Landslide susceptibility mapping of Phewa Watershed, Kaski, Nepal

Abstract

Heavy and incessant rainfall in Nepal, particularly during the monsoon season, leads to water-induced risks like landslides, necessitating the use of Landslide Susceptibility Mapping (LSM) for the prediction of landslide risks. We aim to determine the degrees of connection and connective factors among landslide incidents to generate an updated landslide susceptibility map of the Phewa watershed in Kaski District, Nepal. The most dependable and popular statistical approach for determining LSMs is the frequency ratio model, which was created in ArcGIS 10.7.1 by identifying 46 landslides in the area and analyzing eight causal factors. The LSM categorized the area into five classes, with the low class representing a large percentage (43.27 %) and the high class a small percentage (0.63 %). In FR techniques, slope, proximity to a stream or road, land use/cover, and precipitation were assigned greater weight than aspect, profile curvature, and plan curvature. Using the area under the curve approach, the applied model’s accuracy revealed a good performance value of 0.717. Taken together, the mapping information provides a crucial understanding of risks posed by landslides, ultimately reducing the impacts in the protected watershed of the Phewa Lake area.