The scripts in this folder calculate the grid cell labels from ground-truth measurements.
The raw data used to produce these grid cell labels is available as described in the main repository readme.
The grid cell labels for each application are saved as: outcomes_sampled_[application]_[location]_16_640_[sampling type]_[N]_0.csv
For example: outcomes_sampled_treecover_CONTUS_16_640_UAR_100000_0.csv
Scripts are numbered in the order they should be run (e.g. extractTreecover_1 should be run before extractTreecover_2). To create all the files some scripts must be run twice, once for uniform-at-random sampling (UAR) and again for population-weighted (POP) sampling.
For variables included in the global or super-resolution analyses, respective scripts are labeled _global and _superres, although some extraction scripts produce both continental US and global samples in a single script (e.g. nighttime lights).
The scripts in the ACS/ folder construct the variables used to demonstrate the rapid generalizability of MOSAIKS in Fig. 4B.