Evaluation of MERRA-2 reanalysis for Sulfur dioxide aerosol in China

Considering the uncertainty of the reanalysis data is essential, the uncertainty of Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2) SO2 concentration still is lacking. In this talk, we evaluated the MERRA-2 SO2 concentrations in China. Monthly SO2 concentrations from 100 ground observation during the years 2015-2021 were used to evaluate the monthly MERRA-2 SO2 data. Our results showed that the MERRA-2 SO2 concentrations exhibit a moderate Pearson correlation coefficient (R) with ground-based SO2 measurements (R = 0.62). In terms of root-mean square error (RMSE) and mean absolute error (MAE), biases (≤ 10 μg m-3) were found mainly for extensive regions (about 70%) of the Chinese sites. According to the results, the relative mean bias (RMB) and fractional gross error (FGE) showed values greater than 1 and 0, respectively, in the eastern China, which indicates that MERRA-2 overestimates the SO2 measurements in the urban regions, while underestimation of MERRA-2 SO2 was found in rural regions of China. A parameterized method could be suggested to improve the quality of MERRA-2 SO2