Landslide risk assessment: A case study of Varvasht Watershed, Lorestan province, Iran

Landslides represent one of the most destructive natural hazards worldwide, causing substantial economic losses and human casualties annually. The present study aims to address existing gaps by conducting a comprehensive landslide risk assessment in the Varvasht Watershed, Lorestan Province, Iran, using three machine learning models. The Varvasht Watershed is located in northwest Lorestan Provinces. A landslide inventory map was prepared based on ۱۶۳ landslide polygons systematically surveyed and recorded. Seventeen conditioning factors were selected as independent variables influencing landslide occurrence. The dataset was randomly split into training (۷۰% of the data) and validation (۳۰% of the data) subsets to ensure unbiased model evaluation. Three machine learning models such as Support Vector Machine (SVM), Generalized Linear Model (GLM) and Artificial Neural Network (ANN) were employed. All models were validated using ROC curves. Vulnerability scoring was conducted by integrating the inherent value of elements with hazard classes. The final risk map was produced using the risk equation. Validation results demonstrated that the SVM model achieved the best performance with an AUC of ۰.۹۱۳. According to the SVM model, approximately ۱۲.۶۸% as high, and ۱۱.۹۴% as very high susceptibility. The final risk map revealed that approximately ۱۹% of the area (equivalent to ۲,۹۶۸ hectares) is classified as high and very high-risk classes, necessitating prioritization in management and mitigation measures. The results of this study can serve as a scientific basis for land use planning, sustainable development, and landslide risk mitigation in the mountainous regions of Iran.

کلیدواژه ها:

Landslide risk assessment ، Support Vector Machine (SVM) ، Artificial Neural Network (ANN) ، Generalized Linear Model (GLM) ، Varosht Watershed ، Iran