Simulated sensitivity of seasonal ozone exposure in the Great Lakes region to changes in anthropogenic emissions in the presence of interannual variability

A coupled meteorological and chemical modeling system with a 12-km horizontal grid spacing was used to simulate the evolution of ozone over the Great Lakes region between May and September of 1999 and 2001. The overall temporal and spatial variations in hourly ozone concentrations and ozone exposure from control simulations agreed reasonably well with the observations at most locations with an overall monthly bias computed across all stations ranging from −3.2 to 5.3 ppb, the root mean square difference ranging from 18.2 to 22.2 ppb, and an index of agreement ranging from 0.63 to 0.78. As with the observations, the simulated ozone exposure was higher during most months of the summer of 1999 than during 2001. Sensitivity simulations that increased anthropogenic trace gas emissions were performed to determine the changes in ozone exposure in the presence of meteorological interannual variability. The emission scenario simulations that employed the meteorological conditions of 1999 and increased anthropogenic emissions of NOx and VOCs produced increases in ozone exceeding 80 ppb over the lower peninsula of Michigan, the eastern half of the upper peninsula of Michigan, and over Ontario just north of Lake Superior and Lake Huron. Over the agricultural regions, more ozone between 60 and 80 ppb was produced. The cooler and wetter conditions with more frequent periods of northwesterly flow during the summer of 2001 were not as favorable for ozone production and did not result in increased ozone, despite the increase in anthropogenic emissions. Increases in ozone exceeding 60 ppb occurred only over the lake surfaces and in central Michigan when the meteorological conditions of the summer of 2001 were applied. For both summers, increases in anthropogenic emissions decreased ozone exposure in the immediate vicinity of the largest metropolitan areas. Since anthropogenic emission rate projections depend on assumptions of population, economic development, land-use patterns, and technology, the effect of anthropogenic emission rates on the magnitude and regional-scale distribution of ozone concentrations could be much larger or smaller than indicated by this study. In a subsequent study, the simulated ozone will be used as input to biological models to assess the response of ozone-sensitive tree species in the Great Lakes region to ozone levels produced by various anthropogenic emission scenarios.