Explicit Spatializing Heat-Exposure Risk and Local Associated Factors by coupling social media data and automatic meteorological station data
Zheng Cao,Zhifeng Wu,Shaoying Li,Guanhua Guo,Song Song,Yujiao Deng,Wenjun Ma,Hui Sun,Wenchuan Guan
DOI: https://doi.org/10.1016/j.envres.2020.109813
Abstract:Extremely high temperatures, a major cause for weather-related public health issues, are projected to intensify and become more frequent. To mitigate the adverse effects, a low-cost and effective risk assessment method should be developed. Therefore, we applied automatic meteorological station data and population mobility data to develop a high spatiotemporal resolution temperature risk assessment method. The population mobility analysis results showed the working/residential complex pattern in Tianhe District, with hotspots of spatial clustering located in the north, southwest, and southeast of the study area. Taking the population mobility patterns into consideration, high-temperature risk assessment results with a resolution of 100 m were obtained. The total mortality cases in 2014 and 2015 were used to validate this result. The validation showed that the total mortality in the high-temperature risk areas accounted for over 36% of that in Tianhe District. Thus, the method introduced in this study is capable of reflecting weather-related risk. Furthermore, the high-temperature risk assessment results showed that most of the risky areas were located in the southwest of the study area. Two peak times of the risk areas were determined, being before dawn and in the evening. Compared with the risk areas during weekdays, those at weekends expanded. In addition, we used the geographically weighted regression model to investigate the potential influencing factors. Individual factor contributed more than 22.4% to the spatial distribution of heat exposure. Catering services, transportation services, and living services were higher than others, with mean R2 values of 0.28, 0.23, and 0.25, respectively. More than 47.9% of spatial distribution of heat exposure was attributed to joint function of influencing factors, with global R2 ranged from 0.23 to 0.34. Our research introduces a spatial-specific method to quantitatively assess high-temperature risk. Moreover, the mechanisms behind the spatial distribution of the high-temperature risk were discussed. The theoretical and management implications can help urban designers and energy governors to develop useful strategies to mitigate weather-related public health risks.