The following HTML text is provided to enhance online
readability. Many aspects of typography translate only awkwardly to HTML.
Please use the page image
as the authoritative form to ensure accuracy.
Air Quality Management in the United States
NRC report on I/M programs concluded that such programs provide “much lower benefits” than those estimated by the MOBILE model (NRC 2001c).
Air Quality Modeling
The second step in the development of an attainment-demonstration SIP is to determine the amount of pollutant-emission reductions that will be needed to bring about compliance with the NAAQS. Air quality models generally play the central role in that process. Typically a model is used to simulate one or more of the historical events that contributed to the area’s “design value” (defined in Box 3-2) and, through these simulations, demonstrate that appropriate emission controls will prevent NAAQS violations in the future during similar events.5 Because the model simulations are required to show compliance in the future, they must be run with future projections for a variety of hard-to-predict quantities that affect emissions (for example, socioeconomically driven activity factors, such as vehicle miles traveled and the effectiveness of emission-control technology). The projections and the attainment demonstration based on simulations of the extreme events that contributed to the design value introduce uncertainty into the process beyond that arising from intrinsic uncertainties in the emission inventories and models.
The air quality models used in these analyses are designed to allow policy-makers to link quantitatively pollutant emissions to concentrations of pollutants in the atmosphere. Models of several types have been in development and used for over 30 years and, through a close collaboration between the scientific, engineering, and regulatory communities, have continuously evolved in their capabilities and technical completeness (Russell and Dennis 2000). There are three major classes of air quality models: (1) statistical and empirical models that are based on observed relationships between pollutant concentrations and emission rates with little or no explicit consideration of the underlying physical and chemical processes that determine these relationships; (2) deterministic models that solve mathematical equations that describe the physics and chemistry of air pollutant emissions, formation, transport, and removal; and (3) a hybrid of the former two that, although essentially empirical or statistical in its approach, makes use of physically and chemically based algorithms. An example of the latter
Some recent SIP revisions have not required the exercise of modeling ambient air quality. For example, modeling of ambient air quality was not required for the rate-of-progress SIP revisions that demonstrated that O3 nonattainment areas classified as moderate or above would reduce VOC emissions by 15%. Although they were backed by regional-scale modeling carried out by EPA and the Ozone Assessment Transport Group (OTAG), the NOx SIP call and the I/M SIP revisions did not require air quality modeling by the states.