This article appears in the October 2025 issue of EM magazine, a copyrighted publication of the Air & Waste Management Association (A&WMA)

An overview of the U.S. Environmental Protection Agency’s (EPA) proposed framework for appropriately characterizing background concentrations and how to assess the quality and representativeness of existing ambient monitoring data to inform explicitly modeled nearby sources in cumulative PSD NAAQS impact assessments.

In October 2023, the U.S. Environmental Protection Agency (EPA) proposed updates to the Guideline on Air Quality Models (Guideline)¹ codified as 40 CFR 51, Appendix W. The proposed updates include revisions to
Section 8.3 for characterizing background concentrations. Background concentrations are necessary to quantify the total air quality concentration of a given pollutant for proposed new or modified sources that trigger cumulative National Ambient Air Quality Standard (NAAQS) and Prevention of Significant Deterioration (PSD) increment assessments. Background concentrations must account for all contributing sources of air pollutant emissions and may include nearby sources and other sources, as defined in Section 8.3.1(a) of the Guideline.

Concurrent with the proposed changes to the Guideline, EPA also released in October 2023 a companion document entitled “Draft Guidance on Developing Background Concentrations for Use in Modeling Demonstrations.”² This draft guidance provides a recommended framework comprising four distinct steps:
1. Defining the scope of the cumulative impact analysis for isolated or multi-source situations;
2. Identifying the relevant and available emissions, air quality, and environmental data;
3. Determining representativeness of ambient monitoring data; and
4. Determining nearby sources to be explicitly modeled.

EPA states in the draft guidance that the proposed framework “will facilitate applying best professional judgment to identify appropriately representative monitoring data and select nearby sources to explicitly model to inform a cumulative impact analysis that best represents the local air quality throughout the geographic scope of the analysis in particular near the project source(s) under consideration.”

In the current 2017 Guideline³, EPA recommends using the significant concentration gradient (SCG) approach to inform the explicitly modeled nearby source inventory in multi-source cumulative impact assessments, while contributions from other sources are characterized through adequately representative monitoring data. In the draft guidance, EPA states that the interconnectedness of these two components tends to be overlooked and the SCG approach has proven to be problematic given the lack of a clear definition of that term or specific examples of its implementation. To that end, the framework was developed based on the same general approach and is intended to define the process more clearly for developing appropriate background concentrations.

This article discusses key elements that should be considered when evaluating existing air quality monitoring data, the quality of such data, and how to assess whether the ambient monitoring data captures emissions from nearby sources that may otherwise be subject to explicit modeling in PSD and minor source (when applicable) cumulative NAAQS impact assessments. The draft guidance discusses PSD increment, which is usually assessed via air dispersion modeling without consideration of ambient monitoring background concentrations. PSD increment analyses are outside the scope of this article.

Step 1: Define the Scope of Cumulative Impact Analysis

This step focuses on evaluating the proposed project location relative to “other sources” (as defined in the proposed
Guideline, Section 8.3.1(a)(ii) 1) and stationary emissions sources that may influence the total air quality concentrations within the defined modeling domain. Determining whether the proposed project is an isolated source or is in a multi-source area should be assessed from the following:

• Emissions inventory data within the modeling domain, as available by the state or local air pollution control agency.
• Consideration of the applicable averaging periods that trigger air dispersion modeling to estimate the spatial extent of pollutant impacts to inform Steps 2 and 3.
• Single source impact analyses for the proposed project to identify the most distant location where the project’s significant air quality impacts may occur; see Step 4 for important implications for the 1-hr nitrogen dioxide (NO₂ 1-hr sulfur dioxide (SO₂, and annual fine particulate matter (PM_2.5) NAAQS.

The dispersion environment and influences of terrain and other geographic features on the selected meteorological data. Maps showing the proposed project location and the above information should be prepared and qualitatively analyzed to define the modeling and inform framework Steps 2 through 4.

Step 2: Identify Relevant and Available Emissions, Air Quality, and Environmental Data

In addition to the information assessed in Step 1, the draft guidance recommends identifying and gathering ambient air monitoring data for the relevant pollutants and averaging periods. To characterize background concentrations for isolated sources, EPA places emphasis on the ambient monitoring data and environmental data (i.e., the dispersion environment, terrain, etc., assessed in Step 1). Several options for ambient monitoring data in the vicinity of the project should be identified using sources such as EPA’s AirData website, AirNowTech, or the state’s ambient monitoring website. To characterize background concentrations for multi-source areas, EPA places emphasis the same data needed for isolated sources in addition to the location, emissions, and release parameters of existing stationary sources of emissions that are not adequately characterized by the ambient monitoring data. Any previous permit actions or dispersion modeling for the proposed project, and existing dispersion modeling, active or pending permits or applications, and final permits for potential nearby sources should also be identified.

The ambient monitoring data identified for isolated sources are used to inform Step 3 to characterize background concentrations for cumulative NAAQS assessments. The identified ambient monitoring data for multi-source areas, along with existing and potentially new stationary source information, will inform Step 3 and Step 4 to characterize background concentrations for cumulative NAAQS and PSD increment assessments, as applicable.

Step 3: Determine Representativeness of Ambient Monitoring Data

The draft guidance discusses the differences between isolated sources and multi-source areas and the respective approaches for evaluating ambient monitoring data for use in cumulative NAAQS analyses. Both situations rely on the same underlying data and spatial analyses performed under Step 1. The draft guidance emphasizes that accounting for all source contributions (monitored and explicitly modeled) “is an inherently discretionary exercise with use of best professional judgment.” In addition to the considerations in Step 3, assessing data collection and processing, quality, and completeness is an important step in the process.

When reviewing available monitoring data, it is important to recognize that not all data reported to EPA has the required data quality needed for use in PSD modeling. In most cases, available monitored data will be collected by a state, local, or tribal agency. It is imperative to confirm the monitor parameter codes and type for agency monitors to ensure they were collected with the intent of being compared to the NAAQS or used for another regulatory purpose. If the data being evaluated were collected by a different party, the monitoring objective, monitor type (i.e., an EPA designated Federal Reference Method or Federal Equivalent Method), and the data
collection, validation, and reporting procedures should have been conducted according to an approved Quality Assurance Project Plan (QAPP). The QAPP should be fully reviewed to understand whether the data are appropriate for
use in regulatory modeling.

EPA’s ambient monitoring guidance for PSD⁴ Section 2.4.2 references conformance with 40 CFR Part 58 Appendix A or B requirements and that the monitoring data should achieve at least 80% completeness by quarter for each year of monitoring data. The data completeness provision will require an evaluation of daily or possibly hourly concentrations to understand if completeness is met, and if not, further justification on monitor suitability will be needed. The data should also meet EPA’s quality assurance and quality control requirements⁵.

In conjunction with confirming data quality and completeness, the data should be evaluated for the scale of representativeness and monitoring objective, in addition to the information gathered under Step 1. The scale of representativeness and monitoring objective should be identified when reporting the monitored data to EPA; however, these are generic classifications and may not be descriptive enough to understand whether the monitor is appropriate to use for the proposed project. If a monitor is operated by a state, local, or tribal agency, reviewing the agency’s Annual Network Plan is advised.

Section 3 of the draft guidance provides discussions for isolated source and multi-source areas. Both situations require an assessment of what the ambient monitoring data represents and the maps prepared under Step 1 should be leveraged. The maps, ambient monitoring data, any available emissions inventories, and dispersion characteristics should be carefully reviewed to assess whether any nearby sources (i.e., those that may need to be explicitly modeled) are impacting the monitor. In such cases, it may be appropriate to conclude that the monitoring data captures the source(s) in question, and thus the applicant may propose that the source(s) would not be explicitly modeled. This process should be performed as necessary for all monitors and potential nearby sources.

It is not uncommon for existing monitors to be located in a dissimilar source environment compared to the proposed project. For example, if the available monitors are in a populated urban area, whereas the proposed project is in a less populated or polluted area, the maps prepared under Step 1 should illustrate this. Figure 1 is an example of a proposed project that has no ambient monitors in its vicinity. The available monitors, shown in Figure 2, are in a densely populated urban area with numerous major stationary sources, international shipping port emissions, and interstate highways in the immediate area around the monitors.

In contrast, the area surrounding the proposed project is less populated, has fewer surface roads and highways, and has fewer major industrial sources than the monitoring area shown in Figure 1. The proposed project is downwind of smaller urban communities, light industry, and a handful of distant major stationary sources. Based on the available information and data, the monitors are impacted by much more criteria pollutant emissions than the project site; therefore, the monitors would likely be conservative and therefore adequately representative of the proposed project area. Additional analysis may be needed to finalize the monitor recommendations that include qualitative discussions of direct and secondary pollutants, as applicable.

Figure 1. Example of population density, major highways, Title V facilities, port activity, and airports in the vicinity of the proposed projects

Figure 2. Example of population density, major highways, Title V facilities, port activity, and airports in the vicinity of selected ambient air monitors.

Step 4: Determine Nearby Sources to be Explicitly Modeled

The final step of the framework is determining which nearby sources should be explicitly modeled for multi-source areas. The draft guidance provides discussion of this process and leverages the information gathered in Steps 1 through 3 and other information. As stated in the Guideline, Section 8.3.3(b)(iii) 1, EPA expects the number of explicitly modeled nearby sources to be few except in “unusual situations”. However, since the 2010 promulgation of the 1-hr NO₂ and SO₂ NAAQS, and release of the corresponding recommended significant impact levels (SIL), the spatial extent of the modeling domain for these pollutants as defined by the proposed project’s significance analysis can be quite large and even extend to AERMOD’s 50-km nominal distance.

Contrary to the Guideline, 1-hr NO₂ and SO₂ NAAQS compliance has been the constraining analysis for many major and minor source construction permits due in part to large modeling domains that include many nearby sources in the modeled inventory. The Guideline and the draft guidance emphasize the need for professional judgement when determining the explicitly modeled nearby source inventory. Given the compliance challenges associated with these
short-term pollutants, as well as the recently lowered annual PM_2.5 NAAQS, early consultation with the reviewing authority is recommended.

Environmental Justice Considerations

The draft guidance includes a brief discussion for determining background concentrations in at risk communities, particularly in multi-source areas where there may be the potential for modeled NAAQS or PSD increment violations. EPA provides references to tools that provide population demographics nationwide which may be experiencing disproportionate socioeconomic and environmental burdens. This information can be used to further assess project source impacts, representative ambient monitor locations, and/or whether to include additional nearby sources in a cumulative impact analysis. The draft guidance does not provide recommendations on how to evaluate the data gathered under the framework for unique considerations for at risk communities.

Timeline for the Final Guideline and Background Guidance

As of the writing of this article, EPA indicated publication of the final Guideline and background guidance in September or October 2024, depending on timing of Office of Management and Budget (OMB) review and EPA administrator signature. They emphasized during the 2024 EPA Regional/State/Local (RSL) Dispersion Modelers’ Workshop industry/consultant session on July 30 that the draft guidance is the first iteration for documenting the process of identifying background concentrations. The guidance will evolve over time as EPA gains more insight and information on its implementation by permit applicants and reviewing authorities. EPA also emphasized that applicable portions of the Guideline Section 8.3 and this guidance provide recommendations and are not requirements. EPA gathered state, local, and tribal agency input during the 2024 RSL workshop and in conjunction with public comments received on the draft guidance, will be finalizing the guidance and may include examples of framework implementation.

Patrick McKean, CCM is an air dispersion modeling subject matter expert; and Cara Keslar, P.E., is a managing consultant and ambient air quality monitoring subject matter expert with Trinity Consultants.

References

1. Proposed Revisions to Appendix W to 40 CFR 51, Published October 23, 2023, Federal Register Volume 88 No. 203
2. Draft Guidance on Developing Background Concentrations for Use in Modeling Demonstrations; EPA-454/P 23 001, October 2023.
3. Appendix W to 40 CFR 51, Published January 17, 2017, Federal Register Volume 82 No. 10.
4. Ambient Monitoring Guidelines for Prevention of Significant Deterioration (PSD); EPA 450/4 87 007, May 1987
5. Quality Assurance Handbook for Air Pollution Measurements, Volume II: Ambient Air Monitoring Program; EPA 454/B 17 001, January 2017.