Landslide

Landslides are one of the most recurrent natural hazards occurring each year in the hilly and mountainous regions of Nepal causing massive loss of life and property. Natural hazards such as landslides cannot be avoided completely but the processes and consequences can be mitigated. The main objective of the study was on the application of Geographic Information System (GIS), and statistical calculations for landslide susceptibility modeling of Gulmi District, western Nepal. The models were derived using two different statistical approaches including Frequency Ratio (FR) and Shannon Entropy (SE). A landslide inventory of the Gulmi district was developed. The landslide inventories were used to derive the quantitative relationships between landslide occurrences and landslide causative factors. In this study, ten landslide influencing factors were used which include slope, aspect, curvature, lithology, geology, land use land cover, distance from the river, distance from the road, and distance from fault and soil type. Individual factor maps were prepared as thematic layers. After determining the weights of each class from the proposed two models, the landslide susceptibility maps were ready with five classes (very low hazard, low hazard, moderate hazard, high hazard and very high hazard) using GIS. The values of Area Under Curve (AUC) of success rate for FR and SE methods were found to be 81.8% and 80.6% respectively. The model shows that more than 15 % of the area falls under low and very low susceptibility level while 44% of the area has a high probability of landslide occurrence. The result of the present study indicates that integration of GIS has increased the quality and effectiveness of the overall process of susceptibility modeling and prediction mapping. To enhance the planning strategies for disaster mitigation and ensure sustainable development a reliable landslide hazard forecasting and risk assessment is a key component.