See the latest EHS federal and state regulatory updates due to COVID-19

In 2010, the U.S. EPA adopted a 1-hour sulfur dioxide (SO2) National Ambient Air Quality Standard (NAAQS). In 2013, the agency designated 29 areas as being in nonattainment of the standard and was then directed by court order in 2015 to establish designations for the remaining areas of the country in up to three additional designation rounds. (Two of those three have now been completed; the deadline for the final round is December 31, 2020.) To address the need for additional air quality data for use in making those attainment designations, EPA developed the SO2 Data Requirements Rule (DRR).1 Published in 2015, the DRR requires that agencies monitor or model ambient SO2 levels around sources that emit 2,000 tons per year (tpy) or more of SO2, as well as other areas that the air agency or EPA deems appropriate. The DRR gives air agencies the option of characterizing air quality through either air dispersion modeling, using AERMOD, or ambient monitoring. Alternatively, the agency can adopt enforceable emission limits of 2,000 tpy for subject sources.

The DRR set a January 2016 deadline for air agencies to identify sources that exceed the 2,000 tpy threshold, and a July 2016 deadline for identifying, for each identified source, whether it would monitor air quality, conduct air dispersion modeling, or establish an enforceable limit. Monitoring sites had to be operational by January 1, 2017, and modeling analyses or documentation of enforceable limits had to be submitted by January 13, 2017. Areas using air quality monitoring are required to submit three years of quality-assured data to EPA by early 2020.

Although the bulk of the rule is focused on data for the initial designations, complying with the DRR does not end once a designation is made. The rule also establishes ongoing data and reporting requirements for areas designated attainment. (Areas designated as nonattainment are subject to the requirements of the area's nonattainment plan and are generally not further covered by the DRR.) In areas for which the designation was based on ambient monitoring, monitoring must continue and the data be reported annually unless the monitors have been approved for shutdown by EPA. In areas for which the attainment classification was based on AERMOD modeling of actual emissions, agencies must submit an annual report to EPA by July 1 each year summarizing annual actual emissions of SO2 for sources included in the model. These emissions will be compared with the SO2 emission rates used in the initial designation modeling analyses to determine if a new modeling analysis should be conducted to confirm or change the initial designation. It is therefore important for owners/operators of SO2 sources to understand the basis for their area's designation classification and be aware of how changes in emissions may affect future requirements.

power plant_.jpg

SO2 Designations And DRR Background

The 2010 1-hour SO2 NAAQS was set at 75 parts per billion (ppb). To determine an area's attainment status with respect to that standard, EPA may use data from existing monitors, air dispersion modeling data, or data from new monitors operational by January 1, 2017. For the first round of designations, made prior to the 2015 finalization of the DRR, EPA used 2009-2011 ambient air quality monitoring data documenting exceedances of the 1-hour SO2 NAAQS to designate 29 areas in 16 states as nonattainment.2

The second3 and third4 rounds of designations relied on air dispersion modeling to determine expected SO2 concentrations for areas located outside those initially designated nonattainment areas and meeting the DRR thresholds. Guidance for that modeling was issued by EPA in a Technical Assistance Document (Modeling TAD), last issued in draft form in August 2016.5 Under the TAD, the air dispersion modeling analyses are to be conducted such that modeled concentrations would be equivalent to monitored concentrations if the monitors were sited in a network surrounding the subject sources. Therefore, EPA allowed modeling of actual emissions from sources included in the modeling and required the placement of model receptors only at locations where an ambient monitor could be sited (e.g., not over water). As an alternative, air agencies could model allowable emissions if the resulting modeled concentrations were less than the relevant NAAQS.

The second round of designations, finalized in June 2016 in time to meet the July 2, 2016 deadline, resulted in four areas being designated nonattainment and 57 areas designated as either attainment/unclassifiable or unclassifiable. These designated areas included the largest sources of actual SO2 emissions: those that emitted more than 16,000 tons of SO2 in 2012 or emitted more than 2,600 tons of SO2 and had an annual average emission rate of 0.45 pounds of SO2 per million British thermal units (lb/MMBtu) of heat input in 2012. In addition, a second set of Round 2 designations was made in December 2016, adding one nonattainment area and three unclassifiable/attainment or unclassifiable areas in Texas.6

The third round of designations, made just prior to the December 31, 2017 deadline, included the remaining DRR subject sources not included in Rounds 1 and 2 and for which new ambient monitors were not operational by January 1, 2017. Six of the areas were designated as nonattainment and the remaining areas (approximately 80) were designated as either unclassifiable or attainment/unclassifiable.

The deadline for the fourth and final round of designations is December 31, 2020, and applies to those areas for which new SO2 monitors have been installed and operated. Because the DRR requires that monitors be operational by January 1, 2017, data from 2017 to 2019 will be available for use in making the designations. (Monitor siting and operation should follow the requirements in EPA's Monitoring TAD last issued in draft form in February 2016.7) EPA has indicated that there are 71 areas for which monitoring is being conducted for the purpose of Round 4 designations.8

SO2 Data stackA designation of attainment/unclassifiable means that the DRR assessment found no evidence to indicate that the NAAQS is exceeded; if the available information is insufficient for the analysis to allow classification, the area is designated unclassifiable. For example, there may be questions regarding the representativeness of monitoring data for characterizing peak concentrations or regarding modeling for which non-default options were used in the DRR analysis. (Both attainment/unclassifiable and unclassifiable areas are treated as in attainment for the purpose of major source Prevention of Significant Deterioration permitting applicability-i.e., Nonattainment New Source Review requirements are not applicable for SO2.) There is no formal deadline for areas initially designated as unclassifiable to address questions on the underlying analysis and seek a revision to attainment/unclassifiable or nonattainment, although some states have chosen to address unclassifiable areas through additional monitoring and/or modeling.

DRR Ongoing Requirements

To date, the focus of much of the work by air agencies and industrial facilities related to the DRR has been on air dispersion modeling to determine attainment status and on siting and operation of new ambient monitors. Although there may be a tendency to view DRR requirements as having been addressed once an initial attainment designation is made, the ongoing data requirements found in the DRR have the potential to affect both air agencies and industrial facilities moving forward.

Regulatory Requirements

Ongoing requirements for areas designated as attainment or unclassifiable under the DRR are found in 40 CFR 51.1205; monitored areas are addressed in 40 CFR 51.1205(a). Generally, monitors must continue operation until their shutdown is approved by EPA. The criteria that EPA may use to determine if a monitor can be shut down are provided in 40 CFR 51.1203(c)(3). The air agency can apply to shut down the monitor if the design value is no greater than 50% of the NAAQS in the first three years of operation or in the second three years of operation for a monitor installed and operated for DRR purposes. After the fourth year of operation, the general requirements found in 40 CFR 58.14 for requesting the cessation of monitors for all pollutants apply. For installations where more than one monitor was cited in a network near an industrial source, there is nothing in the DRR regulatory language to preclude a request to discontinue operation of one or more monitors in the network if the criteria specified above are met for one or more monitors in the network, but not all. (It should be noted that both the DRR and 40 CFR 58.14 refer to an SO2 "monitor" rather than a network of monitors.)

Requirements for areas designated attainment or unclassifiable based on modeling are found in 40 CFR 51.1205(b) and (c). For areas where the designation was based on modeling of actual emissions, the air agency must submit an annual report to EPA by July 1, with the first report due by July 1 of the calendar year after the effective date of the area's designation. Because the effective date of the Round 2 designations, which were the first to be based on modeling, was in 2016, those areas were required to submit reports by July 1, 2017, and July 1, 2018. The effective date of Round 3 designations was in 2018, so the first reports for these areas will be due by July 1, 2019. For areas for which the air agency demonstrated that the area would meet the 1-hour SO2 NAAQS based on allowable emissions, no ongoing reporting is required.

In the DRR, EPA requires that the annual report be prepared as either a stand-alone document available for public inspection or as an appendix to the air agency's Annual Monitoring Network Plan, which is also due by July 1 of each year. The report must document the annual SO2 emissions from each applicable source in each modeled area and must also assess the cause of any emissions increase from the previous year and recommend whether additional modeling is needed to characterize air quality in the area. EPA is to review the report and recommendations and may require that the air agency conduct updated modeling for one or more areas, in which case the results of the modeling analysis must be submitted to EPA within 12 months.

An air agency can be exempted from the ongoing reporting requirement if an air quality analysis is submitted and approved showing that ambient impacts are no greater than 50% of the 1-hour SO2 NAAQS. Although the DRR does not specify the requirements for this analysis, it presumably would include current meteorological data, background SO2 concentrations, and actual emissions data for sources from the most recent three-year analysis period, in conformance with the SO2 Modeling TAD. Additionally, an air agency can become exempt from future reporting if it submits revised modeling showing that the 1-hour SO2 NAAQS is met using allowable emissions data. In some cases, EPA has already reviewed and approved requests for exemptions from the ongoing reporting requirements. For example, a request submitted to EPA Region 1 in December 2016 by the Connecticut Department of Energy and Environmental Protection (CT DEEP) contained an air dispersion modeling analysis showing maximum SO2 concentrations of 15.8 ppb or 21% of the 1-hour SO2 NAAQS. In an August 2018 letter, EPA Region 1 provided a written response confirming that CT DEEP had no ongoing reporting requirements under 40 CFR 51.1205(b) because the modeled concentrations were less than 50% of the NAAQS.

If modeling or monitoring data are submitted that indicate the 1-hour SO2 NAAQS is no longer met for a particular area, EPA may take a variety of actions, including requiring the air agency to adopt enforceable emission limits to ensure compliance with the NAAQS, designating or redesignating the area to nonattainment, or issuing a State Implementation Plan (SIP) Call. The DRR does not specify a timeline for these actions to be completed.

Recommendations For SO2 Sources

While the obligation to submit ongoing DRR reports to EPA falls on the air agencies, there could be impacts on sources of SO2 emissions. Sources included in a dispersion modeling analysis used to make an unclassifiable or attainment designation should be aware of whether the analysis relied upon allowable SO2 emissions data or actual emissions data. There is little possibility of additional requirements being triggered under the DRR for areas for which allowable emissions were modeled. However, as discussed below, if actual emissions data were used in the dispersion modeling analysis, future changes in actual emissions could have an impact on requirements.

Sources included in DRR modeling based on actual emissions should be mindful of the SO2 emission rates used in the designation modeling and perform a comparison when reporting actual SO2 emissions in future years to determine if an increase or a decrease in emissions will be reported by the air agency to EPA. If actual emissions appear to be trending upward, sources should consider conducting an internal dispersion modeling analysis to evaluate whether the increases could lead to a formal analysis by the air agency and, if necessary, take mitigation steps to ensure that compliance with the 1-hour SO2 NAAQS is protected. Conversely, if actual emissions are trending downward, an internal analysis could help determine if the source should suggest to the air agency that a report be submitted to EPA documenting that modeled concentrations are below the reporting exemption threshold (i.e., 50% of the NAAQS) and thus qualify for exemption from future reporting requirements.

Sources should also be aware that increases in emissions from other sources in the area could trigger the need for revised dispersion modeling for the area. Ensuring familiarity with the air agency's annual reporting format and the availability of these reports can be helpful to identify trends in levels of annual SO2 emissions that may lead to remodeling requirements due to emissions changes from another nearby source. This information could help sources take proactive steps to understand the possibility and impacts from future nonattainment designations or the imposition of SO2 emission limitations to preclude nonattainment designations.

Additional emissions reporting requirements or a change in the deadline for submitting actual SO2 emissions data could also result from DRR reporting requirements. Although many states require that only the largest sources report annual actual emissions, the DRR and Modeling TAD recommend including smaller sources in modeling analyses if they are located sufficiently close to larger SO2 sources such that the combined concentrations could affect compliance with the 1-hour SO2 NAAQS. Air agencies may therefore issue a special request for SO2 emissions data from these smaller sources outside of normal air emissions inventory reporting requirements. Also, while many states have deadlines for annual actual emissions reporting for the previous calendar year in the first quarter, other states have deadlines falling closer to or at the same time as the submittal date for the DRR ongoing report. Air agencies may therefore also ask certain sources to submit SO2 data earlier in order to validate and evaluate the data in sufficient time to make informed recommendations regarding whether additional modeling is necessary for each area included in the report.


Even after initial attainment designations are finalized, air agencies and sources must comply with DRR requirements for ongoing data gathering and reporting in order to assure continuing attainment of the 1-hour SO2 standard. It is important that sources understand these reporting requirements as well as how possible changes in emissions could trigger additional review. In particular, if the area's attainment designation is based on dispersion modeling using actual emissions data, sources should be aware of how the emissions data used in the initial designation modeling compare with the actual calendar year emissions data not only for its own facility but also for other nearby sources included in the designation modeling. This will help facilities plan proactively to ensure continued attainment of the 1-hour SO2 NAAQS for their area and possibly avoid future reporting requirements or additional emission controls.


1 Data Requirements Rule for the 2010 1-Hour Sulfur Dioxide (SO2) Primary National Ambient Air Quality Standard (NAAQS), 80 FR 51052, August 21, 2015.
2 Air Quality Designations for the 2010 Sulfur Dioxide (SO2) Primary National Ambient Air Quality Standard, 78 FR 47191, August 5, 2013
3 Air Quality Designations for the 2010 Sulfur Dioxide (SO2) Primary National Ambient Air Quality Standard - Round 2, 81 FR 45039, July 12, 2016.
4 Air Quality Designations for the 2010 Sulfur Dioxide (SO2) Primary
5 SO2 NAAQS Designations Modeling Technical Assistance Document, U.S. EPA Office of Air Quality Planning and Standards, August 2016 (DRAFT)
6 Air Quality Designations for the 2010 Sulfur Dioxide (SO2) Primary National Ambient Air Quality Standard- Supplement to Round 2 for Four Areas in Texas: Freestone and Anderson Counties, Milam County, Rusk and Panola Counties, and Titus County, 81 FR 89870, December 13, 2016
7 SO2 NAAQS Designations Source-Oriented Monitoring Technical Assistance Document, U.S. EPA Office of Air Quality Planning and Standards, February 2016 (DRAFT).
8 Air Quality Designations for the 2010 Sulfur Dioxide (SO2) Primary