Abstract:

To enhance the fine-scale susceptibility assessment of landslide-collapse hazards, Paitan Town in Zengcheng District, Guangzhou, was selected as the study area. Based on the geographic information system platform, 2 579 slope units were delineated as the basic assessment units using the “catchment-area overlap method”. Eight key indicators, namely, elevation, slope gradient, slope aspect, slope shape, engineering geological rock group, geological structure, vegetation coverage, and land use type, were selected. The weights of these factors were determined by combining the information value model and the analytic hierarchy process, leading to the susceptibility zoning of geological hazards. The results indicated that the study area can be categorized into four susceptibility levels: high, medium, low, and non-susceptible, accounting for 8.59%, 41.31%, 25.52%, and 24.58% of the total area, respectively. The high-susceptibility areas were concentrated in the northern mountainous region, characterized by steep terrain, fragmented rock structure, and intensive human engineering activities, making them the priority areas for disaster prevention and control. The assessment results were validated using receiver operating characteristic curves, with an accuracy of 85%, indicating that the proposed method is highly applicable to the region and can support geological hazard prevention and territorial spatial planning.

Key words: slope unit, mountain hazards, information value model, susceptibility assessment