By Robynn Andracsek, P.E., Burns & McDonnell and Contributing Editor and Minda Nelson, P.E., Burns & McDonnell

The foundation of the Clean Air Act is the National Ambient Air Quality Standards (NAAQS). It is from these numeric values, which define what is and is not considered "clean" air, that all other air regulations are derived. EPA is required to review the NAAQS every five years and it is uncommon for NAAQS to be swept up into court cases. In contrast, many of the recent new regulations that have impacted the power industry have taken a long time to come into effect or have been reversed or delayed by legal challenges, such as the Mercury and Air Toxics Standard (MATS) and the Cross State Air Pollution Rule (CSAPR). So, it is relatively easy for these regular NAAQS reviews to dramatically impact industry.

The newest NAAQS are the 1-hour standards for sulfur dioxide (SO₂) and nitrogen dioxide (NO₂). Compliance becomes more difficult as the averaging period decreases. Previously, the NAAQS for NO₂ was expressed only as an annual average. Think of air quality as your car's speed. An annual NAAQS would equate to the average speed you drive over the course of a year, including waiting in traffic, highway driving, and driving on residential streets. It would be easy to meet an annual speed limit of 25 mph. A 1-hour NAAQS would then equate to an hourly average speed limit of 25 mph. It would be a lot more difficult to meet this short term limit.

Many states have started determining their 1-hour SO₂ non-attainment areas by proactively modeling significant sources (i.e., those emitting over 100 tons per year). Because, for the first time, dispersion modeling by itself can be used to demonstrate that an area is out of attainment, it is no longer necessary for regulators to wait for three years of monitored data to demonstrate a NAAQS exceedance.

For NO₂, the entire country is assumed to be unclassifiable (meaning, in attainment by default) unless there is an existing monitor showing an exceedance. States are working to implement additional monitoring stations near major roadways because mobile sources are predicted by EPA to be the dominant influence on the 1-hour NO₂ NAAQS. However, a requirement for dispersion modeling can be triggered for any facility due to permitting activities at the plant or merely in the surrounding area.

The problem with dispersion modeling is that it is extremely complex; it's more of an art form than a strict science. There are hundreds of inputs and dozens of crucial decisions that are made based on the modeler's experience and the specific situation. EPA's AERMOD dispersion model predicts ground level concentrations of any generic pollutant without chemical transformation. Thus, the modeled NOx emission rate will give ground level modeled concentrations of NOx. The NAAQS, however, is for NO₂.

The EPA has a three-tier approach to modeling NO₂ concentrations:

• Tier I assumes total conversion of all NOx to NO₂ (NOx = NO₂)
• Tier II uses a default NO₂ / NOx ratio (75 percent for annual and 80 percent for 1-hour)
• Tier III is a case-by-case detailed screening method using procedures such as the ozone limiting method (OLM) and plume volume molar ratio method (PVMRM), which requires EPA approval

Typically, initial NO₂ modeling is performed using either the Tier I or Tier II methodology. The less conservative Tier III methods are used only if needed. Tier III inputs include an in-stack NOx to NO₂ ratio, which is essentially the amount of NO₂ present in the stack gases determined for each piece of equipment being modeled as determined from published data or default values, and hourly or annual background ozone concentrations. The dramatic difference in modeling results for a single plant based on these three tiers of NOx to NO₂ conversion methodology is shown in Figure 1. The procedures used and assumptions made on how much NOx transforms to NO₂ in the atmosphere (based on the amount of ambient ozone available for the chemical reaction) can yield modeled impact reductions that range from 20 to 80 percent, as demonstrated in the three plots.

Adding to the modeling confusion is whether or not "intermittent," emergency, or rarely used sources must be included in the 1-hour SO₂ and NO₂ NAAQS models. An EPA guidance memo supports excluding these sources, but leaves the final determination up to each state.

If compliance with the NAAQS is not on your radar yet, it should be. The 1-hour SO₂ and NO₂ NAAQS have already taken effect; demonstrating compliance through modeling is complicated; and their consequences can be just as severe as anything that MATS or CSAPR might bring.