1 Rule Purpose
1.2 Revise and streamline the electronic reporting requirements of MATS 
Comment 1: Commenters 20593, 20595, 20597, 20599, 20607, 20609, 20610 and 20612 supported EPA's efforts to streamline the MATS reporting requirements. Commenter 20595 supported the sole use of the XML format and ECMPS for MATS reporting. Commenters 20597, 20599, 20610 and 20612 supported the use of the ECMPS as the single reporting system, and agreed that EGU owners or operators would use one familiar electronic reporting system, instead of two separate systems, reducing their burden. The commenters noted that there are further revisions EPA should make. Commenters 20597, 20610 and 20612 stated there are duplicative and unnecessary reporting requirements that are a burden on reporters. Commenter 20610 stated that EPA can still make additional changes that will ease the proposed reporting burden while still providing adequate information to document substantial compliance with the regulations. Commenter 20609 stated there are unresolved issues and inconsistencies from prior rulemakings that have complications for compliance reporting that EPA will have to address in this rulemaking. 
In response to EPA's statement regarding a reduction in the reporting burden through the use of a single system, commenter 20597 requested EPA to reduce the burden of the MATS reporting by the MATS reporting requirements by reporting all the data through the ECMPS Client Tool. The use of ECMPS will benefit both EPA and the industry while improving the quality of the reported emissions and performance stack test data. The commenter also noted that while the revisions may be intended to streamline the original reporting requirements, the revisions represent a substantial change to the way that MATS data are reported. In addition to addressing the existing reporting requirements, the proposal also includes three new appendices and new reporting requirements.
Commenter 20599 noted concerns with certain proposed amendments, particularly the timeline EPA has proposed for sources to begin following the new proposed data collection requirements and changes to the quarterly emissions and CEMS Performance Evaluation Reports.
Commenter 20610 supported the current practice of submitting test reports electronically within the interim reporting structure, and understood the EPA was making improvements to this system. The commenter said this proposal now suggests that this worthwhile and well-designed tool be burdened with many additional requirements beyond what is necessary to demonstrate compliance. 
Response 1: The Agency acknowledges the support from commenters for use of the Emissions Collection and Monitoring Plan System (ECMPS) Client Tool as the sole means for electronic reporting. As described in later responses, the rule has been adjusted to account for many of the commenter's claims of duplicative and unnecessary reporting.  As the commenter notes, some reporting frequencies have been changed to quarterly to better mirror the requirements of the industry-requested ECMPS  Client Tool. As shown in later responses, the extension of quarterly PDF file submissions will continue through December 31, 2020. After that date, MATS electronic reporting via the ECMPS Client Tool will be required. As discussed in later responses, the Agency finds that there are no superfluous reporting requirements: each is necessary for compliance determination. 

Comment 2: Commenter 20606 stated EPA has failed to work with state partners to coordinate any streamlining efforts so that EGU's would not be subject to multiple reporting regimes for the same information. The commenter stated that the Michigan Department of Environmental Quality-Air Quality Division (AQD) confirmed that EGUs would still be required to file hard copy reports based on the current AQD reporting structures and guidelines.
Response 2: The Agency finds the commenter's concerns misplaced: EPA has no ability to change unilaterally state, local, or tribal statutes which may contain separate reporting requirements. The Agency continues to work with other regulatory bodies, sharing the benefits of a single electronic reporting system. The commenter may find that messaging could be successful with its state, local, or tribal authority. 

Comment 3: Commenter 20594 recommended that EPA clarify whether there are paper reporting requirements for MATS in addition to those proposed involving ECMPS. The commenter noted that the proposed rule goes on to discuss requirements for submittal of data in ECMPS, but it is silent on whether the proposed new reporting scheme replaces all paper reporting. The commenter said the rule also does not cover whether the ECMPS submittal will also go to the permitting authority (the state). The commenter recommended EPA should specify whether the new reporting scheme replaces all paper submittals and courtesy copies to EPA (Region and Headquarters) and to the permitting authorities.
Response 3: Use of the ECMPS Client Tool to submit a required MATS report or notification satisfies any requirement in Subpart A of Part 63 to submit that same report or notification to the EPA Regional or State Office, as described in section 63.10021(f). Once submitted to ECMPS, the electronic report will be available for EPA's central data exchange / compliance and emissions data reporting interface (CDX / CEDRI), where only the regulatory authorities and certifiers registered with CDX / CEDRI can view and download the electronic report.  After a 30 or 60 day processing period, the electronic report will be pushed to WebFIRE, where anyone can view and download the electronic report.  It is important to note that while federal reporting requirements are fulfilled when electronic reporting requirements are used, other, nonfederal regulatory authorities operating in accordance with their own statutory authorities may still require reporting via other media.
1.4 Reduce the reporting burden via the use of a single reporting system 
Comment 1: Commenter 20612 disagreed that the proposal will have no significant impact on stakeholders. The commenter stated that the revisions represent a substantial change to the way that MATS data are reported, even though the intent was to streamline the original reporting requirements. The commenter noted the addition of three new appendices. The commenter urged the EPA to provide adequate time to allow changes to be implemented in a reasonable and orderly fashion that will ensure consistent and accurate reporting of the compliance data.
Response 1: The Agency maintains that in merging the separate electronic data reporting systems  -  which is being pursued at industry's request  -  the reporting requirements are shifted from 2 systems to 1 system.  Even though electronic monitoring submission frequency has been increased and provisions for PM, HCl, and HF CEMS, as well as PM CPMS, have been added, in order to align with using the industry-requested ECMPS Client Tool, the Agency finds the change yields an overall burden reduction in reporting and recordkeeping requirements. This annual burden reduction of about $4.2 million per year provides no significant impact on EGU owners or operators, as that reduction amounts to about one-tenth of one percent of projected net utility income for 2015 (see 81 FR 67070, September 29, 2016 and page 2-15, https://www3.epa.gov/ttn/atw/utility/mats_final_ria_v2.pdf).
Comment 2: Commenter 20612 stated that EPA has no basis for saying that the proposed MATS data elements that would be reported electronically for performance tests and CEMS performance evaluations remain unchanged. The commenter said the original requirement to submit data using the ERT is not synonymous with a requirement to provide all the data within the forms represented by all the data input screens in the ERT as EPA seems to be suggesting. It only means that a source would have to report the applicable required performance test data in the format that the ERT generates. The commenter suggested that the situation would be analogous to saying that an individual must report tax information using the forms provided by the Internal Revenue Service, allowing a source to report the data in the ERT that was required by the rule and that applied to their source under the original MATS. The commenter stated EPA needs to recognize that MATS sources were never required to report all the data within the ERT. Rather than just assume that it has to replicate some assemblage of ERT within Appendix E (or any new XML reporting format for stack test data), the EPA should be reflecting on how best to report the data: what data needs to be reported, what portion of that data is critically required in a numeric XML format, and what data is adequately or better addressed within a detailed stack test report that can be submitted electronically in a PDF (or equivalent) file.
Commenter 20602 said the proposal changes the reporting requirement from semiannual to quarterly. This would increase the burden significantly, doubling the annual reporting needed with little change in the flow of information, since the same information would be reported either way. The commenter suggested that the semiannual reporting criteria be maintained rather than switching to quarterly reporting.
Commenter 20606 stated the new reporting requirements significantly increase the amount of information industry has to report to the EPA, as well as the frequency at which such information is submitted (i.e., quarterly versus semi-annually). The commenter said this proposal's shift towards submittal in XML format presents new burdens on the affected sources that EPA has failed to adequately justify. The commenter identified the following information as being submitted in addition to what was required in the February 16, 2012 final rule. 
 Hourly PM CEMS concentrations as mg/wscm, as well as mg/acm, and the temperature and pressure values used to convert from mg/acm to mg/wscm if the PM CEMS measures in units of mg/acm;
 Hourly PM emission rate in units of the applicable standard;
 The results of all PM CEMS daily Quality Assurance (QA) assessments; Hourly average output for PM CPMS or codes indicating that the CPMS was out-of-control, as well as percent monitor availability (PMA) information;
 30- or 90-boiler operating day (or group boiler operating day, for units participating in averaging plans) records for units employing CEMS or CPMS for compliance purposes or participating in an averaging plan;
 Summary records for any stack testing or 30- or 90-boiler operating day Low Emitting Electric Generating Units (LEE) test completed during each calendar quarter. 
The commenter said this proposal will result in a decrease in staff time devoted to MATS reporting obligations as compared to the February 16, 2012 final rule. At best, the proposed rule entails just as much effort as that associated with the final rule. Although some level of efficiency may be gained in the submittal process by using a single EPA electronic reporting tool as opposed to multiple tools, the EPA has expanded the amount of information EGUs are required to submit electronically and added additional electronic summary records that were not initially required. 
Commenter 20608 stated the proposed requirements represent a substantial improvement over the original MATS reporting requirements; but still contain significant amounts of areas where changes are needed to ease the burden of reporting. The commenter did not agree that this was a simple formatting change that will have "no significant impact on stakeholders." The commenter did not support using ECMPS as the single MATS reporting system instead of using the two reporting systems (ECMPS and ERT/CEDRI).
Commenter 20607 supported EPA's effort to streamline reporting, but stated some provisions in the proposed rule would increase the reporting burden on companies and state agencies, and should be amended in the final rule. The commenter said failure to do so would thwart EPA's attempt to simplify and facilitate reporting, as well as consume the cost savings gained by converting to a single reporting system.
Response 2: As mentioned earlier, while the rule increases some submission frequencies and includes provisions for new types of CEMS and CPMS in order to make the electronic reporting system consistent with the industry-requested use of the ECMPS Client Tool, the Agency determined an overall decrease in industry burden resulting from those changes. However, this annual reduction is very small  -  on the order of one-tenth of one percent  -  when compared to annual net income of the industry, so there is no significant change in industry reporting or recordkeeping burden. 
While use of the Agency's Electronic Reporting Tool (ERT), which collects all relevant compliance information for emissions testing, was mandatory in the original promulgated rule, a separate XML schema that identifies components of relevant compliance information has been prepared - at industry's request  -  and is available for use in electronic reporting (see EPA-HQ-OAR-2009-0234-20586 at www.regulations.gov) . Use of the ERT remains a viable option for an EGU owner or operator to submit an emissions test, but its use is no longer mandatory. 
As discussed in the proposal, this rule reduces reporting burden on industry by over $4.2 million per year while maintaining the burden on state, local, and tribal authorities (see 81 FR 67070, September 29, 2016).
2 Interim Reporting Period
2.1 Extension of interim PDF reporting process from April 16,2017 to December 31, 2017 
Comment 1: Commenters 20594, 20596, 20597, 20599, 20600, 20607, 20609 and 20612 stated EPA has not allowed enough time to implement the proposed changes. The proposed rule would require sources to begin following the new proposed requirements for data collected starting on January 1, 2018. Sources and EPA will need more time to implement these changes. The proposed changes will require sources (and/or their software vendors) to make substantial changes to the data acquisition and handing system (DAHS) software that they use for MATS reporting. Programmers will need to interpret the regulations and reporting instructions to ensure that the DAHS software addresses the various reporting options that are allowed under the MATS rule and the different monitoring scenarios that might be used at numerous plants. The software vendors will need to make these changes while supporting the on-going reporting under the existing requirements, which may necessitate cumbersome parallel revisions to their reporting platforms. Iterative rounds of beta testing may be needed to ensure the proper performance of software. The beta testing will require interaction between the software vendors, utility personnel and, at times, EPA to identify and resolve issues prior to the date that actual reporting begins. Stack test companies will also need to develop spreadsheet and/or other tools to accommodate the proposed changes to the way performance test data are reported under Appendix E. All these tasks take time to do properly. Commenter 20595 added that the agency is requesting information on each data set of a RATA, 30-day averages, and other requirements that have not normally been part of the ECMPS software package or reporting requirements within ECMPS. Commenter 20594 stated concern that a shortened transition period would limit vendor availability, making it even harder to complete all of this work in the time frame proposed. 
Commenters 20594, 20597, 20599, 20607, 20609 and 20612 noted that implementation of the changes is also predicated on the Agency both finalizing the rule changes and providing reporting instructions. Sources and software vendors cannot assume that the reporting requirements will stay the same between the proposed and final rules. In addition, EPA did not provide a draft of the reporting instructions for the proposed stack test data for the PM CEMS or HCl CEMS related reporting requirements, which further limits the degree of preparation that can be made prior to the release of final version (and associated reporting instructions). Since the requirements may change, the Agency will need time to develop draft reporting instructions once it finalizes the reporting requirements. And then, EPA will, likewise, need time implement the new requirements within ECMPS and the database systems that will ultimately receive the data. 
Commenters 20597, 20604, 20607, 20608, 20609, and 20612 disagreed that EPA will be able to finalize the rule by April 16, 2017. The commenters said that all stakeholders will need more than six months to get the program up and running. Commenters 20597, 20603, and 20608 urged the agency to carefully consider comments and not rush finalizing this rule by the date proposed. 
Commenter 20612 stated that there are many changes that will need to be made, and suggested that EPA refer to the schedule of activities from when CAMD was implementing the new XML reporting requirements that were included in the original MATS rule. Commenter 20609 said the timeframe will need to include solicitation of feedback on the remaining reporting instructions, discussions to resolve rule ambiguities or inconsistencies, and EPA's revisions to the ECMPS Client Tool. The commenter added all stakeholders (EPA, EGUs, software vendors, and source testing companies) will need time to properly implement the new requirements, including time to resolve new questions that will arise during implementation. Commenter 20604 noted that MATS implementation took over two years from rule publication to the first quarterly reporting compliance requirement. Extra time was needed to address exceptions and special circumstances, via multiple schema, Client Tool, and Reporting Instruction version releases. 
Commenters 20597, 20608, and 20609 urged EPA to ensure that the current transition period is long enough to prevent reversion to the 2012 reporting system while EPA completes this rulemaking, and to reconsider the amount of time it has proposed for implementation of the new requirements. Commenter 20593 requested that the Agency take specific action to delay the expiration of the interim reporting provisions under MATS and provide adequate time for implement reporting requirements in the final rule. The commenter said to ensure the work accomplished by the Agency and industry groups over that last four years are not undone by a timing technicality; the transitional reporting period deadline must be extended. Commenter 20600 noted that based on their experience with past revisions to electronic reporting requirements, they anticipate software development time will be at least six to eight months, followed by approximately three months for implementation and deployment to our customers. 
Commenters 20597, 20608, and 20612 recommended the EPA extend the current interim reporting requirements and mandate the new reporting begin with the data collected eight quarters following the date when the final rule is published. Commenter 20612 disagreed that the transition can take place in six quarters. Commenter 20609 requested that EPA lengthen the implementation period to a minimum of 6 calendar quarters. The commenter suggested that the rule include a start time that is expressed as a number of full calendar quarters after the effective dates of the final rule, instead of January 1, 2018. Commenter 20593 said the Agency should issue a stand-alone, direct final rule to extend the interim reporting period in the rule to January 1, 2020. Commenter 20594 suggested that the transition date be delayed until the first quarter of 2019 with 2018 serving as an optional full-scale trial of the new software, which is similar to the 2008 roll-out of ECMPS. Commenter 20599 encouraged EPA to extend the timeframe for temporary use of PDF submissions in the ECMPS Client Tool to December 31, 2018. Commenter 20603 urged the EPA to extend the interim reporting requirements, providing for submittal of certain reports through ECMPS in PDF format, indefinitely until such time as the agency completes the rulemaking transitioning all reporting to ECMPS. Commenter 20596 supported extending the interim reporting process until January 1, 2020. This will allow time to complete the development of the final reporting process that includes all the necessary compliance reporting elements. The commenter recommended a phased in approach over the next three years (2017  -  2019). This phased in approach is similar to the successful process adopted by the Agency which was implemented back in the 2007-2008 timeline for the transition under the ARP from the old ascii file to the current XML EDR file format: 2017 will be used by the Agency to flush out the requirements, reporting instructions and schemas to complete the electronic reporting process; 2018 will be used by the DAS vendors to develop the appropriate software changes, perform internal testing and to implement beta testing of the software that they will need to develop; and, 2019 will be used to roll out the software by the software vendors to the end users and make whatever configuration changes necessary to the hundreds of software systems which are operating in the field to generate clean, complete and error free submissions of Monitoring Plan records, quality assurance test data and results, and the final hourly and summary records for the recorded emissions. The commenter also recommended an ECMPS beta period in 2019 where vendors and facilities can do trial submissions for testing purposes. Commenters 20597 and 20608 requested this be extended immediately until January 1, 2020, with a brief final rule that makes only that specific change. Once the interim reporting period is extended, EPA would then have more time to complete the other revisions in a second final rulemaking. Just as EPA initially indicated that it would issue these revisions as a direct final rule but later revised its approach when it deemed a proposed rule to be a better option, the Agency should now also modify its original planned deadline. Commenter 20600 requested extending the interim reporting provisions through at least 2018. Commenter 20607 said EPA should extend the deadline to January 1, 2020. Commenter 20604 suggested that EPA extend the compliance deadline to July 1, 2018 at the earliest. To attain this date, two fundamental activities must be accomplished: (1) This proposal must be published by the expiration of the interim reporting rule on 4/17/2017, and preferably earlier, and (2) EPA must publish final schema and ECMPS Reporting Instructions in a timely manner after finalizing this rule. The commenter requested that EPA commit to publishing a final schema with Reporting Instructions within four (4) months of the final rule signature date. This would allow DAHS vendors, along with the stack testers, eight (8) months to work with the final version of schema and reporting instructions, and give EPA eight months to release beta versions of Client Tool and resolve issues with beta xml submissions. The commenter said this is a more reasonable time frame for all parties concerned to achieve an orderly transition to the new reporting requirements. The suggested July 1, 2018 compliance date would need to be delayed accordingly if EPA is unable to meet the above timeline. 
Commenters 20597, 20608, and 20612 noted that extending the interim reporting period will have no impact on either a source's obligation to comply with the MATS emissions limits or the requirement to conduct monitoring or testing to demonstrate such compliance nor will extending the interim reporting prohibit the EPA from using the reported data to assess compliance. 
Response 1: EPA agrees that the proposed December 31, 2017 end date for the interim PDF reporting process did not allow enough time to finalize the rule and implement the new reporting requirements.  In view of this, as suggested by some commenters, on April 6, 2017, the Agency published an extension of the interim reporting process through June 30, 2018 (see 82 FR 16736, April 6, 2017).  New requirements include hourly reporting of PM CEMS and PM CPMS data and reporting of certification and QA tests of HCl CEMS using PS 18 and Procedure 6. Therefore, today's rule further extends the interim PDF reporting through December 31, 2020, to allow enough time for software programming and all of the other activities associated with implementation of the single data system initiative. 
The rule's new deadline for ending interim PDF reporting  -  some 13 calendar quarters after publication  -  amply addresses the commenters suggested implementation dates for electronic reporting. Commenters suggested implementation dates ranged from 2 to 8 calendar quarters after publication of the final rule in order to allow sufficient time so that software can be developed and tested prior to the deadline.  The Agency notes that software developers have been able to start their efforts since September 2016, when the XML schema for performance tests was placed into the docket (see EPA-HQ-OAR-2009-0234-20586 at www.regulations.gov).
3 63.10011-Initial compliance with emission limits and work practice standards
Comment 1: Commenter 20612 stated §63.10011(e) was not revised as described in the preamble. The commenter believed the change described in the preamble is unnecessary, and they recommended that EPA not include such a change in the final rule.
Response 1: The rule corrects this inadvertent oversight.  The revisions to §63.10011(e) described in the preamble of the proposed rule have been finalized.
Comment 2: Commenter 20609 stated that the general provisions require reporting all reports to the delegated authority and the EPA regional office (unless EPA waives the requirement), and after a Title V permit has been issued, reporting the Notification of Compliance Status and performance test results to the appropriate permitting authority. The commenter argued that EPA has no basis to require EGUs that are reporting information directly to a central EPA reporting system to duplicate those reports in separate submissions to states and regions, and that they have access to the relevant information through the SMS. The commenter noted a few places in the proposal where EPA should address this issue. The commenter said EPA must make other revisions to make clear that such duplication is not required. Specifically, the commenter suggested EPA either amend Table 9 to make clear that the duplicative reporting in the general provisions is not applicable to Subpart UUUUU and/or by including in § 63.10011(f) a statement that: "Submission of a report using the ECMPS Client Tool satisfies any requirement under subpart A of this part to submit that report (or the information in that report) to the appropriate EPA Regional Office or State agency of a delegated authority." The commenter noted EPA did this in a similar issue with respect to written "excess emissions" reports under § 63.10(e)(3)(v). The commenter said if EPA does not do this, EPA must explain and solicit comment on its rationale for requiring duplicative reporting of this information and how it can be reconciled with the proposed electronic submission requirements in § 63.10031(b)(2) and (3).
Response 2: See the earlier response to Comment 3 of Section 1.2.  
Comment 3: Commenter 20609 stated that some standardization is needed to harmonize deadlines for the quarterly emissions reports with quarterly reporting under Part 75. This will ensure EGUs have the full period of time intended under all programs. The commenter noted that EPA must address the potential that EPA Host System maintenance outage could adversely impact EGUs' ability to meet fixed reporting deadlines. EGUs are required to include MATS emissions information in the same quarterly report used to satisfy Part 75. The commenter noted EPA has not yet provided any option to submit MATS data separately. To ensure that EGUs are not forced to submit ARP and CSAPR quarterly emissions reports earlier than they would be required to under § 72.11, or to submit a report on a weekend or Holiday, the commenter suggested EPA revise the MATS rule to provide similar deadline extensions. To address Host System outages, the commenter requested that EPA provide in the rule explicit extension of the reporting deadline if the EPA Host System is offline and not available for submissions at any time on the day a report is due. Specifically, the commenter suggested the following (or similar) language to § 63.10011(f) (or some other appropriate provision) to provide similar instructions: "If the final date of any time period (or any deadline) for submission of a report using the ECMPS Client Tool falls on a weekend or a Federal holiday, the time period shall be extended to the next business day. Moreover, if the EPA Host System supporting ECMPS is offline and unavailable for submission of reports for any part of a day when a report would otherwise be due, the deadline for reporting is automatically extended by the number of days affected by the outage." The commenter said if the revision is not made, EPA explain and solicit comment on its rationale for not addressing EPA Host System outages and for imposing different deadlines under Part 75 and MATS for the same quarterly reports. EPA also must explain why such a provision is not appropriate for the new quarterly compliance reports submitted under § 63.10031(g).
Response 3: The final rule includes statements which extend the deadline for required submittals in two circumstances: (1) when a deadline falls on a Federal holiday or weekend; and (2) when the ECMPS host system is in outage on the deadline.  As described in the preamble, in the first case, the deadline is extended to the next working day.  In the second case, the deadline is extended to the first business day after the system outage.  See section 63.10021(f).
Comment 4: Commenter 20609 supported EPA's proposed clarification of the applicability of § 63.10011(g)(3). The commenter specifically agreed the clarification that it only applies to EGUs that rely on paragraph (2) of the startup definition is helpful. The commenter requested that EPA include the same clarifications in § 63.10021(h)(3) and (i), § 63.10031(i), and Table 8 Item 12. Because § 63.10031(c)(5) only applies to EGUs opting to use that definition, the provisions requiring compliance with that section should be identical. Although § 63.10031(c)(5) is clear in its applicability, inclusion of the qualification in some but not all of the relevant provisions that reference § 63.10031(c)(5) could cause confusion.
Response 4: The following sections of the rule have been amended to further clarify that these sections apply only to EGUs that are relying on paragraph (2) of the §63.10042 definition of startup: §63.10031(i); Table 3, Items 3 and 4; and Table 8, Item 12.
4 63.10021-Continuous compliance with EL, operating limits, and WPS
Comment 1: Commenter 20609 supported EPA's revision to replace the general reference to § 63.10031 in § 63.10011(g) and § 63.10021(h) with a more specific reference to § 63.10031(c)(5).
Response 1: The Agency appreciates the commenter's support regarding the revised language.

Comment 2: Commenter 20612 recommended the proposed revisions should clearly indicate that the electronic reporting requirements satisfy general Part 63 reporting requirements. The commenter provided an example stating a source that submits MATS stack test data via ECMPS would not still need to submit that same data to the state and EPA regional office under §63.13 unless specifically required by the state or regional offices to do so. The commenter recommended EPA clarify that submitting the files through ECMPS will meet the requirements to report the same information to the state/region in the general provisions of Part 63. Specifically, the commenter suggested that §63.10021(f) could read: "(f) You must submit the applicable reports and notifications required under §63.10031 (a) through (n) to the Administrator electronically, using EPA's Emissions Collection and Monitoring Plan System (ECMPS) Client Tool. Submission of a report using the ECMPS Client Tool satisfies any requirement under subpart A of this part to submit that report (or the information in that report) to the appropriate EPA Regional Office or State agency of a delegated authority. If the final date of any time period (or any deadline) for submission of a report using the ECMPS Client Tool falls on a weekend or a Federal holiday, the time period shall be extended to the next business day." The commenter said this change would be consistent with the current provisions in §63.10031(f)(5) and would allow sources to benefit from the paperwork reduction that is supposed to be an advantage of electronic reporting and avoid continuing to have to report hardcopy unnecessarily. States and other delegated authorities already have the right to request any report in hardcopy, if deemed necessary, under §63.10031(f)(5). The last sentence of the suggested language is consistent with federal agency practices.
Response 2: See the earlier responses to comments 2 and 3 of section 3.  
Comment 3: Commenter 20609 found error in the provision where Section 63.10021(e)(9) currently specifies reporting of the date in "hard copy" under § 63.10031(f)(5) until April 16, 2017. Section 63.10031(f)(5) is a provision that only applies to reports that are not the subject of other reporting requirements in paragraph (f). However, as noted above, § 63.10031(f)(4) explicitly requires that the semi-annual report, including the date of the last tune-up, be reported electronically. The commenter suspected the erroneous statement is in reference to requirement in § 63.10006(j) of the 2012 rule to report the "results" of boiler tune-ups within 60 days that was removed in the 2016 Technical Corrections Rule. The commenter said EPA proposes to revise § 63.10021(e)(9) to retain the requirement to report the date of the tune-up in PDF format as part of the semi-annual compliance report up until January 1, 2018 and thereafter to require reporting of the date in the quarterly compliance report under § 63.10031(g). The commenter supported this proposed revision, which would remove the current erroneous reference to hard copy reporting. Prior to finalization of this rule, EPA should acknowledge the error in § 63.10021(e)(9) so that delegated states will not be expecting such reports.
Commenter 20609 stated EPA proposes to specify additional detail regarding the substance of the reporting requirement in new Appendix E, sections 10.3 and 10.4. EPA proposes to require reporting of the both the "Date of Last Tune-up" and "Date of Last Burner Inspection." The commenter said this is in direct conflict with § 63.10021(e)(9) and § 63.10031(c)(4), both of which explicitly limit reporting with respect to tune-ups to the date of the optimization and testing in § 63.10021(e)(6) and (7). The commenter said if EPA wants EGUs to also report the date of the last "burner inspection," EPA must revise the other provisions to make the rule consistent. EPA also must explain what it means by "burner inspection" as that relates to the activities described in § 60.0021(e) and identify what date should be reported when the specified activities take place over several days. For example, EPA should specify whether the reported date includes inspection of the combustion controls, and whether it is limited to the initial inspection or includes inspections following repairs. In the draft reporting instructions posted to the docket, EPA explains the requirement as follows:
"Last Burner Inspection Date
Element name: LastBurnerInspectionDate 
Report the completion date of the most recent burner inspection if it was not done in conjunction with the most recent boiler tune-up but instead was delayed until the next scheduled outage."
The commenter said the additional date EPA proposes be reported is the date of completion of some activity, EPA must say that in the rule. EPA cannot use reporting instructions to establish binding requirements. 
Commenter 20611 stated EPA should clarify the date specified on §63.10021(e)(9) to start the reporting of the tune-up dates in the quarterly compliance report. On page 67071 of the proposed revised rule, EPA specifies that "Until January 1, 2018, report the dates of the initial and subsequent tune-up electronically, in PDF format, in your semi-annual compliance reports, as specified in §63.10031(f)(4) and (6)... After December 31, 2017, report the date of all tune-ups electronically in your quarterly compliance reports in accordance with 63.10031(g) and section 10 of Appendix E of this Subpart." The commenter suggested that the following revisions be made to this paragraph to more clearly align with EPA's intent regarding the start date of the quarterly MATS semi-annual report using ECMPS: 
"Report the dates of the initial and subsequent tunes performed on or prior to December 31, 2017, electronically, in PDF format, in your semi-annual compliance reports, as specified in §63.10031(f)(4) and (6)... Report the dates of all tune-ups performed on or after January 1, 2018, electronically in your quarterly compliance reports in accordance with 63.10031(g) and section 10 of Appendix E of this Subpart." 
Response 3:   Section 63.100021(e)(9) has been finalized as proposed, except that the words "Until January 1, 2018" have been replaced with "Prior to January 1, 2021, and the words "After December 31, 2017" have been replaced with "On and after January 1, 2021".   Therefore, prior to, January 1, 2021, tune-up dates must be submitted electronically in the semiannual compliance reports, and hardcopy submittals are required only when specifically requested by the Administrator.  This is consistent with §63.10031(f)(5), which states that the Administrator retains the right to require submittal of reports subject to paragraphs (f) introductory text and (f)(1) through (4) of this section (which includes the semiannual compliance reports) in paper format.   On and after January 1, 2021, the tune-up dates must be submitted electronically in the quarterly compliance reports.  Note, however, that requirements concerning quarterly compliance reports are under §63.10031(g), not (f)---therefore, the provision in §63.10031(f)(5) authorizing the Administrator to request hardcopy submittals is not applicable to the quarterly compliance reports.    
EPA has withdrawn the proposed requirement in section 10.4 of Appendix E for affected EGUs to report the date of the last burner inspection.  The Agency agrees that §63.10021(e)(9) and §63.10031(c)(4) require only the date of the tune-up to be reported.  According to §63.10031(c)(4), the tune up date is the completion date of combustion optimization and measurement of the effluent CO, NOx and O2 concentrations [see §63.10021(e)(6) and (7)]. 
Comment 4: Commenter 20612 recommended §63.10021(h)(3) and §62.10031(i) should be revised to clearly indicate that the reporting only applies if Startup Definition 2 is used.
Commenter 20609 questioned the current purpose of the proposed clarifying revisions to § 63.10021(i). The commenter stated the provision does not impose any obligation that is not already covered in the proposed revisions to both § 63.10021(f) and § 63.10021(h)(3). The commenter quoted the text as promulgated in 2012, showing that § 63.10021(h) adequately addressed recordkeeping for startup and shutdown, and § 63.10021(i) adequately addressed reporting. To avoid confusion, the commenter requested that EPA remove and reserve § 63.10021(i). If EPA does intend to impose some additional obligation through that provision, EPA should explain what that is and provide opportunity for comment on it. If EPA removes § 63.10021(i), all of the reference to it in Table 3 should be changed to § 63.10021(h)(3), which is the more specific provision.  
Commenter 20611 stated EPA should clarify that the emissions data recorded during startup and shutdown periods that should be included in the compliance reports as specified in the proposed §63.10011(g)(3) and §63.10021(h)(3) apply only to facilities that are relying on paragraph (2) of the definition of startup provided in §63.10042. The commenter also suggested EPA should revise §63.10021(h)(3) and §63.10021(i) to clarify the additional startup/shutdown reporting requirements only apply to facilities that are relying on paragraph (2) of the definition of startup provided in §63.10042 as is specified in the proposed §63.10011(g)(3). They should also indicate that the additional information required by §63.10031(c)(5) should be provided in PDF format as an attachment to the quarterly compliance report also as specified in the proposed §63.10011(g)(3).
Commenter 20602 stated EPA needs to provide clarity on requirements associated with use of startup definition (2). A number of provisions in the proposed revisions reference recordkeeping or reporting associated with the information in §63.10031 (c)(5). The requirements associated §63.10031 (c)(5) are only applicable if an affected unit chooses to use startup definition (2) found at 40 CFR 63.10042. The commenter said the revisions should make clear that the information required by §63.10031(c)(5) is only associated with units where startup definition (2) is being used. An example where this is done in the proposed revisions is the second sentence in §63.10011(g)(3). The commenter suggested that this also be clarified in the §63.10011(g)(3) third sentence: If you are relying on paragraph (2) of the definition of startup in § 63.10042, for startup and shutdown incidents that occur on or after January 1, 2018, you must provide the information referenced in§ 63.10031(c)(5) in PDF format as an attachment to the quarterly compliance reports, in accordance with§ 63.10031(i). 
The commenter suggested that this also be clarified in the following sections of §63.10021:
 §63.10021(h)(3): You must report the emissions data recorded during startup and shutdown. If you are relying on paragraph (2) of the definition of startup in § 63.10042, for startup and shutdown incidents that occur on or prior to December 31, 2017, you must also report the supplementary information in§ 63.10031(c)(5) in the semiannual compliance report. If you are relying on paragraph (2) of the definition of startup in § 63 .10042, for startup and shutdown incidents that occur on and after January 1, 2018, the applicable information in§ 63.10031(c)(5) shall be provided quarterly, in PDF format, in accordance with §63.10031(i).
 §63.10021(i): If you are relying on paragraph (2) of the definition of startup in§ 63.10042, you must provide reports concerning activities and periods of startup and shutdown that occur on or prior to December 31, 2017, in accordance with§ 63.10031(c)(5), in the semiannual compliance report. If you are relying on paragraph (2) of the definition of startup in § 63 .10042, for startup and shutdown incidents that occur on and after January 1, 2018, the applicable information in§ 63.10031(c)(5) shall be provided quarterly, in PDF format, in accordance with §63.10031(i).
Response 4: Section 63.10021(i) of the rule has been revised to clarify that EGU owners or operators who choose to use paragraph 2 of the startup definition are to submit reports concerning activities and periods of startup and shutdown.  Regarding the other comments, see the response to comment 4 of section 3.
Comment 5: Commenter 20612 stated the use of the term "notifications and reports" when referencing electronic reporting within the rule may be confusing since not all notifications are submitted electronically (only the Notification of Compliance Status).
Response 5: The term "reports and notifications," not "notifications and reports," is used 3 times in the rule.  While the commenter is correct that the only specific notification mentioned in these 3 instances is the Notification of Compliance Status (NOCS), the EPA does not believe that using the plural of the term will cause confusion with regard to electronic reporting via the ECMPS Client Tool; however, the rule has been revised such that the "and" has been replaced with "or" in the instances of its use.
Comment 6: Commenter 20609 supported § 63.10011(g)(3), § 63.10021(h)(3) and (i), and § 63.10031(i) to require continued submission of all of the information in § 63.10031(c)(5) in PDF format, rather than XML format. The commenter states this information requested is the most transparent and user-friendly format under the circumstances. The commenter questioned the purpose of continuing to require submission of information in PDF format under (c)(5)(iii)-(v) that already has been reported in a quarterly emissions report in XML format. Under the current rule, all EGUs already are reporting hourly average concentration values for Hg and flow in XML format in quarterly emissions reports submitted to ECMPS, and flagging the hours that occur during startup and shutdown. The commenter said all EGUs with CEMS, sorbent trap monitoring systems, or PM CPMS already are or will be reporting this hourly information in XML format for all operating hours with quality assured data and identifying which hours are startup and shutdown. The commenter questioned why EPA would require that the same information be reported in the subsequent quarterly compliance report in PDF format simply because an EGU has opted to rely on paragraph (2) of the startup definition. To eliminate duplication and streamline reporting, the commenter said EPA should remove the requirements in § 63.10031(c)(5)(iii), (iv), and (v) to re-report in PDF format those startup and shutdown emissions-related values that already have been reported in XML format.
Response 6: The rule requires the additional information required for EGUs using paragraph (2) of the definition of startup must be provided along with the quarterly compliance reports, starting with the first quarter of 2021 report.  Until then, the information must be included in the semiannual compliance reports.  In all cases, the information must be reported in PDF format. 
Note, however, that, consistent with the commenter's recommendation, the hourly CEMS readings during startup and shutdown described in §63.10031(c)(5)(iii), (iv), and (v) have not been carried over into the PDF reports that will be submitted on and after January 1, 2021; these paragraphs have been removed and reserved.  Startup and shutdown hours are flagged in the quarterly emissions reports, and the hourly CEMS readings, which are already required by section 63.10011(g)(2) to be collected and recorded, can be identified easily for auditing purposes.
5 63.10031-Reports to Submit
Comment 1: Commenter 20612 indicated that although §63.10031(c)(1) was not in the proposed rule revision it should be removed or revised   They believe the reference to §63.10(e)(3)(v) will be incorrect since the excess emission reporting requirements are will not be required (except for PM CPMS) after the target date year 2018.
Commenter 20612 also indicated that references to ERT and the ERT website in §63.10031(g) should be removed to reduce confusion regarding the use of ERT in MATS. 
Response 1: : The final rule amends Table 9 to clarify that the excess emission reporting requirements of §63.10(e)(3)(v) and (vi) apply only through December 31, 2020, when the semiannual compliance reports are phased out.  On and after January 1, 2021, the 30- or 90- boiler operating day rolling averages must be reported for all CMS, including PM CPMS, and "deviations" must be reported in the quarterly compliance reports.  
While references for mandatory use of the ERT have been removed from the rule, the ERT remains a viable tool for EGU owners or operators to create their XML schema and report their emissions test information.  Indeed, over 15,000 such schemas have been developed using the ERT and are posted in WebFIRE.

5.1 (a)-Submit each report in the applicable section
Comment 1: Commenter 20609 supported EPA's proposal to reflect the requirement to comply with the electronic reporting requirements of Appendix B, when HCl or HF is monitored continuously, in a new paragraph § 63.10031(a)(2).
Response 1: The Agency acknowledges and appreciates the commenter's support regarding this matter.
Comment 2: Commenter 20609 stated that owners and operators and responsible officials should be able to designate an "agent" to make submissions on his/her behalf and that this authorization should be reflected in the rule. The commenter noted that under the ARP and CSAPR EPA developed rules to allow those who are otherwise required to "submit" reports to authorize "agents" to perform that task on their behalf. See 40 C.F.R. § 72.26. Under those rules, a submitter is legally bound by the actions of his/her agent. The commenter provided additional background details and examples supporting why this terminology is necessary.
Response 2:  In the final rule, statements have been added to Appendices A through E, allowing the required reports and notifications to be submitted by an authorized agent of the EGU owner or operator.  See sections 7.2.5.5, 11.5.5, 7.2.5.5, 3.2.6, and 1.0 of Appendices A, B, C, D, and E, respectively. 
Comment 3: Commenter 20611 stated that EPA should clarify if the compliance certification statement currently provided with each quarterly report submitted via ECMPS under Part 75 satisfies the certification statement requirements specified in § 63.10031(a)(5)(E).
Response 3: The certification statement for Part 75 submissions (found at 40 CFR 72.21(a)(2)) is not identical to the statement found in §63.10031(a)(5)(iii)(E) and elsewhere in the MATS rule (see Appendix A, section 7.2.5.4 and Appendix B section 11.5.4).  Therefore, for coal-fired Acid Rain Program EGUs that elect to comply with MATS by monitoring SO2 emission rate as a surrogate for HCl, separate certification statements are required for Part 75 and MATS when the quarterly emissions reports are submitted. 
Comment 4: Commenters 20612 and 20609 expressed opposition to proposed three (3) significant figures reporting requirements and both commenters provided a number of germane references, examples, and background details as to why they are in opposition.
Commenter 20612 stated that the three significant figures reporting requirement in the MATS rule is fundamentally flawed and that a number of "extra" digits might be associated with an electrical or even digital output does not mean that those digits necessarily reflect the actual precision of the measurement. The commenter stated that taken literally, the requirement to report significant digits is equivalent to applying a new precision standard and EPA should correct the MATS reporting requirements that currently state a certain number of significant figures should be reported and instead prescribe that the data be reported to the same level of precision as the emission limit. The commenter stated that If EPA fails to correct the errant three significant figure reporting requirement, sources may decide that EPA meant something other than the conventional meaning of significant figures when it used the term or that one simply does not have to comply with the reporting "requirement" based on the principle of ad impossibilia nemo tentur and that the technical dilemma creates a situation of uncertainty that results in the potential of inconsistent reporting from source-to-source.
Commenter 20609 stated that in § 63.10031(a)(5)(C), EPA proposes to require reporting of hourly SO2 emission rates "rounded to 3 significant figures and expressed in scientific notation." Id. Although EPA already has included such a requirement in its ECMPS Reporting Instructions [ECMPS Reporting Instructions  -  Emissions at 74, 76 (Sept. 14, 2016)], the MATS rule does not appear to require reporting of these SO2 values to ECMPS at all, let alone to three significant figures. And, EPA's reporting instruction do not provide any citation to the rule as support for the instruction. The commenter stated that although it does not object to a requirement to report hourly SO2 emission rates, it does object to a requirement to report to three significant figures. 
Commenter 20609 stated that EPA should not add a new requirement for SO2 because that represents a greater number of significant figures and greater level of precision than the SO2 emission standards require and would conflict with Part 75 reporting requirements. If recording and reporting of SO2 to the nearest tenth is good enough for the ARP and CSAPR, it should be sufficient to support the emission limitations in Subpart UUUUU, which were developed using data reported under those other programs (see ICR Supporting Statement, Part B, at 37, EPA-HQ-OAR-2009-0234-0062).
Response 4:  The EPA reviewed the comments, and in order to reduce confusion, has revised the rounding policy for the rule.  The EPA believes no one is unable to round Hg, HCl, HF, PM, or SO2 emission rates to the precision of the applicable emission standards, by retaining one leading non-zero digit and one decimal place, expressing the result in scientific notation (e.g. 1.5E-2):  this is what the rule requires. The pollutant concentration values for Hg, HCl, HF must also be reported this way; however, in order to remain consistent with existing sulfur dioxide reporting requirements from other programs, SO2 concentration values (in ppm) must be reported to the precision required by Part 75, i.e., to the nearest 0.1 ppm.  Use conventional rounding, i.e., if the first digit to the right of the first decimal place is 5 or greater, increase the number in the first decimal place by one digit; if the first digit to the right of the first decimal place is 4 or less, leave the number in the first decimal place unchanged. 
Comment 5: Commenters 20594 and 20597 expressed opposition to the proposed reporting requirements that PM and SO2 should be "rounded to 3 significant figures, and expressed in scientific notation." (§§ 63.10031(a)(5)(iii)(C) and Appendix C 7.1.7.2). Both commenters provided a number of germane references, examples, and background details as to why they are in opposition.
Commenter 20594 stated that the requirement to report MATS compliance data in scientific notification to the third decimal point is incompatible with CEMS configuration, departs for longstanding reporting practices, and will lead to inconsistent reporting detail among sources. The commenter stated that this is a substantive change in MATS compliance and it disagrees with this approach for the following 3 reasons:  
  (1) Reporting to three significant figures departs from EPA's historical practice of requiring sources to report compliance data to the same number of decimal places as the standard in place. The commenter referenced a consent decree it entered into with the EPA which it stated reflects this longstanding practice. "...Performance standards, emissions limits, and other quantitative standards set by or under this Consent Decree must be met to the number of significant digits in which the standard or limit is expressed...[Commenter] shall report data to the number of significant digits in which the standard or limit is expressed." 
  (2) For all other parameters that are reported through ECMPS, a number of decimal places is specified, not a number of significant figures. Five separate tables in the ECMPS reporting instructions are dedicated to clarifying the number of decimal places that should be used for reporting emissions data. The commenter stated that it cannot identify, nor EPA has provided an explanation for why MATS requires a different standard of data reporting that departs from the standard used for other Clean Air Act regulations and Clean Air Act enforcement tools, such as consent decrees. Sources always have the option of voluntarily reporting to a higher level of precision.  
  (3) Reporting to three significant figures will lead to sources reporting inconsistent detail. In particular, low emitters would need to report values that greatly exceed any reasonable degree of measurement precision. It is obvious that the precision of instrumentation is not the same regardless of the actual emission value, so it is unclear why EPA would require reporting to the same level of precision regardless of the emission level. In addition to requiring inconsistent reporting obligations among sources, the Proposed Rule could subject low emitters to reporting results that are well below the precision of the instrument itself and are therefore inaccurate or misleading at best. Any reference to significant figures should be relative to the precision of the standard. 
Commenter 20597 stated that EPA should correct the MATS reporting requirements that currently state a certain number of significant figures should be reported and instead prescribe that the data be reported to the same level of precision as the emission limit.
Response 5:  See the response to the previous comment.
Comment 6: Commenter 20595 stated that the proposed changes in §§63.10031 (a)(1),(2), and (5) would clarify the electronic reporting requirements for the Hg, HCI, HF, SO2, and auxiliary CMS and that these changes appear to be in addition to that required by other programs (i.e., Part 75, ARP, and CSAPR) where the SO2, flow, CO2 and moisture CEMS are being used. In those programs, the reporting of quarterly QA and the amounts of data required for submittals of RATAs go well beyond what has historically been required. The agency has been silent regarding whether the submittal of RATA results should follow that same submittal schedule. The commenter recommended that the RATA reports and the excessive data elements be included into the uniform schedule and that the agency specifically address the change from the 45-day reporting requirement. 
Commenter 20595 stated that Appendix 9 [sic] should address that § 63.10(e)(3) be excluded since the reporting requirements will follow Subpart UUUUU.
Response 6:  The Agency understands these comments to be seeking clarification whether CEMS data and quality assurance checks obtained for and performed in accordance with other programs can be used for MATS.  As mentioned earlier, section 10031(a)(5)(iii)(E) of MATS already allows use of SO2 data; the rule and its appendices also already allow the use of part 75 instrumentation and procedures for flow, moisture, and diluent data collection (see, for example, 63.10005(h)(3)(i)(A) and (B), 63.10010(c), section 5.3 of Appendix A, and section 5.3 of Appendix B); the Agency believes no additional clarification is warranted.


Comment 7: Commenter 20602 stated that since the MATS rule primarily affects coal-fired EGUs, a majority, if not all, affected units are already subject to the ARP. The commenter suggested that to simplify the requirements under proposed § 63.10031(a)(5), that the rule provide one set of conditions to specify what additional information must be submitted for those units that are already subject to ARP reporting requirements and another set of conditions for those units that are not subject to ARP requirements.
Response 7:  The commenter's suggestion has not been incorporated.  Regardless of whether or not an EGU is in the Acid Rain Program, the same requirements for SO2 reporting apply. Paragraph (a)(5) of §63.10031 clearly states that, ".... you must use the ECMPS Client Tool to submit the following information to EPA (except where it is already required to be reported or has been previously provided under the Acid Rain Program....)."  Coal-fired Acid Rain Program EGUs that are subject to MATS and elect to monitor SO2 as a surrogate for HCl already meet most of the requirements of §63.10031(a)(5). However, the requirement to report an hourly data stream in units of the MATS emission limit is required only by MATS, and the certification statement in §63.10031(a)(5)(iii)(E) is different from the standard Part 75 certification statement (see Response 3 in this section, above).  In addition, Appendix E to MATS requires corresponding Reference Method 6C data to be reported for each run of each RATA of the SO2 monitor. The Agency believes that the provisions of 63.10031(a)(5) and Appendix E are sufficiently clear and that there is no need to revise the rule text to provide two sets of SO2 reporting conditions, i.e., one for Part 75 and one for MATS, as suggested by the commenter.     
Comment 8: Commenter 20611 stated that EPA should revise §63.10031(a)(3) to specify that the electronic reporting of hourly PM CEMS emissions data shall begin with the later of: The first operating hour on or after January 1, 2018 or the first operating hour after completion of the initial PM CEMS correlation test on or after January 1, 2018.
Commenter 20611 stated that EPA should revise §63.10031(a)(4) to specify that the electronic reporting of hourly PM CPMS emissions data shall begin with the later of: The first operating hour on or after January 1, 2018 or the first operating hour after completion of the initial performance stack test that establishes the operating limit for the PM CPMS conducted on or after January 1, 2018. 
Response 8: Table 8 incorporates the commenter's recommendations, except that the January 1, 2018 date is replaced with January 1, 2021.
5.2 (b)-Compliance reports
Comment 1: Commenter 20611 stated that EPA should revise § 63.10031(b)(1) to specify that the first compliance report should cover the period ending on June 30 or December 31, whichever date is the first date that occurs at least 180 days after the compliance date that is specified for your affected facility in §63.9984, or if applicable, the extended compliance date approved under §63.6(i)(4).
Response 1: The commenter's recommendation has been incorporated into the rule.  Revised §63.10031(b)(1) now recognizes that some affected EGUs may have received approval of extended compliance dates under 63.6(i)(4).  Existing sections 63.10000(n) and 63.10005(a)(2) of the rule already acknowledge the possibility that compliance date extensions may be granted in a similar fashion.  
5.2.1 Semiannually to quarterly
Comment 1: Commenter 20595 stated that based on the proposed changes to §63.10031 (b)(5), the current option to follow alternate submission schedules for semiannual compliance reports to coordinate with Title V Semi-annual deviation reporting requirements would be removed, which would lead to additional work being required for modifying Title V Permits. The agency has included several reporting elements of the MATS program into the uniform submission schedule and proposes changing that schedule to quarterly. In some cases, Title V Permits have been amended or modified to allow the alternative semi-annual report submittal schedule. The commenter expressed concern as this proposed change will require another modification to reverse the amendment to the Title V permit where facilities and permitting authorities have worked to get the reporting schedules for semi-annual reports on the same schedule for both Title V and MATS.
Response 1:  In keeping with industry's request to use the ECMPS Client Tool, submissions should be made on a consistent basis.  As the ECMPS Client Tool requires quarterly submissions, MATS electronic reporting requirements have been adjusted to its submission requirements, i.e., from semiannual to quarterly submissions.  In addition, a consistent reporting basis is important, for without one, it would be virtually impossible to track compliance with these reporting requirements if there were a multitude of custom reporting schedules.    
Comment 2: Commenter 20607 expressed support for EPA's proposal to consolidate reports that originally were required to be submitted separately on different time tables into quarterly compliance reports. Commenter stated this change would eliminate the need for companies to track different reporting deadlines, thus reporting would be greatly simplified, as well as more efficient and cost-effective.
Commenter 20607 also expressed support for the proposed compliance deadline for quarterly compliance reports of 60 days after the end of each calendar quarter. Commenter stated that this 60-day period is necessary for companies to compile stack test data. A shorter time period would not allow sufficient time to obtain and verify results prior to submittal to EPA. This is especially important for PM and HCl stack testing, the results of which require additional time for laboratory analysis.  
Commenter 20607 stated that it urges EPA to extend this 60-day period to the submittal of emissions files and quality QA file data. As proposed, this information must be submitted within 30 days after the end of each quarter. This is not sufficient time for data to be collected and reports to be generated. PM and HCl CEMS rely on periodic stack tests for QA. This additional QA testing must be included in the quarterly reports to validate the CEMS results. A 30-day reporting deadline is insufficient to allow companies to perform these QA tests and collect results, particularly when the performance testing is conducted near the end of the calendar quarter. In those situations, it is possible that the test results will not be finalized until after the 30-day submittal period has ended. In addition to affording the time necessary to perform QA tests, extending the deadline to 60 days would align the deadlines for all quarterly reports, decreasing the administrative burden and increasing the overall efficiency of reporting.
Response 2:  EPA appreciates the supportive comments regarding the 60-day period after the end of each quarter in which to submit the results of stack tests performed to demonstrate compliance.  As requested by the commenter, the rule provides a similar extension for submitting the results of HCl and HF monitor RATAs and PM CEMS performance evaluations (RRAs, RCAs, and correlation tests).  For these continuous monitoring system (CMS) performance evaluations, the results may be submitted up to 60 days after the test completion date. 
Comment 3: Commenter 20612 supported the changes to replace the semi-annual reports with the quarterly compliance reports. They also agree, it is absolutely necessary to maintain the 60-day reporting deadline to ensure that testers and sources have adequate time to analyze, review and certify the quality of the emissions data being reported.  
Commenter 20609 generally supported the consolidation of reporting using ECMPS however for each activity involving use of a reference test method  -  i.e., a quarterly performance test or LEE test  -  EPA proposes to require submission of detailed information about the reference method used.  Most of this detailed data will be electronically reported using XML and the remainder using PDF.  This commenter was concerned about the number of details in the proposed reporting changes.
Response 3: The Agency appreciates the commenters' support for quarterly submission of electronic reporting using the ECMPS Client Tool, along with limited PDF file submission. As mentioned earlier, switching to the industry-requested, sole use of the ECMPS Client Tool necessitated some revisions to the existing reporting requirements; however, even with those changes, the overall industry burden decreases by about $4.2 million per year. 
5.3 (c) -CEMS units of measure
Comment 1: Commenter 20597 stated that EPA is proposing that sources that correlate PM CEMS in terms of actual conditions (mg/acm vs. mg/scm) would be required to report hourly PM concentrations in both mg/acm and mg/scm and to report hourly temperature and pressure values as well. The commenter recommended that all PM CEMS, regardless of the type, only be required to report hourly mg/scm values. This approach will simplify the reporting and make the reported data more consistent from unit-to-unit.
Commenter 20607 stated that EPA proposes to require that hourly PM concentrations be reported in both mg/acm and mg/scm. This would require companies to calculate PM emissions using both metrics and report the same data to EPA twice. The commenter recommended that companies be allowed to choose which metric to use when reporting hourly PM concentrations. This approach would simplify and streamline reporting.
Response 1: The rule now requires PM concentrations to be reported in the units of measure that are consistent with the units of measure used for the PM CEMS correlation, rather than requiring reporting PM concentrations in both actual and standard units of measure. This approach is consistent with provisions in Performance Specification 11 (40 CFR part 60, Appendix B) that require one to"...convert the reference method results to units consistent with the conditions of your PM CEMS measurements." Given that measurement units differ among PM CEMS, the rule requires hourly PM concentrations to be recorded and reported in units of measure that are consistent with the PM CEMS correlation (see Appendix C, section 3.2.2).  Moreover, section 7.1.9.5 of Appendix C is consistent with this approach, requiring the reference method readings and the PM CEMS responses obtained in the RRAs and RCAs to be reported in the same units of measure as the PM CEMS correlation curve. For example, if the PM CEMS measures in units of mg/acm and the concentrations used to derive the correlation curve are in those same units, then the hourly PM concentrations must be recorded and reported in mg/acm.   
--------------------------------------------------------------------------------
All PM concentrations must ultimately be converted to standard conditions, in order to calculate the PM emission rates (lb/mmBtu or lb/MWh).  However, the procedures that are followed and/or the equations that are used for the conversions are not necessarily the same for all PM monitoring technologies.  In view of this complicating factor, section 7.1.1.2.2 of Appendix C requires the hard copy portion of the monitoring plan to include an explanation of how the PM concentrations are measured and how they are converted to the units of the applicable emissions limit; in addition, the equation(s) used for the conversions must be documented. 
Comment 2: Commenter 20609 provided a summary of the change to 40 C.F.R. § 63.10031(c)(5)(iii) indicating that EPA proposes changing the phrase "use CEMS to demonstrate compliance with numerical limits" to "use CEMS for compliance purposes." EPA also proposes changing the units for reporting flow values from "standard cubic meters" to "standard or actual cubic feet per hour." This commenter then objects to the change to "use CEMS for compliance purposes" because EPA removing the language adopted in the 2016 Technical Corrections Rule and going back to the language promulgated in 2014 without stating the reason for this change. The commenter requested EPA explain the proposed change so they may adequately provide comments. Commenter 20609 agreed with the change in units of measure to "standard or actual cubic feet per hour" because the revisions would better align with the existing Part 75 requirements. 
Commenter 20609 also indicated that § 63.10031(c) "the compliance report" should be revised to be consistent with the terminology of § 63.10031(b)(5) "the semi-annual compliance report" or add time limiting language to make the rule consistent. If EPA does not do that, the requirement in § 63.10031(c)(1) to comply with excess emissions reporting under § 63.10(e) would continue to apply contrary to EPA's express intent. 
Response 2: The commenter's concern is moot, for, as mentioned earlier, §§63.10031(c)(5)(iii) through (v) of the rule have been removed and reserved.  Hourly CEMS values recorded during startup and shutdown are flagged in the electronic emissions reports and are easily identified for auditing purposes.  Therefore, for owners or operators of EGUs electing to use paragraph (2) of the startup definition, it is unnecessary to report these same values in the semiannual compliance report. 
The word "semiannual" has been inserted into the introductory text of paragraph (c), as well as into paragraph (b), to address the commenter's concern.


5.4 (d)-Excess emissions and monitor downtime for semiannual and quarterly compliance reports
Comment 1: Commenter 20612 stated that §63.10031(d)(2)(iv) should be revised to indicate that the emergency bypass reporting requirements only apply to LEE units.
Response 1: Specific emergency bypass reporting requirements associated with EGUs with LEE status already exist in §63.10031(c)(6), and those requirements have been revised to show the link to section 10.3 of Appendix E; moreover, upon consideration of the comment, §§63.10031(d)(1), (2), and (3) have been replaced by a general requirement for owners or operators of EGUs that use CMS to report excess emissions and monitoring system downtime according to §63.10(e)(3)(v) and (vi) in their compliance reports (see §63.10031(d)).  Submission of those reports is on a semiannual basis through December 31, 2020; however, starting on January 1, 2021, after the semiannual compliance reports are phased out and replaced with quarterly compliance reports under §63.10031(g), reporting of 30- and 90- boiler operating day rolling averages for all parameters that are continuously monitored is required and any "deviations" that occur in the quarter must also be reported.  
Comment 2: Commenter 20612 stated that §63.10031(e)(1) and (2) cause confusion around which reporting requirements apply for the semiannual compliance reports before 2018 and which apply to the reporting requirements for quarterly compliance reports following 2018. Since these requirements are clearly explained in other parts of the rule, these sections should be removed.
Response 2:  EPA agrees with the commenter that proposed paragraphs (e)(1) and (e)(2) of §63.10031 are superfluous, and they have been withdrawn.
Comment 3: Commenter 20609 stated that as an initial matter, the purpose of proposed § 63.10031(d)(2) is not clear. Current § 63.10031(d)(1) establishes semi-annual "excess emissions" and "monitoring system performance" reporting requirements consistent with the general provisions that will end on January 1, 2018 and be replaced by new XML reporting requirements. In other instances where EPA has proposed to replace the current semi-annual reporting requirements with XML reporting requirements in 2018, EPA has reflected the requirements in the data elements in new Appendix E and simply required reporting of them under § 63.10031(g). It is not clear why EPA believes it necessary to address future "excess emissions" reporting both in § 63.10031(d)(2) and in Appendix E. EPA should adopt a consistent approach. If EPA does not change its approach, it must address the following issues with its proposed language. Commenter 20609 stated that EPA provides an exception to the "excess emissions" reporting requirement for EGUs that elect to demonstrate compliance using a PM CPMS [Proposed § 6310031(d)(3) and (k)]. The purpose and scope of this exception are unclear. With respect to scope, if read literally, the use of a PM CPMS would subject an EGU to PDF reporting of excess emissions and monitor evaluation information for all emission limitation and monitoring systems. UARG assumes that EPA only means such EGUs would continue existing excess emissions and monitoring system evaluation reporting with respect to their PM CPMS and its associated operating limit, and not with respect to other emission limitations or CMS. EPA should add language stating as much. 
Commenter 20609 stated that the manner of reporting and the relationship between § 63.10031(d)(2) and the new data elements in Appendix E are unclear. Appendix E data elements clearly must be reported in XML format. Proposed App. E section 1.0. Section 63.10031(d)(2) is silent on the format for reporting. Because most of the information in § 63.10031(d)(2) appears to also be covered in proposed Appendix E, the commenter assumes EPA intends it to be reported in XML format consistent with Appendix E. But the two parts of the rule are not consistent. For example, Appendix E covers exceedances of emission limits and operating limits in sections 12.1.1 and 12.1.2, but labels them "deviations" from an "emission limitation" (see proposed section "12.1.1 Emission limit exceeded"), not "excess emissions." As a result, it is not clear from the rule whether compliance with Appendix E satisfies § 63.10031(d)(2). EPA should not require excess emissions to be reported both as "excess emissions" and "deviations." It should be enough to report once under § 63.10031(g) and Appendix E.
Response 3: The comments are moot, for as mentioned earlier, upon consideration of the comments, §§ 63.10031(d)(1), (2), and (3) have been removed from the rule.
Comment 4: Commenter 20609 stated that the requirements for malfunction reporting are inconsistent. Current § 63.10031(g) only requires reporting of "malfunctions." EPA proposes to revise the "malfunction" reporting in proposed § 63.10031(d)(2)(iv) to include "emergency bypasses." Under the current rule, and proposed Appendix E section 13, "emergency bypass" events are reported only by EGUs with LEE status in order to implement the emergency bypass exceptions in § 63.10000(c)(1)(i)(C)(2). EPA also proposes separate data elements for malfunction reporting in Appendix E section 11.0. At a minimum, to avoid confusion and for consistency, EPA should remove the reference to "emergency bypass events" in § 63.100031(d)(2)(iv). 
Commenter 20609 stated that moreover, EPA should not require the reporting of malfunctions under both § 63.10031(d)(2) and Appendix E. It should be enough to report once under § 63.10031(g) and Appendix E.
Response 4: As mentioned earlier, the reference to emergency bypass events has been removed from paragraph (d) of §63.10031.  Emergency bypass incidents are addressed in Section 10.3 of Appendix E, which makes it clear that these incidents apply only to certain coal-fired EGUs that are using the LEE compliance option.
The rule continues to require reporting of process and control equipment malfunctions in the semiannual compliance reports until those reports are phased out (the final report covers the fourth quarter of 2020).  However, to avoid inadvertently removing the requirement to report malfunctions in the semiannual reports, that requirement has been moved from §63.10031(g) to a new section, §63.10031(c)(10).  New paragraph (g) of §63.10031 requires quarterly compliance reports to replace the semiannual reports, starting in the first quarter of 2021; the quarterly compliance reports retain the requirement to report malfunctions.     


5.5.2 On and after January 1, 2018 compliance based on 63.10031 (g), (i), (k)
Comment 1: Commenter 20611 stated that EPA should revise §63.10031(e)(2)(iii) to specify that new EGUs starting operation on or after January 1, 2018 or EGUs with extended compliance dates approved under §63.6(i)(4) should submit their initial Notification of Compliance Status in accordance with paragraph (h) of §63.10031. It is our understanding that no additional NOCS are required to be submitted for any subsequent tests performed under the MATS rule for existing EGUs with compliance date of April 16, 2015. The results of these tests were being submitted in the semi-annual compliance reports prior to December 31, 2017 and will be submitted in the quarterly compliance reports after January 1, 2018.
Response 1: The rule clarifies in §63.10030(e) and in Table 9 that Notifications of Compliance Status are required only for the initial compliance demonstration.  Sections 63.10000(n) and 63.10005(a)(2) recognize the possibility that compliance date extensions may be given.
5.6 (f)-Interim period PDF reports
Comment 1: Commenter 20612 supported the change from semi-annual to quarterly reporting and maintains that the 60 day reporting deadline must be maintained to provide sufficient time for testers and sources to complete analysis and reporting. 
Commenter 20612 supported submitting detailed performance test report electronically as PDF files but indicates the proposed changes would be simpler if EPA removes the proposed list of requirements for the PDF stack tests reports required under Table 8 and Section 22 of Appendix E and simply references the stack test requirements included in the recently finalized revisions to §60.8(f) and §63.7(g). Without this change, the difference between the content lists will introduce inconsistency between rules.
Response 1: The Agency appreciates the support for quarterly reporting and, as mentioned elsewhere, the rule was adjusted to allow a 60 day reporting deadline for tests performed to demonstrate compliance. The Agency finds that a sole submission of detailed performance test reports via a PDF file would not constitute electronic reporting as described in the Agency's E-Reporting Policy for Regulations. As that policy states  -  and as is provided via the XML schema for this rule  -  electronic reporting is a system in which an electronic tool guides the regulated entity through the reporting process and which has some data quality checks; electronic reporting is not simply submitting a copy of a document, such as a performance test report, to the Agency. In addition, the Agency notes that the XML schema contains rule-specific information to be reported; this information would not be collected if EGU owners or operators were to report just the requirements of the NSPS and NESHAP general provisions, as suggested by the commenter.  Therefore, the changes regarding only PDF file reporting for just the items in the NSPS and NESHAP general provisions are not in the rule. 
5.6.2 CEMS or CPMS must submit quarterly within 60 days
Comment 1: Commenter 20595 stated that the proposed revisions to § 63.10031(f)(2) would require quarterly reporting of all 30- (or 90-) boiler operating day rolling average emission rates for units monitoring Hg, HCI, HF, and/or SO2 emissions, and for units using emissions averaging, starting with a report covering the first quarter of 2018. The agency identifies that this reporting is in place of exception reporting. As such, the agency will need to change the software generally utilized for hourly data submittals to handle a daily value. This change needs to be factored into the timing of completing the changes to ECMPS and allow time for DAHS software providers to make changes to their software to make this reporting successful for the reporting period beginning January 1, 2018.
Response 1: The Agency finds the commenter's concerns misplaced, as ECMPS software will only collect hourly average data from continuous monitoring systems; it will not calculate, or be used as the means of affirmation for  compliance with, the 30- or 90-boiler operating day rolling averages.  Those boiler operating day rolling averages are to be reported in the quarterly compliance reports, and compliance assessment is to be performed by regulatory authorities such as inspectors or auditors.  As an aside, the commenter apparently misunderstands the way in which the rolling average is to be calculated; it is not the average of 30 or 90 "daily" values, but it is the arithmetic average of all of the valid hourly emission rates recorded during the 30- or 90-  boiler operating day period.   
5.6.3 Submittal of PDF
Comment 1: Commenter 20595 stated that in §63.10031(g) through (I) and Table 8, Item 8, the agency identifies that ECMPS will be designated the exclusive data system for reporting. While the data elements are excessive in some situations, the commenter stated that it generally agrees that the sole use of ECMPS and the XML format is preferable.
Commenter 20595 stated that the proposed revisions should clearly indicate that the electronic reporting requirements satisfy general Part 63 reporting requirements (e.g., sources that submit MATS stack test data via ECMPS would not still need to submit that same data to the state and EPA regional office under §63.13 unless specifically required by the state or regional offices to do so). EPA needs to clarify that submitting the files through ECMPS will meet the Part 63 general provision requirements to report the same information to the state/region. The commenter stated that it believes that §63.10031 (f)(5) implicitly allows the existing electronic reporting required under §63.10031(f)(1) through (4) to satisfy the reporting requirements under the Part 63 general requirements. But, the new electronic provisions are introduced in other sections (§63.10031(g) through (I)). The proposed rule should be revised to ensure that the new electronic reporting provisions will satisfy the general provisions to submit such data to state and EPA regional offices (and to make such a statement in a clearer, explicit way). For example, §63.10021(f) could be revised to read: 
"(f) You must submit the applicable reports and notifications required under §63.10031 (a) through (n) to the Administrator electronically, using EPA's Emissions Collection and Monitoring Plan System (ECMPS) Client Tool. Submission of a report using the ECMPS Client Tool satisfies any requirement under subpart A of this part to submit that report (or the information in that report) to the appropriate EPA Regional Office or State agency of a delegated authority. If the final date of any time period (or any deadline) for submission of a report using the ECMPS Client Tool falls on a weekend or a Federal holiday, the time period shall be extended to the next business day." 
Commenter 20595 stated that this change would be consistent with the current provisions §63.10031(f)(5) and would allow sources to benefit from the paperwork reduction that is supposed to be an advantage of electronic reporting and avoid continuing to have to report hardcopy unnecessarily. States and other delegated authorities already have the right to request any report in hardcopy, if deemed necessary, under §63.10031(f)(5).
Response 1: The Agency disagrees that the reporting requirements associated with performance (stack) testing of the original rule or those associated with HCl or PM CEMS or PM CPMS of the proposed rule are "excessive in some situations;" each data element required to be reported (and identified by a separate XML tag) is necessary to determine compliance with a component of the test method, performance specification, or quality assurance or control procedure. With regard to the other comments, see the response to comment 3 of section 1.2 and the response to comment 3 of section 3.
Comment 2: Commenter 20609 stated that references to Notification of Compliance Status ("NOCS") for initial and subsequent compliance demonstrations are confusing since there are no requirements for any NOCS except the initial NOCS.  Currently Section 63.10011(e) reads: "(e) You must submit a Notification of Compliance Status containing the results of the initial compliance demonstration, in accordance with §63.10030(e)." 
Commenter 20609 provided that Section 63.10030(e) currently reads: "(e) When you are required to conduct an initial compliance demonstration as specified in §63.10011(a), you must submit a Notification of Compliance Status according to §63.9(h)(2)(ii)."  They further stated that "language in §63.10031(e)(7) and § 63.9(h)(2)(ii) of the general provisions appear to conflict with those provisions".  
Commenter 20609 provided alternate changes to address these issues:
"EPA should make the following additional changes to ensure that the rule is internally consistent:
  :: Remove and reserve § 63.10030(e)(7)(i), which duplicates current § 63.10031(c)(7); and
  :: In Table 9, under § 63.9, add a note that the reference in (h)(2)(ii) to submission of a NOCS following "completion of any subsequent required performance test" does not apply."  
Response 2: The Agency agrees that section §63.10030(e)(7)(i) is duplicative; that section has been removed from the rule and reserved.  Table 9 and §63.10030(e) in the rule have been revised to clarify that Notifications of Compliance Status are required only for initial compliance demonstrations.

5.7 (g)-Quarterly compliance reporting
Comment 1: Commenter 20609 stated that EPA proposes to significantly expand the information that must be reported for all tests involving reference methods, including RRAs and RCAs for PM CEMS and even if EPA finalizes some or all of the proposed requirements, EPA must reconsider the timing of some of the required submissions.
Commenter 20609 stated that under the current rule, RATA, RRA, and RCA test reports are not required to be submitted until 60 days after completion of the test [40 C.F.R. § 63.10031(f)(1)]. EPA proposes to revise that to require submission no later than with the "relevant quarterly emissions report." [Proposed App. C section 7.2.4; see also Table 8 Items 13 and 14, App. E sections 14.3, 15.2-3]. The commenter stated that if EPA finalizes that requirement, the amount of time between completion of the tests and the electronic reporting requirement would vary based on when the test was conducted. For a test conducted early in a calendar quarter, the revised deadline would provide more time than the current rule. The commenter stated that it does not object to that and agrees that EGUs should be allowed to submit the information earlier, just as EPA has proposed. However, for tests conducted late in the quarter, the rule could require reporting of data as little as 30 days after completion of the test.
Commenter 20609 stated that thirty (30) days may not be adequate for RRAs and RCA conducted on PM CEMS. RRAs and RCAs are much more involved than a typical RATA. They involve isokinetic sampling and require laboratory analysis that might take several weeks for an outside laboratory to process. Once the results are obtained, they need to be integrated with the other stack testing measurements to calculate the final PM emission rates that are provided to the owner/operator. The owner or operator then has to perform the requisite comparison to determine if the test was passed. Requiring completion of that in 30 days is not reasonable, and EPA has provided no rationale for changing the deadline. To the contrary, with respect to performance tests, which also require the processing of laboratory results, EPA acknowledges that 60 days (not 30 days) is a reasonable time period. 81 Fed. Reg. at 67,065. The commenter stated that as a result, rather than require submission of RRAs and RCA results with the quarterly emissions report, EPA should require their inclusion in the quarterly compliance report, which does not have to be reported until 60 days after the end of the quarter.
Response 1: The commenter's suggestion to include RRA and RCA results  in the quarterly compliance reports has not been incorporated.  Instead, the final rule allows up to 60 days after the test completion date for the results of RRAs and RCAs to be submitted.  The quarterly PM emissions data reports must still be submitted within 30 days after the end of each quarter.   
To the extent practicable, the RRA and RCA results should be submitted prior to or concurrent with the relevant quarterly emissions report.  However, when this is not possible (e.g., for tests done near the end of the quarter), the emissions data may be assigned a "provisional" status, pending receipt of the test results.  If the RCA or RRA is successful, then, when the test results are submitted electronically, the status of the "provisional" data changes to "quality-assured" and no further action is required.  If the test is not successful, then, when the results are submitted, data with the provisional status must be invalidated until a subsequent test of the same type is successful; in this case, resubmission of the affected quarterly emissions report(s) is required.   
Comment 2: Commenter 20609 stated that there are two issues with EPA's proposal to replace current § 63.10031(g) with a provision requiring submission of new quarterly compliance reports starting with the first calendar quarter of 2018. The commenter stated that first, the new proposed language contains a typographical error. The reference in the proposed revised section to "appendix B" is a typographical error. It should refer to "appendix E." 
Commenter 206090 stated that secondly, the revision would remove the requirement to report certain malfunction events in semi-annual reports for the period up until December 31, 2017. If EPA wants such information to be reported, EPA needs to include it somewhere else. The commenter suggests that it could be added to the other requirements for semi-annual reports in § 63.10031(c) as a new subsection (c)(10).
Response 2: The rule now corrects the typographical error.  Also, upon consideration, the recommendation to relocate existing paragraph (g) of §63.10031 as a new paragraph (c)(10) has been incorporated in the rule.  The Agency agrees that without the relocation, the revisions to §63.10031(g) would have inadvertently removed the requirement to report malfunctions in the semiannual compliance reports.
5.7.2 Due within 60 days after the quarter
Comment 1: Commenter 20594 stated concern that the proposed rule puts new quarterly reporting obligations in place that are inconsistent with existing MATS reporting time periods. The proposed rule includes a new requirement to report and certify hourly PM CEMS data on a quarterly basis that is not in harmony with the current MATS reporting scheme. 81 Fed. Reg. at 67064. Presently, MATS requires that sources submit quarterly PM CEMS reports without hourly data. The source must submit relative accuracy test audit data or other quality assurance data (RCA and RRA data) within 60 days after the date of completing the test [§ 63.10031(f)]. Since the reference methods used to conduct these tests are wet chemistry methods (as opposed to instrumental analyzer test methods), sources must wait on the test results during this 60-day period after the testing date. A certification problem arises if a source is waiting to receive testing results, but a quarterly report becomes due. The source would be forced to certify PM CEMS data as "true accurate and complete" prior to receiving the testing results that verify the data is indeed accurate and quality assured. The Proposed Rule does not provide a solution for this gap in timing created by the new requirement to submit hourly PM CEMS data through ECMPS. 
Commenter 20594 stated that it recommends revision to this reporting scheme and offered the following actions EPA could take to remedy this disconnect:
:: Exclude hourly PM CEMS emissions data from quarterly reporting.
:: Allow PM CEMS hourly emissions data to be submitted within 60 days after the end of the calendar quarter.
:: Exclude hourly PM CEMS data from certification by revising the certification language.
Response 1: Upon consideration of the comments, the rule maintains hourly PM CEMS emissions data and does not revise the certification language; however, the rule allows PM CEMS RATA data to be submitted in provisional status up to 60 days after the end of the quarter. More specifically, appendix C of the rule requires quarterly reporting of hourly PM CEMS data.  The quarterly emissions reports are due within 30 days after the end of each calendar quarter.    However, to address the commenter's concern about submission of the RRA and RCA results, the rule text in section 7.2.4 of Appendix C reads as follows:  "...Submit the results of the QA test (i.e., RCA or RRA) or, if applicable, a new PM CEMS correlation test, either prior to or concurrent with the relevant quarterly electronic emissions report.  If this is not possible, you have up to 60 days after the test completion date to submit the test results; in this case, you may claim provisional status for the emissions data affected by the QA test or correlation, starting from the date and hour in which the test was completed and continuing until the date and hour in which the test results are submitted.  For an RRA or RCA, if the applicable audit specifications are met, the status of the emissions data in the relevant time period changes from provisional to quality-assured, and no further action is required. For a successful correlation test, apply the correlation equation retrospectively to the raw data to change the provisional status of the data to quality-assured, and resubmit the affected emissions report(s).  However, if the applicable performance specifications are not met, the provisional data must be invalidated, and resubmission of the affected quarterly emission report(s) is required.  For a failed RRA or RCA, you must take corrective actions and proceed in accordance with the applicable requirements found in sections 10.5 through 10.7 of Procedure 2 until a successful QA test report is submitted. If a correlation test is unsuccessful, you may not report quality-assured data from the PM CEMS until the results of a subsequent correlation test show that the specifications in section 13.0 of PS11 are met."
Comment 2: Commenter 20611 stated that EPA should revise §63.10031(g) to specify that the quarterly electronic compliance reports are due "no later than 30 days after the end of each calendar quarter."
Response 2: As mentioned earlier, the rule requires compliance reports described in §63.10031(g) to be submitted within 60, not 30, days after the end of each calendar quarter.  The extra time is needed to receive the results of certain stack tests (i.e., HCl and PM tests) that are performed near the end of a calendar quarter and require laboratory analysis of the collected samples. 
5.9 (i)-Startup and shutdown incidents
Comment 1: Commenter 20612 stated that EPA should not require the data in §63.10020(e), which is referenced in §63.10031(c)(5), to be submitted in XML format. The commenter objected to the requirement to report the startup data, but said there is no practical or meaningful way to make data more "user-friendly." Given the nature of the information required, the data are better suited as a recordkeeping requirement. 
Commenter 20612 stated that if the Agency retains the reporting requirement, a PDF rather than XML format should be used. Trying to report the information in XML format would be difficult because this would cover many additional process parameters and potential configurations of those controls. In contrast to the burden of putting the information in numeric form, there would be little benefit to providing the data in XML since there are few, if any, meaningful checks that could be performed. Given the transient nature of the parameters during startup, screening the data electronically would not only be cumbersome but likely a fruitless effort. The commenter stated that given these factors (and the effort required of the Agency alone to develop XML formats for such data, much less the burden of sources to report the data in XML), PDF reporting would be adequate.
Commenter 20612 stated that EPA published the monitoring and reporting requirements in §63.10020(e) and §63.10030(e)(8) as part of the final startup/shutdown reconsideration rule without having proposed such language or making any indication that it was considering such requirements. The proposal of reopening the comment period for reconsideration of the startup/shutdown (SU/SD) provisions only included requests for additional information or comments on four specific issues: 1) the use of default diluent gas cap during SU/SD, 2) how to calculate SU/SD emissions for affected units sharing a common stack, 3) the use of a default electrical output value for calculating SU/SD emissions when the load is zero, and 4) the results of technical analyses the Agency conducted, which it believed "could support defining the end of startup at coal-fired EGUs as occurring at 25 percent of nameplate capacity plus 3 hours or the start of electricity generation plus 6 hours, whichever comes first." 
Commenter 20612 stated that the requirements that EPA included in §63.10020(e) for sources to monitor (and report such information under §63.10031(c)(5)) could not reasonably be considered a logical extension of any of these four requests; and, therefore, the public was denied the opportunity to comment upon the requirements as required under CAA §307(d)(5). The Agency should, therefore, remove these unlawfully added provisions, which the commenter said provide little benefit in terms of documenting application of the startup work practice standards for sources using Paragraph 2 of the MATS Startup definition, making the question of how to report such information moot. 
Response 1: In the rule, the additional information required for owners or operators of EGUs using paragraph (2) of the definition of startup must be provided along with the quarterly compliance reports, starting in the first quarter of 2021.  Until then, the information must be included in the semiannual compliance reports.  In all cases, the information must be reported in a PDF file.  
The Agency disagrees that the data required when an EGU owner or operator chooses to use paragraph 2 of the definition of startup are not suitable for collection and submission via an XML format.  Those data are numeric values, as opposed to text, and are thus well suited for collection via XML format.  Nevertheless, the requirement in §63.10031(i) to report the information quarterly in a PDF file has been finalized, as proposed.  
The Agency finds the commenter's suggestion that the monitoring and reporting requirements associated with the optional use of paragraph 2 of the definition of startup misplaced, as such comments are beyond the scope of this rule.  In any event, such comments should have been mentioned as part of startup and shutdown reconsideration rulemaking.  In point of fact, few EGU owners or operators have reported using paragraph 2 of the definition of startup (as shown in a recent review, only 20 of 570 EGU owners or operators who reported rely on paragraph 2 of the definition of startup); other EGU owners or operators who choose to continue to use paragraph 1 of the definition of startup are not required to provide such information.  Therefore, EGU owners or operators should consider all implications of their startup definition choice, including reporting and recordkeeping activity, before deciding on an approach.
Note, however, that, consistent with the commenter's recommendation, the hourly CEMS readings during startup and shutdown described in §63.10031(c)(5)(iii), (iv), and (v) have not been carried over into the PDF reports that will be submitted on and after January 1, 2021; these paragraphs have been removed from the rule and reserved.  Startup and shutdown hours are flagged in the quarterly emissions reports, so the hourly CEMS readings for these hours can be identified easily for auditing purposes.
Comment 2: Commenter 20609 stated that in § 63.10011(g)(3), § 63.10021(h)(3) and (i), and § 63.10031(i), EPA proposes to require continued submission of all of the information in § 63.10031(c)(5) in PDF format, rather than XML format like the rest of the compliance report, but solicits comment on whether a more transparent format should be required. 81 Fed. Reg. at 67,066. The commenter stated that it supports the use of PDF format to report the information in (c)(5)(i) and (ii), at least initially, and believes that for the type of information requested it is the most transparent and user-friendly format under the circumstances. Because this information is by its nature unique to the reporting source, XML format may not be sufficiently flexible to accommodate differences in details and there may be little benefit to compiling the information in a single database.
Commenter 20609 stated that it questions the purpose of continuing to require submission of information in PDF format under (c)(5)(iii)-(v) that already has been reported in a quarterly emissions report in XML format. Under the current rule, all EGUs already are reporting hourly average concentration values for Hg and flow in XML format in quarterly emissions reports submitted to ECMPS, and flagging the hours that occur during startup and shutdown. 40 C.F.R. Part 63, Subpart UUUUU, App. A sections 7.1.2.5, 7.1.3.3, 7.1.4.3, 7.2.5. Similar requirements apply for HCl CEMS. Id. App. B. EGUs that use SO2 as a surrogate also are reporting hourly SO2 values in quarterly ECMPS emissions reports, and would continue to do so under the proposed rule. 40 C.F.R. § 75.57(c) and Proposed § 63.10031(5). 
Commenter 20609 stated that under this proposal, EGUs using PM CEMS or PM CPMS would similarly begin reporting hourly PM values on January 1, 2018. Proposed App. C sections 7.1.2.5, 7.1.3.3, 7.2.5.2, and App. D sections 3.1.2.3, 3.1.3.3, 3.2.4. Given that all EGUs with CEMS, sorbent trap monitoring systems, or PM CPMS already are or will be reporting this hourly information in XML format for all operating hours with quality assured data and identifying which hours are startup and shutdown, the commenter stated that it does not understand why EPA would require that the same information be reported in the subsequent quarterly compliance report in PDF format simply because an EGU has opted to rely on paragraph (2) of the startup definition. In light of the purpose of this rulemaking  -  i.e., to eliminate duplication and streamline reporting  -  EPA should remove the requirements in § 63.10031(c)(5)(iii), (iv), and (v) to re-report in PDF format those startup and shutdown emissions-related values that already have been reported in XML format.
Response 2: As mentioned earlier, consistent with the commenter's recommendation, the hourly CMS readings during startup and shutdown described in §63.10031(c)(5)(iii), (iv), and (v), which are required to be included in the semiannual compliance reports, have not been carried over into the PDF reports that will be submitted on and after January 1, 2021; these paragraphs have been removed from the rule and reserved.  Startup and shutdown hours are flagged in the quarterly emissions reports, and the hourly CEMS readings for these hours can easily be identified for auditing purposes.
Comment 3: Commenter 20595 stated if an EGU relies on paragraph (2) of the definition of startup given in §63.10042, the information in §63.10020(e), which is referenced in §63.10031(c)(5), would be reported quarterly in PDF format, as an attachment to the compliance report. The EPA specifically solicited comment on whether this information should be made more transparent and user-friendly. The EPA requested comment on possible techniques to achieve those ends, e.g., by requiring the data to be submitted in XML format. The commenter stated that while it does not utilize paragraph 2 of the definition of start-up, reporting this information appears to be better if handled as a PDF file.
Commenter 20607 supported EPA's proposal to require EGUs relying on paragraph (2) of the definition of startup to report the information required by § 63.10020(e) quarterly in PDF format, as an attachment to the compliance report. The commenter stated that this would continue the current practice of submitting this information in PDF format. Requiring this information to be submitted in XML format would force companies to change the methods by which they collect and report this data; specifically, companies would need to input data into a software application to convert it to XML format prior to submittal. Such a change in practice would be time-consuming and expensive, and would make reporting more burdensome without providing any benefit.
Response 3: As no commenter suggested approaches other than submission of the information in §63.10020(e) in a PDF file, the rule maintains this requirement.  
Comment 4: Commenters 20602 and 20611 stated that a number of provisions in the proposed revisions reference recordkeeping or reporting associated with the information in §63.10031(c)(5) are only applicable if an affected unit chooses to use startup definition (2) found at §63.10042. 
Commenter 20602 stated that these revisions should make clear that the information required by §63.10031(c)(5) is only associated with units where startup definition (2) is being used. An example where this is done in the proposed revisions is the second sentence in §63.10011(g)(3). The commenter suggested that this be clarified in the relevant areas of the proposed revisions including §63.10031(I). The commenter provided the following suggested changes to §63.10031(I):  "If you are relying on paragraph (2) of the definition of startup in§ 63.10042, for startup and shutdown incidents that occur on or prior to December 31, 2017, you must include the information in§ 63.10031(c)(5) in PDF format, in the semiannual compliance report. If you are relying on paragraph (2) of the definition of startup in § 63.10042, for startup and shutdown event(s) that occur on or after January 1, 2018, you must use the ECMPS Client Tool to submit this information in PDF format, as an attachment to each quarterly compliance report starting with the report for the first calendar quarter of 2018. The applicable data elements in paragraphs (f)(6)(i) through (xii) of this section must be entered into ECMPS with each startup and shutdown report."
Commenter 20611 also stated EPA should revise §63.10031(i) to clarify the startup/shutdown reporting requirements only apply to facilities that are relying on paragraph (2) of the definition of startup provided in §63.10042.
Response 4: As suggested by commenters, the rule, at §63.10031(i), clarifies that the information required to be reported under §63.10031(c)(5) and §63.10020(e) applies only to EGUs that are relying on paragraph 2 in the definition of startup.
5.11 (k)-PM CPMS submit excess emissions summary report in PDF quarterly after January 1, 2018
Comment 1: Commenter 20609 indicated that EGUs utilizing PM CPMS have an exception to excess emission reporting in § 6310031(d)(3) and (k) however the proposed purpose and scope of the exception is unclear.  This commenter said that "the use of a PM CPMS would subject an EGU to PDF reporting of excess emissions and monitor evaluation information for all emission limitation and monitoring systems."   Commenter 20609 assumed that EGUs using PM CPMS would continue existing excess emissions and monitoring system evaluation reporting with respect to their PM CPMS and its associated operating limit, and not with respect to other emission limitations or CMS.  The commenter indicated that this reporting method should be better stated.
Response 1: The proposed requirement for PM CPMS to continue the excess emissions report information in §63.10(e)(3)(v) and (vi) after the semiannual compliance reports are phased out has been removed from the rule.  Owners or operators of EGUs that use PM CPMS are required in Appendix D to report the hourly PM CPMS response values, percent monitoring availability (PMA), the 30- boiler operating day rolling averages, and deviations from the operating limit in the quarterly compliance reports.  
7 Table 3-WPS
Comment 1: Commenter 20609 opposed proposes to revise three separate items in Table 3 to add language regarding reporting that mirrors the language in § 63.10011(g)(3) and § 63.10021(h)(3). The commenter said the proposed revision of Item 3(a)(1) to require reporting of the information in § 63.10031(c)(5) is incorrect. Section 63.10031(c)(5) is only applicable to EGUs that opt to rely on paragraph (2) of the definition of startup and Item (a)(1) addresses only paragraph (1) of that definition. The commenter also said the proposed addition of the language to Item 3(d) is incorrect with respect to EGUs that have not elected to rely on the alternative startup definition. Moreover, for EGUs that have elected to rely on paragraph (2) of the definition of startup, the proposed new language is redundant because Items 3(a)(1), 3(d), and 4 already require compliance with § 63.10011(g), § 63.10021(i), and § 63.10031, all of which EPA already has proposed to revise to include the same instructions regarding semi-annual and quarterly compliance reports. The commenter said if EPA does add the language to Item 3(d) or 4, EPA must include the qualifier that it only applies to EGUs that elect to rely on paragraph (2) of the definition of startup. EPA also would have to modify the language in Item 4 to refer to "shutdown" instead of "startup," because Item 4 only addresses shutdown periods. The commenter suggested that EPA should abandon these proposed unnecessary and confusing revisions.
Commenter 20611 stated EPA should make clarifications associated with startup definition (2). The commenter noted a number of provisions in the proposed revisions reference recordkeeping or reporting associated with the information in §63.10031(c)(5). The requirements associated §63.10031 (c)(5) are only applicable if an affected unit chooses to use startup definition (2) found at 40 CFR 63.10042. The revisions should make clear that the information required by §63.1003l(c)(5) is only associated with units where startup definition (2) is being used. An example where this is done in the proposed revisions is the second sentence in §63.1001 l(g)(3). The commenter suggested that this be clarified in the following areas of the proposed Table 3:
 Table 3, paragraph 3(a)(1): The proposed revision to this paragraph relates to §63.10031(c)(5). Since §63 .10031(c)(5) is only applicable to units that rely on paragraph (2) of the definition of startup, there should be no reference to §63.10031(c)(5) in this section of the table. 
 Table 3 paragraph 3.(d): * * * If you are relying on paragraph (2) of the definition of startup in § 63.10042, you must report the applicable information in§ 63.10031(c)(5) concerning startup periods that occur on or prior to December 31, 2017 in PDF format in the semiannual compliance report. If you are relying on paragraph (2) of the definition of startup in§ 63.10042, for startup periods that occur on or after January 1, 2018, you must provide that information quarterly, in PDF format, according to§ 63.10031(i). 
 Table 3 paragraph 4: ***If you are relying on paragraph (2) of the definition of startup in§ 63.10042, you must report the applicable information in§ 63.1003 l(c)(S) concerning startup shutdown periods that occur on or prior to December 31, 2017 in PDF format in the semiannual compliance report. If you are relying on paragraph (2) of the definition of startup in § 63.10042, for startup shutdown periods that occur on or after January 1, 2018, you must provide that information quarterly, in PDF format, according to § 63.10031 (i). 
Response 1: The Agency agrees with the suggestions to clarify and correct Table 3. The rule now clarifies in Table 3 that the requirement to report the information in §63.10031(c)(5) applies only to owners or operators of EGUs that elect to use paragraph (2) of the definition of startup in §63.10042.  The rule also corrects paragraph 4 of Table 3, replacing the word "startup" with the word "shutdown", where appropriate.   
8 Table 8-Reporting Requirements
Comment 1: Commenter 20612 stated the requirement in Table 8 to report pre-2018 PS-11 correlation test results should be revised to allow at least 60 days to report for any test that may occur late in 2017. The commenter suggested the revised version should read: "if, prior to January 1, 2018*, you have begun using a certified PM CEMS to demonstrate compliance with this subpart, you must use the ECMPS Client Tool to submit a PDF report of the existing PS 11 correlation test of the PM CEMS, no later than December 31, 2017* or 60 days after the completion of the PS-11 correlation test, whichever is later." The commenter noted with an asterisk that these data correspond to whatever date the new XLM reporting provisions would become effective.
Commenter 20597 stated the requirement to report pre-2018 PS-11 correlation test results should be revised to allow at least 60 days to report for any test that may occur late in 2017. The revised requirement should only require the current PS-11 correlation test of the PM CEMS be reported.
Response 1: In response to the commenters' suggestions, the rule requires PDF test reports of PM CEMS correlations completed prior to the effective date of the rule to be submitted no later than 60 days after the effective date of the rule.   For correlations completed on or after the effective date of the rule but prior to January 1, 2021, a PDF test report must be submitted within 60 days after the date on which the test is completed.  For correlations completed on or after January 1, 2021, the test data and results must be submitted along with supporting reference method data, electronically, in XML format, as required under section 7.2.4 of Appendix C and the applicable sections of Appendix E.  
Comment 2: Commenter 20609 did not object to the proposed requirement in Table 8 Item 8 for EGUs that elect to demonstrate compliance using a CMS to report rolling average emission rates and related information for all parameters. The commenter suggested with respect Hg, EPA's proposed language does not recognize the option for EGUs that are not participating in a group averaging plan, including EGUs with LEE status, to comply with a 90-boiler operating day average for Hg. The commenter suggested that EPA needs to provide language in Table 8 Item 8 and Appendix E that reflects the availability of the option for EGUs, including those with LEE status, that are not participating in a group averaging plan under § 63.10009. Specifically, Table 8 Item 8 should read: "Starting in the first quarter of 2018, all 30- day rolling averages for all parameters (including Hg, HCl, HF, and/or SO2), or 90-day rolling averages for Hg, must be reported in XML format in the quarterly compliance reports described in § 63.10031(g)." The commenter referenced the revisions made to Table 2 in the 2016 Technical Corrections Rule supporting the 90-day average option. The commenter also added although EPA did not make similar conforming amendments to § 63.10005(a)(2), (d)(3), and (h)(3), which still refer to the 90-boiler operating day averaging only in reference to "certain coal-fired existing EGUs that use emission averaging for Hg," those provisions cannot override Table 5 and the other provisions that explicitly authorize the option. As a result, EPA should take this opportunity to make conforming revisions to those sections as well.
Response 2: Upon consideration of the comments, the rule contains language similar to that suggested by the commenter in Table 8, Item 8, recognizing that owners or operators of EGUs that are not in emissions averaging plans may elect to comply with the applicable Hg emission limit on a 90- boiler operating day rolling average basis.  Regarding the conforming amendments to §63.10005(a)(2), (d)(3), and (h)(3), the words "certain coal-fired existing EGUs that use emission averaging for Hg" appear only in paragraph (a)(2), not in (d)(3) or (h)(3).  These words have been removed from paragraph (a)(2) of the rule. 
Comment 3: Commenter 20612 supported submitting detailed performance test report electronically as PDF files, but suggested EPA remove its proposed list of requirements for the PDF stack tests reports required under Table 8 and Section 22 of Appendix E and simply reference the stack test requirements included in the recently finalized revisions to §60.8(f) and §63.7(g). The commenter said there will be inconsistency between rules if this is not done. The commenter said EPA should revise Table 8 and Section 22 simply to reference the report requirements in §63.7(g), rather than creating a separate list in the MATS Rule, which could introduce inconsistency between rules. The proposed list in §63.7(g) is far more comprehensive than the one included in the MATS revisions and, since EPA has indicated that the revisions to §63.7(g) were made to "ensure that emissions test reporting includes all data necessary to assess and assure the quality of the reported emissions data," the reports will certainly include everything that is needed. The commenter provided an example of how the rule could be changed as follows: "14. For each test described in section 14 of Appendix E to this subpart, a detailed test report in PDF format that meets the requirements of §63.7(g)."
Commenter 20597 stated EPA should remove its proposed list of requirements for the PDF stack tests reports required under Table 8 and Section 22 of Appendix E and simply reference the stack test requirements included in the proposed revisions. Otherwise, the difference between the content lists will introduce inconsistency between rules.
Response 3: As mentioned earlier, the Agency disagrees with the commenters' suggestions to limit the list of items to be reported from performance tests to just those items contained in the NSPS and NESHAP general provisions. The Agency finds that there will be no inconsistency between rules.  The information that is required to be provided in a test report includes not only the information required by 40 CFR §60.8 (g) and §63.7(g) [the general provisions], but also additional information required by this specific subpart.  The required data are needed to replicate the calculations and QA the emissions that have been submitted in order for regulatory authorities to be able to assess compliance with the test methods, performance specifications, or quality assurance procedures.  The rule is unchanged; the submitted report will consist of an XML tags that contain the necessary data elements along with and pdf files containing supporting data, as necessary.
9 Table 9-Applicability of General Provisions to Subpart UUUUU
Comment 1: Commenter 20609 said EPA proposes only one minor change to Table 9 to provide an exception to the applicability of § 63.10(c)(7), which establishes "recordkeeping" requirements for excess emissions, to state that it only applies to EGUs that elect to use PM CPMS. The commenter questioned why the Agency has chosen to provide an exception for recordkeeping, and has not provided any exceptions for reporting. The commenter stated that starting January 1, 2018, EGUs should be exempted from the above listed reporting requirements. That should be made clear in Table 9 and, to cover the possibility that there are other reporting provisions that would duplicate information already reported to ECMPS, by new language in § 63.10011(f).
Commenter 20595 stated the proposed amendments would require all of the 30- (or 90-) boiler operating day rolling averages or WAERs to be included in the quarterly reports. Some excess emissions and monitor availability information would still be included in the compliance reports. Since this information is in place of the excess emission report identified in 63.10(e)(3), the commenter suggested the agency should revise the Table 9 to show that 63.10(e)(3) is not applicable to MATS.
Response 1: Table 9 of the rule has been revised to make it clear that the excess emissions and monitoring system performance recordkeeping and reporting requirements for CMS in §63.10(c)(7) and (c)(8) and §63.10(e)(3)(v) and (vi) apply through December 31, 2020, when the semiannual compliance reports are phased out.  Starting with January 1, 2021, 30- and 90- boiler operating day rolling averages and any deviations (as defined in §63.10042) must be reported for all CMS, in the quarterly compliance reports as described in §63.10031(g).
Comment 2: Commenter 20609 stated that the general provisions require reporting all reports to the delegated authority and the EPA regional office (unless EPA waives the requirement), and after a Title V permit has been issued, reporting the Notification of Compliance Status and performance test results to the appropriate permitting authority. The commenter argued that EPA has no basis to require EGUs that are reporting information directly to a central EPA reporting system to duplicate those reports in separate submissions to states and regions, and that they have access to the relevant information through the SMS. The commenter noted a few places in the proposal where EPA should address this issue. The commenter said EPA must make other revisions to make clear that such duplication is not required. Specifically, the commenter suggested EPA either amend Table 9 to make clear that the duplicative reporting in the general provisions is not applicable to Subpart UUUUU and/or by including in § 63.10011(f) a statement that: "Submission of a report using the ECMPS Client Tool satisfies any requirement under subpart A of this part to submit that report (or the information in that report) to the appropriate EPA Regional Office or State agency of a delegated authority." The commenter noted EPA did this in a similar issue with respect to written "excess emissions" reports under § 63.10(e)(3)(v). The commenter said if EPA does not do this, EPA must explain and solicit comment on its rationale for requiring duplicative reporting of this information and how it can be reconciled with the proposed electronic submission requirements in § 63.10031(b)(2) and (3).
Commenter 20608 stated the proposed revisions should clearly indicate that the electronic reporting requirements satisfy general Part 63 reporting requirements (e.g., a source that submits MATS stack test data via ECMPS would not still need to submit that same data to the state and EPA regional office under 63.13 unless specifically required by the state or regional office to do so). EPA needs to clarify that submitting the files through ECMPS will meet the requirements to report the same information to the state/region in the general provisions of Part 63.
Response 2: See the response to comment 3 of section 1.2.  
10 Appendix A-Hg. Monitoring Provisions
Comment 1: Commenter 20609 supported the clarifications EPA proposed for Appendix A.
Response 1: The Agency appreciates the commenter's support for the clarifications.
Comment 2: Commenter 20609 stated that throughout existing Appendices A and B, and proposed new Appendices C and D, EPA refers to "quarterly reports." The commenter said although these quarterly reports are different from the "quarterly compliance reports" required under proposed § 63.10031(g) and Appendix E, the fact that the reports are both submitted quarterly, but according to different deadlines, can be confusing. In proposed revised Table 8, EPA refers to the reports under Appendices A, B, and C as the "quarterly emissions report" to distinguish it from the "quarterly compliance report. The commenter suggested that for consistency, EPA use that same term in the appendices themselves. To avoid confusion, the commenter suggested EPA should make conforming revisions so that the term "quarterly report" refers to the quarterly report of PM and PM CPMS averages submitted for the period ending on December 31, 2017, the term "quarterly emissions report" refers to the quarterly reports submitted under Appendices A-D, and "quarterly compliance report" refers to the quarterly reports submitted under Appendix E.
Response 2: The Agency disagrees with the suggestion to make a clearer distinction regarding quarterly reports.  As a result, the rule has not been revised such that different terms are used throughout the rule to distinguish among the reports.  The Agency believes that the recommended changes are unnecessary and that the rule in able 8 provides sufficient clarity regarding the content of each type of report and its submission deadline.   
11 Appendix B-HCl and HF Monitoring Provisions
Comment 1: Commenter 20607 said the MATS rule provides a daily assessment startup grace period for CO2, SO2, and NOx, which allows a unit to complete a daily calibration during startup without losing any data. The commenter urged EPA to expand this provision to HCI and Hg to make the program consistent and facilitate calibration for all pollutants.
Response 1: The Agency agrees that the MATS provides for a daily calibration assessment startup grace period for CO2, SO2, and flow monitoring systems which are used under the rule. This and other Part 75 monitoring provisions are allowed under MATS as these monitoring systems were already installed for use in meeting Part 75 monitoring requirements. For HCl monitoring systems, Sections 5.1 and 5.2 of Appendix B set out the provisions for on-going quality assurance depending on whether the affected facility has chosen to comply with Performance Specification 15 (PS-15, 40 CFR 60, Appendix B) or Performance Specification 18 (PS-18, 40 CFR 60, Appendix B), respectively. For an HCl monitoring system certified under PS-15, calibration error must be assessed `on a daily basis' as per Section 5.1.1 of Appendix B. For an HCl monitoring system certified under PS-18, calibration drift must be assessed `at least once daily (approximately every 24 hours)' as per Procedure 6 referenced in Section 5.2 of Appendix B. As there are no limitations as to whether these calibrations are conducted on- or off-line and EGU owners or operators have some latitude in scheduling these calibrations within a 24-hour period, the Agency sees no need to adopt the Part 75 provision's daily calibration assessment startup grace period for HCl monitoring systems in MATS. For Hg monitoring systems, Section 5.1.1 of Appendix A of the MATS mandates that calibrations be done with the affected unit operating (on-line). The comments regarding this provision for Hg monitoring systems are beyond the scope of this rulemaking because we did not propose to allow either the startup grace period or the off-line calibration for these systems. No changes to the rule have been made.
Comment 2: Commenter 20607 urged EPA to expand the on-line/off-line daily calibration procedures in 40 C.F.R. Part 75, Appendix B, Sections 2.1.1 and 2.1.5.1 to apply to Hg and HCl monitors. This would allow offline daily calibrations on infrequently operated units to be used to validate data on a limited basis. It also would decrease emissions by minimizing the amount of times the bypass stack would have to be brought online just for calibration purposes.
Response 2:  For the same reasons noted in the response to the previous comment, the rule does not adopt the 40 CFR Part 75 provisions for on-line / off-line daily calibration procedures for HCl and Hg monitoring systems.

11.1 Monitoring of HCl and/or HF Emissions
Comment 1: Commenter supported the correction of an error in the existing Appendix B section 2.3 heading by removing the reference to "FTIR," so that the section heading simply reads "Monitoring System Equipment, Supplies, Definitions, and General Operation." The commenter said that is necessary because the subsections address both FTIR and traditional CEMS.
Response 1:  The Agency appreciates the supportive comment.  The amendment to the heading of Appendix B, section 2.3 now appears in the rule as proposed.
11.2 Recordkeeping Requirements
Comment 1: Commenter 20612 stated the proposed change to Appendix B 10.1.8.1.9 for HCl and HF monitoring is not necessary. The commenter said this section discusses the performance of the interference test of the HCl monitor. The last sentence requires the user to "keep records to document the quantity and quality of gases, gas volume/rate, temperature, and pressure used to conduct the test." The commenter said there are no specifications for any of those parameters in Performance Specification 18, and simply puts an additional burden on the user and layer of complexity to the interference check with no apparent benefit in the quality of the data. This sentence is unnecessary and should simply be removed.
Response 1: The Agency notes that there are no specific criteria in the rule set in PS-18 for the items required to be documented in Section 10.1.8.1.9; however, Section 11.1.7 of PS -18 specifically requires documentation (not reporting) of the `gas volume/rate, temperature, and pressure used to conduct the interference test' because all these parameters, as well as the gas quality, are germane to proper scientific documentation of the test. No changes to the rule have been made.
Comment 2: Commenter 20612 suggested EPA clarify and revise under the heading "Miscellaneous Clarifications/Revisions" by correcting the word order in Section 10.1.8.1.3 of Appendix B. The commenter said the text should read "...relevant process operating data, and all the calculations used to determine the relative accuracy."
Response 2: The Agency agrees with the commenter's suggestion, and the rule has incorporated the suggestion.  The word order in Appendix B, section 10.1.8.1.3 has been corrected.
11.3 Reporting Requirements
Comment 1: Commenter 20612 requested simplifying the references for calculation of standard deviation and confidence coefficient since both Performance Specifications, PS-2 and PS-18 provide the same calculations. The commenter said section 11.4.3.11 of Appendix B refers to the calculation of standard deviation during a relative accuracy test audit (RATA). It states the user should use Equation 2-4 in Performance Specification (PS) 2 in Appendix B of Part 60, unless the user is following PS 18. For those using PS 18, it states to calculate standard deviation following the procedures in Section 12.6 of PS 18. Both methods/equations will give the user the exact same value. The commenter said this section is unnecessary or should be rewritten to cite either Performance Specification. Similarly, the commenter said section 11.4.3.12 of Appendix B refers to the calculation of confidence coefficient during a relative accuracy test audit (RATA). It states the user should use Equation 2-5 in PS 2 in Appendix B of Part 60, unless the user is following PS 18. For those using PS 18, it states to calculate the standard deviation following the procedures in Section 12.6 of PS 18. Both methods/equations will give the user the exact same value. This section is unnecessary or should be rewritten to cite either Performance Specification.
Response 1:  While the Agency understands that simplification of the language in sections 11.4.3.11 and 11.4.3.12 of Appendix B could be desirable, we do not view reference to both PS-2 and PS-18 as unnecessary. The references to the PS-2 calculations for performing RATAs of an FTIR CEM certified under PS-15 are in keeping with the language of section 3.1.2 of Appendix B which specifically refers to section 11.1.1.4 of PS-15 for conduct of a RATA which, in turn, refers to PS-2 for calculations needed to compare the reference method results to the FTIR CEMS results. We find the commenter's suggestion to cite the calculations of either performance specification as appropriate for calculation of the RATA standard deviation and confidence coefficient is a reasonable compromise and sections 11.4.3.11 and 11.4.3.12 of Appendix B of the rule have been revised to provide that option.
Comment 2: Commenter 20612 disagreed with EPA's statement that the new reporting and recordkeeping requirements to Appendix B will not increase the burden. The commenter said the revisions do indeed increase the reporting requirements for sources using HCl CEMS. The commenter disagreed that EPA should make this type of statement in the first place. Instead of making statements such as if one option is too burdensome then a source can choose another, the Agency should strive to improve all of the monitoring and reporting options under any rule and eliminate unnecessary burdens whenever possible. EPA should always consider the balance of the burden that it is imposing with the benefits and provide meaningful justification for its actions.
Commenter 20609 stated that the proposed sections 10 and 11 would significantly increase the amount of information required to be recorded and reported. The commenter did not object to recording the information specified in section 10.0, but questioned the usefulness of reporting all of the information specified in section 11.0, especially in XML format. Because much of the information will have to be hand entered into the DAHS so that it can be formatted to XML, such reporting will significantly increase the burdens of using HCl CEMS. The likelihood of transcription errors also is high, thus increasing the possibility that EPA may have to spend time investigating reported information to determine the source of the error (whether in transcription or in the test itself). The commenter said EPA must better explain the need for the reported information and include that in an ICR to comply with the PRA. To the extent EPA can justify submission of the information, the commenter suggested that EPA allow submission in PDF format. That way, EPA and delegated states would have records they could audit on a periodic basis, without imposing the significant burdens associated with the proposed XML reporting.
Response 2: The Agency remains committed to assessing monitoring costs and reducing their burden where possible, in addition to bringing new measurement technologies, such as those HCl CEMS covered by PS 18 and Procedure 6. In this case, the Agency finds no increase in burden due to HCl CEMS use as the rule specifies in sections 10 and 11 of Appendix B the data elements that must be recorded and reported for each of the tests required by PS 18 and Procedure 6.  Electronic reporting of these tests through ECMPS would have been required if PS 18 and Procedure 6 had been in place when the original MATS rule was published.  In view of this, for EGU owners or operators electing to use HCl CEMS, the amendments to section 11 of Appendix B introduce no unnecessary reporting burden, and these provisions of the rule are unchanged.  The results of certification and on-going QA tests must be reported electronically for all CEMS required under this rule in order for ECMPS to assess the quality-assured status of the emissions data.  Further, not all of the tests described in section 11 are required for all HCl monitors.  For example, some of the tests ( beam intensity, temperature and pressure verifications) are specific to IP-CEMS, and Procedure 6 offers a choice among three different types of audits (cylinder gas audits, relative accuracy audits, and dynamic spiking audits) for the required quarterly QA tests. In addition, the Agency notes that the rule's reporting requirements for the interference check (which is not necessarily performed on each individual analyzer) have been simplified.  
12 Appendix C-PM Monitoring Provisions (CEMS)
Comment 1: Commenter 20609 stated that promulgation of the Appendix C will cause redundancy with Section 63.10010, which addresses installation, operation and maintenance requirements for all CMS and CPMS. The requirements for PM CEMS are covered in § 63.10010(i). The commenter said this could be confusing given its absence of recognition of the more specific instructions in Appendix C. To avoid this, and to maintain consistency with how Hg CEMS are treated in § 63.10010(g), the commenter suggested that EPA revise § 63.10010(i) to refer to the new requirements in Appendix C, rather than refer directly to PS 11 and Procedure 2, and eliminate duplicative details. Mirroring § 63.10010(g) for Hg, such a provision would read: 
"(i) If you use a PM CEMS you must install, certify, operate, maintain and quality assure the data from the monitoring system in accordance with appendix C to this subpart. You must calculate and record a 30-boiler operating day rolling average PM emission rate, in units of the standard, updated after each new boiler operating day. Each 30- boiler operating day rolling average emission rate, calculated according to section 6.2 of appendix C to the subpart, is the average of all of the valid hourly PM emission rates in the preceding 30-boiler operating days." 
The commenter said this revision would simplify the rule and avoid the possibility for inconsistency between that provision and other provisions of the rule, like § 63.10020(b)-(d) and Appendix C section 6.1, which already cover data collection and reduction. The commenter said when EPA makes this revision, the details in § 63.10010(i)(2)(i) and (ii), establishing the frequency for conducting the RRA and RCA required under Procedure 6, will need to be moved to Appendix C.
Response 1: The Agency agrees with the commenter's suggestion to simplify §63.10010(i), and the rule now incorporates it.  Rule sections of Appendix C pertaining to certification, operation, and maintenance of the PM CEMS, as well as the recordkeeping and reporting sections of Appendix C, are simply cross-referenced in revised paragraph (i). 
12.1 General provisions
Comment 1: Commenter 20609 noted that proposed Appendix C, and current §63.10010(i), refer to "certification" of a PM CEMS according to PS 11. Appendix C similarly refers to "recertification" of PM CEMS. PS 11 addresses the "initial installation and performance procedures that are required for evaluating the acceptability of a PM CEMS" PS 11 ¶ 1.0. Those are accomplished through calibration procedures and development of a site-specific correlation to an appropriate reference method, referred to as the "performance specification test." PS 11 ¶¶ 2.0, 8.0. Procedure 2 establishes "the minimum requirements for evaluating the effectiveness of QC and QA procedures" and the quality of data produced by a PM CEMS. Neither PS 11, nor Procedure 2, refer to "certification" or "recertification" of a PM CEMS. The commenter suggested to bridge this gap, while maintaining consistency of the new Appendix C with the terms used in other parts of the rule, that EPA replace the language in Appendix C section 1.2, requiring the owner or operator to "comply with the initial certification and recertification procedures of [PS 11]," with a requirement to "certify and, as applicable, recertify the PM CEMS by complying with the calibration and correlation requirements in [PS 11]." That language would equate the terms "certification" and "recertification" and with compliance with the requirements of PS 11, even when PS 11 does not use those terms.
Response 1: The Agency finds the commenter's suggested clarifications helpful, and section 1.2 of Appendix C of the rule has been modified.  In addition, section 1.2 of Appendix C states that the owner or operator must certify, and, if applicable, recertify the PM CEMS according to Performance Specification 11.

12.3 PM Emissions Measurement Methods
Comment 1: Commenter 20612 summarized that EPA is proposing that sources that correlate PM CEMS in terms of actual conditions (mg/acm vs. mg/scm) would be required to report hourly PM concentrations in both units of "mg/acm" and "mg/scm" and to report hourly temperature and pressure values as well. The commenter recommended that all PM CEMS, regardless of the type, instead only be required to report hourly "mg/scm" values. This approach will simplify the reporting and make the reported data more consistent from unit-to-unit. Reporting the data in terms of standard conditions would be consistent with how the hourly mercury concentration (ug/dscm or ug/wscm) and volumetric flow (scfh) data are already reported under the MATS Rule. For both stack flow and mercury concentrations, the data are corrected based on the measurement temperature and pressure to EPA standard conditions. The temperature and pressure, which are not themselves reported, are simply interim components of the reported concentration or flow values. The same straightforward calculations are used to determine the standardized PM concentration in units if "mg/scm", and the same standardized only reporting should be applied.
Commenter 20606 stated in cases where the PM CEMS measurement basis is milligram per actual cubic meter (mg/acm), the EPA proposes to require the reporting of PM concentrations as both mg/acm and mg/wet standard cubic meter (wscm) [there are also similar requirements at 3.2.2.3]. The commenter said EPA provides no basis from reporting PM concentrations on both bases. It is further noted that this approach is inconsistent with the reporting of other CEMS based measurements that are corrected to a stipulated reference condition. For example, 40 CPR Part 75 flow data must be reported in units of standard cubic feet per hour. The commenter said EPA does not also require the reporting of flow data in units of actual cubic feet per hour. The commenter suggested the EPA should simply require reporting of PM CEMS concentration in units of mg/wscm, with a requirement to record other fundamental readings used to determine the mg/wscm concentration if the PM CEMS measures on a different basis.
--------------------------------------------------------------------------------
Response 1:  See Response 1 of section 5.3(c).
12.5 Ongoing QA and Data Validation
Comment 1: Commenter 20612 suggested to add "as applicable" for PM CEMS SVA test references since the SVA test does not apply to all PM CEMS but only to extractive monitors that use a volume measurement to calculate the PM concentration.
Commenter 20606 stated section 5.3.2.1 pertains to daily and quarterly QA assessments and currently includes the following language: "You must also keep records of the procedures used to perform quarterly ACA and SVA audits." The preceding language would suggest that both ACA and SVA audits apply to all CEMS, but only extractive PM CEMS are subject to an SVA audit. Thus, the language should be revised as follows: "You must also keep records of the procedures used to perform quarterly ACA and SVA audits, as applicable."
Response 1: The Agency agrees with the commenters' suggestions, and section 5.3.2.1 of Appendix C of the rule now states that, "You must also keep records of the procedures used to perform quarterly ACA and (if applicable) SVA audits." 
Comment 2: Commenter 20612 stated that in addition to including conditionally valid data and grace period provisions similar to those it already allows for other CEMS, Appendix C should include revisions to conduct ACAs and SVAs no less than 30 days apart to the extent practicable (rather than the 60 days requirement in Procedure 2) and allow for adjustment to PM CEMS response range as long as an ACA is conducted within 168 unit or stack operating hours. The commenter said just as the MATS requirements for mercury CEMS do not rotely follow the specifications in Appendices B and F of Part 60, the MATS requirements for PM CEMS can allow greater flexibility where appropriate. The later allowance to include some sort of accommodation that would allow sources to revise the range of a PM CEMS is particularly needed since, because the PM CEMS responses do not directly correlate with the emissions, it is difficult to know what range will be adequate prior to conducting the correlation test.
Response 2: The commenter's suggestions have not been incorporated in the rule.  In order to ensure that quarterly audits (i.e., ACAs and SVAs) are performed at regular intervals, the Agency continues to require that the quarterly audits "must occur no closer than 2 months apart" as specified by Section 10.3 of Procedure 2 (40 CFR part 60, Appendix F).  In order to allow more flexibility when scheduling the quarterly audits, the Agency advises that, unlike Part 75, these audits may be performed when the EGUs are not operating.  Additionally, the Agency disagrees that an EGU owner or operator should be allowed to revise the response range of a PM CEMS.  Modifying the response range of a PM CEMS after the correlation curve has been developed may modify the correlation relationship between the PM CEMS output and the PM emissions readings; therefore it is not allowed.
Comment 3: Commenter 20609 stated EPA excluded from Appendix C any discussion of the required frequency for conducting RRAs and RCAs. The commenter noted that in the preamble, EPA notes only that: "in contrast to appendices A and B, the familiar QA operating quarter and grace period scheme would not apply to the on-going QA tests of the PM CEMS." The commenter said the proposed rule would explicitly provide in section "5.1.2 Out-of-Control Periods" that a PM CEMS is out of control "when a required QA test is not performed on schedule." Under § 63.10020(c), when a CMS is out-of-control its data cannot be used to compute compliance averages. 
Commenter 20609 stated that the first major flaw is EGUs are free to apply any reasonable interpretation of the phrases "annually" and "every 3 years," EGUs may not be performing the tests or reporting the results on a consistent basis. To the extent EPA wants to avoid that result, the commenter suggested EPA must propose for comment, and then promulgate, rule language that is clear. Commenter 20609 stated that the another flaw is whatever interpretation an EGU employs, it is inevitable that at some point the EGU will miss the deadline for an RRA or RCA either because it was not operating or because some event (including a weather event or an unscheduled outage at the time of a previously schedule test) prevented the test from being scheduled or conducted prior to the deadline. When this happens, the EGU will have to report the data as out-of-control for all of the operating hours until the test can be scheduled and successfully completed. That could result in days or weeks of invalidation of otherwise valid data. The commenter stated EPA has never promulgated a rule that declares data to be out of control as a result of a missing a deadline for a QA test without also providing a "grace period" for performing that test if the EGU misses a deadline. This rule should be no different. The commenter said EPA wants to encourage the use of PM CEMS, it must promulgate rules that not only are clear, but that can be reasonably implemented. EPA's proposed rule would punish EGUs that opt to use PM CEMS by exposing them not only to uncertainty regarding their testing deadlines, but also to the potential for significant loss of otherwise quality assured data. The commenter requested that EPA harmonize the rule by providing "grace periods" for RRAs and RCA that are similar to those already provided under this rule for Hg, HCl, and SO2 RATAs. This request is necessary to accommodate the realities of how EGUs operate. 
Commenter 20612 stated EPA should add conditional data and grace period provisions for PM CEMS. Appendix C should allow similar conditional data and grace period requirements for PM CEMS as the Agency allows for mercury and HCl CEMS under Appendices A and B. While not as important for some of the other QA tests (since Procedure 2 allows ACA and SVA tests to be performed while the unit is off-line), Appendix C should allow grace periods for PS-11 correlation, RCA and RRA tests. The commenter disagreed with the restriction on like-kind analyzers provided that the analyzer is indeed of identical design and operation and that the same setup configuration and correlation as the original are applied, and said EPA provided no justification for this. The commenter also disagreed with the restriction on conditionally valid data, and stated if one bases the validity of the data from the analyzer on being able to demonstrate satisfactory performance via a series of quality assurance tests and if no adjustments are made either before or during the test, why would the analyzer responses be considered representative after the test but not for the period before the test? The commenter added the exclusion of the conditionally valid data provisions would provide sources with a way to avoid unnecessary "out-of-control" periods of monitor downtime when no compliance information would be reported even though representative data are available. The commenter stated that including grace period provisions in Appendix C, like those included in Appendices A and B, would be consistent with EPA's statements about not requiring sources to operate a unit solely to conduct performance tests. 
Commenter 20597 said EPA should allow similar conditional data and grace period requirements for PM CEMS as the Agency allows for mercury and HCl CEMS under Appendices A and B. While not as important for some of the other QA tests (since Procedure 2 allows ACA and SVA tests to be performed while the unit is off-line), Appendix C should allow grace periods for PS-11 correlation, RCA and RRA tests.
Commenter 20599 suggested that EPA allow a grace period and conditional data for performing PM CEMS quality assurance tests similar to the current requirements for Hg, HCI, and S02 CEMS.
Commenter 20591 stated Subpart UUUUU, Appendix A, Section 5.1.2.2 only requires Hg linearity tests during QA operating quarters, with at least one linearity test required every four calendar quarters. Appendix A, Section 5.1.2.4 sets the Hg RATA test frequency to once every 4 QA operating quarters up to a maximum of eight calendar quarters. Appendix A, Section 5.1.3 allows grace periods for Hg linearity checks, 3-level system integrity tests, and RATA tests. Appendix B allows these same basic extensions and exemptions for HCl RATA testing and linearity testing. The commenter suggested that these same extensions and exemptions should be made available for PM systems for RRA and RCA testing. Specifically, an RRA should be performed every 4 QA operating quarters with it being required at least once every 8 calendar quarters. For an RCA, the commenter suggested that it would only have to be performed every 12 QA operating quarters and at least every 20 calendar quarters. The commenter suggested allowing the 720 operating hour grace period for RRA and RCA testing if not completed by the end of the required testing period. For infrequently operated units/stacks this should allow accurate PM data without requiring possibly higher emitting units/stacks to operate simply to complete required QA testing of the PM monitoring systems. 
Commenter 20607 stated while the proposed rule would provide conditional data validation and grace period provisions for Hg and HCl CEMS, it would not apply these same provisions to ongoing QA tests of PM CEMS. The commenter urged EPA to apply these provisions to PM CEMS. The commenter said EPA provides no basis for why these provisions are not allowed for PM CEMS, and they noted EPA states only that "for technical reasons, the use of temporary like-kind replacement PM analyzers and the conditional data validation provisions in §75.20(b)(3) would not be allowed." Indeed, there is no reason why they should not be available for PM CEMS. 
Commenter 20607 urged EPA to revise the proposed rule to require PM RRA reports and RCA testing to be based on QA operating quarters. Under the Proposed Rule, an RRA test must be performed every year, and an RCA test must be performed every three years. In contrast, Hg and HCl RATA tests must be performed every four QA operating quarters, so that operators are not required to run a test every year on a bypass stack. This testing schedule would be similarly beneficial for PM RRA and RCA tests on infrequently operated units and stacks. Accordingly, the commenter recommended that RRAs be completed every four QA operating quarters (with a maximum extension of two years), and the RCA be completed every 12 QA operating quarters (with a maximum extension of every six years). 
Commenter 20594 stated that the proposed rule omits provisions for PM CEMS for the QA operating quarter, grace period, and conditional validation. EPA provides no rationale for the exclusion of these provisions. Curiously, EPA provided these provisions for mercury (Hg) CEMS and hydrogen chloride (HCl) monitoring systems. QA operating quarter, grace period, and conditional validation provisions are particularly essential for RRAs and RCAs, which must be performed while the source is operating. Disallowing the use of QA operating quarters or grace periods will force sources to operate and, therefore, have additional emissions just to conduct testing. In the past, EPA has articulated a contrary position to not force operation for the sole purpose of testing. The commenter recommended that the frequency for an RRA of the PM CEMS be on an annual basis (once every four QA operating quarters) and the frequency for an RCA of the PM CEMS be once every twelve QA operating quarters. If an RRA or RCA is not conducted during the quarter in which it is due, the commenter proposed that a 720-hour unit or stack operating hour grace period be allowed. These suggestions are consistent with the QA Operating Quarter Language for Hg CEMS in Appendix A, 5.1.2.4 and Table A-2.
Commenter 20594 stated the use of conditionally valid data is vital to avoid long periods of missing data from PM CEMS. To certify or recertify a PM CEMS, sources must conduct and extensive series of tests. For a PM CEMS correlation varying control device, performance must occur. The testing is at least as complex as a RATA and as such should be afforded the same option of using conditional data validation. If a PM CEMS experiences a failure that necessitates a replacement, EPA has effectively set up a situation in which no valid data would be reported until the testing could be scheduled after the repair occurs. Since certification tests typically require coordination with outside stack testing firms, the source may have many weeks without valid PM data. During this time the source would experience a drastic reduction in its PMA, and EPA would have an extended period of time with no data to assess compliance with the standards.
Commenter 20594 stated the conditional validation procedures are not punitive, but rather allow a source to report valid data from a monitor contingent on the successful completion of the required QA tests within the specified testing window. Once a PM CEMS is in service, the raw response of that PM CEMS is adjusted by the mathematical correlation established during the PS-11 correlation testing. The commenter did not understand why it would not be appropriate to apply the mathematical correlation retroactively to the instrument raw response assuming that no other adjustments are made following the probationary calibration error test. Thus, the commenter recommended that the allotted window of time to complete an RRA or RCA mirror that for a RATA, as specified in 40 CFR § 75.20(b)(3)(iv) (720 consecutive unit or stack operating hours after the initiation of a probationary calibration error test). These procedures would be consistent with the conditional validation procedures EPA proposed for Hg CEMS in Appendix A, Section 5.1.5 and for HCl CEMS in Appendix B, Section 5.3.3.
Response 3: Upon review of the comments, the Agency maintains that to the extent practicable, the results of an RRA, RCA, or correlation test of a PM CEMS should be submitted electronically prior to or concurrent with the relevant quarterly emissions report.  However, when this is not possible (e.g., for tests conducted near the end of the quarter), the rule at section 7.2.4 of Appendix C now allows the results to be submitted up to 60 days after the test completion date, and the emissions data may be assigned a "provisional" status, pending receipt of the test results.  If the test is successful, then, when the test results are submitted electronically, the status of the provisional data changes to "quality-assured" and no further action is required.  If the test is unsuccessful, then, when the results are submitted, data with the provisional status must be invalidated until a subsequent test of the same type is successful; in this case, resubmission of the affected quarterly emissions report(s) is required.  
The rule at section 5.1.2 of Appendix C now specifies that an RRA must be done once every four calendar quarters and an RCA must be done once every twelve calendar quarters.  For both tests, section 5.1.3 of Appendix C provides a grace period, which may be used in cases where unforeseen circumstances prevent a test from being completed on schedule.  The grace period begins immediately after the end of the calendar quarter in which the test is due.  The length of the grace period is the lesser of 720 operating hours or one calendar quarter.  Note that when a grace period RRA or RCA is completed, the deadline for the next test is four (for the RRA) or twelve ((for the RCA) calendar quarters after the calendar quarter in which the test was originally due, rather than the calendar quarter in which the grace period test is completed. 
Comment 4: Commenter 20597 supported changes to replace the semi-annual reports with the quarterly compliance reports, but urged the EPA to maintain the 60-day reporting deadline to ensure that tester and sources have adequate time to analyze, review, and certify the quality of the emissions data being reported. The 60-day reporting requirement is critical to provide adequate time for PM and HCl stack test data. Unlike for Part 75 related stack testing (where sources generally rely on instrumental reference methods), the 60-day period is necessary to provide sources adequate time to get PM and HCl test results back from the labs and review the results. However, in addition to sources that demonstrate compliance based on quarterly stack testing, those that use PM and HCl CEMS must also rely on periodic stack tests (RATAs, PS-11 correlation tests, RCAs and RRAs) using the same non-instrumental reference methods for quality assurance. The commenter said with the proposed rule, any CEMS data would need to be reported in the quarterly Emissions reports within 30 days after the end of the quarter. The associated QA test data (including data for RATAs, PS-11 correlation tests, RCAs and RRAs) must be reported in the QA files prior to submitting the emissions report to validate the emissions data and, naturally, the results of those QA tests can affect the reported emissions values. Thus, there is a contradiction in the proposed reporting requirements since EPA proposed to allow 60 days for sources to report stack test related data in Appendix E, but the proposal only allows sources 30 days to report similar (and in some cases the same) stack test related data in other provisions. As for the performance test results, the proposed 30-day reporting deadline will not be adequate for the PM and HCl stack test results used to certify and quality assure CEMS. EPA should revise the MATS Rule to allow sources to submit the MATS related Emissions and QA file records within 60 days after the end of the quarter. The commenter said the Agency will need to make some modifications to the current ECMPS submission process to address this change since sources will still have a 30-day reporting deadline for reporting ARP, CSAPR related emissions and QA File data under Part 75, but a 60-day MATS reporting deadline is necessary to ensure that sources will have time to both receive and appropriately review the results before submittal.
Commenter 20597 provided two alternatives to accommodate this change. First, they said EPA could change the reporting deadline in Part 75 to be consistent with the need for a 60-day deadline for MATS. This would be the simplest approach to implement and would have little impact on the timeliness of the reported data under Part 75. This would also benefit the quality of the review of such data, particularly given the voluminous increase in reporting that has and is being mandated by this and other rules. 
Commenter 20597 said, alternatively, sources could be required to submit provisionally quality assured MATS emissions data with the Part 75 data during initial 30 days after the end of the quarter pending reporting of the final RATA or PM CEMS QA test results for HCl or PM CEMS. This change would be consistent with the current provisions §63.10031(t)(5) and would allow sources to benefit from the paperwork reduction that is supposed to be an advantage of electronic reporting and avoid continuing to have to report hardcopy unnecessarily. States and other delegated authorities already have the right to request any report in hardcopy, if deemed necessary, under §63.10031(f)(5).(
Response 4:  Upon consideration of the commenter's suggestions, the Agency agrees that in some instances additional time for submission of quarterly reports may be warranted and has adjusted the rule to identify those instances and provided procedures to be used when those instances occur. The requirement to submit the quarterly compliance reports within 60 days after the end of each quarter remains in the rule, as proposed. However, as mentioned earlier, instead of incorporating the commenter's suggestion to move the submission deadline for emissions reports from 30 days to 60 days after the end of each quarter, the rule allows the results of HCl RATAs and RRAs, RCAs, and correlation tests of PM CEMS to be submitted up to 60 days after the test completion date, when necessary.  To the extent practicable, the results of these tests should be submitted prior to or concurrent with the relevant quarterly emissions report (i.e., no later than 30 days after the end of the quarter).  But when this is not possible (e.g., for tests conducted near the end of the quarter), the emissions data may be assigned a provisional status (as suggested by the commenter), pending receipt of the test results.  If the test is successful, then, when the test results are submitted electronically, the status of the provisional data changes to "quality-assured" and no further action is required.  If the test is unsuccessful, then, when the results are submitted, the provisional data must be invalidated until a subsequent test of the same type is successful; in this case, resubmission of the affected quarterly emissions report(s) is required.   
Comment 5: Commenter 20609 opposed the suggestion that a "monitoring system malfunction" always meets the definition of "out-of-control" period. Section 63.8(c)(7) defines an "out of control" period as one in which a monitoring system fails to meet a performance specification. Although an "out of control" period clearly could be caused by a "monitoring system malfunction," some monitoring system malfunctions may not coincide with QA activity. The commenter referenced other rules that recognize that potential, such as the requirement of maintenance records in § 63.10(b)(2)(vi). The commenter was concerned with the proposed wording of Appendix C section 5.1.2 is that it could be read to mandate that a PM CEMS malfunction that does not involve failure of a performance specification be characterized in monitor downtime reports as an "out of control" period rather than monitoring system malfunction, contrary to the definition of "out of control" period and contrary to the way such periods would be reported for SO2, Hg, and HCl. The commenter recommended that EPA should remove the reference to monitoring system malfunctions in the section. This would make clear that such periods must be reported as monitor downtime and that any data collected by a malfunctioning monitor cannot be used in compliance calculations. The commenter noted that Subpart UUUUU already includes another provision that could be viewed as inconsistent with treatment of the two as potentially different types of events. Specifically, the definition of "monitoring system malfunction" in § 63.10042. The commenter said this definition was in the original rule but was ever proposed. The commenter stated the definition recognizes that a monitoring system malfunction can result in an out-of-control period. Any other reading would be inconsistent with the definition of out-of-control period in § 63.8(c)(7). The commenter suggested, to avoid further confusion and to promote consistency, that EPA also remove the phrase "out-of-control" from the definition of "monitoring system malfunction" in § 63.10042.
Response 5: Upon consideration of the comment, the Agency has removed the language discussing monitoring system malfunction in section 5.1.5 (formerly Section 5.1.2) as suggested by the commenter.  Section 5.1.5 has been changed to make it similar to the language already in Appendix A and Appendix B.
Comment 6: Commenter 20597 stated concern with reporting of similar files for both performance tests and QA tests since EPA has suggested that the files would be used for emission factors development and "other purposes." The CEMS QA test are not performance tests and, therefore, do not necessarily represent periods where the unit is performing optimally. This is particularly true for PS-11 correlation and RCA tests where some data may represent intentionally elevated emissions. Comingling the information within the same database could lead to errant emission factors or emission limits. The commenter suggested that only actual performance tests and 30-day average CEMS data used to demonstrate compliance should be used for emission factor purposes.
Response 6: The Agency appreciates the commenter's caution regarding the potential use of emissions data obtained from performance or quality assurance testing. As the commenter suggests, typical emission factors represent national average emissions from sources operating at normal conditions and are used in the national emissions inventory. However, emissions data can be useful for purposes other than the national emissions inventory.  For example, it may be informative for rule writers, regulators, PM instrument developers, or diagnostic technicians to be aware of and to have emissions data from operational conditions other than normal. Moreover, some existing emission factors have been developed for varying load conditions. In any event, as the commenter should know, the public is afforded an opportunity to review and comment on emissions factors before they are finalized, so specific concerns such as those expressed by the commenter can be raised as part of the emissions factor development process.    
12.6 Data reduction and Calculations
Comment 1: Commenter 20606 stated that Appendix C Section 6.2.2.2 relates to the F-factor used to convert PM concentration to units of lb/mmBtu. The commenter said the EPA stipulates that the EGU must use the applicable F-factor as listed in Table 19-2 of Reference Method 19, if available. If not available, the EGU may then use F-factors from 40 CFR Part 75, Appendix F, Sections 3.3.5 or 3.3.6. The commenter noted the preferential listing of Table 19-2 of Reference Method 19 could be problematic in that there may be instances in which it conflicts with the options available in 40 CPR Part 75. The commenter provided an example: a coal-fired EGU with fuel oil startup could elect (per 40 CFR 75, Appendix F, §3.3.6.5) to use the worst case F-factor for any operating hour, regardless of whether fuel oil, coal, or a combination of these fuels were being combusted in the clock hour. the commenter asked if the preceding option would still be available if the EGU were firing fuel(s) with an available F-factor in Table 19-2 of Reference Method 19. The commenter suggested to eliminate this possible inconsistency, this section should simply state that the F-factor must be consistent with Table 19-2 of Reference Method 19 or 40 CFR Part 75, Appendix F, Sections 3.3.5 or 3.3.6.
Response 1:  Upon review of the comment, the Agency agrees that clarification could be helpful. The rule at section 6.2.2.2 of Appendix C now allows F-factors obtained either from Table 19-2 of Method 19 or from Part 75, Appendix F, section 3.3.5 (or 3.3.6) to be used in the emissions calculations. 
12.7 Recordkeeping and Reporting
Comment 1: Commenter 20612 said the quarterly MATS emission and QA file submittal deadline should be extended to 60 days. The commenter noted the 60-day reporting requirement is critical to provide adequate time for PM and HCl stack test data. Unlike Part 75-related stack testing (where sources generally rely on instrumental reference methods), the 60-day period is necessary to provide sources adequate time to get PM and HCl test results back from the labs and review the results. However, in addition to sources that demonstrate compliance based on quarterly stack testing, those that use PM and HCl CEMS must also rely on periodic stack tests (RATAs, PS-11 correlation tests, RCAs and RRAs) using the same non-instrumental reference methods for quality assurance. Under the proposed rule, any CEMS data would need to be reported in the quarterly Emissions reports within 30 days after the end of the quarter. The associated QA test data (including data for RATAs, PS-11 correlation tests, RCAs and RRAs) must be reported in the QA files prior to submitting the Emissions report to validate the emissions data and, naturally, the results of those QA tests can affect the reported emissions values. Thus, there is a contradiction in the proposed reporting requirements since EPA proposed to allow 60 days for sources to report stack test related data in Appendix E, but the proposal only allows sources 30 days to report similar (and in some cases the same) stack test related data in other provisions. 
Commenter 20612 stated EPA should revise the MATS Rule to allow sources to submit the MATS related Emissions and QA file records within 60 days after the end of the quarter to allow "sufficient time to receive the results of tests performed at or near the end of the quarter" consistent with what it allows for other reports that rely on the same type of stack test results. The commenter noted that the Agency will need to make some modifications to the current ECMPS submission process to address this change since sources will still have a 30-day reporting deadline for reporting ARP and CSAPR related Emissions and QA File data under Part 75, but a 60-day MATS reporting deadline is necessary to ensure that sources will have time to both receive and appropriately review the results before submittal. The commenter suggested that EPA could change the reporting deadline in Part 75 to be consistent with the need for a 60-day deadline for MATS. Specifically, the simplest approach would likely be to have sources submit the Part 75 data within the 30-day window, but just leave the ECMPS "submission window" open for another 30 days to allow sources to overwrite the quarterly Emissions file with the MATS data. Alternatively, the commenter recommended sources could be required to submit provisionally quality assured MATS emissions data with the Part 75 data during initial 30 days after the end of the quarter pending reporting of the final RATA or PM CEMS QA test results for HCl or PM CEMS. The commenter suggested another option that may be more difficult to implement: having sources that cannot submit a "full" combined Part 75 and MATS Emissions file within the first 30 days to report an initial file with just the Part 75 related records and then import an additional MATS-only records file (i.e., additional MATS emissions records could be reported in a separate MATS Emission file separate from the standard mission file already submitted). The commenter said it would be burdensome if a source might have to avoid conducting a PM or HCl test in the later portion of the quarter to avoid potential reporting problems because the rule allows such QA tests to be performed at any time during the quarter. Also, conducting a QA test at the beginning of quarter is sometimes impractical or even impossible given the operating constraints of the unit. MATS sources need a 60-day reporting deadline for reporting MATS Emission and QA Files, just as for stack test reports, the proposed Compliance Reports, and the proposed supplemental stack test XML files. An automated approach for making such submissions is also needed. 
Commenter 46348 stated that it agrees with the above comments provided by Commenter 20612. In addition, Commenter 46348 stated it has discussed the above issues with multiple stack testing firms that it has worked with and they all agree that the 30-day timeframe is not adequate to provide the required data and would be burdensome and costly to schedule all tests for sources early on in the quarter. Even at 60 days to report (currently allowed) we have experienced difficulty in providing timely submittals that have had adequate review and authorization signatures. [EPA-HQ-OAR-2009-0234-20608]
Response 1: See response 4 to section 12.5.   
Comment 2: Commenter 20609 opposed EPA's proposal to expand the information that must be reported for all tests involving reference methods, including RRAs and RCAs for PM CEMS. The commenter added that even if EPA finalizes some or all of the proposed requirements, EPA must reconsider the timing of some of the required submissions. Under the current rule, RATA, RRA, and RCA test reports are not required to be submitted until 60 days after completion of the test, and EPA proposes to revise that to require submission no later than with the "relevant quarterly emissions report." The commenter said if EPA finalizes that requirement, the amount of time between completion of the tests and the electronic reporting requirement would vary based on when the test was conducted. For a test conducted early in a calendar quarter, the revised deadline would provide more time than the current rule. The commenter did not object to that and agreed that EGUs should be allowed to submit the information earlier; however, for tests conducted late in the quarter, the rule could require reporting of data as little as 30 days after completion of the test. The commenter said thirty days may not be adequate for RRAs and RCA conducted on PM CEMS. RRAs and RCAs are much more involved than a typical RATA. The commenter stated requiring completion of that in 30 days is not reasonable, and EPA has provided no rationale for changing the deadline. To the contrary, with respect to performance tests, which also require the processing of laboratory results, EPA acknowledges that 60 days (not 30 days) is a reasonable time period. See 81 Fed. Reg. at 67,065. The commenter suggested that rather than require submission of RRAs and RCA results with the quarterly emissions report, EPA should require their inclusion in the quarterly compliance report, which does not have to be reported until 60 days after the end of the quarter.
Response 2:  See response 4 to section 12.5.   
Comment 3: Commenter 20609 stated no objection to the deadline for newly certified CEMS. However, disagreed that 21 day prior to the start of electronic reporting is an appropriate deadline for EGUs that already are using PM CEMS. The purpose of advance submission of monitoring plan information is to ensure that the information is available to evaluate new PM CEMS certification information when it is initially collected or submitted to ECMPS. That rationale does not apply to previously certified PM CEMS. 
Commenter 20609 did not understand what EPA means by "21 days before reporting of PM emissions data via ECMPS is required to begin." The commenter asked if EPA meant 21 days before January 1, 2018, 21 days before the end of the first calendar quarter, or 21 days before the deadline for submission of the first quarterly emissions report of 2018, which would be 21 days before April 30, 2018. The commenter said none of those deadlines makes sense and EPA has provided no rationale for any of them. It should be enough that the electronic monitoring plan information for a previously certified CEMS is submitted concurrent with submission of the first quarterly emissions report under Appendix C so that it is available for evaluation in ECMPS. Requiring submission before that date also would accelerate the need for EPA and EGUs to be fully prepared to make the new ECMPS submissions, something that already is a concern to UARG. 
Commenter 020609 stated the proposed rule appears inconsistent with respect to submission of the hard copy portion of the monitoring plan. The commenter noted that Appendix C sections 7.1.2.2 and 7.2.3.3 require retention of hard copy monitoring plan records, including schematics and blue prints, but do not require submission using ECMPS or any other method, while proposed section 7.2.3.1 requires "submission" of both the hard copy and electronic portions. (This language mirrors language in Appendix A, which also appears to be conflicting). The commenter stated that EPA does not intend to require submission of the hard copy portion of the plan in any format, because it includes material that may be of irregular size and difficult to scan or print. The commenter suggested that EPA should resolve the inconsistency between the provisions in proposed Appendix C (as well as in existing Appendix A section 7.1.1.2.2 and Appendix B section 11.3.1) by removing reference to "submission" of the hard copy portion of the plan.
Response 3: The Agency considered the commenter's suggestions and made revisions to the rule as described below. On the one hand, for EGUs that are already continuously monitoring PM emissions, the rule incorporates the commenter's suggestion to allow the initial electronic monitoring plan to be submitted at any time prior to or concurrent with the first required quarterly PM emissions report, rather than 21 days prior to the date on which emissions reporting is required to begin (see section 7.2.3.1 of Appendix C).  ECMPS can evaluate the first PM emissions report provided that the monitoring plan is in place when the report is submitted.  For consistency, parallel changes have been made to section 7.2.3.1 of Appendix A, section 11.3.1 of Appendix B, and section 3.2.3.1 of Appendix D for EGUs that are already monitoring Hg, HCl, or HF emissions or using the PM CPMS compliance option.  
On the other hand, for new EGUs, or for EGUs that switch to continuous monitoring after having previously implemented another MATS compliance option, the rule requires the initial monitoring plan to be submitted at least 21 days prior to the start of initial certification testing (or, for PM CPMS, 21 days prior to the PM test that establishes the initial operating limit).  This provides the regulatory agencies with an opportunity to review the information and to arrange to observe the tests.  
Regarding submission of hard copy monitoring plan information, the Agency's intent has always been for the hard copy portion of the monitoring plan to be submitted to the State and EPA Regional offices.  However, the Agency acknowledges that this is not clear because, as noted by the commenter, the reporting sections of proposed Appendix C and existing Appendices A and B are not internally consistent.  For instance, section 7.2.3.1 of Appendix C appears to require the hardcopy information to be submitted along with the electronic monitoring plan, whereas section 7.2.3.3 appears to require only that records of the hardcopy information be kept---and there is parallel language in Appendices A and B.   In view of this, statements have been added to sections 7.2.3.3 of Appendix A, 11.3.3 of Appendix B, and 7.2.3.3 of Appendix C of the rule, clarifying that the hard copy portion of the monitoring plan must be submitted to the appropriate State and EPA Region, and the statements in those sections that required hard copy monitoring plan information to be kept on file have been removed from the rule, as they are duplicative of the recordkeeping requirements in sections 7.1.1.2.2 of Appendix A, 10.1.1.2.2 of Appendix B, and 7.1.1.2.2 of Appendix C.  
Comment 4: Commenter 20609 summarized that EPA proposes to include in new Appendices D and E a requirement for submission of a "compliance certification" that largely mirrors language in existing Appendices A and B. Although those provisions have been in place for several years, and are being implemented, that implementation has presented some challenges as a result of the rule's failure to adequately recognize the practicalities associated with the separate acts of "certifying" and "submitting." The commenter stated that owners and operators and responsible officials should be able to designate an "agent" to make submissions on his/her behalf and that this authorization should be reflected in the rule. The commenter noted that under the ARP and CSAPR EPA developed rules to allow those who are otherwise required to "submit" reports to authorize "agents" to perform that task on their behalf. See 40 C.F.R. § 72.26. Under those rules, a submitter is legally bound by the actions of his/her agent. The commenter provided additional background details and examples supporting why this terminology is necessary. With the transition to ECMPS reporting, the commenter said that authorization should be reflected in the rule. The commenter suggested this could easily be accomplished with a minor revision to the current and proposed compliance certification language to add the words "or his or her agent" or "or your agent" after the reference to "owner or operator." Although the commenter would that change, they are not seeking, and would object to, any other revisions to the compliance certification language.
Response 4: The Agency agrees with the commenter's suggestion. Statements allowing an authorized agent of the owner or operator to submit the required reports and notifications have been added to section 7.2.5.5 of Appendix A, section 11.5.5 of Appendix B, Section 7.2.5.5 of Appendix C, section 3.2.6 of Appendix D, and section 1.0 of Appendix E of the rule.
Comment 5: Commenter 20606 stated Appendix C Sections 7.1.3.3.2 and 7.1.3.4 appear to require the recording of stack temperature and pressure values if such values are used to convert the PM CEMS measurement from units of mg/acm to mg/wscm. The commenter suggested to eliminate this duplicity, Section 7.1.3.3.2 should only reference the correction in Equation C-2 (as the correction associated with Equation C-1 is already addressed in Section 7.1.3.4).
Commenter 20606 stated Appendix C Section 7.1.7.3 relates to a code for the formula used to convert PM concentration into units of the applicable standard and includes an incorrect internal reference. The commenter said specifically, in the parenthetic language in this section, the reference to Equation C-2 should be replaced with a reference to Equation C-3. 
Commenter 20606 stated that in relation to the electronic quarterly report submitted in XML format, Appendix C Section 7.2.5.3.3 states the report must include all applicable information as specified in the recordkeeping requirements of Section 7.1.2 through 7.1.7 of proposed Appendix E. The preceding data elements include PM concentrations in native measurement units if the measurement basis is other than mg/wscm (see 7.1.3.3.2 and 7.1.3.3.3), as well as PM CEMS measurement temperatures and pressures (see 7.1.3.3.2) and stack temperatures and pressures (see 7.1.3.4), as applicable. Again, while it makes sense to require EGUs to maintain records of the preceding information, the EPA offers no explanation for why submittal of such information is required for each and every unit operating hour. As noted previously, the reporting of such information is inconsistent with EPA' s collection of other data which is reported at specified reference conditions. 
--------------------------------------------------------------------------------
Response 5: Upon consideration of the comments, the Agency revised the rule so that PM CEMS that measure concentration at actual conditions (e.g., mg/acm) need only record, but not report, the temperatures and pressures used to convert the measured hourly PM concentrations to standard conditions (in Equation C-1 or C-2).
--------------------------------------------------------------------------------
Moreover, the incorrect reference in section 7.1.7.3 of Appendix C to Equation C-2 of the rule has been replaced with a reference to Equation C-3.
--------------------------------------------------------------------------------
Regarding the remainder of comments, please see Response 1 of section 5.3(c).
13 Appendix D-PM CPMS Monitoring
13.3 Recordkeeping and Reporting
Comment 1: Commenter 20609 stated the proposed deadline for submission of the electronic monitoring plan in section 3.2.3.1 has the same flaw as Appendix C section 7.1.1.2. Specifically, the requirement to submit existing monitoring plans 21 days before the Administrator specifies reporting of hourly PM CPMS data to ECMPS is both unclear and unreasonable. The commenter recommended EPA should revise the provision to require reporting of existing electronic monitoring plans with submission of the first quarterly emissions report.
Response 1: As mentioned above, the Agency considered the commenter's suggestions and made revisions to the rule as described below. For EGUs that are already using the PM CPMS compliance option, the rule incorporates the commenter's suggestion to allow the initial electronic monitoring plan to be submitted at any time prior to or concurrent with the first required quarterly report, rather than 21 days prior to the date on which reporting is required to begin (see section 3.2.3.1 of Appendix D).  ECMPS can evaluate the first quarterly report provided that the monitoring plan is in place when the report is submitted.  For EGUs switching to PM CPMS after having previously implemented another MATS compliance option, the rule requires the initial monitoring plan to be submitted at least 21 days prior to the particulate matter test that establishes the initial operating limit for the CPMS. This provides the regulatory agencies with an opportunity to review the information and to arrange to observe the tests.  
Comment 2: Commenter 20609 stated EPA did not the explain the purpose of the exception to the "excess emissions" reporting requirement for EGUs that elect to demonstrate compliance using a PM CPMS. The commenter said it is not clear why EPA believes it necessary to address future "excess emissions" reporting both in § 63.10031(d)(2) and in Appendix E. The commenter agreed that PM CPMS operating limit exceedances would not be "excess emissions" because some exceedances are "permitted" under § 63.10021(c)(2). However, because the rule already defines the exceedance of a PM CPMS operating limit as a "deviation," see, e.g., § 63.10021(b), the commenter assumed that such exceedances would be reported as a deviation under section 12.1.2 of new Appendix E. As defined in § 64.10042, "deviations" are not necessarily violations. Requiring continued reporting under § 63.10(e)(3) interjects significant duplication with that requirement and with the new XML monitor downtime reporting requirements for PM CPMS in proposed Appendix D.
Response 2: Upon consideration of the comment, the Agency has removed from the rule the proposal that would have required EGUs that elect to demonstrate compliance with PM CPMS to continue submitting excess emission and monitoring system performance reports under §63.10(e)(3)(v) and (vi), even after completion of the transition to a single data system.   As mentioned earlier, the requirement in Appendix D to report the hourly PM CPMS response values and PMA, taken together with the requirements to report the 30- boiler operating day rolling averages and deviations from the operating limit in the quarterly compliance reports provides sufficient information to assess compliance. 
14 Appendix E-Data Elements
Comment 1: Commenter 20606 stated the EPA provided context for the data elements required to be reported in Sections 1 through 13 via the development of detailed draft reporting instructions (i.e., Instructions for MATS Compliance Reports (June 2016)). The commenter disagreed that there was sufficient reporting instructions for the data elements the EPA is proposing to collect in Sections 17 through 21. The commenter said there was not enough information to provide meaningful comments on those aspects of Appendix E. The commenter said there is not a clear explanation of what each data element is supposed to represent. To the extent that the EPA does not intend to abandon the collection of detailed reference method data as reflected in proposed Appendix E, Sections 17 through 21, the commenter suggested the EPA should augment the regulatory docket with detailed reporting instructions for the associated data elements and provide opportunity for public review and comment.
Commenter 20612 indicated that the proposed revisions of Appendix E include significant amounts of duplicative and unnecessary reporting requirements that should be removed. The burden is particularly true with respect to RATA, PS-11, RCA and RRA test, which the proposed rule would require to be reported both within the QA files as well as the new XML stack test data files (with sources reporting some of the same data in both places). Availability values are also required in the quarterly compliance report even though values are reported hourly in the emissions file. This commenter believes that these duplicative reporting requirements should be removed. Since EPA will be collecting all this information via ECMPS, the data does not need to be reported in more than one file for the Agency to populate whatever databases may use this information.  
Commenter 20594 said Appendix E is deficient because the data elements identified to be reported in ECMPS are not adequately defined, which will lead to inconsistent interpretations and reporting outcomes among sources. The commenter said there is lack of detail in this list. Many terms have multiple interpretations since they are not defined. For example, Section 10.4 requires identification of the "last burner inspection," which is listed after 10.3 that is entitled "Date of Last Tune-up." It is not clear if the inspection entry is only in reference to burner inspections performed as part of MATS tune-ups. EPA has provided "Instructions for MATS Compliance," issued in June 2016 (the Instructions). These Instructions provide more details concerning how to report the data elements and essentially shape how sources will report information for MATS. The Instructions should be incorporated into the Proposed Rule. The Instructions provide sources information on how to report data elements, which cannot be extrapolated from Appendix E. Even more troubling, EPA is not bound by the Instructions. In other words, the Instructions can change at EPA's whim. The federal rulemaking process is in place to prevent agencies such as EPA from implementing substantive requirements that are a moving target. 
Commenter 20594 stated that proposed Appendix E is deficient because the data elements identified to be reported in ECMPS are not adequately defined, which will lead to inconsistent interpretations and reporting outcomes among sources. 
Response 1: The Agency does not agree with the commenters' concerns over the performance test information to be reported. That information was required by the original rule to be reported via CEDRI using the Electronic Reporting Tool (ERT). As explained elsewhere, in order to merge the separate electronic reporting systems into the ECMPS Client Tool, a task that was undertaken at industry's request, the information required to be reported via the ERT now needs to submit via the ECMPS Client Tool. While the ERT can be used for this purpose, industry representatives sought an alternative submission approach using EXMPS to submit the XML schema. The proposed rule provided last year included a methodology, but commenters then sought additional identification of the data elements in a separate appendix, which was provided in the proposal as Appendix E. In order to further assist commenters, the Agency has reviewed the data elements in Appendix E, and this response to comment document, as well as the rule in Appendix E, now includes additional clarifying information to eliminate the inconsistent interpretations and confusion.
The Agency has not taken the commenter's suggestion to incorporate the Quarterly Compliance Report Instructions into the rule.  However, the applicable rule citations from sections 2-13 of Appendix E have been entered as footnotes into each section of the Instructions.  
Reporting Instructions have been developed for the MATS Reference Method Test Data from sections 17-30.
Regarding the example cited by a commenter, the Agency has withdrawn from the proposed rule the requirement in section 10 of Appendix E for affected EGUs to report the date of the last burner inspection.  The Agency agrees that §63.10021(e)(9) and §63.10031(c)(4) require only the date of the tune-up to be reported.  According to§63.10031(c)(4)  , the tune up date is the completion date of combustion optimization and measurement of the effluent CO, NOx and O2 concentrations [see§63.10021(e)(6) and (7)]. 
Comment 2: Commenter 20595 stated a note is provided in § 63.10031(g), for all cases in which the EPA reference methods supported by the ERT are used, the EGU owner or operator would be required to provide the data elements specific to the test method(s) used, in XML format, as an attachment to the compliance report. A review of the data elements in Appendix E indicates that several do not support compliance demonstrations. An example is the identification of the plant contact for stack testing, which can change test to test and is better to be identified as the responsible official. Should questions regarding any submittal of data elements, contacting the responsible official is proper as he(she) may then answer the concern or delegate the response to one of his(her) employees or contractors.
Response 2: The Agency has reviewed the list of data elements and a few data elements have been removed. The remaining data elements are standard and necessary information in a stack test report. Regarding the example offered by the commenter, identification of the plant contact is already required by ASTM D7036-15A, the standard to which part 75 EGUs must already adhere (all stack tests performed for the acid rain program must be conducted by Air Emission Testing Bodies, which must adhere to ASTM D7036-15A, per requirements promulgated March 10, 2011 at 76 FR 17288). These and non-part 75 EGUs are required to identify the plant contact per the NESHAP general provisions at §63.7(g)(2); moreover, this information is included as an item to be reported in Guideline Document 043. The Agency finds that the information supports compliance demonstrations and is necessary for communication purposes should questions arise regarding any submittal of data elements. Delegated authorities need this information in case there are questions about stack test reports. In order to avoid any potential confusion on the part of the commenter, the rule now clarifies in Appendix E the Facility Point of Contact may be an owner, operator, or responsible official. Note that the EGU owner or operator is to determine who the best contact is and how that contact will be available for communicating with the delegated authority.
Comment 3: Commenter 20609 stated EPA has provided as part of the current rulemaking docket Draft Compliance Report Instructions that covers certain of the Appendix E data elements in the quarterly compliance reports described in the proposed rule. The commenter stated that ECMPS, even though it is a complicated reporting program, can be effectively implemented if stakeholders participate in the development. The commenter was concerned with the potential of the ECMPS format to impose substantive requirements that have not been adopted through rulemaking. One way this could occur is if the format specifies data elements that are not required to be reported under the applicable rules. This also could occur if EPA attempts to use reporting "instructions" to add details or gloss to rule requirements that otherwise are ambiguous or not clear. The commenter said that EPA's reporting instructions and schema do not include citations to the applicable rules, which makes it difficult for stakeholders to understand the reported element in context. The commenter said this issue has been raised many times in the past regarding Part 75 and other MATS ECMPS issues, and the Agency has been asked to provide documentation to confirm the consistency of its ECMPS reporting format with the substance of the rules. The commenter said although the Agency did at one point in the past conduct a comparison of the ECMPS format to the Part 75 rules, and provide the commenter a summary of its findings (including a list of ECMPS data elements that were not supported by any rule requirement), EPA has never provided the commenter with the full results of that comparison or revised its instructions and schema to incorporate citations to the applicable rules. The commenter said stakeholders cannot confirm from examination of the draft documentation that the proposed data elements discussed in the instructions and schema are limited to the information required to be reported under MATS. To the extent EPA believes that it has authority under the MATS rule to require reporting of the data elements specified in the draft instructions and schema, it is the Agency's responsibility to identify that authority. The commenter said stakeholders cannot fully evaluate the appropriateness of the format and instructions without this information. The commenter requested that the Agency either revise the draft instructions to identify a rule citation for each specified data element discussed in those instructions, or provide other documentation that clearly and transparently achieves the same result. Under other EPA electronic reporting programs, the Agency has addressed similar concerns by providing the information in Table format. Because the Agency presumably relied on the proposed MATS rule provisions to support the data elements in the draft instructions and schema, providing documentation of that analysis should not impose any significant additional burden on the Agency.
Response 3:  The Agency has incorporated the commenter's recommendation to revise the Quarterly Compliance Report Instructions by identifying the applicable rule citations from sections 2-13 of Appendix E for the data elements in the XML schemas.  The rule citations have been entered as footnotes into each section of the Instructions.   
14.1 Quarterly compliance reports
Comment 1: Commenter 20612 supported the change from semi-annual to quarterly reporting and maintains that the 60 day reporting deadline must be maintained to provide sufficient time for testers and sources to complete analysis and reporting.
Response 1:  The final rule retains the 60-day reporting deadline for the quarterly compliance reports, as proposed.
14.1.1 Data elements in sections 2-13
Comment 1: Commenter 20612 stated that the proposed XML stack test data records include far more than is needed to recalculate the test results.  The commenter stated that the ERT includes a lot of information that is not needed if the objective is simply to recalculate the results and that could be better addressed within the PDF stack test report.
Commenter 20612 stated that their understanding was that the principle reason EPA is proposing to require sources to report point-by-point reference method test data in numeric format is to allow states to recalculate the results. The commenter stated that it sees little benefit to a state or local agency recalculating the results of every test and believes that perceived benefits of the ERT outweigh its costs. But, even if the objective of collecting the XML stack test data is to allow the results to be recalculated, there are still numerous data points in Appendix E that are not needed for this task. 
Commenter 20612 stated that many of the data elements required in Appendix E are unnecessary because the information has already been provided in other MATS reports or the data may be obtained from the PDF version of the stack test report that is attached to the data submittal and other data elements are too vague and require additional clarification.
Commenter 20612 stated that while EPA has provided draft reporting instructions in the docket for the new quarterly MATS compliance reports, the Agency has not provided similar draft instructions for the supplemental test data reports in Appendix E.
Commenter 20612 provided examples and additional background details in conjunction with the above statements.
Response 1: As the Agency has mentioned before, the main objective in collecting emissions test data is to have the data available in electronic format to demonstrate compliance with emissions testing requirements, not to just recalculate test results, although that is an appropriate use of the data. While the ERT does include optional information, those portions are clearly marked as `optional.' Even so, the ERT is a tool that an EGU owner or operator could use to develop a complete test report and, as required by MATS, an XML file. The MATS preamble identifies two ways that the XML file can be generated to meet the reference method data submission requirements: either via the existing ERT or via another program that provides the data in an appropriate XML file format (using the same xml tags that are in the ERT xml.)  

As mentioned earlier, third party software developers and programmers have had the full XML schema available since September 2016, when it was posted on the ERT website and was included in the docket (EPA-HQ-OAR-2009-0234-20586). To the extent that a third party program or system that can create the XML containing all the data tags and data necessary to complete all calculations in a stack report and listed in Appendix E, MATS allows such a program or system to be used to submit data. In addition to compliance determination, the data elements listed in Appendix E are useful for quality assurance and emission factor development. The User manual of the ERT gives instructions on entering data and creating the XML schema; that manual is available on the ERT webpage and is included in the docket. 
Comment 2: Commenter 20609 stated that EPA's  proposal in Appendix E sections 10.3 and 10.4 to require reporting of the both the "Date of Last Tune-up" and "Date of Last Burner Inspection." is in direct conflict with § 63.10021(e)(9) and § 63.10031(c)(4), both of which explicitly limit reporting with respect to tune-ups to the date of the optimization and testing in § 63.10021(e)(6) and (7). The commenter stated that If EPA wants EGUs to also report the date of the last "burner inspection," EPA must revise the other provisions to make the rule consistent. EPA also must explain what it means by "burner inspection" as that relates to the activities described in § 60.0021(e) and identify what date should be reported when the specified activities take place over several days.
Commenter 20609 referenced the draft reporting instructions posted to the docket, [Draft Compliance Report Instructions § 5.0 (June 2016), EPA-HQ-OAR-2009-0234-20589] and stated that if the additional date EPA proposes be reported is the date of completion of some activity, EPA must say that in the rule as EPA cannot use reporting instructions to establish binding requirements.
Response 2:  Upon consideration of the comment, the Agency agrees that §63.10021(e)(9) and §63.10031(c)(4) require only the date of the tune-up to be reported. The rule has been revised to remove this proposed requirement Note that according to §63.10031(c)(4), the tune up date is the completion date of combustion optimization and measurement of the effluent CO, NOx and O2 concentrations [see §63.10021(e)(6) and (7)]. 
Comment 3: Commenter 20609 stated that proposed Appendix E Sections 3.0 and 4.0 correspond roughly to the summary information currently submitted in a semi-annual report and that it does not object to these requirements, but the commenter requests that EPA include a better description in the rule for "Timing of Test" (section 3.6). The commenter stated that it understands from section 3.0 of the Draft Compliance Report Instructions that EPA intends this to indicate whether the test was performed "late," (e.g., performed after the test deadline pursuant to § 63.10005(f)(3)), but that explanation needs to be in the rule as well.
Commenter 20609 stated that proposed Appendix E Sections 5.0, 6.0, and 7.0 correspond roughly to the type of information currently required in a performance test report under § 63.7(g), which requires reporting of the "results," including "the analysis of samples, determination of emissions, and raw data" and that also does not object to these requirements, which are not unlike the information currently reported under Part 75 for RATAs. 
Commenter 20609 stated that it objects to much of this additional reporting EPA has proposed to require in the quarterly compliance report of significant additional information for each run of a reference method used in a performance stack test, including LEE tests. The commenter stated that this would require submission of the same reference method information for each RATA, RRA, and RCA.
Response 3: The Agency appreciates the commenter's suggestions. The final rule clarifies that the "timing of test" data element in section 3.6 of Appendix E indicates whether the test was done on schedule according to §63.10006(f), or was late. 
The Agency acknowledges no objections to the proposed Appendix E Sections 5.0, 6.0, and 7.0 since, as suggested by the commenter, the sections correspond roughly to the type of information currently required in a performance test report under §63.7(g); however, the Agency notes that the objection by the commenter over information to be reported is at odds with the requirements of §63.7(g)(2)(v), which specify that where a test method requires one to record or report information, that information shall be included in the test report. Consistent with the earlier discussion, the data elements viewed by the commenter as "additional" are specific to each run and need to be provided electronically so the calculations easily reviewed. Reporting in this fashion makes the electronic data available to the delegated authority and the public, providing data transparency for compliance, quality assurance, and emission factor development via the complete data set.
Comment 4: Commenter 20609 stated that it objects to the proposed new requirement in Appendix E section 10.5 to report the first and last use of each fuel during the month because for EGUs that burn different types of fuels, this would impose significant additional burden by requiring manual review of daily fuel use records to determine start and end dates during the month. On the other hand, EGUs that primarily combust the same fuel may not have a begin or end date during the month, and some EGUs might have overlapping periods (e.g., during co-firing of coal and gas).
Commenter 20609 also stated that it cannot fully evaluate the above proposed new requirements, because EPA has not provided a reason for them, or explained how EPA would use the additional information. EGUs are not limited in the type of fuel they combust as long as it is not classified as a solid waste. As a result, the information is not needed to demonstrate compliance. To the extent a new fuel is combusted, or use of a fuel is permanently discontinued, that information already is reported in the Part 75 electronic monitoring plan, but using different Unit Fuel Codes and Descriptions [40 C.F.R. § 75.53(e)(1)(D); ECMPS Reporting Instructions  -  Monitoring Plan at section 4.3 Fuel Use Data (Sept. 14, 2016)]. The commenter stated that rather than add new duplicative requirements, EPA should take advantage of the consolidation under ECMPS and rely on those existing records to determine new fuel use. Any other use EPA may have for tracking fuel use can be satisfied by the current requirements.
Commenter 20606 disagreed with the requirement to enter in a fuel type "start and end date" in the quarterly report. Although not entirely clear in proposed Appendix E, when examined in conjunction with draft Instructions for MATS Compliance Reports (June 2016), it becomes apparent that the EPA expects both a begin date and end date to be reported for every fuel fired within a given calendar month. Further, it appears as though this requirement applies to clean fuels in addition to those fuels which are directly targeted by MATS. The commenter said this an example of the reporting tool overreaching what the regulation requires. As an example, if an existing coal fired units fired 100% sub-bituminous coal, blended mixtures of sub-bituminous coal and 100% bituminous coal, as well as fuel oil for startup in a given month, one would need to manually parse daily fuel usage records to determine the first and last instance in the month when each fuel/fuel blend was fired. The preceding would be time consuming for units with multiple fuel types/blends, especially across a fleet of MATS affected units. The commenter suggested in the alternative that the fuel type(s) for a given unit be treated as a monitoring plan record, with the "active" fuels consisting of those with a begin date but no end date. The monthly fuel usage records could then be used to assess when each "active" fuel was being fired in a given unit. These monitoring plan style records could also be used to non-waste fuel type information that the EPA is proposing to collect as part of each quarterly compliance submittal. Further, the EPA already has a Fuel Type record as part of the 40 CFR Part 75 Monitoring Plan data elements that could likely be adopted for this approach. The requirement in §63 .10032( d)(l) is to document monthly usage by type; therefore we object to section 10.5 .1 and 10.5 .2 in Appendix E which requires identification of "start date" and "end date" to be included in the quarterly report.
Response 4: The Agency finds the commenters' concerns somewhat misplaced, as the rule at §63.10020(e)(1)(i), (iv), and (v) and (e)(2)(i), (iv), and (v), already requires owners or operators of EGUs that choose to use paragraph (2) of the definition of startup to provide the date and time that clean fuel use begins and ends and that non-clean fuel use begins. The rule requires reporting of start and end dates and times of clean fuel use and start of non-clean fuel use for EGUs that choose to use paragraph (2) of the definition of startup.  This information must be reported separately, as a PDF submittal that accompanies the quarterly compliance report, not as part of the compliance report (see §63.10031(i)).  Therefore, the proposed requirement to report the start and end date for the combustion of each fuel type during each month has been removed from the quarterly Compliance Report Instructions document.  Owners or operators of EGUs that choose to use paragraph (1) of the definition of startup are only required to provide monthly usage by fuel type in the quarterly compliance report. 
Comment 5: Commenter 20609 stated that EPA explain the purpose of the proposed exception to the "excess emissions" reporting requirement for EGUs that elect to demonstrate compliance using a PM CPMS [Proposed § 6310031(d)(3) and (k)]. The commenter stated that it assumes it is based on a concern that PM CPMS operating limit exceedances are not "excess emissions" (or deviations). The rule defines "excess emissions" as the: "results of any required measurements outside the applicable range (e.g., emissions limitations, parametric operating limits) that is permitted by this subpart." [40 C.F.R. § 63.10042]. 
Commenter 20609 stated that it agrees that PM CPMS operating limit exceedances would not be "excess emissions" because some exceedances are "permitted" under § 63.10021(c)(2). However, because the rule already defines the exceedance of a PM CPMS operating limit as a "deviation," see, e.g., § 63.10021(b), the commenter stated that it assumes that such exceedances would be reported as a deviation under section 12.1.2 of new Appendix E. As defined in § 64.10042, "deviations" are not necessarily violations. Requiring continued reporting under § 63.10(e)(3) interjects significant duplication with that requirement and with the new XML monitor downtime reporting requirements for PM CPMS in proposed Appendix D.
Commenter 20609 stated it is concerned that it may not be possible to identify duration with that level of specificity with respect to EPA's proposal to implement the requirement to report the "duration" of a malfunction by reporting the date and hour in which the malfunction incident began and ended. In some instances, a malfunction may not be immediately apparent and the operator may need to look to other factors to estimate the hour in which the incident began. Similar to the current rule, EPA should implement the requirement more flexibly and require a textual explanation of the malfunction duration including the date and hour when that information is discernible. EPA also proposes to require a description of corrective action taken and the commenter stated it does not object to that requirement.
Commenter 20609 stated that current § 63.10031(g) and proposed § 63.10031(d)(2)(iv) only require reporting of a malfunction if it caused (or may have caused) an exceedance of an emission limitation. Proposed Appendix E § 11.0 requires reporting of all malfunctions. Because section 2.5 of the Draft Compliance Report Instructions includes the qualifier regarding exceedances, this may have been a drafting error and the commenter requests that it be corrected. The commenter stated that it would object to an expansion of the malfunction reporting requirement in Appendix E to include all malfunctions. Reporting malfunctions that do not affect compliance is overly burdensome and EPA has provided no rationale to justify that.
Commenter 20595 stated in Appendix E to subpart UUUUU of 40 CFR part 63, Sections 2 through 13 list the data elements that must be reported in XML format in the quarterly compliance reports that cover the period beginning January 1, 2018, and are required under proposed § 63.10031(g). The information data elements in Section 11 of Appendix related to reporting malfunctions are no longer relevant. The same information would be provided in Section 12 regarding deviations where, in 12.2 the source "A description of the deviation, including the date (or range of dates), the cause (if known), and any corrective actions taken." Whether a deviation is caused by a malfunction or not was changed when the agency removed section 63.10010 earlier in the technical correction and start-up rule revision in April of 2016. AEP suggests that section 12 of Appendix E be eliminated.
Commenter 20606 stated malfunction reporting in this proposed section should be limited to only those malfunctions that caused (or may have caused) an exceedance of an emission limitation, consistent with 40 CFR §63.10031(g). The proposed language in Appendix E, § 11.0 would require the reporting of all malfunctions, and this is not supported by the underlying rule language.
Commenter 20594 urged the EPA to provide clarity in section 11, malfunction information. The commenter said it is unclear whether there will be formatting to allow a source to adequately describe the event, cause and corrective action. The Instructions suggest that formatting will be allowed but Appendix E does not provide this detail. 
Response 5: See response 2 of section 13.3.
Comment 6: Commenter 20609 stated that in section 12.1, EPA proposes to require reporting of the "nature of the deviation" and sets out a list of types, which EPA then proposes to implement in its Draft Compliance Report Instructions using the codes in Table 17. The commenter stated that it believes the list is too limited and prescriptive. Deviations can take many forms. At a minimum, EPA should include an "Other" category.
Commenter 20609 stated that one of the types of deviations listed in section 12.0 is "Monitoring requirement not met." [App. E section 12.1.5]. In section 6.0 of the Draft Compliance Report Instructions, EPA expands upon the rule and instructs that, for periods when an unmonitored bypass stack was used, the EGU should report the "total number of hours of bypass stack usage during the quarter." The commenter stated that it objects to the use of the reporting instructions to establish new requirements. If EPA wants that information to be reported, rather than the "date or range of dates" for the deviation, EPA must put that in the rule.
Commenter 20609 stated that in section 12.2, EPA's instructs EGUs to report "monitor downtime incidents" and percent monitor data availability ("PMA") under the label of "deviation." EPA thus incorrectly implies that all periods of monitor downtime, including periods when data are not collected because of monitoring system malfunctions, repairs, required QA or quality control, as well as periods when a monitoring system is out of control, are deviations from monitoring requirements. They are not.  The rule does not require that monitoring systems operate or collect data during such periods and the rule clearly states that such periods are not "deviations" (see 40 C.F.R. § 63.10020(b) and (d)).
Commenter 20609 stated that although these periods are not reported as "deviations," EGUs currently are required to maintain records of such periods under 40 C.F.R. § 63.8(b)(2)(vii) and § 63.10(c), and to report monitoring system downtime in the semi-annual compliance report under § 63.10(e)(v)-(vi) and § 63.10031(c). EGUs also already keep records of monitoring system repairs and maintenance, identify hours when quality-assured data are not recorded, and report PMA for Hg CMS and HCl/HF in quarterly emission reports under Appendices A and B. (See, e.g., App. A sections 5.4.1.3, 7.1.3.4-5 and 7.1.4.4, 7.1.4.8, and App. B sections 8.1.3, 10.1.3.4-5). EPA has proposed similar requirements for PM CEMS and, where appropriate, for PM CPMS [Proposed App. C sections 5.3.1.3, 7.1.3.5-6 and App. D section 3.1.3.5]. There is no basis for reporting monitor availability or identifying monitor downtime a second time. Moreover, inclusion of such information in a "compliance report" is misleading, because there are no minimum availability requirements in the rule. EPA should remove all references to PMA from Appendix E.
Commenter 20609 stated EPA should revise Appendix E section 12.0 to make clear that the only periods of monitor downtime that would be reported as a "deviation" under 12.1.5 and described under 12.2 are those that do not fall within the above exceptions. If, after the end of the transition period, EPA wants to continue to receive reports on the causes of monitoring system downtime similar to those currently required under § 63.10(e)(v)-(vi) and § 63.10031(c), EPA should address that requirement is a separate section in Appendix E.
Commenter 20606 noted EPA suggests in section 12 that monitoring downtime incidents and percent monitor availability be reported under the label of "deviation." The preceding does not recognize that 40 CFR §63.10020(b) and (d) clearly acknowledge that loss of monitoring data due to monitoring system malfunctions, out-of-control periods, and required quality assurance or quality control activities are not deviations from the monitoring requirements of the rule.
Response 6: The Agency has reviewed the commenters' suggestions and has incorporated the suggestion to add "Other Requirement Not Met" as a type of deviation to be reported under the rule in section 13.2.6 of Appendix E.  
Also, the Agency has withdrawn the proposed rule requirement to report the lowest percent monitor availability (PMA)) and the PMA at the end of the quarter; these values are easily obtained from the quarterly emissions reports. In addition, a note has been added to section 6.0 of the instructions for the quarterly compliance reports, to clarify that periods of monitoring system downtime for  required QA activities and routine maintenance are not to be reported as deviations. 
Comment 7: Commenter 20609 stated that in Appendix E Section 2.7, EPA proposes to require reporting of a "Compliance Indicator." The term "compliance indicator" is not defined anywhere in the rule. The commenter stated that it understands from section 1.0 of EPA's Draft Compliance Report Instruction that EPA intends this to be an indicator of whether any performance stack test result or calculated boiler operating day rolling average shows an exceedance. The commenter stated that it objects to this section as EPA cannot use reporting instructions to establish binding requirements. If EPA wants that information reported, EPA must explain in the rule what it means. EPA also has not established any purpose for the additional record, which does not provide any information that is not already reported in section 12.1.1 and, if EPA does not remove it, § 63.10031(d)(2). To the extent no tests were performed or no average calculated, that also would be obvious from other required records in the report.
Commenter 20606 stated per the draft Instructions for MATS Compliance Reports, a Yes, No, or Not Applicable must be reported for each calendar quarter to indicate whether all data collected during quarter indicates compliance or noncompliance, with a zero reported if no compliance determinations were conducted during the quarter. With the shift from reporting by exception to directly reporting all values needed to directly assess compliance, this record is not needed. 
Commenter 20594 stated that in section 2.7, the term "compliance indicator" is ambiguous. The Instructions create further uncertainty. The Instructions provide, "Report `N' in this field if any stack test results and/or 30- (or 90-) day rolling average(s) show noncompliance." See Instructions at 6. If a source has a deviation, in particular one that may be caused by a malfunction, it is not clear whether this is an event of noncompliance that would necessitate an "N" entry. 
Response 7:  The Agency agrees with the commenters that the term "compliance indicator" is unclear and that in some cases could be redundant. Therefore, that data field has been removed from both Appendix E of the rule and the quarterly compliance report instructions.  
Comment 8: Commenter 20606 stated in cases where stack tests or LEE tests are used for compliance purposes, the draft Compliance Report Instructions require reporting a code (I or 0) to indicate whether the test was conducted in a timely manner. The commenter disagreed that this information needs to be reported, as there should be sufficient information reported via ECMPS for a determination of whether such tests were conducted in a timely manner. This is fundamentally no different from ECMPS assessing whether periodic tests such a linearity tests, RATAs, etc. are conducted in a timely manner.
Response 8: EPA disagrees with the commenter's assessment. The quarterly, annual, and triennial performance stack tests must be conducted within certain date ranges, as specified in §63.10006(f). As the ECMPS Client Tool will not be programmed to check compliance with those requirements, its software will perform only basic checks of the information in the quarterly compliance reports, e.g., for completeness, range, and formatting. As mentioned earlier, compliance assessment will be left to those delegated authorities reviewing the reports. Therefore, to assist the reviewers, the "timing of test" data field has been retained in section 3.6 of Appendix E of the rule and in the quarterly compliance report instructions.  
Comment 9: Commenter 20606 stated in section 10.2 the EPA requires that each quarterly compliance report include unit type information (also see Section 5.0 draft Instructions for MATS Compliance Reports). The commenter said there is nothing within the MATS rule that requires the reporting of unit type. The commenter said this information is duplicative of the information already collected in the EPA's Certificate of Representation Form for EGUs subject to the ARP, CSAPR or MATS Rule. Lastly, it makes no sense to report unit type on a quarterly basis, as this is a fundamental unit characteristic that is not likely to change periodically, if ever.
Commenter 20606 stated that the section 10.5.7 reporting obligation is confusing in the draft Instructions for MATS Compliance Reports. The data element FuelType includes a code of NHNWF (see Table 14) for "Non-hazardous Non-waste Secondary Fuel'', and there is yet another data element NonWasteFuelType which again requires the reporting of a specific fuel type (see Table 16). The reporting of fuel type would be streamlined if the specific NHNWFs listed in Table 16 were simply included directly in Table 14. This two-step reporting process for any NHNWFs is inefficient, especially considering that this information needs to be entered for every single calendar month during which such a fuel is fired.
Response 9:  Upon consideration of the comments, the Agency agrees that clarification is warranted. The proposed requirement to report unit type has been removed from section 10.2 of Appendix E of the rule and from the quarterly compliance report instructions. Also, the Agency has accepted the commenter's suggestion to streamline the two-step process for reporting combustion of non-hazardous, non-waste (NHNW) fuel types.  However, rather than adding the NHNW fuels to Table 14 of the instructions, the NHNWF code has been removed from Table 14 and all of the NHNW fuels have been consolidated in Table 15 of the instructions. The reporting instructions now direct EGU owners or operators to report the type(s) of fuel combusted during the calendar month, using the appropriate code(s) from Table 14 and/or Table 15, as applicable. Revised Table 16 of the instructions now is used to report the units of measure for each type of fuel combusted during the calendar quarter.   

14.2 Performance stack tests/RATA/PM CEMS correlations/RRAs/RCAs
Comment 1: Commenter 20612 indicates that Appendix E Section 16.1 specifies that EPA Reference Method 2 must be used to determine average stack gas velocity if the EGU is complying with a heat input based standard.  This commenter stipulates that the flue gas velocity is also available from the Reference Method and does not require a separate Method 2 test, for example Method 5 and that the language should reflect this.  This commenter also wants EPA to allow certified stack gas flow monitors measurement of gas flow to be allowed to calculate output-based emission values.  
Commenter 20612 also comments that the statement regarding the use of moisture data in Section 16.3 of Appendix E for conversion from wet to dry basis is incorrect and that the moisture value is used instead to determine molecular weight of the gas.
Response 1: The Agency does not understand the commenter's concern, as every isokinetic method (Method 5, 26A, or 29) includes the use of Method 2 for determining average stack gas velocity, (vs), by design. This information is needed for isokinetic calculations. Therefore, as suggested by the commenter, anytime one uses an isokinetic method, Method 2 is performed. The commenter need only transfer the requisite information in order to fulfill this requirement.
Upon examination arising from an alternative test method request, the Agency has already disallowed use of a certified stack monitor instead of performance test data; therefore, the Agency disagrees with the commenter's suggestion. As noted in the disapproval, certified stack monitors do not necessarily produce results adequate for the Administrator's determination of compliance. More specifically, the principal concerns stem from the fact that the quality assurance requirements for the reference test methods are more stringent than the quality assurance requirements for the certified stack gas flow monitors. 

The Agency notes that Methods 2, 2C, 2F, and 2G, require a multi-point traverse be performed during ach test run to determine stack gas velocity and volumetric flowrate. This traverse
provides a more representative measurement of facility emissions than that provided by certified stack gas flow monitors. Additionally, since there are no flow audit
materials parallel to calibration gases, the accuracy of an EGU's flow monitor
is only verified, at most, twice a year during the performance of the relative accuracy test audit, which includes a comparison to a multi-point traverse.  The Agency agrees with the commenter that a clarification of section 16.3 of appendix E of the rule is needed, and the rule is revised to reflect that the moisture value is also used to determine molecular weight of the gas.

14.2.1 Data elements in sections 17-21
Comment 1: Commenters 20597 and 20612 said Sections 14-21 of Appendix E impose a significant amount of redundant reporting, which would force sources to duplicate information that is already reported electronically in the proposed compliance reports and the QA reports as well as in the detailed PDF stack test reports. Having duplicative reporting requirements would seem to be in direct contradiction with the stated intent of the MATS proposal to streamline the electronic reporting. 
Commenter 20597 also suggests that the proposed compliance reports require sources to provide a host of performance test information that can meet the "essential" objectives. Such values could be readily reviewed by agency personnel and can be validated as part of the submission process to ensure that compliance is achieved to verify that the test was properly performed and to estimate the potential uncertainty of the results for emission factors purposes. In short, the information in the compliance report provides all the information needed to meet the Agency's stated objectives and the other performance test data, including the point-by-point run values can be fully documented by electronically reporting complete stack test reports as PDF files. The burden of providing point-by-point stack test data in XML format simply outweighs its benefit.
Commenter 20606 suggested that to help achieve EPA's stated goal of streamlining the MATS e-reporting requirements, EPA should abandon the current effort to effectively require ERT style electronic reporting for RATAs, RCAs, RRAs and stack tests (i.e., proposed Appendix E, Sections 14-21). In the alternative, EPA should retain the current process whereby complete QA and stack test reports, including all of the information required within the proposed Appendix E for such activities, continue to be repotted in PDF format through ECMPS. The preceding, as augmented by the quarterly reporting in proposed Appendix E, Sections 1-13, would result in a substantial reduction in the burden associated with electronic repotting as compared to the February 16, 2012 final rule. Relative to the EPA's stated goal of increasing public transparency, such a format for the QA and test reports is far more transparent than an electronic submission consisting of various XML records. This approach strikes a reasonable balance where EGUs are reporting additional electronic data elements in certain regards (i.e., PM CEMS and CPMS hourly data, continuous compliance data), while also concurrently relieving the electronic reporting burden in regard to the proposed requirements for detailed XML format reference method data for QA tests and stack tests.
Response 1: The Agency finds the commenters' view that information submitted via a PDF file somehow equates to electronic reporting mistaken. As mentioned earlier, according to the Agency's E-Reporting Policy Statement for EPA Regulations , E-reporting is not simply a regulated entity e-mailing an electronic copy of a document (e.g., a PDF file) to the government. Further, a PDF file cannot generate tables and run calculations.
The electronic data are essential to conduct the calculations; such activity is not possible through use of data in a PDF file. When using PDF file submissions, the data would need to be reentered into a spreadsheet or database with the correct formulas to replicate the calculations and to check the values submitted. As mentioned earlier, EGU owners or operators can use the already available ERT which can accept the data and conduct those calculations.  
The Agency notes that development of an XML schema, prepared at industry's request, as well as its use, can reduce - if not eliminate - duplicative reporting, since those data elements unlikely to change can be referenced by separate reports. This information can be streamlined for the delegated authorities (region, tribes, state, or local entities) using a uniform XML format and data tags. 
Utilizing the same submitted XML data file will eliminate transcription errors, increase transparency, and allow the data to be reviewed without requiring cumbersome reentry procedures.  The rule remains unchanged in this regard.
Comment 2: Commenter 20609 has listed a number of the proposed data elements that are unclear, incorrect, or duplicative. For example, the following terms are not defined and are unclear: "state facility ID" (section 17.19), "process parameter" (section 17.23), "action on process material (e.g., burned)" (section 17.30), "project number" (section 17.33), "units of process parameter" (section 17.36), "pump vacuum" (section 19.17), "process run ID" (section 19.18), "nozzle calibration" (section 19.22), "filter/oven exhaust temperature" (section 19.29), and "process parameter run data" (section 21.8). He also indicates that there are terms that may appear clear to the reviewer, but may be unclear when implemented. For example, they ask if the requirements to report "quantity of fuel" (section 17.37) and "type of fuel" (section 17.38) for each test means quantity/type during the entire testing period, which could be days (a value of questionable value) or is it just when runs are being performed (something that may difficult to accurately estimate). Further the proposed rule uses different terms for what are most likely the same values for example "clock time start" and "clock time end" used in section 18.0 and "run start time" and "run end time" in the other sections. 
Commenter 20595 stated there is excess data information required in Appendix E. The ECMPS Client Tool would also be used to attach the XML and PDF files that contain the applicable data elements and other information from sections 17 through 22 of proposed Appendix E (as applicable), which provide details of the reference method(s) used for each test, along with the electronic test results. A review of the reporting instructions shows that there is a lot of information required to be in the record that is not necessary to confirm compliance. Careful consideration of the data needed for compliance demonstration and also repeated entry requirements of data for the compliance report and the stack tests, RATAs etc. should be eliminated where possible. The commenter suggested the agency rethink the requirements in the data elements Appendix E since the QA and certification requirements of the test contractors are already provided through their certifying bodies. Having to report this information for each stack test, and in some cases for each run, does not provide additional support of compliance demonstrations. 
Commenter 20594 stated there are data elements not clearly defined in Appendix E. Specifically, the commenter noted the following topics: 
 §17.28. Control device description. Appendix E should clarify whether this term means the "applicable" control device.
 §17.30. Action on Process Material. Appendix E needs to clarify what this term means. 
 §17.37. Quantity of Fuel. Appendix E needs to clarify what this term means. Since MATS requires that testing occur during maximum operating conditions, reporting maximum fuel consumption is not directly relevant to MATS compliance. 
 §§ 19.28, 19.29. Filter/Oven Temperature and Filter/Oven Exhaust Temperature. The terms "filter/oven" should be separated. Filter temperatures and oven temperatures are different metrics for Method 5. The Instructions do not add any clarification on this point. 
 §§ 19.8, 20.16, 21.8. Process Parameter Run Data. Appendix E needs to clarify what this term means. 
 §21.16. Associated Process Run Rate. Appendix E needs to clarify what this term means. 
Commenter 20606 stated that EPA provided unclear intent and no guidance on several data elements within a number of sections of Appendix E. For example, are the clock start time and clock end time supposed to reflect local time, Eastern Standard Time (EST), etc.? The commenter said the EPA should clarify whether there are any reporting obligations under Section 18.0 in the event that an alternative to Method 4 is used to determine moisture content during a MATS QA test or performance test (i.e., EPA Alternative Method (ALT) 008; EPA ALT-060; EPA ALT-091; case specific approvals to use Method 320 for moisture determination in conjunction with HCl measurements, etc.). Additional specific comments on this section are as follows: 
 Section 17.22 Test Method: This field is unclear if this record is specific to a given performance stack test or QA test. Essentially, nearly any performance stack test or QA test will likely involve the use of multiple Reference Methods (RMs). Is the expectation that only the test method directly associated with the pollutant of interest will be identified, or are all test methods, including those for auxiliary measurements (i.e., diluent gas concentration, exhaust flow, moisture, as applicable), supposed to be identified? 
 Section 17.23 Process Parameter: It is not clear what the process parameter is supposed to represent. The primary "process parameter" as it relates to MA TS performance stack testing is load, with a general requirement to maintain load at 90-110% of rated capacity (unless an alternate load is more representative of normal operation). If load is the only possible process parameter, then this field should specifically require the reporting of load in a specified unit of measure. 
 Section 17.28 Control Device Description: It is not clear what this data element is deigned to capture. If this is simply a broad description of the control devices installed on a given EGU, such information is already identified with the 40 CFR Part 75 Electronic Monitoring Plan (see the Unit Control Data record). 
 Section 17.30 Action on Process Material (e.g., burned): Again, the intent of this field is not clear. As MA TS regulates the combustion of fossil fuels, if the action on process parameter would always be burned, then there is no inherent reason to submit such a static value for every performance stack test and QA test. 
 Section 17.33 Project Number: The intent of this field is not clear. Item 17.21 already requires that each test be provided with a unique identifier. If this field refers to a project number assigned by the test contractor, what is the rationale for requiring that it be reported electronically to the EPA? 
 Section 17.35 Percent 0 2/C02 Correction: The intent of this field is not clear. While certain regulatory standards are expressed as a concentration corrected to a specified diluent level, none of the emission limits under MATS fall into the preceding category. Thus, this field should be eliminated. 
 Section 17.38 Type of Fuel: This information should only be required for performance stack tests, not QA tests. In the case of QA tests, the type of fuel is not relevant as the "test" is not linked to the compliance demonstration; for CEMS equipped units, compliance is instead based upon direct CEMS data.
 Sections 18.13 Dry Basis F-Factor (Fd) and 18.14 Wet Basis F-Factor (Fw): It is unclear why Fd and Fw fuel factors would need to be provided in the context of Reference Methods 1-4. The preceding factors are generally used in conjunction with pollutant concentrations in order to express the pollutant emissions rate in terms of mass per unit of heat input. Further, it is unclear why the EPA presumes that only a Fd or Fw fuel factor would be relevant. Many coal-fired EGUs monitor C02 as the diluent, and any conversion of Reference Method pollutant concentrations could also be based upon use of C02 as the diluent, in conjunction with the applicable C02 based fuel-factor (Fc). 
 Section 18.21 Manometer Used: It is not clear what the EPA hopes to collect with this data element. Is the EPA simply trying to collection information on the type of manometer used? 
 Section 18.22 Run Elapsed Time at Start (=O): The purpose of this data element is not clear. Parenthetically, the EPA seems to be suggesting that this field will always contain a value of zero. If the preceding is accurate, then this field should be eliminated. There is no reason what an affected EGU should be required to report a static data element for every single MATS related test event. 
 Section 19.0 Data Elements for Methods 5, SD, 26, 26A and 29: The intent of several of the data elements within this section is not clear, and the EPA has also provided no guidance in relation to the convention to be used for time reporting. For example, are the clock start time and clock end time supposed to reflect local time, Eastern Standard Time (EST), etc.? Additional specific comments on this section are as follows. 
 Section 19.8 Process Parameter Run Data: The intent of this data element is not clear. If this field is to be used to collect load data for those tests used to demonstrate compliance with an output based emission standard, such guidance should be provided, as well as a direct reference to load (rather than the more generic reference to process parameter). Further, if the preceding is accurate, then EPA should clarify whether this field applies for those compliance tests associated with mass per unit heat input emission limits or QA tests. 
 Sections 19.18 Process Run ID and 19.19 Process Run Parameter ID: The intent of these fields is not clear. Any process data (i.e., load data) collected during a test would be paired with the Reference Method data collection and the associated RM Run Numbers/IDs. Thus, it is unclear why there would be a separate set of "Process Run IDs" for a given test event. 
 Section 20.0 Data Elements for Methods 6C and 3A: The intent of several of the data elements within this section is not clear, and the EPA has also provided no guidance in relation to the convention to be used for time reporting. For example, are the clock start time and clock end time supposed to reflect local time, Eastern Standard Time (EST), etc.? Additional specific comments on this section are as follows. 
 Sections 20.8 Cylinder ID; 20.9 Gas Level (Zero, Low, Mid, High); 20.11 Compound (Analyte) and 20.12 Cylinder Gas Units of Measure: There are no instructions as to what info1mation is to be entered in these fields. Often, one set of gas cylinders (zero/low, mid and high) will be used for multiple gases. For example, consider a case where the zero/low level gas is a bottle of pure nitrogen, and there are then a single bottle for each of the mid and high level gases (with each bottle containing S02, C02, 02 and a balance ofN2). The data elements should clearly note that only a zero or low-level record need be submitted (not both). Further, with the "analyte" field, it seems as though the EPA may require a unique set of cylinder IDs for every "analyte" regardless of whether multiple analyzes are contained within a single gas cylinder. If the preceding is an accurate understanding, this is a very inefficient way to enter the calibration gas data and will be burdensome when this data is associated with testing that occurs every calendar quarter. 
 Sections 20.20 Calibration Span Concentration and 20.24 Calibration High-Level Concentration: Requiring submittal of values for both of these parameters makes no sense, as per Sections 3.3.3 and 3.4 of RM 7E, these values must be the same ("3.3.3 High-Level Gas means a calibration gas with a concentration that is equal to the calibration span."). Entry of the high-level gas concentration alone should be sufficient. 
 Sections 21.21 Calibration Zero-Level Concentration and 20.22 Calibration Low-Level Concentration: Again, neither RM 6C or 3A requires the use of both a zero-level and low-level calibration gas; one of the two gas levels would be employed. Thus, these data elements should be clarified to note that only 1 of the two records is required. 
 Sections 20.25 Zero Gas Response and 20.26 Low Level Response: It is presumed that these refer to the responses during the analyzer calibration error test as required in RM 7E, Section 8.2.3. Again, neither RM 6C or 3A requires the use of both a zero-level and low-level calibration gas; one of the two gas levels would be employed. Thus, these data elements should be clarified to note that only 1 of the two records is required. 
 Sections 20.29 Span Zero Response and 20.30 Span High Response: It is not clear what these data elements are supposed to represent. How are these data elements any different from those associated with Items 20.25 and 20.28? Are these data elements specific to use of dilution sampling, whereas the data elements in Items 20.25-20.28 apply to both straight extractive and dilution sampling systems?
 Sections 20.31 Pre-test Zero Response and 20.32 Pre-test Bias Response: Again, it is not readily apparent what the EPA is attempting to collect in these data elements. It is suspected that these data elements correspond to the initial bias test as required in RM 7E, Section 8.2.5 (at least for straight-extractive sampling systems). If the preceding understanding is correct, it is suggested that these data elements be renamed as follows: 20.31 =Initial Low-Level Bias/System Calibration Error Response; 20.32 =Initial Upscale Bias/System Calibration Error Response. Also, note that the bias test is conducted at two different calibration gas levels: 1) zero or low-level, and 2) mid-level or high-level (i.e., upscale). Thus, referring to a "Zero" response is not appropriate, as either a zero-level or low-level calibration gas could be used for the test. 
 Sections 20.33 Post Zero Response and 20.34 Post Span Bias Response: Again, it is not readily apparent what the EPA is attempting to collect in these data elements. It is suspected that these data elements correspond to the post-run bias test as required in RM 7E, Section 8.5. If the preceding understanding is correct, it is suggested that these data elements be renamed as follows: 20.33 =Post-run Low-Level Bias/System Calibration Error Response; 20.34 =Post-run Upscale Bias/System Calibration Error Response. Also, note that the bias test is conducted at two different calibration gas levels: 1) zero or low-level, and 2) mid-level or high-level (i.e., upscale). Thus, referring to a "Zero" response and "Span" response is not appropriate, as the preceding suggests that a zero-level and high-level gas must be used for these tests (i.e., it would also be permissible to use low-level and mid-level calibration gases).
 Sections 20.35 Raw Measured Concentration (Cavg) and 20.36 Raw Measurement Units: Neither of these data fields capture where the measurement is on a dry basis or wet basis. Further, for RM 6C and 3A, the measurement units should simply be stipulated (i.e., ppm for RM 6C and % by volume for RM 3A). 
 Sections 20.37 Zero Gas Percent Error and 20.38 Low Gas Percent Error: Again, neither RM 6C or 3A requires the use of both a zero-level and low-level calibration gas; one of the two gas levels would be employed. Thus, these data elements should be clarified to note that only 1 of the two records is required. Furthermore, there is a presumption that the result will be reported as percent error. These checks are also acceptable if the absolute difference is less than or equal to 0.5 ppm (for S02) or 0.5% C02 or 02. In cases where the alternate performance criteria are employed, it is still necessary to report the result as percent error? Sections 20.41 System Zero Level Calibration Error and 20.42 System High Level Calibration Error: Yet again, the desired information is not readily apparent. It is suspected that these data elements are specific to dilution sampling systems, and the reported values should correspond to the Initial Low-Level System Calibration Error Check and Initial Upscale System Calibration Error Check results (per RM 7E, Section 8.2.5). 
 Sections 20.43-20.50, Pre-run and Post-run Bias and Drift Information: These data elements relate to pre- and post-run bias and drift check results for each of the two required calibration gas levels. Yet again, the EPA incorrectly refers to "Zero" and "High" level checks even though RM 6C and 3A permit the use of either zero or low-level gases and either mid-level or high-level gases. Furthermore, it seems like there is a presumption that the results will be reported as percent error (i.e., 20.45 Pre-run High Level Bias, Percent). These checks are also acceptable if the absolute difference is less than or equal to 0.5 ppm (for S02) or 0.5% C02 or 02. In cases where the alternate performance criteria are employed, it is still necessary to report the result as percent error? 
 Section 20.51 Calculated Average Dry Emissions Concentration (Cgas): In cases where dilution sampling is employed for RM purposes, why would there be a need to report a dry gas concentration? This data element should note that it applies only when straight-extractive sampling is employed. 
 Sections 20.52 Measurement Units of Cgas (Dry) and 20.53 Measurement Units of Cgas (Wet): For RM 6C and 3A, the measurement units should simply be stipulated (i.e., ppm for RM 6C and % by volume for RM 3A). 
 Section 21.0 Data Elements for Method 30B: The intent of several of the data elements within this section is not clear, and the EPA has also provided no guidance in relation to the convention to be used for time rep01iing. For example, are the clock start time and clock end time supposed to reflect local time, Eastern Standard Time (EST), etc.? Also, this section does not differentiate between Reference Method 30B employed for Relative Accuracy Test Audit (RATA) purposes or RM 30B employed for Low Emitting EGU (LEE) testing, and there are some unique considerations depending upon the context. 
 Section 21.8 Process Parameter Run Data: The intent of this data element is not clear. If this field is to be used to collect load data for those tests used to demonstrate compliance with an output based emission standard, such guidance should be provided, as well as a direct reference to load (rather than the more generic reference to process parameter). Further, if the preceding is accurate, then EPA should clarify whether this field applies for those compliance tests associated with mass per unit heat input emission limits or QA tests. 
 Section 21.9 Area of Stack: It is not clear if the "Area" being sought by the EPA is the area at the RM 30B sampling location or the area at the exhaust gas discharge point. The location for which the area is to be provided should be clarified. 
 Sections 21.12 %02 and 21.13 %C02 : For RATAs using RM 30B, the comparison between the RM data and CEMS or sorbent trap monitoring system data is performed on a ug/scm basis (see 40 CFR 63, Subpart UUUUU, Appendix A, Section 4.1.1.5.2). Therefore, for such RAT As, there is no need for a diluent gas value and these records should note that they only apply to LEE testing. Furthermore, per §63. l 0005(h)(3)(i), the EPA recognizes that a diluent gas value is only need is electing to show compliance with the MATS lb/TBtu limit, and only 0 2 or C02 is needed (not both). These additional nuances on the reporting of 0 2 and C02 data in the context of RM 30B must be captured. Lastly, no moisture basis is specified in regard to the reporting of the diluent concentration. 
 Section 21.14 Stack Gas Volumetric Flow Rate (dry, standard conditions): There is no context for the reporting of this data. As described above for Items 21.12 and 21.13, this data is simply not needed when using RM 30 for RATA purposes. Furthermore, this data is only required for LEE testing purposes when demonstrating compliance with an output based emission limit. Lastly, §63.10005(h)(3)(i)(B) notes that the flow data can be comprised of data collected from a 40 CFR Part 75 certified flow monitoring system. As many such flow monitoring systems measure on a wet basis, the proposed Appendix E should allow reporting on either a scf or dscf basis, along with a field for unit of measure. 
 Section 21.16 Associated Process Run Rate: The intent of this data element is not clear (if further is not clear how this information would differ from the Section 21. 8 Process Parameter Run Data). 
 Sections 21.17 Start Minutes (cumulative); 21.18 End Minutes (Cumulative) and 21.19 Actual Clock Time: The intent of these fields is unclear. Run start and end times are already required in Items 21.6 and 21.7, there is no obvious meaning for "cumulative" start and end minutes, and no explanation of what is meant by Actual Clock Time. 
Response 2: As mentioned earlier, the Agency disagrees that these data elements are unnecessary and that a PDF file submission is equivalent with electronic reporting. Please see Response 1 of Section 14.2.1.
The Agency appreciates the commenters' suggestions for clarification on specific data elements. 
To assist with the clarification of the data elements, the data elements have been reorganized.  The original layout was by general information and methods.  The EPA has restructured the data elements into smaller groupings to assist with clarifying the data elements.  Those grouping are facility and testing information, test location information, isokinetic method run-level data element and point-level elements, Method 30B run-level data element and point-level elements, instrumental pre-test calibrations, Method 3A and 6C run-level data elements, Method 2 run-level data element and point-level elements, and Method 4 run-level data elements.  Our responses to those suggestions are shown by section number below. Where appropriate, the section in Appendix E of rule has been clarified as follows: 
"17.14 State Facility ID - The state identification number as provided by a state air pollution control agency.  Many state and local agencies have an alphanumeric identifier for individual process operating units with an associated name describing the unit." This data element allows the delegated authority to link the source and the report to the state source identifier;   
"17.25 Test Method  -  the test method directly associated with the parameter being reported."  These methods correspond to the methods in 63.10007 and Table 5.
Particulate Matter  -  Method 5 or 5D
Non-Hg HAP metals  -  Method 29
Hydrogen chloride and hydrogen fluoride  -  Method 26, 26A, Method 320, or ASTM 6348-03
Mercury  -  Method 30B, ASTM D6784, or Method 29
Sulfur dioxide  -  Method 6C
According to §63.10007(e)(3) if data is being reported in the units of lb/MWH or lb/GWh, the emission data must be in lb/h  If an isokinetic method is not used, Method 2, 2A, 2C,  2F, or 2G may be needed to determine stack gas volumetric flow rate (scfh) and, in some cases, Method 4 or an approved alternative may be required for moisture content;  
"17.23 Process Parameter  - Gross load, as the average electrical load during the performance test (megawatts or gigawatts), or heat input (mmBTU).   If gross load is chosen, this value is required to be recorded during the performance test and is needed for the calculation of the emission limits expressed in lb/MWh or lb/GWh.;  
"17.20 Control Device Description  - a description of all emissions control devices used in-line with the process flow that is being tested." This is important information needed for future technology reviews and emission factor development.  According to CAA §112(d)(6), developments in practices, processes and control technology must be reviewed every 8 years.  Also, according to the general provisions at §63.7(g)(2)(iii), description of control devices is required in performance test reports;  This information is taken from the NEI. 
"17.30 Action on process material (e.g., burned or generated)  -  For this subpart, the action could be steam or electricity generated or heat input (heat derived from combustion of fuel, coal or oil burned."  This information is needed to know which standard is relevant to the testing data;
"17.2 Subpart  -  subpart for which the emissions test is being conducted." A performance test can be conducted for multiple subparts. It is important to identify all of the EGU's the applicable regulations as required in the general provisions §63.7(g)(2)(ii). For instance, a MATS performance test report may also be utilized for an NSPS regulation (e.g., subpart Da) or for a state's PSD requirements. Moreover, testing for multiple subparts may be conducted at the same time, and the data for a specific subpart resides in one complete report;
"17.15 SCC  -  source characterization code for the EGU." Sources can have multiple SCCs for an EGU, depending on the type of fuel that is being utilized.  Also, the general provisions at §63.7(g)(2)(iii), require the appropriate SCC to appear in performance test reports. There are over 40 SCCs in the Electric Generation section. There is no SSCs for blends of fuel.  Select the most representative fuel type to determine which SCC is appropriate, These SCC are listed in the Reporting Instructions MATS Supplemental Stack Test Data.
"17.21 Project number  -  Unique identifier for the test report." EGU owners or operators who are required to comply with ASTM D7036, Section 6.10.3(c) are also required to provide a unique identifier of the test report. This data element refers to that unique identifier for the test report. Use of an unique identifier allows the XML schema and corresponding PDF files to be identified and grouped. Our state, local, and tribal partners have also requested unique project numbers for test reports. Reports with unique identifiers may be utilized to help delegated authorities sort quarterly reports for different EGUs at the same or at different sites. The project number can be referenced easily in communications to ensure the same report is being discussed. Unique identifiers are also important because stack testers can test multiple units in a day and have multiple reports submitted with the same date; 
"17.34 Emission Concentration Units." The rule will clarify that this is the units of measurement that are needed for the calculation of the emissions. For instance, ppm and % are emission concentration units used for SO2 and O2; gr/dscf is a PM emission concentration unit;  
17.35 Percent O2/CO2  -  Diluent Correction Percent is not required by MATS to be reported, and it will be removed from the rule;
"18.35 Units of Process Parameter - Megawatts/hour, gigawatts/hour, or million British Thermal Units/hour (mmBTU/hr)".  These units need to be stated so the delegated authority and the public know which compliance strategy is being followed.
"17.36 Quantity of Fuel Combusted  -  tons of coal, gallons of oil, tons of petroleum coke, or thousand cubic feet of synthetic gas combusted during the stack test." This data element is the value or measure of the process parameter that is applicable to the compliance strategy- normal operating load.  The quantity of fuel used must be capable of providing between 90 and 110% of the design load capacity but should be representative of site specific normal operations during each run.  This amount will be needed to complete the calculations in §63.10007(e)(3) for emission limits in terms of lb/MWh or lb/GWh;
"18.22 Type of Fuel Combusted  -  The type of fuel combusted during the test." Fuel type indicates what F-factor will need to be used if using that compliance strategy. Fuel type also needs to be identified for SCC selection and emission factor development;
 "18.2 Run Date  -  This is the date the test run is started." While the commenter suggested adding an end date added to each section where the Run date is listed, the rule remains unchanged as this addition is unneeded since each run has a clock start and end time;
Note that the Dry and Wet Basis F-Factors (Fd)  and  (Fw), of Sections 18.13 and 18.14 of Appendix E, respectively, can be used in EPA Method 4 if the equation in Section 16.4 of Method 4 is used. That value will be the volume of wet combustion components per unit of heat content at zero percent oxygen, dscf/10[6] BTU (scm/J).
"18.26 Percent Moisture  -  Actual  -  obtained from Method 4 or default value. Note that when the default value is provided, it must be identified as a default value." Where Methods 1-4 were used determine the flow rate, the actual moisture found from Method 4  needs to be reported.  If a default value is used for the emission calculation, it needs to be documented in the report that the actual moisture determined by Method 4 was not used and the default value was used;
"18.21 Manometer Used  -  the unique identifier associated with the manometer." The identification of the manometer used is needed to relate the QA calibration data to a specific manometer;
 "Clock Time Start  -  standard time (hour and minute) that a test run begins." and "Clock Time End  -  standard time (hour and minute) that a test run ends." The Agency notes that initially these data elements were to calculate the duration of the sampling run; however, in order to clarify and be consistent with the previous section, these section titles are changed in the rule to mirror section titles for 18.3 and 18.4.  Note that part §75.59 (5)(ii) (A) and (B) refer to the similar data elements - (A) involves the Date, hour and minute of beginning of test run while (B) involves the Date, hour and minute of end of test run. The Agency also notes that these data elements are  to be reported as the start and end times at the start of the test run in standard time, not accounting for daylight saving time;
"19.8  Process Parameter Run Data  -  heat input (mmBTU/hr) or the load average recorded during the test run." The value selected depends on the compliance strategy selected;
"19.17 Final Train Leak Check Vacuum  -  value from the final train leak check." This data element replaces "Pump Vacuum" from the rule. The Agency notes that pump vacuum values must be on the field data sheets for each sampling point in the PDF file to ensure the sampling trains pass the leak check at the maximum vacuum used to pull gas through the sampling train. Also note that this data element is needed as a QA check for a valid test run;
"19.18 Process Run ID  -  identifier for the process run number." This identifier is necessary so that an individual run can be linked to corresponding sampling test runs;
"19.19 Process Run Parameter ID - electrical load during the performance test (megawatts or gigawatts) or heat input (mmBTU)." The parameter selected is dependent on the selected compliance strategy;
"18.21 Nozzle Calibration  -  nozzle diameter in inches." This is the value obtained from the calibration sheets for the diameter of the nozzle, and it allows the area of the nozzle to be calculated for isokinetics;
"18.29 and 18.30, Initial and Final Volumes of Dry Gas Meter -  actual meter box reading recorded during the test run." The difference between the values represent the volume that will not exclude mid-run leak checks. Those meter readings at the leak checks will need to be recorded so that they can be subtracted from the difference. Note that the difference will exclude a post run leak check;
"19.7 Filter Temperature  -  sample gas temperature through the filter." This data element will be changed in the rule to remove "/Oven" to reflect the requirement of Section 8.5 of Method 5, Figure 5-3 of Method 5, and Table 5 of MATS. The Agency notes that the front half temperature include probe temperature and the sample gas temperature through the filter;
"23.31 and 23.33 Filter Train Leak Check Rate  -  leak rate of filter train." The rule will be changed from Filter/Oven Exhaust Temperature. The Final Train Leak Check Rate is needed as a QA check for a valid train test run;
"18.16 Cumulative Elapsed Sampling Time  -  Total sampling time per isokinetic run." The Agency notes that the cumulative elapsed sampling time was inadvertently omitted from the data elements of section 19 of Appendix E of the rule. As the cumulative elapsed sampling time differs from the period between the clock start and stop times, as that period includes port changes and possible leak checks that may take place between port changes, the rule now includes the cumulative elapsed sampling time data element. The Agency also notes that for isokinetic sampling, the difference between clock start and end times is generally longer than the duration of the sampling train collection and that the cumulative elapsed sampling time will be the total minutes that the sampling train collected sample only (so leak checks, port changes and sampling disruptions will not be included);The Agency agrees with the commenter, and the rule will remove section 20.1 Sampling Location, since this data element appears in Section 17 and does not need to be repeated;
As mentioned in sections 19.5 and 19..6 above, the titles for sections 20.11 and 20.13 will be changed in the rule to "Clock Time Start" and "Clock Time End," respectively. Consistent with sections 19.5 and 19.6, these data elements should be reported as the time at the start or the end of the test run in standard time, not accounting for daylight saving time. The Agency notes that section "25.6, 25.9, 25.12 Cylinder ID" is the cylinder ID that is on the gas certification sheet for the calibration gas cylinder;
The Agency agrees with the commenter that zero gas can be used as a low level gas, so the title of section 20.9 of the rule will be revised to "25.7 Calibration Low-Level (or Zero Gas Concentration), Mid, High);"
"25.3 Calibration Gas Cylinder Analyte  -  name of the compound in the cylinder." This information serves as a quality assurance check on the cylinder; 
"26.35 Run Number of Process Parameter" - heat input (mmBTU/hr) or the load average recorded during the test run." The value selected depends on the compliance strategy selected;
The Agency notes that "20.18 Clock time" is redundant, as pointed out by a commenter. The rule will be revised to remove this data element;
"25.13 Calibration Span (High-Level) Gas Concentration  -  high concentration value on the gas certificate." The Agency agrees that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references section 3.3.3 of Method 7E, the span concentration is equal to the High-level concentration;
The Agency agrees that section "20.20 Calibration Zero-level Concentration" is redundant; this data element will be removed from the rule;
"25.7 Calibration Low-level (or Zero) Gas Concentration  -  low or zero concentration value on the gas certificate." The Agency agrees that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Method 7E, the Low-level concentration may be the zero-level concentration;
The Agency agrees that section "20.22 Calibration High-level Concentration" is redundant; this data element will be removed from the rule.
The Agency agrees that section "20.23 Zero Gas Response" is redundant; this data element will be removed from the rule;
"25.164 Low-Level (or Zero) Gas Response in Direct Calibration Mode  -  measured concentration of the low (or zero) calibration gas when introduced in direct calibration mode." The Agency agrees a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E,  the Low Gas Response is the measured concentration of the low (or zero) calibration gas when introduced in direct calibration mode (ppmv);
"25.19 Mid-Level Gas Response in Direct Calibration Mode - the measured concentration of the mid calibration gas when introduced in direct calibration mode (ppmv)." Note that according to Method 6C which references Section 12.1 of Method 7E, the Mid Gas Response is the measured concentration of the mid calibration gas when introduced in direct calibration mode (ppmv);
"25.22 Span (High-Level) Gas Response in Direct Calibration Mode - the measured concentration of the high (span) calibration gas when introduced in direct calibration mode (ppmv)." The Agency agrees a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, the High (Span) Gas Response is the measured concentration of the high (span) calibration gas when introduced in direct calibration mode (ppmv);
"20.27 System Low (or Zero) Response - the measured concentration of the low (or zero) calibration gas when introduced in system calibration mode (ppmv). The Agency agrees a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references section 12.1 of Method 7E, Cs, the System Low (or zero) Response is the measured concentration of the low (or zero) calibration gas when introduced in system calibration mode (ppmv);
"20.28 Span High Response - the measured concentration of the calibration gas when introduced in system calibration mode (ppmv)." The Agency agrees that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, Cs, the System Upscale (Mid or High) Response is the measured concentration of the calibration gas when introduced in system calibration mode (ppmv);
"26.15 Pre-Run Low-Level (or zero) System Bias / System Response - the actual initial system calibration concentration of the low (or zero) calibration gas." The Agency notes that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, COA, the Pre-test Low (or zero) System Response is the actual initial system calibration concentration of the low (or zero) calibration gas;
"26.18 Pre-Run Upscale (mid or span-high)System Bias / System Response - the actual initial system calibration concentration of the upscale (mid or high) calibration gas." The Agency notes that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, CMA, the Pre-test Bias /System Response is the actual initial system calibration concentration of the upscale (mid or high) calibration gas;
"26.21 Post-Run Low-Level (or zero) System Bias / System Response - the actual final system calibration concentration of the low (or zero) calibration gas." The Agency notes that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, COA, the Post-test Low (or zero) Bias / System Response is the actual final system calibration concentration of the low (or zero) calibration gas;
"26.26 Post-Run Upscale (mid or span-high) System Bias / System Response - the actual final system calibration concentration of the upscale (mid or high) calibration gas." The Agency notes that a title change would offer clarification, so the rule has been revised. Note that according to Method 6C which references Section 12.1 of Method 7E, CMA, the Post-test Bias /System Response is the actual final system calibration concentration of the upscale (mid or high) calibration gas;
"26.30 Unadjusted Raw Emissions Average Concentration  - the average of the unadjusted gas concentration indicated by data recorder for the test run.";
"26.31 Unadjusted Raw Emission Units of Measure (dry)  -  the units used to measure the unadjusted gas concentration which must indicate if the sampling was on a wet basis or a dry basis (for instance: ppmvd, ppmvw, %vd, %vw).";
The Agency agrees that "20.34 Zero Gas Percent Error" is redundant and will be removed from the rule;
"25.17 Low-Level (or zero) Calibration Error  -  the initial low (or zero) gas analyzer calibration error calculation (Eq. 7E-1)." Note that his value needs to be within 2% of the calibration span or less than or equal to 0.5 ppm absolute difference, according to section 13.1 of Method 7E;
"25.20 Mid-Level Calibration Error  -  the initial mid gas analyzer calibration error calculation (Eq. 7E-1)." Note that this value needs to be within 2% of the calibration span or less than or equal to 0.5 ppm absolute difference, according to section 13.1 of Method7E ;
"25.23 Span (High-Level) Calibration Error  -  the initial span (high) gas analyzer calibration error calculation (Eq. 7E-1)." Note that this value needs to be within 2% of the calibration span or less than or equal to 0.5 ppm absolute difference, according to section 13.1 of Method 7E;The Agency agrees that "20.41 System Zero Level Calibration Error" and "20.42 System High Level Calibration Error" are redundant and will be removed from the rule;
"20.38 Pre-run Low (or Zero) System Bias - the calculated Pre-test Low (or zero) System Bias in percent of the calibration span." Note that this data element is from Method 6C which references SBi from Eq 7E-3 of Method 7E;
The Agency agrees that "20.44 Pre-run Zero Drift" is not necessary and will be removed from the rule;
"20.45 Pre-run Upscale (mid or high) System Bias - the calculated Pre-test upscale (mid or high) System Bias is in percent of the calibration span." Note that this data element is from Method 6C which references SBi from Eq 7E-3 of Method 7E;
The Agency agrees that "20.46 Pre-run Zero Drift" is not necessary and will be removed from the rule;
"26.21 Post-Run Low (or Zero) System Bias - the calculated Post-test Low (or zero) System Bias is in percent of the calibration span." Note that this data element is from Method 6C which references SBfinal from Eq 7E-3 of Method 7E; 
"26.23 Post-Run Low-Level (or zero) Drift - drift assessment in percent of the calibration span." Note that according to Method 6C, which references Eq 7E-4 D of Method 7E, the calculated Post-test Low (or zero) must be less than 3% or the bias responses must not differ more than 0.5 ppmv per section 13.3 of Method 7E ; 
"20.49 Post-run Upscale (mid or high) System Bias - the calculated Post-test Upscale System Bias in percent of the calibration span." Note that this data element is from Method 6C which references SBfinal from Eq 7E-3 of Method 7E;
"26.25 Post-Run Upscale (mid or span-high)Drift - drift assessment in percent of the calibration span."  Note that according to Method 6C, which references Eq 7E-4 D of Method 7E, the calculated Post-test Low (or zero) must be less than 3% or the bias responses must not differ more than 0.5 ppmv per section 13.3 of Method 7E; 
"26.32 Calculated Average Dry Concentration, Adjusted for Bias (Cgas)  -  average effluent gas concentration adjusted for the bias, ppmv.";
"26.33 Concentration Units of Measure (Dry) - ppmvd or %vd.";
"20.53 Measurement Units of Cgas (Wet) - ppmvw or %vw.";
The Agency agrees that "21.1 Sampling Location" is redundant and will be removed from the rule;
As mentioned in sections 19.5 and 19..6 above, the titles for sections 23.9 and 23.11 will be changed in the rule to "Run Start Time (Clock Time Start)" and "Run End time (Clock Time End)," respectively. Consistent with sections 19.5 and 19.6, these data elements should be reported as the time at the start or the end of the test run in standard time, not accounting for daylight saving time; 
As mentioned in  sections 19.8 and 20.16, section 23.45's description is "23.45 Run Number of Process Data  -  heat input (mmBTU/hr) or the load average recorded during the test run." Note that the unit selected depends on the EGU's compliance strategy;
The Agency agrees that "21.9 Area of Stack" is redundant and will be removed from the rule;
With regard to the explanation for "23.43 Percent O2," As stated in 1.0 of Appendix E, you must record the electronic data elements in this appendix that apply to your compliance strategy under this subpart.  If %O2 is needed to calculate your emissions and is applicable to your compliance strategy, then you need to report it when conducting Method 30B;
With regard to the explanation for "23.44 Percent CO2,"As stated in 1.0 of Appendix E, you must record the electronic data elements in this appendix that apply to your compliance strategy under this subpart.  If %CO2 is needed to calculate your emissions and is applicable to your compliance strategy, then you need to report it, when conducting Method 30B;
With regard to the use of certified flow monitoring systems for "21.14 Stack Gas Volumetric Flow Rate (dry, standard conditions), please see the detailed discussion in Response 1 in Section 14.2;
"23.45 Run Number of Process Data  -  unique identifier for a process run." Note that this data element replaces the redundant Associate Process Run Rate title; 
The Agency agrees "21.17 Start Minutes (cumulative)" is not necessary and will be removed from the rule;
"23.15 Cumulative Elapsed Sampling Time (A);" 

The Agency agrees that "21.19 Actual Clock Time" is redundant and will be removed from the rule; and

The Agency notes that "21.21 Pre-Leak Check Vacuum (in. Hg)" is a QA element required by section 8.3.1 of Method 30B. This pre-test leak check must be conducted to about 15 inches of Hg, and the value needs to be recorded to confirm that the appropriate vacuum was pulled..

14.2.2 XML
Comment 1: Commenter 20597 said even if XML stack test data are reported for performance tests, the files should not be required for MATS CEMS performance evaluation tests. All the information needed to document the QA of the CEMS will be reported in the "Quality Assurance and Certification" XML data files, which sources must submit via ECMPS. The QA files currently include tests such as the RATA results for mercury CEMS and HCl CEMS, and, under the proposal would include PS-11 correlation tests, RCAs, and RRA for PM CEMS. These files already document significant run level detail in XML format along with information such as start/end times, CEMS concentrations, reference method concentrations, and unit load a long with detail s on the test results such as the mean difference, standard deviation, confidence coefficient, and relative accuracy values. The Agency has successfully used these types of QA records to document CEMS quality assurance test data for over two decades under 40 CFR Part 75 for the ARP and NOx Budget/CSAPR Programs. These records have been sufficient and remain sufficient for that task. Similarly, the RATA records (and the analogous QA records proposed for PM CEMS) are adequate for the purpose of documenting the quality assurance of CEMS used for the MATS Rule. In the case of the SO2 CEMS, the same exact equipment that is used for Part 75 monitoring is used for MATS with no additional QA/QC requirements. So, why would there be the need for additional reporting? Likewise, when the Agency finalized very similar mercury monitoring requirements for compliance under the Clean Air Mercury Rule (before the courts vacated the rule for reasons completely unrelated to either the monitoring or reporting requirements), nearly identical QA records were deemed sufficient under that rule. There is simply no reason that the additional XML reporting proposed in Sections 14 through 22 of Appendix E would now need to be required for RATAs, PS-11 correlation test, RCA or RRAs. Removing the requirement for the CEMS QA related stack test data will resolve the duplicative reporting.
Commenter 20610 stated they've operated a coal-fired power plant for many years. During EPA's interim reporting period they filed electronic PDF files in accordance with the interim requirements. These interim reporting requirements methods seemed adequate to provide EPA with means to verify a source's adherence to EPA's procedures. EPA now proposes, in the name of necessity for EPA to assure compliance, that the source incorporate raw data taken from performance test reports and other evaluations into the electronic data report. For many decades now, vendors have been successful in employing EPA-approved methods and calculations to sources and providing competent and timely emission evaluation reports to the source for subsequent transfer to regulatory agencies as a means of assuring compliance. In recent years, EPA, has developed qualified vendors to further assure conformance with their methods. EPA is always presented with such reports to confirm the accuracy of presented results. Sources, consultants, and state agencies monitor and verify that vendors follow the procedures now. EPA would now have sources transfer the raw data received from test vendors into XML format so it can duplicate the entirety of the test report to assure that the source has faithfully conducted an evaluation in accordance with EPA methods. The commenter disagreed that EPA will perform the computations into the future any more frequently than it has in the past. If EPA has questions or concerns about certain test results presented in a report it has the option to compel a source to test again, or it may conduct the test. Unless we misunderstand the proposal, much of it seems wholly unnecessary and an extra burden from which no substantive errors will be determined and for which no environmental benefit will be gained. The commenter requested that EPA either withdraw the proposal or receive the helpful suggestions offered by vendors, software engineers, and source operators in such a way as to reduce to the minimum the changes necessary to demonstrate that compliance is achieved by the sources. The commenter said this is nearly adequately done now. Regardless, it is important that EPA provide enough time to incorporate the software changes that are necessary to fully implement the final revisions to MATS reporting requirements. The commenter disagreed this can be concluded by January 1, 2017. The implementation date needs to be consistent with the magnitude of changes ultimately required by EPA, so while the commenter suggested the new reporting for data collected be eight quarters following the date when the final rule is published, a lessening of the burden now proposed by EPA might make a somewhat earlier date more possible.
Commenter 20606 stated the original MATS rule reporting requirements for performance tests are different than what is required for CEMS performance evaluations. The results of a performance tests shall include the analysis of samples, determination of emissions, and raw data. While raw data is not necessarily required as part of the performance evaluation results, some raw data is required for performance test results. But, even for the performance tests, the required reported results would not represent a complete, detailed stack test report since §63.7(g)(3) states that the performance test recordkeeping requirements include "the records or results of such performance test and other data needed to determine emissions from an affected source." §63.10007(g) likewise indicates that "such records as may be necessary to determine whether the performance tests have been done according to the requirements" shall be made available "upon request." This implies that the "results" which are required to be reported do not necessarily include all the data needed to determine the emissions. The commenter said it is important for EPA to recognize that MATS sources were never required to report all the data within the ERT, decide how best to report the data and what data needs to be reported, what portion of that data is critically required in a numeric XML format, and what data is adequately or better addressed within a detailed stack test report that can be submitted electronically in a PDF (or equivalent) file. Since the ERT is no longer used for reporting under the proposed MATS revisions, all the references to the ERT need to be removed. If the references to the ERT are not removed it would be confusing and result in duplicate reporting. 
Response 1: The Agency finds the commenter's suggestion that raw data will not be calculated more frequently mistaken. One aspect of electronic reporting is that every emissions test will have all of the underlying calculations performed; as has been mentioned earlier, this automatic calculation is one reason that PDF file submissions, which provide data that must be reentered, are unsuitable. While the Agency has worked to accept as much data from other programs, such as the acid rain program, differences between program objectives and statutes preclude data incorporation from other programs into MATS. With respect to comments concerning use of third party programs to assist EGU owners or operators with their submission requirements, the Agency continues to support such efforts by developing an XML schema that can be used by anyone to supply the necessary data, even though use of the ERT already ensures that the performance testing data can be submitted appropriately.  

As mentioned earlier, the Agency maintains that the ERT is a tool that could be used to develop a complete test report and XML. It already identifies the data required to be reported in the recently-promulgated general provisions updates to parts 60 and 63; however, while the ERT remains available for use, its use is not mandatory. As discussed in the preamble, the XML file to be submitted electronically can be generated to meet the reference method data submission requirements either via the existing ERT or via another program that provides the data in an appropriate XML file format.  

Also as mentioned before, the Agency maintains that complete emissions testing information was and is required to be submitted via MATS. The Agency relies on the data elements listed to determine the emissions, verify compliance and develop emission factors. Moreover, in order to provide enough information to the delegated authority to determine if the results submitted pass the quality assurance provisions of the reference method and are valid emissions data, the raw data are needed, just as the raw data are needed by the acid rain program (see §75.21(f) and Appendix A 3.1.2(a) of part 75).
Comment 2: Commenter 20612 disagreed with references to use of ERT for any purpose in MATS and indicates that they instead support EPA proposal to specifically define the data elements that are to be reported for MATS in an XML format for performance tests. They support reporting in terms of an actual format making reporting more easily automated and integrated using the tools sources and testers already have in place. 
Commenter 20609 indicated that EPA has proposed language limiting the obligation to submit the detailed data elements for reference method information in Appendix E sections 17-22 to "performance stack tests" conducted using a method (or methods) supported by the ERT. This commenter said the purpose and implications of ETS's applicability is not clear and the commenter objects to it. EPA suggests that the purpose is to ensure that EGUs have the option of using the ERT to generate the require XML files. The commenter said that EGUs that want to use the ERT to generate reports can do that without MATS rule language authorizing that. 
Commenter 20607 urged EPA to clarify that companies are not required to use the ERT to generate XML files. Instead, companies should have the option to use their own data acquisition systems to collect data for transmission to EPA. This will reduce the administrative burden of MATS reporting by allowing companies to continue using software they already have in place. It also allows companies the flexibility to determine the best and most efficient system of data collection for them.
Commenter 20597 opposed the suggestion that the ERT should be used to generate stack test data XML files. The idea of using the ERT and CEDRI is not preferable in any way. The commenter said going back to using ERT and CEDRI was inconsistent with the current objective. After engaging in a single data system initiative designed to replace the original patchwork of MATS reporting requirements, the commenter disagreed that EPA would suggest that continuing to use the ERT and CEDRI in addition to ECMPS. The commenter supported electronic reporting, but said the ERT is a poor tool for reporting emissions test data. The commenter appreciated that EPA has now proposed to specifically define the data elements that are to be reported for MATS in an XML format for performance tests. Sources and emission testers already have spreadsheets and software tools that they have personally verified over time. Defining the reporting in terms of an actual format allow the reporting to be easily automated and integrated using the tools sources already in place. The commenter said that reimposing the ERT and CEDRI would void these benefits and make the reporting more, not less burdensome. To streamline and automate the emission reporting, the commenter requested clearly defined reporting format rather than being required to create the data using a specific reporting tool. Ignoring the many issues with the ERT and the fact that developing an alternative to the original ERT and CEDRI reporting was a fundamental objective of the revisions to the MATS electronic reporting requirements, the ERT cannot actually produce an XML file that provides the proposed stack test data required in Appendix E.
Commenter 20595 stated the agency is recommended to remove the suggestion that the ERT should be used to generate stack test data XML files. The proposal removes all requirements to report data using the ERT and CEDRI. However, the preamble suggests that "the existing ERT can produce a single XML file that includes all of the detailed reference method information necessary for the stack test and QA test reports" under Appendix E of the proposed rule and solicits comments "on whether submitting the detailed reference method data to ECMPS will actually reduce the reporting burden on EGU owners or operators, or whether submitting the data (produced using the ERT) directly to CEDRI would be preferable." The suggestion of going back to using ERT and CEDRI is completely inconsistent with the current objective. After engaging in a "single data system initiative" designed to replace the original patchwork of MATS reporting requirements, the commenter questioned why the Agency now suggests that continuing to use the ERT and CEDRI in addition to ECMPS (the situation that existed before starting the initiative) would somehow be preferable. The ERT is a cumbersome tool that EPA needs to abandon. It does not make sense to continue to report some data using CEDRI in comparison to submitting all the data through ECMPS.
Response 2: The Agency appreciates the commenters who support using an XML format for reporting of the data for performance test reports. As mentioned earlier, the preamble identifies two ways that an XML file can be generated to meet the rule's data submission requirements: either via the existing ERT or via another program that provides the data in an appropriate XML file format. The Agency notes that while the ERT creates an acceptable XML file, its use is optional, not mandatory. The ERT is a convenient, helpful piece of software in use when electronic reporting is required by other NESHAP and NSPS; the Agency would be remiss to not include its availability for public use. Many emission testing companies have embraced the ERT, linking their proprietary spreadsheets easily to tool templates. That ability makes importing the isokinetic and other data simple since those data have already been placed in a spreadsheet. The Agency's the ERT is not cumbersome, nor will it be abandoned, nor will references to the ERT be removed from the rule; like many other software programs, once one becomes familiar with its operation, one finds it invaluable.  
Comment 3: Commenter 20609 indicated that EPA should remove the Appendix E "percent O2/CO2 correction" in section 17.35 because there are no MATS standards to which that applies. At a minimum, EPA should clarify that data elements are only required if they are required for the particular EGU to demonstrate compliance. For example, some proposed data elements, like F-Factors, flow rates, and stack diameters, would only apply if the EGU was demonstrating compliance with a particular type of standard  -  i.e., input-based or output-based. The commenter suggested EPA could avoid all of these issues by simply requiring submission in PDF format of the test report that would contain all of the information required to be recorded to support an applicable test method run. 
Response 3: The Agency thanks the commenter for his suggestions for rule changes.  The Agency agrees that percent O2/CO2 correction is not an item that needs to be reported, and it will be removed from the rule's required data element list.  
As mentioned in section 1.0 of Appendix E, the reportable electronic data elements are determined by the chosen compliance strategy; therefore the Agency believes no additional clarification on this topic is necessary. Note that some data elements may or may not be necessary to demonstrate compliance depending on this strategy. Also, that that if a data element is listed in the general provisions at §63.8(g), in the test method, in this subpart, or is required by the delegated authority, and is needed to determine compliance, then the applicable data element needs to be reported. 
Also, as already discussed earlier, PDF file submissions do not constitute electronic reporting.  
Comment 4: Commenter 20609 disagreed with EPA's assertion that the proposal will have "no significant impact on stakeholders." The commenter said that EGUs will have to spend significant resources developing new procedures and software to record in their DAHS significant amounts of previously unreported data so that they now can be reported in XML format.
Commenter 20608 stated that EPA inaccurately stated the proposed MATS data elements that would be reported electronically for performance tests and CEMS performance evaluations remain unchanged. The original requirement to submit data using the ERT is not synonymous with a requirement to provide all the data within the forms of the ERT as EPA seems to be suggesting. It only means that a source would have report the applicable required performance test data in the format that the ERT generates. MATS sources were never required to report all the data within the ERT. Rather than just assume that it has to replicate some assemblage of ERT within Appendix E (or any new XML reporting format for stack test data), the Agency should be reflecting on how best to report the data: what data needs to be reported, what portion of that data is critically required in a numeric XML format, and what data is adequately or better addressed within a detailed stack test report that can be submitted electronically in a PDF (or equivalent) file. Careful consideration will benefit both sources and streamline eventual reporting.
Commenter 20606 stated EPA's proposal will introduce significant additional burden in regard to the MATS reporting obligations, but will not result in any meaningful reduction in the level of effort associated with the electronic data elements that must be reported to EPA. This includes the electronic data elements associated with stack tests and QA test reports. Initial industry suppliers for using ECMPS as a single electronic reporting tool was largely driven by a desire to avoid using the EPA's ERT. The desire was that all required MATS electronic data entry would be handled through existing DAHS or ECMPS. However, the EPA's current proposal essentially continues to require the use of ERT or some alternate data entry scheme employed within the DAHS (assuming the DAHS vendors commit to building such functionality) of the same complexity as that found within ERT. The commenter noted additional language in the preamble suggests that the burden associated with reporting all of the electronic data elements for stack tests and QA test reports will not help EGU owners/operators to quickly identify errors which could impact the quality assured status of the associated data. EPA recognizes that errors in QA test submittals could impact the validity of the CEMS data, they failed to recognize that any electronic checks on the QA test data should be an inherent capability within ECMPS. In regard to existing 40 CFR Part 75 or Hg CEMS QA test data submitted via XML format via ECMPS, any standard electronic evaluations regarding data validity can be performed at any time after the data is imported into ECMPS, and at the latest, is performed at the time of data submittal via ECMPS. The preceding allows an EGU owner/operator to identify errors in QA test data as quickly as possible and minimize the potential loss of CEMS data in the event that an error is found within the QA test data. In contrast to this historic approach for evaluation of QA test data reported in XML format, EPA now proposes to conduct electronic checks on the data external from ECMPS. From an EGU perspective, all of the effort associated with electronic submittal of stack test and QA test data elements would be no different than submitting a hard copy or PDF version of the QA test data. Under this proposal, the EGU has no ability to directly perform the electronic checks that may be performed by EPA, will be unaware of what specific checks are being performed by EPA, and will have no assurance of a date certain by which EPA will complete their evaluation of the stack test and/or QA test electronic data. Any data evaluation process should be developed in a manner that can easily be recreated by all parties; the commenter said this proposal fails to meet that standard. 
Commenter 20597 said the proposed revisions contain a large amount of duplicative reporting that is unnecessary. This is the case with respect to RATA, PS-11, RCA and RRA test, which the proposed rule would require some of the data be reported to both the QA file as well as the new XML stack test data file. Monitor availability values are also required in the quarterly compliance report even though values are reported hourly in the emissions file. These duplicative reporting requirements should be removed. Since EPA will be collecting all this information via ECMPS, the data does not need to be reported in more than one file as the Agency may use ECMPS to populate whatever databases they may use this information. The requirement in Appendix E for sources to report the fuel usage during performance tests, RATAs, PM CEMS correlation tests, RRAs, and RCAs should be removed. Gross load and/ or heat input, which is already reported on an hourly basis, would serve as much more meaningful process indicators. The supplemental XML stack test data reporting requirements in Sections 14-21 of Appendix E are unnecessary and should be removed. The proposed quarterly compliance reports contain all information needed to assess compliance and in sure that the performance tests were performed properly. The data compliance report can be combined with the heat in put or load information already reported by sources in the quarterly emissions files to develop emission factors. For the Agency to truly evaluate the reported values a review of the detailed test report would be required. MPC would suggest this information be submitted in a PDF formant upon request. 
Response 4: The Agency disagrees with the commenters' assertions. As mentioned earlier, the Agency finds the result of implementation of this rule yields a burden reduction, as opposed to a significant impact  -  see Response 1 to Section 1.4. Also as mentioned earlier, MATS required use of the ERT for those emissions tests included in the ERT. As all MATS emissions tests are included in the ERT, all emissions tests were to have been reported using the ERT. Now that an XML schema which identifies the information already included in the ERT has been developed at industry's request, data from this XML schema can suffice for the electronic submission requirement  -  but so can data collected via the ERT. As part of the XML schema development, the Agency thoroughly reviewed the ERT, clarified it, and in some cases, pared it back. The data identified in Appendix E and contained in the XML schema was given careful consideration, and, as mentioned elsewhere, is necessary for emissions determination, calculation replications, compliance verification, and emission factor development  -  see Response 1 to section 14.2.2.  
While a commenter may claim to find not much difference between file submission via PDF or the ECMPS Client Tool, the Agency suggests expanding such a myopic view. Submitted data serve many purposes, including many that are beyond the scope of an individual EGU owner or operator. In addition, as mentioned earlier, PDF file submissions, by themselves, do not constitute electronic reporting. The rule allows a careful balance of data submission via the ECMPS Client Tool and, in cases where data are not suited for electronic reporting, e.g., EGU CEMS installation schematics, via PDF files. Those limited instances in which PDF file submissions supplement the XML schema and are acceptable can be found in section 30 of Appendix E of the rule. Finally, to the extent duplicative reporting requirements exist  -  and the Agency does not believe such requirements exist  -  having data identifiable via XML tags would simplify data collection and submission, thus reducing reporting burden.
14.3 Reference methods in sections 14-16

Comment 1: Commenter 20609 stated EGUs currently report RATA results for Hg CMS in XML format and RATA, RRA, and RCA test data in PDF format. The commenter said this information is sufficient to ensure that tests were performed correctly. For all tests involving the use of reference methods, EPA's proposal expands the required information to include detailed information on the testing company, source physical characteristics, individual traverse sample point data, and other elements specific to various test methods. Proposed App. E sections 14.2-3, 15.2-3, 17.0-21.0. The commenter objected to these additional requirements. Providing this information, much of which will have to be entered by hand, imposes significant burdens with little benefit. As with the new information that EPA proposes be reported under Appendix B, the likelihood of transcription errors is high, thus increasing the possibility that EPA may have to spend time investigating reported information to determine the source of the error (whether in transcription or in the test itself). The commenter said EPA has not provided any rationale for requiring this information other than to assert that it already is being submitted by sources only in a different format. The commenter said it assumed EPA's statement is based on the information currently requested when creating a report using the ERT. However, the content of the ERT is not a rule requirement. The content of reports is dictated by the applicable rules and methods, and EPA has previously confirmed that when the ERT requests information not required by the rules to be submitted, sources are free to ignore that request. See EPA Response to Comments on the Industrial Boiler MACT Reconsideration at 670 (Dec. 2012), EPA-HQ-OAR-2003-0058-3846. EPA must submit a new ICR for these proposed new reporting requirements. 40 C.F.R § 1320.3 (definition of "collection of information").
Response 1: As mentioned earlier, the Agency is surprised that the commenter objects to the efforts to better align the RATA requirements for CMS to the industry-requested ECMPS Client Tool. Also as mentioned earlier, the data sought by the rule is necessary for compliance determination, as well as for other activities, including calculation checks and emissions factor development. Even with the increased submission frequency of some reports  -  also necessary to better match the requirements of the ECMPS Client Tool - the Agency finds collection and submission of these data to be a burden reducer, not increaser. While the Agency agrees that the ERT need not be used when it does not include a specific test method and that clearly-identified optional information need not be reported, an EGU owner or operator remains able to use the ERT as the basis for submitting emissions testing information; of course, the EGU owner or operator is also able to use the XML schema for emissions testing and other reporting requirements.
Comment 2: Commenter 20609 said in proposed Appendix E section 16, EPA discusses the need for information from various test methods when conducting performance stack tests, RATAs, RRAs and RCA. This discussion appears to respond to a request by UARG to clarify the rule with respect to the need for collection of data using a separate reference method when the equivalent data are generated by a reference method that already is being conducted. The commenter appreciated and supported this discussion, but requested further minor clarification in the final rule. 
Specifically, in the 2011 MATS proposal, EPA proposed to require use of Method 2, 2F, or 2G to determine velocity and the flow rate of the stack gas in conjunction with performance testing for PM, non-Hg HAP metals, HCl, HF, and Hg, and for LEE testing using Method 30B. Proposed Table 5, 76 Fed. Reg. at 25,129-31. In comments, UARG pointed out that in many cases independent use of those methods is not necessary, because most of the methods specified in Table 5 to determine pollutant concentrations (e.g., Methods 5, 26A and 29) also provide measurements of stack gas velocity and subsequent computation of volumetric flow. UARG Comments on 2011 Proposed MATS Rule at 194-95 (Aug. 4, 2011), EPA-HQ-OAR-2009-0234-17775. In its response to comments, EPA stated that the rule had been amended to clarify that "the requirement of Method 2 (or equivalent) [is] only as necessitated by the calculation of the emission limit." EPA Response to Comments, Vol. 2 at 5-6 (Dec. 2011), EPA-HQ-OAR-2009-0234-20126. However, Table 5 did not itself reflect that clarification. As a result, it was not clear where EPA actually addressed UARG's concern.
In comments on the Technical Corrections Rule, UARG again requested that EPA revise Table 5 to clarify that independent application of Method 2 (or an equivalent method) is not mandatory. UARG similarly requested clarification that Method 4 does not need to be performed independently from, or in addition to, Methods 5, 26A or 29 to obtain moisture values. UARG Technical Corrections Comments at 100-101. EPA did not respond to those comments because they were outside the scope of the proposal. 2016 RTC at 124-26. 
The commenter appreciated that EPA now is addressing this concern in this proposal. Specifically, in proposed section 16.1, EPA explains that for output-based standards stack gas flow rate is needed during each run of Methods 5, 5D, 26, 26A and 30B in order to convert concentration to units of the standard, but notes that methods that require isokinetic sampling  -  like Methods 5, 5D, 26A, and 29  -  already "provide the necessary data for the Method 2 flow rate calculations." 81 Fed. Reg. at 67,084. EPA also states that for input-based standards "you must still use Method 2" to determine stack gas velocity. In proposed section 16.3, EPA explains that measurement of stack gas moisture using Method 4 also may be required for conversion to units of the standard, but notes that when using Method 5, 5D, 26A, or 29 for a "performance test" the "Method 4 moisture determination may be made using the water collected in the impingers" together with other data that must be recorded. Id. at 67,085. In contrast, because Methods 26 and 30B do not require isokinetic sampling, EPA says that when gas flow rate is needed to determine compliance, such as with output-based standards, EGUs "must make a separate Method 2 determination during each test run." Similarly, EPA says that separate Method 4 measurements must be made with Methods 26 and 30B. Id. 
The commenter supported these new provisions but believes that they could be worded more clearly. For example, when EPA refers to "Method 2 flow rate calculations" "us[ing] Method 2," and "the Method 4 moisture determination," the commenter believed that EPA means the data collected under Method 5, 5D, 26A, or 29 that is equivalent to what is collected under Methods 2 and 4. The commenter requested that EPA confirm that and consider using terms like "Method 2-equivalent" and "Method 4-equivalent" data, procedures, or determinations. In addition, in section 16.3, the commenter was concerned that where EPA says that Method 4 determinations may be made under Methods 5, 5D, 26A, and 29 "for the performance test," someone might interpret that to suggest that the same is not true for other types of tests (like PM CEMS correlations or HCl CEMS RATAs). To avoid that possibility, the commenter suggested that EPA remove the phrase "for the performance test." 
Response 2: The Agency appreciates the commenter's suggestions and notes that every isokinetic method (i.e. Method 5, 26A, or 29) includes the use of Method 2 and Method 4 for determining average stack gas velocity, (vs) and moisture content. The data collected in the performance test must be used for the calculations of the emissions; data may not be substituted from any other source or method. With respect to this issue, the rule language in sections 16.1 and 16.3 has been clarified to avoid the apparent confusion the proposed wording caused.
Comment 3: Commenter 20609 requested that EPA consider broadening its discussion to authorize collection of the necessary data using a certified stack monitor. Although integral to Methods 5, 26A and 29, Method 2-equivalent flow measurement data can lead to volumetric flow calculations that are biased high when the stack flow has significant non-axial components. EPA recognized this problem and promulgated Methods 2F, 2G and 2H in 1999. Most EGUs affected by the MATS Rule are also subject to the CEMS requirements codified in Part 75 and are required to install and certify continuous stack flow monitors. Many EGU owner/operators have elected to choose among Methods 2F, 2G and 2H when certifying their stack flow monitors in order to obtain as accurate volumetric flow data as possible. As a result, the commenter requested that EPA consider revising the rule to accept use of volumetric flow data from continuous stack flow monitors that are certified in accordance with Part 75 requirements. The flow data should not consist of the hourly averages that are routinely submitted to EPA's Clean Air Markets Division. Rather, the volumetric flow rate used would be the average value computed from DAHS-stored data over the time period of each stack test run. Although EPA already is considering such a CEMS alternative in response to a source-specific alternative test method request, authorizing such a procedure in this rule would provide EPA the opportunity to build into the rule electronic reporting data elements to support use of the alternative.
Response 3: In April and May of 2016, the Agency received a request under §63.7(e)(2)(ii) for use of an affected facility's flow monitoring system to measure stack gas velocity and volumetric flow rate in lieu of Method 2, 2A, 2C, 2F, 2G, or 2H when conducting quarterly testing to demonstrate compliance with the MATS Rule. In the decision issued on December 6, 2016 to disapprove that request, the Agency explains that the reference test methods provide a measurement more representative of the source at the time of the performance test than the installed flow monitoring system for two reasons: (1) use of the multipoint traverse and (2) the fact that the accuracy of the facility's flow monitoring system is only verified, to most, twice a year during the relative accuracy test audit. For these reasons, the Agency has not revised the rule to authorize use of the continuous flow monitoring data for quarterly testing.
14.4 Section 22
Comment 1: Commenter 20606 provided commenter on Appendix E section 22, specifically regarding the other information for each test. The commenter said in the examples of test information to be submitted via PDF versus XML format, the EPA references "stack testers' credentials". The meaning of the preceding reference is not clear. The only possible reporting under MATS for which there is a specific "credential" is the Air Emission Test Body (AETB) requirement for RATAs of 40 CFR Part 75 monitoring systems. However, the preceding "credential" information is already submitted via ECMPS and need not be submitted via PDF. The EPA should strike the reference "stack testers' credentials".
Response 1: The Agency agrees that the AETB requirement applies only to Part 75 monitoring systems. Although Hg, HCl, and HF CEMS are not Part 75 monitoring systems, the Agency notes that an SO2 monitoring system that is used as a surrogate for HCl under MATS must conform to Part 75 requirements. Therefore, any RATA report for the SO2 monitoring system must include evidence that a "Qualified Individual" was on-site during the test. No change was made to the rule. 
Comment 2: Commenter 20612 supported submitting detailed performance test report electronically as PDF files. However, EPA should remove its proposed list of requirements for the PDF stack tests reports required under Table 8 and Section 22 of Appendix E and simply reference the stack test requirements included in the recently finalized revisions to §60.8(f) and §63.7(g). Otherwise, the difference between the content lists will introduce inconsistency between rules. For Appendix E, the commenter suggested: "22.0 For each test, you must submit a detailed test report for each test in PDF format that meets the requirements of §63.7(g) as in addition to the XML data file required by this appendix." The commenter said this revision would also replace the incorrect notion conveyed by Agency's proposed language that the PDF reports would include the test information that was somehow "incompatible with electronic reporting." Submitting the test report in PDF format through ECMPS would, in fact, be electronic reporting. The simple truth that some information may be better suited to present in a PDF file rather than XML does not detract from the fact that reporting such information electronically in that format provides ready access to the information in the full context of the report. The commenter advocated that the rule should clearly require sources to submit a complete, detailed PDF stack test report rather than the unclear list of items in the originally proposed language. Providing a complete stack test report as a PDF file would give sources an efficient way to satisfy the requirements of §63.7(g), would allow for complete transparency, would document a full record that can be preserved electronically, and would give agency personnel ready access to the information in the full context of the whole report in situations where a more detailed review is deemed necessary. In addition, the commenter said to avoid confusion, the proposed reporting in Sections 14-21 of Appendix E should not be considered a "test report" but only a supplemental collection of XML data reported in addition to the complete test report. The complete test report should meet the requirements of §63.7(g) and provide any other data that the source and/or testers believe is necessary to fully document the details, performance, and results of the test. The PDF version of the test report should document the items in §63.7(g), but it is not necessary that all these items be reported in XML file since many of the items are more appropriately documented in the full PDF version of the test report.
Response 2: The Agency finds the commenter's suggestions inconsistent. On one hand, the commenter has stated that the supplemental XML stack test data reporting requirements are unnecessary for compliance but on the other hand, the commenter maintains that submitting a complete PDF file which would contain all stack test data  -  including the XML stack test data - would provide full context for activities such as compliance determination. The commenter cannot have it both ways, especially when the commenter seeks elimination of many of the stack test data elements. The rule will maintain submission of the Agency-developed, pared-down to the essentials version of test report, which was tailored for use by EGUs, in addition to other material best submitted via a PDF file format (see section 30 of Appendix E).
Note that, as mentioned before, the commenter's suggestion that submission of PDF files constitutes electronic reporting is incorrect; therefore, supplying such data in that manner only remains incompatible with electronic reporting. However, the commenter and others are welcome to submit PDF files of their test reports in addition to the rule's requirements.

The Agency agrees that current language in section 30 of Appendix E could be confusing with respect to the requirements of §63.7(g); therefore, the introductory part of that section will be revised to say: "You must provide for each test included in the XML data file described in this appendix with supporting documentation in a PDF file submitted concurrently with the SML file, such that all  the data required to be reported by §63.7(g) are provided. That supplemental data includes but is not limited to diagrams. . ."
Comment 3: Commenter 20594 stated EPA must provide further guidance in section 22 regarding other information for each test. The commenter said Appendix E asks for, "All information pertaining to the test that is ordinarily included in a comprehensive test report . . . ." Id. at § 22.0. Since testers and report information can vary, the information requested by Appendix E will differ. In addition, Appendix E begins with a list of "other information" but says "including but not limited to," which opens up the data metric to include an unspecified amount of information for each test. EPA must list the testing information that it is requiring for MATS.
Response 3: The Agency agrees with the commenter that the information pertaining to each test can vary, depending on the method(s) used. However, for that very reason, the Agency maintains that the type of information described in section 30 of Appendix E should not be overly prescriptive. Apart from the new rule language, described above, which ensures that all of the information required by §63.7(g) is submitted, emission testers are best able to identify what other specific information may be needed to be included in each test report.  
15 Statutory and Executive Order Reviews
15.1 EO 12866: Regulatory Planning and Review; and, EO 13563: Improving Regulation and Regulatory Review
Comment 1: Commenter 20594 stated stakeholders have not had a meaningful opportunity to comment. The Proposed Rule was published in the Federal Register on September 29, 2016. Comments are due on October 31, 2016. The comment period is only 30 business days. EPA recently extended the comment period 15 additional days after receiving numerous requests from prospective commenters. Executive Order 12866 directs a minimum of 60 days for comments on any regulatory action to allow for meaningful public comment. 
The commenter said EPA proposed changes to a highly technical reporting schema that are much more than clerical. Substantive changes include a different frequency in reporting, type of data to be reported, omission/inclusion of data, and other issues discussed infra. These changes require that sources have more time to review and comment on the Proposed Rule to evaluate its practical and financial impacts. 
The commenter disagreed with EPA's statement, "While the changes in this proposed rule will take time to implement, no significant impact on stakeholders is expected." 81 Fed. Reg. 67062, 67064 (Sept. 29, 2016). EPA explains, "Only the reporting format and mechanism would change." Id. The commenter disagreed these assertions could be accurate if the actual information to be reported and frequency of reporting is proposed to change by the Proposed Rule. The two years it took EPA to develop the Proposed Rule after EPA's first notification that it would streamline MATS reporting requirements is indicative of the substantive nature of the Proposed Rule -- See 79 Fed. Reg. 68795 (Nov. 19, 2014). 
The commenter added the Executive Order also requires EPA to submit the Proposed Rule to OMB before issuing the proposal. Exec. Order No. 12,866, 58 Fed. Reg. 51735 (Oct. 4, 1993). This requirement applies to any significant regulatory action. The commenter said this rule defines significant regulatory action as any rule that may have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities. EPA conclusively states that the Proposed Rule "is not a significant regulatory action" and did not submit the rule to OMB for review. 81 Fed. Reg. at 67069. The Proposed Rule does not identify the costs of the action. The commenter suggested this rule may be significant enough to warrant OMB review due to additional reporting obligations.
Response 1: The Agency disagrees with the commenter's assertions regarding not having a meaningful opportunity to comment on the rule. Executive Order 12866 (EO 12866) does not mandate a minimum comment period; rather, it requires each agency to develop its own rules, regulations, or procedures to provide the public with meaningful participation in the regulatory process, and it says each agency should afford the public a meaningful opportunity to comment on any proposed regulation, which in most cases should include a comment period of not less than 60 days (italics added for emphasis). Relevant to this rulemaking, the Agency has defined the threshold which triggers the EO 12866 protections as having an annual effect of on the economy of $100 million or more or having an adverse effect in a material way on the EGU sector. As mentioned earlier, merging the separate electronic data reporting systems  -  which is being pursued at industry's request  -  is not a significant regulatory action, as it yields an overall burden reduction of about $4.2 million per year in reporting and recordkeeping requirements, which amounts to about one-tenth of one percent of projected net utility income for 2015 (see 81 FR 67070, September 29, 2016 and page 2-15, https://www3.epa.gov/ttn/atw/utility/mats_final_ria_v2.pdf). Note that the Agency has been forthright regarding the significance level of this rulemaking; see the not significant determinations in preambles at 80 FR 15514 (March 23, 2015), 81 FR 67069 (September 29, 2016), and 82 FR 16738 (April 6, 2017). Because the rulemaking is not significant, no 60 day comment period is required. Even without a significant rulemaking, the Agency afforded commenters 45 days to provide their comments.
The Agency does not agree with the commenter's suggestion that additional time beyond the first 15 day extension was required for meaningful comment to be provided: as evidenced by the responses, other commenters were able to review the proposal thoroughly. 
As noted earlier, the Agency's efforts to better align rule reporting with the industry-requested ECMPS Client Tool result in increased submission frequency and in provisions for PM, HCl, and HF CEMS, as well as PM CPMS; even so, annual burden is reduced. The Agency notes that the commenter should not look upon the delay of this rulemaking as indicative of some particular difficulty in merging the electronic monitoring systems; Agency activity  -  including rulemaking  -  is a function of priority and resources. 
As mentioned earlier, the Agency agrees that rules categorized as significant must undergo OMB review; however, this rule is not significant. Nonetheless, the Agency may choose to obtain OMB review. The preamble of the promulgated rule will describe what review, if any, OMB provided.   
16 Other
Comment 1: Commenter 20612 offered a suggestion for clarification and revision under 63.10010, which was not amended with this proposal. The commenter said §63.10010(h)(6)(i) and (ii), §63.10010(i)(5)(i)(A) and B, and §63.10010(j)(4)(i)(A) and (B) make reference to reporting information in "your annual deviation reports." However, the rule does not require sources to submit an annual deviation report. The commenter interpreted that EPA intended this information to be included in the semi-annual compliance reports and suggested that the rule should be revised to correct this error.
Response 1: Upon consideration of the comments, the Agency agrees that clarifications would be useful. Paragraphs (h)(6), (i), and (j)(4) of §63.10010 of the rule have been revised. References to "annual deviation reports," which are not required by Subpart UUUUU, have been removed from all of these paragraphs. Note that deviations are required to be reported in the compliance reports.
Comment 2: Commenter 20609 stated EPA needs to provide clarification of the testing requirements when switching between input- and output-based standards. The commenter noted EGUs must identify in their NOCS the emission limit with which they have complied. 40 C.F.R. § 63.10030(e)(2). In the 2016 Technical Corrections Rule, EPA added to § 63.10030(e) a provision explaining how EGUs can switch from an input- to an output-based standard. Id § 63.10030(e)(7)(iii). The provision requires (1) submission of a request at least 30 days before the switch, (2) a demonstration of compliance with both limits through "performance stack tests results" conducted within 30 days prior to submission, and (3) revision and submission of new monitoring and/or averaging plans. The commenter provided background on past comment and response statements. In response to UARG comments on the proposed rule, EPA agreed that "stack test" results are not necessary for those EGUS demonstrating compliance using CEMS and said that the rule would be revised to remove the words "stack test," so that a stack test would only be required for any EGU using quarterly stack testing to comply. 2016 RTC at 110, 111. EPA also said that it was revising the requirement to allow tests to be conducted up to 45 days prior to submission of the request, in order to allow time to get testing results. Id. at 111. Neither revision was reflected in the final rule language. EPA should correct those oversights when it finalizes this rule.
Response 2: As mentioned in response to comment 2 of section 5.6.3, the rule allows data from CMS, as well as stack test data, to be used to justify switching from one standard to the other. Note that the rule allows EGU owners or operators to use data from as far back as 45 days prior to the submission date of the request for the switch, as part of the request.  See section 63.10030(e)(7)(iii)(A)(3). 

The Agency also notes that section 63.10030(e)(7)(i) has been removed from the rule and reserved because the requirement is for an initial Notification of Compliance Status to include summarized results of annual and triennial performance tests which have not yet been performed is incorrect; the requirement to submit these test summaries belongs in §63.10031, not §63.10030; and text similar with that of §63.10030(e)(7)(i) does, in fact, exist in §63.10031(c)(7), which requires the annual and triennial test results to be summarized in the semiannual compliance reports. Be advised that when the semiannual reports are phased out in the last quarter of 2019, summaries of these tests must be included in the quarterly compliance reports under §63.10031(g).  
Finally, a note has been added to §63.10030(e)(7)(iii), clarifying that requests to switch from one standard to another are made subsequent to, and are not part of, the initial Notification of Compliance Status.  
Comment 3: Commenter 20609 stated EPA proposes to add a requirement for EGUs, including those participating in averaging plans, to report rolling average emission rates for Hg. Proposed § 63.10031(f)(2) and App. E section 9.0. EPA also proposes to require reporting of additional reference method data for all performance tests, including "Hg LEE tests." Proposed App. E section 14.1. The commenter said in the 2016 Technical Corrections Rule, EPA revised § 63.10009(a)(2) to make clear that the alternative 90-group boiler operating day rolling average standard for Hg is an option and that the 30-group boiler operating day rolling average standard also may be used. Unfortunately, the Technical Corrections Rule introduced two inadvertent errors that warrant correction before implementation of these new more detailed reporting requirements. In the proposed revisions to the section, EPA inadvertently created an error by listing only CEMS and manual performance testing, and not including sorbent trap monitoring, among the options for demonstrating compliance with the 30-group boiler operating day Hg standard. When UARG asked EPA to correct that oversight, EPA responded: "We agree that sorbent trap monitoring data should have been added to the option to use a 30-boiler operating day rolling average for Hg. The commenter disagreed that subpart UUUUU does not provide a manual performance testing option for Hg. Method 30B is listed in both Table 2 and Table 5. Section 63.1009(a)(2) is revised to include sorbent trap monitoring." (2016 RTC at 88). However, rather than list sorbent trap monitoring among the options, EPA changed "manual performance testing" to "manual performance (LEE) testing." The commenter stated this revision is confusing because LEE testing is not conducted using continuous sorbent trap monitoring under Appendix A, and § 63.10009(a)(ii) appears to prohibit EGUs with LEE status from participating in facility averaging. The commenter suggested EPA should use this opportunity to clarify § 63.10009(a)(2) by listing "sorbent trap monitoring" as an option for calculating rolling averages under an averaging plan, and removing the reference to LEE testing. The commenter was concerned that if this is not corrected, ECMPS might be programmed to expect Method 30B test data from EGUs participating in averaging plans, rather than data from the sorbent trap monitoring systems required for those EGUs under Appendix A if they do not opt to use Hg CEMS. 
Commenter 20609 added while making clear in § 63.10009(a) that the 30-group boiler operating day rolling average standard is available for Hg, EPA did not revise the provision that sets out the equation for calculating such averages. Specifically, the applicability of § 63.10009(b)(2), which provides equation 2a for calculation of the weighted 30-boiler operating day rolling average emission rates, is limited to "pollutants other than Hg." Although commenters on the proposed rule asked EPA to make that correction, EPA deemed the comment outside the scope of the rulemaking. 2016 RTC at 94-95. Because EPA now has proposed that EGUs report 30-group boiler operating day averages for Hg, EPA should take this opportunity to make clear in the rule that the equation in § 63.10009(b)(2) should be used to calculate those averages.
Response 3: Upon consideration of the commenter's suggestions, the Agency agrees that clarifications would be helpful. Paragraph (a)(2) of §63.10009 of the rule has been revised to clarify the types of data that may be used to determine weighted average Hg emission rates (WAERs). In addition, the rule now allows data from Hg CEMS, sorbent trap monitoring systems, and, consistent with paragraph (a)(1)(ii), the results of quarterly Hg emission tests for oil-fired EGUs (but not annual LEE tests) to be used for Hg emissions averaging. Finally, the words "pollutants other than Hg" have been removed from the introductory text of paragraph (b)(2) of §63.10009 of the rule, and in the rule's Equation 2a nomenclature, the words "or sorbent trap monitoring" have been added to the definition of "Heri".  
Comment 4: Commenter 20609 stated Sections 63.10010(h)(6)(i), (i)(5)(A) and (B), and (j)(4)(i)(A) and (B) of the current rule specify what data from PM CPMS, PM CEMS, and HAP metal CEMS must be excluded from compliance determinations. The provisions currently require reporting of any "monitoring system malfunctions or out of control periods" in "annual deviation reports," and reporting of "any monitoring system qualify assurance or quality control activities " in the semi-annual compliance report "under § 63.10031(b)." The commenter said that EPA should revise § 63.10010(i) to simply refer to Appendix C, and remove the other details that might conflict with the more detailed requirements. In any event, to ensure that the rule remains current after January 1, 2018, EPA should replace any references to "annual deviation reports" and "under § 63.10031(b)" in those provisions with a requirement to report "under § 63.10031," which would include the existing reports as well as the new appendices.
Response 4: As mentioned earlier, upon review of the commenter's suggestions, the Agency agrees some clarifications are warranted. Paragraphs (h)(6), (i), and (j)(4) of §63.10010 of the rule have been revised. Likewise, references to "annual deviation reports," which are not required by Subpart UUUUU, have been removed from the rule. Note that deviation reporting remains a component of compliance reports.
Also note that in paragraphs (h)(6)(i), (i), and (j)(4)(i), the requirement to report QA/QC activities per §63.10031(b) has also been removed from the rule because §63.10031(b) concerns the schedule for semiannual compliance reports submissions, not QA/QC activities; this removal addresses the commenter's recommendation to replacing it with a more general reference to §63.10031. While considering this comment, the Agency found inconsistency between the text of §63.10010(h)(6)(i), (i)(5)(i)(A), and (j)(4)(i)(A) and (B) that calls for reporting of out of control periods in annual deviation reports and the and the text in §63.10010(h)(7), (i)(5)(ii), and (j)(4)(ii) that calls for out of control period reporting "upon request." The Agency believes that these latter paragraphs are incorrect because reporting of QA test results is mandatory for all CMS. Therefore, three revisions to rule have been made.  First, the last sentence in paragraphs (h)(6)(i), (j)(4)(i)(A), and (j)(4)(i)(B), containing the incorrect reference to §63.10031(b), has been removed. Second, paragraphs (h)(7) and (j)(4)(ii) now require the QA/QC activities for PM CPMS and HAP metals CEMS to be reported. Third, a new paragraph, k, has been added to §63.10031, with a corresponding summary in Table 8, requiring the QA/QC activities for PM CPMS and HAP metals CEMS to be reported quarterly, in a PDF file. These reports will be due within 30 days after the end of each calendar quarter, starting with a report for the first quarter of 2021. Reporting in a PDF file is appropriate in this instance because there are no standardized QA test procedures for these CMS in the Code of Federal Regulations; their QA test requirements are found only in source-specific MATS monitoring plans, and will likely vary from source-to-source.
Finally, for consistency with new Appendix C, the various requirements for certification, operation, maintenance, etc. of PM CEMS that were specified in paragraph (i) of the rule have been replaced with cross-references to the applicable sections of Appendix C.  
17 Request for Extension
Comment 1: Commenter 20594 stated that it appreciates EPA's consideration of its comments [on the proposed rule amendments]. The commenter urged EPA to provide sources additional clarification and modifications of some areas of the proposed rule [Note: These are summarized in the appropriate sections of this document]. The commenter stated that it further reserves the right to comment on additional areas of the rule given that the time for comment has not been adequate.
Response 1: The comment is moot, as the Agency extended the public comment period by 15 days, in response to the requests for an extension (see 81 FR 75365, October 31, 2016).
Comment 2: Commenters 20587, 20590, and 20589 all requested an extension of the current October 31, 2016 deadline for public comment on the MATS Completion of Electronic Reporting Requirements proposal that was published in the Federal Register on September 29, 2016 (81 Federal Register 67062). 
Commenter 20587 stated that the proposal represents a substantial change to how the data are reported under the MATS Rule. Given the scope of the changes, the proposal warrants additional time for review in order to provide meaningful comments to this complex rule. In the preamble of this proposal, the Agency appears to categorize these changes as simple format changes, stating that "no significant impact on stakeholders is expected." The commenter stated that it disagrees with this categorization. The proposed changes restructure the way much of data is reported and, in some cases, add new reporting requirements and accelerate the frequency that the data are reported. In light of these considerations, the commenter requested that EPA extend the comment period for this proposed rule by at least an additional 30 days. This extension will allow time needed to thoroughly review how the changes will impact current operations and procedures, as well as how data compiled by third parties may need to be modified, and provide the most meaningful comment on this proposal.
Commenter 20590 stated that it made this request on behalf of a variety of electric utilities that are participating in an electric utility funded workgroup that is focused primarily on addressing reporting issues related to the MATS Rule. The commenter stated that it supports the Agency's attempt to simplify and streamline MATS reporting requirements, but the proposal represents a substantial change to how the data are reported under the MATS Rule. Given the scope of the changes, the proposal warrants additional time for review. The proposed changes restructure the way much of the data is reported and, in some cases, add new reporting requirements and accelerate the frequency that the data are reported. In addition to new reporting requirements for HCl and HF CEMS in Appendix B of the MATS Rule, the proposal includes three completely new appendices which add appreciable monitoring and data reporting.
Commenter 20590 stated that while the Agency has provided draft reporting instructions in the docket for the new quarterly MATS compliance reports, the EPA has not provided similar draft instructions for the supplemental test data reports in Appendix E, which is largely just a list of data elements that sources must report. Lacking draft instructions for the proposed supplemental test data reports, it is not always clear precisely what information would be required based on the brief description in the proposed rule or exactly how that information might be reported in various circumstances. This void will make it harder to determine the underlying intent of the proposed requirements and more difficult to provide meaningful comments. 
Commenter 20590 suggested that the complexity of the rule revisions can also be inferred from the amount of time that the Agency has spent on developing the proposal. It has been about two years since EPA indicated that it would "promulgate an additional data reporting revision to the MATS rule in the second phase" of the effort to streamline the MATS reporting requirements that it initiated by proposing interim reporting procedures in November of 2014 (79 Federal Register 68795). After allowing its staff such a long period to develop the proposal, it would seem inconsistent for the Agency to shortchange the rulemaking process by not providing adequate time for thorough public review and comment. 
Commenter 20590 stated that the timing of EPA's proposal also exacerbates the difficulty for electric utilities to provide the thorough review within the current 30-day comment period. The current comment period coincides with the 30-day window during which sources must compile and submit quarterly reports under the MATS Rule as well as the ARP and CSAPR. The situation is compounded by the fact that comment period also coincides with the end of the MATS compliance demonstration deadline for facilities that received a one-year MATS compliance extension. Thus, in addition to being occupied with other quarterly reporting requirements, many sources will be busy reviewing and preparing the initial MATS performance test reports, the initial CEMS certification reports and the notifications of compliance status reports associated with those activities. Given their expertise, the personnel involved in preparing the reports are the same individuals who are needed to review and provide comments on proposed changes to the MATS electronic reporting requirements. The commenter stated that in light of these considerations, it requests that EPA extend the comment period for this proposed rule by 30 days (November 30th). This extension would allow sources and the public the time needed to review the changes thoroughly and to provide meaningful comment on this proposal. The commenter stated that it believes the extension will benefit EPA's final rulemaking.
Commenter 20589 stated that it is an ad hoc, unincorporated association of individual electric companies and industry trade associations that participates on behalf of its members collectively in administrative proceedings under the Clean Air Act, and in litigation arising from those proceedings and that it appreciates the opportunity to submit comments on this MATS proposal, but requires additional time to evaluate and develop those comments. Although the Agency may view its proposal as a simplification of current requirements, the proposal is quite complex. It involves almost a complete restructuring of the manner and schedule under which facilities will report compliance information to the Agency. It also contains a significant amount of new detail not only in the sections it revises and in the three new appendices it would add, but also in the draft reporting instructions EPA has made available in the docket. EPA's final action will benefit from submission of comments that are thorough and well considered.
Commenter 20589 stated the thirty day comment period EPA has provided is not adequate. That inadequacy is compounded by the complete overlap of the current comment period with deadlines for several other complicated tasks that must be performed by the same personnel who are needed to evaluate and assist in preparation of comments on the current proposal. Specifically, as the Agency is well aware, the third quarter 2016 reporting period for the ECMPS begins on October 1 and runs through October 31, 2016, the day now established as the deadline for comments on this proposal. During that period, the personnel who would be responsible for the new ECMPS reporting EPA has proposed will be fully occupied preparing and quality assuring quarterly reports to meet their existing reporting requirements, which include not only MATS data elements, but also reporting under the ARP and CSAPR. In addition, the comment period also coincides with the end of the 180-day period for completion of compliance demonstrations for those facilities that operated under a one-year MATS compliance extension. As a result, in addition to ECMPS reporting, many facilities also will be busy preparing and submitting reports associated with those activities, including MATS performance test reports and notifications of compliance status. In order to allow companies time to provide meaningful comment on this proposal, without compromising their compliance with existing reporting requirements, the commenter requested that EPA extend the current comment deadline by 30-days.
Commenter 20589 stated that in the event the Agency does not provide the requested extension, it hereby requests that the Agency hold a public hearing on its proposal and, in compliance with Clean Air Act § 307(d)(5), hold the comment period open for 30 days following that hearing. If the Agency holds such a hearing, one of Commenters 20589 representatives will attend and present testimony. However, if EPA provides the requested extension of the current comment deadline, the commenter will focus its efforts on its written comments and withdraw its request for a hearing.
Response 2: See response to comment 1 of section 17.  With regard to other comments, the Agency is disappointed that the commenters who represent the industry which requested changes to make the MATS electronic reporting system more like the existing ECMPS Client Tool suggest that the revisions are now burdensome, rather than consistent.  In keeping with our pledge for industry-requested consistency, we will continue on the path for one electronic reporting system.
The Agency notes that EGU owners or operators are required to provide the information in Appendix E regardless of whether draft  -  or final  -  reporting instructions have been prepared.  Even so, we recognize that such instructions can ease implementation and enhance uniformity across responders; therefore, we have prepared and posted draft reporting instructions for Appendix E. The Agency finds that commenters who believe this rule's alleged complexity has delayed its promulgation are incorrect; Agency activity  -  including rulemaking  -  is a function of priority and resources.  
While the Agency understands the effort needed to meet reporting obligations, EGU owners or operators have known about these requirements for over 4 years.  Prudent EGU owners or operators would have been practicing and delegating report review, development, and submission and would be prepared to meet those obligations.
Comment 3: Commenter 20609 provided comments on the expiration of the existing transition period on April 16, 2017 (§ 63.10031(f)). The commenter stated that as a result of the expiration of this transition period, if EPA does not promulgate another rule revision that is effective prior to that date, electronic reporting will revert back to the two system structure originally adopted in the 2012 rule. In the March 2015 final rule, EPA reasoned that this deadline was necessary in order to ensure that data submitted in the future "is consistent with the database accessibility that is associated with information reported in structured XML formats" in the event the conversion to ECMPS cannot be completed. 80 Fed. Reg. at 15,512. 
Commenter 20609 stated if EPA plans to move forward with the concepts in its current proposal, which the commenter assumes it will, EPA must take the time to carefully review comments and make adjustments necessary to make the rule clear and consistent. Although no rule is perfect, EPA cannot expect EGUs to comply if the rule is internally inconsistent or ambiguous. If EPA does not have time to complete those tasks and get a final rule (taking into account all comments) in effect prior to expiration of the current transition period, EPA must be prepared to extend it. EPA has not, and cannot, justify requiring EGUs to both spend resources to implement an entirely new expanded reporting program while at the same time temporarily complying with a different system with which EGUs have little familiarity and no process in place to implement. In short, if EPA needs more time to finalize the rule, EPA must extend the transition period for the same reasons it created that period in the first place. 
Commenter 20609 stated that although all aspects of this proposal are not noncontroversial, it believes that extension of the transition period to allow for orderly finalization of this rule would be noncontroversial. As a result, that could be accomplished using the same direct final rule procedure EPA used in 2015. The commenter stated that it also does not believe that continuation of the transition period for a few months would significantly impact the quality of the data available to the public.
Response 3: See the response to comment 1 of section 17.  The rule further extends the PDF reporting through December 31, 2020, to allow sufficient time to develop and test the necessary software, and to draft reporting instructions.
Comment 4: Commenter 20609 provided comments on the period for implementation of new ECMPS reporting requirement(s). Commenter 20609 stated that once a final rule is issued, both EGUs and EPA will need time to prepare software and procedures necessary to implement the new XML formats. Although EPA does not commit to finalizing a rule by any date, EPA proposes to require submission of the new reports and information in XML format starting on and after January 1, 2018, with one exception. That exception is that a report covering a period or event that occurred before that date would still be submitted in PDF format (e.g., a semi-annual report for the period July 1 to December 31, 2017 would still be submitted in PDF format by the current deadline of January 31, 2018. 81 Fed. Reg. at 67,065.
Commenter 20609 stated that in the past when EPA has implemented major changes to Part 75 or ECMPS reporting requirements, the Agency has provided a significant period of time for implementation. For example, when EPA transitioned from the old Part 75 reporting format to the current XML format in ECMPS, EPA made the new format optional for an entire year to allow for a phased transition. Similarly, when EPA finalized the MATS rule, EPA began circulating and taking informal comment on draft reporting instructions and XML schema for the new ECMPS reporting requirements in April 2013, a full two years before EGUs were required to start submitting quarterly emissions reports through ECMPS for MATS.
Commenter 20609 stated that this rule, when finalized, will require similar resources and time for implementation. The implementation period must be sufficient to cover not only the software programming by DAHS vendors, and EPA's ECMPS contractor, but also the development by EGUs of procedures to ensure that all of the recorded information (including information recorded by hand) is properly input into the DAHS and quality assured. And, EGUs have to do that while still complying with the existing reporting requirements of the rule. 
EPA has essentially proposed an implementation timeframe that would provide EGUs, EPA, and their software vendors the equivalent of 3 calendar quarters before the beginning of the calendar quarter in which they will begin complying with the rule's new requirements. That assumes that EPA will finalize the rule, and other documentation, before the end of the current transition period, which EPA has not proposed to alter. If EPA cannot do that, EPA also will need to extend the starting date for the new requirements to preserve the proposed implementation period. However, the commenter is concerned that the amount of time EPA proposes will not be enough time even if the rule and material are released before April 16, 2017. For example, that timeframe does not appear to provide time for solicitation of feedback on the remaining reporting instructions, or for discussions to resolve rule ambiguities or inconsistencies that may be discovered in the final rule. Moreover, if EPA is not able to complete revisions to the ECMPS Client Tool in time for EGUs to test it, EGUs will be forced to use a tool that is not ready. As a result, the commenter requested that EPA lengthen the implementation period to a minimum of 6 calendar quarters. 
Finally, to ensure that the implementation period is fully accommodated even if there is a delay between completion of the rulemaking package and its signature by the Administrator, Commenter 20609 suggested that the rule include a start time that is expressed as a number of full calendar quarters after the effective date of the final rule rather than the January 1, 2018 date EPA proposed. 
Response 4: See response 1 to section 17 and response 3 to section 17.  
18 Incorporation by Reference
Comment 1: Commenters 20595, 20598, and 20603 supported comments submitted by the Utility Air Regulatory Group (UARG)-docket ID 20609. 
Commenters 20593, 20595, 20598, 20600, 20601, 20603, and 20608 supported comments submitted by RMB Consulting and Research, Inc-docket ID 20612.
Response 1: See previous responses to those comments.



