Exposure Beyond Mass

Abstract

PDS 76: Source specific outdoor air pollution studies, Exhibition Hall (PDS), Ground floor, August 26, 2019, 4:30 PM - 5:30 PM Background: Most epidemiological studies of health effects of atmospheric particulate matter use conventional filter measurements. These measurements quantify total mass concentration but may oversimplify air pollution exposure estimates as they are blind to many critical physiochemical properties of atmospheric particles. These properties likely have larger exposure variability than particulate matter mass. Objectives: This study is designed to measure and predict the spatial variations of particulate matter exposure using state-of-the-art single-particle mass spectrometry to resolve detailed physicochemical properties, namely number, size, source and chemical mixing state of individual particles in a populous urban area to study their population exposure variability and the contrasts with mass exposure. Methods: We performed mobile sampling using an advanced single-particle mass spectrometer to measure the within-city variability of number concentration of source-resolved 50 – 1000 nm particles and particle mixing state in Pittsburgh, PA, USA. We built land use regression (LUR) models to estimate their spatial patterns. We coupled demographic data to estimate population exposure. Results: Particle number concentration has a much larger spatial variability (factor of ~5) than mass concentration (factor of 2) within the city. Fresh primary particles emitted from traffic and cooking drive the exposure variability on the number concentration basis, but the mass exposure is dominated by aged background particles composed of secondary material. In addition, we show that people exposed to elevated number concentrations of atmospheric particles are also exposed to more externally mixed, primary particles. Conclusions: There is a large intra-city spatial heterogeneity of exposure to traffic and cooking particles on the number basis. Such heterogeneity is not captured by conventional mass exposure measurements. In addition, the resulting strong spatial patterns of particle chemical mixing state may potentially be significant when evaluating health effects of particulate matter exposure.