Ms. Darcy Bybee
Director, Air Pollution Control Program
Missouri Department of Natural Resources
P.O. Box 176
Jefferson City, MO 65102-0176

Dear Ms. Bybee,
The U.S. Environmental Protection Agency appreciates the opportunity to participate in the Missouri Department of Natural Resources' State Implementation Plan development process. Enclosed are the EPA's comments regarding the MoDNR's public noticed SIP revision titled "Interstate Transport Provisions for the 2015 Ozone Standard". 
The EPA is committed to working with the MoDNR to ensure that the SIP submission contains the information required to meet the Clean Air Act requirements of Section 110(a)(2) and is protective of the 2015 Ozone National Ambient Air Quality Standard.
 							Sincerely,
							
                                    Mark A. Smith, Director
                                    Air and Waste Management Division
Enclosure


Background:

Section 110(a)(1) of the Clean Air Act requires states to submit a State Implementation Plan revision to the U.S. Environmental Protection Agency within 3 years after the promulgation of a new or revised National Ambient Air Quality Standard. The EPA revised the 8-hour Ozone NAAQS in October 2015 prompting the requirement for the state to submit a SIP revision by October 2018. The purpose of the SIP revision is to provide for the "implementation, maintenance, and enforcement" of the new or revised NAAQS. The EPA has historically referred to these revisions as "infrastructure SIPs" or "iSIPs" to distinguish them from other types of SIP submittals (such as nonattainment area SIPs, maintenance SIPs, etc.).
Section 110(a)(1) addresses the timing and general requirements for an iSIP submission, and section 110(a)(2) provides details concerning the required content of the submission (often called iSIP "elements"). The comments in this document are specific to "elements" of section 110(a)(2)(D)(i)(I). 
Comments:
1)  In Section 3.5.5.1, "2009-2011 Meteorology was Conducive to Ozone Formation" the discussion regarding the Allegan, Michigan monitor includes an analysis of whether meteorology during the 2009-2011 period was conducive to ozone formation. This analysis states that although the meteorological conditions in 2009 were not conducive to ozone formation, the meteorology in both 2010 and 2011 was "clearly conducive" to ozone formation and that therefore "the inclusion of these two years in the alternative base period means that the base period as a whole did include meteorology that was conducive to ozone formation.". The EPA recommends the MoDNR include additional support for the statement that the 2009-2011 base period "as a whole" was conducive to ozone formation despite the inclusion of 2009, which according to Figure 1 was the 5[th] coldest summer on record for Michigan.
2)   In Section 3.5.2, "Other Considerations Regarding the Allegan Monitor", begins a discussion related to the Allegan, Michigan monitor, and specifically an analysis of Missouri's "share of emission reduction obligations" in comparison to other upwind states. Rather than suggesting a remedy for Missouri's proportional share of the downwind air quality problem, Missouri instead proposes to conclude that the state's share is sufficiently small and within a 5 ppb margin of error of the modeling.  EPA recommends the MoDNR further explain how they derived the 5ppb modeling error as it does not match region-specific or state-specific performance stats in Appendix A of the Air Quality Modeling Technical Support Document (TSD) for the Updated 2023 Projected Ozone Design Values.

 In Table 7 of Section 3.5.2.1, MoDNR lists a number of anticipated retirements that were
apparently not reflected in EPA's 2023 modeling.  EPA recommends the MoDNR provide a basis for asserting that these retirements are anticipated before 2023 (e.g., public announcements or filings) and if this information is not available, EPA recommends removing the table from the SIP. Moreover, while MoDNR lists the anticipated tons of reduction from each unit, the SIP would be further strengthened if MoDNR evaluated potential shifts in generation resulting from the anticipated retirements and how that might affect the state's overall emission trends and downwind contribution to the Allegan maintenance receptor. 

  In Section 3.5.2.2 "International and Initial and Boundary (I&B) Contributions", EPA recommends MoDNR address why it is technically and legally supportable to subtract 100% of anthropogenic and non-anthropogenic contributions from Canada and Mexico from the projected design value (and why it is appropriate to also subtract 2% of initial and boundary conditions), especially when the total contribution from these sectors is barely greater than 1 ppb in Missouri's calculation.

 EPA encourages Missouri to include a comprehensive analysis of the existing controls on sources, NOx reduction potential, and cost of controls for sources, including EGU and non-EGU sources or facilities.  An assessment of factors that influence EGU emissions (such as but not limited to allowance prices, fuel prices, and enforceable limits) in the future is recommended.   EPA encourages Missouri to include a comprehensive assessment of all of the NOx emitting non-EGU facilities, including their NOx emissions sources/units, reduction potential, and cost of any potential control options. Showing that no further cost-effective reductions are available, if that is the case, will bolster the strength/approvability of the submittal.

 EPA recommends MoDNR strengthen its SIP by showing that the relied upon emission rates used in the 2023 modeling for SCR controlled units (that reflect operation) are reasonable estimates for future expected operations of those controls (in particular for units that have shown high degrees of variability in past control operation (e.g., Thomas Hill, New Madrid)). Adding certainty to its emission projections through these types of considerations is useful for demonstrating no risk of emission reductions being offset by changes in control operation.  

 In Section 1.1, "Good Neighbor Obligations for the 1997 and 2008 Ozone NAAQS, the MoDNR states that the NOx SIP Call and the NOx Budget Program addressed the 1997 ozone standard.  While the NOx SIP Call was originally promulgated to address both the 1979 and the 1997 ozone standards, the EPA subsequently stayed the implementation of the NOx SIP Call for purposes of addressing the 1997 standard.  65 Fed. Reg. 56,245.  Accordingly, when the NOx Budget Program was implemented, it only addressed the good neighbor provision obligations for the 1979 ozone standard.  EPA recommends MoDNR revise the language in this section to reflect the clarification.  

 In Section 1.2.1, "CAA 110(a)(2)(D)(i)(I) and EPA's 4-Step Process", the MoDNR incorrectly recites the third step of the four-step process that the EPA has used to evaluate interstate ozone transport.  EPA recommends the MoDNR revise the description of step 3 as follows: "Identify the requisite level of emission control necessary to address the upwind state's significant contribution to the air pollution problem or interference with maintenance in the downwind states."

 In Section 1.2.4 "EPA Memo-Alternative Methods of Identifying Maintenance Monitors", the MoDNR incorrectly describes the EPA's October 2018 memorandum as stating that states "may not need to address identified maintenance monitors" if they meet certain criteria based on an alternative base year. EPA recommends the MoDNR revise the language to reflect the memorandum outlines alternative approaches (or flexibilities) that states may use to identify a monitor as a maintenance receptor, and with an appropriate technical demonstration, states may eliminate a site as a maintenance receptor based on currently measured clean data or use a design value from the base period that is not the maximum design value. 

 In several locations throughout the document, the MoDNR incorrectly refers to 1 ppb and 1% of the NAAQS as "significance thresholds" or "significant contribution thresholds." The EPA notes that such as thresholds are more correctly referred to as a contribution threshold and that a contribution threshold alone is not intended to represent the level of a "significant contribution," but rather a contribution that merits more consideration to determine whether a state impacting a downwind receptor above that threshold will significantly contribute to nonattainment or interfere with maintenance of the NAAQS.  EPA recommends the MoDNR revise the language throughout the document to reflect the change.    

