Robust Flood Risk Management Strategies Through Bayesian Estimation and Multi-objective Optimization

• Published in: International Journal of Environmental Research Vol. 15; no. 6; pp. 1057 - 1070
• Main Authors: Sobhaniyeh, Zahra, Niksokhan, Mohammad Hossein, Omidvar, Babak, Gaskin, Susan
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.12.2021; Springer Nature B.V
• Subjects: Bayesian analysis; Climate change; Decision analysis; Decision making; Earth and Environmental Science; Environment; Environmental Engineering/Biotechnology; Environmental Management; Environmental risk; Evolutionary algorithms; Flash floods; Flood control; Flood hazards; Flood management; Flooding; Floods; Geoecology/Natural Processes; Landscape/Regional and Urban Planning; Levees; Mathematical models; Multiple criterion; Multiple objective analysis; Natural disasters; Natural Hazards; Optimization; Parameter sensitivity; Performance assessment; Research Paper; Risk management; Risk reduction; Robustness; Rural areas; Sensitivity analysis; Uncertainty; Iran; Deep uncertainty; Bayesian estimation; Robust optimization; Uncertainty analysis; Flood risk control
• ISSN: 1735-6865; 2008-2304
• DOI: 10.1007/s41742-021-00370-w

Management and control of flood hazards, the most frequent natural disaster worldwide, has become a greater challenge due to the increasingly unpredictable precipitation and runoff due to climate change. As many rural areas in Iran are vulnerable to flash floods occurring mainly in the spring, more accurate plans are needed to help reduce the risk of related damage. To address this concern, a robust methodology using multi-objective optimization is proposed, which incorporates the large uncertainties in the modeling parameters defining the risk of flooding. The proposed framework has been implemented in the upper catchment of the Taleghanrood river in the Taleghan district in Iran, which is vulnerable to flooding. The results provide a detailed performance assessment of alternative infrastructure designs, which will help to increase the efficiency of flood management strategies. The optimization uses multi-criteria optimization evolutionary algorithms (MOEA) and Bayesian estimation concepts. The resulting specific design plans, as levees’ height increases over a 50-year time horizon, for controlling floods under given scenarios reflect the uncertainty in the parameters. Graphic Abstract Article Highlights A robust decision making model has been developed to address flood management scenarios. The proposed methodology addresses deep uncertainties in decision parameters. Sensitivity analysis of the plausible scenarios and discovers vulnerable scenarios has been done.