An Application of Geospatial Based MCDM Technique to Identify Landslide Susceptibility Zones in the Ragnu Khola River Basin of Darjeeling Himalayan Region, India

Abstract

Several natural disasters are taking place on the earth, and landslide is one of them. Darjeeling Himalaya is one of the world's young fold mountainous area, often suffering from landslide hazards. Hence, the study identifies the landslide susceptibility zone in the Ragnu Khola river basin of the Darjeeling Himalayan region by applying the geospatial-based MCDM technique. This research's major goal is to identify whether this GIS-based multi-criteria decision-making (MCDM) technique is validated or not for landslide susceptibility zones (LSZ); if validated, then how much manifest for describing the LSZ in the study area. MCDM evaluation applies to determining weight value to integrate different thematic layers of river morphometry like Drainage Diversity (DD) parameters and Relief Diversity (RD) parameters. Both DD and RD have significant impacts on landslide intensity. Hence, both layers are combined using the analytical hierarchy process (AHP) of the MCDM technique for the final LSZ. The final result has been validated by ROC analysis using landslide occurring point data obtained from the Geological Survey of India (GSI). The outcome of the study shows that1.45% and 17.83% areas of the region fall in 'very high' and ‘high' LSZ, which belongs to near Mull Gaon, Sanchal forest, and Alubri basty. Most of the area (47.70%) is observed in 'moderate' LSZ. Only 1.32% and 31.7% are kept in ‘very low’ and ‘low’ LSZ, respectively, through the study area. The description capability of the technique for LSZ is significant as the area under the curve (AUC) is 72.10%. The validation of the study using the frequency density of the landslides (FDL) also indicates the 'very high' LSZ is associated with the maximum (2.19/km2) FDL. The work will be needful to develop the overall socio-economic condition of such kind of tectonically sensitive region by proper effective planning.