Evolution of Organic Aerosol Molecular Fingerprints in Beijing: Impact of Anthropogenic‐Biogenic Emissions and Humidity‐Induced Aqueous Chemistry

Abstract Organic aerosol (OA) continues to pose a significant challenge in the study of fine particulate matter due to its complex sources and composition. By utilizing the Extractive Electrospray Ionization Time‐of‐Flight Mass Spectrometer (EESI‐TOF‐MS), OA can be identified at a near‐molecular level, which facilitates a deeper understanding of its composition. Intercomparison of OA and specific source tracers between EESI‐TOF‐MS and an Aerodyne Mass Spectrometer (AMS) confirms that EESI‐TOF‐MS can track OA evolution, while the results indicate that the EESI‐TOF‐MS only measures 30%–43% of the OA quantified by the AMS. During the summer campaign conducted in Beijing, CHO was identified as the predominant species, representing 66% of OA. CHON and CHN groups contributed for 27% and 2%, respectively. Molecular fingerprints indicated significant contributions from both anthropogenic (e.g., biomass burning, aromatic oxidation) and biogenic (e.g., monoterpene and sesquiterpene oxidation) sources. By comparing the concentration and composition of OA during polluted and clean periods, it is evident that the pollution was predominantly characterized by organic nitrate (ON) and CHO 4‐5 species, linked to enhanced anthropogenic emissions and potentially biogenic sources. Furthermore, machine learning and SHAP analysis identified relative humidity as a critical driver of most OA species. Cluster analysis further resolved these RH‐enhanced ions, which are mainly short‐chain organic acids, into two distinct categories: Cluster 1 driven by early‐morning photochemical and aqueous‐phase processes, and Cluster 2 associated with regional chemistry and sulfate‐mediated reactions. Our findings highlight the synergistic roles of anthropogenic emissions and meteorological conditions in shaping the OA burden and composition, providing crucial molecular‐level insights for refining climate and air quality models.