Landslide susceptibility index map generation based on geologic and geomorphologic factors using artificial neural network

This study focuses on generating landslide susceptibility index using artificial neural networks with the data obtained in Southern Japanese Alps. The training data include the geomorphic parameters (altitude, slope and aspect) generated using ASTER satellite images obtained during summer, winter and fall of 2004. The geologic parameters (rock type, distance from geologic boundary and geologic dip-strike angle) obtained both in areas with and without landslide are also used. The most significant factors affecting landslide occurrence were determined after the spatial analysis of the data. The artificial neural network structure and training scheme are formulated to generate the index. Data from areas with and without landslide occurrences are used to train the network. The network is trained to output 1 when the input data were obtained from areas with landslides occurrence, and 0 when the input data were obtained from sites without landslide. The trained network generates an output ranging from 0 to 1, reflecting the probability of landslide occurrence, based on the inputted data. Output values nearer to 1 means higher probability of landslide occurrence. The artificial neural network model is incorporated into the GIS software to generate a landslide susceptibility index map.