Remote sensing and GIS-based landslide susceptibility mapping using LNRF method in part of Western Ghats of India

Abstract

Every year, the Western Ghats region experiences devastating landslides, resulting in significant loss of life and damage to both private and public assets. To mitigate these losses, it is essential to identify the areas most susceptible to landslides. This research aims to create an accurate landslide susceptibility map for a section of the Western Ghats region by employing a comprehensive approach that combines remote sensing and geographical information systems (GIS). The Landslide Numerical Risk Factor (LNRF) method is utilized to determine landslide susceptibility zones. The LNRF model incorporates a landslide inventory based on geographic object-based image analysis and influencing factors. The resulting LNRF model generates a landslide susceptibility map, classifying over 35% of the study region's land area as having a high probability of landslides. The model's accuracy is assessed using the receiver operating characteristic (ROC) method, which yields an area under the curve (AUC) value of 0.77. This indicates that the LNRF model exhibits good predictive performance, resulting in a reliable landslide susceptibility map.