GIS Based Landslide Susceptibility Mapping-A Study from Darjeeling-Kalimpong Area, Eastern Himalaya, India

Macro scale (1:50,000) Landslide Susceptibility Map (LSM) of 290.33 sq km area in part of Survey of India Topographical sheet 78A/8 in Darjeeling district categorizes the hill slope into five susceptibility classes based on their estimated landslide susceptibility. The LSM was prepared by facet-wise integration of thematic maps on six causative geofactors (lithology, structure, slope morphometry, relative relief, landuse and land cover and hydrogeology) using a knowledge-based rating system proposed by Bureau of Indian Standards (BIS). The facet and related slope morphometry parameters were derived from the Digital Elevation Model (DEM) which was derived from elevation data using both raster and vector GIS techniques. The thematic maps of geofactors were prepared through detailed field studies and augmentation of existing database. The ARC/INFO 9.1 software was extensively used to handle a large volume of thematic database for multiple retrieval, operation and generation of LSM. The spatial distribution of thematic classes vis-à-vis landslide incidence of the area indicates (a) relative abundance of landslide area is maximum in case of phyllite-phyllitic quartzite-schist assemblages of the Daling Group and minimum in the quartzo-feldspathic gneiss of Central Crystalline Gneissic Complex (CCGC), (b) a slightly higher concentration of landslide incidences in and around the prominent regional thrust (MCT), (c) a progressive increase of landslides abundance in successive higher categories of slopes, (d) highest relative abundance of landslides in the "barren" landuse and landcover categories and (e) higher relative abundance of landslides in 'dripping' and 'flowing' categories of hydrogeological situations. The LSM database reveals that 18.08% of the studied area comes under high and very high susceptibility categories, while the moderately susceptibility class and low and very low susceptibility classes constitute 42.94% and 38.98% of the study area respectively. The landslide incidence map when validated against the prepared LSM indicates (a) no landslide incidence in very low susceptibility zone, (b) a progressive increase in the relative abundance values of landslide for successive higher categories of susceptibility zones and (c) high value (64.80) for high and very high susceptibility zones together. It is felt that factors like antecedent rainfall, erodability of the drainages, large-scale anthropogenic interferences are also important in inducing instability in an area. Therefore, locally, the present susceptibility status of the existing slope may undergo significant changes due to the effect of the above-mentioned factors.