Chapter 11
Monitoring and Compliance
Contents
11.1	Initial Compliance Demonstration	3
Initial Compliance Determination	3
Hybrid Approach	5
Should Conduct Initial Performance Testing	5
Should Not Conduct Initial Performance Testing	6
Best Approach for Initial Performance	8
11.2	Proposed CO2 Monitoring Methods	9
11.2.1	The CEMS Option	9
Allow O2 Monitoring	11
Other Monitoring	11
11.2.2	The Heat Input and Fuel Factor Option	12
Recommend fuel flow monitoring	12
Support Use of Appendix D and Appendix G	15
11.3	Monitoring of CO2 Mass Emissions from Integrated Equipment	16
11.4	Monitoring and Reporting of Gross Generation	16
11.5	Quality Assurance and Quality Control (QA/QC)	18
11.6	Additional Measurement Requirements	20
11.7	The Monitoring Plan	24
11.8	Schedule for Monitoring System Certification	25
11.9	Annual or Multi-Year Compliance Periods	25
Annual vs. Multi-Year Basis	25
Support Multi-Year Compliance	26
Oppose Multi-Year Compliance	27
Support 12-month Compliance	27
Include an Option for 12-month Compliance	28
Not Set a More Stringent Standard for Multi-Year Compliance	28
Include Stringent Measures for 84-month Compliance Option	29
NGCC	30
Hourly, Monthly, Daily	31
Daily Violations	32
Format of the Standard, Total Emissions over Total Electrical Output	33
Include 0 for Emission Rates if Not Operating	34
Compliance Issues with 3-Year Applicability Threshold	34
Compliance Issues Related to Applicability	34
11.10	Valid Data, Backup Monitors, Missing Data Procedures, and Minimum Data Availability	35
Valid Data	35
Backup Monitors	36
95 Percent Data Availability	37
Missing Data	37
11.11	Violation of Monitoring Requirements	38
11.12	Carbon Capture and the Calculation of lb/MMBtu Emission Values	40
11.13	Part 75 Monitoring Methods	41



Initial Compliance Demonstration
Initial Compliance Determination
Commenter 9514 stated EPA must take steps to improve near-term compliance.  Commenter 9514 stated EPA should establish a requirement for an early initial compliance demonstration.
Commenter 9514 stated that ignoring emissions that predate the applicability threshold is unwarranted. Commenter 9514 stated that one impact of the extended averaging period for the applicability determination is a significant delay in the date by which certain sources must comply with the proposed standards. Commenter 9514 gave an example where, for a source that exceeds the applicability threshold in January, this language appears to allow the source to ignore all emissions prior to that month, with the source's first 12-month compliance period concluding in December of that year, so the proposed rule would not require a compliance determination until up to 12 months after the applicability threshold is crossed.
Commenter 9514 stated there is no reason to wait up to a full year after applicability is triggered to make the first compliance demonstration, and EPA should require the initial compliance determination in the first month after a source crosses the applicability threshold. Commenter 9514 stated the initial compliance dates should reflect that most plant operators will know the ultimate applicability status of their sources well in advance of crossing the applicability threshold. Commenter 9514 stated EPA should factor into the initial averaging period to determine a source's compliance the operating days that predate the point at which the source crosses the applicability threshold. Commenter 9514 further stated that including in a compliance determination the emissions that predate the point at which a source crosses the applicability threshold is consistent with Congress's timing scheme for applying NSPS to new sources. Commenter 9514 stated the CAA defines a "new source" for the purposes of section 111 as "any stationary source, the construction or modification of which is commenced after the publication of regulations (or, if earlier, proposed regulations) prescribing a standard of performance under this section which will be applicable to such source."  Commenter 9514 stated this unique use of the date of proposed regulations to define the sources to which an NSPS applies shows that Congress meant to require compliance by new sources without delay. Commenter 9514 further stated, EPA should therefore not base applicability determinations on a source's actual generation.
The EPA agrees with commenter 9514 that the sources subject to the regulation will need to start monitoring and reporting under the regulation. In doing so the affected sources will need to certify all new monitoring equipment. The EPA will require reporting of these certifications prior to the affected source completing a full compliance period. Additionally affected sources will be required to report data after every quarter to the ECMPS Client Tool. The data that will be submitted after every quarter will not always include a compliance period average but will include all of the hourly data that the source will use to calculate the compliance averages when applicable. Additionally the EPA will only determine compliance with respect to the appropriate reported 12
With respect to the first performance test or compliance period, Commenter 9666 stated that EPA's reference to the Part 75 reporting provision in section 75.64(a) is unnecessarily confusing. Commenter 9666 stated if compliance is based on a 12-month rolling average, the period should begin with the first operating month following the month in which monitoring system certification is complete, or the month when the 180-day period in section 75.4 expires, whichever is earlier, but it should not begin in the month in which that event occurred. Under the proposal, Commenter 9666 stated if the EGU did not certify the monitoring system, or the period did not expire, on the first day of the month (i.e., if the monitor was not certified in the middle or the end of the month), the EGU would be required to calculate their first 12-month rolling average with less than 12 full operating months of data. Commenter 9666 stated EGUs should not be required to artificially schedule the certification of their monitoring systems to occur on the first day of the month in order to obtain a full 12 operating months of data.
The EPA appreciates the commenter for pointing out the need for additional clarification under the regulation with respect to the certification of instruments under Part 75.64(a). The EPA is requiring in the final rule that all CO2 CEMS must be certified within 180 days after startup and that monitoring must begin immediately after certification.
Commenter 9666 stated that if compliance is determined annually, and for EGUs that become subject to the NSPS as a result of a change in capacity factor, the first compliance period or performance test would begin with the first operating month of the first calendar year following monitoring system certification or expiration.  Commenter 9666 noted that alternatively, for new units, rather than discard the data for the first partial year of commercial operation, EPA could require that the data recorded during the first partial year be combined with data from the first full calendar year of operation for the initial compliance demonstration.
Commenter 9780 asked EPA to clarify that "before sources have completed the first 12 or 84 months of operation...there is no calculation of a rolling average and thus no basis for finding noncompliance with the emissions rate standard." 
Commenter 9427 asked, "when does the first compliance period begin and end? Does the combustion turbine wait until the first 3-year applicability period is satisfied, and then comply going forward? Or, does compliance look back to the beginning of the 3-year period?" 
Commenter 10095 asked, "does the affected facility have a compliance obligation during the first three years of operation?" and "when is the first compliance determination for a CT?" 
In the final rulemaking the applicability has been clarified so that a source cannot go "in and out" of being subject to the rule.  Compliance with the emission standards is based on a 12-operating month average where the averaging period starts with the first operating month a source becomes subject to the NSPS. Compliance is to be assessed solely on the basis of whether the average CO2 emission rate over the specified 12-operating month averaging period meets the standard. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble.
 
Hybrid Approach
Commenter 10389 recommended a hybrid approach that combines (1) initial performance standard, and (2) 12-month average performance standard. Commenter 10389 stated that requiring an initial compliance test that is numerically more stringent than the annual standard for new combined cycle facilities would insure that the most efficient stationary combustion turbines are installed and the right investment decisions are made at the outset. Commenter 10389 stated the less stringent 12-month rolling average standard would be set at a level that would take into account actual operating conditions. Commenter 8957 stated daily emissions performance matters less than performance over many years or decades.
According to §60.5508 of Subpart TTTT, the rule applies only to new units that commence construction after the specified date. After careful consideration of the comments received, the initial applicability criteria in the final rule have been simplified. Applicability is based on the size of the EGU (>= 250 mmBtu/hr heat input capacity for fossil fuel combustion) and the size of the generator (> 25 MW) producing electricity for sale. It is therefore not necessary for a unit to operate for up to 3 years before it can be determined that it is subject to the rule. A unit that meets the applicability criteria in §60.5509(a)(1) or (a)(2) and is not exempted under §60.5509(b) will remain an affected unit; no "exit ramps" are provided.
Should Conduct Initial Performance Testing
Commenter 9514 stated an initial performance test is needed for all sources covered by these NSPS. Commenter 9514 stated EPA should require an initial performance test before the source begins operating, to ensure compliance with the NSPS. Commenter 9514 stated that failure to require performance testing is inconsistent with section 111(b)(1)(B) and 111(e).
Commenter 9514 stated the need for early compliance determinations is especially great in light of the long averaging times proposed. Commenter 9514 stated there is no reason not to require sources to demonstrate their ability to perform to the annual average emission limits when the unit is new and under controlled conditions.
Commenter 9514 noted that EPA departs from many years of consistent practice by failing to provide for an early initial compliance demonstration for either fossil-fuel EGUs or NG-fired CTs and CCGTs, and EPA does not provide adequate explanation.
Commenter 8957 believed an initial performance test for all sources is consistent with DEP's requirement that all new sources, including turbines, must perform initial compliance tests after the shakedown period.
Commenter 9514 stated, for sources relying on the 84-month compliance alternative, it is essential to demonstrate the ability to meet the applicable emission limits, and provide sufficient documentation to demonstrate at startup that the CO2 capture system functions properly and that it will be able to sequester CO2 on-site at a properly permitted facility or assure long-term containment through transfer to an offsite facility. Commenter 9514 noted that where sequestration operations are under common ownership or control, there is no reason testing could not be conducted as normally required by NSPS. Commenter 9514 noted that where sequestrations operations are not co-located or under common control, there may be unique challenges in conducting an initial performance test (where not already operating), however, an initial test of the EGU and its CO2 capture equipment would provide valuable information. 
Commenter 9514 stated that years of excess CO2 emissions could be avoided. Commenter 9514 indicated that early initial performance tests for all regulated units will avoid periods of noncompliance, avoid situations where "equities on the ground" make injunctive relief less probable, and resolves issues of responsibility for any problems between vendor and operator.
Commenter 9514 noted that for transfer to a sequestration facility that is already in use, no initial performance testing may be necessary, but suggested documentation of the facility's permit and most recent report be included in compliance materials.
Commenter 9514 stated that initial performance testing should be conducted using the most carbon-intensive fuels permitted at the source and with the source operating at load levels, including part-load operation, that reflect what the agency calls "realistic worst case conditions.?
Commenter 10681 stated EPA could use a single, multi-load set of stack tests using Part 60, Appendix A methods, and noted this initial compliance determination under 60.8 is different than meeting the 12-month rolling average limit of the NSPS. Commenter 10681 suggested EPA consider Washington state rule WAC 173-407, sections 140, 150, and 230, for a possible compliance approach. Commenter 10681 indicated Washington State utilizes a quarterly, 4 load emission test program to demonstrate compliance with GHG emissions standards. 
The final rule retains the exemption from performing initial performance stack tests. Because compliance with the emission standards is based on a 12-operating month average, it is not reasonable to hold the sources to these standards based on the results of a stack test. Over a long period of time, the CO2 mission rate will fluctuate and may exceed the numerical value of the standard at times. If a 3-hour emission test is performed during one of these incidents, it may appear that the EGU is out of compliance. But this is not necessarily so. Compliance is to be assessed solely on the basis of whether the average CO2 emission rate over the specified 12-operating month averaging period meets the standard. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble.
Should Not Conduct Initial Performance Testing
Commenter 9666 stated that while EPA proposes that the requirement to perform an initial stack test under section 60.8 would not apply (Proposed Subpart TTTT, Table 2, identifies section 60.8 as inapplicable), EPA also solicits comment on requiring such a test, and thus initial compliance with a more stringent standard, for stationary combustion turbines to ensure that only the most efficient combustion turbines are installed. Commenter 9666 objected to this proposal and noted EPA has not provided any data to demonstrate that such a standard is achievable.
Commenter 9426 supported the proposal that an initial performance test should not be required under section 60.8 for utility boilers and IGCC units or for NG combustion turbines. Commenter 9426 stated an initial performance test provides no added environmental benefit but does add cost. Commenter 9426 stated initial performance will be demonstrated when the facility completes its first period of operation using the approved monitoring system, as is consistent with Subpart Da compliance demonstration for SO2 and NOx, where the initial compliance test consists of operating the certified CEMS for 30 operating days and reporting those results.
Commenter 9426 stated the initial design efficiency or short-term measurement during a stack test would not be representative of the form of the emissions standards proposed and would not be representative of the data the EPA evaluated to establish the standards. Commenter 9426 stated EPA has not provided any evaluation of short-term averages in developing BSER for combustion turbines, and there is no basis for establishing this "secondary" BSER. Commenter 9426 stated that if EPA set an initial testing requirement under 60.8, the short-term standard equivalent to the long-term BSER would need to be less stringent, not more stringent. 
Commenter 9426 stated that setting an initial design efficiency by requiring an initial performance test would be difficult to set, given that new combustion turbines and combined cycle units may have significant variations in performance for many factors beyond EPA's stated intention of insuring the most efficient units are installed, such as ambient conditions, fuel supply, elevation, physical arrangement and size of power generation block, choice of cooling system, amount of duct burning, etc.
Commenter 9426 stated that if EPA decides to set an initial design efficiency, it must reissue the NSPS for NG combustion turbines and provide the basis for public review and comment. 
Commenter 8952 stated there is no value in conducting initial performance measurements at or near plant commissioning, prior to the 12-month compliance period. Commenter 8952 stated compliance with the standard will be strongly dictated by the operational profile of the plant, not the base load emission rate at or near commissioning. Commenter 8952 indicated the additional testing is unnecessary because the project owner always requires a performance validation.
Commenter 8952 recommended an exemption for turbines from initial performance testing.
Commenter 8952 stated the performance test should only apply to combined cycles using ASME PTC 46 (separate performance codes are defined for simple cycle gas turbines), and noted that no simple cycle turbines operate below 1,000 lb/MWh, and most operate above 1,100 lb/MWh. 
Commenter 9426 stated that conducting an initial performance test provides no added environmental benefit but adds cost. Commenter 9426 further stated that BSER is the appropriate mechanism to establish the standard. Commenter 9426 also stated that setting a short-term standard "equivalent to" the long-term BSER would need to be less stringent. 
Commenter 10554 did not support an initial performance test.
The final rule retains the exemption from performing initial performance stack tests. Because compliance with the emission standards is based on a 12-operating month average, it is not reasonable to hold the sources to these standards based on the results of a stack test. Over a long period of time, the CO2 mission rate will fluctuate and may exceed the numerical value of the standard at times. If a 3-hour emission test is performed during one of these incidents, it may appear that the EGU is out of compliance. But this is not necessarily so. Compliance is to be assessed solely on the basis of whether the average CO2 emission rate over the specified 12-operating month averaging period meets the standard. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble.
Best Approach for Initial Performance
Commenter 8952 recommended EPA determine compliance with a CO2 lb/MWh standard at the maximum continuous rating point of the gas turbine and steam turbine during the initial performance testing on natural gas using a fuel measurement method noted in the text (reference ASME PTC 46).
Commenter 8952 believed ideal way to ensure the best technologies are deployed is to determine the overall performance as part of the performance testing conducted during the commissioning phase of a combined cycle facility, using already in place performance test standards ASME PTC 46, typically conducted within the first 180 days of first-fire. Commenter 8952 stated basing compliance evaluation at the maximum continuous rating point corrected to ISO standard conditions, uncertainties related to part load operation and transients can be avoided. Commenter 8952 stated this is essentially an efficiency evaluation, which is what a performance test is.
Commenter 8952 stated the performance test should only apply to combined cycles using ASME PTC 46.  Commenter 8952 stated performance compliance testing would be limited to natural gas operation only.
Commenter 7976 stated that a simple economical way to monitor compliance is to use accurately tested rates of fuel consumption and concurrent MW output of the plant when operating at base load under strict test conditions. Commenter 7976 stated this data accurately establishes the plant capacity. Commenter 7976 noted that after initial compliance testing, regulators can monitor changes in plant heat rate and emission rate but they will not know why changes have occurred; review of maintenance will usually be the more effective way to monitor a plant, and regulators should review plant maintenance records to determine whether CO2 emission can be reduced by effective maintenance and repairs. Commenter 7976 stated re-testing is an option but the results need to be corrected for degradation. Commenter 7976 stated that limiting fuel burn until remedial maintenance is completed is the best remedy because it reduces emissions but avoids shutdown.
The final rule retains the exemption from performing initial performance stack tests. Because compliance with the emission standards is based on a 12-operating month average, it is not reasonable to hold the sources to these standards based on the results of a stack test. Over a long period of time, the CO2 mission rate will fluctuate and may exceed the numerical value of the standard at times. If a 3-hour emission test is performed during one of these incidents, it may appear that the EGU is out of compliance. But this is not necessarily so. Compliance is to be assessed solely on the basis of whether the average CO2 emission rate over the specified 12-operating month averaging period meets the standard. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble.
Commenter 0585 stated energy is a large economic driver for much of Colorado, especially on the Western Slope and the proposed regulations will have impact on the economy and livelihoods. Commenter 0585 asked that EPA consider including compliance flexibility in the rule, and this should include full consideration of distinct regions and would include individual stakeholders and their unique circumstances.
The EPA thanks the commenter for their comment and the EPA believe that the final rule provides compliance flexibilities for all types of new affected sources.
Commenter 8952 stated emissions should only be reported during periods of natural gas operation. Commenter 8952 stated that during periods when gas supplies are curtailed, plants with interruptible gas supplies may use alternative or backup fuel, commonly No.2 distillate or similar fuels, which have significantly higher CO2 emission profiles. 
Commenter 9661 stated EPA does not address compliance for units utilizing oil as a backup fuel, and stated this type of operation should be exempt, as units forced to burn oil for extended shutdowns of NG would emit at significantly higher CO2 emissions rates than allowed by the NSPS standard. Commenter 9661 stated many plants may fire oil as an emergency backup fuel, during period of NG curtailment such as hurricane recovery, other natural disaster disruption, and these periods are infrequent, as economics dictates it is disadvantageous to use oil except in emergency situations.
The EPA disagrees with the commenters as reporting is required by affected EGUs for all fuel types in the final rule along with the associated fuel that was utilized during the compliance/reporting period. The affected source must comply then with the appropriate standard in the rule.
Proposed CO2 Monitoring Methods
The CEMS Option
Commenter 9514 stated EPA should require CEMS for emissions from all units, regardless of fuel type. Commenter 9514 stated direct monitoring of emissions, especially using CEMS, is generally more accurate than estimating emissions using fuel consumption, as EPA has previously acknowledged (cited RIA for MRR). 
Commenter 10554 supported installation of CEMS as the only method to determine compliance with the CO2 standard, noting it is a standard system that follows many checks each day, quarter, and year to ensure the data are correct. Commenter 10554 noted fuel-based calculations have room for error in the analyses.
Commenter 9514 stated EPA should require all new plants to use CEMS to calculate CO2 emissions, and should require plants to include periods of ramping and low load in the compliance demonstration for any monitoring methods that are included in the final rule.
Commenter 9514 stated EPA should require all new plants to use CEMS to calculate CO2 emissions. Commenter 9514 stated the information used to set the standards for new gas turbines is largely CEMS data, which does not exclude periods of ramping activities and low load. Commenter 9514 noted the fuel sampling procedure allows a source to exclude ramping up or down or low load, which effectively exempts periods of low or changing load from regulation. Commenter 9514 stated that allowing sources to disregard these modes of operation from a compliance demonstration weaken the standard below BSER and violate the requirement to demonstrate continuous compliance. Commenter 9514 should require plants to include periods of ramping and low load in the compliance determinations for any monitoring methods.
Commenter 10869 stated that CO2 CEMS will be a significant technological improvement in EPA's ability to track emissions and will come at very little, if any additional cost. Commenter 10869 stated CEMS data should be made available to the public, and it will help inform the impacts of different policies and inform the design and operation of future climate policies. Commenter 10869 stated mandating CEMS will create a more transparent and independent emissions-tracking system that will help improve the quality and credibility of GHG reporting and inventories.  Commenter 10869 noted that CEMS are already used at many stationary sources, including power plants, and under the Acid Rain program, many power plants are already required to report SO2, NOx, and CO2 and O2. Commenter 10869 noted these existing CEMS may be used or upgraded and in some cases new devices may need to be installed. Commenter 10869 stated EPA should ensure robust tracking for CO2 emissions from plants with CCS and for non- CO2 GHGs.
Commenters 9425 and 9666 stated they do not oppose requiring use of CEMS and a stack flow monitor for solid fuel-fired units that employ carbon separation technology. Commenter 9666 stated they do not oppose allowing other units to use such monitoring.
The final rule requires the use of CEMS for coal-fired EGUs and IGCC units and allows all other affected units to use CEMS. However, units that combust only liquid and/or gaseous fuel may opt to use Equation G-4 in Appendix G of Part 75 to quantify the CO2 mass emission rate.
Commenter 9666 supported several monitoring options in the proposed rule but believed EPA has unreasonably eliminated several other Part 75 options. Commenter 9666 supported requiring new EGUs that combust solid fuel, and to allow other new EGUs, to monitor CO2 mass emissions using a CO2 concentration monitor, flow monitor, and (if applicable) a moisture monitoring system in accordance with Part 75 (Proposed Subpart TTTT, Subpart Da, Subpart KKKK), and as an alternative, supported allowing EGUs that combust only liquid or gaseous fuel to determine CO2 mass emissions under the provision in Part 75, Appendix G section 2.3 that uses heat input determined under Part 75, Appendix D and a carbon-based F-factor to determine CO2 mass emissions. (Proposed Subpart TTTT, Subpart Da, Subpart KKKK). 
Commenter 9780 supported used of CEMS data but EPA should consider fuel-based measurements as an alternative.
Commenter 1005 stated EPA should not require use of CO2 CEMS. Commenter 9426 stated use of CEMS would not be feasible for stationary combustion turbines.
Commenter 9661 stated EPA should not require stack flow monitors and CO2 analyzers and should allow fuel flow meters and other mass measurement technologies. 
Commenter 9426 stated use of CEMS would not be feasible for stationary combustion turbines.
Commenter 9425 opposed a requirement to use CO2 CEMS for units that do not use carbon separation. 
The final rule requires the use of CEMS for coal-fired EGUs and IGCC units and allows all other affected units to use CEMS. However, units that combust only liquid and/or gaseous fuel may opt to use Equation G-4 in Appendix G of Part 75 to quantify the CO2 mass emission rate.
Allow O2 Monitoring
Commenter 9425 stated that while most Part 75 coal-fired units use CO2 CEMS, EPA should include use of O2 CEMS and F-factors as alternatives, as Part 75 and Subpart Da allow these. Commenter 9666 noted that while Part 75 and Subpart Da allow use of an O2 CEMS in lieu of a CO2 CEMS, an O2 CEMS cannot obtain accurate CO2 measurement for EGUs that use carbon separation because the calculation relies on F-factors.  Commenter 9425 noted that EPA did not propose to allow the option in 75.10(a)(3)(iii) to use an O2 concentration monitor to determine CO2. Commenter 9425 advocated for an option to use an O2 concentration monitor.
Commenter 9666 did not understand why EPA has required use of a CO2 CEMS and not allowed use of an O2 CEMS to obtain CO2 at an EGU that does not use carbon separation, as both existing NSPS Subparts Da and KKKK allow use of either a CO2 concentration monitor or an O2 concentration monitor as the diluent monitor to calculate emissions in lbs per MMBtu. Commenter 9666 further noted that Part 75 explicitly allows use of an O2 concentration monitor to measure CO2 concentration to obtain CO2 mass emissions. Commenter 9666 stated there is no valid reason not to allow both types of monitoring systems under this rule as well for EGUs without carbon capture. Commenter 9666 noted that gas- or oil- fired EGUs might prefer an O2 monitor since O2 analyzers generally cost less, are more stable, and are less affected by interferences (e.g., H2O and CO) that can impact CO2 measurements.
Commenter 9666 stated EPA must issue a rulemaking for eliminating the O2 monitoring option for public comment. Commenter 9666 stated that eliminating the option in the NSPS effectively eliminates it for NSPS units subject to Part 75.
Commenter 9666 stated EPA could add an option for use of an O2 monitor simply by adding a reference in Proposed Subpart TTTT section 60.5535(b)(1), or in Proposed Subpart Da section 60.46Da(f)(2)(i) and Subpart KKKK section 60.4373(b)(1), to section 75.10(a)(3)(iii).
The final rule (Subpart TTTT) allows the use of data from a certified O2 monitor to calculate hourly CO2 concentrations, provided that the affected EGU does not use carbon separation.
Other Monitoring
Commenter 10098 urged EPA to adopt regulations to address facilities that cannot meet CEMS certification requirements. Commenter 10098 stated that some facilities are unable to comply with the 20 percent variability range for flow RATA for CEMS certification.  Commenter 10098 urged EPA to include additional provisions that would apply in lieu of the certification requirements, in the event that a source is unable to obtain CEMS certification.
The EPA thanks the commenter for their concern however the EPA points to the data collected over the last 20 years from the Acid Rain Program that modern CO2 and flow rate monitors can easily meet Part 75 relative accuracy standards. 
Commenter 7976 stated that compliance monitoring needs improvement. Commenter 7976 stated it is impossible to know the cause of changes in an operating plant's CO2 emission rates simply by monitoring the CO2 emission rate. Commenter 7976 noted that normal heat variations diminish the value of long term CO2 emission rate monitoring for compliance, and a regulator will not know the reason for the change in emission rate of a plant. Commenter 7976 stated if faulty or lax maintenance is not the cause for increased CO2 emissions, then it is unreasonable for regulators to take enforcement action against the plant. Commenter 7976 stated GHG regulators will find that long term online monitoring efforts will rarely uncover significant CO2 emission rate increases due to lax maintenance. Commenter 7976 stated most combined cycle plants are well maintained for reasons of economy and reliability, plus insurance companies, public utility commissions, FERC, and NERC.
The EPA thanks the commenter for their concern however the EPA is finalizing a 12 operating month rolling average discussed more in section IV.C of the preamble. 
The Heat Input and Fuel Factor Option
Recommend fuel flow monitoring
Multiple commenters (8023, 9591, and 3862) supported the option of fuel flow meters for gas or liquid fuels. 
Commenter 9033 recommended that fuel flow composition and fuel flow rate are sufficient for reporting CO2 emissions.
Commenter 3862 stated fuel monitoring may be more accurate than available monitor in some cases.
Commenters 8937, 9426, and 9780 supported fuel-based measurements as an acceptable option or alternative to CEMS. Commenters 9426, 9780 stated utilities currently use fuel-based measurements to accurately measure fuel burned by a unit for their public utility commissions for both ratemaking and fuel supplier invoice true-up. Commenters 9426, 9780 stated fuel-measurements coupled with chemical analysis of fuels burned provide an alternative methodology. Commenter 9780 stated that fuel measurements can provide GHG measurements with an average error of 1.6 percent, making this alternative potentially as accurate as the use of CEMS-based data.
Commenter 9661 stated there is no rational basis for EPA to limit monitoring options to stack flow monitors to the exclusion of fuel flow meters; fuel flow metering under 40 CFR 75 is widely utilized and provides high quality CO2 emissions data, and there is no rational basis not to allow the use of proven technology to calculate CO2 mass emissions as provided in 40 CFR 75, including fuel flow meters. Commenter 9661 noted, for example, proposed section 60.5525(c)(1) cross references 40 CFR 75.10(a)(3)(i), which would require the use of a stack flow monitor and would exclude other options, including fuel flow meters, that are provided within other provisions of 40 CFR 75.10(a)(3). Commenter 9661 stated EPA should update this and any other references to make clear that all of the monitoring options provided within 40 CFR 75.10(a)(3) are available. 
Commenter 9665 stated periodic fuel sampling should be allowed as an acceptable/accurate monitoring method for compliance.
Commenter 8952 recommended fuel flow monitoring where applicable.
Commenters 8952, 9665 recommended the fuel monitoring method be the preferred compliance methodology, and Commenter 8952 recommended CEMs be included as an alternative methodology. Commenter 8952 stated fuel consumption monitoring should be the preferred method for gas turbines with CEMS being the alternative.
Commenter 9591 stated that fuel sampling is currently used to demonstrate compliance for gas- and oil-fired units. Commenter 9591 stated that CEMS add to the cost of the rule and reduce flexibility.
Commenter 8952 recommended that fuel flow composition (weight percent carbon in the fuel) and fuel flow rate are sufficient metrics for reporting CO2 emissions. Commenter 8952 stated fuel analysis (fuel composition and consumption) is a more accurate, cost-effective method and maybe also report a higher output value than the CEMS, making it the more reasonable method for measuring CO2. 
Commenter 8937 noted the GHGRP for the power section relies on reporting using CEMS, and indicated more accurate emissions can be calculated from data reported to EIA-923 power plant operations surveys, and there is strong evidence to support the option of using fuel-based measurements as an alternative to CEMS. Commenters 9665, 10095 indicated that not requiring CEMS would be consistent with other EPA regulation (cited GHG MRR). Commenter 9665 stated that fuel monitoring is superior to CEMS because (1) assumes all carbon in the fuel is oxidized to CO2, and (2) fuel flow data is very precise, since plants are charged for fuel consumption.
Commenter 9665 stated CEMS have inherent uncertainties, especially during transient operation (startup, shutdown, ramping). Commenter 9665 stated CEMS require both CO2 concentration measurement and volumetric flow rate measurement, which results in additional cost and equipment to maintain and certify. Commenter 9665 indicated that one valid data reading is required every 15 minutes (per PS in part 60) but plant parameters can vary significantly over 15 minutes, resulting in significant uncertainty.
Commenter 8952 presented data for combined cycle CEM data that reveals emission levels below 1,000 lb/MWh emission output (average was 880 lb/MWh) however few simple cycle units can match this emissions performance. Commenter 8952 noted that using the recommended fuel flow technique suggests that actual emissions values are higher than the CEMS data presented. Commenter 8952 stated that combined cycle data (NG, using high capacity factor) give an emission average of 880 lb/MWh, and assuming 95 percent compliance and 2 sigma statistical calculation on the historical operation dataset, NG operation would have a recommended benchmark of 1,092 lb/MWh (880 + 2 x 106 = 1,092) or approximately 1,100 lb/MWh.
Commenter 8952 compared the CO2 produced based on total fuel consumption and the reported CO2 based on CEMS measurements (for gas-fired systems that are not cogeneration or CHP, and with extreme emission data points removed). Commenter 8952 noted that total fuel consumption is expected to be more indicative of total carbon released to the environment, including startup and shutdown. Commenter 8952 noted that for combined cycle, the median value [emission rate] is somewhat higher when determined using fuel consumption. Commenter 8952 stated that engineering principles suggest that the fuel flow measurement method would be more accurate, and the data reveal that emission levels [from fuel measurement] are nominally higher than reported by the CEMS. Commenter 8952 stated that differences in the measured values provided are believed to result from inherent inaccuracies in the CEMS measurement systems, most notably during periods of startup and shutdown (although it is not clear whether this CEMS data was operational during startup). Commenter 8952 further noted that during transient events such as startup, there is a time delay between the fuel consumption and the combustion parameters and the CO2 level measured by the emission measurement device. Commenter 8952 explained the delay is caused by the time required for the gases to leave the combustor, reach the exhaust stack, travel down the flow extraction umbilical tube, and then reach the monitor; in some installations this can be a minute or more. As a result, CO2 concentration is measured and converted to a CO2 mass flow based on exhaust flow measurements that may differ by several minutes. Commenter 8952 noted this time lag is not significant during steady operation, but can be significant during transient events, especially start-up or shutdown. Commenter 8952 stated that measurement error is reduced if the fuel flow measurement technique is used. Commenter 8952 stated EPA should adopt a fuel flow method for measuring CO2 for any gas turbine cycle due to its greater accuracy and ease of adoption by those units. Commenter 8952 further noted that these data support their claim that the actual benchmark should be above the proposed 1,000 to 1,100 lb/MWh range.
The final rule allows the use of Appendices D and G (specifically, Equation G-4) for units that combust only liquid and/or gaseous fuel.
Commenter 8952 stated that exempting startup and shutdown period from reporting and averaging would allow fuel flow and CEM methods to be better aligned.
Commenter 9514 stated EPA's "average fuel consumption per unit of generation" procedure allows a source to exclude as nonrepresentative any hours in which the unit is ramping up or down or is operating at low-load. Commenter 9514 stated this procedure exempts period of low or changing load, which are high emissions periods. Commenter 9514 stated that if EPA include a monitoring option based on fuel consumption, the ramping up or down periods and operation at low-load periods should not be excluded from the emissions calculation. Commenter 9514 stated the best CCGT designs provide for a broader range of efficient operation than poorer designs, a fact that should be recognized in BSER and in compliance obligations of the rule.
The EPA is requiring monitoring at all times and the EPA recognizes the concern with startup and shutdown which is why a 12-operating month rolling average is being finalized. The startup and shutdown events will have a small impact on emissions during this long averaging time.
Support Use of Appendix D and Appendix G
Commenter 10095 supported the option of using any part 60 and 75 methods for monitoring and calculating CO2 mass emissions. Commenter 10095 stated monitoring and reporting should be as flexible as the existing requirements under parts 60 and 75, including the ability to use part 75 Appendix D and G methodologies.
Commenters 8023, 9666 agreed with use of Appendix G of part 75 as an alternative to CEMS for liquid and gas-fired EGUs. [4562] Commenter 10095 stated EPA allow use of Appendix D and G for units that do not combust solid fuel or burn exclusively gaseous or liquid fuels.
Commenter 9427 supported the alternative to use Appendix D to part 75, noting that they currently use Appendix D and find it to be accurate, reliable, better data collection, and fewer equipment malfunctions than CEMS.
Commenter 9425 supported use of fuel flow as an option. Commenter 9524 indicated that most new gas- and oil-fired combined cycle units will comply with the ARP to monitor and report hourly heat input and SO2 mass emissions by using Part 75 Appendix D fuel measurement, sampling, and analysis; the heat input value calculated under Appendix D can be used to calculate CO2 mass emissions under Part 75, Appendix G section 2.3 (9425) and subpart KKKK for NOx in lb/MWh. Commenter 9666 noted that EPA previously considered and rejected requiring such units to install stack flow monitoring systems in lieu of Appendix D. Commenter 9666 noted that fuel flow meters are not subject to interferences that can bias measurement. Commenter 9666 stated requiring diluent CEMS at gas and oil-fired combustion turbines would be unjustified.
Commenter 9666 agreed with the option to use, but not require, site-specific F-factors under Equation G-4, noting that Appendix F provides F-factors for most commonly used fuels and includes a procedure for annual determination of a site-specific F-factor for fuels that are not listed, or if a source wants to develop its own F-factor. Commenter 9666 indicated that the variability in commonly used fuels is not sufficient to warrant determination of a site-specific F-factor for all fuels, and that to the extent there is small variability, it is unlikely that variability would be captured in an annual determination. Commenter 9666 further noted that EPA has relied on F-factors identical to those in Part 75, Appendix F in NSPS compliance determinations since the mid-1970s, and requiring development of a site-specific F-factor for commonly used fuels would be a waste of time and resources. 
Commenter 7990 stated EPA should align monitoring, testing, compliance, and reporting requirements with current practices. Commenter 7990 supported use of Equation G-4 to calculate mass emissions, as is allowed under EPA's GHGRP and ARB MRGHG.
Commenter 7990 asked that the monitoring, testing, compliance, and reporting requirements in the NSPS align with current practice established by EPA and ARB. 
Commenter 9425 supported the option for oil- and gas-fired units of using heat input calculated under Part 75 Appendix D and carbon based F-factors as provided in Appendix F using Equation G-4, as the use of Equation G-4 is appropriate for Appendix D units. Commenter 9425 requested EPA include the option of Equation G-4A, and noted that for Appendix D units that fire multiple fuel types simultaneously, ECMPS requires use of Equation G-4A to calculated total CO2 from the individual fuels combusted in the same hour.
Commenter 10952 encouraged EPA to rethink its rejection of optimized reasonable heat rate efficiencies approach, as it is the only rational and legally permissible option at this time.
The final rule allows the use of Appendices D and G (specifically, Equation G-4) for units that combust only liquid and/or gaseous fuel.
Monitoring of CO2 Mass Emissions from Integrated Equipment
Commenters 9666 and 10023 noted that if integrated equipment were used, continuous monitoring likely would not be warranted, as integrated equipment has very few emissions or operates very infrequently. Commenters 9666 and 10023 believed that because there is uncertainty about their use, EPA should not spend resources specifying monitoring requirements. Commenters 9666 and 10023 suggested a provision allowing submission and approval of a petition for exemption of emissions that are de minimis from the compliance calculation, or for an alternative means of accounting for any resulting emissions. Commenter 9666 suggested, mass balance calculation or engineering estimate) from any integrated equipment that are not otherwise accounted for.
 The EPA is finalizing the use of all applicable continuous monitoring requirements that are currently in the Acid Rain Program's part 75 requirements to comply with the emission limits. The EPA believes that there is flexibility provided in the final rule that still maintains the stringency required to determine compliance with the appropriate emission standard. 
Monitoring and Reporting of Gross Generation
Commenter 10243 agreed that compliance should be based on gross generation. Commenter 3493 stated that gross generation is consistent with reporting requirements under Part 75. Commenter 3495 stated gross generation does not factor into the emissions rate the parasitic load from emissions control technologies.
Commenter 9666 stated EGUs already are required to record and report hourly gross unit load in MWge under Part 75 (72.2, 75.57(b)(3), and 75.64(a)(2)). Commenter 9666 stated all new EGUs likely already measure and record MWs for all generation. 
Commenter 9666 stated that EPA's ECMPS instructions say that the value reported under Part 75 represents an hourly rate and not the total load for the hour, so that for hours of partial operation, the value reported will not represent the actual output for that hour. 
Commenter 9666 stated that units on common stacks report the weighted sum of the hourly unit loads for all units that exhaust through the stack. Commenter 9666 stated that reported steam load for an hour is reported at the measured temperature and pressure and is not corrected to ISO conditions.
Commenter 9666 stated that EPA proposal requires calculation and use of actual gross electrical output for each hour and useful thermal energy relative to ISO conditions. Commenter 9666 stated that if EPA includes partial unit operating hours, the values already reported under Part 75 will need adjustment.
Commenter 9666 stated EPA should clarify in the rule partial unit operating hours are included. Commenter 9666 stated that the Part 60 General Provisions make clear that partial unit operating hours are only used in emission calculations when that use is specified (60.13(h)(2)); subpart Da does not included partial unit operating hours for compliance calculations; and subpart KKKK does require inclusion of partial unit operating hours. Commenter 9666 stated that if EPA intends to include, EPA must ensure the data analysis to develop the final emission standard also includes such partial operating hours.
Commenter 9425 stated the requirement should be to report the gross electric output generated during the year and the potential electric generation of the facility. Commenter 9425 stated that reporting the gross electric output sold to an electric grid during the year can't really be done, would have to report the net MW load, since the gross load is not sold to the grid. Commenter 9425 stated the rule should define the potential electric output of the facility. Commenter 9425 stated EPA should consider the methodology defined in part 75, i.e., nameplate load rating multiplied by 8,760 hours.
Commenter 9427 stated stationary combustion turbines are already required by Part 75 to measure heat input, including heat input to duct burners at associated HRSGs, and allocating gross output (MWh) of a common steam turbine electric generator by using heat input to combustion turbines and associated HRSGs would be a very straightforward process and should be allowed without the need for a plan approved by the Administrator.
Commenter 9425 stated the standard would need to allow for the additional CO2 produced for the number of starts a unit could potentially have. Commenter 9425 stated coal units combusting startup fuel or small amounts of coal, the minimum air flow through the boiler dilutes the measured percent CO2 to approximately zero; Commenter 9425 notes the accuracy of determining CO2 mass emissions during these period needs to be evaluated against the small increase in CO2 that would be reported. 
Commenter 10681 questioned the basis for including only 75 percent of the "useful thermal output", on page 1450 of the proposed rule.
Commenter 10095 supported the proposed strategy to only monitor the common exhaust for units that share an exhaust stack, but stated that if a single unit uses multiple stacks or ducts that each should be monitored. 
Commenter 9666 stated that while EPA proposes to require EGUs with common stacks or multiple stack to monitoring "stack operating time", there is not a requirement in the rule to record or report the adjusted electrical and thermal energy output values and if EPA wants those values, EPA must include a requirement in the rule.
The final rule retains the requirement to use gross output (Pgross) in the compliance calculations in addition to allowing sources with permit requirements to monitor net energy output. For partial operating hours, the gross MW values reported under Part 75 must be adjusted using the unit or stack operating time, to convert them to MWh. 
EPA agrees with Commenter 9427 that it is unnecessary to submit a plan to apportion the gross output at a common steam turbine. The final rule allows the combined gross electrical load to be apportioned to the individual EGUs based on the fraction of the total steam load contributed by each unit. If the units are identical, the combined hourly gross electrical load may be apportioned to the individual EGUs according to the fraction of the total heat input contributed by each unit.          
Commenters 9425 and 9666 stated EPA should include a provision in the rule allowing use a value of 1 or 2 MW for any hour during which monitoring gross load is zero. 
Commenter 9666 noted EPA previously developed a policy to allow use of default values for reporting under Part 75; because some such hours were included in the Part 75 data EPA used to support the emission limit, EPA should explicitly allow use of default values for the NSPS.
EPA believes it is unnecessary to report a 1 or 2 MW default electrical load during operating hours when the actual gross output is zero, because the CO2 emission rate (kg/MWh) is not calculated on an hourly basis. Rather, zero gross output is considered to be a valid value for use in the compliance calculations.
Commenter 1899 suggested the use of mass-based compliance instead of rate-based compliance.
The EPA disagrees with the commenter and the final rule retains the rate-based compliance approach.
Quality Assurance and Quality Control (QA/QC)
Commenter 9666 supported reliance on Part 75 QA/QC.
Commenter 9666 does not understand EPA's reference to compliance with part 75 Appendix D, as Appendix A is CEMS certification requirements, Appendix B is ongoing QA/QC requirements, and Appendix D is calculating heat input and mass emissions for SO2 units without CEMS. Commenter 9666 stated EPA should replace Appendices B and D with Appendices A and B.
Commenter 9593 supported exclusion of CEMS bias adjustment and data substitution provision.
Commenters 9471, 9666, and 9780 do not support bias adjustments from Part 75. Commenters 9471 and 9666 noted the bias adjustment factor is a one-way (positive) adjustment that was justified based on the ARP's status as a market-based program, but it has not been used for other NSPS. 
Commenters 8937, 9426, 9661, 9780, 10095, and 10243 supported the use of unadjusted stack gas flow rate values in CEMS calculations, instead of the bias adjustment factors. Commenter 9661 stated there is little margin with the compliance limits proposed by EPA, and requiring use of data that has a positive BAF applied to adjust for uncertainty in the measurement process would be punitive and inappropriately lead to situations of noncompliance.
Commenter 10095 stated EPA should allow grace periods from part 60 and part 75 for QA activities, such as for linearities and RATAs, as was incorporated into MATS.
For units using CEMS to demonstrate compliance, the final rule requires unadjusted stack gas flow rate values to be used in the compliance calculations. Also, the standard frequencies for performing the required QA tests and the allowable grace periods specified in Part 75 apply to all affected units.
The incorrect reference to Appendix D of Part 75 for CEMS certification has been removed and replaced with a reference to Appendix A.
Commenter 9666 stated that to ensure that EGUs are not subject to duplicative and inconsistent QA/QC requirements, EPA must exempt EGUs using CEMS from the requirement in 60.13(a) to comply with performance specifications in Part 60, Appendix B and ongoing QA/QC under Part 60, Appendix F.
Commenter 9666 stated that if EPA wants CO2 mass emissions under this NSPS to be quality-assured according to Part 75, EPA must identify the QA/QC requirements in §60.13(a) as not applicable in Table 1 to Subpart TTTT, and should explicitly state in the applicable rules that those requirements do not apply.  Commenter 9666 stated if there are specific elements of Part 60, Appendices B and F that EPA believes should apply in lieu of Part 75, EPA should identify them and solicit comment on that alternative approach. Commenter 9666 stated EPA's proposal to require compliance with both QA/QC rules would result in significant overlap in testing and confusion regarding the validity of monitored CO2 CEMS data, contrary to EPA's intent and assertions in its proposed ICR. Commenter 9666 noted that EPA previously provided an option to use Part 75 QA/QC in lieu of Part 60 requirements for CO2 and O2 monitors used to monitor SO2 and NOx in lb/MMBtu under Subpart Da (60.49Da(w)(3) and (4)).
Table 3 of the final rule explicitly states that §60.13 is inapplicable, and that all emissions monitoring under Subpart TTTT is to be done according to Part 75.
Multiple commenters (2471, 9426, 9591, 9661, 9666, and 9780) requested that the current Part 75 calibration drift and RATA criteria remain the same. Commenter 10095 supported the calibration drift in Part 75 and the applicable alternative specifications.
Commenter 9666 opposed any divergence from the calibration drift or RATA criteria in Part 75, and in particular the stringent criteria given for comment (daily calibration error of 0.3 percent for CO2 and O2 CEMS, and RATA specification of 2.5 percent for CO2 CEMS and stack flow monitors), as neither criterion are consistently achievable. Because EGUs have reported initial and daily CO2 and O2 calibration drift, and initial and annual stack flow RATA results since 1994, EPA has robust database from which to determine whether the increased stringency is warranted and supported under subpart TTTT, and EPA would have provided this data demonstration if they had it.
Commenter 10554 stated that EPA should not require a daily calibration drift cap no greater than 0.3 percent CO2 for CO2 CEMS because it is infeasible. Commenter 10554 asked that EPA retain the 7.5 percent (annual) and 10 percent (semiannual) limits as in Part 75.
Commenter 10095 stated that the relative accuracy standard of 2.5 percent is not supported by current data, methods, or monitoring system equipment. 
Commenter 9666 further noted that EPA evaluated CO2 and flow monitor RATA results reported under Part 75 (results showed 3.06 percent of 2736 CO2 RATAs and 5.78 percent on 3019 flow RATAs), and determined no revision was warranted; EPA concluded that tightening the standard would not result in any appreciable improvement in results, and also expressed concern that a standard closer to the achieved results would approach the variability of the reference method itself and be beyond the limits of current monitoring technology.  Commenter 9666 stated that EPA must issue a proposal supported by data showing the criteria are necessary and achievable if EPA believes the criteria should be tightened.  Commenter 9666 stated that more stringent QA/QC criteria in subpart TTTT would impact reporting under other NSPS programs, including ARP, since inconsistent QA/QC requirements would require separate databases for the different validation criteria.
Commenters 9426 and 9666 stated that EPA must evaluate the impact of data loss on compliance determinations due to additional QA/QC test failures under the more stringent criteria. 
Commenters opposed the more stringent daily calibration drift (9425) and relative accuracy requirements (8023, 9425). Commenter 9425 stated the criteria are lower than any other EPA rule. Commenters (8023, 9425) stated EPA must provide data to show these standards are achievable. Commenter 9425 stated EPA cannot promulgate new requirements without data and analysis. Commenter 9425 stated these criteria would result in missing data. Commenter 9425 stated lower QA/QC would require separate data sets for Part 75 and for Part 60.
The final rule defers to the existing Part 75 CEMS performance specifications for calibration error and relative accuracy. If EPA believes, at a later date, that these performance specifications need to be revised, changes will be proposed in a separate rulemaking.
Additional Measurement Requirements
Commenters 9666 and 10023 objected to all additional measurement requirements on which EPA requested comment. Commenters 9666 and 10023 noted EPA did not provide any justification for additional requirements, and EPA must propose the requirements and identify the supporting data. Commenters 9666 and 1023 also noted that EPA did not rely on any of these in establishing the proposed emission standards.
Commenter 2471 has no objection to the use of Methods 1, 2F, 3A, and 4. 
Commenter 10095 stated that none of the additional test methods are likely to improve the precision or accuracy of CEMS data. 
Response: The additional and alternative measurement requirements described in the preamble of the proposed rule have not been finalized. The final rule defers to the test methods allowed under Part 75.
Comment: 
Commenters 8023, 9425, and 9666 stated EPA should allow use of M2F and M2H or CTM-041 but should not require them. Commenter 9666 explained that M2F was developed to eliminate high bias of stack flow measurements for cyclonic flow, and M2H and CTM-041 to correct for velocity decay near a stack or duct wall that is not present at other measurement points. Commenters 9425, 9666 stated any EGU concerned about overestimation of measured flow due cyclonic flow conditions or wall effects can opt to use the methods but there is no reason to require their use.
Commenter 10095 asked that M2F not be mandatory. 
Commenters 9425, 9666 stated EPA should allow use of M2G, and can opt to use the best flow method for a particular application but should not be required to use.
Commenter 9425 stated that M2 should be allowed, if the facility can demonstrate compliance with the standard even with overstated flow and the resulting higher emission rate estimates.
Commenters 8023, 9425, and 9666 noted the methods impose additional burden. Commenter 10554 stated that EPA should not require M2F because it is expensive, time-consuming, difficult to calibrate, and calculate flow during RATA testing. Commenter 10554 offered M2G as an alternative. 
The final rule allows the use of Method 2, 2F, or 2G at the discretion of the owner or operator after a source has examined the flow patterns in the stack or duct to see if significant flow angularity is present; if so, the use of Method 2F or 2G must be used as applicable.
Commenters 8023, 9666 stated that Part 75 allows use of 3, 3A, and 3B for dry molecular weight. Commenter 8023 recommended that these methods be an option. Commenters 9425, 9666 stated EGUs should be allowed to use any method allowed under Part 75. Commenter 9666 stated EPA has not provided any explanation for the restriction to M3A or M3B only.
Commenter 10095 stated that M3A and M3B should be allowed but not required. Commenter 9425 opposed a requirement to use M3A. Commenter 1054 asked that EPA not require Method 3B. 
The final rule defers to the test methods required and allowed for under Part 75.
Commenter 9666 did not understand why EPA solicited comment on use of M4, because these sources are already required to use M4 for flow RATAs (M2, M2F, or M2G for performing flow RATAs and each require use of M 4), and asked that EPA explain how use of M4 would differ from what is already required under Part 75. Commenter 9666 indicated that Part 75 allows use of both the reference method and the approximation method in M4.
Commenter 10095 supports the use of M4. 
The final rule defers to the test methods required and allowed for under Part 75.
With respect to ambient air Argon concentration, Commenter 9666 stated the negative error of approximately 0.4 percent suggested in M3 is trivial and not worth correcting, and noted that EPA has not considered the error significant enough to require correction in the past. Comment 9666 stated EPA must provide explanation and propose a procedure for including argon in the analysis. Commenter 9666 stated EPA would best make a proposal for revision to M3 rather than in an individual NSPS. 
Commenter 10095 stated dry gas molecular weight requirements should be consistent with Part 75 and not require the change to use ambient air argon concentration and molecular weight. Commenter 10095 stated this change would minimally improve accuracy given the insignificant percentage of argon in ambient air, but further complicates compliance with additional calculations than that required in Part 75.
The additional and alternative measurement requirements described in the preamble of the proposed rule have not been finalized.  
With respect to requiring use of a value for pi of 3.14159, Commenter 9666 stated that EPA provides no explanation for this proposed requirement. Commenter 9666 further noted that to the extent EPA proposes this value to obtain greater accuracy, that effort is misguided. Commenter 9666 stated the impact on CO2 mass emissions would be insignificant and would suggest a level of accuracy that simply does not exist in the required stack measurements.
The additional and alternative measurement requirements described in the preamble of the proposed rule have not been finalized.  
Commenter 9666 noted that language for installing moisture monitoring systems for dry basis CO2 CEMS could be interpreted as eliminating use of default moisture values (see 75.11(b), Proposed Subpart Da section 60.46Da(f)(2)(1), Subpart KKKK section 60.4373(b)(1), Subpart TTTT section 60.5535(b)(1)). Commenter 9666 stated they can think of no reason why EPA would eliminate use of default moisture values for new units, and hope it was not intended to eliminate use of default values.
Commenter 9666 stated EPA must issue a rulemaking for eliminating moisture default value option for public comment. Commenter 9666 stated that eliminating the option in the NSPS effectively eliminates them for NSPS units subject to Part 75.
If CO2 is measured on a dry basis, commenter 9593 stated the rule should require a continuous moisture monitoring system or allow a fuel-specific default moisture value for each unit operating hour, as is consistent with Part 75.11(b). 
Commenter 10100 stated EPA should eliminate the requirement to monitor stack moisture content using CEMS, noting that nearly all solid fuel CEMS systems are dilution extraction systems, so moisture is not an issue. Commenter 10100 noted that moisture monitoring capability is unusual in CEMS.
The final rule allows the fuel-specific moisture defaults in §75.11(b) to be used in lieu of continuously monitoring the stack gas moisture content.
Commenter 9666 objected to the requirement to measure the stack or duct dimensions using a laser device (see Proposed Subpart TTTT section 60.5535(b)(3), Subpart Da section 60.46Da(f)(2)(iii), Subpart KKKK section 60.4373(b)(3)). Commenter 9666 stated the record contains no evidence to support that existing methods are not sufficiently accurate or that laser devices are more accurate. Commenter 9666 stated the requirement imposes additional cost and burden with no benefit.
Commenter 10095 stated there is no mention of the conditions under which measurements should be made; off-line, full-load, ambient conditions affect expansion and contraction of ducts and stacks. Commenter 10095 stated EPA should provide explanation how the laser devices improve the measurement accuracy, and otherwise, EPA should not require this measurement. 
Commenter 9593 stated the requirement to measure stack dimensions at 3 distinct locations should be reduced to 2 distinct locations to prevent installation of an additional sample port just for stack dimension measurement. Commenter 9593 stated the stack dimension requirement should be to within 0.25 inches at the flow monitor location.
Commenter 9426 supported the use of a laser device to make stack diameter measurements. Commenter 9426 stated that the EPA should make it clear in the final rule that the laser measurements are to be taken only at the stack testing or flow monitoring elevation on the stack, and that only a single set of measurements should be required if the testing elevation and the flow monitoring elevation are closely coupled in the stack. Commenter 9426 stated the requirement to measure stack dimensions at 3 distinct locations should be reduced to 2 distinct locations because the cost of installing two additional ports to accommodate a third measurement location does not justify the potentially small increase in measurement precision.
The EPA thanks the commenters for the comments and is finalizing the rule with the requirement to measure stack gas flow according to Appendix A or part 75 and §75.10(a)(3)(i). Additionally the EPA is finalizing measurements of diameter to be required at only 2 locations for circular stacks and 2 locations for each length and width of rectangular ducts.
The Monitoring Plan
Four commenters (8023, 9425, 9666, and 10023) had no opposition to the Monitoring Plan requirement. Commenters 9425, 9666, and 10023 indicated that new EGUs already have to develop these plans under Part 75 and are submitted by Part 72 designated representative (DR); Commenters 9425 and 9666 noted that information for NSPS must be submitted by facility owner or operator, which may be someone other than the DR. Commenters (8023, 9425, 9666, and 10023) requested EPA clearly state that the Part 75 plan submitted by the Part 72 DR would constitute compliance with NSPS requirement. Commenters 9425 and 9666 requested that to avoid having two individuals certify the same information, where the NSPS relies on information already submitted and certified under Part 75, no further submission or certification should be required. 
For Acid Rain Program units, the final rule simply affirms the existing requirement for monitoring plan submittals to be made by either the Designated Representative (DR), the Alternate DR, or a delegated agent of the DR. To ensure consistent implementation of the rule, the owner or operator of an affected unit that is not in the Acid Rain Program (ARP) must, at a minimum, appoint a DR and the DR must register with the Clean Air Markets Division (CAMD) Business System. The monitoring plan submittals for a non-ARP unit must be made by the DR, or, if applicable, by the ADR, or by a delegated agent of the DR.
Commenter 9666 stated the Monitoring Plan rule language is too broad in that section 75.53(g) and (h) are not limited to CEMS for CO2 mass emissions only, and EPA should limit the Monitoring Plan information to CO2 emissions monitoring.
The final rule requires reporting of only those monitoring plan data elements in §75.53(g) and (h) that are needed to quantify the hourly CO2 mass emission rate (tons/hr). This clarification is especially important for units that are not in the Acid Rain Program.
Commenter 10095 supported the use of ECMPS to collect reports, CO2-related data, and gross energy output data for a unit, provided that EPA can supply any necessary updates to ECMPS in a timely manner. Commenter 10095 noted that ECMPS already contains megawatt data and should be modified to also allow the reporting of steam flow rather than pulling generation data from another source. Commenter 10095 stated that EPA should allow permitting authorities the flexibility to collect reports and data through different mechanisms if desired or needed.
The final rule requires the quarterly CO2 emissions reports and the compliance reports to be submitted electronically using ECMPS.	
Commenter 10029 suggested that EPA require EGUs to prepare monitoring, reporting, and verification (MRV) plans to ensure proper monitoring and reporting of the disposition or fate of CO2 once transferred from an affected EGU, similar to the MRV plans under 40 CFR 98.448.
 The EPA thanks the commenter for their comment the EPA is requiring in the final rule reporting under subpart RR of part 98 and points the commenter to see Chapter 6 of this document for responses relating to 40 CFR Part 98 Subpart RR.
Schedule for Monitoring System Certification
Commenter 9666 supported the 180 days allowed for CEMS certification. Commenter 9666 pointed out that while the preamble indicates all "monitoring systems" for CO2 may be certified in timeframe for 40 CFR 75.4(b), they could not find rule language that implements this; the commenter noted that the two rules should have the same certification deadlines, and the regulatory text should refer specifically to "monitoring systems required to monitor CO2 mas emissions" rather than to CEMs, to reflect that Appendix G monitoring systems rely on certified fuel flow meters and not CEMS. 
Commenter 9514 stated EPA needs to reduce the 180 days allowed for CEMS certification, noting that the 180 day allowance in Part 75 reflects the more difficult task of retrofitting monitoring equipment onto existing sources and new sources should not face similar challenges. Commenter 9514 recommended 90 days for CO2 CEMS certification. 
The EPA disagrees with Commenter who stated that the window of time allowed to complete certification of the continuous monitoring systems should be shortened. The 180 day certification window specified in §75.4(b) is the existing requirement for new Acid Rain Program (ARP) units. The Agency believes it is inappropriate to deviate from that established certification timeline. Nearly all of the affected units under Subpart TTTT will also be in the Acid Rain Program.  
Annual or Multi-Year Compliance Periods
Annual vs. Multi-Year Basis
Commenter 9514 stated that even the 12-month compliance will leave the compliance status of newly constructed sources uncertain for an extended period. 
Multiple commenters (8964, 8969, 8970, 8973, 8995, 9318, 9671, 9672, 9736, 10392, and 10876) did not support a short-term standard but if EPA decides to adopt a short-term standard the commenters asked that it be based on vendor guarantees of unit performance during pre-commissioning testing.
The EPA disagrees with the commenter 9514 and compliance is to be assessed solely on the basis of whether the average CO2 emission rate over the specified 12-operating month averaging period meets the standard. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble
Support Multi-Year Compliance
Commenter 10239 stated EPA must include a multi-year compliance option.
Commenters 9515, 10031, 10089, 10095, 10098, 10239 supported multi-year compliance periods. Commenters 10098, 10239, and 10788 supported the 84-month compliance period. Commenter 10031 supported a 24 month calendar average.
Commenter 9666 recommended that EPA establish a single compliance period on a calendar year or longer basis for both coal-fired and NGCC units. Commenter 9666 stated that requiring more frequent compliance demonstrations would hinder operational flexibility without providing any additional environmental protection. 
Commenter 9664 stated 84-month compliance option increases operational flexibility. Commenter 9964 indicated this is helpful in earliest stages of facility operation and early years of the performance standard, where captured CO2 is delivered offsite to sequestration not under owner control. Commenter 9471 stated compliance under a sliding scale over a 7-year period provides some additional flexibility for new coal units with CCS, and it also implicitly acknowledges the uncertainties as to when or if CCS technology will be developed.
Commenters 9666 and 10023 stated a multi-year compliance period, such as the proposed 84-month compliance period for coal-fired EGUs, would be justified because of the nature of the alleged effects on public health and welfare. Commenters 9666 and 10023 stated this approach would be consistent with the year-to-year changes in operations and emissions of EGUs. 
Commenter 9664 stated the 84-month compliance option minimizes rule effects on electricity prices to a certain extent, but unlikely to translate to significant cost benefits associated with former 30-year alternative.
Commenter 9471 supported the regulatory flexibility for the 84-month compliance period, but does not view it as a meaningful provision that preserves future coal-fired generation.
Commenters 2470 and 9035 asked that EPA consider compliance flexibility to allow time for CCS technology development. Commenter 9035 offered two options: 1) allow a percentage of capture over an extended period of time (e.g., 20 or 30 years), or 2) require a set percentage of capture by a nearer set date but allow some time for new plants to operate without CCS. 
Commenter 9317 stated that the compliance period should be no shorter than a 12-month rolling average but an 84 month averaging period could create problems with enforcement. Commenter 9317 stated that a limit more consistent with the five year term of a title v permit would be more logical. 
Commenters 3862, 8967, and 10525 supported a choice to maintain the 84-operating month rolling average or the 12-operating month rolling average.
Commenter 9765 asked that EPA clarify in detail why the 84 month averaging period was proposed and why it is preferred over the 30 year period originally proposed in April 2012. 
Commenter 1900 asked that EPA consider 24 months as the compliance period. 
A 12 operating month compliance period has been retained for the final rule for all affected EGUs. More discussion on this can be found in section IV.C of the preamble.
Oppose Multi-Year Compliance
Commenter 8957 believed the 84-month compliance standard runs counter to the Agency's argument that CCS is commercially available, and why would an 84-month period be needed to dampen short-term excursion if the technology is already available. Commenter 8957 believed the 12-month standard is sufficient.
Commenter 7977 stated that enforcement of the 84-operating month rolling average is problematic because operating permits issued under Part 70 or 71 are five years.
Commenter 10869 noted that the 84-operating month compliance option is problematic and should be tightened to "ensure that it does not merely serve as a loophole for new high-emitting power plants to continue to be built with a high risk that those emissions may never be significantly reduced." Commenter 10869 further stated, "Given the fact that many U.S. coal plant operators have resisted installing pollution controls for sulfur dioxide (SO2 scrubbers), their assurances that CCS will be added at some future date must be viewed with some degree of skepticism. Finally, if a plant's backers fail to include the cost of a future CCS retrofit in the plant's original price tag that could prevent regulators, ratepayers, and investors from knowing its true cost and judging how the proposed plant compares with cleaner energy alternatives."
Commenter 10869 asked that the compliance period be as short as possible and also recommended "including a provision requiring power plant owners who want to employ this option to simultaneously invest in renewable energy, energy efficiency, or other low carbon technologies that would compensate for the emissions released during the first years of operation as a conventional coal plant."
Commenter 8924 disagreed with the 30-year average compliance period. 
A 12 operating month compliance period has been retained for the final rule for all affected EGUs. More discussion on this can be found in section IV.C of the preamble
Support 12-month Compliance
Commenters 2470, 9591, 10100, 10554, and 10788 recommended EPA to allow a choice of either 12-month rolling average or an annual basis (calendar year). Commenter 10100 noted that some facilities will prefer 12-month rolling average (peaking units, extreme seasonal weather), and some facilities will prefer a calendar year average (less variability over year, extreme summer weather). Commenters 9591, 10095, 10100 noted that calendar year basis is consistent with ARP and MRR.
Commenter 10239 believed a 12-month compliance period is necessary.
Commenters 8957, 9033, 9591, 9678, 9780, 10681, and 10693 supported the 12-month rolling average calculation for compliance. Commenter 10389 noted the rolling 12-month compliance provides critical flexibility for start and stops or burning oil.
Commenter 9425 believed the 12-month method of compliance is an improvement. Commenter 9425 stated EPA should clarify that the compliance determination does not begin until there is  12 months of data to make the calculation, citing precedent for this (310 CMR 7.29 in Massachusetts).
Commenters 7990, 9042, 9194, 9426, 9592, 9666, 9777, 10023, 10031, 10095, 10098, and 10239 supported a 12-month compliance based on an annual standard (calendar year), not a rolling average. Commenters 9592, 10095, 10239 stated an annual basis for compliance will be less costly to industry, reduce burden, easier to administer, and will provide the same level of compliance assurance. Commenter 10095 stated EPA has not identified any environmental benefits that warrant rolling 12-month compliance. Commenter 10095 stated that if EPA chooses a rolling 12-month compliance, EPA must clearly identify why that compliance period is a better balance of economic, environmental, and energy considerations that a block calendar compliance period. Commenter 10031 preferred a 24 month calendar average but 12 month calendar year average was the second choice.
Commenter 10095 stated EPA must revise the definition of "operating month" to provide a threshold exclusion for months when minimum amounts of fuel are combusted (e.g., only fuel combusted was for testing, equipment exercising, post maintenance, compliance testing, etc.)
Commenter 9666 stated that the additional language in Proposed Subpart Da section 60.46Da(d) is confusing and should be deleted; although compliance is determined at the end of each operating month, compliance is not determined "for the month." 
A 12 operating month compliance period has been retained for the final rule for all affected EGUs. The compliance determination is not made until 12 operating months has occurred, however the hourly CO2 data will be reported quarterly via part 75's ECMPS client tool during those first 12 operating months and thereafter. More discussion on this can be found in section IV.C of the preamble
Include an Option for 12-month Compliance
Commenter 9666 supported providing owners or operators with an option of complying on a 12-operating month rolling average basis if they so choose.
The 12 operating month compliance period has been retained for the final rule for all affected EGUs. More discussion on this can be found in section IV.C of the preamble.
Not Set a More Stringent Standard for Multi-Year Compliance
Commenters 9515, 10023, 10098, 10239 disagreed with a lower emissions limit for multi-year compliance timeframe. Commenter 10239 supported an emissions limit that is the same as the limit for 12-month compliance.
Commenter 9666, 10095 did not believe that a compliance period longer than 12 months justifies setting a more stringent standard for either coal- or gas-fired EGUs.
Commenter 10095 stated intermediate standards are not warranted for longer compliance periods.
Commenter 9780 stated it is not clear why EPA imposes a 12-month operating average that undercuts flexibility, if the goal of the 84-month compliance option is to provide additional flexibility to units installing a new technology.
Commenter 9780 stated that if EPA holds a unit liable for every day of the 84-month compliance period if the CO2 standard is exceeded, there is already more than ample incentive to ensure careful compliance with an additional 12-month operating average standard.
A 12 operating month compliance period has been retained for the final rule for all affected EGUs. More discussion on this can be found in section IV.C of the preamble
Include Stringent Measures for 84-month Compliance Option
Commenter 9660 indicated that States are concerned that the 84-month option may pose compliance and enforcement issues if a power plant fails to meet the NSPS. Commenter 9960 stated EPA should include effective deterrence mechanisms, adequate contingency measures in the event that the plant violates the NSPS, and mandatory reporting of emission averages at specified intervals to be able to gauge whether the plant is on track to comply with the NSPS and allow for timely corrective action as needed.
Commenter 5537 noted the 84-month compliance option could create difficulties in effective enforcement if the unit fails to meet the NSPS. Commenter 5537 stated the longer term compliance option should be done to seek greater total CO2 reductions over the lifetime of the EGU relative to the 12-month option, provided effective deterrence mechanisms at the outset to avoid later noncompliance, scaled penalties for violating the standard, and should be uniformly applied across all states and regions. Commenter 5537 stated noncompliance penalties should continue to be assessed according to the more stringent CO2 emission limit of the 84-operating-month option on a rolling monthly basis. Commenter 5537 stated the option should promote greater technology innovation for installation of measures that go beyond that needed for 12-month compliance standard. 
Commenter 5537 stated the standard should be more stringent relative to the 12-month compliance standard.
Commenter 9514 stated the agency must include interim demonstrations for the 84-month compliance option, to ensure compliance. Commenter 9514 stated EPA should limit the option's application only to those sources that may need the additional flexibility it provides.
Commenter 9514 stated EPA must require milestones to ensure that regulated sources take all necessary steps to prepare for, and operate under, the 84-month emission limitation period.
Commenter 9514 stated EPA must require that the operator demonstrate at the end of each 12-operating month period that the source will meet the requirement. Commenter 9514 further stated the permitting authority would be required to approve the certification and demonstration of compliance; regulators and the public could bring an enforcement action if the demonstration was not sufficient to establish compliance with the NSPS. Commenter 9514 noted these steps will avoid situations where sources find themselves ultimately unable to achieve sufficient emission reductions to make up for excess emissions during the initial months of operations.
Commenter 9514 indicated the key question to answer for interim analysis is whether they retain the capability to comply with the 84-month limit given the time that remains in the compliance period. Commenter 9514 suggested a 3-step process that would enable permitting authorities to annually evaluate the ability to meet the 84-month compliance:  (1) Step 1 - Determine the projected 84-month MWh output of the source; (2) Step 2 - Determine the minimum practicable 84-month CO2 emissions for the source; and (3) Step 3 - Calculate the source's minimum practicable 84-month CO2 emission rate. Commenter 9514 stated if the "minimum rate" exceeds the standard, then the source is in violation and must cease operating until it is able to come into compliance.
Commenter 9514 stated EPA must ensure that information on the progress of sources subject to the 84-month compliance option is available. Commenter 9514 stated this information is needed for EPA to review and revise this NSPS and for other source categories where CCS may be appropriate, and for citizens concerned about health and environmental impacts of power plants. 
Commenter 9514 stated the rule needs to include specific deadlines and required filings with the permitting agency to ensure all carbon capture equipment, necessary infrastructure and sequestration agreements, and any other components are in place. Commenter 9514 stated these measures must be incorporated into a source's Title V permit, to ensure they are binding and enforceable.
Commenter 9514 stated the rule should clarify the 84-month compliance option is available only during a source's initial 84 months of operation after construction; after this, EPA's rationale for the extended compliance period no longer applies, and the flexibility provided by the 84-month option is no longer warranted. Commenter 9514 stated that after their initial 84 operating months, all sources should be subject to the 12-operating month standard.
Commenter 9514 stated the 84-month option should automatically terminate for new plants commencing construction eight years or more after the proposed rule, i.e., in 2021. Commenter 9514 stated that automatic termination of the provision will prevent unwarranted expectations that the option will be renewed, but does not precluding EPA from renewing the provision if it is still determined to be appropriate in 2021.
A 12 operating month compliance period has been retained for the final rule for all affected EGUs. More discussion on this can be found in section IV.C of the preamble
NGCC
With respect to NGCC units, Commenter 9666 supported basing compliance requirements on an annual (calendar year) average basis because a single compliance period each year simplifies compliance. More frequent compliance demonstrations would hinder operational flexibility without providing any additional environmental protection. 
Commenter 9666 suggested that EPA provide owners and operators of NGCC units with the option of complying on a 12-operating-month rolling average basis.
For NGCC units, commenter 10243 supported the methodology of calculating 12-month average emission rates based on hourly mass CO2 emissions and would also support a multi-year averaging period. 
Commenter 10466 asked that any standard for large natural gas combined cycle turbines be implemented on a 3-year average basis. Commenter 10466 noted that the 3-year average ensures that efficiency declines due to blade fouling are accounted for.
The 12-operating month rolling average compliance periods have been retained in the final rule and for a detailed discussion on this topic please see section IV.C of the preamble. Initial performance stack tests and interim compliance assessments are not required. However, the regulatory agencies and other interested persons can monitor the progress of the affected EGUs by reviewing the quarterly Part 75 electronic emissions reports. These reports include hourly CO2 mass emission rates (tons/hr), hourly unit or stack operating times, and hourly gross electrical loads. This information is sufficient for regulatory agencies and auditors to make general assessments of the units' ability to meet the emission limits. 
Hourly, Monthly, Daily
Commenter 9033 suggested that hourly reporting data is a significant burden to the operator and not required.
Commenter 8952 stated hourly reporting is not a necessity because the total fuel flow corresponds exactly to the total CO2 mass emissions and these data are already reporting through regulatory channels (FERC). Commenter 8952 stated that where fuel monitoring reporting is already underway, the rule should allow these data to be used as substitute in lieu of collecting hourly information.
Commenter 8952 saw no reasonable justification for the requirement to quantify each hourly emission period when the averaging period is 12 or 84 months. Commenter 8952 stated the requirement imposes burden without any environmental or other benefit.
Commenter 8952 noted there is no public health or environmental protection benefit from compliance on a monthly basis or even to require continuous hourly data collection; these requirements are therefore not reasonable. Commenter 8952 stated it would be far more accurate, at the end of each month, to compute a new rolling average by summing all CO2 emitted in the past 12 months and dividing by the total MWhs generated in the past 12 months.
Commenter 10239 did not support compliance determinations each month for a rolling compliance period.
Commenter 8952 stated that monthly data collection would be more than sufficient for reporting purposes. 
Commenter 9514 stated daily computation poses no additional burden on permitting authority audits, since those authorities have to review hourly and daily emission data in order to properly audit the monthly averages, and software applications can be used. Commenter 9514 also noted the utility industry is already required to monitor and electronically report hourly CO2 emissions data under Part 75. See 40 C.F.R. section 75.64(a)(6). 
 The EPA is utilizing the Acid Rain Program part 75 monitoring and reporting procedures as much as practicable for compliance with the emission standards set in this rule.  The EPA believes that many sources will already be reporting under part 75 however additional detail is needed in the final rule for those units potentially not subject to the applicability requirements under the Acid Rain Program.
Daily Violations
Commenter 9514 stated EPA should require daily, rather than monthly, calculation of rolling annual average, as is consistent with CAA penalty provisions for "each day of violation", not merely those days on which compliance happens to fall. Commenter 9514 stated that once a facility is in violation of the standards, penalizing each additional day spent operating in excess of the applicable CO2 limit will better ensure that the source comes into compliance as soon as possible. Commenter 9514 noted that sources will likely calculate the rolling average each day even if only monthly data is submitted to EPA, and therefore this change would likely impose little additional burden.
Commenter 9033 suggested the requirement to report daily violations is unreasonable since the basis of compliance is a 12 month rolling average.
Commenter 10100 supported calculating the number of daily violations within an averaging period, which prevents double-counting of violations across different rolling averaging periods.
Commenter 8952 stated determining daily violations is not reasonable. Commenter 8952 stated that the inclusion of hourly or daily emissions rates will become a determinant factor in establishing any potential penalty.
Commenter 10681 stated that the discussion of daily violations is inappropriate and if EPA wants to provide for daily violations, EPA should develop and propose a "not to exceed" daily emission limitation. 
Commenter 9317 did not agree with EPA's proposal on determining daily violations and stated, "There could only be one violation for each month where the rolling average exceeds the standard." 
Commenter 10048 asked that EPA consider assessing violations on a case-by-case basis. Commenter 10048 suggested that instead of counting all days within the operating period as violations, EPA could determine the specific days that contributed to the actual exceedance and only count those days. 
Affected units must begin using ECMPS to submit quarterly electronic emissions reports as soon as the monitoring systems have been certified. These reports, which are also required for new units that are subject to the Acid Rain Program, include hourly CO2 mass emission rates (tons/hr), hourly unit or stack operating times, and hourly gross electrical loads. The vast majority of the units in Subpart TTTT are also in the Acid Rain program---therefore, the requirement in Subpart TTTT to report hourly CO2 emissions data introduces no new reporting burden. Note that affected sources are not required to calculate hourly, daily, or monthly CO2 emission rates in units of the standard (kg/MWh)---rather, a single CO2 emission rate is calculated at the end of each compliance period, as the ratio of the total CO2 mass to the total gross output for the valid operating hours in that period
If the standard is exceeded for a particular compliance period, it counts as one violation. The EPA has removed the concepts of "excess emissions" over the 12 operating month averaging period.
Format of the Standard, Total Emissions over Total Electrical Output
Commenters 8952, 9033, 9591, 9678, 10095, and 10389 supported compliance calculation as total CO2 emissions over total generation data over the period. Commenter 8952 noted the calculation is an improvement.
Commenter 9678 stated this ensures that plants that cycle more frequently in some months have additional compliance flexibility.
Commenter 10100 stated that for 12-month calendar year average, compliance should be based on annual emissions over annual MWhs (instead of an arithmetic average of 12 monthly values).
Commenter 9514 supported the abandonment of the April 2012 strategy of using the "average of 12 monthly averages" to determine compliance.
Commenter 9780 should address whether the average is a straight average of the monthly emissions rates, or is the average weighted in any way by output in the month, and any weighting in determining the average must be specified.
Commenter 8952 suggested simplifying the method for computing the 12-month rolling average. For fuel consumption monitoring, commenter 8952 stated monthly data collection would be more than sufficient for reporting purposes.
Commenter 10389 stated the only potential problem is where a facility is burning fuel in reserve, but not delivering "useful energy output" to the grid; by counting the emissions, but no megawatt hours, a plant owner would artificially increase its average emission rate.
Commenter 10095 stated EPA's calculation for longer compliance periods must continue to be the sum of emissions for all operating hours over the compliance period over the sum of all gross energy output over the same compliance period.
In the final rule the EPA is finalizing the method to sum all hourly emissions and energy output for the compliance period and then doing the calculation for an emission rate to be compared to the applicable emission standard. The EPA believes this method provides the greatest consistency of comparing emission rates for the long averaging times to the applicable emission standards.
Include 0 for Emission Rates if Not Operating
Commenter 9780 stated EPA should clarify that if a unit is not operating during a calendar month, an emissions rate of 0 lbs/MWh is to be used for that month in the rolling average calculation.
The EPA does not agree with the commenter and compliance is assessed on a 12- operating month basis. Non-operating months are disregarded when determining compliance.
Compliance Issues with 3-Year Applicability Threshold
Commenter 9427 stated EPA needs to address multiple issues related to the 3-year application threshold:  (1) What if a combustion turbine owner/operator complies with requirements such as reporting, but it turns out that the requirements were not applicable because the 3-year thresholds were not met? Alternatively, what if a combustion turbine owner/operator assumes the requirements would not apply but then at the end of a 3-year period finds the applicability thresholds were met?  (2)  In Title V annual compliance certifications, how can an owner/operation certify compliance when applicability is not known?  (3) How is compliance handled if a combustion turbine may or does routinely operate above and below applicability thresholds over consecutive but overlapping 3-year periods?  Commenter 9427 recommended the following to address these issues:   (1) Owners/operators of new stationary combustion turbines that meet design applicability criteria but do not meet all operating applicability criteria could certify that their facility is not subject to the standard; and (2) By notification or certification, owners/operators should be able to revise the applicability status to reflect changes in operating status over successive "three consecutive calendar years" periods.
The 3-year thresholds have been removed from the final rule. Additionally, compliance for all EGUs subject to emission limits is over a 12 operating month rolling average. For more discussion on the 12-operating month compliance period please see section IV.C. of the preamble
Compliance Issues Related to Applicability
Given the applicability criteria for fossil fuel-fired boilers and IGCCs, Commenter 10095 stated that EPA needs to clarify how the applicability criteria interact with the 84-month compliance period, and whether the timeframes within an 84-month period that do not meet the applicability criteria count towards compliance.
Given the applicability criteria for combustion turbines, Commenter 10095 stated that EPA needs to clarify how the applicability criteria interact with the 12-month rolling average compliance period, and if a CT meets all three of the applicability criteria below listed over a three year average but fails to meet one of the criteria for a year during the same three year period, from which 12-operating month rolling average compliance determinations are the affected facility excluded, and to which 12-operating month rolling average compliance determinations do the affected facility have compliance obligations? (If a CT does not meet the 10.0 percent fossil fuel criteria, from which 12-operating month rolling average compliance determinations are the affected facility excluded? If a CT does not meet the 90 percent natural gas criteria, from which 12-operating month rolling average compliance determinations are the affected facility excluded? If a CT does not meet the one-third and/or 219,000 MWh sales criteria, from which 12-operating month rolling average compliance determinations are the affected facility excluded?)
Commenter 9427 recommended a "three consecutive calendar years" compliance timeframe to address variability issues, shortcomings in EPA's BSER database, and there is no scientific reason to require a 12-operating month compliance period for carbon dioxide.
Commenter 3862 noted that an option based on a 36 month operating month rolling average would be consistent with other 3 year applicability periods in the proposed rule. 
Commenter 7977 recommended a 3-year averaging period to determine applicability of the rule to allow for usage during emergencies regardless of the type of unit, purpose, or fuel.
According to §60.5508 of Subpart TTTT, the rule applies only to new units that commence construction after the specified date. After careful consideration of the comments received, the initial applicability criteria in the final rule have been simplified. Applicability is based on the size of the EGU (>= 250 mmBtu/hr heat input capacity for fossil fuel combustion) and the size of the generator (> 25 MW) producing electricity for sale. It is therefore not necessary for a unit to operate for up to 3 years before it can be determined that it is subject to the rule. In the final rule, the annual and 3-year rolling average assessments of the unit's status have been eliminated. A unit that meets the applicability criteria in §60.5509(a)(1) or (a)(2) and is not exempted under §60.5509(b) will remain an affected unit; no "exit ramps" are provided.
Valid Data, Backup Monitors, Missing Data Procedures, and Minimum Data Availability
Valid Data
Commenter 9666 stated EPA does not define the term "valid data."  Commenter 9425, 9666 indicated EPA must be clear about what it means by "valid data".
Commenter 9666 indicated one interpretation is that data are valid as long as the MS is not "out of control" (consistent with Part 75). Commenters 9425, 9666 stated that use of part 75 definition for "out of control" would allow EGUs to rely on the same definition for both rules. Commenters 9425, 9666 stated Subpart Da and Subpart KKKK definitions are not consistent with Part 75. 
Commenter 9666 objected to EPA's use of "out of control" period beginning and ending with the corresponding "quadrant" of an hour instead of the corresponding clock hour under Subpart Da, revised without notice and comment, and the revision to the Subpart KKKK "out of control" definition.
Commenters 9425, 9666 stated that having inconsistent definitions for valid data make programming difficult, results in inconsistent data sets, and inappropriate invalidation of data. Commenter 9666 stated facilities would have to have systems in place to track the different data validation rules and how data under each rule are affected.
Commenter 9666 believed that the Subpart Da definition of out of control would not result in a greater level of data availability than the Part 75 definition, citing issues with calibration error tests and RATA.
Commenters 9425, 9666 stated EPA should revise the definition in subpart Da to be consistent with Part 75, adopt a consistent definition in Subpart KKKK, and defer to Part 75 definition in Subpart TTTT.
Commenter 9666 stated EPA must clarify whether additional provisions apply for invalid data in Part 75, such as when certain types of changes are made to the monitoring system or if QA test is not conducted on time.
Commenter 9661 supported use of data only from valid operating hours. Commenter 9661 stated that since the substitute data provisions are intentionally punitive in nature to provide an incentive for companies to maximize the availability of CEMS, use of substitute data would lead to intentionally overstated CO2 emissions, unnecessarily putting compliance demonstration at risk. Commenters 9426 and 10243 supported using data only from valid operating hours and not using substitute data provisions of part 75. Commenter 9426 supported the use of only operating hours for which you have valid data for all the parameters used to determine the hourly CO2 mass emissions and gross output data.
Commenter 9426 noted the rule is silent on the matter of how to handle substituted flow or CO2 data that are applicable to Part 75, and is also silent on how to handle flow or CO2 data that have had "full scale range exceedance" values applied to hourly averages, which are also applicable to data used for compliance with Part 75. Commenter 9426 recommended the rule should stipulate that neither substituted data (flow or CO2) to be used in the emissions calculation, nor any hourly data containing "full scale range exceedance" data, should be used in the emissions calculation. Commenter 9426 stated those hourly averages should be considered monitor downtime and so identified in the quarterly summary of emissions and monitoring system performance. 
The EPA has included in the final rule a definition of valid data and additionally has required that only valid data be used in determining compliance with an affected EGUs emission standard. The EPA is also adding more detail with respect to the monitoring provisions that will be used in accordance with part 75 procedures already in place. The EPA has additionally in the final rule expanded upon the requirements of the rule with respect to the requirement to not use substitute data that is allowed for in part 75 provisions as the EPA found that substitute data is not appropriate to use when determining compliance with a rate-based emission standard.  
Backup Monitors
Commenters 9425, 9666 stated EPA should allow, but not require, use of backup monitoring systems under Part 75.
Commenter 3862 asked for an allowance to use backup monitoring equipment in case of a primary monitoring failure. 
The final rule does not prohibit the use of back up monitoring systems allowed for under part 75.
95 Percent Data Availability
Commenter 9514 supported the proposed 95 percent requirement.
Commenter 9666 noted that CO2 MS may not be able to meet the 95 percent data availability requirement if the Part 75 data validation/invalidation provisions are used. Commenter 9666 stated they did not object to invalidation of data under Part 75 because  it does not treat missing data or missed QA test as a violation, and because it allows facilities to petition EPA for alternative data validation and substitution procedures.
Commenter 9425 stated EPA requires use of "all valid data" collected by the monitoring systems (including data recorded during startup, shutdown, and malfunction) to assess compliance, however, only those operating hours during which valid data are collected for all required parameters would be used. Commenter 9425 noted that having to distinguish between "valid" and "non out-of-control" data types will impose a significant burden, without gaining much additional data to report for GHG as compared to Part 75 reporting. Commenter 9425 stated they would support the option to report "'non-out-of-control" data if we need to find "valid" data to maintain > 95% data capture.
Commenter 9666 stated a rule with 95 percent data availability requirement should have an alternative validation procedure.   Commenter 9666 indicated EPA should either (1) make clear that data are invalidated under this rule unless a performance test is failed (i.e., the MS is out of control), or (2) include a procedure to validate data that does not otherwise meet Part 75. Commenter 9666 stated EPA must take into account in the ICR the development and maintenance of separate databases for the two rules.
Commenters 9425, 9666 stated EPA should allow a petition to validate data, similar to Part 75 with respect to alternative data substitution procedures. 
Commenter 9425 indicated the rule should provide an alternative allowed number of missing data hours for units that meet 72.2 definition of peaking unit or other limited operation.
The EPA is finalizing the requirement of the use of valid data in all calculations for compliance with an applicable emission standard and is retaining the requirement of 95% data availability.  Additionally the EPA is not allowing in this final rule for the use of any substitute data as the EPA does not believe it is appropriate to use for complying with the rate-based emission standards put in place by the final rule.
Missing Data
Commenter 9666 supported reliance on part 75 monitoring procedures, without the missing data substitution.
Commenter 8937 supported the use of data only from valid operating hours, avoiding the incorporation of Part 75 substitute data in the calculations.
Commenters 9425, 9666 agreed EPA should not use Part 75 missing data procedures/substitution. Commenters 9425, 9666 stated EPA should allow, but not require, use of backup monitoring systems.
Commenter 9471 opposed mandatory use of missing data procedures. Commenters 2590, 9471, and 9666 stated missing data procedures can overstate emissions, and EPA has not required missing data for EGU NSPS in the past. Commenter 9471 stated EPA has not provided justification for its use.
Commenter 10095 stated EPA should not allow substitute data, and should adopt the same approach as MATS.
Commenter 9425 should provide an alternative allowable number of missing data hours for units that meet section 72.2 definition of a peaking unit or otherwise has limited operation.
Commenters 9593 and 10100 urged EPA to allow use of Equation G-4 in Appendix G of Part 75 during CO2 CEMS downtime.
The final rule includes a definition of "valid data", which defers to the data validation criteria in Part 75. Section §60.5540(a)(1) also defines a "valid  operating hour" as one in which: (1) "valid data" are obtained for all parameters that are used to calculate the CO2 mass emission rate (tons/hr); and (2) the corresponding gross output value is available. Part 75 substitute data values are not considered to be valid data. Also, for the purposes of Subpart TTTT, an operating hour in which there is a full-scale exceedance for a parameter that is used to calculate the CO2 mass emission rate (tons/hr) is not considered to be a valid operating hour.
The final rule retains the 95 percent data availability requirement and ties it to the definition of "valid operating hours". For units that operate frequently, at least 95 percent of the operating hours in each compliance period must be "valid operating hours".
The rule includes a provision allowing the use of backup monitors, in accordance with §§75.10(e) and 75.20(d).  
Violation of Monitoring Requirements
Commenter 9666 objected to EPA's assertion that failure to collect and record required data is a violation of monitoring requirements, indicating that EPA includes exceptions for monitoring system malfunctions, repairs associated with monitoring system malfunctions, and required monitoring system QA/QC activities. Commenter 9666 stated that not all maintenance can be identified in a QA/QC plan, EPA has never included such a definition in an NSPS, and EPA has not provided a reason for doing so in this subpart. Commenter 9666 stated that if EPA includes such a provision, then both periods of scheduled and unscheduled monitoring system maintenance must be excluded. Commenter 9666 further noted that if such maintenance is defined as a deviation, EGUs may be forced to wait until a monitoring system malfunctions or fails a QA/QC test before taking action.
The final rule does not include the statement to which the Commenter objects---that statement was only in the preamble, not in the proposed rule text. The final rule requires the continuous monitoring systems to be installed, certified, maintained, and operated according to Part 75. Section 75.5(e) allows for temporary interruptions of emissions monitoring for various reasons, including times when maintenance must be performed.
Commenters 9780 and 10095 supported not double counting operating days that were determined as violations under the previous compliance averaging period. Commenter 10095 stated that EPA must clarify, for all sources, how civil penalties for violations will interact with the proposed applicability criteria and compliance period. Commenter 10095 further insists that EPA first clarify their intent and then allow the public to review and comment on EPA's implementation of the proposal.
  Section 113 of the Clean Air provides statutory criteria for the evaluation of a violation and the assessment of civil penalties. The EPA evaluates any instance of noncompliance on a case-by-case basis to determine the appropriate response and resolution.
Commenter 10100 supports the exceptions to continuous monitoring obligations during the periods identified in the Proposal but does not believe EPA should limit the repair exception to repairs associated with monitoring system malfunctions. Commenter 10100 noted that requiring demonstration of a malfunction for each period excluded due to CEMS repairs would be burdensome and is not appropriate since CEMS may incur downtime for a variety of reasons that are not related or traceable to a specific malfunction. Commenter 10100 further noted that there are no environmental or human health impacts associated with the failure of a CEMS, making any malfunction demonstration unnecessary. Commenter 10100 supports EPA's proposed exception for monitoring during QA/QC periods, but stated that EPA should also include other exceptions that are allowed under the Acid Rain Program, including periods of preventative maintenance and repair.
The final rule requires the continuous monitoring systems to be installed, certified, maintained, and operated according to Part 75. Section 75.5(e) allows for temporary interruptions of emissions monitoring for various reasons, including times when maintenance must be performed.
Commenter 9425 does not support EPA's approach to define deviations of the monitoring requirements, in particular EPA's failure to account for monitor downtime due to maintenance activities. Commented 9425 noted that not all maintenance can be reasonably identified in a QA/QC plan and further noted that EPA had not included any rule language associated with this proposal in proposed Subpart TTTT.
The final rule requires the continuous monitoring systems to be installed, certified, maintained, and operated according to Part 75. Section 75.5(e) allows for temporary interruptions of emissions monitoring for various reasons, including times when maintenance must be performed.
Carbon Capture and the Calculation of lb/MMBtu Emission Values
Commenter 9666 noted that while Part 75 and Subpart Da allow use of an O2 CEMS in lieu of a CO2 CEMS, an O2 CEMS cannot obtain accurate CO2 measurement for EGUs that use carbon separation because the calculation relies on F-factors.  Commenter 9666 stated any equation that uses an F-factor will not provide accurate measurements downstream of any device that removes CO2 from the gas stream.
Commenter 9666 noted an F-factor represents the ratio of the gas volume of the products of combustion to the heat content of the fuel; the calculation for lb/MMBtu from CEMS data is dependent on the known relationship between the percent diluent gas produced from complete combustion of a fuel and the fuel's heat content. Commenter 9666 further stated that when one of the assumptions is changed, for example fuel is not completely combusted or CO2 is removed by CCS prior to measurement, the methodology is no longer accurate. Commenter 9666 indicated that EPA addressed these issues in ARP by allowing a diluent cap in place of O2 or CO2.
Commenter 9666 provided additional information in an attachment to its comment letter; with removal of CO2, the stoichiometric combustion assumption is invalid, the heat input is underestimated, and the lb/MMBtu rate is overestimated. Commenter 9666 indicated that removal of CO2 invalidates the use of diluent O2 monitors for CO2 concentration determinations, because this approach also uses F-factors and the stoichiometric combustion assumption is invalid. Commenter 9666 noted that pollutant concentrations are acceptable for determining lb/hr emissions if volumetric flow is measured because equations in Appendix F4 of Part 75 do not depend on F-factors.
Commenter 9666 stated if EPA promulgates an NSPS that can only be met by removing CO2 from the gas stream, EPA must recognize that the compliance procedures mandated in EPA and state CAA rules and permits will no longer be valid and overstate actual emissions in lb/MMBtu. Commenter 9666 further indicated that if either the diluent concentration will also have to be measured upstream of CO2 removal or the percent CO2 removed will also have to be measured and added back in, then EPA must account for the costs of that additional monitoring in the ICR.
Commenter 10039 noted that by requiring carbon capture from coal fired EGUs, the diluent ratios in the stack gases are changed and cannot be monitored in the manner required under ARP. Commenter 10039 noted that the rule does not account for this impact.
Commenter 10039 also noted that "monitoring for a minimum of two separate flue gas conditions" will be required. Commenter 10039 further stated that "One to include carbon capture systems functioning (i.e. reduced CO2 levels in flue gas) and the other would be with carbon capture systems off (i.e. normal CO2 levels in flue gas)."
EPA agrees that when CO2 is removed from the flue gas, accurate CO2 concentrations cannot be determined using an O2 monitor and Equation F-14a or F-14b in Appendix F of Part 75. Therefore, the final rule allows the use of data from a certified O2 monitor to calculate CO2 concentration only if the unit does not use carbon separation (e.g., carbon capture and storage).
Part 75 Monitoring Methods
Commenter 9666 supported use of Part 75 monitoring procedures, and has noted their comments on EPA's language, which in some cases is overly broad or duplicative.
Commenters 9666 and 10023 objected to elimination of some Part 75 monitoring options and the imposition of additional measurement requirements on new units. Commenter 9666 and 10023 stated EPA has provided no justification for requiring EGUs to perform monitoring that is different from, or more stringent than, Part 75. Commenter 9666 and 10023 noted EPA has explicitly asserted in its ICR that monitoring is not different or more stringent.
Commenter 9666 did not object to the proposal to report the calculated CO2 emission rate on a quarterly basis.
Commenter 9666 did not object to the proposal to report the calculated CO2 emission rate on a quarterly basis.
Commenter 10100 requested annual reporting instead of quarterly reporting. 
Commenter 7977 asked that EPA streamline all monitoring, recordkeeping, and reporting requirements to be consistent with other air quality programs, such as the ARP.
Commenter 10681 noted that, "instead of utilizing the preexisting monitoring, recordkeeping, reporting, and calculations, EPA has instead added to the complexity for a source and regulator to determine compliance with all requirements" of subparts Da, KKKK, and Part 75.
The final rule defers to the monitoring, recordkeeping, and reporting requirements in Part 75. 
