Soft computing and GIS for landslide susceptibility assessment in Tawaghat area, Kumaon Himalaya, India

摘要: This paper mainly presents a case study of landslide vulnerability zonation along Tawaghat-Mangti route corridor in Kumaon Himalaya, India. An attempt is made to predict susceptibility using back-propagation neural network (BPNN) and pro- pose suitable model for that zone, which can be successfully implemented the prevention slides. Various affecting parameters such as lithology, slope, aspect, structure, geotechnical properties, land use, inventory, distance from recorded epicenter are used susceptibility. The database on above derived satellite imageries, topographic maps, field work integrated GIS generate an information layer. Database this layer train, test, validate BPNN model. A three-layered with input layer, two hidden layers, one output found optimal. developed demonstrates promising result, prediction accuracy has been 80 % field.