Mapping gully erosion susceptibility in Babai Basin, lumbini Province, Nepal, using machine learning algorithms

A study identifies areas highly susceptible to gully erosion in the Babai Basin, Lumbini Province, Nepal, using optimized machine learning models. The research leverages high-resolution satellite imagery and multi-source geospatial data, including 14 key predictors like geomorphological, topographical, and hydrological variables. Among the models tested—Random Forest (RF), Extreme Gradient Boost, and Support Vector Machine—the RF model demonstrated superior performance with an AUC of 0.82, effectively detecting complex spatial patterns. The resulting susceptibility map provides actionable insights for targeted land-use planning and erosion mitigation strategies, enhancing the precision of conservation efforts in this UNESCO World Heritage Site.

Parajuli, B., Schröder, D., & Omran, A. (2026). Mapping gully erosion susceptibility in Babai Basin, lumbini Province, Nepal, using machine learning algorithms. Natural Hazards, 122(5). https://doi.org/10.1007/s11069-025-07822-y