Measuring the Emission Changes and Meteorological Dependence of Source‐Specific BC Aerosol Using Factor Analysis Coupled With Machine Learning.

Reducing ambient black carbon (BC) relies on the targeted control of anthropogenic emissions. Measuring emission changes in source‐specific BC aerosol is essential to assess the effectiveness of regulatory policies but is difficult due to the presence of meteorology and multiple co‐existing emissions. Herein, we propose a data‐driven approach, combining dispersion‐normalized factor analysis (DN‐PMF) with a machine learning weather adjustment (deweathering) technique, to decompose ambient BC into source emissions and meteorological drivers. Six refined BC sources were extracted from the factor analysis of aethalometer multi‐wavelength BC and concurrent observational datasets. In addition to the widely reported dominant sources, such as vehicular emissions (VE) and coal/biomass burning (BB), a discernible port and shipping emission source were identified with potential impacts on coastal air quality. The source‐specific BC showed abrupt changes in response to interventions (e.g., holidays) after separating weather‐related confounders. Significant reductions in deweathered coal and BB, VE, and local dust verified the effectiveness of policies, such as clean winter‐heating and support for the Clean Air Actions. As revealed by a post‐hoc model explanation technique, the evolution of the boundary layer was the predominant meteorological driver exerting the opposite impact on local sources with respect to distant regional‐wide sources, that is, the port and shipping emissions. Plain Language Summary: Throughout the course of human history, substantial efforts have been made worldwide to alleviate air pollution. Quantifying how a regulatory action (intervention) can impact air pollutant emissions and air quality is crucial, but complicated owing to multiple confounders (e.g., meteorology). Black carbon (BC) is a primary air pollutant that warms the atmosphere, worsens air quality, and impairs human health. Herein, we applied a data‐driven approach, combining factor analysis with machine learning techniques, to attribute ambient BC to its emission sources and meteorological‐related drivers. Using this coupled approach, the impacts of a series of human interventions (e.g., holiday effects, the COVID‐19 lockdown, clean‐heating policy, and the Chinese Clean Air Action) on source‐specific BC can be quantified. Our analyses highlight that this approach is also applicable to other regulated air pollutants, with great potential to facilitate air quality accountability studies. Key Points: Six refined black carbon (BC) sources were resolved from the factor analysis of aethalometer BC combined with concurrent measurementsPositive matrix factorization (PMF) coupled with the deweathering technique enables the detection of emission changes in source‐specific aerosol over timeImpacts of human interventions (holidays and the COVID lockdown, winter heating policy, Clean Air Actions) on BC aerosol were quantified [ABSTRACT FROM AUTHOR]