Machine learning and GIS-based multi-hazard risk modeling for Uttarakhand: Integrating seismic, landslide, and flood susceptibility with socioeconomic vulnerability

The increasing complexity of multi-hazard risk (MHR) demands a comprehensive understanding of how multiple natural and human-induced hazards interact. These can occur simultaneously or sequentially within a vulnerable region, potentially amplifying their impacts. Uttarakhand, situated in the Indian Himalayan region, is highly prone to MHR due to topographical, meteorological, hydrological, socio-economic, and environmental factors. This study presents a district-level MHR assessment for Uttarakhand, integrating seismic, landslide, and flood susceptibilities with socio-economic (i.e., population density, vulnerable population, literacy rate, and employment rate) and built environment (i.e., building density and road density) vulnerabilities. Employing machine learning algorithms, individual hazard maps are generated and overlaid to prepare multi-hazard susceptibility maps, which are further reclassified into multi-hazard index (MHI). The vulnerability index (VI) was quantified using the Analytical Hierarchy Process (AHP), and the multi-hazard risk index (MHRI) was subsequently calculated by integrating MHI and VI. About 21.01 % of the total area of Uttarakhand faces simultaneous risks from earthquakes and landslides, with 22.36 % classified as being in high to very high MHR zones. Tehri Garhwal and Rudraprayag districts are identified as highly vulnerable and susceptible to multi-hazard risk. Unlike traditional single-hazard assessments, the present study comprehensively assesses overlapping hazard zones to understand multi-hazard dynamics. These findings offer critical insights for disaster risk reduction, informing strategies for economic planning, risk financing, and land-use planning to enhance the resilience of infrastructure and communities in Uttarakhand. • Integrates seismic, landslide, and flood risks with socio-economic vulnerabilities. • Machine learning coupled with remote sensing for precise multi-hazard risk mapping. • Identifies 22.36 % of Uttarakhand as high to very high multi-hazard risk zones. • Tehri Garhwal and Rudraprayag are most vulnerable to multi-hazard risks. • This study enhances resilience and disaster management and advances SDGs.