The empirical correlations between PM2.5, PM10 and AOD in the Beijing metropolitan region and the PM2.5, PM10 distributions retrieved by MODIS

Abstract

We observed PM2.5, PM10 concentration, aerosol optical depth (AOD), and Ångström exponents (α) in three typical stations, the Beijing city, the Xianghe suburban and the Xinglong background station in the Beijing metropolitan region, from 2009 to 2010, synchronously. The annual means of PM2.5 (PM10) were 62 ± 45 (130 ± 88) μg m−3 and 79 ± 61 (142 ± 96) μg m−3 in the city and suburban region, which were much higher than the regional background (PM2.5: 36 ± 29 μg m−3). The annual means of AOD were 0.53 ± 0.47 and 0.54 ± 0.46 and 0.24 ± 0.22 in the city, suburban and the background region, respectively. The annual means of Ångström exponents were 1.11 ± 0.31, 1.09 ± 0.31 and 1.02 ± 0.31 in three typical stations. Meanwhile, the rates of PM2.5 accounting for PM10 were 44%–54% and 46%–70% in the city and suburban region during four seasons. The pollution of fine particulate was more serious in winter than other seasons. The linear regression functions of PM2.5 (y) and ground-observed AOD (x) were similarly with high correlation coefficient in the three typical areas, which were y = 74x + 18 (R2 = 0.58, N = 337, in the City), y = 80x + 25 (R2 = 0.55, N = 306, in the suburban) and y = 87x + 9 (R2 = 0.64, N = 350, in the background). The functions of PM10 (y) and ground-observed AOD (x) were y = 112x + 57 (R2 = 0.54, N = 337, in the city) and y = 114x + 68 (R2 = 0.47, N = 304, in the suburban). But the functions had large differences in four seasons. The correlations between PM2.5, PM10 and MODIS AOD were similar with the correlations between PM2.5, PM10 and the ground-observed AOD. With MODIS C6 AOD, the distributions of PM2.5 and PM10 concentration were retrieved by the seasonal functions. The absolute retrieval errors of seasonal PM2.5 distribution were less than 5 μg m−3 in the pollutant city and suburb, and less than 7 μg m−3 in the clean background.