Air quality modeling for control strategy development

With the onset of National Ambient Air Quality Standards (NAAQS) for fine particulate matter (PM2.5) and more stringent NAAQS for ozone, states are mandated to develop State Implementation Plans (SIPs) detailing strategies for achieving those standards in non-attainment regions. Sensible formulation of attainment strategies is crucial, as the costs and benefits of pollution control are sizable and the relative cost-effectiveness of various options may vary by orders of magnitude. Control strategy selection for PM2.5 and ozone is particularly challenging because both pollutants originate from complex interactions of a variety of precursor compounds emitted both locally and regionally.

This poster will explore how air quality models can be applied, in conjunction with control cost and health benefit assessments, to enhance their relevance to the selection of abatement measures for PM2.5 and ozone. While SIPs have always relied on air quality modeling as a final determination of whether an overall strategy is sufficient, we suggest that air quality models should also be applied early in the SIP development process to help identify and prioritize control measures. Through sensitivity analysis techniques such as brute force or the decoupled direct method, models can be applied to estimate the responsiveness of ambient pollutant concentrations to controls of various precursor compounds, source categories, and emission regions. Because computational limitations preclude simulating each possible control measure for each source, sensitivity analysis must be tailored to maximize its relevance for assessing a wide variety of potential measures within the constraints of time and computational resources. We will illustrate, based on SIP development underway for ozone and PM2.5 in the State of Georgia, how the impacts of various potential control measures can be represented as linear combinations of atmospheric sensitivities to a limited number of generic controls, with separate assessment of potential controls for the largest point sources. If conducted early in the SIP development process, with close coordination between modelers and policy developers, sensitivity analysis can both guide the selection of abatement measures and facilitate an iterative search for additional measures if later modeling deems an overall strategy to be insufficient. Linkage of sensitivity analysis with assessments of control costs and health benefits could foster the evaluation of various abatement options for multiple pollutants on a common metric and enable a more integrated and multi-faceted approach to policy development.