Landslide Susceptibility Analysis Using Analytical Hierarchy Process and Frequency Ratio Method: A Case Study of Bhotekoshi Rural Municipality, Nepal

Abstract

Landslides are a recurring natural threat in Nepal, often causing significant harm to human life and infrastructure. This damage can be mitigated if the cause-and-effect relationships of the events are known. This study focuses on analyzing landslide susceptibility in Bhotekoshi Rural Municipality, an area acknowledged for its vulnerability to landslides. A landslide inventory map of the area was prepared using temporal information from Google Earth Pro over the past ten years. Approximately 56 landslides were identified and mapped, with 80% of them being randomly selected for model training, and the remaining 20% were used for validation purposes. To comprehend the factors contributing to landslides and predict future occurrences, landslide susceptibility mapping of this region was carried out using frequency ratio (FR) and Analytical Hierarchy Process (AHP) models. The data of slope, aspect, curvature, rivers, roads, geology, and landslides are used as causative factors for landslides. After the complete analysis, two different maps of susceptible areas for landslide based on the AHP and FR method are obtained. Finally, the results are compared and validated with the training data using the approach of Receiver Operating Characteristics (ROC) and Area Under the Curve (AUC). From the analysis, it is seen that both the models were equally capable of predicting the region's landslide susceptibility (AHP model (prediction rate = 0.610); FR model (prediction rate = 0.710)). The obtained landslide susceptibility map can serve as a major tool for engineers and planners to carry out development works in the study area.