Determinants on ambient PM2.5 infiltration in non-heating season for urban residences in Beijing: Building characteristics, interior surface coverings and human behavior

Abstract

Ambient fine particulate matter (PM2.5) pollution is currently a major public health concern in Chinese urban areas. However, assessment of ambient PM2.5 exposure and its health effects is challenging in China because the exposure primarily occurs indoors. There is large inter-home variability of the fraction of ambient PM2.5 that penetrates indoors and remains airborne (Finf), and the factors influencing this variability are unknown. In this study, 24-h real-time indoor and outdoor PM2.5 mass concentrations were concurrently collected for 41 urban residences in the non-heating season. The Finf were estimated with steady-state and dynamic models derived from mass balance considerations. Multivariate statistical analyses were employed to examine the associations between Finf and 78 factors related to building characteristics, motor vehicle traffic, human behavior, meteorology, furnishings, and atmospheric/indoor chemistry. The estimate of Finf over the 24-h monitoring period with the steady-state model was 0.72 ± 0.01; the Finf estimate for single residences, using the dynamic model, were 0.59 ± 0.13 (N = 33). Two predictive models for Finf were constructed with categorical and numerical variables, respectively. The results revealed that building characteristics, traffic, wall and floor coverings, and human behavior had substantial influence on Finf in the non-heating season. The variance contributions of the determinants of traffic, wall and floor coverings, and human behavior were comparable to or even greater than those of the building characteristics.