Comparative urban flood hazard mapping using GIS-integrated multi-criteria decision analysis: A remote sensing approach for Colombo, Auckland, and Valencia

Urban flooding (UF) has increased significantly over the past decade, leading to substantial economic, social, and health losses. A key factor driving this escalation is the continuous urban development that converts open ground surfaces into impervious surfaces. Additionally, natural flood susceptibility factors (FSFs) such as slope, curvature, elevation, river proximity, soil types, drainage density, and rainfall also contribute to the rise of UFs. This study aims to assess the relative influence levels of selected FSFs using remote sensing data, including the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM), Landsat 8 Operational Land Imager (OLI), and real-time precipitation data. The ArcGIS software enabled a GIS-based multicriteria decision-making (MCDM) approach to identify urban flood hazard (UFH) zones in Colombo (Sri Lanka), Auckland (New Zealand), and Valencia (Spain). Thematic map layers representing topographical, geological, climatic, and hydrological features are utilized to evaluate UFH zones. This study uniquely applies a cross-regional hybrid Analytical Hierarchy Process (AHP) framework across three diverse global cities, which has not been explored in previous flood hazard studies. A hybrid process is utilized here to reduce the subjectivity and uncertainty of susceptibility weights. The resulting hazard maps classified study areas under four flood risk levels: Extremely High, High, Moderate, and Low, revealing that 13.64% of Colombo, 25.64% of Auckland, and 17.63% of Valencia fall under extremely high hazard levels for flooding. This approach demonstrated its effectiveness in identifying UFH zones and producing flood hazard maps, offering valuable insights for urban planners in implementing area-specific sustainable flood mitigation strategies.