Dissertation/Thesis Abstract

Novel approaches to spatio-temporal modeling of air contaminants in Fresno, California
by Noth, Elizabeth Marianne, Ph.D., University of California, Berkeley, 2009, 454; 3383606
Abstract (Summary)

The community of Fresno, California is notable for a high prevalence of asthma among an ethnically diverse population, and for high levels of ambient air pollution, especially PM, making it an appropriate location to address questions of air pollution's impact on this vulnerable population. The Fresno Asthmatic Children's Environment Study (FACES) is focused on quantifying the relationship between air pollution and asthma in young children who reside in Fresno, California. An underlying premise of FACES is that observed health effects are associated with specific exposures or sets of exposures, and that there are subsets of the population of asthmatic children who are more responsive to different exposures. To define the exposure characteristics of the children who comprise these subsets, the exposure assessment program is targeted to accurately estimate the individual level exposures daily over several years. This dissertation applies spatial regression to model the concentrations of three types of air pollution data for use in personal exposure assessment: ambient elemental carbon (EC), ambient polycyclic aromatic hydrocarbons (PAH), and PAH in pine needles. The spatial and temporal distribution of EC samples and PAH samples collected during the FACES study were each evaluated and modeled for all participant locations and dates in the FACES study using mixed effects modeling in land use regression models. These models use a central monitoring site as the primary temporal source data for both EC and PAH, and supplement the information on temporal variability with meteorological data, e.g., wind speed, and season indicator variables. Spatial data included land use, traffic intensive and proximity, point source locations, and agricultural burning. Third, pine needles were collected in Fresno at 91 locations and analyzed for PAHs. This biomonitoring data was used to build a model of the spatial distribution of PAHs on a single day in Fresno. Cross-sectional land use regression models that were built from the pine needle data show that both traffic and other non-mobile sources are important for PAH estimation in the urban environment. The models from filter EC and PAH were applied to estimate daily personal exposures for all days and for all participants in the FACES study.

Indexing (document details)
Advisor: Hammond, S. Katharine
Commitee:
School: University of California, Berkeley
School Location: United States -- California
Source: DAI-B 70/10, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Public health, Environmental science
Keywords: Air pollution, Asthma, Elemental carbon, Polycyclic hydrocarbons
Publication Number: 3383606
ISBN: 978-1-109-46072-8
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest