Reliability-Based Flood Management in Urban Watersheds Considering Climate Change Impacts

Abstract

Research has revealed the impacts of climate change on the hydrologic cycle. These impacts are intensified in the urban water system due to special characteristics of urban watersheds, such as a high percentage of impervious areas and the presence of a microclimate. A slight change in rainfall intensity and duration can cause severe floods, with a significant incremental increase in life and property losses. Therefore, it is a key issue to modify the urban watershed characteristics in a systematic way to deal with these probable changes. In this paper, a multicriteria optimization model is developed to select best management practices (BMPs) for flood management in urban watershed systems. The addressed criteria include maximizing reliability of the drainage system, as well as minimizing flood damages and BMP costs. A new reliability index for dealing with floods is suggested in this study that considers both severity and duration of flooding. The optimization model is solved using a two-stage multicriteria decision making method which employs the nondominated sorting genetic algorithm II (NSGA II) and data envelopment analysis (DEA) for the selection of best alternatives. The performance of the suggested model is tested in an urban watershed located in the northeastern part of Tehran, the capital of Iran. The drainage system of this watershed is simulated using the storm water management model (SWMM) developed by United States Environmental Protection Agency (USEPA). The climate characteristics of this region are simulated under a climate change scenario based on Hadley Center Coupled Model, version 3 General Circulation Model (HadCM3 GCM) outputs. The rainfall events, with flooding potential, are determined and the optimization model is used to select the best BMPs for flooding conditions. The results show the important value of using the proposed algorithm for reducing flooding volume and increasing the drainage system’s reliability. The results show further investigations are needed for real case applications to incorporate the uncertainties in climate change impacts.