RF_GWL_projections_climate

This repository includes all the Python programming language scripts developed for long-term groundwater level projections using the random forests (RF) method in combination with ordinary kriging in Finney County in southwest Kansas under various climate scenarios. The Scikit-learn library is used to construct the RF model and the ArcPy package is utilized for all geospatial and geostatistical analyses. The climate scenarios are developed based on the downscaled climatic data of 20 GCMs for the RCPs of 4.5 and 8.5. The repository also includes the required data for running the scripts. All the scripts and data are uploaded as a single 7z file. To project future GWLs, initially change the home folder pathname in all 3 included python scripts, namely \"all_calculations_in_arcpy.py\", \"projecting_water_level_variations_2017_2099_using_RF.py\", and \"removing_redundant_rasters.py\". Then, run the \"projecting_water_level_variations_2017_2099_using_RF.py\" file.