Risk Assessment And Management Strategies Of Urban Rainstorm Waterlogging

With increased stresses from global climate change and urbanization, flooding events in urbanized regions occur more frequently due to the limited capacity of drainage systems, seriously affecting the development of social economy and public safety. Conventional urban management that focused on emergency response and past-impact analysis has moved to placing the emphasis on preparedness planning towards reducing overall flood risk. Risk assessment has become a hot spot of waterlogging prevention study because it provides theoretical support for the sustainable management of drainage system. However, urban drainage system is a complex, nonlinear system with a long service life, confronted with a wide range of uncertainties. This study aimed at investigating the risk assessment of urban storm-water drainage systems by using an example network in Anhui province. Meanwhile, the risk assessment was combined with rehabilitation decision-making, thereby yielding optimum rehabilitation strategies and reducing the damages associated with urban flooding, providing theoretical support and basis for making decisions for city’s sustainable development.1) A spatial fuzzy clustering algorithm was established for flood risk assessment of drainage systems with the purpose of discovering spatio-temporal distribution patterns of risks of the study region. This method can reduce the complexity of risk assessment and support differentiation strategies for different regions according to the characteristics of study area. The results showed that the spatial fuzzy clustering algorithm could discover the distribution of flood risks according to the spatial pattern analysis and the locations of risky areas under different risk levels were consistent with the actual situation recorded by the drainage office.2) To analyze different types of uncertainties in risk assessment of drainage system and improve the reliability of risk analysis, a new uncertainty analysis approach based on the maximum entropy principle was constructed to identify the most sensitive factors of risk assessment. The obtained result became more accurate with the larger integral interval and higher integral moments with a longer calculation time. Compared with Monte Carlo and Latin Hypercube sampling, the proposed approach can improve the calculation efficiency and the results meet the requirement.3) A rehabilitation strategy based on Low Impact Development (LID) and Best Management Practices (BMPs) was proposed taking the flood risk into account, and thus provides a scientific basis and theoretical support for stable operation and rehabilitation decision making of drainage systems susceptible to local flooding. The result showed that, the BMPs (storage pond) were effective for the storm events with higher return period while the impact coverage of storage pond was small. The storage located in the upper stream of disasters region with large volume achieved a better modification outcome. The LID measures were effective under the storm events with short return period with large impact coverage. Combination of LID with BMPs measures yielded the optimal strategy for reducing the urban flood risk.