Focusing on the main urban area of Badong County,14 static landslide factors were selected from four aspects,namely topography,geology,humanities and human engineering activities.Additionally,time series InSAR technology was employed to extract the surface deformation rate in the study area as the dynamic landslide factor.To eliminate the dimensional differences among different landslide factors,the factors were classified in ArcGIS and the frequency ratio(FR)of each grade was calculated,and the map of each factor was spatially correlated with landslides to construct a landslide susceptibility evaluation dataset.In addition,a random forest evaluation model was established to evaluate landslide susceptibility,and the accuracy of the model was analyzed.The evaluation results of landslide susceptibility were compared with and without deformation factors.The results demonstrated that introducing the surface deformation factor increased the AUC value of the model from 0.86 to 0.87,and the susceptibility zoning was more reasonable.The high landslide-prone areas in the urban area of Badong County were mainly concentrated along the river,while the low landslide-prone areas were characterized by dense vegetation and higher altitude,located farther from the river.Through time-series analysis of the deformation rate of typical landslides,it was observed that actively changing landslides exhibited slow deformation over time,necessitating closer monitoring.The landslide susceptibility evaluation method provided in this study can offer guidance for regional development and provide insights for related research.
landslide susceptibility evaluationInSAR deformation raterandom forestArcGISurban area of Badong County
NETL
NSTL
万方数据
