Regional landslide susceptibility zonation utilizing bivariate statistical techniques in the northwestern Himalayas, Jammu and Kashmir, India

Abstract

This research focuses on assessing landslide susceptibility in the Jammu and Kashmir (J&K) region of the northwestern Himalayas, which is known for its high incidence of landslides. Utilizing advanced geographic information system (GIS) techniques, 18 influencing factors, including terrain characteristics, land use, rainfall, and lithology, were incorporated to create a comprehensive landslide susceptibility map (LSM). Leveraging a robust database comprising 6669 landslides, with 70% utilized for modelling and 30% for validation, the study utilized a Yule's coefficient (YC). The resulting LSM, categorized into five susceptibility zones, indicates that one third of the study area is highly susceptible to landslides, with 9.9, 23.9, 27.9, 23.1, and 15.2% falling into very high, high, moderate, low, and very low susceptibility zones, respectively. The model’s accuracy was validated with an 80.9% success rate through receiver operating curve (ROC) analysis. This LSM serves as a crucial tool for regional planning and management, providing valuable insights to mitigate landslide hazards. It facilitates informed decision-making and proactive measures and enhances resilience in landslide-prone areas, thereby contributing to the sustainable development and safety of the J&K Himalayan region. © Indian Academy of Sciences 2024.