Long-term particulate matter modeling for health effect studies in California - Part 2: Concentrations and sources of ultrafine organic aerosols

Abstract

Organic aerosol (OA) is a major constituent of ultrafine particulate matter (PM0.1). Recent epidemiological studies have identified associations between PM0.1 OA and premature mortality and low birth weight. In this study, the source-oriented UCD/CIT model was used to simulate the concentrations and sources of primary organic aerosols (POA) and secondary organic aerosols (SOA) in PM0.1 in California for a 9-year (2000-2008) modeling period with 4 km horizontal resolution to provide more insights about PM0.1 OA for health effect studies. As a related quality control, predicted monthly average concentrations of fine particulate matter (PM2.5) total organic carbon at six major urban sites had mean fractional bias of -0.31 to 0.19 and mean fractional errors of 0.4 to 0.59. The predicted ratio of PM2.5 SOA/OA was lower than estimates derived from chemical mass balance (CMB) calculations by a factor of 2-3, which suggests the potential effects of processes such as POA volatility, additional SOA formation mechanism, and missing sources. OA in PM0.1, the focus size fraction of this study, is dominated by POA. Wood smoke is found to be the single biggest source of PM0.1 OA in winter in California, while meat cooking, mobile emissions (gasoline and diesel engines), and other anthropogenic sources (mainly solvent usage and waste disposal) are the most important sources in summer. Biogenic emissions are predicted to be the largest PM0.1 SOA source, followed by mobile sources and other anthropogenic sources, but these rankings are sensitive to the SOA model used in the calculation. Air pollution control programs aiming to reduce the PM0.1 OA concentrations should consider controlling solvent usage, waste disposal, and mobile emissions in California, but these findings should be revisited after the latest science is incorporated into the SOA exposure calculations. The spatial distributions of SOA associated with different sources are not sensitive to the choice of SOA model, although the absolute amount of SOA can change significantly. Therefore, the spatial distributions of PM0.1 POA and SOA over the 9-year study period provide useful information for epidemiological studies to further investigate the associations with health outcomes.