GIS-based Quantitative Analysis Approach of Landslide Hazard Assessment: A case study of NH-513 from Pasighat to Mariyang, Arunachal Pradesh, India

Abstract

Hill slope failure along the roadways is a major problem in hilly and mountainous areas. Landslide hazard zonation maps produced in GIS environment can be utilized for planning and tackle of disasters associated with landslides. The aim of the study is to create landslide susceptibility maps along National Highway 513 from Pasighat to Mariyang, Arunachal Pradesh, India using quantitative approaches such as frequency ratio and Shannon entropy technique. Various factors causing landslides were taken into account for the study namely slope steepness, aspect, curvature, relative elevation, geology, terrain wetness index, lithology, drainage density, land use and land cover, distance to road, distance to water source distance to contour. Landslide susceptibility maps were generated by calculating the relationship between factors influencing landslides and locations of past landslide occurrences using frequency ratio and Shannon entropy technique. The results obtained are validated by receiver operating characteristic (ROC) curves. For the area from Pasighat to Mariyang along NH-513, the accuracy on the ROC curve was found to be maximum for frequency ratio i.e.0.748, while the prediction ratio for Shannon curve entropy turned out to be 0.736. The ROC curve for frequency ratio was determined to have the highest ratio among the two analysis techniques, positioning it as the optimal choice.