Background and Objective: Landslides are among the most common natural hazards, causing extensive economic, environmental, and human losses. This study aims to assess and map landslide susceptibility in the Siah Rud watershed located in Gilan Province, using a hybrid approach that integrates the Analytic Network Process (ANP) model with Geographic Information Systems (GIS).
Methodology: This applied, descriptive-analytical research utilizes the ANP model within the GIS environment to delineate landslide hazard zones in the Siah Rud watershed. Ten influential factors were selected for the analysis: land use, elevation, slope, slope aspect, vegetation cover, precipitation, geological formations, and distances from roads, rivers, and faults. The relevant thematic layers were derived using various data sources, including DEMs, Landsat 8 and Sentinel-2 satellite imagery, geological maps, and precipitation maps. Analytical tools in ArcGIS were employed to extract certain layers, such as distance from rivers and roads. The ANP model, which surpasses AHP in modeling interdependencies between criteria, was used for weighting and final analysis.
Results and Findings: According to the ANP weighting results, the variables with the highest influence on landslide occurrence were slope (0.27), geological formations (0.24), and precipitation (0.16). Spatial analysis further revealed that areas with steep slopes, weak volcanic and sedimentary formations, high rainfall, and proximity to faults and rivers had greater instability potential. In contrast, regions with dense vegetation, forest land use, and significant distances from roads and faults showed higher stability. The final GIS-based hazard zonation map classified the area into five risk levels, with 36% of the watershed falling into high and very high hazard zones. These high-risk areas were primarily concentrated in the upstream parts of the watershed, where the combination of steep topography, high rainfall, sparse vegetation, and unstable geological conditions contributes to widespread landslide susceptibility.