ABSTRACT Landslides are recurrent and destructive geological hazards, particularly prevalent in the Garhwal Himalayas of India. Landslide susceptibility mapping (LSM) serves as a critical tool for identifying potentially unstable slopes and mitigating risks. This study utilised the Analytical Hierarchy Process (AHP) and Frequency Ratio (FR) models to generate landslide susceptibility maps of the Tehri District, Uttarakhand, using geospatial tools. We considered 15 causative factors including slope, elevation, aspect, relative relief, curvature, geomorphology, stream power index (SPI), geology, normalised difference vegetation index (NDVI), distance to lineament, land use and land cover (LULC), distance to stream, topographic wetness index (TWI), distance to road, and annual rainfall for thematic map preparation. A comprehensive landslide inventory of 2612 landslide events has been developed. For experimentation, 70% of the inventory was used to train the AHP and FR model, while the remaining 30% was used to validate the two resultant landslide susceptibility maps. The AHP model achieved a success rate of 78.7% and a prediction rate of 76.5%, while the FR model outperformed with a success rate of 81.3% and a prediction rate of 80.3%. Subsequently, risk assessment was performed by overlaying susceptibility maps with population and building footprint data. The AHP model indicated that 57% of the population and 63% of buildings fall within high to very high‐risk zones, while the FR model estimated 47% of the population and 55% of buildings in similar zones. These findings emphasise the superior predictive accuracy of the FR model and highlight the importance of integrating susceptibility with exposure data to identify critical hotspots for disaster risk reduction and resilient land‐use planning in the Garhwal Himalayas.
Chandra et al. (Wed,) studied this question.