Slope-Unit Scale Landslide Susceptibility Mapping Based on the Random Forest Model in Deep Valley Areas

Landslide susceptibility evaluation is critical for landslide prevention and risk management. Based on the slope unit, this study uses information value method- random forest (IV-RF) model to evaluate in deep valley area. First, based historical data, a inventory was developed by using remote sensing technology (InSAR optical sensing) field investigation methods. Twelve factors were then selected as input data model. Second, units with different scales obtained r.slopeunits method used susceptibility. Finally, spatial distribution characteristics of grade under optimal scale are analyzed. The results showed that unit when c = 0.1 3 × 105 m2, internal homogeneity/external heterogeneity 8425 extracted best, an AUC 0.905 F1 0.908. In case, accuracy highest well; it shown finer would not always lead higher results; necessary comprehensively consider homogeneity external units. Under scale, number landslides highly extremely susceptible areas map accounted 82.60% total landslides, which consistent actual landslides; shows method, combining model, can obtain high accuracy.