An Assessment of Air Pollution Databases to Inform their use in Epidemiological Studies
Date of Award
Master of Science (MS)
Ambient particulate matter (PM) is a complex mixture of chemical components with well documented adverse health effects. Science-based standards for PM are reassessed on a five-year cycle with epidemiological studies often carrying significant weight. Such studies typically do not measure individual exposure but instead use a proxy such as data from a central monitoring site or emissions within a geographic area; however, the spatiotemporal variability of PM and especially its chemical components makes it difficult to characterize individual exposures.
This thesis focuses on informing the use of air pollution datasets in epidemiological studies. Two forms of data are explored: stationary monitoring networks and emissions inventories. In stationary networks, the sampler precision is a necessary consideration when assessing variability between sites and whether a single monitor adequately represents individual exposure. For emissions inventories, the role of both inventory reporting structure and inclusion criteria on the study population are important considerations.