Landslide Susceptibility Assessment Using Bivariate Method: A Case Study from River Neelum and Jehlum Catchment Area

Abstract

Landslide is a frequently occurring natural calamity in the northern areas of Pakistan. The current study is aimed to assess the susceptibility of landslide hazard to highlight the vulnerable areas for the purpose of risk reduction along Neelum and Jehlum rivers in district Muzaffarabad. A data-driven predictive approach was adopted to conduct this study by using Weight of Evidence (WOE) model along with eleven conditioning factors. A spatial distribution map of landslides was prepared using orthophoto, previous records, and derivatives (hill shad, topographic openness, slope, aspect, curvature) of Digital Elevation Model (DEM). The results show that the roads, lithology, and rivers are the most important triggering factors for landslides in both valleys. Approximately 30% of the area is under low susceptibility zone in Jhelum valley while only 13% of the area falls under low susceptibility zone in Neelum river valley. In Jhelum river valley the medium susceptibility zone covers 35% of the total area whereas, Neelam river valley has 26% of the total area under medium susceptibility zone.  Around 61% of the land in the Neelam river valley and 35% of the land in the Jhelum river valley are under high susceptibility zone. The area under high hazard lies in the north-east of the district due to multiple conducive factors to trigger landslides including weak lithology (mudstone, sandstone, shales, and clays), high altitude along steep slopes and excessive precipitation (1800 mm/ year). Furthermore, the high hazard zone in study area is not suitable for construction purpose but was suitable for plantation. The validation result (89.41%) is justifying the performance of this model.