Assessment and mitigation of tangible flood damages driven by climate change in a tropical city: Hat Yai Municipality, southern Thailand

Abstract

Climate change-induced floods in tropical urban areas have presented a serious global challenge because of failed conventional stormwater management practices. This research aims to develop a comprehensive methodological framework for flood damage estimation and mitigation, particularly in a tropical urban city. In this study, interdisciplinary fields were integrated through statistical downscaling, hydrologic-hydraulic modeling, and the development of flood damage curves. Relationships between tangible flood damage and flood-borne outbreak with flood depths were elucidated to predict future damage. Various flood mitigation strategies were evaluated. Herein, Hat Yai Municipality in Southern Thailand was selected as the study area. The flood simulation was conducted for 2010 and the highest flood damage sensitivity was exhibited by non-commercial buildings due to significant commercial stock damage, which was followed by that observed for detached houses. There was a strong linear relationship between the number of patients infected with leptospirosis and flood depth (R2 = 0.85). For climate change studies, flood maps for storms with 20-, 50-, and 100-year return periods under the A2/RCP8.5 scenario were generated using hydrological-hydraulic 1D/2D model; these maps were applied with the developed flood damage curves for damage estimation. It was found that reducing flood damage by implementing agroforestry and expanding the main bypass channel provides comparable damage reductions of −25.5% and − 27.5%, respectively, under the worst-case scenario of a 100-year return period in 2040–2059. Therefore, to deal with uncertain climate change situations, the incorporation of structural and non-structural measures is recommended. Such a combination when coupled with an eight-hour flood awareness time can result in a damage reduction of −59.9%. A flood warning system was in high demand by residents in the area; however, damage reduction from this measure alone was not high (approximately −17.0%) when compared to that obtained with other measures; consequently, additional measures were needed.