Application of frequency ratio (FR) model in spatial prediction of landslides in the Balason river basin, Darjeeling Himalaya

Abstract

The present study aims to prepare landslide susceptibility map using frequency ratio (FR) model in the Balason river basin of Darjeeling Himalaya. Along with keeping in mind that, can the produced landslide susceptibility map yields acceptable landslide prediction accuracy or not. For that reliable landslide inventory map was prepared with a total of 295 landslide locations. The chosen landslide conditioning factors were altitude, aspect, slope angle, curvature, geology, geomorphology, soil, land use/land cover, normalized differential vegetation index, drainage density, lineament density, distance from lineament, distance to drainage, stream power index and topographic wetted index. To estimate FR value for each class of all the landslide conditioning factors pixels affected by landslide (%) and total pixels (%) were taken into account. FR model was applied to integrate all the data layers on GIS platform. The derived susceptibility map was divided into five categories i.e. very low, low moderate, high and very high which cover 17.52, 18.71, 30.35, 24.49 and 8.92% area of the basin respectively. The area under curve value of receiver operating characteristics indicates the prediction accuracy of the prepared map was 94.2% that is highly desirable. FR plots represent that there were positive relationship between landslide susceptibility classes and FR value. The prepared map will helping developers and planners to implement slope management plans, land use plans and other development action plans in this area.