| Background:The Urban heat island effect(UHI)is a regional climate change phenomenon mainly caused by human aggregation activities.In the context of global warming and accelerated urbanization,the UHI is becoming obvious.From a health perspective,the UHI increases the risk of heat-related mortality and morbidity.However,previous studies focused on the heat-related mortality variation between different cities,and lack of directly quantified the magnitude of the UHI on mortality within a city.In this study,Guangzhou,a highly urbanized city,was selected as a research site to explore the temporal and spatial distribution characteristics of UHI.The attributable fraction and the number of mortalities from UHI were quantified on a fine-scale spatial scale,and high-risk areas were identified.This will help to accurately assess the risk of the UHI and provide a scientific reference for relevant departments to make scientific disaster prevention decisions and optimize health resource allocation.Methods:This study first used the k-means algorithm to spatially cluster 161 streets/towns in Guangzhou,and scientifically merged areas with fewer deaths.Subsequently,the inverse distance weight interpolation method was used to obtain the average daily temperature in each cluster area from 2011 to 2017,and the temporal and spatial distribution characteristics of temperature were analyzed.The central and suburban areas of the city were divided by the population density,and the UHI intensity(UHIi)were calculated and the temporal and spatial distribution characteristics of UHI were analyzed.Based on the temperature and non-accidental death data,the distribution lag nonlinear model was used to analyze the exposure response relationship between temperature and non-accidental death in Guangzhou,urban center region,suburban area and each cluster area of the center region.Combined with the UHIi,the attributable fraction and the number of mortalities from UHI in the total urban center region and each central cluster-area were calculated during high-temperature season(Jun.-Sep.).Results:During 2011-2017,the total number of non-accidental death in Guangzhou was 298 990,the average daily death toll was 117.04,and the average weekly death toll was 819.15.Based on 30 indicators,161 streets/towns in Guangzhou were clustered into 105 cluster-areas.The average daily temperature in Guangzhou was 22.86±6.27°C,which showed the variation characteristics of winter and spring lower than summer and autumn.The average daily temperature difference of central cluster-areas was 2.85°C,and the spatial characteristics as follows: the northern part of Guangzhou is lower than the south,and the area with the highest temperature mainly concentrated in the central-western parts of Guangzhou.The daily UHIi in the urban center of Guangzhou has no obvious periodic characteristics.However,the UHIi in the low-temperature season(Dec.-Mar.)showed the growth trend.The cluster-areas with high UHIi were mainly concentrated in the central-western parts of the urban center region.The highest is 1.13±0.34°C,which located in the cluster area of Nanyuan Street.The minimum mortality temperature(MMT)in the total city,urban center and suburbs of Guangzhou was 24.90 °C,25.50 °C and 23.50 °C,respectively.Above this temperature,the risk of non-accidental death of residents increases 1.55%(95% CI: 0.78%,2.32%),1.46%(95% CI: 0.65%,2.28%)and 2.00%(95% CI: 1.05%,2.08%)for per degree increase in weekly average temperature.The risk of death among different cluster-areas were quite different.Among the 46 statistically significant areas,the highest one is the cluster-area of Dasha Street + Wenchong Street,which showed 26.46%(95% CI: 9.16%,46.51%)increased risk for per degree increase in weekly average temperature.The attributable number of mortalities resulting from UHI in the urban center in high-temperature season of Guangzhou was 565 cases(95% CI: 251 cases,882 cases),and the percentage of attribution was 1.04%(95% CI: 0.47%,1.63%).Among the 46 cluster-areas in the central region,which performed statistically significant,the average number of deaths due to UHI was 28.85±26.00.The highest three areas were the cluster-area of Shiqiao Street,Nanshi Street + Xingang Street and Yayao Town + Jianggao Town + Xinhua Street,with 121 cases(95% CI: 54 cases,192cases)? 95 cases(95% CI: 48 cases,143 cases)and 91cases(95% CI: 54 cases,130 cases),respectively.The average attributable fraction due to UHI in each cluster-areas was 3.82+2.24%.The highest three areas were the cluster-areas of Dasha Street + Wenchong Street,Jiahe Street + Junhe Street and Shiqiao Street,with 10.12%(5% CI: 3.64%,17.10%),9.82%(95% CI: 0.72%,19.84%)and 8.77%(95% CI: 3.93%,13.88%),respectively.Conclusions:The temperature,urban heat island effect,heat death risk and the attributable fraction and attributable number of mortalities resulting from UHI during high-temperature season in Guangzhou were found to be spatial variation,which make the geography-targeted UHI effect reduction strategy be important for improving public health in the climate change scenarios. |
PDF Full Text Request