Multi-Source Geospatial Analysis for Disaster Risk Management in Smart Cities: Integration of GIS & Remote Sensing

Abstract. This study presents a geospatial framework for earthquake risk assessment in Türkiye’s Marmara Region, one of the country’s most densely populated and hazard-prone areas. Integrating multi-source datasets within a GIS and Remote Sensing (RS) environment, the approach synthesizes hazard, exposure, and vulnerability layers into a composite risk index at 100 m spatial resolution. Hazard modelling was conducted using fault proximity data from the General Directorate of Mineral Research and Exploration (MTA) and lithological susceptibility maps, both normalized and weighted to reflect seismic amplification potential. Exposure was quantified through demographic and infrastructural density, combining LandScan Global 2023 population data and OpenStreetMap (OSM) building footprints processed via kernel density estimation. Vulnerability was represented using building density as a proxy for structural fragility. All layers were normalized into a 0–1 scale and spatially aligned using GDAL-based resampling. The resulting risk map identifies Istanbul, Kocaeli, Bursa, and Sakarya as high to very high-risk zones, aligning with historical earthquake events such as the 1999 İzmit earthquake. Findings confirm that risk is driven not only by seismic hazard but also by demographic exposure and urban vulnerability. The proposed workflow demonstrates the applicability of open and national geospatial datasets in disaster risk management and offers a reproducible methodology for smart city resilience planning.