This component contains Google Earth Engine scripts to extract the dataset described in the manuscript.
The project folder contains the following scripts, in the order which they were run:
1. `delineate_regions.js` : divides the study area into polygons and exports the polygons to assets
2. `export_region_terrain.js`: Coarsens elevation (strm) and surface water occurence (jrc) from 30 m to 1 km resolution as saves to assets.
3. `gfc_export_region.js` : exports the yearly Global Forest Change (GFC) data to assets, coarsened from 30 m to 1 km scale and reprojected to MODIS crs (Hansen et al, 2013)
4. `gfc_export_edge_reproject.js`: uses the Canny edge detector to identify forest edges for a given year and region (Canny, 1986), coarsens the forest edges from 30 m to 1 km, and reprojects to MODIS projection. The 1 km edge density is saved to assets for each year.
5. `export_annual_mod11.js` : For a given year and region, identifies assessment and control areas, computes the differences-in-differences metric and covariates for the assessment pixels, and exports to Drive.
6. `gfc_hist.js`: This script creates histogram of tree canopy cover % for the year 2000, using the Hanse GFC data
(Supporting information Figure 1 in manuscript)
#### Modules
The GEE code also contains the following modules:
1. `changeModule.js` : Useful functions to compute the monthly LST difference between years. Needed for the script 'export_annual_mod11.js', to compute the differences in differences metric
2. `didModule.js` : code relevent to extracting the differences-in-differneces dataset (see project/export_annual_mod11.js)
3. `gfcModule.js`: functions to compute yearly forest cover and loss from the GFC data (Hansen et al, 2013)
4. `modModule.js` : contains functions to import the Modis 8-day LST data, and apply basic quality filters and analysis.
\
**References**:
Canny, J. (1986). A computational approach to edge detection. IEEE Transactions on pattern analysis and machine intelligence, (6), 679-698.
Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A., ... & Townshend, J. (2013). High-resolution global maps of 21st-century forest cover change. science, 342(6160), 850-853.
