Approaches to estimating exposure levels of diesel exhaust particles (DEP) in an urban airshed: Model development and applications.

摘要

Epidemiological studies have indicated positive associations between exposure to ambient fine particulate matter (PM) and adverse health effects. Emissions from vehicular sources including diesel exhaust particles (DEP), is an important source of fine PM in urban areas. To evaluate the health effects of DEP, one of the important aspects is to assess the levels of exposure. This dissertation presents an approach to estimating the exposure levels to DEP from an established ambient PM2.5 monitoring network in the Greater Cincinnati Region (GCR).; The database consists of ambient PM2.5 mass and chemical speciation data collected from 2002 through 2006 in the GCR air shed. Furthermore, collocated indoor/outdoor PM2.5 samples were collected in spring and fall seasons, 2004, at six houses with varying distance from the highway. Multivariate receptor models including Unmix, PMF and PARAFAC, were applied to the data to estimate the traffic source contributions to ambient PM2.5 aerosols. Elemental carbon (EC) attributed to traffic sources (ECAT) is derived using Unmix. A modified kriging spatial interpolation method that incorporated Euclidean distance and traffic counts was developed to estimate ECAT at non-measured locations. This dissertation also discusses two important issues in multivariate receptor modeling - the handling of missing values and the effect of data set size on the robustness of the modeling results.; Results demonstrate that the inner city regions generally have higher levels of DEP than the suburban areas. DEP exposure levels are not solely dependent on the proximity of highways, as not all sites close to highway were equally high, but also depend on activity patterns such as the traffic volumes and meteorological conditions.; The traffic component was significant in the ambient air of the six houses. However, no significant correlation between traffic source contribution to indoor environment and distance of house from the highway was observed. The data suggest that for houses located in close proximity to major highways - the structure of the house envelope and ventilation pattern play a more important role in traffic related aerosol penetration than the distance from the highway and traffic intensity factors.