Machine Learning Models and Spatial Distribution of Landslide Susceptibility

Abstract

The present study is dealt with the preparation of landslide susceptibility map of the Balason river basin of Darjeeling Himalaya with the help of GIS tools machine learning model i.e. support vector machine (SVM) and artificial neural network model (ANNM). Fifteen landslide causative factors i.e. slope, aspect, curvature, elevation, geology, geomorphology, soil, distance to drainage, drainage density, distance to lineaments, lineament density, land use and land cover, stream power index (SPI), topographic wetness index (TWI) and rainfall were considered to produce the landslide susceptibility zonation maps. To generate all these factors map topographical maps, geological map, geomorphological map, soil map, satellite imageries, and google earth images were processed and constructed into a spatial data base using GIS and image processing techniques. SVM classification algorithm was performed for each factor by using the RBF kernel to prepare landslide susceptibility map. And, the back-propagation method was also applied to estimate factor’s weight and the landslide hazard indices were derived with the help of trained back-propagation weights using ANN model. Then, the landslide susceptibility zonation map of the Balason river basin was made using GIS tool and classified into five i.e. very low, low, moderate, high and very low landslide susceptibility. To validate the prepared landslide susceptibility map landslide inventory was used and accuracy result was obtained after processing ROC curve.