PM_2.5 Estimation Using Remote Sensing Data And Risk Assessment Of Population Exposure To PM_2.5 In Shanghai-Hangzhou Bay Area

Due to frequent occurrence of haze weather,fine particle pollution has seriously affected people’s physical health and normal life.With the rapid development of satellite remote sensing monitoring technology,it has become crucial means for scholars at home and abroad to study PM2.5 pollution because of advantages in wide coverage,high frequency of updates,fast and convenient access and so on.The reasonable assessment of exposure to PM2.5 pollution for residents is an important part of effective air pollution prevention and control.Therefore,this paper takes ShanghaiHangzhou Bay area as the research area,adopts MODIS L1 B remote sensing image data of 827 scenes in 2016 and Dark Target aerosol inversion algorithm,and remote sensing inversion results in aerosol optical thickness(AOD)with spatial resolution at 1km.Based on the measured data of PM2.5 concentration from environmental air quality monitoring stations,the optimal relationship model between AOD and PM2.5 was established,and the spatio-temporal distribution results of PM2.5 concentration in the study area were estimated by AOD.Combined with the demographic data of statistical yearbook and point of interest(POI)data,the spatial simulation of population density was carried out,and the population grid data of 2016 was obtained with the spatial resolution at 1km.Finally,the relative exposure risk model was used to evaluate and analyze the average annual PM2.5 exposure risk and seasonal changes in 7 cities.The main conclusions are as follows:(1)Based on remote sensing images of MODIS L1 B and Dark Target algorithm,AOD(the spatial resolution at 1km)achieved a verification accuracy of 0.781 with AERONET global aerosol data,which proved that the feasibility and reliability of using MODIS images to retrieve AOD and the spatial scale was suitable for ShanghaiHangzhou Bay area.(2)Among the regression models constructed by retrieval AOD and measured PM2.5 concentration,Power Model was the optimal model in spring,autumn and winter,and Index Model was the best one in summer.The coefficient of determination(R2)respectively reached 0.551 in summer,0.524 in autumn,0.511 in spring,0.504 in winter.Based on four-season optimal models and inversion AOD,the distribution of PM2.5 concentration in four seasons of 2016 was obtained from estimation and simulation.Its spatial scale was 1km ×1km,which was more spatio-temporal continuous than measured PM2.5 concentration data from monitoring stations.(3)The spatio-temporal distribution of PM2.5 concentration by remote sensing estimation in Shanghai-Hangzhou Bay area had obvious regularity,and the seasonal characteristics showed winter > spring > autumn > summer.High-value concentrated in urban area in Shanghai,Hangzhou and Jiaxing,low-value mainly in the southwest of Hangzhou and Zhoushan.Except Zhoushan,the PM2.5 concentration in other cities’ urban area exceeded the national annual average limit of level-2(35μg/m3),indicating that the PM2.5 pollution situation is still serious at a fine scale,and efforts should be made to control PM2.5 emission in key cities and special seasons.(4)The seasonal risk of population exposure to PM2.5 in each city did not change much,and dangerous levels were mainly distributed in urban area.The higher risk was in Shanghai(3.57)and Jiaxing(1.24),and lower in Zhoushan(0.46)and Huzhou(0.44).There was a strong spatial autocorrelation of exposure risk in Shanghai-Hangzhou Bay area: high-value clustering in Shanghai urban area(Hongkou,Jingan,Huangpu,Xuhui District,etc.),Hangzhou urban area(Shangcheng,Xiacheng,Xihu,Jianggan District,etc.),Ningbo urban area(Jiangbei,Zhenhai District),Haishu,Yinzhou,Beilun and Cixi,and Yuecheng District of Shaoxing;low-value clustering in higher elevation area,as well as Chongming District of Shanghai and Zhoushan.Therefore,it is necessary to strengthen the control of population size and total pollutant emission in high-risk area according to local conditions,and the maintenance of air quality and ecological protection in low-risk area.This study produced AOD and PM2.5 concentration products with high spatial resolution applicable to Shanghai-Hangzhou Bay area,focused on solving the problem of spatial scale and spatio-temporal continuity of population and PM2.5 concentration data,and completed the assessment on risk of population exposure to PM2.5 at a fine scale.In addition,it proved that the exposure risk assessment based on the spatial distribution of population density was far more accurate than only based on PM2.5 concentration.Research results could comprehensively reveal the real PM2.5 population exposure in Shanghai-Hangzhou Bay area,and could reasonably provide scientific reference for reducing fine particulate pollution and improving living environment.