Urban Flood Simulation And Resilience Analysis Based On Multi-Source Information

In recent years,the frequency of urban flooding is increasing due to climate change and rapid urbanization.Flood disasters can not only result in loss of social economic production and bring threats to urban residents' life and health.In the face of growing impacts of urban floods,adaptation measures could help to improve the drainage capacity,such as the expansion of the drainage facilities.Apart from adaptation measures,conducting study on urban flood simulation and flood resilience could also improve the level of urban disaster prevention and mitigation,which is an important non-engineering measure to reduce urban flood disaster damage.Although urban flood simulation has played an important role in urban flood management at present,there is still a need to further study high resolution urban flood modelling.In addition,the research on flood resilience and the relationship between flood resilience and risk are relatively limited,and it is difficult to meet the needs of urban flood management.London and Dalian are chosen as the study area.This thesis studies high resolution urban flood modelling considering multiple information sources and urban features,regional scale flood resilience,grid scale flood resilience and the relationship between grid flood resilience and risk,and urban surface water flood risk management based on real options.The main research achievements are summarized below:(1)This thesis presents a new holistic framework for high resolution 2D urban flood modelling that utilizes information from multiple sources and takes into account the influences of critical urban features on flood propagation.The new framework integrates the methodology to address the key challenge in improving the accuracy of urban flood simulations,including extraction of flood information,handling of urban key feature and model assessment methods.First,photos were collected from social media featuring the flood event in the area of Wallington station at 14:50 on 7 June 2016 in order to extract flood information and set model parameters.Second,rainfall deduction approach and constant infiltration approach are studied when there is lack of soil infiltration data and drainage capacity data.At last,the urban features in terrain data are revised and how the urban feature representations in different terrain data settings affect flood modelling is investigated.The results show that the identified flood information obtained from social media is a useful source for setting the model parameters.The constant infiltration approach can describe the recession process better and should therefore be used for urban flood modelling.Urban micro-features,including buildings and underpasses,have significant influence on the inundation extent and depth.Urban flood risk analysis requires these important features to be identified and included in flood modelling.Based on the fast Cellular Automata Dual-DraInagE Simulation(CADDIES)flood modelling,the framework considering multiple information sources and urban features could improve the accuracy of urban flood modelling.(2)The approach to calculate regional scale flood resilience is proposed to identify low resilient sub-catchments for informed flood management decision making.The Siergou drainage area in Dalian was selected as the study area.First,the study area was divided into several natural sub-catchments using ArcGIS.Second,flood resilience of 31 sub-catchments were calculated with the CADDIES hydrodynamic simulation results,and uncertainty analysis of the relevant parameters and input data were investigated.The results show that the proposed method for regional scale flood resilience is feasible,and it can be used to identify the vulnerable regions.However,the proposed regional scale flood resilience is affected by the threshold of water depth,there is a need for decision makers to use reasonable threshold of water depth for flood resilience analysis.(3)In order to better manage urban floods and relieve urban flood problem,this thesis proposed a new approach for calculating urban grid scale flood resilience,and compared the grid flood risk and resilience.In the case of Waterloo in London,the flood resilience and risk of 2753,369 grids(horizontal precision 5 m x 5 m)in the study area were calculated using the CADDIES hydrodynamic simulation results.The results show that the proposed approach for grid flood risk and resilience can be used for the research of high resolution urban resilience and flood risk,which can help decision makers easily to identify spatial distribution of high risk and low resilience grids.Grid flood risk and resilience both show different relationships,i.e.,high risk-low resilience(?),high risk-high resilience(?),low risk-low resilience(?)and low risk-high resilience(?),which is helpful for urban flood management,improving the level of urban disaster prevention and reduction.The grid flood resilience is affected by the threshold of water depth when the grid performance is completely lost,so the decision makers should use the reasonable threshold to analyze the grid flood resilience according to the purpose.(4)This thesis develops a real options approach to assess Sustainable Urban Drainage Systems(SuDS)adaptation options(green roof,bio-retention,permeable pavement)in urban surface water flood investment and reconstruction under climate change.In order to reduce urban flood loss and fully utilize the rainwater resources,SuDS measures could be used to mitigate urban drainage systems.However,developing an adaptation option is challenging for long-term urban flood risk management due to uncertain future climatic conditions.This thesis uses a real options approach to reduce the economic loss brought by the uncertainty of climate change in urban surface water flood long investment and reconstruction.The trinomial tree model is used to calculate the changes of rainfall intensity and corresponding transition probabilities combined with UKCP09 climate change data.The Waterloo urban catchment in London is used as a case study,and the results from real options are compared to those from a fixed adaptation approach.The results show that the real options approach with higher net present value is more cost effective compared to the fixed adaptation approach,and the real options approach achieves a bigger advantage than the fixed adaptation approach with an increasing cost of adaptation measures.The real options approach is able to handle the uncertainty of climate change in assessing SuDS measures for surface water flood risk management.