Ozone Sensitivity to Emissions and Changes of Limiting Reagents

Regional control strategy options for reducing ozone change temporally and spatially in Central California where air pollution is particularly serious. The rate of ozone production is a complex function of the concentrations of volatile organic compounds (VOCs) and nitrogen oxides (NOx) as well as meteorological conditions. As a result, ozone formation exhibits a very non-linear dependence on its precursors. Determining the relative benefits of controlling NOx emissions or VOC emissions remains a challenging problem. Current practice of modeling 3 to 5 day episodes does not capture the changes in limiting reagents since they represent a limited sample of the diverse meteorology and human behavior that affect air pollution.

We are using CMAQ, the EPA's Community Multiscale Air Quality Model, to model a season of air quality in Central California for the summer of 2000 to illustrate some limitations of current practice. We have modeled a 15-day period and, in concert with the modeling, have also used the Decoupled Direct Method to compute ozone sensitivities to NOx emissions and VOC emissions. The emissions have been disaggregated differently to extract mechanistic information regarding limiting reagents, and to explore issues of long range transport. We have computed ozone sensitivities to total NOx emissions and VOC emissions for the entire modeling domain, NOx emissions and VOC emissions from specific air basins, as well as emissions from specific air basins for specific time intervals. We demonstrate how the computed sensitivity coefficients of ozone to the various emission types and consideration of meteorology may be used to demonstrate and understand limiting reagent changes throughout the modeling domain.