Spatiotemporal Evolution and Drivers of Soil Erosion in the Yarlung Tsangpo River Basin: A RUSLE-Based 30-Year Analysis and Future Scenario Projections

This dataset includes gridded soil erosion data for the Yarlung Tsangpo River Basin from 1990 to 2019, derived using the Revised Universal Soil Loss Equation (RUSLE) model. It also contains projected rainfall erosivity (R-factor) rasters for four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100) under three CMIP6 emission scenarios (SSP126, SSP245, and SSP585), based on the bias-corrected outputs of 25 GCMs using the quantile delta mapping (QDM) method. In addition, the dataset includes C-factor rasters simulated using the XGBoost machine learning algorithm, trained with historical WordClim climate variables for each future scenario and time slice. Finally, projected average soil erosion rasters for each period and scenario across the entire basin are also provided.