AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Spatial Correlation of Ultrafine Particle Number and PM2.5 Mass Concentrations: Implications for Health Assessment
PROVAT SAHA, Shayak Sengupta, Joshua Apte, Peter Adams, Allen Robinson, Albert Presto, Carnegie Mellon University
Abstract Number: 280 Working Group: Aerosol Exposure
Abstract There are growing concerns that adverse health effects of ambient ultrafine particles (UFP, diameter < 100 nm) may differ from those of currently regulated fine particulate matter (PM2.5) mass concentrations. However, the epidemiological evidence of whether UFP have chronic health effects independent of PM2.5 mass remains inconclusive or insufficient. The correlation between PM2.5 mass and UFP concentrations influences the ability of epidemiological studies to differentiate their health effects. To assess the spatial correlation of UFP and PM2.5 mass, we integrate insights from temporally and spatially rich measurements, land-use regression (LUR), and a high-spatial-resolution chemical transport model (CTM). We conducted 3-6 weeks of continuous measurements of particle number concentrations (PNC; a proxy for UFP) at 32 sites, and year-long continuous measurements of PM2.5 at 50 sites in Pittsburgh, Pennsylvania covering a wide range of urban land-use attributes. While PNC are more spatially heterogeneous than PM2.5 mass, they are spatially moderately correlated in our data set (R2 ~ 0.3, N= 30 sites). LUR models derived from these measurements explains about 80% of the spatial variability in PNC using traffic, commercial land use, and population density as predictors; PM2.5 LUR model explains about 60% variability with commercial and residential land use, population and point emission sources. The LUR-predicted PNC and PM2.5 concentration surfaces for the city of Pittsburgh are moderately correlated (R2~ 0.4). High-spatial-resolution (1km) CTM simulations show that PNC and PM2.5 are highly correlated (R2~ 0.8) in intra-urban scale. Our simulations and measurements suggest that the intra-urban spatial differences in PNC and PM2.5 mass concentrations are driven by primary emissions. The implication is that the PNC-PM2.5 mass correlations will make it difficult to isolate the independent health effects of UFP in epidemiological studies at the intra-urban scale.