Accuracy Verification And Spatiotemporal Distribution Analysis Of Aerosol Products In The Yangtze River Delta Region

In recent years,with the development of China's economy,people's pursuit of quality of life has been continuously improved.Air pollution,as an important environmental problem,has become the focus of attention.The Yangtze River Delta region is one of the most economically active areas in China,with the characteristics of large population density,industrial agglomeration,developed transportation network,etc.,with high emissions of industrial waste gas,automobile exhaust and other polluting gases.Therefore,strengthening the monitoring of air quality indicators,including the concentration of atmospheric particulate matter,is conducive to the research and control of regional air pollution.Aerosol optical depth is a key factor to characterize atmospheric turbidity,and its value can reflect the concentration of near-surface particles to a great extent.In this study,the ground-based observation data of AERONET and CCCES were used to evaluate MODIS and VIIRS data products,and to analyze the applicability of these products in this region in 2011-2020,so as to provide reference for AOD related research.The main conclusions of this study are as follows:Terra/Aqua C6.1's Dark Target algorithm(DT),Deep Blue algorithm(DB)and the combined algorithm(DT&DB)have been used in Hangzhou,Ningbo,Hefei,Shangxing,Nanxin and Taihu sites,the result showed that the smaller sampling window has better fitting effect,so the combination of 3×3 pixels as spatial window and±15min as time window was selected for evaluation.In this combination verification result,the accuracy of DB algorithm and DT algorithm for two farmland-based surface sites was good,but the correlation coefficient of DT algorithm was relatively high.Only DB algorithm can reverse the effective observation points at Taihu site,although the observation year is long,the number of reverse effective observation points is small,which is related to its proximity to Taihu Lake,so water pixels cannot reverse AOD.The correlation coefficient of DB algorithm in three urban sites are the highest,and the data of DB algorithm falling within MODIS expected error are all over 60%.The validation results of VIIRS EDR products showed that the accuracy of DT algorithm was poor,R~2 is less than 0.6,and the accuracy of DB algorithm is high(R~2 ranges from 0.79 to95,and more than 50%of the data fall within the expected error).Compared with Aqua products,the accuracy of VIIRS DB algorithm at shangxing site was higher than that of Aqua DB algorithm,while the accuracy of Aqua products at other two sites were higher.On the whole,MODIS products were more applicable,and the error of VIIRS DB algorithm was smaller,but the error of DT algorithm was larger.As far as time variation is concerned,the monthly mean value of AOD has obvious fluctuation and regularity,with the high value appearing in March and April(spring)and the low value appearing in July and August(summer).The annual average value of AOD was higher in 2011 and 2014,while it showed a downward trend from 2011 to 2020.The monthly average value of?is in the range of 0.8-1.6,and the main pollution particles are mixed particles and fine particles,which indicates that human activities have great influence.In terms of spatial changes,the distribution of AOD values in the Yangtze River Delta was obviously different,which was mainly manifested as high in the north and low in the south.The pollution in plain areas and coastal riverfront areas was serious,and the air quality in mountainous and hilly areas was better.