum number, now run least try times. reduces risk trapped local optimum (issue #154 Github). convergent solutions, metaMDS now tabulate stopping criteria (trace = TRUE). can help deciding criteria made stringent number iterations increased. documentation monoMDS metaMDS give detailed information convergence criteria. summary permustats prints now P-values, test direction (alternative) can changed. qqmath function permustats can now plot standardized statistics. partial solution issue #172 Github. MDSrotate can rotate ordination show maximum separation factor levels (classes) using linear discriminant analysis (lda MASS package). adipart, hiersimu multipart expose argument method specify null model. RsquareAdj works cca allows using ordiR2step model building. code developed Dan McGlinn (issue #161 Github). However, cca still used varpart. ordiellipse ordihull allow setting colours, line types graphical parameters. alpha channel can now given also real number 0 … 1 addition integer 0 … 255. ordiellipse can now draw ellipsoid hulls enclose points group. ordicluster, ordisegments, ordispider lines plot functions isomap spantree can use mixture colours connected points. behaviour similar analogous functions vegan3d package. plot betadisper configurable. See issues #128 #166 Github details. text points methods orditkplot respect stored graphical parameters. Environmental data Barro Colorado Island forest plots gained new variables Harms et al. (J. Ecol. 89, 947–959; 2001). Issue #178 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-4-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.4-0","text":"Function metaMDSrotate removed replaced MDSrotate. density densityplot methods various vegan objects deprecated replaced density densityplot permustats. Function permustats can extract permutation simulation results vegan result objects.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-5","dir":"Changelog","previous_headings":"","what":"vegan 2.3-5","title":"vegan 2.3-5","text":"CRAN release: 2016-04-09","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-5","text":"eigenvals fails prcomp results R-devel. next version prcomp argument limit number eigenvalues shown (rank.), breaks eigenvals vegan. calibrate failed cca friends rank given.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-4","dir":"Changelog","previous_headings":"","what":"vegan 2.3-4","title":"vegan 2.3-4","text":"CRAN release: 2016-02-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-4","text":"betadiver index 19 wrong sign one terms. linestack failed labels given, input scores names. Reported Jeff Wood (ANU, Canberra, ACT).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-3-4","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.3-4","text":"vegandocs deprecated. Current R provides better tools seeing extra documentation (news() browseVignettes()).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vignettes-2-3-4","dir":"Changelog","previous_headings":"","what":"Vignettes","title":"vegan 2.3-4","text":"vignettes built standard R tools can browsed browseVignettes. FAQ-vegan partitioning accessible vegandocs function.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"building-2-3-4","dir":"Changelog","previous_headings":"","what":"Building","title":"vegan 2.3-4","text":"Dependence external software texi2dvi removed. Version 6.1 texi2dvi incompatible R prevented building vegan. FAQ-vegan earlier built texi2dvi uses now knitr. , vegan now dependent R-3.0.0. Fixes issue #158 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-3","dir":"Changelog","previous_headings":"","what":"vegan 2.3-3","title":"vegan 2.3-3","text":"CRAN release: 2016-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-3","text":"metaMDS monoMDS fail input dissimilarities huge: reported case magnitude 1E85. Fixes issue #152 Github. Permutations failed defined permute control structures estaccum, ordiareatest, renyiaccum tsallisaccum. Reported Dan Gafta (Cluj-Napoca) renyiaccum. rarefy gave false warnings input vector single sampling unit. extrapolated richness indices specpool needed number doubletons (= number species occurring two sampling units), failed one sampling unit supplied. extrapolated richness estimated single sampling unit, now cases handled smoothly instead failing: observed non-extrapolated richness zero standard error reported. issue reported StackOverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-3","text":"treedist treedive refuse handle trees reversals, .e, higher levels homogeneous lower levels. Function treeheight estimate total height absolute values branch lengths. Function treedive refuses handle trees negative branch heights indicating negative dissimilarities. Function treedive faster. gdispweight works input data matrix instead data frame. Input dissimilarities supplied symmetric matrices data frames robustly recognized anosim, bioenv mrpp.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-2","dir":"Changelog","previous_headings":"","what":"vegan 2.3-2","title":"vegan 2.3-2","text":"CRAN release: 2015-11-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-2","text":"Printing details gridded permutation design fail grid within-plot level. ordicluster joined branches wrong coordinates cases. ordiellipse ignored weights calculating standard errors (kind = \"se\"). influenced plots cca, also influenced ordiareatest.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-3-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.3-2","text":"adonis capscale functions recognize symmetric square matrices dissimilarities. Formerly dissimilarities given \"dist\" objects produced dist vegdist functions, data frames matrices regarded observations x variables data confuse users (e.g., issue #147). mso accepts \"dist\" objects distances among locations alternative coordinates locations. text, points lines functions procrustes analysis gained new argument truemean allows adding procrustes items plots original analysis. rrarefy returns observed non-rarefied communities (warning) users request subsamples larger observed community instead failing. Function drarefy similar returned sampling probabilities 1, now also issues warning. Fixes issue #144 Github.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-1","dir":"Changelog","previous_headings":"","what":"vegan 2.3-1","title":"vegan 2.3-1","text":"CRAN release: 2015-09-25","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-1","text":"Permutation tests always correctly recognize ties observed statistic result low P-values. happen particular predictor variables factors (classes). changes concern functions adonis, anosim, anova permutest functions cca, rda capscale, permutest betadisper, envfit, mantel mantel.partial, mrpp, mso, oecosimu, ordiareatest, protest simper. also fixes issues #120 #132 GitHub. Automated model building constrained ordination (cca, rda, capscale) step, ordistep ordiR2step fail aliased candidate variables, constraints completely explained variables already model. regression introduced vegan 2.2-0. Constrained ordination methods cca, rda capscale treat character variables factors analysis, return centroids plotting. Recovery original data metaMDS computing WA scores species fail expression supplied argument comm long & got deparsed multiple strings. metaMDSdist now returns (possibly modified) data frame community data comm attribute \"comm\" returned dist object. metaMDS now uses compute WA species scores NMDS. addition, deparsed expression comm now robust long expressions. Reported Richard Telford. metaMDS monoMDS rejected dissimilarities missing values. Function rarecurve check input cause confusing error messages. Now function checks input data integers can interpreted counts individuals sampling units species. Unchecked bad inputs reason problems reported Stackoverflow.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-1","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-1","text":"Scaling ordination axes cca, rda capscale can now expressed descriptive strings \"none\", \"sites\", \"species\" \"symmetric\" tell kind scores scaled eigenvalues. can modified arguments hill cca correlation rda. old numeric scaling can still used. permutation data can extracted anova results constrained ordination (cca, rda, capscale) analysed permustats function. New data set BCI.env site information Barro Colorado Island tree community data. useful variables UTM coordinates sample plots. variables constant nearly constant little use normal analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-23-0","dir":"Changelog","previous_headings":"","what":"vegan 2.3-0","title":"vegan 2.3-0","text":"CRAN release: 2015-05-26","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-3-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.3-0","text":"Constrained ordination functions cca, rda capscale now robust. Scoping data set names variable names much improved. fix numerous long-standing problems, instance reported Benedicte Bachelot (email) Richard Telford (Twitter), well issues #16 #100 GitHub. Ordination functions cca rda silently accepted dissimilarities input although analysis makes sense methods. Dissimilarities analysed distance-based redundancy analysis (capscale). variance conditional component -estimated goodness rda results, results wrong partial RDA. problems reported R-sig-ecology message Christoph von Redwitz.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"windows-2-3-0","dir":"Changelog","previous_headings":"","what":"Windows","title":"vegan 2.3-0","text":"orditkplot add file type identifier saved graphics Windows although required. problem concerned Windows OS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-and-functions-2-3-0","dir":"Changelog","previous_headings":"","what":"New Features and Functions","title":"vegan 2.3-0","text":"goodness function constrained ordination (cca, rda, capscale) redesigned. Function gained argument addprevious add variation explained previous ordination components axes statistic = \"explained\". option, model = \"CCA\" include variation explained partialled-conditions, model = \"CA\" include accumulated variation explained conditions constraints. former behaviour addprevious = TRUE model = \"CCA\", addprevious = FALSE model = \"CA\". argument effect statistic = \"distance\", always show residual distance previous components. Formerly displayed residual distance currently analysed model. Functions ordiArrowMul ordiArrowTextXY exported can used normal interactive sessions. functions used scale bunch arrows fit ordination graphics, formerly internal functions used within vegan functions. orditkplot can export graphics SVG format. SVG vector graphics format can edited several external programs, Illustrator Inkscape. Rarefaction curve (rarecurve) species accumulation models (specaccum, fitspecaccum) gained new functions estimate slope curve given location. Originally based response R-SIG-ecology query. rarefaction curves, function rareslope, species accumulation models specslope. functions based analytic equations, can also evaluated interpolated non-integer values. specaccum models functions can evaluated analytic models \"exact\", \"rarefaction\" \"coleman\". \"random\" \"collector\" methods can use finite differences (diff(fitted())). Analytic functions slope used non-linear regression models known fitspecaccum. Species accumulation models (specaccum) non-liner regression models species accumulation (fitspecaccum) work consistently weights. cases, models defined using number sites independent variable, weights means observations can non-integer numbers virtual sites. predict models also use number sites newdata, analytic models can estimate expected values non-integer number sites, non-analytic randomized collector models can interpolate non-integer values. fitspecaccum gained support functions AIC deviance. varpart plots four-component models redesigned following Legendre, Borcard & Roberts Ecology 93, 1234–1240 (2012), use now four ellipses instead three circles two rectangles. components now labelled plots, circles ellipses can easily filled transparent background colour.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-1","dir":"Changelog","previous_headings":"","what":"vegan 2.2-1","title":"vegan 2.2-1","text":"CRAN release: 2015-01-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-1","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-1","text":"maintenance release avoid warning messages caused changes CRAN repository. namespace usage also stringent avoid warnings notes development versions R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-1","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-1","text":"vegan can installed loaded without tcltk package. tcltk package needed orditkplot function interactive editing ordination graphics.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-1","text":"ordisurf failed gam package loaded due namespace issues: support functions gam used instead mgcv functions. tolerance function failed unconstrained correspondence analysis.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-1","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-1","text":"estimateR uses exact variance formula bias-corrected Chao estimate extrapolated number species. new formula may unpublished, derived following guidelines Chiu, Wang, Walther & Chao, Biometrics 70, 671–682 (2014), doi:10.1111/biom.12200, online supplementary material. Diversity accumulation functions specaccum, renyiaccum, tsallisaccum, poolaccum estaccumR use now permute package permutations order sampling sites. Normally functions need simple random permutation sites, restricted permutation permute package user-supplied permutation matrices can used. estaccumR function can use parallel processing. linestack accepts now expressions labels. allows using mathematical symbols formula given mathematical expressions.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-22-0","dir":"Changelog","previous_headings":"","what":"vegan 2.2-0","title":"vegan 2.2-0","text":"CRAN release: 2014-11-17","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-2-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.2-0","text":"Several vegan functions can now use parallel processing slow repeating calculations. functions argument parallel. argument can integer giving number parallel processes. unix-alikes (Mac OS, Linux) launch \"multicore\" processing Windows set \"snow\" clusters desribed documentation parallel package. option \"mc.cores\" set integer > 1, used automatically start parallel processing. Finally, argument can also previously set \"snow\" cluster used Windows unix-alikes. Vegan vignette Design decision explains implementation (use vegandocs(\"decission\"), parallel package extensive documentation parallel processing R. following function use parallel processing analysing permutation statistics: adonis, anosim, anova.cca (permutest.cca), mantel (mantel.partial), mrpp, ordiareatest, permutest.betadisper simper. addition, bioenv can compare several candidate sets models paralle, metaMDS can launch several random starts parallel, oecosimu can evaluate test statistics several null models parallel. permutation tests based permute package offers strong tools restricted permutation. functions argument permutations. default usage simple non-restricted permutations achieved giving single integer number. Restricted permutations can defined using function permute package. Finally, argument can permutation matrix rows define permutations. possible use external user constructed permutations. See help(permutations) brief introduction permutations vegan, permute package full documention. vignette permute package can read vegan command vegandocs(\"permutations\"). following functions use permute package: CCorA, adonis, anosim, anova.cca (plus associated permutest.cca, add1.cca, drop1.cca, ordistep, ordiR2step), envfit (plus associated factorfit vectorfit), mantel (mantel.partial), mrpp, mso, ordiareatest, permutest.betadisper, protest simper. Community null model generation completely redesigned rewritten. communities constructed new nullmodel function defined low level commsim function. actual null models generated simulate function builds array null models. new null models include wide array quantitative models addition old binary models, users can plug generating functions. basic tool invoking analysing null models oecosimu. null models often used analysis nestedness, implementation oecosimu allows analysing statistic, null models better seen alternative permutation tests.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-2-2-0","dir":"Changelog","previous_headings":"","what":"Installation","title":"vegan 2.2-0","text":"vegan package dependencies namespace imports adapted changes R, trigger warnings notes package tests. Three-dimensional ordination graphics using scatterplot3d static plots rgl dynamic plots removed vegan moved companion package vegan3d. package available CRAN.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-2-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.2-0","text":"Function dispweight implements dispersion weighting Clarke et al. (Marine Ecology Progress Series, 320, 11–27). addition, implemented new method generalized dispersion weighting gdispweight. methods downweight species significantly -dispersed. New hclust support functions reorder, rev scores. Functions reorder rev similar functions dendrogram objects base R. However, reorder can use (defaults ) weighted mean. weighted mean node average always mean member leaves, whereas dendrogram uses always unweighted means joined branches. Function ordiareatest supplements ordihull ordiellipse provides randomization test one-sided alternative hypothesis convex hulls ellipses two-dimensional ordination space smaller areas randomized groups. Function permustats extracts inspects permutation results support functions summary, density, densityplot, qqnorm qqmath. density qqnorm standard R tools work one statistic, densityplot qqmath lattice graphics work univariate multivariate statistics. results following functions can extracted: anosim, adonis, mantel (mantel.partial), mrpp, oecosimu, permustest.cca (corresponding anova methods), permutest.betadisper, protest. stressplot functions display ordination distances given number dimensions original distances. method functins similar stressplot metaMDS, always use inherent distances ordination method. functions available results capscale, cca, princomp, prcomp, rda, wcmdscale.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-2-0","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.2-0","text":"cascadeKM one group NA instead random value. ordiellipse can handle points exactly line, including two points (warning). plotting radfit results several species failed communities species one species. RsquareAdj capscale negative eigenvalues now report NA instead using biased method rda results. simper failed group single member.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-2-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.2-0","text":"anova.cca functions re-written use permute package. Old results may exactly reproduced, models missing data may fail several cases. new option analysing sequence models . simulate functions cca rda can return several simulations nullmodel compatible object. functions can produce simulations correlated errors (also capscale) parametric simulation Gaussian error. bioenv can use Manhattan, Gower Mahalanobis distances addition default Euclidean. New helper function bioenvdist can extract dissimilarities applied best model model. metaMDS(..., trace = 2) show convergence information default monoMDS engine. Function MDSrotate can rotate k-dimensional ordination k-1 variables. variables correlated (like usually case), vectors can also correlated previously rotated dimensions, uncorrelated later ones. vegan 2.0-10 changed weighted nestednodf weighted analysis binary data equivalent binary analysis. However, broke equivalence original method. Now function argument wbinary select method analysis. problem reported fix submitted Vanderlei Debastiani (Universidade Federal Rio Grande Sul, Brasil). ordiellipse, ordihull ordiellipse can handle missing values groups. ordispider can now use spatial medians instead means. rankindex can use Manhattan, Gower Mahalanobis distance addition default Euclidean. User can set colours line types function rarecurve plotting rarefaction curves. spantree gained support function .hclust change minimum spanning tree hclust tree. fitspecaccum can weighted analysis. Gained lines method. Functions extrapolated number species size species pool using Chao method modified following Chiu et al., Biometrics 70, 671–682 (2014). Incidence based specpool can now use (defaults ) small sample correction number sites sample size. Function uses basic Chao extrapolation based ratio singletons doubletons, switches now bias corrected Chao extrapolation doubletons (species found twice). variance formula bias corrected Chao derived following supporting line material doi:10.1111/biom.12200 differs slightly Chiu et al. (2014). poolaccum function changed similarly, small sample correction used always. abundance based estimateR uses bias corrected Chao extrapolation, earlier estimated variance classic Chao model. Now use widespread approximate estimate EstimateS variance. changes functions similar EstimateS tabasco uses now reorder.hclust hclust object better ordering previously cast trees dendrogram objects. treedive treedist default now match.force = TRUE can silenced verbose = FALSE. vegdist gained Mahalanobis distance. Nomenclature updated plant community data help Taxonstand taxize packages. taxonomy dune data adapted sources APG III. varespec dune use 8-character names (4 genus + 4 species epithet). New data set phylogenetic distances dune extracted Zanne et al. (Nature 506, 89–92; 2014). User configurable plots rarecurve.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-2-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.2-0","text":"strata deprecated permutations. still accepted phased next releases. Use permute package. cca, rda capscale return scores scaled eigenvalues: use scores function extract scaled results. commsimulator deprecated. Replace commsimulator(x, method) simulate(nullmodel(x, method)). density densityplot permutation results deprecated: use permustats density densityplot method.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-10","dir":"Changelog","previous_headings":"","what":"vegan 2.0-10","title":"vegan 2.0-10","text":"CRAN release: 2013-12-12","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-10","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-10","text":"version adapted changes permute package version 0.8-0 triggers NOTEs package checks. release may last 2.0 series, next vegan release scheduled major release newly designed oecosimu community pattern simulation, support parallel processing, full support permute package. interested developments, may try development versions vegan GitHub report problems user experience us.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-10","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-10","text":"envfit function assumed external variables either numeric factors, failed , say, character strings. Now numeric variables taken continuous vectors, variables (character strings, logical) coerced factors possible. function also work degenerate data, like one level factor constant value continuous environmental variable. ties wrongly assessing permutation P-values vectorfit. nestednodf quantitative data consistent binary models, fill wrongly calculated quantitative data. oecosimu now correctly adapts displayed quantiles simulated values alternative test direction. renyiaccum plotting failed one level diversity scale used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-10","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-10","text":"Kempton Taylor algorithm found unreliable fisherfit fisher.alpha, now estimation Fisher α based number species number individuals. estimation standard errors profile confidence intervals also scrapped. renyiaccum, specaccum tsallisaccum functions gained subset argument. renyiaccum can now add collector curve analysis. collector curve diversity accumulation order sampling units. interesting ordering sampling units allows comparing actual species accumulations expected randomized accumulation. specaccum can now perform weighted accumulation using sampling effort weights.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-9","dir":"Changelog","previous_headings":"","what":"vegan 2.0-9","title":"vegan 2.0-9","text":"CRAN release: 2013-09-25 version released due changes programming interface testing procedures R 3.0.2. using older version R, need upgrade vegan. new features bug fixes. user-visible changes documentation output messages formatting. R changes, version dependent R version 2.14.0 newer lattice package.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-8","dir":"Changelog","previous_headings":"","what":"vegan 2.0-8","title":"vegan 2.0-8","text":"CRAN release: 2013-07-10","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-8","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-8","text":"maintenance release fixes issues raised changed R toolset processing vignettes. also fix typographic issues vignettes.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-8","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-8","text":"ordisurf gained new arguments flexible definition fitted models better utilize mgcv::gam function. linewidth contours can now set argument lwd. Labels arrows positioned better way plot functions results envfit, cca, rda capscale. labels longer overlap arrow tips. setting test direction clearer oecosimu. ordipointlabel gained plot method can used replot saved result.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-7","dir":"Changelog","previous_headings":"","what":"vegan 2.0-7","title":"vegan 2.0-7","text":"CRAN release: 2013-03-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-7","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-7","text":"tabasco() new function graphical display community data matrix. Technically interface R heatmap, use closer vegan function vegemite. function can reorder community data matrix similarly vegemite, instance, ordination results. Unlike heatmap, displays dendrograms supplied user, defaults re-order dendrograms correspondence analysis. Species ordered match site ordering like determined user.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-7","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-7","text":"Function fitspecaccum(..., model = \"asymp\") fitted logistic model instead asymptotic model (model = \"logis\"). nestedtemp() failed sparse data (fill < 0.38%).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-7","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-7","text":"plot function constrained ordination results (cca, rda, capscale) gained argument axis.bp (defaults TRUE) can used suppress axis scale biplot arrays. Number iterations nonmetric multidimensional scaling (NMDS) can set keyword maxit (defaults 200) metaMDS.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-2-0-7","dir":"Changelog","previous_headings":"","what":"Deprecated","title":"vegan 2.0-7","text":"result objects cca, rda capscale longer scores u.eig, v.eig wa.eig future versions vegan. change influence normal usage, vegan functions need items. However, external scripts packages may need changes future versions vegan.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-6","dir":"Changelog","previous_headings":"","what":"vegan 2.0-6","title":"vegan 2.0-6","text":"CRAN release: 2013-02-11","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-6","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-6","text":"species scores scaled wrongly capscale(). scaled correctly Euclidean distances used, usually capscale() used non-Euclidean distances. graphics change redone. change scaling mainly influences spread species scores respect site scores. Function clamtest() failed set minimum abundance threshold cases. addition, output wrong possible species groups missing. problems reported Richard Telford (Bergen, Norway). Plotting object fitted envfit() fail p.max used unused levels one factors. unused levels result deletion observations missing values simply result supplying subset larger data set envfit(). multipart() printed wrong information analysis type (analysis correctly). Reported Valerie Coudrain. oecosimu() failed nestedfun returned data frame. fundamental fix vegan 2.2-0, structure oecosimu() result change. plot two-dimensional procrustes() solutions often draw original axes wrong angle. problem reported Elizabeth Ottesen (MIT). Function treedive() functional phylogenetic diversity correctly match species names community data species tree tree contained species occur data. Related function treedist() phylogenetic distances try match names .","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-6","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-6","text":"output capscale() displays value additive constant argument add = TRUE used. fitted() functions cca(), rda() capscale() can now return conditioned (partial) component response: Argument model gained new alternative model = \"pCCA\". dispindmorisita() output gained new column Chi-squared based probabilities null hypothesis (random distribution) true. metaMDS() monoMDS() new default convergence criteria. importantly, scale factor gradient (sfgrmin) stricter. former limit slack large data sets iterations stopped early without getting close solution. addition, scores() ignore now requests dimensions beyond calculated instead failing, scores() metaMDS() results drop dimensions. msoplot() gained legend argument positioning legend. Nestedness function nestednodf() gained plot method. ordiR2step() gained new argument R2scope (defaults TRUE) can used turn criterion stopping adjusted R2 current model exceeds scope. option allows model building scope overdetermined (number predictors higher number observations). ordiR2step() now handles partial redundancy analysis (pRDA). orditorp() gained argument select select rows columns results display. protest() prints standardized residual statistic squared m12 addition squared Procrustes correlation R2. calculated, latter displayed. Permutation tests much faster protest(). Instead calling repeatedly procrustes(), goodness fit statistic evaluated within function. wcmdscale() gained methods print, plot etc. results. methods used full wcmdscale result returned , e.g., argument eig = TRUE. default still return matrix scores similarly standard R function cmdscale(), case new methods used.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-5","dir":"Changelog","previous_headings":"","what":"vegan 2.0-5","title":"vegan 2.0-5","text":"CRAN release: 2012-10-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-5","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-5","text":"anova(, ...) failed = \"axis\" = \"term\". bug reported Dr Sven Neulinger (Christian Albrecht University, Kiel, Germany). radlattice honour argument BIC = TRUE, always displayed AIC.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-5","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-5","text":"vegan functions permutation tests now density method can used find empirical probability distributions permutations. new plot method functions displays density observed statistic. density function available adonis, anosim, mantel, mantel.partial, mrpp, permutest.cca procrustes. Function adonis can return several statistics, now densityplot method (based lattice). Function oecosimu already density densityplot, now similar vegan methods, also work adipart, hiersimu multipart. radfit functions got predict method also accepts arguments newdata total new ranks site totals prediction. functions can also interpolate non-integer “ranks”, models also extrapolate.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-5","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-5","text":"Labels can now set plot envfit results. labels must given order function uses internally, new support function labels can used display default labels correct order. Mantel tests (functions mantel mantel.partial) gained argument na.rm can used remove missing values. options used care: Permutation tests can biased missing values originally matching fixed positions. radfit results can consistently accessed methods whether single model single site, models single site models sites data. functions now methods AIC, coef, deviance, logLik, fitted, predict residuals.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"installation-and-building-2-0-5","dir":"Changelog","previous_headings":"","what":"Installation and Building","title":"vegan 2.0-5","text":"Building vegan vignettes failed latest version LaTeX (TeXLive 2012). R versions later 2.15-1 (including development version) report warnings errors installing checking vegan, must upgrade vegan version. warnings concern functions cIndexKM betadisper, error occurs betadisper. errors warnings triggered internal changes R.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-4","dir":"Changelog","previous_headings":"","what":"vegan 2.0-4","title":"vegan 2.0-4","text":"CRAN release: 2012-06-19","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-4","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-4","text":"adipart assumed constant gamma diversity simulations assessing P-value. give biased results null model produces variable gamma diversities option weights = \"prop\" used. default null model (\"r2dtable\") default option (weights = \"unif\") analysed correctly. anova(, = \"axis\") cases failed due ‘NAMESPACE’ issues. clamtest wrongly used frequencies instead counts calculating sample coverage. detectable differences produced rerunning examples Chazdon et al. 2011 vegan help page. envfit failed unused factor levels. predict cca results type = \"response\" type = \"working\" failed newdata number rows match original data. Now newdata ignored wrong number rows. number rows must match results cca must weighted original row totals. problem concern rda capscale results need row weights. Reported Glenn De’ath.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-4","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-4","text":"Functions diversity partitioning (adipart, hiersimu multipart) now formula default methods. formula method identical previous functions, default method can take two matrices input. Functions adipart multipart can used fast easy overall partitioning alpha, beta gamma diversities omitting argument describing hierarchy. method betadisper biased small sample sizes. effects bias strongest unequal sample sizes. bias adjusted version developed Adrian Stier Ben Bolker, can invoked argument bias.adjust (defaults FALSE). bioenv accepts dissimilarities (square matrices can interpreted dissimilarities) alternative community data. allows using dissimilarities available vegdist. plot function envfit results gained new argument bg can used set background colour plotted labels. msoplot configurable, allows, instance, setting y-axis limits. Hulls ellipses now filled using semitransparent colours ordihull ordiellipse, user can set degree transparency new argument alpha. filled shapes used functions called argument draw = \"polygon\". Function ordihull puts labels (argument label = TRUE) now real polygon centre. ordiplot3d returns function envfit.convert projected location origin. Together can used add envfit results existing ordiplot3d plots. Equal aspect ratio set exactly ordiplot3d underlying core routines allow . Now ordiplot3d sets equal axis ranges, documents urge users verify aspect ratio reasonably equal graph looks like cube. problems solved future, ordiplot3d may removed next releases vegan. Function ordipointlabel gained argument select items plotting. argument can used one set points.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-3","dir":"Changelog","previous_headings":"","what":"vegan 2.0-3","title":"vegan 2.0-3","text":"CRAN release: 2012-03-03","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-3","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-3","text":"Added new nestedness functions nestedbetasor nestedbetajac implement multiple-site dissimilarity indices decomposition turnover nestedness components following Baselga (Global Ecology Biogeography 19, 134–143; 2010). Added function rarecurve draw rarefaction curves row (sampling unit) input data, optionally lines showing rarefied species richness given sample size curve. Added function simper implements “similarity percentages” Clarke (Australian Journal Ecology 18, 117–143; 1993). method compares two groups decomposes average -group Bray-Curtis dissimilarity index contributions individual species. code developed GitHub Eduard Szöcs (Uni Landau, Germany).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-3","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-3","text":"betadisper() failed groups factor empty levels. constrained ordination methods support functions robust border cases (completely aliased effects, saturated models, user requests non-existng scores etc). Concerns capscale, ordistep, varpart, plot function constrained ordination, anova(, = \"margin\"). scores function monoMDS honour choices argument hence dimensions chosen plot. default scores method failed number requested axes higher ordination object . reported error ordiplot R-sig-ecology mailing list.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-3","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-3","text":"metaMDS argument noshare = 0 now regarded numeric threshold always triggers extended dissimilarities (stepacross), instead treated synonymous noshare = FALSE always suppresses extended dissimilarities. Nestedness discrepancy index nesteddisc gained new argument allows user set number iterations optimizing index. oecosimu displays mean simulations describes alternative hypothesis clearly printed output. Implemented adjusted R2 partial RDA. partial model rda(Y ~ X1 + Condition(X2)) component [] = X1|X2 variance partition varpart describes marginal (unique) effect constraining term adjusted R2. Added Cao dissimilarity (CYd) new dissimilarity method vegdist following Cao et al., Water Envir Res 69, 95–106 (1997). index good data high beta diversity variable sampling intensity. Thanks consultation Yong Cao (Univ Illinois, USA).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-2","dir":"Changelog","previous_headings":"","what":"vegan 2.0-2","title":"vegan 2.0-2","text":"CRAN release: 2011-11-15","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-2","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-2","text":"Function capscale failed constrained component zero rank. happened likely partial models conditions aliased constraints. problem observed anova(..., =\"margin\") uses partial models analyses marginal effects, reported email message R-News mailing list. stressplot goodness sometimes failed metaMDS based isoMDS (MASS package) metaMDSdist use defaults step-across (extended) dissimilarities metaMDS(..., engine = \"isoMDS\"). change defaults can also influence triggering step-across capscale(..., metaMDSdist = TRUE). adonis contained minor bug resulting incomplete implementation speed-affect results. fixing bug, bug identified transposing hat matrices. second bug active following fixing first bug. fixing bugs, speed-internal f.test() function fully realised. Reported Nicholas Lewin-Koh.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-2","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-2","text":"ordiarrows ordisegments gained argument order.gives variable sort points within groups. Earlier points assumed order. Function ordispider invisibly returns coordinates points connected. Typically class centroids point, constrained ordination groups LC scores.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-1","dir":"Changelog","previous_headings":"","what":"vegan 2.0-1","title":"vegan 2.0-1","text":"CRAN release: 2011-10-20","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-1","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-1","text":"clamtest: new function classify species generalists specialists two distinct habitats (CLAM test Chazdon et al., Ecology 92, 1332–1343; 2011). test based multinomial distribution individuals two habitat types sampling units, applicable count data -dispersion. .preston gained plot lines methods, .fisher gained plot method (also can add items existing plots). similar plot lines prestonfit fisherfit, display data without fitted lines. raupcrick: new function implement Raup-Crick dissimilarity probability number co-occurring species occurrence probabilities proportional species frequencies. Vegan Raup-Crick index choice vegdist, uses equal sampling probabilities species analytic equations. new raupcrick function uses simulation oecosimu. function follows Chase et al. (2011) Ecosphere 2:art24 [doi:10.1890/ES10-00117.1], developed consultation Brian Inouye.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"bug-fixes-2-0-1","dir":"Changelog","previous_headings":"","what":"Bug Fixes","title":"vegan 2.0-1","text":"Function meandist scramble items give wrong results, especially grouping numerical. problem reported Dr Miguel Alvarez (Univ. Bonn). metaMDS reset tries new model started previous.best solution different model. Function permatswap community null models using quantitative swap never swapped items 2x2 submatrix cells filled. result permutest.cca updated ‘NAMESPACE’ issue. R 2.14.0 changed accept using sd() function matrices (behaviour least since R 1.0-0), several vegan functions changed adapt change (rda, capscale, simulate methods rda, cca capscale). change R 2.14.0 influence results probably wish upgrade vegan avoid annoying warnings.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-1","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-1","text":"nesteddisc slacker hence faster trying optimize statistic tied column frequencies. Tracing showed cases improved ordering found rather early tries, results equally good cases.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"vegan-20-0","dir":"Changelog","previous_headings":"","what":"vegan 2.0-0","title":"vegan 2.0-0","text":"CRAN release: 2011-09-08","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"general-2-0-0","dir":"Changelog","previous_headings":"","what":"General","title":"vegan 2.0-0","text":"Peter Minchin joins vegan team. vegan implements standard R ‘NAMESPACE’. general, S3 methods exported means directly use see contents functions like cca.default, plot.cca anova.ccabyterm. use functions rely R delegation simply use cca result objects use plot anova without suffix .cca. see contents function can use :::, vegan:::cca.default. change may break packages, documents scripts rely non-exported names. vegan depends permute package. package provides powerful tools restricted permutation schemes. vegan permutation gradually move use permute, currently betadisper uses new feature.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-functions-2-0-0","dir":"Changelog","previous_headings":"","what":"New Functions","title":"vegan 2.0-0","text":"monoMDS: new function non-metric multidimensional scaling (NMDS). function replaces MASS::isoMDS default method metaMDS. Major advantages monoMDS ‘weak’ (‘primary’) tie treatment means can split tied observed dissimilarities. ‘Weak’ tie treatment improves ordination heterogeneous data sets, maximum dissimilarities 1 can split. addition global NMDS, monoMDS can perform local hybrid NMDS metric MDS. can also handle missing zero dissimilarities. Moreover, monoMDS faster previous alternatives. function uses Fortran code written Peter Minchin. MDSrotate new function replace metaMDSrotate. function can rotate metaMDS monoMDS results first axis parallel environmental vector. eventstar finds minimum evenness profile Tsallis entropy, uses find corresponding values diversity, evenness numbers equivalent following Mendes et al. (Ecography 31, 450-456; 2008). code contributed Eduardo Ribeira Cunha Heloisa Beatriz Antoniazi Evangelista adapted vegan Peter Solymos. fitspecaccum fits non-linear regression models species accumulation results specaccum. function can use new self-starting species accumulation models vegan self-starting non-linear regression models R. function can fit Arrhenius, Gleason, Gitay, Lomolino (vegan), asymptotic, Gompertz, Michaelis-Menten, logistic Weibull (base R) models. function plot predict methods. Self-starting non-linear species accumulation models SSarrhenius, SSgleason, SSgitay SSlomolino. can used fitspecaccum directly non-linear regression nls. functions implemented found good species-area models Dengler (J. Biogeogr. 36, 728-744; 2009).","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"new-features-2-0-0","dir":"Changelog","previous_headings":"","what":"New Features","title":"vegan 2.0-0","text":"adonis, anosim, meandist mrpp warn negative dissimilarities, betadisper refuses analyse . functions expect dissimilarities, giving something else (like correlations) probably user error. betadisper uses restricted permutation permute package. metaMDS uses monoMDS default ordination engine. Function gains new argument engine can used alternatively select MASS::isoMDS. default use stepacross monoMDS ‘weak’ tie treatment can cope tied maximum dissimilarities one. However, stepacross default isoMDS handle adequately tied maximum dissimilarities. specaccum gained predict method uses either linear spline interpolation data observed points. Extrapolation possible spline interpolation, may make little sense. specpool can handle missing values empty factor levels grouping factor pool. Now also checks length pool matches number observations.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"deprecated-and-defunct-2-0-0","dir":"Changelog","previous_headings":"","what":"Deprecated and Defunct","title":"vegan 2.0-0","text":"metaMDSrotate replaced MDSrotate can also handle results monoMDS. permuted.index2 “new” permutation code removed favour permute package. code intended normal use, packages depending code vegan instead depend permute.","code":""},{"path":"https://vegandevs.github.io/vegan/news/index.html","id":"analyses-2-0-0","dir":"Changelog","previous_headings":"","what":"Analyses","title":"vegan 2.0-0","text":"treeheight uses much snappier code. results unchanged.","code":""}]