Prep some things for using OrdinaryDiffEq.jl

[charleskawczynski]

This PR preps some things for using OrdinaryDiffEq.jl.

[charleskawczynski]

All of the cases look great, but GABLS has some noticeable changes. Is this tune-able? See any issues with these changes, @yairchn?

[yairchn]

All of the cases look great, but GABLS has some noticeable changes. Is this tune-able? See any issues with these changes, @yairchn?

what is happening is that you remove the condition on the sign of surface buoyancy flux in the updraft BC, it was here

TurbulenceConvection.jl/src/Turbulence_PrognosticTKE.jl

Line 495 in 3d353fb

if Case.Sur.bflux > 0.0

but its not there anymore

TurbulenceConvection.jl/src/Turbulence_PrognosticTKE.jl

Line 495 in 33b8689

function set_updraft_surface_bc(edmf::EDMF_PrognosticTKE, grid, state, up)

so now you are getting an updraft in the first grid point (that never grows) and disrupts the surface BC for GABLS

for all other cases bflux>0 and this is not an issue

[charleskawczynski]

but its not there anymore

It's still there, I renamed it to compute_updraft_surface_bc-- to differentiate between that and a new function I added, set_updraft_surface_bc, which acts on the state and is needed in multiple places.

[charleskawczynski]

bors r+

[bors]

Build succeeded:

• buildkite/turbulenceconvection-ci
• docbuild
• format