Landslides hazard, vulnerability and risk mapping in the data-poor region of northern Pakistan

Abstract

Landslides are devastating geological disasters in mountainous terrains around the world. The Chitral Valley, situated in the Eastern Hindu Kush Mountain ranges, is highly exposed to recurrent landslides posing a significant threat to urban growth, society, and the surrounding ecosystem. To assist in understanding and mitigating the landslide impacts, this study utilized statistical and geospatial modeling tools to develop a landslide hazard map and subsequently evaluate the vulnerability of the infrastructure and settlements to derive a risk map. A comprehensive landslide inventory was developed through on-screen digitization and field verification and compared with the predominant causative parameters to establish a landslide susceptibility index (LSI). The LSI map was analyzed with the precipitation and peak ground acceleration (PGA) data to generate the landslide hazard index (LHI). A semi-quantitative multi-criterion evaluation (MCE) technique was applied for the vulnerability assessment. Physical, environmental, and social vulnerability indexes were determined using direct and pairwise comparison matrices and integrated into a landslide vulnerability index (LVI). The landslide risk index map was developed by combining the hazard and vulnerability. The findings reveal that 37.25% of the region is categorized as shallow risk, with the remaining sections classified as low (62.22%), moderate (0.37%), high (0.11%), and very high risk (0.05%). The study findings are critical for the local authorities and policymakers to develop plans and policies for landslide-induced disaster risk reduction, landuse planning, and public involvement to reduce the vulnerabilities.