Study On Vulnerability Characteristics Of Urban Waterlogging Disaster

With the acceleration of urbanization,waterlogging disaster has become an important factor restricting urban security and sustainable development.The research on urban waterlogging disaster vulnerability is of great significance for improving urban anti-risk ability and emergency planning ability of waterlogging disaster and realizing sustainable development of urban public safety and economic and social health.In order to provide a more systematic assessment method for urban waterlogging disaster vulnerability research and provide strategic value for urban waterlogging disaster prevention in the southern section of the Yellow River,this paper takes the cities in the southern section of the Yellow River as the research object,through literature collection analysis and frequency analysis,from the exposure,sensitivity and adaptability of 15 evaluation indicators selected to build an evaluation system,entropy method and factor analysis method to determine the index weight,based on BP neural network to build an urban waterlogging disaster vulnerability assessment model.On this basis,the embrittlement characteristics of the eight cities were analyzed from the three dimensions of exposure,sensitivity and adaptability by Arc GIS,and the spatio-temporal evolution of waterlogging disaster vulnerability in the eight cities was analyzed from the two dimensions of time and space change,and the following conclusions were drawn:(1)Factors affecting urban waterlogging disaster vulnerability can be classified into natural factors and social factors through regional analysis and sorting out related studies.Based on this,the evaluation system constructed by selecting 15 evaluation indexes from three aspects of exposure,sensitivity and adaptability is comprehensive and systematic,and can meet the requirements of comprehensive evaluation.(2)The assessment model of urban waterlogging disaster vulnerability built by BP neural network can be effectively applied to the study of urban waterlogging disaster vulnerability.From the output results of the model,the error of the evaluation results obtained is far less than the maximum error,and within the acceptable range,indicating that the BP neural network model designed in this paper has a certain effectiveness,and can be used to evaluate the vulnerability of urban waterlogging disaster.(3)Through the analysis of vulnerability exposure,sensitivity and adaptability of waterlogging disaster,it is found that the embrittlement characteristics of 8 cities are as follows:Zhengzhou,Xinxiang,Jiaozuo and Luoyang are high exposure and high sensitivity cities;Kaifeng and Jiyuan are medium adaptability and low exposure cities;Puyang and Sanmenxia are low sensitivity and low adaptability cities.(4)Through the assessment model and Arc GIS analysis,it is found that the temporal and spatial evolution characteristics of waterlogging disaster vulnerability in the 8 cities are as follows: From the temporal changes,the waterlogging disaster vulnerability index of the 8cities shows an overall decline and local fluctuation trend from 2011 to 2020.The fluctuation range of high-exposure cities is large,and the vulnerability to waterlogging disaster is unstable.The vulnerability of low-exposure cities to waterlogging disaster decreased significantly in the past ten years,indicating that the urban disaster prevention and mitigation system was constantly improving and progressing.The decrease rate of low-sensitivity cities is small,indicating that the progress of vulnerability reduction of cities to waterlogging disaster is not significant.In terms of spatial distribution evolution,from 2011 to 2020,the vulnerability of eight cities to waterlogging disaster changed from low adaptability to high exposure.The spatial accumulation effect of waterlogging disaster vulnerability was obvious,and the vulnerability characteristics generally showed a distribution law of “high vulnerability in the center and decreasing outward in the circle layer”.