Landslide Susceptibility Mapping Using Remote Sensing and Bivariate Frequency Ratio Method in Bahrabise Municipality of Sindhupalchok District, Nepal

Sindhupalchok district, located in the northeastern region of Bagmati Province, Nepal, is highly susceptible to landslides. Every year during the monsoon season, many people lose their lives and properties due to landslides. Bahrabise municipality, in particular, has experienced repeated landslide events in recent years, yet remains under–researched in terms of localized susceptibility mapping. The area’s steep terrain, fragile geology, and proximity to settlements make it a critical zone for detailed landslide risk assessment. The present study focuses on mapping areas susceptible to landslides in Bahrabise Municipality, which lies in the Sindhupalchok district of Nepal. An integrated methodology was adopted in this study, incorporating Geographic Information System (GIS), Remote Sensing (RS), and the Frequency Ratio (FR) model. To prepare the landslide susceptibility map for the area, various spatial and environmental datasets were utilized, including precipitation, slope gradient, elevation, proximity to drainage networks, topographic wetness index, geological features, distance from roads, land use and land cover (LULC), terrain curvature, and slope orientation. These factors and thematic layers were derived from both remotely sensed data and ground – based information such as GPS field surveys, visual inspections, and household interviews, and analyzed using ArcGIS software. The Frequency Ratio (FR) model was applied to assign weights to each thematic layer, representing their influence on landslide occurrence. These weighted factors were then combined within the ArcGIS environment to generate the final landslide susceptibility map for the study area. The resulting landslide susceptibility map classified the study area into four zones, with 47.9% falling under high and very high susceptibility. The model achieved a prediction accuracy of 81.3%, indicating strong reliability in identifying landslide-prone areas. The study examines the feasibility of employing more comprehensive methodologies to identify the landslide susceptibility in the region.