GIS-MCDA Based Flood Hazard, Vulnerability, and Risk Assessment in the Lower Nilwala Floodplain, Sri Lanka

Floods are among the most severe climate-induced disasters that severely affect human life, the social and physical environment as well as ecosystem stability. This study aims to model flood hazard, vulnerability, and risk in the lower Nilwala River Basin, Sri Lanka using a geographic information system (GIS) integrated multi-criteria decision analysis (MCDA) approach. Twenty-two sub-criteria under two criteria domains were used to derive final flood indices coupling with the analytic hierarchy process (AHP) and weighted linear combination (WLC) method. The results revealed that hazard criteria play a more decisive role in risk evaluation than vulnerability criteria. Out of the total area 7.9% (32.7 km²) falls under very high and high flood risk categories while 48.7% (200.6 km²) is at moderate risk. Very high risk areas concentrated in and around Matara city since high population density with low elevation and slopes whereas Thihagoda DSD faces high risk. Majority of areas in Kirinda-Puhulwella, Akuressa, and Athuraliya DSDs are in very low to moderate risk areas. The model validation through AUC values of 0.71, 0.65, and 0.70 for FHI, FVI, and FRI has also proved that the reliability of findings. Thus, despite the inherent limitations of the GIS-MCDA based research that utilize remote sensing (RS) data findings offer valuable insights for flood hazard mitigation in the study area. It is recommend that to minimize limitations of these kind of research though integrating high-resolution RS data and improved comparison methods to achieve high accuracy in future studies.