Assessing climate and land use impacts on surface water yield using remote sensing and machine learning.

Climate and land use changes are critical factors affecting watershed water yields, with significant implications for water resources at both local and regional levels. This study examined the combined effects of temporal and spatial climate variability and land use/land cover (LULC) changes on surface water yield and availability in the Gilgel Gibe watershed, Ethiopia, from 1993 to 2023. Utilizing the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) hydrological models, machine learning, and remote sensing techniques, this study assessed variations in water resources and their impacts on basin water yield. This study utilized Landsat (30 m), MODIS (500 m-1 km), and 4 km resolution climate datasets from the United States Geological Survey (USGS) and NASA POWER for large-scale climate and land-use analyses from 1993 to 2023. An ensemble of machine learning models, including Random Forest (RF), Support Vector Machine (SVM), and XGBoost (XGB), were used to evaluate the effects of climate variability and land use on annual water yield. The study revealed significant land cover changes over a 30-year period. Shrubland decreased from 1,108.37 km