The Evaluation of Landslide Sensitivity using Frequency Ratio and Fuzzy Logic Models (Case Study: Khorramabad-Arak Freeway)

Landslide risk zoning plays a significant role in the development of safe and sustainable infrastructure, urbanization, land use, and environmental planning. Identifying and determining sensitive and landslide-prone areas not only prevents damages but also provides a basis for the implementation of slope stabilization plans. Landslide risk zoning is done through different methods including statistical, expert evaluation and definitive methods. Choosing appropriate zoning method depends on the type of analysis, study area, experts’ skills and knowledge, and the type of geological and geomorphic parameters affecting landslide risk.  The main goal of this study were preparing a landslide distribution map, identifying the factors affecting landslides, and zoning its danger in Khorramabad-Arak Freeway (Khorramabad to Boroujerd). Using satellite images and field studies as well as frequency ratio (FR) model, a landslide distribution map was prepared, and the factors influencing landslides including slope, lithology, slope direction, elevation classes, land use, rainfall, distance from fault factors and the network of waterways were analyzed. Fuzzy gamma (0.9) was used for zoning the landslide risk. The landslide hazard map was divided into very low (18.55%), low (30.67%), medium (26.51%), high (18.15%) and very high category (6.12%) and finally validated by ROC curve. The results of ROC curve analysis for Fuzzy GAMMA showed that the landslide sensitivity map in the study area has excellent predictive power with area under the curve of AUC=0.94. Therefore, it is suggested to apply the obtained results for freeway security and regional planning.