Evaluating personal PM2.5 and black carbon exposure variability in Beijing's rural communities

Abstract

Exposure to air pollution is a major public health concern, with PM2.5 and black carbon (BC) linked to adverse health outcomes. To reduce emissions of PM2.5 and BC, the Chinese government implemented the Coal-to-Clean Energy Policy (CCEP) in 2016, reducing indoor PM2.5 concentrations. However, its effect on personal exposure remains unclear. This study evaluated the role of time-activity patterns in personal exposure to PM2.5 and BC in the context of the policy. Data from the Beijing Household Energy Transition study (winters of 2018-2022) included 252 participants with concurrent indoor and personal PM2.5 measurements and GPS- based time-activity data. Geofencing classified participant locations, and generalized linear models assessed exposure determinants. Model performance was evaluated using indoor PM2.5 data with and without time-activity adjustments. Personal PM2.5 exposure averaged 52.9 μg/m3, while BC averaged 1.6 μg/m3. The best PM2.5 model used indoor PM2.5 over the full sampling period (AIC: 489.06, adjusted R2: 0.59). The top BC model used indoor PM2.5 averaged only while participants were home (AIC: 407.59, adjusted R2: 0.25). On average, participants spent 20.0 hours at home per day (95% CI: 19.4, 20.7). Despite these time-activity trends, the lack of reductions in personal exposure were not explained by time-activity patterns, indicating that other influential factors may be impacting exposure, or the available data was insufficient to fully capture exposure variability. Enhanced time-activity monitoring is necessary to improve exposure assessments and better inform air quality interventions.