Knowledge-Driven Geospatial Techniques for Landslide Susceptibility Mapping: A Case Study in West Siang District, Arunachal Pradesh, India

Abstract

Landslides are frequent natural hazards that cause substantial damage to life, infrastructure, and ecosystems, especially in mountainous areas. The term "landslide" encompasses various processes involving the downward and outward movement of slope-forming materials, including natural rock, soils, artificial fills, or combinations of these materials. This study aims to delineate the landslide susceptibility zones in the West Siang District of Arunachal Pradesh, using a knowledge-driven heuristic approach. By integrating eight geo-environmental factors—lithology, landforms, lineaments, soil texture, drainage, land use/land cover (LULC), slope, and aspect—the weighted overlay method was employed to create a comprehensive landslide susceptibility map in a Geographic Information System (GIS) environment. The rank and weight of each factor, assigned based on expert knowledge, reflect their influence on landslide occurrence, with higher values indicating greater impact. The results categorised the area into five susceptibility classes: very low, low, moderate, high, and very high. The model validation, achieved by overlaying the historical landslide inventory, showed that 80.82% of the historic landslides fell within the high and very high susceptibility zones. The study is a valuable tool for planning, hazard management, and infrastructure development in landslide-prone areas.