Application of advanced machine learning algorithms and geospatial techniques for groundwater potential zone mapping in Gambela Plain, Ethiopia

Abstract Groundwater availability is one of the key anxieties in most semi-arid regions Ethiopia. The purpose this study was to investigate groundwater potential zone map alluvial plain Gambela. applied analytic hierarchy process (AHP) models with four different machine learning algorithms: random forest classifier (RFC), gradient boosting (GBC), decision tree (DTC), and K-neighbor (KNC). features that are used as predictors include geology, geomorphology, slope, soil, lineament density, drainage land use cover (LULC), normalized difference vegetation index (NDVI), topographic wetness (TWI), roughness (TRI), rainfall. final output classified low, moderate, high, very high zones. authentication through receiver operating curve (ROC) shows 78.2, 93.4, 92.5, 72.4, 87.7% values area under (AUC) for AHP, RFC, GBC, DTC, KNC, respectively. results show RFC GBC best GWPZ estimator. also rainfall geomorphology primary factors influencing GWPZ. outcome might promote improved management alternatives other areas country a comparable climate.