An Ensemble of Weight of Evidence and Logistic Regression for Gully Erosion Susceptibility Mapping in the Kakia-Esamburmbur Catchment, Kenya

Gully erosion is the most intensive type of water and it leads to land degradation across world. Therefore, analyzing spatial occurrence this phenomenon crucial for management. The objective research was predict gully susceptibility in Kakia-Esamburmbur catchment Narok, Kenya, which badly affected by erosion. GIS ensemble techniques using weight evidence (WoE) logistic regression (LR) models were used map First, 130 gullies detected study area portioned out 70:30 training validation, respectively. Nine conditioning factors selected as predictors. relationships between locations identified quantified WoE, LR WoE–LR models. results show that use/cover, distance road, sediment transport index (STI) topographic wetness (TWI) are have influence on catchment. Additionally, model performed better than WoE models, producing an AUC value 0.88, higher model, 0.62 0.63. useful mapping help decision makers land-use planning.