New insight into PM2.5 pollution patterns in Beijing based on one-year measurement of chemical compositions.

In recent years, air pollution has become a major concern in China, especially in the capital city of Beijing. Haze events occur in Beijing over all four seasons, exhibiting distinct characteristics. In this study, the typical evolution patterns of atmospheric particulate matter with a diameter of less than 2.5μm (PM2.5) in each season were illustrated by episode-based analysis. In addition, a novel method was developed to elucidate the driving species of pollution, which is the largest contributor to the incremental PM2.5 (ΔPM2.5), not PM2.5. This method revealed a temporal variation of the driving species throughout the year: nitrate-driven spring, sulfate-driven summer, nitrate-driven early fall, and organic matters (OM)-driven late fall and winter. These results suggested that primary organic particles or volatile organic compounds emissions were dominant in the heating season due to residential heating, while NOx and SO2 emissions dominated in the other seasons. Besides, nitrate formation seemed more significant than sulfate formation during severe pollution episodes. It was also found that the pollution formation mechanism in the winter showed some unique features in comparison with the other seasons: aqueous reactions were more important in the winter, while multiple pathways coexisted in the other seasons. Furthermore, this study confirmed that the PM2.5 in Beijing was moderately acidic despite a fully neutralized system. In addition, the acidity variation during pollution episodes displayed different patterns between seasons and was driven by both the variation of aerosol water and chemical compositions. These results provide a new perspective to understand the characteristics and mechanisms of aerosol pollution in Beijing. However, more accurate measurements are necessary for effective air pollution control that depends on the seasonal variation of fine particle formation in Beijing and the surrounding areas.