| This dissertation presents a science framework relevant to evaluating impacts of land use policy scenarios, energy technologies, and climate on urban and regional air quality. Emerging from collaboration with urban planners, this work provides a means for employing atmospheric chemical transport modeling to understand environmental ramifications of long-term, spatially disaggregated changes in population and automobile emissions at the census tract level, and to assess the sensitivity of these changes to densification strategies. Toward these goals, the framework is used to evaluate model skill in resolving contemporary characteristics of ozone (O3) and speciated fine particles (PM2.5) in the Great Lakes region of North America, and to quantitatively explore meteorological processes that bring about observed features of these pollutants in the region.;The Great Lakes were chosen due to a population concentrated in sprawling metropolitan areas, consistently high and widespread pollutant burdens, and seasonal effects of the lakes on the atmosphere. In annual simulation at 36 km resolution, the Community Multiscale Air Quality model is evaluated using speciated PM2.5 measurements taken at regulatory monitoring networks orientated to sample urban, rural, and remote areas. Performance relative to ad-hoc regional modeling goals and prior studies is average to excellent for most species throughout the year. Both pollution episodes and seasonality are captured. The Great Lakes affect pollution seasonality: strong winds advect aerosols around the deep marine boundary layer to lower surface PM2.5 in fall and winter, while O3 over the lakes is enhanced throughout the year, driven by temperature in the cool seasons and lake breeze circulation in spring and summer.;Simulations confirm observational evidence that rural and small-city sources are responsible for most regional PM2.5. Sensitivities to urban and rural reductions are of comparable magnitude on a percentage basis. Higher horizontal resolution is necessary to capture changes in urban-suburban-rural pollution gradients, given the non-linear O3 response to emissions changes in NOK-limited urban environments (where vehicle NO, reductions increase O3) vs. VOC-limited rural environments (where reductions decrease O3). Densification scenarios produce small-scale responses. O 3 response in the suburbs is of opposite sign to O3 response in urban cores, while emissions reductions consistently reduce PM2.5 . |
