Author's School

School of Engineering & Applied Science

Author's Department/Program

Energy, Environmental and Chemical Engineering


English (en)

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Jay R Turner


An understanding of spatial and temporal variability in ambient particulate matter: PM) is important for effective air quality management and for assessing potential exposure misclassification in epidemiological and exposure studies used to support health-based standards. Spatiotemporal variability of PM in urban areas can be influenced by many factors, such as local sources of primary PM; source locations and their emission profiles; topographic barriers; meteorological patterns; behavior of semi-volatile components; and measurement errors. Intraurban variability is often gauged by conducting measurements at a network of monitoring stations across the region of interest. While certain statistical metrics are commonly used and interpreted in a relative sense across site-pairs, there is no standardized framework for analyzing such datasets.

This dissertation presents systematic data analysis approaches applicable to a variety of monitoring networks for assessing intraurban variability in PM and its components. Interpreting patterns in statistical metrics for a network with a large number of sites can be particularly challenging, and calculating these metrics for each site with respect to a reference concentration time series may better reveal the variability. In the absence of a representative background site, the network itself can be utilized to generate baseline and site-specific excess concentration time series to semi-quantitatively differentiate urban- and larger-scale contributions from local-scale emissions. Utilizing this approach for interpretation of patterns in the statistical metrics provides insights into the factors influencing the baseline and the monitoring sites displaying greater variability.

Apportionment of measured concentrations at each site into baseline and site-specific excess concentrations towards refined application of wind regression tools for estimating local emission source regions are also discussed. The approach is also utilized for identifying meteorological and geographic factors that modulate the spatial and temporal PM trends. It also provides a weight-of-evidence to conventional source apportionment tools used for estimating local and regional source impacts. The strengths and limitations of the proposed approaches are discussed for a variety of networks measuring PM and/or its components on varying spatial and temporal scales. Issues regarding measurement uncertainty estimation and precision in data reporting which can influence interpretation of variability are also discussed.


This work is not available online per the author’s request. For access information, please contact or visit

Permanent URL:

Included in

Engineering Commons