Landslide susceptibility mapping of Darjeeling Himalaya, India using index of entropy (IOE) model

The assessment of landslide susceptibility is closely associated with the spatial distribution of previous landslides and landslide causative factors. In the present study, index of entropy (IOE) model was applied to map landslide susceptibility zones of Darjeeling Himalaya. This model is dealt with the relationship between landslide phenomena and landslide causative factors. To perform the model, 15 landslide conditioning data layers, i.e. elevation, aspect, slope, curvature, geology, soil, lineament density, distance to lineament, drainage density (DD), distance to drainage, stream power index (SPI), topographic wetted index (TWI), rainfall, normalised differential vegetation index (NDVI), and land use and land cover (LULC) were considered. Each and every class of landslide causative factor contributes a relative importance in landslide occurrences. To obtain the relative significance of each class of landslide causative factors and individual factor weight, Pij and Wj values of IOE model were estimated respectively to prepare landslide susceptibility map of Darjeeling Himalaya in GIS platform. The results suggested that the most important factor of landslide occurrences was soil type, whereas NDVI played least role in landslide susceptibility assessments. The prediction accuracy of the prepared landslide susceptibility map using ROC curve was 78.2%. Estimated frequency ratio value increased from very low to very high susceptibility zones, which showed the reliability and authenticity of IOE model.