Climate Change-Induced Regional Landslide Hazard and Exposure Assessment for Aiding Climate Resilient Road Infrastructure Planning: A Case Study in Bagmati and Madhesh Provinces, Nepal

Abstract

Nepal’s hilly and mountainous regions are highly susceptible to landslides triggered by extreme precipitations. The prevalence of such landslides has increased due to climate change-induced extreme hydro-meteorological conditions. These recurring landslides have significantly impacted the road transport infrastructure, which is the economic lifeline for cities and socio-economic mobility of rural communities in the hilly and mountainous regions of the country. This study modelled extreme rainfall scenarios for the current 1976–2005 baseline and future horizons of 2030, 2050, and 2080 to develop high-resolution 1 km × 1 km mean precipitation datasets under RCP4.5 and RCP8.5. Based on these extreme precipitation scenarios, we developed high-resolution landslide hazard models adopting integrated weighted index by combining the Frequency Ratio (FR) and Analytical Hierarchical Process (AHP) methods using multi-variate factors. The multi-variate factors included three terrain parameters—slope, aspect, and elevation; two soil parameters—lithology and soil type; two Euclidean distance parameters from the likely sources—distance from the lineaments and distance from the stream/river; an anthropogenic parameter—land use; and the climate parameter—the mean annual rainfall for four-time horizons and two RCPs. These parameters were spatially modelled and combined using the weighted overlay method to generate a landslide hazard model. As demonstration case studies, the landslide hazard models were developed for Bagmati and Madhesh provinces. The models were validated using the Receiver Operating Characteristic curve (ROC) approach, which showed a satisfactory 81–86% accuracy in the study area. Spatial exposure analysis of the road network assets under the Strategic Road Network (SRN) was completed for seven landslide hazard scenarios. In both Bagmati and Madhesh provinces, the exposure analysis showed that the proportion of road sections exposed to landslide hazard significantly increases for the future climate change scenarios compared to the current baseline scenario.