18 July 2023
This code repository is an attachment for the resource on the Minnesota Geospatial Commons entitled Land Surface Temperature 2022, Twin Cities. This repository contains javascript code to create and processes a high-resolution (10 meter) dataset on land surface temperatures from September 1, 2022 for the Twin Cities.
This repository contains a folder “javascript_codes” where you can find JavaScript Google Earth Engine (GEE) code used to process and create the downscaled land surface temperature derived from Landsat thermal sensing using the spectral bands of Sentinel-2. The “r_postprocessing” file contains R scripts to post-processes the results of the downscaled raster data.
High resolution (10 meter) land surface temperature (LST) from September 1, 2022 is mapped for the seven-county metropolitan region of the Twin Cities. The goal of the map is to show the heat differences across the region and is not intended to show the maximum temperature that any specific area can reach. The raster dataset was computed at 30 meters using satellite imagery from Landsat 9 and downscaled to 10 meters using Copernicus Sentinel-2. These datasets were integrated using techniques modified from Ermida et al. 2020 and Onačillová et al. 2022). Open water was removed using ancillary data from OpenStreetMap and 2020 Generalized Land Use for the Twin Cities (Metropolitan Council).
First, Landsat 9 imagery taken at 11:59 am CDT on September 01, 2022 was processed into 30-meter resolution LST (based on Ermida et al. 2020). At this time, the air temperature was 88° F at the Minneapolis-St. Paul International Airport (NOAA). A model predicting LST based on spectral indices of Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Normalized Difference Built-up Index (NDBI) was created and applied to 10-meter Sentenel-2 imagery. Sentinel-2 imagery was also taken on September 1, 2022, and this resulted in a 10-meter downscaled LST image (based on Onačillová et al. 2022). To account for anomalies in NDVI on the primary image date of September 1 (e.g., recently harvested agricultural fields), maximum NDVI occurring between July 1, 2022 and September 1, 2022 was used for both Landsat and Sentinel image processing. Water bodies were removed for all processing steps (OpenStreetMap 2023, Metropolitan Council 2021).
This dataset is an update to the 2016 LST data for the Twin Cities Region (Metropolitan Council).
Ermida, S.L., Soares, P., Mantas, V., Göttsche, F.-M., Trigo, I.F., 2020. Google Earth Engine open-source code for Land Surface Temperature estimation from the Landsat series. Remote Sensing, 12 (9), 1471; https://doi.org/10.3390/rs12091471.
Metropolitan Council. 2021. Generalized Land Use 2020. Minnesota Geospatial Commons. https://gisdata.mn.gov/dataset/us-mn-state-metc-plan-generl-lnduse2020
Metropolitan Council. 2017. Land Surface Temperature 2016, Twin Cities. Minnesota Geospatial Commons. https://gisdata.mn.gov/dataset/us-mn-state-metc-env-cva-lst2016
NOAA, National Oceanic and Atmospheric Administration, National Centers for Environmental Information, station USW00014922. September 1, 2022.
Onačillová, K., Gallay, M., Paluba, D., Péliová, A., Tokarčík, O., Laubertová, D. 2022. Combining Landsat 8 and Sentinel 2 data in Google Earth Engine to derive higher resolution land surface temperature maps in urban environment. Remote Sensing, 14 (16), 4076. https://doi.org/10.3390/rs14164076.
OpenStreetMap contributors. 2023. Retrieved from https://planet.openstreetmap.org on April 12, 2023.