Landslide susceptibility evaluation based on BPNN and GIS: a case of Guojiaba in the Three Gorges Reservoir Area

摘要: A landslide susceptibility evaluation is vital for disaster management and development planning in the Yangtze River Three Gorges Reservoir Area. In this study, with support of remote sensing Geographic Information System, 4 factor groups comprising 10 separate subfactors landslide-related data layers were selected to establish a model based on back-propagation neural network including slope, aspect, plan curvature, strata lithology, distance faults, land use/land cover, Normalized Difference Vegetation Index, Water from roads, effect rivers. During development, three-layered interconnected structure input layer × 20 hidden layer × 1 output layer was used evaluating Guojiaba. At same time, algorithm applied calculate weights between layer. The results showed that slope has highest weight value 0.2051, which more than two times other factors, followed by lithology 0.1213 then rivers 0.1201. end evaluation, area divided into four zones such as very high, moderate low susceptibility. For verification, receiver operating characteristic curve network-derived drawn, under 0.8790 prediction accuracy 88%. Furthermore, obtained article verified comparing existing historical multiple field-verified results. Lastly, map will help decision makers risk management, site selection, planning, design control engineering.