A Quantitative Approach for Improving the BIS (Indian) Method of Medium-scale Landslide Susceptibility

Abstract In India, the Bureau of Indian Standards (BIS) recommends a heuristic method for medium-scale (1:25,000/ 1:50,000) landslide susceptibility mapping. This is based on fixed ratings of geofactors, without the inclusion of landslide inventory information. In BIS method, the pre-defined ratings of geofactors are applied over diverse areas, irrespective of the terrain-specific spatial inter-dependence of geofactors and landslide types, which leads to rather moderate prediction. In this paper, we evaluate the effectiveness of the existing BIS method in Darjeeling Himalaya through a quantitative method adapting weights of evidence (WofE) modeling. The quantified spatial associations between specific geofactors for different landslide types and failure mechanisms that were generated, using this method showed improved prediction rates as compared to the BIS method of fixed ratings of geofactors. We therefore recommend adjusting the existing BIS guidelines by inclusions of weights, derived locally through quantitative spatial analysis of landslide inventories and geofactor maps.