Aim: Landslides are major hazards in mountainous and semi-arid regions, causing serious damage to infrastructure, resources, and lives. This study aims to map landslide-prone areas in the Qarnaveh watershed, eastern Golestan Province, using the probabilistic Weight of Evidence (WoE) model within a GIS environment.
Material & Method: This study identified 179 landslides through field surveys and databases; 70% were used for modeling and 30% for validation. Fourteen conditioning factors including elevation, slope degree, slope aspect, distance to faults, fault density, distance to roads, distance to rivers, drainage density, slope curvature, stream curvature, geology, land use, LS index, and topographic wetness index were extracted and analyzed using ArcGIS and SAGA-GIS software. The significance of each factor was assessed through the WoE model, which was then used to generate a landslide susceptibility map for the watershed. Model accuracy was evaluated using the Receiver Operating Characteristic (ROC) curve.
Finding: Results indicated that drainage density had the highest influence, while fault density had the least impact on landslide occurrence in the study area. The southern and southwestern parts of the watershed were identified as zones with very high landslide susceptibility. The WoE model demonstrated an accuracy of 79.7% in predicting landslide-prone areas.
Conclusion: Based on the results, the Weight of Evidence model shows a reasonably high capability in identifying landslide-prone areas with acceptable accuracy. It can serve as an effective tool in natural resource management, watershed planning, disaster risk reduction, spatial planning, and reducing vulnerability in high-risk areas.
Innovation: This study employed the statistical Weight of Evidence (WoE) model along with a spatially explicit and comprehensive approach, incorporating key influencing factors to map landslide susceptibility in a semi-arid region. The results provide an effective tool for identifying high-risk landslide areas and prioritizing prevention, aiding natural resource, road, and disaster management agencies in damage reduction.