Climatology and calibration of MERRA-2 PM2.5 components over China

Due to the lack of ground monitoring data, current research on ambient fine particulate matter (PM2.5) is limited to mass concentration, and few studies have investigated climatology of PM2.5 components in China. The Version 2 Modern-Era Retrospective analysis for Research and Applications (MERRA-2) provides a series of surface PM2.5 component datasets, including black carbon (BC), organic carbon (OC), and sulfate. We validated the MERRA-2 PM2.5 components by ground-based data from three monitoring sites in Nanjing, China and found there is a good correlation between them. We then used MERRA-2 data to characterize the spatiotemporal distributions of BC, OC, and sulfate concentrations over China from 1980 to 2018. We identified the hot spots of BC, OC, and sulfate pollution in China. Time series analysis shows that BC, OC, and sulfate pollution kept rising before 2007. Then the pollution were suppressed and at a stable level during 2007-2013, and decreased sharply after 2013, which were closely related to air pollution control policies in the last two decades. In addition, a random forest model was developed to preliminarily calibrate the MERRA-2 PM2.5 component data using ground measured PM2.5 components and MERRA-2 meteorological data. Modeling results achieved overall daily cross-validation (CV) R values of 0.85, 0.73 and 0.69 for BC, OC and sulfate, respectively, and the monthly R values reached 0.96, 0.85 and 0.84, which are higher than validation results of original MERRA-2 PM2.5 components. Our results show that MERRA-2 provides an effective tool for exposure assessment of PM2.5 components in China.

Automated summary

This study analyzed the PM2.5 components in China using MERRA-2 data, revealing a significant decrease in BC, OC, and sulfate pollution levels after 2007. Additionally, a random forest model was developed to calibrate the MERRA-2 PM2.5 component data, achieving higher calibration accuracy compared to the original MERRA-2 data.