Research On Economic Loss Assessment Of Urban Rainstorm Flood Disasters

With the acceleration of climate warming and urbanization,urban rainstorm flood disasters have become increasingly severe and frequent,causing huge economic losses to the world,especially for China.With the convening of the Third World Conference on Disaster Reduction and the formulation of the Sendai Framework for Disaster Risk Reduction in 2015-2030,governments around the world have fully realized the importance of reducing disaster risks and formulating scientific disaster prevention and mitigation plans.A comprehensive understanding of the economic impact of disasters is an important part of carrying out disaster risk analysis and formulating effective disaster prevention and mitigation strategies.It has become a hot and difficult issue in the field of meteorological disaster research.In the context of the in-depth development of economic integration,the interrelationship between industrial sectors has deepened.Therefore,to fully reflect the impact of disasters,we must not only pay attention to the direct economic impact,but also measure its indirect economic losses.Under the realistic background of the lack of basic disaster data and the accumulation of multi-source data,how to establish economic loss assessment models of urban rainstorm flood disasters based on the perspective of data fusion,to assist government departments and industrial sectors to improve disaster response speed and formulate disaster prevention and mitigation strategies in time,has become an urgent and practical problem.This study focuses on the economic loss assessment of urban rainstorm flood disasters based on data fusion,constructs the loss assessment system,and discusses the direct and indirect economic losses respectively.Among them,the former is measured through social media big data,and the latter is calculated by using Input-Output Model(IO)and Computable General Equilibrium Model(CGE).On this basis,this study incorporates the disaster resilience factor to improve the direct and indirect economic loss assessment models,and to expand and optimize the loss assessment methods.Finally,this study sorts out the research conclusions systematically,and proposes countermeasures and suggestions to reduce the losses of urban rainstorm flood disasters.The main research contents are as follows:(1)Using social media big data(Weibo),this study develops an assessment model for rapid loss assessment of urban rainstorm flood disasters,conducts an empirical analysis of an urban rainstorm flood disaster that happened in Chongqing in August 2020 as an example,and describes the whole process from loss text extraction to disaster loss assessment.First,the physical losses caused by disasters are divided into five categories,including affected people,affected productive activities,damaged houses/buildings,lost properties,and damaged infrastructure.Based on the results of physical losses,the overall severity of waterlogging in each region is measured.Second,the labor factor is an important resource.In addition,emotional damage is also an important part that needs attention.Based on this,machine learning classification algorithms are used to analyze the degree of emotional losses of individuals in various regions,and filter out vulnerable victims emotionally.(2)Based on the CGE model and IO model,a comparative analysis of the indirect economic losses of urban rainstorm flood disasters is conducted.The CGE model and IO model are commonly used methods to assess indirect economic losses,but these two methods have their own advantages and disadvantages.The comparative study of the two methods can provide a more accurate indirect loss interval and reduce the uncertainty of assessment results.However,the comparative study of the two methods mainly stays at the theoretical level.Taking the "7.21 severe rainstorm" in Beijing in 2012 as a typical case,based on the same disaster impact scenario,that is,starting from the disaster reduction of direct losses in agriculture and transportation,this study introduces the disaster impact module and disaster impact parameters of agriculture and transportation,and uses these two methods to calculate the indirect economic loss rate of industrial sectors and the output losses of the industrial economic system,and then provide a range of disaster indirect losses.(3)The resilience factor is an important factor to be considered in disaster loss assessment.However,there is little literature that quantitatively measured the effect of resilience factor on reducing disaster losses.Based on this,this study incorporates the disaster resilience factor to improve the direct and indirect loss assessment model.In the direct economic loss assessment model,taking the flood disaster in Chongqing in August 2020 as an example,by evaluating the recovery speed of various regions of the disaster-stricken city from the impact of the disaster,the resilience performance of each region under the disaster scenario is measured.In the indirect economic loss assessment model,taking the "7.21 severe rainstorm" in Beijing in 2012 as an example,the disaster resilience factor is included in the production module of the CGE model,and the change in the elastic parameter value caused by the reduction of labor and capital factors in the production function is examined,to evaluate the extent of resilience factor to reduce the indirect impact of disasters quantitatively.The results of this study are as follows.First,real-time information on the direct losses can be obtained from social media data,which enables us to assess the impact of disasters in real time and dynamically,and can provide support and a basis for real-time response to disasters.In the Chongqing flood,Yuzhong District,Yubei District,Nanan District,Shapingba District and Jiangbei District suffered the most serious effects.Morever,local residents and students tended to express more negative emotions.Second,compared with the IO model,the construction principle,evaluation process and data requirements of the CGE model are more complicated,and the simulation accuracy of the actual economic situation is also higher.Thus,the IO model is suitable for rapid assessment of the indirect economic losses,and the CGE model is more suitable for comprehensive analysis of indirect economic losses.Third,based on social media data,we can quickly explore and monitor the recovery of the disaster-stricken regions from the perspective of affecting public life.As of 23 August 2020,the disaster scenarios such as water outage,power outage,and traffic conditions caused by the Chongqing flood have basically recovered.Fourth,the loss rate of various economic indexes has been reduced after considering the resilience factor.For example,the reduction rate of residents' income is 2.517%,the reduction rate of total investment is 3.940%,and the recovery degree of service sectors such as wholesale and retail industry are the highest.The main innovations of this study are as follows.First,this study conducts research on the direct economic losses of urban rainstorm flood disasters based on social media,and expands the general model of disaster loss assessment.Second,on the basis of analyzing the mechanism of mainstream indirect economic loss models systematically,namely CGE model and IO model,the interval of disaster indirect losses is proposed for the first time based on case analysis,which enhances the accuracy of assessment results.Third,this study incorporates the disaster resilience factor,improves the direct and indirect loss assessment model,and expands the scope of application of the models.