Spatial And Temporal Distribution Of Aerosol In Anhui Province In Recent 15 Years And The Relationship Between AOD And PM2.5

MODIS AOD data are characterized by large scale and continuous space and social media data are easy to acquire,cheap and have geographical coordinates.Using this two kinds of data to retrieval ground PM2.5 can effectively make up the defects of lacking PM2.5 observation station.The MODIS Leve2 AOD data are widely used in the world,but the newly released C6.1 version has not been validated and evaluated in China.In this paper,ground based observation data will be used for validation of MODIS C6 and C6.1 Leve2 aerosol optical depth retrievals over Anhui province.Based on the first part discusses,suitable MODIS aerosol products will be used to retrieval aerosol optical properties of Anhui in recent fifteen years.Then,analyzed the relationship between the PM2.5 and the aerosol optical depth will be analyzed in Hefei,and the feasibility of adding social media data to the relational model between PM2.5 and aerosol optical depth will be explored as well.The main results indicated that:(1)The suitable MODIS Leve2 aerosol data for Anhui province is Terra C6.1 version at 10 km resolution with deep blue algorithm and best data quality.Compare to the Terra,Aqua is the better choice for 3km resolution data.In county,retrieval aerosol optical depth at 3km resolution is better than 10 km resolution.The improvement of the C6.1 version brings more retrieval promotion to Terra than Aqua.(2)The distribution of aerosol optical thickness in Anhui Province in the last 15 years shows a decreasing trend from north to south,and its changes adapted to the changes of terrain.The AOD characterized by a seasonal pattern,with the highest in summer and the lowest in spring and winter.The Angstrom in Anhui Province is small in north and south,and the middle is large.In summer,Angstrom is large in the north and small in the south.In autumn,Angstrom is large in the south and small in the north.In spring,Angstrom shows an increasing trend from north to south.From 2002 to 2016,the highest AOD appears in 2007,and the lowest AOD appears in 2016.After 2012,the annual average AOD and PM2.5 have similar variation tendency.The monthly average distribution shows a single peak,with the highest AOD in June,0.99.(3)The relationship between aerosol optical thickness and PM2.5 shows regional differences.And the correlation is better in the area far away from the center of the city,The MLR shows seasonal differences,with the better fitting effect in spring.After adding microblog data,the fitting effect of the model can be effectively improved.In the spring model,the adjusted R2 is 0.67,and the multiple R is 0.81.