Abstract:
Objective Against the background of rapid urbanization and global warming, Wuhan is frequently hit by extreme heat events, which not only poses a serious threat to the health status of local residents, but also brings great losses to socio-economic development. Mapping the risk of high-temperature disasters in the urban development area of Wuhan and analyzing the risk of high-temperature disasters and influencing factors at the local scale can provide an important basis for the prevention of high-temperature disasters in the city.
Methods Based on the “hazard-exposure-vulnerability” high-temperature disaster risk assessment framework proposed by the Intergovernmental Panel on Climate Change, we constructed an assessment system with utilizing multi-source data, and then pre-processed all the indicators to make them dimensionless. Then, a combination of the analytic hierarchy process and principal component analysis is used to assign weights to each indicator, and finally the weights of each indicator are superimposed to obtain the hazard map, exposure map and vulnerability map, respectively, on this basis, high temperature disaster risk map in Wuhan urban development area was synthesized to identify the distribution characteristics of high-temperature disaster risk in the study area. Then Landsat 8 remote sensing images with SAGA GIS software, Google Earth Pro software, and Random Forest algorithm were used to classify Wuhan urban development zones into 17 local climate zone (LCZ) types according to the remote sensing image classification method of World Urban Database and Access Portal Tools (WUDAPT). With 70% random samples used for drawing and 30% random samples used for checking, LCZ map that meet the requirements of classification accuracy are obtained and analyzed for site identification at the local scale. The LCZ map were superimposed on the high temperature disaster risk map to identify the local-scale high temperature disaster risk characteristics, analyze the degree of high temperature disaster risk for each LCZ type and the differences in high temperature disaster risk among different LCZ types, and explore the reasons for the risk differences. Finally, eight types of LCZ landscape pattern indices were preliminarily selected from the type and landscape scale levels, and the optimal study size was obtained using the moving window method in Fragstats 4.2 software, LCZ types with strong correlations were screened out under the optimal size, the multicollinearity of all LCZ landscape pattern indices was examined and those with multicollinearity were excluded. Finally, geographically weighted regression models were used to explore the effect of LCZ landscape patterns on spatial heterogeneity of heat hazard risk.
Results The characteristics of high temperature disaster risk in each district do not differ much, and the overall spatial presentation of the development center of each district gradually decreases from high to low, with high-risk areas mainly located in the south-central part of Caidian District, the west and north of Jiangxia District, the dense industrial parks in the south of East and West Lake Districts, the Wuhan Iron and Steel Factory in Qingshan District, and the Tianhe Airport in Huangpi District, and the low-risk areas are mainly in the watershed part. Jianghan, Qiaokou, and Qingshan districts have relatively high mean values of heat hazard risk due to high population density or dense buildings, while Wuchang and Hongshan districts have relatively low mean values of heat risk due to the presence of large areas of water and green areas within the districts. Overlaying the LCZ maps with the normalized high-temperature disaster risk maps, it can be seen that, among the building types, sparse built-up area (LCZ 9) has the lowest average value of high-temperature disaster risk value, and the average value of high-temperature disaster risk is significantly higher than that of the other building categories in large low-rise buildings (LCZ 8) and heavy industrial buildings (LCZ 10), which are mainly industrial plants and heavy industrial zones with large building areas. Among the natural environment types, waters (LCZ G) has the lowest average value of heat disaster risk, which indicates that water can effectively mitigate the risk of heat disaster, and the three types of bare rock (LCZ E), exposed sand (LCZ F), and construction building (LCZ H) are exposed to the outdoor, absorbing solar radiation for a long time and have higher values of heat disaster risk. percentage of landscape types (PLAND) has a higher degree of influence on the risk of heat disaster than aggregation (AI).
Conclusion Based on the results of the study, strategies to cope with high-temperature disasters are proposed. First, the area of vegetation and water should be increased. Secondly, the building layout should be rationally planned. At the same time, anthropogenic heat source emissions should also be controlled. Finally, high temperature service facilities should be improved to enhance the city’s coping ability.