New PM2.5 SIL Released

On April 30, 2024, EPA released "Supplement to the Guidance on Significant Impact Levels for Ozone and Fine Particles in the Prevention of Significant Deterioration Permitting Program," (the 2024 Supplement), which establishes a new PM2.5 annual "significant impact level" (SIL) of 0.13 ug/m3 and new Class I increments for that pollutant. It did not recommend any changes to the 24-hour PM2.5 or ozone SILs.

ANALYSIS

EPA states that in its 2018 "Technical Basis for the EPA's Development of Significant Impact Thresholds for PM2.5 and Ozone," it adopted an approach that considers the "inherent variability in the historical ambient monitoring data and the level of the corresponding standard." Because EPA revised the PM2.5 annual NAAQS in 2024, it was appropriate to reassess the historic variability in light of the new monitoring data. It was not necessary to reevaluate the SILs for the ozone, 24-hour PM2.5 NAAQS and increments "because the level of these NAAQS remains unchanged and inherent variability in the new ambient data was not meaningfully different than the results presented in 2018." 2024 Suppl. at 2.

EPA describes its methodological approach as applying the bootstrapping statistical approach. "The bootstrap technique, as applied in the analysis, models multiple scenarios to quantify the degree of air quality variability at an ambient monitoring site to account for the inherent spatial and temporal variability in observed ambient data." 2024 Suppl. at 5. Multiple iterations are run and a confidence interval is developed. The recommended SIL is then based on 50 percent of the confidence interval "to quantify the bounds of change in air quality that can be considered an 'insignificant impact' for the purposes of meeting requirements under the PSD program." EPA then set the SIL at the median of the relative variability at all monitors across the United States and multiplying it by the level of the associated NAAQS. Id.

Based on this analytical approach, EPA recommended the following SIL values for ozone and PM2.5:

For the 24-hour PM2.5 NAAQS, EPA states that its methodology gave a result of 1.6 ug/m3 but that it decided to retain the 1.2 ug/m3 because that value appears in 40 CFR 51.165(b)(2). EPA determined that it could not, as a matter of guidance, vary from its prior regulatory determination but that it could "conclude that impacts below 1.2 ug/m3 are insignificant at any location and can generally be determined not to cause or contribute to a violation of the NAAQS." 2024 Suppl. at 7.

Conversely, EPA noted that under its view, 40 CFR 51.165(b)(2) "does not address whether an impact below 0.3 ug/m3 causes or contributes to a violation of the NAAQS, the EPA and other permitting authorities retain discretion under this provision to determine on a case-by-case basis whether an impact below 0.3 ug/m3 will cause or contribution to a violation of the primary PM2.5 NAAQS." EPA then went on to state that "the EPA's judgement is that an impact below 0.13 ug/m3 generally will not be significant and can therefore be determined not to cause or contribute to any violation of the primary annual PM2.5 NAAQS that is identified." 2024 Suppl. at 7.

Turning to the increments, EPA recommended the following values:

As can be seen, EPA recommended using the NAAQS SILs for Class II and Class III areas. For Class I areas, EPA calculated the new SILs "by applying the ratios of the Class I and Class II allowable PSD increments to the NAAQS SIL values derived in the technical analysis. 2024 Suppl. at 7.

EPA provided additional details of its technical approach in the appendix.

COMMENTARY

In the final PM2.5 NAAQS rule, EPA stated that it intended to complete a reanalysis of the PM2.5 SILs. The 2024 Supplement delivers on that commitment. While many had anticipated that the SIL would be reduced by the level of reduction in the NAAQS, which would have resulted in a value of 0.15 ug/m3, EPA chose to continue with its approach of looking at variability. This is probably a more defensible approach than merely applying the a priori reduction in the NAAQS to the SILs as well given the role that the SILs play in the program. NSR Law Blog appreciates the EPA's effort to maintain a science-based approach to the SILs. With that said, it is disappointing that EPA chose not to apply that approach to the 24-hour SIL when the science suggests that the current SIL is too restrictive given the variability in background conditions. NSR Law Blog acknowledges that the presence of the regulatory determination is limiting, but urges EPA to consider a rulemaking to revise 40 CFR 51.165(b)(2) to reflect the current science.

As the NAAQS continue to come down, however, EPA will need to consider whether multiplying the variability against the NAAQS level continues to be appropriate. If, at some level, natural variability is an appropriate application of the SIL - which NSR Law Blog believes is likely true - then multiplying the natural variability by the NAAQS level may become indefensible as the NAAQS creep lower and it becomes harder and harder to distinguish the source's impact from natural variation. At some point, it can no longer be argued that causation exists, at which point EPA's administration of the modeling program may fall on the wrong side of Alabama Power and New York.