Landslide Hazard Assessment in Minjiang River Basin Based on GIS and Random Forest Algorithm

Disaster risk assessment is an effective tool for landslide disaster management. Machine learning models can effectively avoid the subjectivity of determining the weights of factors in disaster risk assessment, which is a hotspot of current research. We evaluate the risk of landslide disaster in Minjiang River Basin by Random Forest Algorithm based on characters include rainfall, slope, vegetation cover, distance to river and distance to fault. The results show that landslides in the Minjiang River Basin are closely related to the distance to the river and the distance to the faults. About 11.5% of the total area of the Minjiang River Basin is susceptible to small landslides.