Chapter 7
Standards for Stationary Combustion Turbines

Contents
7.1	General	2
7.2	BSER	2
7.2.1 Alternatives	9
7.3	Proposed Emission Limits	10
7.3.1 Proposed Standards Are/Are Not Achievable	21
7.3.2 Proposed Standards Should Be Based on Actual Emissions Data	25
7.3.3 Addressing Factors Beyond Operator's Control	27
7.4	Other Subcategories	33
7.4.1 Peaking Units, Intermediate Load Units, and Base Load Units	33
7.4.2 Simple Cycle Units	36
7.4.3 Other Fuels	40


General 
Multiple commenters (9665, 9666, 9678, 9770, 9779, 10023, 10103, 10693, and 10952) offered general support for the proposed standards.
Commenter 9665 stated, "The Agency correctly predicts that natural gas-fired electric generating units (EGUs) likely will be the 'facilities of choice' for new electric power generation for the foreseeable future." Compared to other fossil fuel choices for new generation, NGCC produce lower overall emissions of CO2, criteria pollutants, and hazardous air pollutants. Similarly, NGCC generation eliminates the difficult environmental issues of safely disposing of the solid waste by-products that necessarily come with new coal-fired units.
Commenter 9678 encouraged EPA to "continue to update its assumptions with the most recent assessments from the U.S. Energy Information Administration (EIA) before finalizing the rule."
Commenter 10693 supported "the increased use of natural gas as a transitional fuel for the production of electricity as a cleaner alternative to biomass, coal and other fossil fuels" and "requiring the installation and operation of state-of-the-art pollution control systems at new and existing natural gas-fired power plants."  
 We thank the commenters for their support. The final rule establishes separate standards of performance for fossil fuel-fired electric utility steam generating units and natural gas or multi-fuel fired stationary combustion turbines. Where possible, EPA updated its assumptions to utilize more recent data from the EIA and other data sources. 
BSER
Multiple commenters (3862, 8952, 9201, 9381, 9425, 9591, 9648, 9656, 9660, 9665, 9666, 9678, 10095, 10098, and 10239) supported EPA's conclusion that natural gas combined cycle (NGCC) represents the best system of emission reduction (BSER) for natural gas-fired stationary combustion turbines. 
 We appreciate the commenters' support. We are finalizing BSER determinations for the three subcategories of stationary combustion turbines: base load natural gas-fired units, non-base load natural gas-fired units, and multi-fuel-fired units. For newly constructed and reconstructed base load natural gas-fired stationary combustion turbines, the BSER is the use of efficient NGCC technology. For newly constructed and reconstructed non-base load natural gas-fired stationary combustion turbines, the BSER is the use of clean fuels (i.e., natural gas with an allowance for a small amount of distillate oil). For multi-fuel-fired stationary combustion turbines, the BSER is also the use of clean fuels (e.g., natural gas, ethylene, propane, naphtha, jet fuel kerosene, distillate oils 1 and 2, biodiesel, and landfill gas). Section IX.C.4 of the preamble to the final rule provides additional details. 
Multiple commenters (3862, 8952, 8973, 9201, 9381, 9425, 9426, 9591, 9648, 9656, 9665, 9666, 9678, 9723, 9770, 9780, 10023, 10095, 10098, 10239, 10392, and 10660) agreed with EPA's conclusion that carbon capture and sequestration (CCS) does not represent BSER for natural gas-fired stationary combustion turbines.
Commenter 8952 noted that the gas turbine has been improved over the last 50 years, but "none of the innovations has included technology to efficiently and cost-effectively extract carbon dioxide from the exhaust of either a simple cycle or a combined cycle gas turbine." 
Commenter 9426 noted that "A finding that partial or full CCS is BSER for NGCC units would effectively prevent the development [of] new NGCC projects because [the commenter] does not believe any company will construct a NGCC unit that requires CCS, nor would financing for such a project likely be available." Commenter 9426 further stated that "taking NGCC technology out of the mix would leave the power sector with few options for new base load and intermediate generation, which would create serious consequences for the delivery of reliable supplies of electricity."
Related to the use of CCS with natural gas combustion turbines, commenter 9665 stated the following:
 EPA needs to ensure that before it imposes a CCS requirement that the stability of electric supply will not be impacted.
 Enhanced oil recovery (EOR) cannot be implemented nationally
 "Lack of regulatory clarity surrounding long-term stewardship and liability for stored CO2"
 "If CCS were established as BSER for natural gas, significant adverse increases in electricity prices are certain. This is due to the following factors: (1) costs would be relatively higher for natural gas-fired CCS versus coal-fired CCS on a cost/ton of CO2 basis; (2) EPA's cost analysis is dependent on little or no CCS deployment until the mid-2020s; and (3) any future demonstration project using partial CCS for NGCC will depend heavily on government funding."
 
Commenter 9666 stated that "no company will construct an NGCC EGU if required to incorporate CCS, nor would financing likely be available to do so even if a company had such an interest. Such a decision would have serious consequences for the delivery of reliable supplies of electricity in this nation because the predominant means of providing coal and gas-fired generation would become unavailable."
Commenter 9678 agreed with EPA's conclusion that requiring CCS on natural gas units would adversely affect the nation's energy needs and the environment. The commenter stated that the EPA has properly recognized that imposing partial CCS on natural gas-fired combustion turbines would potentially delay such projects more than at a coal-fired EGU. Additional factors that make CCS more challenging for natural gas combustion units compared to coal-fired EGUs include the time it would take to complete the CCS project and the water use requirements. According to the commenter, delays to the construction of new NGCC units would in turn impede the transition to a lower-emissions national electricity fleet and impede the overall reduction of CO2 emissions that is the object of this rulemaking. The commenter stated that NGCC units are projected to provide a significant share of new power generation, EPA should recognize that requiring partial CCS on NGCC units would have proportionally higher impact on electricity prices. 
Commenter 9780 noted that "requiring all new natural gas CTs to use CCS would have significant adverse effects on national electricity prices, electricity supply, and the structure of the power sector." Commenter 9780 asked that EPA "bolster the record as to the technical infeasibility and unreasonable costs of CCS on natural gas-based units." 
Commenter 10392 agreed with EPA's determination that "cycling/load change operations of CTs would negatively impact the CCS process, that the CO2 concentration in the exhaust of CTs is too low to be effective, and that increased CT unit cooling requirements as a result of CCS would lower unit efficiency and raise emissions." 
Commenter 10660 noted that CCS cannot be BSER for natural gas fired turbines due to the following:
 "unresolved issues of disposal (i.e., storage and sequestration) of CO2"
 "the feasibility of national implementation"
 "available data show the cost of CCS for gas turbines would be unreasonable"
 "if EPA were to determine CCS is BSER for natural gas fired units, the action would destabilize the nation's electricity grid, by effectively preventing the construction of new natural gas fired units"
As explained further in Section IX.C.4 of the preamble to the final rule, we evaluated partial CCS as an option for base-load natural gas-fired combustion turbines and concluded that CCS does not represent the BSER for the universe of combustion turbines that are covered by this final rule. 
Multiple commenters (7537, 8501, 9513, 9646, 10087, 10119, and 10869) raised questions about the determination that CCS is not BSER for stationary combustion turbines.
Commenter 7537 noted that "the President's budget includes $25-102 million to fund CCS for natural gas projects. If one of these projects becomes operational, would that be sufficient for EPA to begin requiring CCS as part of the NSPS or the PSD permitting process? What is the goal of these efforts? Will EPA be working with DOE on these projects?"
Commenter 8501 stated that the justification for rejecting CCS as BSER for NGCC is "not supported by facts of the record." Further, Commenter 8501 noted that "EPA attempts to support its decision to treat NGCC differently than coal-fired facilities on the basis that NGCC facilities start up and shut down more frequently than coal fired facilities, and thus are not good candidates for CCS. This operating characteristic of NGCC facilities, however, is not an impediment to using partial CCS. NGCC operators could bypass the carbon capture system during startup and shutdown modes (which are typically shorter and less intensive efforts compared to startup or shutdown of a coal facility) and then employ the carbon capture system when operating normally."
Commenter 9513 noted that CCS is "technically feasible at this time for all electric generating units...including gas-fired plants" and that CCS is now "demonstrated and even commercially available for gas plants." Commenter 9513 further noted that "requiring CCS at all new gas-fired generating facilities immediately might prove challenging for the testing, permitting, monitoring and implementation activities that would need to occur in a fairly short timeframe to develop adequate sequestration capacity. Furthermore, the economics of requiring CCS at gas plants at this time might discourage a transition to newer, more efficient gas-fired units in favor of maintaining older, less efficient fossil fuel-fired plants."
In relation to the BSER analysis for gas fired combustion turbines, commenter 9513 noted the following:
 "It may be more manageable to designate CCS as BSER for plants regulated under subpart Da in this rulemaking, and defer until the subsequent NSPS revision whether CCS is BSER for subpart KKKK sources. However, we strongly urge EPA to revisit this question during the next phase of NSPS revisions. At that time, it should consider the actual, on-the-ground experience of CCS deployment, particularly at new subpart KKKK facilities for which CCS has been determined to be the best system of emissions reduction (BACT) under the Prevention of Significant Deterioration (PSD) program, 42 U.S.C. Sections 7470-7479, in the intervening years. The case-by-case analysis of BACT is a reasonable and productive way to introduce new technology in an industrial category without creating the economic and environmental problems EPA has identified.
 "To the extent that the factors discussed above support EPA's determination that CCS is not the best system of emission reduction (BSER) for gas-fired EGUs at this time, they do not preclude a finding that it should be the basis for BACT determinations for specific new plants. We are concerned that EPA's rationale for rejecting CCS as CCGT, which contains technical inaccuracies and is overly broad, may be relied on by sources and permitting authorities as an excuse to reject CCS as BACT in all future permits. We therefore ask EPA to correct those inaccuracies in this record."
Commenter 9646 stated that "CO2 removal is just as technologically feasible for a Natural Gas Combined Cycle (NGCC) EGU as a coal-fueled EGU. Some literature suggests 40% removal capabilities, and there would be no expected difference with the sequestration issues between the two types of EGUs." 
Commenter 10087 noted that EPA did not cite the DOE/NETL study which explains that CCS for natural gas is cheaper than coal. Commenter 10087 further stated that "EPA does not provide any justification...nor does EPA explain why DOE/NETL's assessment is incorrect on the issue of CCS and natural gas."
Commenter 10119 noted that "NGCC with CCS need not be actually in use in order to be considered 'adequately demonstrated.'" Commenter 10119 stated that the key question is "whether evidence supports a reasonable projection that the technology is actually available for installation at new facilities." Commenter 10119 noted that EPA did not address whether costs or other nonair quality environmental impacts would preclude its use. Commenter 10119 noted that "information in the docket suggest that the cost might not be unreasonable."
Commenter 10869 stated, "In the future EPA should also consider setting a standard for natural gas with CCS. After all, natural gas is a fossil fuel and is contributing to the buildup of atmospheric carbon dioxide. In order to meet long-term emission reduction targets that will be needed to avoid the worst impacts of climate change, natural gas will eventually need to be deployed with CCS or replaced with other lower and zero carbon technologies." 
As explained further in Section IX.C.4 of the preamble to the final rule, we considered the commenters' arguments, and have concluded that there is not sufficient information at this time for us to determine that CCS is adequately demonstrated for all base load natural-gas fired combustion turbines. Many NGCC units do not operate at the steady state conditions under which CCS has been adequately demonstrated, and we are not aware of any pilot-scale CCS projects that have demonstrated how fast and frequent starts, stops, and cycling will impact the efficiency and reliability of CCS. Unlike for coal-fired units, the Department of Energy has not yet funded a CCS demonstration project for a NGCC unit, and no NGCC-with-CCS demonstration projects are currently operational or being constructed in the U.S. 
The NSPS sets the minimum level of control for new sources. We expect that state air agencies and other air permitting authorities will evaluate CCS when permitting new NGCC power plants, taking into consideration case-specific parameters, like operating characteristics, to determine whether CCS could be BACT or LAER in specific instances. While the NGCC-with-CCS units that currently are in the planning stages do not provide us with enough assurance to determine that CCS is adequately demonstrated for combustion turbines, it is our expectation that these units and others to come will provide additional information for both permitting reviews and the next NSPS review in eight years.
Commenter 9191 stated that the EPA's asserted rationale for distinguishing between natural gas and coal fired generating units is illogical and does not reflect decision-making considerations consistent with Section 111 of the Clean Air Act. The commenter stated that each factor identified by EPA as a basis for requiring CCS technology for new coal-fired electric generating units but not for new gas-fired units reflects an inherent EPA bias toward gas-fired units rather than an evaluation of the best system of emission reduction adequately demonstrated, which is the appropriate Clean Air Act criterion.
The EPA disagrees with this comment. The EPA's determination that partial CCS is adequately demonstrated for coal-fired EGUs, but not natural gas-fired EGUs, is based on a thorough technical record explaining the differences between these two types of sources, not bias.
Multiple commenters (7976, 8952, 9194, 9201, 9396, 9427, 9471, 9514, 9665, and 10119) disagreed with the BSER analysis, suggesting other factors that should have been or should later be considered, and noting areas where clarification is needed in the BSER analysis for NGCC.
Commenter 7976 offered the Gas Turbine World Handbook as a source of technologies that could have been included in the BSER analysis. According to commenter 7976, "many underestimate the value of modern combined cycle" and "it is important to consider what technology is available and reasonably feasible going forward, as compared to only considering older, less efficient technology in currently operating plants. The Gas Turbine World Handbook is an example of a reasonably forward looking list of available and feasible technologies."
Commenter 8952 stated that "In its BSER analysis, EPA should start with the premise that the market itself will drive the use of the most efficient power, such that regulatory constraints must be carefully crafted not to interfere with these incentives."
Commenter 9201 noted that EPA did not examine new technology or different configurations of NGCC. 
Commenters 9513 and 9514 asked that EPA establish standards for gas plants based on "highly efficient CCGT technology." According to commenter 9513, "EPA has not proposed to adopt as BSER the most efficient technologies employed within the CCGT category." Commenter 9514 included a table listing efficient CCGT models from recent years, noting the capacity in MW, efficiency, and heat rate.
Commenter 9514 noted a flaw in the EPA's reference to a figure labeled "Heat Rates for Commercially Available CCGT Units in 2011" from the 2011 GTW Handbook (which was used in the initial rule proposal). Commenter 9514 stated that the differences in efficiencies in the figure are "the consequence of deliberate decisions by designers to incorporate features and systems that enhance combustion and permit a greater amount of electricity generation per unit of fuel" and do not represent "inherent limitations in technology that prevent better performance." 
Commenter 9514 stated that "better-performing CCGT units have adequately demonstrated that lower emissions are achievable," referencing the Technical Support Document (TSD). Commenter 9514 noted that the lower emission rates were achieved even though:
 "the operators of existing sources were not attempting to meet a defined level of CO2 emissions performance;
 EPA's study pool included a number of inefficient designs; 
 given permitting, financing, and construction schedules, all plants included in EPA's data pool/study must have been between six and 17 years old; 
 EPA employed the 'average of monthly averages' discussed elsewhere in these comments in deriving the proposed emissions rate from the study data."
Commenter 9665 stated that "EPA should be designing these emission standards to encourage the deployment and market penetration of lower-emitting technologies. Otherwise, any emission benefits obtained by a tighter standard will likely be offset by the negative impact of reducing the amount of investment in new, more efficient NGCC units." 
Commenter 10119 stated that the analysis failed to consider "what may fairly be projected for the regulated future" and instead relies only on emissions data from existing facilities. 
Commenter 10119 stated that the rule "lacks any analysis weighing the potential emission benefits of more stringent BSER and emission limits against specific energy, economic, or environmental impacts." It is the responsibility of EPA to demonstrate the "operational flexibility" concerns of the rule that prevent a more stringent BSER determination rather than asking that "parties recommending lower emission limits address general concerns about 'operational flexibility.'" The data in the TSD Memo demonstrates that "variations in temperature, pollution controls, elevation, and even cycling were found to not have a substantial effect on CO2 emission rates." The commenter concluded that "a less stringent standard may not be necessary to accommodate a wide range of 'operational flexibility' concerns."
Commenter 10119 stated that the rule "relies explicitly on power sector models that predict increasing use of cheap natural gas obtained using unconventional drilling methods." The commenter noted that the rule failed to consider the environmental and other impacts of these methods. 
Commenter 10119 stated that the BSER analysis should consider new, advanced NGCC or more efficient NGCC due to the following:
 advanced NGCC technology is "capable of achieving lower emission rates than 'conventional facilities'"
 advanced NGCC facilities have a base load rating emission rate of 760 lb CO2/MWh, 8% lower than conventional NGCC facilities 
 EPA did not explain any "negative impacts associated with advanced NGCC technology that might outweigh the possible emissions benefit"
The EPA has concluded that it is appropriate to use historical NGCC operating data for establishing a reasonable emission standard for multiple reasons. First, the base load subcategory includes both fast-response and traditional NGCC designs. Fast-start NGCC units are specifically designed to startup and ramp quicker than traditional NGCC units. While these characteristics make fast-start NGCC units useful for backing up intermittent renewable generation, the units are not as efficient operating at steady state base load conditions as traditional NGCC designs. The final emission standard includes enough compliance margin to assure that NGCC units can continue to provide backup power and support additional generation from intermittent renewable sources. Second, incremental improvements to NGCC designs are not unique technologies. These "advanced" NGCC designs were included in the EPA's analysis. Operating parameters can have a larger influence on the emission rate of an NGCC unit than incremental technology improvements. For these reasons, our data set includes a large population of technologies and load conditions to ensure that new NGCC units can achieve the final emission standard in all regions of the country. See Sections IX.C.3 of the preamble to this final rule for additional information regarding the EPA's rationale.
Commenter 9427 noted that the BSER analysis did not include "existing combined heat and power (CHP) facilities or other facilities of the type that would be subject to the proposed standard if they were new facilities." Commenter 9427 added that EPA should "complete the analysis to include data for substantially more existing stationary combustion turbine facilities in the database and adjust the compliance timeframe and/or level of the emission standard accordingly." 
CHP units are not currently required to submit both electric and useful thermal output data to CAMD. Therefore, we were unable to do a detailed analysis specific to CHP units. However, one of the benefits of CHP units is that they are substantially more efficient than electric-only stationary combustion turbines. The final rule gives 100 percent credit for useful thermal output when calculating the emission standard, and the corresponding emission rate for a well-designed CHP unit is approximately 600 lb CO2/MWh-g. We have therefore concluded that the final emission standard is achievable for any potentially affected new CHP units. 
Commenters 9194, 9396, and 9427 remarked that the EPA's BSER analysis is inadequate because a majority of the units in the database were not included in the BSER analysis:
 Database contains 1,027 units
 about 282 combustion turbine units would meet the proposed standards. 
 about 25 units would not have met the standard
 720 units were not considered in the BSER analysis because emission rate data were not developed 
Commenter 9471 stated, "With respect to the proposed gas standard, vendor information reviewed in planning and designing our current natural gas combined cycle (NGCC) project indicates that the standard is not continuously achievable for new NGCC units. In addition, the EPA's proposal sets a bad precedent for any future standards that may be promulgated for existing or modified sources and creates serious problems with respect to Best Available Control Technology (BACT) determinations for modified units under the Prevention of Significant Deterioration (PSD) program." 
 The EPA analysis included all non-CHP NGCC units that reported the output from the steam turbine. The EPA has concluded that its analysis of the actual emissions data for these 345 NGCC units demonstrates that the final standard is achievable. These units are diverse in size, operating parameters, and site-specific factors (e.g., ambient conditions and cooling type), and the results are appropriate for a nationwide standard. In addition, our analysis of design efficiency data in conjunction with recent GHG permits for NGCC units indicate that the final standard is achievable by the vast majority of existing NGCC designs.
Commenter 9665 asked that EPA "clarify that the NSPS for combined cycle units will not set the BACT floor for simple cycle units under the PSD program." 
The NSPS establishes the least stringent standard (i.e., the BACT floor) for new units that require permitting under the PSD program. Any new combustion turbine that sells electricity in excess of the percentage electric sales threshold would have to comply with an emission standard of at least 1,000 lb CO2/MWh-g. A developer of a new natural gas-fired facility would decide at the time of construction if they intend to exceed this sales threshold. New simple cycle turbines that limit their electric sales will only have to comply with the clean fuels standard. Therefore, the BACT floor for new simple cycle units is the use of clean fuels. A site-specific analysis would be performed to determine if a more stringent standard is appropriate.
Commenter 9665 noted that "forcing generators to deploy combined cycle units in lieu of simple cycle units may force the operator to run the plant during low load periods to be prepared to rapidly respond to grid demands instead of shutting the unit down completely."
 The EPA disagrees that the subcategory approach will force operators to deploy NGCC technologies. The non-base load natural gas-fired unit subcategory provides adequate flexibility for new simple cycle turbines to continue to operate as they historically have.
According to Commenter 10119, EPA "cannot establish BSER for a subset of new natural gas plants based on the least efficient existing technology, and further leave emissions from other facilities entirely uncontrolled despite the existence of achievable methods for reducing those emissions."
 The final rule establishes appropriate emission standards for the vast majority of new stationary combustion turbines that sell significant quantities of electricity to the grid. Specifically, the final applicability is a modified version of the broad applicability approach the EPA solicited comment on. In conjunction with this applicability, the EPA established an appropriate BSER for three subcategories of stationary combustion turbines. 
7.2.1 Alternatives
Multiple commenters (2984, 3175, 7976, 9513, 9514, 9771, and 9780) offered alternatives to the proposed BSER.
Commenter 2984 requested one of the following modifications "for states with an RPS exceeding 25%:
 Change the performance standard for smaller natural-gas fired units (equal to or less than 250 MW) to be measured at full capacity rather than over all hours that include stops and starts. This would likely enable simple cycle combustion turbines to meet the performance standards.
or
 Increase the 33% capacity factor exemption to 40%. This would likely exclude simple cycle combustion turbines from the performance standards."
Commenter 3175 recommended, "For affected units with a heat input of less than or equal to 850 MMBtu/h...we would support an annual test at base load to demonstrate compliance to an emissions rate lower than 1,200 lbs. CO2/MWh, as opposed to the proposed gross emissions divided by net energy exported over a rolling 12 month period."
Commenter 9771 noted the following:
 "Some combined cycle EGUs that came on line in the last ten years have been designed to be efficient load-following units. They are part of a transition that uses a combination of hydropower, older steam boilers and new flexible units to integrate the first waves of variable renewables. As California has rapidly added variable renewable generators as part of its plan to de-carbonize the electricity sector, the incremental need for flexibility, which is critical for maintaining grid reliability, has also increased. "
 "In California, more than 7,000 MW of central-station and distributed solar capacity now serves California loads [reference California ISO, California Solar Initiative data]. Another 5,000 MW of variable output central-station renewable capacity has been contracted for by the state's investor-owned-utilities and is expected to come on line within the next three years [reference: California Public Utilities Commission, Renewable Portfolio Standard Project Status Table, February 2014]. Another 3,000 MW or more of new distributed generation, mostly solar, is forecasted through 2020 [reference: California Energy Demand 2014 - 2024, California Energy Commission, Form1-4, Statewide Peak Demand 2014-2024 Baseline Forecast]. Integration of these variable output resources requires a thermal generation fleet that is both flexible enough to sit idle or at a very low output to allow absorption of the solar energy generated at midday, and quickly contribute up to 13,000 MW of firm, dispatchable resources between 4:00 and 8:00 PM, as the sun sets and peak evening demand occurs "
 "Flexible EGUs with the ability to cycle and ramp as described above will have different emissions profiles compared to EGUs that are operated as either base load or peaker units. Frequent cycling and ramping will decrease thermal efficiency and increase GHG emissions per MWh compared to units providing base load energy. Even when the flexible resource is operated at a full steady state, the very engineering design that enables the flexible operation may also reduce thermal efficiency. Newer EGU projects include multiple modules that allow incremental dispatch of the modules, each able to start and shut down quickly once or more often each day and rapidly ramp up and down. Existing units are being similarly dispatched, demonstrating the increasing need for flexibility while maintaining grid reliability as California de-carbonizes the electricity sector with greater amounts of variable renewable resources. "
Commenter 9780 noted that the standard should provide more flexibility and referenced a few PSD permits that have CO2 emission limits on a ton-of-CO2-per-year basis. For example, Commenter 9780 referenced the PSD permit for the Oregon Energy Center in Oregon, Ohio; the Sevier plant in Utah, the Riverton Power Station project in Kansas, the Garrison Energy project in Delaware, etc.
 The EPA considered the alternative systems of emission reduction, sales thresholds, and emission standards suggested by the commenters before arriving at the agency's final decision on these issues. See Sections IX of the preamble to the final rule for additional details on the EPA's rationale for the final applicability criteria, subcategories, BSER determinations, and emission standards for stationary combustion turbines.
Commenter 9514 noted that the BSER analysis only considered technologies that were "part of the fossil-fuel fired generating unit" and recommended that EPA consider concentrated solar power (CSP) technology in the BSER analysis. 
As noted in the preamble to the proposed rule, the EPA considers hybrid power plant technology to be promising. However, solar hybrid technology is not a viable means of generating additional steam in many areas of the country, so the EPA did not consider this technology as the basis for a nationwide standard.   
Proposed Emission Limits
Multiple commenters (2984, 8952, 9646, 9665, 10136, 10390, and 10606) commented or questioned specific aspects of the emission limit determination.
Commenter 2984 asked that the standard be clarified "so that the applicable emission limits, either 1,000 pounds or 1,100 pounds of CO2/MWh, are based solely on size measured in MW. The 1,000 pounds of CO2/MWh should apply to units larger than 250 MW and the 1,100 pounds of CO2/MWh to units equal to or smaller than 250 MW. The "850 MMBtu/h" criterion in parentheses of the emission limits corresponds not to 250 MW, but to approximately 100 MW as noted in Footnote 85 of the Proposed Rule. This makes the applicability of the standards confusing."
Commenter 9425 and 9780 noted that smaller combustion turbines (i.e., less than 300 MW) may have higher emissions profiles.
Commenter 8952 noted that future market designs are expected to require a high degree of component flexibility, which is expected to come at the expense of the annually averaged combined cycle efficiency, even if the gas turbine is a state-of-the-art machine." Additionally, commenter 8952 stated, "the operational requirements of the system -- not the technical capability of the gas turbine or the combined cycle -- will determine the final emission profile of the facility. In short, the proposed standard does not so much influence the plant (or gas turbine) design, which is already designed to achieve high efficiency operation, but how the plant can be operated. "
Commenter 8952 evaluated the CO2 emission rates of only combined cycle plants and noted "The improvement in performance (using CO2 lb/MWh as the benchmark) is significant when the steam cycle is included in the power generation. This is evident in the published data shown in Figure 5 [page 9 of comment letter]. We also comment that these are not necessarily representative of guarantees provided by the OEM. In fact, for most OEM's the only guarantee will typically be associated with just the gas turbine component (represented by the simple cycle gas turbine emission performance depicted in Figure 4 [page 9 of comment letter]). For many turbine suppliers, the final end-use configuration (a single or multiple gas turbines with steam turbine heat recovery) evident in Figure 5 is beyond their scope of supply. Also, in this figure we note that the simple average of all the calculated performance is not too different from the CEM data that was used to summarize the historical fleet performance. This may be indicative that most of the historical data were obtained when the combined cycles were at are the maximum rated power point."
Commenter 9646 noted these areas for potential improvement/optimization:
 "Various makes/models/sizes/designs of combustion turbines exhibit higher thermal operating efficiencies. Assuming all modem designs, combustion turbines with higher ratings tend to have higher efficiency. Some manufacturers offer better compressor designs, combustor configuration, and higher temperature capabilities than others that tend to improve the combustion turbine (CT) efficiency. The temperature capabilities must be matched to optimize the heat recovery steam generator (HRSG)/steam turbine capabilities in combined cycle configuration, however, General Electric (GE) offers a range of frame mid-to-large size CTs that are advertised at efficiencies ranging from 38% to 40% in simple cycle configuration, and 59% to 61% in combined cycle configurations GE also advertises an aeroderivative CT with a simple cycle efficiency of 42% and a combined cycle unit/combined heat and power (CHP) efficiency of 84%. (Hitachi offers a wide range of sizes and configurations with efficiencies ranging from 34% for small simple cycle CTs to 55% for large combined cycle units.) (Other manufacturer combined cycle unit offerings include up to 59% efficiency from Mitsubishi, and 60% from each of Alstom and Siemens.)
 The combustion turbine/HRSG configuration can impact overall unit efficiency. A configuration where multiple CTs supply exhaust heat to a single HRSG/steam turbine tends to have a higher overall efficiency than a configuration where a single combustion turbine supplies exhaust heat to a single HRSG/steam turbine. (Hitachi combined cycle configuration in a 1x1 is shown to be approximately 3% lower than the same combustion turbines in a 2X1 configuration.)
 Integration of a HRSG and combustion turbine to achieve high and intermediate pressure sections of the HRSG to improve the thermal efficiency of the steam turbine cycle (have a steam turbine with high, intermediate, and low pressure sections).
 Selection of a steam turbine with a dual-flow LP section that facilitates improved optimization of last stage blading to achieve reduced exhaust losses, thereby improving (the steam turbine efficiency).
 Selection of combustion turbine and HRSG design and materials that permit faster startups to reduce the amount of time the unit must operate at low loads and low efficiencies. This is more of an impact of units that would be expected to frequently start and stop.
 Selection of various combustion turbine inlet cooling methodologies to optimize combustion turbine efficiency and minimize impact of higher than design ambient air inlet temperatures.
 Selection of steam turbine condenser cooling equipment and methodology to improve steam turbine efficiency and reduce negative impact of higher ambient temperatures. (Wet cooling usually better than air cooling, once through wet cooling usually better than the use of cooling towers, etc.). "
Commenter 9665 stated that the current emission standards would have the "unintended consequence of limiting operational flexibility essential to the continued reliable deployment of renewable electricity generation. Specifically, other sources of electricity generation need to come on line quickly when wind and solar plants are unable to generate electricity to help ensure an adequate electricity supply and grid reliability. This will, in turn, result in the need to deploy additional electric generating units to meet existing demand, having both economic and environmental effects that are counter to the intended effect of regulating GHGs from the power sector and other sources--to reduce the overall global warming footprint of the country. To explain how this effect would occur, when plant owner/operators bid on opportunities to provide power generation to the grid operators, they must include min-load and max-load levels along with various other cost components. If a bid is successful, the grid operator will comply with the bid's minimum and maximum load specifications. If the proposed 1,000 lb. CO2/MWh limit is promulgated, however, plant operators will need to bid a narrower range of load to ensure that compliance is maintained. This will result in the need for additional capacity that is likely to be less efficient in order to avoid the load restrictions. The impact will be that additional capacity will need to be utilized--which by definition will be less efficient (and higher CO2-emitting) capacity, increased variable electricity cost, and higher total CO2 emissions. These economic, environmental, and energy implications are both necessary and appropriate factors for EPA to consider in the BSER analysis. In fact, just as simple cycle units are critical to supporting renewables, there is also a role for some combined cycle capacity in this regard, which can be an important economic market for these units. If NGCC units must continually narrow their bids, they will likely lose the ability to support that aspect of the renewable market for which they are appropriately dispatched. On the other hand, if EPA sets the limit at 1,100 lb. CO2/MWh for larger NGCC units and 1,200 lb. CO2/MWh for smaller units, deployment of renewable energy will increase and overall CO2 reductions from the system would be expected to decrease." 
Commenter 10390 asked, "Why can or should turbines with heat input below 850 MMBtu/hr meet a limit of 1,100 lb CO2/MWh, but those with a heat input below that level meet a limit of 1,000 lb CO2/MWh? What happens to turbine performance at that threshold that makes the performance change?"
See sections IX.B and IX.D.3 of the preamble to the final rule for EPA's response to this comment. The EPA is not promulgating size-based subcategories for stationary combustion turbines. Methane leakage from natural gas extraction and conveyance is outside the scope of this rulemaking.
Multiple commenters (6501, 6505, 8909, 8952, 9035, 9202, 9591, 9666, 10098, 10103, 10620, and 10991) supported the proposed combustion turbine standards.
Commenter 6505 stated "Both mature and recently commercialized MHPSA combustion turbine technologies are capable of meeting the 1,000 lb-CO2/MWh standard for combustion turbines firing greater than 850 MMBtu/h heat input. Even under a wide range of operating and degradation assumptions, MHPSA's turbine technologies will be capable of meeting the proposed limit."
Commenter 8909 agreed that "larger natural gas-fired units should be subject to a more stringent standard" because "smaller capacity natural-gas units are less efficient than larger units."
Commenter 8952 agreed with a standard between 1,000 and 1,100 lb CO2/MWh. 
See sections IX.B.1 and IX.D.3 of the preamble to the final rule for EPA's response to this comment. After considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion.
Commenter 10136 noted, "Given the information that is emerging about methane leakage from natural gas extraction and conveyance, it is now very clear that stack emissions of natural gas plants do not reflect their actual climate impact. Requiring the lowest stack emissions standard possible would help mitigate the lifecycle and leakage emissions that are not considered under the rule. EPA should for this reason, and many others, set the emission standard for new natural gas plants at the level achieved by the newest, most efficient plants." 
This comment is out of the scope of this final rule.
Commenter 9666 suggested that "EPA set the standards using scientific notation that would clarify the number of significant digits for compliance purposes, which is ambiguous as proposed."
The EPA has clarified that emission standards of 1,000 or above have three significant figures and that emission standards less than 1,000 have 2 significant figures.
Three commenters (9593, 9581 and 9780) suggested some changes in determining compliance with the combustion turbine standards.
Commenter 9591 agreed with the proposed standards, "including the 12-operating month averaging time" and noted that EPA "has provided sufficient flexibility."
Commenter 9593 asked that the "calendar year averaging period where all operating months as defined in the proposal, are included" be used for compliance. 
Commenter 9593 asked the compliance be determined by "dividing the annual total mass of CO2 by the annual MWhs, instead of using the arithmetic average 12 monthly values"
Commenter 9593 agreed with basing the limit on "gross, not net generation; this is consistent with Acid Rain Program, 40 CFR part 75 (Part 75)."
Commenter 9780 asked that compliance be measured on a rolling average basis. 
 As noted in Section IX.D.3.a of the preamble, the final standards for base load combustion turbines will be calculated on a 12-operating-month rolling average basis by dividing the total mass of CO2 emitting during the 12-operating-month period by the total generation during the 12-operating-month period. The final standards include both a gross limit and an optional net limit.
Three commenters (9665, 10102, and 10390) noted that EPA's analysis must distinguish between high heating value (HHV) and lower heating value (LHV) in relation to heat input rates.
According to commenter 9665, the currently-effective Subpart KKKK NSPS standard "included a size cut-point of 850 MMBTU/h heat input, LHV basis." However, commenter 9665 noted that "the final version of NSPS Subpart KKKK included the same cut-point, [but] the basis of the cut-point value changed from LHV to HHV. As a result of this change, the cut-point was shifted to a smaller machine heat input by 11 percent and certain turbines within the gas turbine fleet that were defined as small under the proposal were transformed into large turbines in the final rule."
Commenter 10102 asked that the "850 MMBtu/hr applicability standards be adjusted to account for HHV as the basis for determining compliance with the emission limits."
Commenter 10390 noted, "Why EPA selected either the 110 MW output threshold or 44 percent as a proper assumption for converting that intended target output back to heat input is not mentioned in the final rule's promulgation.
Commenter 10390 noted that the EPA analysis assumed an energy conversion efficiency of 44% LHV "as though it were HHV-derived." Commenter 10390 stated that "the proposed rule appears to use a different heat input calculational basis for setting the limits than it would for identifying the limits that would apply under the rules."
 The policy of the EPA is to use the HHV of a fuel unless specifically stated otherwise. Because the final rule does not include subcategorization of combustion turbines based on size, the comment is no longer relevant. 
Commenter 9514 commented on the difference between one large unit and multiple smaller units when complying with subpart KKKK standards. Commenter 9514 asked, "If EPA determines that different subcategories (for CCGTs) according to size are warranted, it should clarify that multiple small units built together should be treated as a large unit if the combined output of the facility exceeds 850 MMBtu/h." Commenter 9514 noted that "facilities should not be permitted to meet less stringent emission limits merely by building a series of smaller units rather than a single larger unit."
See sections IX.B.1 and B.2 of the preamble to the final rule for EPA's detailed response to this comment. After considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas-fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion. In contrast, non-base load natural gas-fired combustion turbines are subject to a separate heat input-based emission standard of 120 lb CO2/MMBtu. Because emission standards apply to both subcategories of natural gas-fired combustion turbines, there is little incentive for companies to construct a series of smaller units in place of one larger unit.    
Multiple commenters (2984, 7990, 7994, 8349, 8964, 8969, 8970, 8973, 8995, 9033, 9194, 9318, 9320, 9382, 9396, 9408, 9425, 9654, 9665, 9666, 9671, 9672, 9678, 9723, 9734, 9780, 10023, 10031, 10048, 10052, 10083, 10095, 10392, 10554, 10621, 10929, and 10952) noted that the proposed standards for stationary combustion turbines are too stringent and many offered specific cases which should be considered in setting the standards.
Commenter 2984 asked that the standards "be modified in states that rely heavily on renewable energy." 
Commenter 7994 asked for flexibility to account for the full range of foreseeable operating conditions. 
Multiple commenters (8964, 8969, 8970, 8973, 8995, 9318, 9671, 9672, and 10392) noted that the current limit seems to exclude frame-design units and asked that a higher standard be established. 
Multiple commenters (8964, 8969, 8970, 8973, 9318, 9671, 9672, and 10392) noted that "the optimum design for most NGCC units is based on use of both a frame-design CT and a heat recovery system generating unit." 
Commenters (8964, 8969, 8970, 8973, 9318, 9671, 9672, 10392) stated that it presents a dilemma for smaller entities that need additional generation: "a near-term solution to power supply needs--absent the aforementioned 'disfavoring'--could be to install a frame-design CT, with the goal of 'stepping up' that unit to an NGCC unit at a later date by adding a HRSG unit." 
Commenter 9033 cited a study performed by University of California Berkley (Matthew J. Kotchen and Erin T. Mansur, How Stringent is the EPA's Proposed Carbon Pollution Standard for New Power Plants, University of California Center for Energy and Environmental Economics, April 2012) and asked that EPA establish the standard in the "1100-1200 pound range to allow for low load operation in support of reliability in the grid, support of renewables, and other unplanned events."
Multiple commenters (9425, 9666, 10023, 10031, 10048, and 10083) asked that the standard be set so that it is achievable by 99% of units.
Commenter 9425 stated that "EPA's own analysis demonstrates that a less stringent standard is necessary to address factors beyond the control of NGCC operators that effect emissions rates." Commenter 9425 requested that EPA "modify the standard to at least 1,100 lb CO2/MWh for units that burn natural gas."
Commenter 9780 analyzed GHG BACT analyses and PSD permits and discovered "several recently issued PSD permits contain GHG BACT limits very close to, or even above, the proposed standards." For example, commenter 9780 noted that "the PSD permit issued for the Cheyenne Prairie Generating Station in Wyoming (a 220-MW NGCC plant) includes a GHG BACT limit of 1,100 lb CO2e per MWh (and 187,318 tons per year of CO2e per turbine), which [is] higher than the proposed standard."
Commenter 10052 asked that EPA "either increase the standard or implement exemptions or alternative compliance limits for certain, specified operating parameters such as low-load (similar to an exemption for simple cycle turbines), startup and shutdown frequency, and other conditions."
Commenter 10952 asked that the limits be set "at the highest end of the proposed respective ranges."
 Based on the EPA's analysis of emissions data, GHG permits, and design efficiency data, the EPA has concluded that the final emission standard is achievable for new base load natural gas-fired stationary combustion turbines. See Section IX.D.3.a of the preamble to the final rule for a more detailed discussion.
Multiple commenters (1747, 2329, 2512, 9406, 9513, 9514, 9646, 10025, 10087, 10090, 10093, 10119, 10136, and 10620) asked that the emissions limits for stationary combustion turbines be more stringent.
Commenter 2512 urged EPA to take "aggressive action to reduce greenhouse gas emissions from all sources, with emissions reduction targets of at least 20 percent below 1990 levels by 2020 and 80-100 percent below 1990 levels by 2050."
Commenter 9406 noted that "EPA's analysis confirms that existing units are performing below EPA's proposed new unit limits." Commenter 9406 further noted that "the higher emissions rate during startup has little overall effect on the 12 month rolling average emissions rate, which EPA has proposed in the Rule. The 12 month rolling average is calculated by dividing the total pounds of CO2 emitted over a 12 month period by the total generation output in that same period. While a unit's emissions rate is higher during startup, the total pounds of CO2 actually emitted during startup is small compared to the total emissions of CO2 and amount of energy produced as the unit approaches and operates at its targeted capacity. Moreover, large NGCC combustion turbines are base load or intermediate units, which are unlikely to startup more than once or twice per day, and often will shut down and start up even less frequently. These machines were not designed as peaking plants, and there are economic and structural limitations to how often these units can startup and shutdown. Excessively starting and stopping a unit will severely stress the operating equipment and cause wear and tear that will significantly reduce the unit's lifespan and lead to higher operating costs."
Commenter 9513 stated that such a "lenient limit cannot be reflective of the best system of emission reductions that can be found for the subcategory." 
Commenter 9514 stated that EPA adopted an emission limit "that is far too lenient in light of the level at which ['modern, efficient' CCGT] plants can actually perform."  
Commenter 10119 stated, "the fact that the proposed standard could be achieved by virtually every existing unit of a type to which the Proposed Rule would apply underscores EPA's failure to conduct a forward-looking analysis, its failure to adopt a standard that advances the development and installation of better emissions control technology, and its failure to achieve meaningful emissions reductions." 
Commenter 10119 noted that EPA "fails to explain how the standards meet Section 111's requirement that the agency look forward to ascertain what emission rates would be achievable in new facilities. Nor do the standards do anything to advance technological development. Moreover, because the standards are already being met by virtually the entire NGCC fleet, they do not appear to reduce emissions at all, much less to the maximum practicable degree. Finally, EPA did not present any evidence or analysis indicating that the potential cost of compliance precluded adopting more stringent standards."
Commenter 10119 stated, "EPA's own data show a far more stringent standard is readily achievable by a large number of more efficient existing facilities." 
Commenter 10136 stated that EPA should "set the emission standard for new natural gas plants at the level achieved by the newest, most efficient plants." 
Commenter 10620 supports the rule but "would encourage even stricter emissions limits."
The EPA does not agree that the final emission standard is not sufficiently stringent. The final base load natural gas-fired subcategory includes high-efficiency NGCC designs intended for strictly base load operation, fast-start NGCC designs, as well as large and small units. Because the differences between fast-start and traditional NGCC designs are incremental (i.e., a design could include certain aspects that allow for faster starting, but not use all available technologies), we cannot precisely distinguish between these two types of NGCC designs. In addition, NGCC units, regardless of design, will likely operate over a range of capacity factors that depend on the particular electric demand and generating resources in the specific transmission region. Therefore, we cannot precisely distinguish between these two designs using either technology or percentage electric sales. Because of this, the final emission standard is appropriate for what is achievable for the range of units in our base load subcategory over their operational life. The EPA notes that since we have determined this is what is achievable by the identified BSER there is not requirement to cost out theoretical more stringent emission standards. Furthermore, the NSPS serves as an emission rate floor for BACT determinations. Permitting authorities will perform an individual analysis taking into consideration site-specific conditions when establishing an appropriate emission standard for new units. See Section IX.D.3.a of the preamble to the final rule for additional discussion.
Multiple commenters (7976, 8952, and 9665) questioned the cut-off between small and large gas turbines or asked that the cut-off be changed.
Commenter 7976 noted that "Differentiation of the NSPS based on GT size will attract litigation that seems likely to result in an extended period of counterproductive regulatory uncertainty with eventual change."
Commenter 8952 recommended a cut-off of 1,500 MMBtu/hr HHV based on Gas Turbine World fleet data, using the "largest of the 'small' gas turbine heat input ratings, with added margin to allow for growth... as technology advancements become commercially available."
Commenter 8952 and 9665 noted that the 850 MMBtu/h is consistent with the value designated in Subpart KKKK for regulation of NOX but argued that "the design parameters that influence NOX formation fundamentally differ from those that affect CO2 formation, making the proposed cut-point inappropriate for CO2."
Commenter 9665 stated that the cut-point is too low and recommended a heat rate cut-point of 1,500 MMBtu/h (HHV).
Commenter 9665 noted that "the differentiation between aero-derivative turbines and frame turbines is not an indicator of plant efficiency. As a result, this justification is simply not appropriate in establishing the cut-point." Commenter 9665 further stated, "the presence of reheat steam cycles depends more on the thermal energy available within the exhaust gas of the gas turbine than on the overall size of the turbine. Large turbines have more exhaust energy and a reheat steam cycle is most efficient, whereas small turbines do not have sufficient exhaust energy to warrant the reheat cycle. Although it is true that aero-derivative turbines tend to have lower exhaust energy than frame units, it is not the fact of being an aero-derivative instead of a frame turbine that makes the difference. It is the totality of the combined cycle system, gas turbine efficiency in conjunction with the steam cycle efficiency, that determines the plant level efficiency."
As discussed in Sections IX.B.1, IX.B.2 and IX.D.3 in the preamble to the final rule, after considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion.
Multiple commenters (7976, 8937, 9033, 9425, 9514, 9592, 9596, 9597, 9646, 9654, 9661, 9723, 9780, 10083, 10389, 10617, 10621, and 10693) asked that EPA revise the proposed standards so that one limit applies to all combustion turbines, regardless of size. The commenters suggested a standard ranging from 800 lb CO2/MWh to 1,200 lb CO2/MWh.
Commenter 7976 stated that the "differentiation of 'small' and 'large' gas turbines cannot be justified because 'small' GTs can also be very efficient in combined cycle configurations." Commenter 9514 agreed and noted that "the same technologies that reduce the emission rates of larger units could be incorporated into smaller units."
Commenter 9646 recommended an emission rate limit of no higher than 800 lb CO2/MWh for all subject stationary combustion turbines. 
Commenter 10617 urged EPA to consider a standard set in New York State that "limit[s] carbon emissions to 925 lbs of CO2/MWH" for  "facilities regardless of fuel source that generate 25 megawatts or greater."
Commenter 10693 asked that EPA set a limit of, at most, 950 lb CO2/MWh.
Multiple commenters (8937, 9033, 9425, 9592, 9597, 9661, 9780, 10083, 10389, and 10621) recommended a standard of 1,100 lb CO2/MWhr to accommodate different operating modes, startup and shutdown, use of dry cooling technologies, effects of location, and renewable resources.
Commenter 10389 stated, "an ongoing performance standard of 1,100 lb/MWh (with reasonable accommodations for alternative fuel use and startup and shutdown) would ensure that the plant is operated efficiently and well maintained, while providing a reasonable compliance margin if market conditions change and the facility needs to cycle more." Commenter 10389 noted that the technical basis and the data used "do not support the establishment of separate standards for large and small turbines" and "undermine the incentive for technology innovation and discourage investment in more efficient technologies."
Commenter 9033 recommended a standard of 1,100 lb CO2/MWh "provided start-up and shut downs and other market interruptions beyond operator control are exempted from the calculation." Otherwise, commenter 9033 asked for a standard of 1,200 lb CO2/MWh.
Commenter 9654 asked EPA to "Raise the CO2 level to at least 1,200 lbs/MWh to account for expected operating scenarios including part-load, rapid ramping and cycling, performance degradation and differences in site and atmospheric conditions and fuel quality."
Commenter 9723 noted support for a limit of not less than 1,200 lb/MWh for all gas turbines. 
 After considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas-fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion. In contrast, non-base load natural gas-fired combustion turbines are subject to a separate heat input-based emission standard of 120 lb CO2/MMBtu. See Sections IX.B.1, IX.B.2, and IX.D.3 of the preamble to the final rule for the EPA's detailed response to this comment.
Multiple commenters (2329, 3175, 9406, 9426, 9471, 9497, 9514, 9595, 9665, 9666, 9730, 10023, 10100, 10119, 10243, and 10394) recommended that EPA adopt a different standard for small turbines. The suggestions ranged from 950 lb CO2/MWh to 1,200 lb CO2/MWh.
Commenter 10119 asked for a limit of 950 for small units.
Commenter 9406 recommended a limit of 1,000 for small turbines.
Commenter 9514 asked for a limit of 1,000 or less for small units, but would prefer no separate standard for small units.
Commenter 9730 recommended a limit of 1,050 or lower for small turbines.
Commenter 10394 recommended 1,050 for smaller combined cycle units. 
Multiple commenters (3175, 9471, 9497, 9595, 9665, 9666, 10023, 10100, and 10243) recommended a limit of 1,200 for small turbines.
Commenter 9426 recommended a limit of 1,200 lb CO2/gross MWh for small units but would prefer that EPA "undertake further assessment and repropose the standard." 
After considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas-fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion. In contrast, non-base load natural gas-fired combustion turbines are subject to a separate heat input-based emission standard of 120 lb CO2/MMBtu. See Sections IX.B.1, IX.B.2, and IX.D.3 of the preamble to the final rule for additional details on the EPA's rationale.
Multiple commenters (2329, 7990, 8023, 8909, 8952, 8974, 9381, 9406, 9424, 9426, 9471, 9497, 9499, 9593, 9595, 9660, 9665, 9666, 9730, 9770, 10023, 10024, 10092, 10095, 10100, 10119, 10243, 10393, 10394, 10466, 10665, 10681, and 10869) recommended that EPA adopt different emission standards for large gas turbines. The suggestions ranged from 800 lbs/MWh  to 1,250 lbs/MWh
Commenter 10869 recommended a level of 800 lb CO2/MWh. 
Commenters 10119 and 10394 recommended a limit of 850 for large units. 
Commenter 9406 recommended a limit of 925 lb CO2/MWh for larger units. 
Commenter 10681 recommended a standard of "970 lb CO2e/MWh (approximately 940 lb CO2/MWh)." 
Commenters 8909, 9660, and 9730 recommended a standard of 950 for larger units.  Commenter 9660 asked that EPA lower the emission limit to 950 lb CO2/MWh "based on the determination by the New York State Department of Environmental Conservation that a 925 lbs. CO2/MWh emission rate is feasible" and "the equivalent CO2 emission standard established by Washington State" at 940 lb CO2/MWh. 
Commenter 8023 stated that the standard for large gas combined cycle systems should be no less than 1,100 lbs/MWh, but did not object to the proposed emissions limit of 1,000 lbs/MWh for large combustion turbines when applied to combined cycle units used primarily to produce base load power.  
Commenter 9593 recommended setting the limit between 1,000 and 1,100 for large units. 
Multiple commenters (7990, 9471, 9381, 9426, 9497, 9499, 9595, 9665, 10023, 10095, 10100, 10243, and 10466) recommended a standard of 1,100 for large units.
Commenter 10466 stated, "The standard for large CCTS ((natural gas) combined cycle turbines) should be 510 Kg/MWh (1100 lb/MWh) on a net output basis. This standard and basis is consistent with the 1100 lb/MWh(net) standards already adopted by several States including; Washington, Oregon, and California." 
Commenter 9666 asked for a standard between 1,100 and 1,200 to increase the achievability level of the standard.
Commenter 8952 recommended a limit of 1,200 lb/MWh "if no exemption for part load operation is provided for the combined cycle." 
Multiple commenters (9424, 9770, 10024, 10092, 10393, and 10665) requested an increase from 1,000 lb CO2/MWh to 1,250 lb CO2/MWh.
 After considering public comments, we are not finalizing the subcategories for large and small combustion turbines that were contained in the proposal. Instead, all base load natural gas-fired combustion turbines must meet an emission standard of 1,000 lb CO2/MWh-g. Base load natural gas-fired combustion turbines include units that actually burn more than 90 percent natural gas and exceed the final percentage electric sales criterion. In contrast, non-base load natural gas combustion turbines are subject to a separate heat input-based emission standard of 120 lb CO2/MMBtu. See Sections IX.B.1, IX.B.2, and IX.D.3 of the preamble to the final rule for additional details.
Multiple commenters (7976, 9513, 9514, and 10100) recommended that the standard be set up in a different way.
Commenter 7976 asked that multiple NSPS levels be considered. Commenter 7976 suggested the following NSPS formula and notations: 
"NSPS (lb. CO2 / MWh) = 1,000 - {[(Maximum Allowed % CF - 25%) /25%] x 100}
for CF levels ranging from 10% through 65% using emission calculations based on new and clean, base load performance tests that are corrected for plant back pressure and other ambient conditions as previously discussed. CF levels outside the range are unregulated."
Commenters 9513 and 9514 recommended a three tier system: base load units, intermediate load units, and peaking units.  More detail is in Section 7.4 (Other Subcategories). Commenters 9513 and 9514 recommended the following standards (on a net output basis):
 825 lbs CO2/MWh for base load units (those operating over 4,000 hours annually), 
 875 lbs CO2/MWH for intermediate and load-following units (those operating between 1,200 and 4,000 hours annually), and 
 1,100 lbs CO2/MWh for peaking units (those operating less than 1,200 hours per year).
Commenter 10100 asked that if a higher standard is not adopted EPA should consider using a "sliding scale based on generation thresholds to determine the performance standard for EGUs on an individualized basis.  For example, the highest limit should apply to units operating at low capacity factors (i.e., in the 33 percent to 45 percent range).  The emissions limit could be reduced incrementally as the generation threshold increases."
 While a three-tiered approach to subcategorization is theoretically appealing, as explained in the preamble, the EPA has concluded that we do not have sufficient information at this time to establish emission standard for three tiers. Instead, the EPA is finalizing the sliding-scale approach to subcategorize non-base load and base load units. Because the sliding-scale approach bases the transition from the non-base load to the base load subcategory on an amount of electric sales equal to the design efficiency of the combustion turbine, it could be considered a modified three-tiered approach. The sliding scale incentivizes the installation of the most efficient non-base load units by allowing them to sell more electricity to the grid before being subject to the base load standard. While the non-base load emission standard is not especially stringent, the approach will achieve some minor emission reductions from non-base load units. See Sections IX.B.2 and IX.D.3 of the preamble to the final rule for a more detailed discussion.
Commenter 10025 asked that the emission limit be set "to the level necessary to bring atmospheric carbon concentration back to 350 ppm by 2100."
Commenter 10087 stated that "EPA provides no evidence that it should set an emission limit because EPA's limit, according to this rule itself, it will do nothing. If EPA chooses to set an emission limit, then the limit should be set at a level that has discernible impact on the climate change factors EPA identifies, namely 'heat waves; more intense precipitation events and storm surges, less precipitation and more prolonged drought in the West and Southwest, more fires and insect pest outbreaks in American forests, especially in the West; and increased ground level ozone pollution.'" 
No emission standard on natural gas-fired EGUs would restrict CO2 emissions sufficiently to stop the increase in atmospheric CO2 concentrations, let alone reduce atmospheric CO2 back down to 350 ppm, without additional reductions from other sectors, existing sources, and other countries. This does not mean, however, that reducing CO2 emissions and incentivizing the development of cleaner and more efficient technologies is not important. The EPA applied the BSER criteria to determine the BSER for this category of sources and set an emission standard that is achievable by new sources. The EPA will review the NSPS for this source category in eight years and increase the stringency of the emission standard if warranted at that time.
7.3.1 Proposed Standards Are/Are Not Achievable
Multiple commenters (9514, 9660, and 10098) agreed that the proposed standards of performance for stationary combustion turbines are achievable.
Commenter 9660 discussed a recent analysis whereby the New York State Department of Environmental Conservation (NYSDEC) is in the process of finalizing Part 251, which "will establish a New York State CO2 performance standard for new sources with a capacity of at least 25 megawatts (MW), as well as for existing sources that increase capacity by at least 25 MW. Part 251 will establish an output-based CO2 emission limit for most base load EGUs of 925 lbs/MWh."
Commenter 10098 stated, "By focusing on actual emissions from recently constructed facilities that are currently in operation, EPA has ensured that the proposed standard is based on technology that is adequately demonstrated."
 We appreciate the commenters' support. Section IX.D.3 of the preamble to the final rule provides additional detail on our conclusions regarding the achievability of the final standards for base load natural gas-fired stationary combustion turbines.
Commenter 9033 referenced a study performed by University of California Berkley (Matthew J. Kotchen and Erin T. Mansur, "How Stringent is the EPA's Proposed Carbon Pollution Standard for New Power Plants", University of California Center for Energy and Environmental Economics, April 2012) that was "based on actual emission data as opposed to EPA's predicted and adjusted data." According to Commenter 9033, the study found "approximately 16-29% of the NGCCs (depending on calculation methods) will fail the 1000 pounds per MWh standard as opposed to the 5% failure rate EPA has suggested."
 The commenter is incorrect. NGCC units are not required to report output from the steam turbine to CAMD when reporting gross output. Because the steam turbine of an NGCC unit contributes about one-third of total output, any unit that does not report its steam turbine output will have a reported emission rate that appears to be closer to that of a typical of simple cycle unit. For this reason, the EPA's analysis filtered out any NGCC units not reporting their steam turbine output to CAMD. The University of California study, on the other hand, did not filter out these units. Therefore, the study does not provide an accurate assessment of whether these units could achieve the final standard and should be disregarded.
Commenters 9425, 9666, 10023, 10031, and 10083 stated that EPA did not follow its own process for establishing NSPS. Commenters 9425 and 10023 noted, "Since the 1990s, EPA has calculated statistically achievable emission limits for NSPS by allowing exceedance frequencies of either one percent, one per year, or one per ten years, which allows at least 99 percent of the units to comply." Commenter 9666 noted that EPA "generally requires that the standards must be achieved by at least 99 percent of affected units." Commenter 10083 asked that "the emission rate for new combined cycle combustion turbines be set at a limit that is achievable by at least 99% of units." 
These commenters are mistaken about how the EPA has calculated achievable emission standards for combustion turbines and coal-fired EGUs in the past. In previous NSPS reviews, the EPA first identified the best performing unit(s) (i.e., the unit(s) with the lowest emission rate) and then based the emission standard on a 99 percent confidence level using the data from those units. However, for multiple reasons (discussed in the preamble), the EPA adopted a different approach for this rulemaking. The primary reason is that the base load natural gas-fired unit subcategory includes units with significantly different operating characteristics. These operating characteristics can influence emission rates to such an extent that it is difficult to identify the best performing units based solely on design and maintenance practices. Therefore, we concluded that it was appropriate to consider operating data from existing NGCC units that commenced operation starting in 2000 and evaluate the range of operating conditions, turbine sizes and models, and other factors that our final standard would need to account for. As we explain in the preamble, the vast majority of NGCC units that have commenced operation since 2000 are already achieving the final standard.
Multiple commenters (8937, 8952, 9033, 9035, 9396, 9425, 9426, 9471, 9597, 9665, 9666, 9769, 9780, 10023, 10031, 10052, 10083, 10095, 10105, 10243, and 10621) disagreed that the proposed standards of performance are achievable.
Commenter 8937 stated that EPA's own data shows that not all new NGCC units can meet the standard.
Commenter 8952 noted that the standard is "only achievable through complex time averaging, not through any specific emissions reduction capability inherent in the design or operation of the gas turbine."
Commenter 9396 stated, "EPA has not demonstrated that such a standard could be met under a range of operating conditions, design criteria, and other factors, and over the lifetime of the stationary combustion turbine." Commenter 9396 noted that EPA's analysis and conclusion are inadequate because only a subset of data were used, no analysis was performed on 70% of the data entries, and there are unanswered questions about "efficiency degradation, cycling and startups/shutdowns, part load operation, and secondary fuel use."
Commenter 9425 questioned whether NGCC "has been adequately demonstrated to be achievable continuously under normal, real-world operating conditions."
Commenter 9471 stated that the standard is achievable only if an NGCC is operated as a base load unit. According to Commenter 9471, many NGCC units are required to operate at lower capacities to "accommodate integration of renewables and other system, grid or electricity market needs, or to experience multiple startup and shutdowns over seasonal operating periods. If EPA wants to encourage the continued development of intermittent renewables such as wind and solar, it will be critical for NGCC units to have emission limits that would allow them to perform this function." 
Commenter 9665 asked EPA to consider the "counterproductive impacts a standard that does not properly account for achievability under the range of anticipated operating conditions could have on total CO2 emissions and to ensure within the legal construct of Section 111 that it is exercising its discretion to avoid such impacts."
Commenter 9666 stated, "EPA has not demonstrated that the proposed CO2 emissions limits are achievable" for the following reasons:
 "If EPA increases the CO2 standard to 1100 lb CO2/MWh and uses the 1/3 sales criterion exemption, the available data shows that combined cycle turbines could meet an achievability level of approximately 99 percent."
 "On the other hand, if EPA decreases the CO2 standard below 1,000 lb CO2/MWh for large turbines, the sales criterion will have to be increased above the proposed 1/3 level in order for the achievability level to be in the historic range of 99 percent." 
 "With respect to large NGCC EGUs, UARG's consultant estimates that only 96.7 percent of the turbines analyzed would achieve the proposed NSPS."
 "An even higher percentage of small NGCC EGUs fail to meet the proposed NSPS, although this estimate is based on a very small sample size so UARG has less confidence in this conclusion." 
Commenter 9666 conducted an "independent analysis of CO2 emissions starting with the identical dataset of natural gas-fired combined cycle combustion turbines," added emissions data through September 30, 2013, and added additional turbines "if they satisfied EPA's general filtering criteria." The final analysis "comprises 325 instead of 307 turbines and analyzes 22,653 months of rolling average emission data compared to EPA's 15,670 months." 
Commenter 9666 stated that for large turbines at a sales criterion of 0.333, the achievability level of 97 percent approaches the historic achievability level for a CO2 limit of 1,000 lb CO2/MWh. A slight increase in the proposed limit would likely result in an achievability in the range of 99 percent. Commenter 9666 provided a table that shows the result of analyzing the achievability for a range of sale criteria for a CO2 limit of 1100 lb CO2/MWh (see Table 3 of the comment letter). As can be seen from this table, at a sales criterion of 0.333 the achievability level is 98.7 percent - very close to the historic achievability level. Therefore, setting the CO2 standard at 1100 lb CO2/MWh would satisfy the historic achievability level at a sales criterion of 0.333 Since the small turbines have only one exceedence at 1175 lb CO2/MWh, out of only 6 turbines that are subject to the rule, the achievability level for this set would be 83 percent. The only way for the achievability level to approach 99 percent for small turbines would be to set the standard at 1,200 lb CO2/MWh.
COMMENTER 9666 STATED THAT UNCERTAINTY ON THE PRECEDING IMPLEMENTATION ISSUES MADE IT DIFFICULT TO COMMENT ON BOTH THE ASSOCIATED RANGE OF SALES CRITERION SUGGESTED BY THE EPA AND ACHIEVABILITY ACROSS THE RANGE OF CO2 limits included in the proposal. The commenter proceeded to present a series of assumptions regarding how the EPA may choose to implement the sales criterion; the commenter stated that the assumptions presented were not a preferred approach for compliance determinations but rather the assumptions were necessary to allow a first order evaluation of the impact of various sales criteria on achievability of the standard. The commenter's assumptions implementation of the sales criterion were applied to a data set similar to the data set provided by the EPA at Docket Item -0082, however, since the commenter's analysis was completed 21 months after the EPA's analysis, the commenter's data set was larger than the EPA's with an additional 7,000 months of combustion turbine operating data. The population in the EPA's data set was 307 turbines; the population in the commenter's supplemented data set was 325 turbines. Based on the commenter's assumed methodology for implementing the sales criterion with compliance calculations, the commenter presented a table that indicated 31 out of 320 large turbines did not achieve the 1000 lb CO2/MWh standard in the supplemented historical data set assuming no sales criterion exemption; 23 of the large turbines did not achieve the 1000 lb CO2/MWh standard based on a 20 percent sales criterion; 18 large turbines did not achieve the standard based on a 25 percent sales criterion; 8 large turbines did not achieve the standard based on the proposed one-third sales criterion; and 6 of the 320 large turbines (approximately 2 percent) did not achieve the 1000 lb CO2/MWh standard based on a 40 percent sales criterion. Commenter 9666 also provided an analysis for the 15 turbines included in the small turbine data set; only one turbine did not achieve the 1,100 lb CO2/MWh standard. According to the commenter's analysis, this small turbine would not have met the standard even with a 40 percent sales criterion. Subsequently, commenter 9666 provided a similar analysis based on achievability of a 1,100 lb CO2/MWh standard for large turbines; this analysis indicated that only 3 of the 320 large turbines included in the supplemented historical analysis did not achieve the 1,100 lb CO2/ MWh standard with a corresponding one-third sales criterion exemption. For large turbines, commenter 9666 concluded that setting the CO2 standard at 1,100 lb CO2/MWh would satisfy the historic achievability level at a one-third sales exemption level. For small turbines, commenter 9666 concluded that setting the CO2 standard at 1,200 lb CO2/MWh is the only way to approach the historic achievability level (referring to achievability of the 1998 Subpart Da NOx standard).
Commenter 9780 urged the use of representative data to support an achievable standard and stated that, "EPA must reject the use of 'average' emissions data in support of lower rates unless these 'averages' were computed using the proposed compliance averaging formula. Otherwise, these averages cannot be used to support any assertions of achievability."
Commenter 10095 stated that EPA must explain how the standard is achievable under the most adverse conditions which can reasonably be expected to occur. According to commenter 10095, historical operating data show that many units operate above the proposed standard when experiencing adverse conditions.
 See Sections IX.B.2 and IX.D.3 of the preamble to the final rule for an explanation of how the sliding-scale and final standards for non-base load and base load combustion turbines operate together to ensure that all new and reconstructed combustion turbines will be able to achieve the final standards.
Commenter 9769 provided a real-world example of a simple cycle power plant (that began commercial operation in January 2011) which would not meet the proposed standard. Commenter 9769 stated that Dave Gates Generating Station (DGGS) "operates at an average capacity factor over 30 percent and, as the attached graphs show, does not operate under steady state conditions. The variability in generation increases heat rate which results in a greater CO2 emissions rate when compared to steady state operations. The CO2 emissions rate for DGGS currently averages about 1,450 pounds of CO2 per MWh on an annual basis. Because DGGS is controlled by the transmission system operator and must operate 24/7, there are few, if any, options to limit operations to reduce or control the capacity factor, or to lower the CO2 emissions rate and ensure NorthWestern Energy complies with mandatory reliability and control performance standards."
 The commenter incorrectly focuses on capacity factor. The appropriate criterion in the proposed and final rules is the percentage electric sales threshold. The EPA acknowledges that the Dave Gates Generating Station (formerly known as Mill Creek) has sold more than the one-third percentage electric sales threshold on an annual basis. In fact, of the six simple cycle turbines at the facility, the maximum annual percentage electric sales was 39 percent, and the maximum 3-year average percentage electric sales was 35 percent. However, even if these units were subject to the requirements of this final rule, they would not have to change how they are operated. The subcategorization threshold between base load and non-base load units is based on the design efficiency of the unit. Currently, available simple cycle turbine models have design efficiencies of greater than 40 percent. In theory, a new simple cycle turbine serving the same load as the Mill Creek using the most efficiency simple cycle turbine available could generate 20 percent more electricity prior to moving into the base load subcategory. In addition, the system emergency exclusion would permit the simple cycle turbine to continue to operate during emergencies because those sales do not count toward the percentage electric sales threshold. Finally, a new unit intended to sell this high of a percentage of its potential electric output could be designed to use fast-start NGCC technology. As described in the preamble, NGCC technology is generally cost effective at this level of electric sales and fast-start NGCC offers similar operating flexibilities as simple cycle turbines.     
7.3.2 Proposed Standards Should Be Based on Actual Emissions Data
Multiple commenters (8952, 9033, 9425, 9665, 9780, 10017, 10095, and 10243) stated that the analysis for stationary combustion turbines should have been based on actual emissions data.
Commenter 8952 noted that based on collected CAMD data from SNL Energy for the years 2001 through 2012, "significant aspects of EPA's proposed rule lack technical support and rationality." Commenter 8952 also stated that the "data from the 2007-2011 time periods could under-represent gas turbine utilization since the period included such a severe economic downturn." Commenter 8952 referenced a "measureable increase in generation that took place just in the last six weeks of 2013."
Commenter 9425 stated that actual variable emissions rates are a more accurate representation of unit performance than design performance specifications.
Commenter 9425 stated, "According to the Standard of Performance for Natural Gas-Fired Combustion Turbines Technical Support Document (NG TSD) included in the docket for this Proposal, EPA analyzed the actual CO2 emissions and other operating characteristics for a study population consisting of 307 NGCC units to propose the standard for new units, which is more appropriate than the design specifications used in the 2012 Proposal." 
Commenter 9665 noted that it is "impermissible to base the standard on assumed operating conditions that are not reflective of how plants actually do operate." 
Commenter 9780 noted that design specifications do not "reflect normal, real-world operations." 
Commenter 10017 noted that the analysis relies on "tentative, future projects to set a standard instead of using actual data from existing plants to ensure the promulgated standard is achievable."
Commenter 10095 stated that "EPA's data set is insufficient to justify the proposed standard and noted the following issues with the analysis:
 "all of the units analyzed are relatively new and thus have experienced limited degradation. As these units continue to operate, they will experience degradation-potentially up to eight percent-which would cause many of the units to operate above the proposed standard." 
 "between 2007 and 2011, the years analyzed by EPA, increasing demand due to declining natural gas prices led many NGCCs to operate at high capacity factors and thus higher average efficiencies and lower CO2 emission rates. EPA must analyze the potential range of real-world operations, which must include the influence of higher natural gas prices and fluctuations in demand."
 "New power plants are large upfront investments built to provide benefits spread over several decades and, therefore, will likely experience a wide range of operational scenarios over their lives. EPA, subsequently, needs to set a standard based on operational scenarios that may occur over this same period.
Commenter 10243 stated, "EPA should review actual operating data under all expected operating conditions for NGCCs and promulgate a standard that is achievable."
The commenters are mistaken that the EPA did not use actual operating data in the proposed rulemaking. The EPA analyzed eight years of operating data from existing NGCC units that commenced operation in 2000 or later when determining an appropriate emission standard that will account for a wide range of operating conditions and future degredation. We also used design efficiency data was as support in in the final rule to confirm that the final emission standard is achievable. See Section IX.D.3 of the final rule preamble to the final rule for further details.
Commenter 9780 noted that recent GHG BACT determinations do not support the proposed standards. Commenter 9780 stated, "A closer examination of these and other recently issued permits reveals that EPA's reliance on these BACT limits is misplaced. For example, some of these permits do not contain enforceable emissions rate standards. Others contain provisions that make it clear that continuous compliance with the emissions rates cited by EPA is not required. In addition, several other recently issued PSD permits contain GHG BACT limits very close to, or even above, the proposed standards."
The EPA reviewed recent BACT permit limits for GHG emissions to evaluate the structure of these limitations (e.g., as mass emission rates or heat rate limitations), the averaging periods, treatment of startups and shutdowns, and associated testing and monitoring requirements. We found a number of variations in the way permitting authorities had structured these BACT limits. Some of the BACT permit conditions had separate numeric limits that applied during startups and shutdowns with associated limitations on the total duration of startups and shutdowns. The most common compliance period was 12-months. Some permits only included heat rate type limits (i.e., MMBtu heat input per MWh output or lbs CO2/MWh) with ongoing monitoring based on periodic performance tests at greater than 90 percent of maximum output with associated numeric limits that were significantly more stringent than the final NSPS. The EPA disagrees with Commenter 9780's characterization of the Oregon Clean Energy Center's BACT permit limits  as an example of a recently issued BACT permit that does not "support the reasonableness" of the final standard for combustion turbines. The EPA also reviewed this permit which includes a limit of 833 lbs CO2/MWh for the power block based on installation of the Siemens turbine. The EPA directs the commenters attention to page 84 of 86 under permit condition (gg)(2)(d) which states, 
      "[Compliance] test(s) shall be conducted under those representative conditions that challenge to the fullest extent possible a facility's ability to meet the applicable emissions limits and/or control requirements, unless otherwise specified or approved by the appropriate Ohio EPA District Office or local air agency. Although this generally consists of operating the emissions unit at its maximum material input/production rates and results in the highest emission rate of the tested pollutant, there may be circumstances where a lower emissions loading is deemed the most challenging control scenario. Failure to test under these conditions is justification for not accepting the test results as a demonstration of compliance."

The EPA has concluded that this testing condition in conjunction with the associated numeric limit of 833 lbs CO2/MWh is sufficient to ensure that the design and operation of the permitted facility will result in an emission limit that does not exceed 1000 lbs CO2/MWh. Additionally, the EPA disagrees with the commenters statements that the permit limits expressed in the form of a heat rate (e.g., 7,605 Btu/kWh) are significantly different than the form of the final NSPS (lbs CO2/MWh) because all recently permitted combined cycle combustion turbines primarily or strictly fire pipeline natural gas. Given this fuel a limitation expressed as a heat rate can be converted directly to the form of the NSPS using the standard methods of 40 CFR Part 75 as allowed by the final rule. 

Finally, while the EPA did review recently issued permits when evaluating the level of the standard, the EPA also reviewed actual performance data to determine the degree of limitation that is actually being achieved after the commencement of operations (as required by section 111(a)(1)). To directly address commenters concerns that the permit for the Cheyenne Prairie Generating Station included a limit of 1,100 lbs of CO2 per MWh compared to the final standard of 1000 lbs of CO2/MWh, the EPA reviewed the most recently available operating data for this new load-following unit. During February and March of 2015 this unit operated a total of 54 hours with an actual emission rate of 983 lbs of CO2/MWh. It is important to note that the percentage electric sales of this unit during this operating period would have classified this unit as a non-base load unit if the unit sustained this percentage of electric sales for 12-months. However despite this low capacity factor the unit was capable of operating below the 1,100 lbs of CO2/MWh permit limit and below the level of the final NSPS. 


7.3.3 Addressing Factors Beyond Operator's Control
Multiple commenters (6505, 8952, 9033, 9194, 9381, 9396, 9408, 9422, 9425, 9497, 9591, 9592, 9654, 9661, 9665, 9666, 9777, 9779, 9780, 10023, 10031, 10052, 10095, and 10105) noted that fuel disruptions and the resulting need to burn back-up fuel influence CO2 emissions and should be addressed in EPA's analysis.
Commenter 8952 asked for an exemption when back-up fuel is used or allow for 500 hours annually. Commenter 8952 referenced EPA's exemption of full-time operation using liquid fuel where natural gas is not available and stated that exemption during emergency operation should also apply. 
Commenter 9033 noted that many gas turbines experience fuel disruptions in the winter and at times of gas market delivery malfunctions. Commenter 9033 asked that the use of alternate fuels be exempted from the averaging or provide a 500 hour exemption. 
Commenter 9592 stated that back-up fuel must be used during gas curtailments as a result of hurricanes and other natural disasters. Commenter 9592 asked that EPA exempt the use of back-up fuel during disruptions.
Commenter 9654 asked that EPA consider the implications of advanced equipment that will allow for easier transition to back-up fuels. Commenter 9654 noted the existence of "a new technical approach to improve the flexibility of the gas turbine equipped with a DLE/DLN combustion system, so that it can operate on either natural gas or an alternative liquid fuel, with no change in the combustion equipment and no need to retrofit or replace any of the hot gas path hardware." 
Commenter 9666 asked that operations during fuel disruptions be excluded from compliance determinations or "analyze CO2 emissions during oil-fired operations, propose an NSPS, and allow the public to comment on whether such NSPS is achievable."
Commenter 9780 noted, "Assuming the same heat input while firing on natural gas or firing on oil to achieve the same electrical output, emissions during emergency operations while firing on oil would be 37 percent higher than when firing on natural gas." 
Commenter 10095 noted that "units must regularly exercise the equipment, conduct post maintenance testing, and perform environmental compliance testing while burning oil."
Commenter 10095 asked that EPA exclude oil-burning operations from the averaging standard. Commenter 10095 also noted that Georgia Power Company's McDonough Combined-Cycle Facility has a CAA permit that allows up to 1,000 hours of oil operations for critical turbines. Commenter 10095 stated that many units are "required to have oil backup fuel available for reliability needs, customer contract and contingency response requirements, offsetting natural gas service loss during inclement weather or other non-routine events, and meeting critical demands including grid stability support."
The natural gas-fired base load standard only applies when a unit is operating above the percentage electric sales threshold and burning over 90 percent natural gas during the applicable 12-operaitng month period. When a combustion turbine is not meeting both of these criteria, the applicable emissions standard is based on the use of clean fuels. The clean fuels standard is readily achievable for combustion turbines burning distillate oil and other non-natural gas fuels. In the worst case scenario, a base load natural gas-fired unit would be burn only 10 percent non-natural gas. This operating scenario would increase the unit's emission rate by a maximum of 4 percent (when burning 10% distillate oil) and more typically by only 2 percent (when burning gaseous fuels containing heavier hydrocarbons). The final standard is conservative and accounts for multiple operating conditions, including the periodic use of non-natural gas fuels. As we explain in the preamble, the vast majority of existing NGCC units, including those that burn non-natural gas fuels from time to time, are operating significantly below the 1,000 lb CO2/MWh-g standard. Therefore, there is no reason to exclude periods of fuel disruptions.
Multiple commenters (7976, 9194, 9381, 9396, 9422, 9425, 9426, 9427, 9471, 9592, 9665, 9666, 9780, 10052, and 10621) noted that unit efficiencies degrade over time leading to increased CO2 emissions. The factor is beyond the operator's control and should be addressed in EPA's analysis.
Commenter 9425 and 9381 noted that units have variable rates of degradation over time.
Commenter 9592 noted that efficiency declines with normal aging "even when adhering to the manufacturer's recommended maintenance program."
Commenter 9665 noted that combined cycle plant degradation impacts efficiencies and emissions.
Commenter 9666 noted that heat recovery steam generator (HRSG) and the auxiliary components become less efficient over time due to "fouling of the heat transfer surfaces."
Commenter 9666 noted that routine inspections to assess the condition of the turbine hardware are commonly conducted "at intervals ranging from 8,000 to 12,000 hours of operation...[wherein] components are repaired or replaced that show wear that may affect continued efficient operation." According to commenter 9666, between inspections "overall efficiency may drop by as much as 2 percent", increasing CO2 emissions.
 See Section IX.D.3.a of the preamble to the final rule for an explanation of why the EPA's final standard for base load natural-gas fired combustion turbines provides an adequate compliance margin for future degradation. The EPA also notes that owners and operators can limit or prevent future degradation by adopting good operating and maintenance practices and conducting efficiency upgrades.
Commenter 9381 noted that "emissions (on an output basis) increase significantly as load is reduced.  Many NGCC units, even those designated as base load, are required to operate at lower capacities to accommodate integration of renewables and other system, grid or electricity market needs.  Combined-cycle units that have a two-on-one or three-on-one configuration may have occasion to operate while one CT is out of service for maintenance, which reduces efficiency due to reduced heat input into the steam cycle."
Commenter 9665 noted that units operate at low load during overnight periods as well.
The EPA's emissions data analysis did not attempt to filter out NGCC units using two-on-one and three-on-one configurations and inherently accounted for periods of operation when one or more combustion turbines is out of service. In addition, while two-on-one and three-on-one configurations may occasionally operate with one combustion turbine out of service, which may reduce the efficiency of the steam cycle, the overall design efficiency of these power blocks is higher than comparable one-on-one configurations. According to design data submitted to Gas Turbine World, using multiple combustion turbines in a single power block reduces the emission rate by an average of 0.9 percent. This would likely more than account for any increases in the emissions rate due to operating with combustion turbine(s) out of service. 
Multiple commenters (7976, 9381, 9425, 9666, 9780, 10052, and 10083) noted that higher elevations impact emissions, are beyond the operator's control, and should be addressed in EPA's analysis. 
Commenter 9381 stated that "higher elevations...have a noticeable negative impact on the generation of electricity, which would increase CO2 emissions per MWh." 
Commenter 9666 noted that "in some regions of the country, a new NGCC unit may have minimal or no access to water for controlling condenser temperatures." According to commenter 9666, the units will require air cooled condensers, which leads to fouling and a lowering of performance efficiency.
Multiple commenters (9499, 9665, 10052, 10083, and 10105) noted that low water use technologies have higher heat rates. Commenters 9665, 10052, and 10105 noted that utilizing dry cooling systems to reduce water consumption negatively impact the heat rate and CO2 emission profile of the unit. Commenter 10083 noted that "air-cooled [units] should not be penalized for adopting a low water use technology." 
Commenter 9665 noted multiple factors that affect the performance and emissions of a new NGCC unit:
 deployment of supplemental duct firing
 steam turbine condenser cooling
Commenter 9780 noted that seasonal temperature variations affect the operational efficiency and also the emissions rate. 
Commenter 10083 noted that humidity affects gas turbine performance. 
Multiple commenters (9381, 9425, 9780, 10052, and 10105) noted that dusty environments influence emissions.
Three commenters (9381, 9425, and 9780) stated that the standards do not account for the fact that each unit is specific in design (e.g. number and efficiency of inlet filters) and operating size conditions (dusty open plants, high ambient temperature), and the fact that the percent degradation over time is variable.
The commenter is mistaken that the EPA analysis did not account for site-specific conditions. The data set of NGCC units in the EPA's analysis is diverse and includes NGCC units operating at different ambient conditions, elevations, cooling types, and accounts for factors such as the number and efficiency of inlet filters. While we acknowledge that duct firing reduces the overall efficiency of an NGCC unit, the final standard includes sufficient flexibility for the periodic use of supplemental firing to provide additional output during periods of peak electric demand. See Section IX.D.3.a of the preamble to the final rule for further explanation of why the EPA's final standard for base load natural-gas fired combustion turbines adequately accounts for these factors. 
Multiple commenters (9320, 9381, 9425, 9665, 9666, 9780, 10052, 10083, and 10105) noted that combined cycle facilities in locations with high ambient temperatures are less efficient.
Commenter 9406 noting that, although EPA's analysis suggested that combustion turbines operate less efficiently at higher ambient temperature, the analysis also showed that "the vast majority of units with higher actual monthly CO2 rates in summer still meet the standard without any exceedances." Commenter 9406 further noted that, by proposing a 12-month rolling average compliance period, units will be able to offset higher emissions rates during warmer months with operations during the rest of the year.
The EPA agrees with the commenter that the impact of temperature has been accounted for when determining the achievability of the final emission standard.
Multiple commenters (7994, 8937, 9194, 9396, 9408, 9592, 9661, 9666, and 10031) noted that EPA's standard does not account for the operation of combined cycle units in simple cycle mode.
Commenters 9194 and 9396 offered the following examples:
 a facility is licensed and designed to operate in both simple cycle and combined cycle modes.
 there are times when combustion turbines at existing NGCC plants are licensed and designed to operate in simple cycle mode when maintenance is being performed on the steam cycle components
 The steam cycle segment of the NGCC is complex, and there are many components that can require maintenance
 Operating the combustion turbine in simple cycle mode for limited periods of time may be the most appropriate manner for producing electricity from reliability, environmental, and consumer cost perspective.
Commenter 9194 also offered the following examples:
 The commenter noted that one facility has "a 110 MW combined cycle gas-fired unit (CT generates 75 MW (100% load), ST generates 35 MW (100% load) for 110 MW total) that is sometimes required to follow the GRU system load requirement. If the CT is operated at 40 MW and ST at 20 MW, the CT will be in violation of NOx limits since the low NOx combustion controls require at least 50 MW to operate. The only solution is to operate the CT at 60 MW in simple cycle mode. To meet Title V permit limits the steam cycle is taken off line. It should be noted that the units are then serving as peaking units in these cases."
 "there are times when CCs are run in simple cycle mode due to equipment breakdowns with the steam cycle components. The steam cycle segment of the CC machine is actually quite complex and there are many components that can fail causing the CC to have to function in simple cycle mode. This vulnerability is especially critical for smaller public power systems since they do not have a large fleet of units to choose from. Operation in simple cycle mode may often be required to meet reliability constraints."
 Commenter 9592 added examples of 1) maintenance, 2) equipment breakdowns, and 3) reliability constraints and that EPA should account for the instances.
Commenters 9666 and 10023 asked EPA to "exempt simple cycle mode operations from compliance when an NGCC unit falls below the 33 percent capacity factor in that mode."
The EPA disagrees with the commenters that an NGCC unit operating in simple cycle mode should be treated differently than an NGCC unit operating in combined cycle mode. The final rule does not set separate standards for simple cycle and NGCC units. Instead, the final rule sets separate standards for base load and non-base load units, which are subcategorized based on the amount of electricity sold to the grid. The final standard for base load units includes sufficient compliance margin to provide flexibility to NGCC units that might operate in simple cycle mode for brief periods throughout the year. If a NGCC unit operates in simple cycle mode for a significant period of time, then it is essentially operating as a non-base load unit and will be subject to the input-based standard for non-base load units as long as its electric sales are below the sliding scale. If the owner or operator wants to operate the unit to serve intermediate or base load demand, then they should operate the unit in combined cycle mode, consistent with the BSER for base load units.
Multiple commenters (7976, 9194, 9320, 9381, 9396, 9408, 9425, 9426, 9427, 9471, 9592, 9597, 9665, 9666, 9780, 10052, 10083, 10243, and 10621) noted that CO2 emissions are higher as a result of cycling and times of startup and shutdown.
Commenter 9592 stated that "startup creates time periods where fuel is combusted inefficiently and little or no electricity is generated which will cause CO2 emissions in excess of 1000 lb CO2/MWh." Commenter 9592 also noted that a similar situation occurs during shutdown.
Commenter 9665 noted that an increased amount of CO2 and criteria pollutants and higher in-use degradation occur when operating during conditions of rapid cycling and lower overall capacity factors because combined cycle units produce more emissions during startup and shutdown. Commenter 9665 noted that "emissions during startup are also variable and differ from startup to startup at the same plant due to their dependence on site- and time-specific factors, such as equipment temperature at the initiation of startup."
Commenter 9666 noted that "Docket Item -0082 [contended] that cycling does not impact the CO2 emissions levels but EPA's methodology does not demonstrate the validity of this conclusion." 
Commenter 9666 noted that there are different types of cycling on different configurations of combined cycle turbines. Commenter 9666 stated, "startup and shutdown (SU/SD) times on CC combustion turbines are dependent upon the temperature gradients in the intermediate pressure steam turbine (IPST) of the HRSG, and not necessarily the combustion turbine itself. If the IPST is heated or cooled too rapidly it may warp the steam turbine rotor shaft and render the steam turbine inoperable."
 See Section IX.D.3.a of the preamble to the final rule for an explanation of why the EPA's final standard for base load natural-gas fired combustion turbines adequately accounts for startup, shutdown, and cycling. The EPA also notes that NGCC units with less than 50 percent electric sales (i.e., those units that start up, shut down, and cycle most frequently) will be subject to the input-based standard for non-base load units. The EPA further notes that the way the final base load emission standard is calculated (i.e., total emissions divided by total output over the applicable 12-operating-month period and not an average of the hourly emission rates) minimizes the impact of periods of startup and shutdown and part load operation.
Multiple commenters (3175, 7990, 8937, 9194, 9382, 9422, 9425, 9497, 9591, 9597, 9678, 9780, 10052, 10083, 10095, 10105, 10243, 10621, 10665, and 10929) noted that increases in the use of renewable energy sources will impact CO2 emissions from natural gas turbine units.
Commenter 9425 stated, "EPA cannot rely on a retrospective analysis of NGCC emissions, but instead must assess whether known increases in renewables penetration will affect emissions rates and the ability of NGCC units to meet a 1,000 lb CO2/MWh standard consistently."
Commenter 9678 noted that "EPA has not fully assessed the standard against projected increase in renewable units." Commenter 9678 noted that, "according to EIA, 37 states have renewable portfolio standards (RPS) or renewable targets. As a result, system operators will increasingly have to manage the bulk electric system to accommodate the electricity from renewable sources. Whereas conventional thermal resources such as coal-fired power plants and natural gas turbines turn output up and down by increasing or decreasing fuel consumption in response to electricity demands, variable renewable energy sources such as wind and solar increase and decrease output based on wind and daylight conditions. These factors may or may not be correlated with demand and are outside the control of system operators. In order to keep supply and demand of electricity evenly balanced, grid operators must address the variability of output from these energy sources, and the uncertainty associated with the timing and magnitude of that variability." 
 Commenter (9772) stated that because large-scale wind generation is non-dispatchable, it is very difficult for a regulating unit to predict when it may need to operate to balance drops in wind generation. The commenter stated that in response to this variability, a regulating unit may be required to operate in hot standby, ready to respond in minutes to a drastic drop in renewable generation. The heat rate achieved while operating a combustion turbine in hot standby results in a disproportionately large output efficiency-based emission rate of CO2. The commenter stated that even the most efficient plants on the market claim that they cannot meet the proposed efficiency-based emission rate of 1,100 lb CO2/MWh.
Commenter 10052 asked that EPA "avoid the unintended consequence of utilities merely building and operating a combined-cycle unit rather than building renewable generation that is backed up by a natural gas combined-cycle unit" by setting an appropriate standard.  
Commenter 10095 asked that EPA "account for increased cycling of NGCCs due to the potential influx of more renewable generation." 
Commenter 10665 noted, "it is very unclear whether new NGCC units can sustain that high level when they are used to firm up delivery of wind and other intermittent generation." 
Commenter 10052 noted, "Emissions increase on an average megawatt basis as load is reduced; even base load units are required to operate at lower capacities periodically to accommodate integration of renewables and other system, grid, or electricity market needs." 
See Section IX.B.2 of the preamble to the final rule for an explanation of why the sliding-scale and non base-load unit subcategory provide sufficient flexibility for simple cycle and fast-start NGCC turbines to back up future intermittent renewable generation. 
Commenter 8952 asked that EPA account for fuel quality in setting the combustion turbine emission limit. Based on "published performance data from Gas Turbine World for a range of gas turbines from 3 MW up to 300 MW," the minimum required efficiency to meet the standard is "essentially a function of the presence of heavier hydrocarbons in the fuel supply."
 The final output-based standard of 1,000 lb CO2/MWh-g only applies to base load combustion turbines that burn over 90 percent natural gas. If a NGCC unit is mixing in less than 10 percent heavier hydrocarbon fuels, the efficiency impact would be minimal. If the units burns 10 percent or more non-natural gas fuel, the unit will fall into the multi-fuel unit subcategory and comply with a standard based on the use of clean fuels, which does not consider the efficiency of the combustion turbine. Therefore, the final standards will not limit the use of heavier hydrocarbon fuels. 
Multiple commenters (9381, 9425, 9426, 9499, 9665, 9780, 10052, 10083, and 10105) offered miscellaneous suggestions regarding factors that are outside the operators' control that should be addressed in EPA's standard for stationary combustion turbines.
See Section IX.D.3.a of the preamble to the final rule for an explanation of why the EPA's final standard for base load natural-gas fired combustion turbines can be met by all turbine sizes and is sufficiently conservative to account for a variety of factors, including those mentioned by the commenters.
Other Subcategories
7.4.1 Peaking Units, Intermediate Load Units, and Base Load Units
Multiple commenters (2984, 8952, 8974, 9034, 9042, 9194, 9382, 9396, 9403, 9407, 9424, 9425, 9471, 9499, 9513, 9514, 9591, 9592, 9602, 9661, 9665, 9667, 9723, 9730, 9734, 9770, 9777, 9780, 10023, 10031, 10043, 10083, 10095, 10098, 10239, 10242, 10243, 10389, 10394, 10395, 10518, 10520, 10660, 10665, 10929, and 10950) recommended other subcategories be included in the stationary combustion turbine standards.
The main three suggestions were as follows:
 peaking unit, intermediate load units, and base load units
 simple cycle units
 firing of other fuels
These subcategories are covered in separate subsections below.
Other commenters asked for the following subcategories:
 "combined cycle units operating at loads below 50% or with multiple starts"  (Commenter 8952)
 projects under development (commenter 9403)
 % of natural gas combusted (commenter 9513)
 operating hours less than 2920 per year (commenter 9513)
 "the addition of a heat recovery steam generator to an existing CT to convert it to more efficient NGCC operation would not be covered by the proposed rule" (commenter 9382)
 "grid-connected power plants equipped with new innovative technologies for research and development purposes, while allowing potential short-term exceedances of air emissions standards during testing, including greenhouse gas performance standards" and test facilities (commenter 10389)
Commenters 9513 and 9514 recommended a three-tiered subcategory structure for the stationary combustion turbine standards: peaking units, intermediate/load following units, and base load units.  According to Commenters 9513 and 9514, the recommended standards (on a net output basis) are 825 lbs CO2/MWh for base load units (those operating over 4,000 hours annually), 875 lbs CO2/MWH for intermediate and load-following units (those operating between 1,200 and 4,000 hours annually), and 1,100 lbs CO2/MWh for peaking units (those operating less than 1,200 hours per year).
According to commenter 9514, "the difference in proposed emission limits is specifically designed to account for any loss in efficiency due to the cycling that is necessary for load-following operations." 
Commenter 9514 stated that "published pricing data shows no significant upfront cost difference between more efficient and less efficient CCGTs or between more efficient and less efficient CTs within the same size range. Since operating costs are lower at more efficient units there is no cost to the industry, or for that matter to individual units, that can be assigned to the emission rates we have proposed, and indeed, plant operators would likely recognize cost savings over the long run."
Commenter 9514 stated, "data already included in the record indicates that units such as those described above (combined cycle peakers) can meet stringent CO2 performance standards even when they undergo frequent cycling. As part of its study of the performance of over three hundred NGCC units, EPA evaluated whether units that cycle more frequently exhibit higher CO2 emission rates. Although the units included in the study pool had a wide range of cycling behavior, ranging from to 1,553 starts per year, EPA found "limited correlation" between the number of starts and CO2 emission rates. In addition, EPA found that the average CO2 emission rate of the ten units that cycled most frequently was 883 lb/MWh, which is very close to our recommended standard for intermediate load units. These results confirm that load-following units are capable of meeting an emission standard that is much more stringent than the 1,000 and 1,100 lb/MWh standards that EPA has proposed." 
Commenter 9514 noted that the "intermediate tier should be set at a limit that modern fast-response CCGT units are capable of meeting, as a forward-looking matter, even if CT units cannot meet the limit. The performance capabilities of simple cycle turbines should not provide the base assumption for intermediate units because they are inherently less efficient, and are currently used primarily as peakers."
While a three-tiered approach to subcategorization is theoretically appealing, as explained in the preamble, the EPA has concluded that we do not have sufficient information at this time to establish emission standard for three tiers. Instead, the EPA is finalizing the sliding-scale approach to subcategorize non-base load and base load units. Because the sliding-scale approach bases the transition from the non-base load to the base load subcategory on an amount of electric sales equal to the design efficiency of the combustion turbine, it could be considered a modified three-tiered approach. The sliding scale incentivizes the installation of the most efficient non-base load units by allowing them to sell more electricity to the grid before being subject to the base load standard. While the non-base load emission standard is not especially stringent, the approach will achieve some minor emission reductions from non-base load units. See Sections IX.B.2 and IX.D.3 of the preamble to the final rule for a more detailed discussion. 
In regards to combustion turbines involved in research and development activities, they would only be subject to the base load standard if the applicable percentage electric sales criterion is met. Research and development turbines would otherwise be subject to the clean fuels standard. We have concluded that this approach provides adequate flexibility for activities that result in short-term increases in CO2 emission rates.
Commenter 9514 offered multiple reasons that a combined cycle unit is capable of providing fast-response generation. 
Commenter 9514 stated, "For the purposes of reliability and renewable integration, combined-cycle units are fully capable of providing fast-response generation. They are therefore fully capable of matching variable renewable output, and can satisfy load-following and immediate dispatch needs in manner comparable, if not identical, to simple cycle units. Siemens has published documentation showing that its Fast Start 30 is capable of 10 minute starts after an overnight shutdown. (Commenter 9514 provided documentation as an attachment and as a link: http://www.energy.siemens.com/hq/pool/hq/power-generation/gas-turbines/downloads/SGT6-5000F_ApplicationOverview.pdf) Longer times necessary to reach full load are limited to circumstances where an operator elects to shut the unit down for more than 48 hours. There is no technological limitation requiring a unit to shut down for that period of time, but an operator may elect to do so if the unit will not be needed for that duration. However, even under this scenario, full output of the combustion turbines that are components of these units are available within 10 minutes."
Commenter 9514 stated, "combined cycle units can act as peakers or load-following units by ramping up their combustion turbines very quickly, while still meeting full load simply by warming up the heat recovery steam generator in anticipation of increased demand. This point is important because the peak is rarely a surprise. Utilities are quite good at estimating peak demand based on weather and usage patterns. Thus, operators have sufficient time to warm up a combined-cycle unit to meet full-load needs, while at the same time having sufficient flexibility to dispatch units quickly at more than half of their full load capacities within 10 minutes if an urgent need arises."
Commenter 9514 noted that "the proposed Oakley Generating Station in California has been designed with the capability to start up and dispatch quickly with GE's Rapid Response package [which] will allow the plant to start up from warm or hot conditions in less than 30 minutes." 
Commenter 9514 noted, "it is factually inaccurate to claim that combined-cycle units are incapable of meeting the technical function of a load-following unit. Advances in HRSG technology have allowed for faster response times with reduced or even eliminated thermal penalties." 
The EPA agrees that fast-start NGCC units have operating flexibilities that allow them to provide rapid power when it is urgently needed and to operate as load following units. However, while these newer NGCC designs are promising emission reduction technologies, they have not been deployed in the numbers necessary for use to have confidence that they can provide sufficient backup power to accommodate the potential rapid growth of renewable generation. While we anticipate that state and federally issued permits for new EGUs will consider the CO2 benefits of fast-start NGCC technologies compared to simple cycle turbines, the EPA has concluded at this time that it is appropriate to finalize a percentage electric sales threshold that provides additional flexibility for simple cycle turbines. The EPA notes that part of the rationale for capping the sliding-scale percentage electric sales threshold at 50 percent, and not 45 percent, is to avoid introducing a regulatory disadvantage for fast-start NGCC units intended for peaking and load following applications. This threshold provides additional flexibilities to fast-start NGCC units relative to simple cycle units because they can sell a greater percentage of their potential electric output prior to being subcategorized as base load units.
7.4.2 Simple Cycle Units
Commenter 10098 noted that the three year averaging period to determine 33% sales criteria delays the applicability determination.
 As discussed in Sections IX.A and B in the preamble to the final rule, the percentage electric sales criterion is no longer used to determine the applicability of the rule. The percentage electric sales threshold in the final rule is used to distinguish between the base load and non-base load subcategories of natural gas-fired combustion turbines. Both subcategories are subject to CO2 emission standards under the final rule.
Multiple commenters (2984, 8952, 8974, 9034, 9042, 9194, 9382, 9396, 9407, 9424, 9425, 9471, 9499, 9591, 9602, 9661, 9665, 9667, 9723, 9730, 9734, 9770, 9777, 9780, 10023, 10031, 10043, 10083, 10095, 10098, 10239, 10242, 10243, 10390, 10394, 10395, 10518, 10520, 10660, 10665, 10788,10929, 10950, 10965, and 10991) stated that standards should not apply to simple cycle turbines. 
Commenter 8952 noted that gas turbines can only achieve the standard if "they are used with energy recovery in a combined cycle mode."
Commenters 8952, 9194, 9592, 9602, 9667, 9780, 10098, 10390, 10660, and 10950 added that a separate BSER is required for simple cycle units. Commenters 9667, 9780 and 10098 noted that EPA has the authority under Section 111(b)(2) to establish subcategories.
Commenters (8952, 9194, 9592, 9602, 10098, 10390, and 10660) did not specify whether the capacity factor should remain as part of the separate BSER, only that the current BSER is not sufficient for simple cycle units. 
Commenter 9425 noted that NGCC cannot be BSER for simple cycle turbines because "NGCC is a technology choice and is not a 'system of emission reductions' whereas the efficient operation of a NGCC could be considered a 'system of emission reduction.'"
Commenter 9499 noted that simple cycle units use less water than combined cycle units and stated that "incentivizing CC over SC units therefore has the perverse effect of increasing EGU water consumption or requiring expensive water conserving upgrades to CC units."
Commenters 9592 and 9602 noted that "EPA cannot propose a standard that would apply to simple cycles without such a [BSER] determination."
Commenter 9602 noted that "EPA did not provide any analysis of what the actual historical emissions rates have been for the many simple cycle combustion turbines installed throughout the United States, nor has EPA attempted to determine what BSER is for simple cycles." 
Commenter 9665 noted that "EPA's proposed treatment of simple cycle units could easily lead to the perverse outcome of sources electing to use smaller, less efficient, higher-emitting turbines simply to avoid the NSPS--with the result being higher CO2 emissions."
Commenter 9667 noted "It is also important for EPA to ensure that it has sufficient record evidence demonstrating that 'significant variables relevant to the standard's achievability' were taken into account because 'promulgation of standards based upon inadequate proof of achievability would defy the Administrative Procedures Act's mandate against action that is 'arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.'"
Commenter 9667 noted that "Although the proposed standard may be achievable by some of the most efficient simple cycle turbines under optimum load and ISO conditions, a variety of real world operating factors make the standard unachievable." Commenter 9667 offered the following examples:
 "Simple cycle turbines critical role as peak and backup power units mean they often operate at reduced loads and are required to quickly ramp up and down production." 
 "Operation at reduced loads is less efficient for any unit resulting in an increase of carbon emissions at lower load levels."
 "Similarly, load cycling to meet peak demand also can lead to inefficiencies."
 "The performance of simple cycle units is also decreased when they operate on backup fuel during times when natural gas is unavailable."
Commenters 9667 and 10950 noted that the standards "are not achievable for even the most advanced simple cycle units." Commenter 9667 and 10950 further stated "If EPA does include simple cycle units supplying one-third or more of their potential output to the grid, the proposal needs to include separate standards for simple cycle units that are reasonably achievable" and noted that the limit "would be substantially higher than the proposed standard for NGCC." 
Commenter 9780 stated, "If EPA intends to subject simple-cycle CTs that fail a sales-based test to some emissions standards, EPA must take into consideration their ability to achieve those standards."  
Commenter 10098 noted the following reasons that the BSER for combined cycle turbines should not apply to simple cycle turbines:
 simple cycle turbines operate at lower, less efficient loads and cannot use a heat recovery steam generator; nor are there add-on controls available to control CO2 emissions
 EPA should "perform a new BSER analysis based on actual, real-world simple cycle emissions data instead of a summary of average emissions (which cannot be found) for unidentified 'advanced' units."
 EPA did not identify a single peaking plant that uses combined cycle technology; therefore peaking plants that employ simple cycle turbines should be regulated separately
 "EPA relies on data showing that 0.2% of simple cycle turbines sold more than one-third of their potential electric output to the grid over a three-year averaging period."  However, commenter 10098 noted that "simple cycle turbines may operate more frequently as renewable power generation increases amid base load plant retirements" and EPA should regulate them in a way that allows the greatest flexibility.
 "EPA's only consideration of BSER for simple cycle turbines was 'high efficiency' or 'advanced' simple cycle turbines. These units are never described or defined."
Commenter 10390 noted that the performance standards "could even consider that aeroderivative SCTs are more efficient at low loads." 
Commenter 10660 stated that "EPA's proposed rule is factually unsupported and not grounded in the statute...because it cannot propose to apply the standard to simple cycle turbines without first showing that the standards reflect a reasoned BSER determination." 
Commenter 10950 noted that "EPA could exempt simple cycle units operating at or below a capacity factor of 50 percent."
Multiple commenters (8952, 9661, 9730, and 10394) recommended a separate limit for simple cycle turbines, rather than an exemption:
 Commenter 10394 recommended 1,300. 
 Commenter 8952 recommended "significantly above 1,100 lb/MWh."
 Commenter 9730 "supports a separate emission limit, rather than an exemption, for simple cycle turbines that provide grid stability, (a.k.a., peaking units)."
 Commenter 9730 recommended working "with California air districts...to develop simple cycle turbine limits based on currently permitted operations."
 Commenter 9661 asked to exempt simple cycle units that supply "less than a certain portion of its potential electric output to the grid on a 3-year rolling average."
 The commenters are mistaken that the EPA proposed that simple cycle turbines must meet an emission standard based on the performance of NGCC units. The EPA is finding that NGCC technology is the BSER for base load natural gas-fired combustion turbine applications. The implication of this is that an owner or operator of a new unit that intends to sell electricity in excess of the percentage electric sales threshold, which divides the base load and non-base load subcategories, will need to meet an emission standard based on NGCC technology. In other words, new simple cycle turbines will be limited to selling a percentage of their potential electric output equal to or less than their design efficiencies so that they can comply with the non-base load standard.  
As stated in the preamble, we would ideally establish an output-based (i.e., lb CO2/MWh) standard for non-base load units. However, we do not have sufficient information at this time to set a meaningful output-based standard that could be achieved by the universe of simple cycle turbines operating as peaking and cycling units. Therefore, the final standard is on an input basis (i.e., lb CO2/MMBtu) and based on the use of clean fuels.

The commenter is also mistaken that NGCC units require more water than simple cycle units. NGCC units using dry cooling towers do not require any more water than a comparable simple cycle combustion turbine.
Commenter 8952 noted that a gas turbine is not the only power conversion system available, and the rule may be a "disincentive for end-users to select the cleanest and most efficient technology (the gas turbine) in many applications." 
The final percentage electric sales threshold is sufficiently flexible to avoid creating a regulatory disincentive to use stationary combustion turbines in lieu of other technologies (e.g., reciprocating engines). Further, even though competing technologies are not subject to the requirements of this final rule, they would likely have to get a PSD GHG permit. During the review process, the permitting authority has the ability to compare the GHG mission rates of competing technologies that could serve the same role.
Multiple commenters (2984, 9499, 9734, 9779, 9780, 10083, and 10098) noted that the future will show increased demand for simple cycle turbines due to the increased use of renewables.
The EPA did an extensive evaluation of the impact of additional generation from intermittent renewable sources on the operation of simple cycle combustion turbines. Based in part on this analysis, the final rule incorporates additional flexibilities for simple cycle units to increase generation beyond historical baselines. See Section IX.B.2 of the preamble to the final rule for additional details.
Multiple commenters (9471, 9591, 9665, 10098, 10239, and 10952) noted that there are differences between the roles of simple cycle and the roles of combined cycle units in the electricity market.
Commenter 9665 stated, "simple cycle units will fulfill an important role that cannot be cost-effectively met with NGCC." 
Commenter 10098 stated that combined cycle turbines are very efficient, have high utilization rates, and are designed for base load power. Commenter 10098 noted that simple cycle turbines have unpredictable hours of operation, rarely operate at full load, and only provide peaking power. Commenter 10098 noted that the "decision to regulate them together [is] arbitrary and capricious." 
Commenter 10098 stated "Relying solely on EIA's Advanced Energy Outlook 2013 report, EPA found that 'advanced simple cycle combustion turbines have a base load rating CO2 emission rate of 1,150 lb CO2/MWh...' 79 Fed. Reg. at 1485. The commenter noted that simple cycle turbines are not used for base load generation, providing further evidence that the EPA fails to understand the difference between the two types of turbines and their respective uses in the power system. If the figure the EPA relied upon reports average CO2 emissions from simple cycle turbines at full load, then it should clarify this statement as well as provide an explanation of why it would rely on that figure given that simple cycle turbines generally run at lower loads."
The EPA recognizes that, in general, simple cycle combustion turbines serve a different role than NGCC units. The subcategorization approach in this final rule recognizes that difference. The non-base load subcategory provides adequate flexibility for owners and operators of new simple cycle combustion turbines to operate them to fulfill the same role simple cycle turbines have historically filled, while providing room for growth to back up intermittent renewables. 
Three commenters (9780, 10098, and 10950) noted consequences of including simple cycle turbines in the NSPS.
Commenter 9780 noted negative impacts to the stability and reliability of the electrical grid. 
Commenter 10098 noted increased costs, increased risk of electricity disruption, and significant uncertainty related to grid reliability.
Commenter 10950 noted that including simple cycle turbines in the rule "could actually increase overall CO2 emissions by incentivizing the purchase of smaller, lower-efficiency turbines and reciprocating engines that EPA would not regulate in this rule." 
 See Section IX.B.2 of the preamble to the final rule for a discussion of why the EPA is subcategorizing combustion turbines based on their electric sales using the sliding scale, which will alleviate the concerns noted by these commenters.
Multiple commenters (7976, 8911, 9514, and 10693) asked that EPA not exempt simple cycle turbines from the standards.
Commenter 8911 noted that dealing with grid emergencies is not a sufficient reason to exempt simple cycle turbines from the standard. Commenter 8911 stated, "if a dire grid emergency requires rapid new capacity additions, the solution is to temporarily waive the NSPS to permit the simple cycle phase for a higher than usual capacity factor until the balance of the combined cycle plant can be commissioned, typically 6 to 18 months later."
Commenter 9514 noted that simple cycle turbines operating over a 1,200 hour threshold (the commenter asked that EPA "define peaking units as units that operate fewer than 1,200 hours per year") should meet the NGCC BSER.
Commenter 9514 noted, "the feasibility of fast-start and quick-ramping combined-cycle turbines has advanced substantially. It is factually inaccurate to claim that combined-cycle units are incapable of meeting the technical function of a load-following unit. Advances in HRSG technology have allowed for faster response times with reduced or even eliminated thermal penalties. In short, CTs are unnecessary-and unnecessarily dirty-options for intermediate and load-following services, and EPA should not dilute the performance standard for gas plants in order to accommodate those less efficient technologies."
The EPA did not exempt simple cycle turbines in the final rule. All combustion turbines that meet the general applicability criteria are subject to an emission standard.
Commenters discussed the risk of facilities using simple cycle turbines instead of combined cycle turbines if simple cycle turbines were exempted from the rule.
Commenter 10693 noted that the exemption of simple cycle turbines from the standard could lead to plants using simple cycle turbines instead of combined cycle turbines, leading to increased emissions.
Commenters 9591 and 9780 countered that argument.
Commenter 9780 noted that it is not economically sound to use simple cycle turbines instead of combined cycle turbines. 
Commenter 9591 stated that excluding simple cycle turbines from the standard would not provide an incentive to build less efficient simple cycle turbines because combined cycle and simple cycle units are not interchangeable within the electricity market. 
 The final rule includes all stationary combustion turbines (both simple and combined cycle turbines) that meet the general applicability criteria. We have concluded that our approach does not provide a regulatory advantage for either technology. See Section IX.A of the preamble to the final rule for a discussion of why the EPA chose to adopt the broad applicability approach.
7.4.3 Other Fuels
Multiple commenters (8974, 9426, 9665, 9666, and 10043) commented on oil-fired combustion turbines.
Multiple commenters (8974, 9426, and 10043) asked that oil-fired units be exempt from the rule. 
Commenter 9665 stated, "EPA should clarify that the exclusion of stationary combustion turbines that operate on non-natural gas fuel applies to all periods of non-natural gas operation." 
Commenter 9666 asked EPA to "analyze CO2 emissions during oil-fired operations, propose an NSPS for such a period of operation, and allow [commenters] to comment on whether such NSPS are achievable." 
  As discussed in Section IX.B.3 in the preamble to the final rule, we are finalizing a separate fuel-based subcategory for multi-fuel-fired combustion turbines. Combustion turbines that burn 90 percent or less natural gas on a 12-operating-month rolling average basis will be included in the multi-fuel-fired subcategory and be subject to a separate emission standard. That emission standard is based on the use of clean fuels. 
Two commenters (7977 and 9654) commented on using other liquid fuels in combustion turbines.
Commenter 7977 noted, "EPA has not adequately documented why the proposed rule is applicable to liquid fuel when combusted in steam generating units or IGCC units, but not when combusted in combustion turbines. The fact that liquid fossil fuels are rarely used in combustion turbines currently is not determinative of future use. Liquid fuels usage is based on economics rather than technical limitations, which will necessarily change if the proposed regulation goes into effect as written."
Commenter 9654 asked that EPA "Exempt operation with alternative fuels such as No. 2 distillate fuel (and its analogs-kerosene, kero-jet, Jet-A, Naphtha, etc.)." 
As discussed in Section IX.B.3 in the preamble to the final rule, we are finalizing a separate subcategory for combustion turbines that burn 90 percent or less natural gas on a 12-operating-month rolling average basis will be included in the multi-fuel-fired subcategory and be subject to a separate emission standard.
