Robust Decision Making In Mitigating Pluvial Flood Risk Under Climate Change Scenarios

With the intensification of global warming caused by climate change and the combined effects of social-environmental changes such as sea level rise,land subsidence,and rapid urbanization,the frequency and magnitude of the extreme inundations events in future coastal megacities may increase in the future,which imposes severe challenges on the urban safety and sustainable development.Traditional research method,which based on the idea of predict-then-act,pays lots of attention at the disaster risk assessment,yet neither rarely assess the relative effectiveness of alternatives for reducing flood risk under the uncertainty of future changing climate nor incapable to solve the decision making under deep uncertainty scenarios.This study builds a decision-making framework foundation by coupling the Robust Decision Making(RDM)and Dynamic Adaptation Policy Pathway(DAPP)under the internationally advanced theories of Decision Making under Deep Uncertainties(DMDU).The synthesized flood model of Shanghai Urban Inundation Model(SUIM)is developed to carry out the performance evaluation of adaptive solutions under future uncertain scenarios.The main research contents and results are as follows:(1)Summarizes the theoretical basis and research progress of decision-making methods under the background of deep uncertainty.Comprehensively analyzes the similarities and differences of the DMDU methods from the five dimensions: policy structure,scenario generation,alternatives generation,robustness and vulnerability.Three decision-making theories,RDM,Info-Gap Decision Theory(IGDT),and DAPP that are widely used in the domain of flood disasters management,are chosen to carry out the comparative analysis.The research framework is created by coupling the advantages of both RDM and DAPP which help to build the theoretical foundation.(2)Taking Shanghai as an example,the study simulates the historical "913" torrential rainfall event based on the Soil Conservation Service(SCS)model and verified the accuracy of the model's simulation.According to the research conclusion of climate prediction and urban social-environmental changes,three uncertain factors of future precipitation,urban rain island effect and decrease of drainage capacity are used to construct future extreme rainstorm scenarios via Latin Hypercube Sampling(LHS)method.The simulations of the future pluvial flood are conducted under future extreme rainstorm scenarios.The results show that the low-lying areas in the city center would always be the place that exposed to the huge risk in future scenarios.The maximum waterlogging depth in extreme rainfall scenarios may exceed 1.5m,and the accumulated depth in most affected low-lying areas has always been over 1m even in the mild scenarios.Correlation analysis shows that the contributions of future precipitation and urban rain islands are inapparent,while the decrease of drainage capacity will be the main influencing factor leading to future extreme waterlogging.(3)This study develops a comprehensive SUIM that integrates the process of flooding simulation,risk assessment,and solution evaluation.The distribution of disaster-bearing asset value is composed of the physical building damage,indoor property losses,and economic interruption caused by waterlogging.Risk modeling and risk assessment are carried out based on the “three elements” of hazard,exposure and vulnerability in future waterlogging scenarios.The results show that the risk of waterlogging caused by extreme rainfall in the future would be concentrated in the lowlying downtown commercial and residential areas,which are also asset-value-intensive areas.The severity of waterlogging risks and spatial inundation distribution behave differently among scenarios.The inundation area caused by the mild scenario is relatively small and the loss is acceptable,while the inundation area and loss caused by the medium and extreme scenarios are increased significantly.(4)The study assesses both the risk reduction rate and cost-effectiveness of all solutions.Three adaptation solutions,including increase public green area,drainage system improvement,and construction of deep tunnel are selected based on local urban drainage improvement plan and the process of knowledge co-creation discussed with policy makers and experts.The solutions are subsequently parameterized in the SUIM models.The performance of different solutions and their combinations under various scenarios are evaluated separately.The result shows that "drainage + green area + 30%absorption capacity of deep tunnel" can reduce the future flooding risk by 85%(±8%).The cost of each solution and their combinations are evaluated via Life Cycle Cost Analysis(LCCA),and the overall risk reduction rate was estimated as well.The results show that compared with the existing defense measures,the average risk reduction rate of the two solutions "30% absorption capacity of deep tunnel" and "70% absorption capacity of deep tunnel" have high benefits of disaster mitigation yet high construction costs.The "50% absorption capacity of deep tunnel" is therefore deemed as the highest cost-benefit ratio due to the high performance of risk reduction capabilities and relatively low-cost investment.(5)This study evaluates the robustness of all solutions and formulates the dynamic pathway.The Patient Rule Induction Method(PRIM),deemed to be the core idea of scenario exploration of RDM to analyze vulnerability scenarios,and the control standard of average risk reduction rate is used to evaluate the number of successful scenarios for each solution combination as the robustness measure.Both of them help to identify the sell-by date of the pathway plan through Tipping Point Analysis,weight the performance and cost-effectiveness of the measures,and formulate the path of adaptation countermeasures.The results of the study show that the two measures of "increase drainage capacity" and "increase public green area" failed to reach the 70%average risk reduction rate control standard.The robustness and failure time of "30%absorption capacity of deep tunnel"," drainage + green area","50% absorption capacity of deep tunnel","drainage + green area + 30% absorption capacity of deep tunnel" and "70% absorption capacity of deep tunnel" appears to be comparatively enhanced.In consideration of comprehensive multi-dimensional factors,"drainage + green area + 30%absorption capacity of deep tunnel" is a convertible pathway plan for short,medium and long-term adaptation strategy.This study develops the research method by coupling RDM and DAPP theory under the DMDU's robust decision-making framework,and build the comprehensive inundation assessment tools SUIM.The future extreme rainstorm scenarios are created,and then evaluate the cost-effectiveness of waterlogging prevention solutions in combining with the Shanghai local drainage improvement plan.The research results can not only provide scientific tools for Shanghai's adaptation strategic policies plan in coping with extreme weather and climate events under the background of climate change,but also provide both the theoretical foundation of decision-making methods and best practices to support the decision-making process for other coastal mega cities to adapt to the changing climate.