Spatially gap free analysis of aerosol type grids in China: First retrieval via satellite remote sensing and big data analytics

Abstract

Spatially contiguous aerosol type grids were rarely available for air quality management in the past. To bridge the gap, we developed an integrated remote sensing and big data analytics framework to generate spatially gap-free aerosol type grids between 2000 and 2020 in China. The effect of emission control via environmental management on haze reduction was fully realized for the first time with the aid of satellite-based gap-free aerosol type data. Daily gap-free aerosol fine mode fraction (FMF) data were first derived via a data-driven regression model based on remote sensing big data. According to empirically determined FMF probability distributions over regions with typical emission sources, aerosols in China were classified into eight major types, including typical/atypical/mixed anthropogenic aerosols, typical/atypical/mixed dust, and typical mixed and multiple modes. The results indicated that the gridded FMF estimates derived in this study agreed well with FMF retrievals from AERONET, with correlation coefficient of 0.81 and root mean square error of 0.13. The long-term variations in major aerosol types showed that in China the territory covered by typical anthropogenic aerosols was reduced from 21.38% to 11.76% over the past two decades, while dust aerosols decreased from 6.99% to 2.15%. The declining trend in anthropogenic aerosols could be attributed to reduced coal consumption and/or favorable dispersion conditions, whereas decreasing dust aerosols were largely associated with increased vegetation cover and/or weakened wind speed in the west. Overall, such advancements provide fresh evidence to improve our understanding of the emission control effect on haze pollution variations in China.