Chemical characterization and source apportionment of PM2.5 personal exposure of two cohorts living in urban and suburban Beijing

Abstract

In the study, personal PM_2.5 exposures and their source contributions were characterized for 159 subjects living in the Beijing Metropolitan area. The exposures and sources were examined as functions of residential location, season, vocation, cigarette smoking, and time spent outdoors. Sampling was performed for two categories of volunteers, guards and students, that lived in urban and suburban areas of Beijing. Samples were collected using portable PM_2.5 monitors during summer and winter. Exposure measurements were supplemented with a questionnaire that tracked personal activity and time spent in microenvironments that may have impacted exposures. Simultaneously, ambient PM_2.5 data were obtained from national network stations located at the Gucheng and Huairouzhen sites. These data were used as a comparison against the personal PM_2.5 exposures and produced poor correlations between personal and ambient PM_2.5. These results demonstrate that individual behavior strongly affects personal PM_2.5 exposure. Six primary sources of personal PM_2.5 exposure were determined using a positive matrix factorization (PMF) source apportionment model. These sources included Roadway Transport Source, Soil/Dust Source, Industrial/Combustion Source, Secondary Inorganic Source, Cd Source, and Household Heating Source. Averaged across all subjects and seasons, the highest source contribution was Secondary Inorganic Source (24.8% ± 32.6%, AVG ± STD), whereas the largest primary ambient source was determined to be Roadway Transport (20.9% ± 13.6%). Subjects were classified according to the questionnaire and were used to help understand the relationship between personal activity and source contribution to PM_2.5 exposure. In general, primary ambient sources showed only significant spatial and seasonal differences, while secondary sources differed significantly between populations with different personal behavior. In particular, Cd source was found to be related to smoking exposure and was the most unpredictable source, with significant differences between populations of different sites, vocations, smoking exposures, and outdoor time.

Keywords: Personal behavior; Personal exposure; Smoking; Source apportionment.