Simulation of Ozone in the Great Lakes Region

It is well-documented that elevated surface ozone concentrations can have an adverse effect on many different types of vegetation in the upper Great Lakes region. Ozone exposure estimates have typically been based on measurements made at ozone monitoring stations. However, most ozone monitoring stations are located in urban areas and the measurements made at these sites are not representative of values in remote forested regions

In this study, a coupled mesoscale meteorological and chemical model is used to simulate the formation, transport, and fate of ozone in the Great Lakes region for 1-month period during the summer of 1999. The modeling system employs a nested grid configuration with the outermost grid encompassing North America and nested grids centered over the Great Lakes region. The meteorological model includes four-dimensional data assimilation that incorporates observed meteorological data to limit forecast errors in the wind field. A high spatial and temporal resolution of the resulting mesoscale meteorological analyses describe the evolution of the boundary layer and its interaction with the larger-scale synoptic circulations that affect ozone. The ozone fields simulated by the chemical model are evaluated using data obtained from operational ozone monitoring stations as well as air chemistry data collected at the surface and aloft during a field campaign in the vicinity of Philadelphia as part of the North East Oxidant and Particulate Study (NE-OPS). After the model has been evaluated, we will determine the ozone transport pathways into the remote forested regions surrounding the Great Lakes. Sensitivity studies will be made to distinguish between ozone produced by natural biogenic and anthropogenic precursor emissions, and will lay the foundation for assessing future landscape change impacts on ozone concentrations in the region. It will be shown that the model results can also be used to derive ozone exposure estimates in remote forest areas where no observational data exist.