Proposed Revisions to the Guideline on Air Quality Models (Appendix W)

On July 14, 2015, the U.S. Environmental Protection Agency’s (U.S. EPA’s) Office of Air Quality Planning and Standards’ (OAQPS) Air Quality Modeling Group provided a pre-Federal Register (FR) publication of the proposed revisions to the Guideline on Air Quality Models (Appendix W to 40 CFR Part 51) (Guideline) on the U.S. EPA’s Support Center for Regulatory Atmospheric Modeling (SCRAM) website. In addition, a fact sheet summarizing the revisions made to the Guideline was also provided. As previously announced on SCRAM, the 11th Conference on Air Quality Models (Modeling Conference) is scheduled for August 12 and 13, 2015 at U.S. EPA’s Campus in Research Triangle Park, NC. The Modeling Conference will serve as a public hearing for the Guideline’s revisions, which I will be attending. Written comments will be accepted for 90 days after the final publication in the FR.

Below, I have outlined the revisions to the Guideline that have the biggest impacts on the regulated community. First, it should be noted that the revisions propose a one (1) year transition period during which time previous guidance contained in the 2005 version of the Guideline will be accepted. The Guideline is currently scheduled to be finalized in spring of 2016, allowing U.S. EPA time to address comments received during the 90-day comment period and Modeling Conference public hearing. The proposed revisions fall into the following two (2) categories:

1. Substantive changes to address enhancements to the formulation and applications of the AERMOD air dispersion modeling and the incorporation of a tiered demonstration approach to address the secondary chemical formation of ozone (O3) and fine particulate (PM2.5) associated with precursor emissions [nitrogen oxides (NOX) and sulfur dioxide (SO2) for PM2.5 and NOX and volatile organic compounds (VOC) for O3] from single sources.
2. Editorial changes to update and reorganize the Guideline.

Proposed AERMOD formulation updates include the incorporation of default settings options to mitigate over predictions in modeled concentrations related to low wind speeds (especially relevant to the 1-hour standards) and horizontal and capped release points. ADJ_U* is an option available in AERMOD’s meteorological preprocessor AERMET that allows the user to adjust the minimum friction velocity (u*). The minimum friction velocity is a critical value for calculating mixing height, initial horizontal and vertical dispersion. My own evaluation of ADJ_U* presented at the National Council for Air and Stream Improvement, Inc. (NCASI) Northern Regional Meeting in May 2015 indicated very favorable results in mitigating over predictions in modeled concentrations especially from low release and fugitive sources. Another proposed AERMOD option to assist in low wind speed over predictions is LOWWIND3. LOWWIND3 increases the minimum lateral turbulence intensity (sigma-v) and adjusts the dispersion coefficient to account for the effects of horizontal plume meander. Both LOWWIND3 and ADJ_U* are welcomed updates to AERMOD that will mitigate over predictions in low wind speed conditions. Another useful update to AERMOD’s default settings are the options that address plume rise for horizontal and capped stacks through the use of the POINTCAP and POINTHOR. Current conservative guidance is to assume no initial velocity associated with horizontal and capped stacks. The POINTCAP and POINTHOR options will also help to mitigate over predictions in modeled concentrations from capped and horizontal stack releases. When using these options the model performs the necessary adjustments to account for plume rise. For sources with rain caps the plume radius is adjusted to account for initial spread from the rain cap and for horizontal sources the initial horizontal momentum in the downwind direction of the wind is utilized.

Next under the substantive changes revisions category is proposed model application updates. First, is the long-time coming revision to officially designate AERSCREEN as the preferred screening model, replacing SCREEN3. Based on this revision applicants in States that have air toxics modeling requirements should take note. Usually States rely upon screening models for air toxics modeling, therefore it is more likely that States will update their guidance to utilize AERSCREEN instead of SCREEN3 for air toxics modeling.

Two (2) other welcomed application updates are related to nitrogen dioxide (NO2) modeling. The first revision involves the replacement of the ambient ratio method (ARM) with the ARM2 method as the Tier 2 option. ARM2 applies a NO2/NOX ratio to modeled concentrations based on an evaluation of measured NO2/NOX ratios from U.S. EPA’s Air Quality System (AQS) record of air quality data. The second revision is the incorporation of the Tier 3 Ozone Limiting Method (OLM) and Plume Volume Molar Ratio Method (PVMRM) as default options in AERMOD. This means that the Tier 3 options will no longer be treated as an alternative model and, therefore, U.S. EPA Regional Office approval will no longer be necessary as part of regulatory modeling applications such as when conducting NO2 National Ambient Air Quality Standards (NAAQS) modeling.

The proposed revisions to the Guideline related to precursor impacts represent additional analyses that will have to be incorporated into air quality modeling evaluations in support of Prevention of Significant Deterioration (PSD) permitting requirements. The proposed revisions for addressing impacts of secondary formation of PM2.5 and O3 go back to a January 4, 2012, U.S. EPA-granted petition submitted on behalf of Sierra Club on July 28, 2010. This petition required U.S. EPA to initiate rulemaking for establishing air quality models to evaluate impacts of PM2.5 and O3 precursors to be used by all major sources applying for a PSD permit. The revisions to the Guideline are the first step in the initiated rulemaking.

The revisions lay out a two (2)-tiered demonstration approach for addressing single-source impacts of secondary formation of PM2.5 and O3.

1. Use of technically credible relationships between precursor emissions and a source’s impacts that may be published in peer-reviewed literature; developed from modeling that was previously conducted for an area by a source, a governmental Agency, or some other entity that is deemed sufficient; or generated by a peer-reviewed reduced form model.
2. Case-specific chemical transport models (e.g., photochemical grid models) to be determined in consultation with the U.S. EPA Regional Office. The suggested preferred photochemical grid models include:

a. Comprehensive Air Quality Model with Extensions (CAMx)
b. Community Multiscale Air Quality (CMAQ) model

These two (2) tiers are consistent with the second and third tiers laid out in US. EPA’s “Guidance for PM2.5 Permit Modeling”.  The revisions to the Guideline propose a separate rulemaking to establish new PM2.5 and O3 Significant Impact Levels (SIL) and to introduce a new demonstration tool for PM2.5 and O3 precursors referred to as Model Emissions Rates for Precursors (MERP). A MERP will represent a level of emissions of precursors that are not expected to contribute significantly to the concentrations of O3 or secondarily-formed PM2.5 and will have ambient impacts that will be less than the SIL and therefore will not cause or contribute to a violation of the PM2.5 or O3 NAAQS or PM2.5 PSD increments. The MERP demonstration tool will replace the first tier qualitative assessment outlined in U.S. EPA’s “Guidance for PM2.5 Permit Modeling.” A new guidance document outlining the MERP approach will be published on the docket once the Guideline rulemaking is published in the FR entitled “Guidance on the Use of Models for Assessing the Impacts from Single Sources on Secondarily Formed Pollutants Ozone and PM2.5.”

In other unexpected revisions, U.S. EPA is proposing to remove CALPUFF as the preferred long-range transport model citing “concerns about the management and maintenance of the model code given the frequent change in ownership of the model code since promulgation in the previous version of the Guideline.” The current need for long range transport modeling in PSD modeling evaluations is associated with assessing Class I PSD increments. In its place U.S. EPA outlined a screening approach to evaluate Class I PSD increments and recommends evaluating Class I SIL impacts at or about 50 kilometers with U.S. EPA preferred near-field models. CALPUFF may still be used as a screening approach if further analysis is required. However, if a multi-source Class I PSD increment analysis is required there will be no preferred model for distances beyond 50 kilometers and applicants will have to select an alternative model for determining Class I PSD increments impacts on a case-by-case basis with the reviewing authority. It should be noted that the proposed action to remove CALPUFF as the preferred long-range transport model does not affect its use under Federal Land Managers (FLMs) guidance regarding Air Quality Related Value (AQRV) analyses (i.e., visibility and deposition impacts) or Best Available Retrofit Technology (BART) regional haze implementation plans.

Another interesting proposed revision is related to the proposed use of prognostic meteorological data in limited situations. Specifically, U.S. EPA states in the revisions to the Guideline that the use of prognostic meteorological data may be used “where there is no representative National Weather Service (NWS) station, and it is prohibitive or not feasible to collect adequately representative site-specific data.” In these situations the Mesoscale Model Interface (MMIF) prognostic meteorological data processing tool would be used to develop an AERMOD meteorological dataset using the three (3) most recent years of prognostic data. Unfortunately, I doubt that the cost associated with an on-site meteorological monitoring program qualifies as “prohibitive or not feasible.”

The proposed revisions to the Guideline represent the first changes since AERMOD replaced ISC as the preferred near-field model in the 2005 amendments and represents a significant rework. Most of the revisions reflect changes necessary as a result of 2010 1-hour NO2 and sulfur dioxide (SO2) NAAQS and incorporate guidance contained in U.S. EPA memorandums released since this time period. In addition, how applicants account for impacts from PM2.5 precursors continues to evolve and for the first time addressing impacts from O3 will work its way into air quality modeling evaluations. This will be the first time that information contained in the guidance memorandums will be available to the public for comments so I encourage all those affected by air quality modeling requirements to take this opportunity to supply comment. I will be attending the public hearing as part of the 11th Conference on Air Quality Models on August 12 and 13, 2015 so look for another update from me on the information I learn.