Landslide Susceptibility Mapping and Vulnerability Assessment in the Badigad Watershed of Western Nepal

The landslide is a major and common natural hazard in the Nepalese Himalayas. This phenomenon is because of its active tectonic setting, extensively deformed rock formation, steep mountain terrain, frequent seismic events, high-intensity rainfall during monsoons, and high rate of weathering it occurs every year, especially during the monsoon season. The landslide susceptibility assessment using the Geographic Information System (GIS) and remote sensing tools helps to display the hazards, and vulnerable zones providing the required knowledge of the susceptibility of landslides in a specific region. Landslide susceptibility with vulnerability assessment covering the comprehensive parameters provides an important understanding of the areas that are more likely to the damaging effect of the landslide hazards and is useful for mitigation, management, and avoiding threats. A total of 339 landslides were identified in the Badigad watershed from the satellite images on Google Earth, out of which 20% were used for model validation. The study was carried out using the Weight of Evidence (WoE) model. Eleven factors were considered as possible causative factors for the hazard assessment. Based on the landslide susceptibility map, 11% to 29 % of the study area is found to be in the range of very high to shallow landslide susceptibility. From the final map prepared it is found that many areas of Satyawati Rural Municipality (RM), Ruru RM, Chandrakot RM, and Chhatrakot RM, along with Musikot Municipality in Gulmi and Badigad RM in Baglung district, exhibit a notable high hazard level. The Receiver Operating Characteristics (ROC) graph with the Area under the curve (AUC) value was used to check the performance of the WoE model which shows that the model has 80.4% prediction accuracy for future events which means, the model shows very good performance. The vulnerability within the study area is assessed by obtaining vulnerability scores. To obtain the vulnerability score, fourteen indicators were analyzed following the Local Disaster and Climate Resilience Planning (LDCRP) guideline. The result obtained from the vulnerability assessment showed that two wards, namely Thulolumpek in Satyawati RM of Gulmi and Bobang in Dhorpatan Municipality of Baglung are highly vulnerable. By analyzing and comparing the vulnerability obtained from susceptibility mapping (using GIS) and social methods (using LDCRP guideline), it is observed that they do not converge based on the factors studied.