Graduation Year

2024

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Qiong Q. Zhang, Ph.D.

Co-Major Professor

Yu Y. Zhang, Ph.D.

Committee Member

Mauricio Arias, Ph.D.

Committee Member

Hadi Gard, Ph.D.

Committee Member

Mahmood Nachabe, Ph.D.

Committee Member

Yi Qiang, Ph.D.

Keywords

Flooding control, Interdependency, Low impact development, Optimization, Stormwater management

Abstract

Flooding can cause extensive and devastating damage to human properties, including damage to homes, businesses, infrastructure systems, and personal belongings and it has become a threatening challenge during the past decades due to the massive damage brought to communities, infrastructures, and ecosystems. As climate change intensifies, the frequency and severity of flooding issues are expected to increase, exacerbating the existing flooding risks for vulnerable communities. Various approaches of modeling, optimizing, and designing for multiple levels of flooding problems have been adopted to deal with flooding control studies and research to alleviate the flooding risks on multiple levels of communities. This dissertation did research for evaluating the performance of urban stormwater drainage systems under future climate and land use conditions, including the stormwater runoff and flooding areas, applying simulation optimization approaches to figure out the optimal strategies to relieve the stress of increasing surface runoff and flooding areas, thus improving the performance of interdependent stormwater drainage systems and road transportation systems based on the interconnection between them.

In objective 1, the coupled effects of future rainfall and land use on urban drainage systems in terms of surface runoff quantity and flood area changes was investigated using EPA SWMM. Future downscaled and bias-corrected precipitation projections for 2040-2060 and 2080-2099 from “Downscaled CMIP3 and CMIP5 Climate and Hydrology Projections” Archive and future land use for 2050s and 2090s from EPA ICLUS project were used for the City of Tampa in Florida (USA). It was shown that potential runoff volume and flood area changes ranged from -40% to 160%, and -40% to 400% due to rainfall change, while potential changes induced by land use change ranged from 0% to 3.5%, and 0% to 18% respectively. Additionally, this study examined the coupled effects of future rainfall and land use changes on flood area variations, considering the capacity of drainage infrastructure and elevation variation of study site. Resualso,showed that flood area variations are not solely influenced by runoff quantity changes due to the two external drivers, but also by other factors. Specifically, in low-relief areas with inadequate drainage infrastructure and less elevation variation, these two drivers had no additive effects on the percentage of flood area changes. In contrast, in areas with adequate drainage infrastructure and greater surface elevation variation, there were additive effects from rainfall and land use changes on flood area expansions. These findings have important implications for stakeholders involved in city planning and flooding control, particularly for urban areas facing future precipitation and urbanization challenges.

In objective 2, the strategies for reducing surface runoff and flooding areas were investigated through multi-objective optimization (MOO). Low impact developments (LIDs) and pipe replacement were selected strategies to manage surface runoff and flooding in the optimization problem design. Six scenarios of rainfall intensities were designed to explore the trends and solutions in terms of minimizing surface runoff and flooding areas with different budgets for a study site and the adjustments were applied to those scenarios to reflect the future nonstationary conditions. It was found that there were non-linear increases for both runoff volume and flooding area reductions as more money was invested. The runoff volume reduction had a different trend from flooding area reduction among six scenarios and the highest reduction percentage of runoff volume occurred in 2.33y-24h scenario while the highest reduction percentage of flooding areas occurred in 25y-24h scenario. Through the calculation of benefit-cost ratios of optimal solutions under different scenarios, the results showed that the 25y-24h scenario has the highest benefit-cost ratio in terms of flooding areas reduction. After considering the future nonstationary condition, the current 25y-24h were revised and converted to 24-year design storm in the future as the future design reference for drainage system modeling and flooding control planning with limited budgets. The spatial analysis of LIDs implementation for different scenarios where the highest benefit-cost ratios were achieved, especially the results in the 25y-24h scenario, showed that the LIDs implementation was clustered in subcatchments with large available land areas and highly impermeable surfaces. The zoom-in map under the 25y-24h scenario also revealed that LIDs and pipe replacement working together could reduce flooded water depth and flooding areas efficiently.

In Objective 3, the interdependency between stormwater drainage and road transportation systems was considered and the visualizations for the road network structures under different design storm scenarios were presented. The spatial layouts of optimal strategies to enhance the performance of interdependent systems under disruptions were investigated through multi-objective optimization. LIDs and pipe replacement were selected solutions to manage surface runoff, flooding problems, and road network performance improvement in the optimization problem design. The co-location of flooding areas and overlapped road segments and their mutual effects were discussed under different design storms, including the topological structures and performance changes of road networks. The results showed that with the increasing storm intensity, more road segments were closed due to rising flooded water depth, and changed the road network structure, which contributed most to the decline of network connectivity, even worse, rendering the original OD demand matrix unsuitable for evaluation and causing network malfunctions. What was worse, due to the structure change, the road network performance had suffered from a steep increase in total travel time (TTT) and total travel mileage (TTM) in 25y-24h scenario. The performance enhancements after applying the proposed strategies suggested that those strategies were effective in resuming the functionality of road segments under disruptions and increased network connectivity. The LIDs near the flooded road segments with high vehicle flow could reduce the flooded water depth effectively and increase the capability of road systems to deal with flooding disruptions. Finally, distributing LIDs implementation worked better in enhancing the performance of road networks than the centralized LIDs implementation regarding TTT and TTM reduction benefits.

This dissertation comprehensively investigated the response of urban drainage systems under future rainfall and land use changes, explored the optimal LIDs and drainage pipe replacement solutions to improve the performance of urban drainage systems through simulation optimization, and provided strategies to advance the performance of interdependent stormwater drainage and road transportation systems. The methods and findings can help stakeholders make decisions in terms of stormwater management and flooding control considering future extreme storm events.

Download

Share

COinS