New Source Performance Standards for Greenhouse Gas Emissions From New, Modified, and Reconstructed Fossil Fuel-Fired Electric Generating Units; Emission Guidelines for Greenhouse Gas Emissions From Existing Fossil Fuel-Fired Electric Generating Units; and Repeal of the Affordable Clean Energy Rule

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

The Environmental Protection Agency (EPA) is finalizing multiple actions under section 111 of the Clean Air Act (CAA) addressing greenhouse gas (GHG) emissions from fossil fuel-fired electric generating units (EGUs). First, the EPA is finalizing the repeal of the Affordable Clean Energy (ACE) Rule. Second, the EPA is finalizing emission guidelines for GHG emissions from existing fossil fuel-fired steam generating EGUs, which include both coal-fired and oil/gas-fired steam generating EGUs. Third, the EPA is finalizing revisions to the New Source Performance Standards (NSPS) for GHG emissions from new and reconstructed fossil fuel-fired stationary combustion turbine EGUs. Fourth, the EPA is finalizing revisions to the NSPS for GHG emissions from fossil fuel-fired steam generating units that undertake a large modification, based upon the 8-year review required by the CAA. The EPA is not finalizing emission guidelines for GHG emissions from existing fossil fuel-fired stationary combustion turbines at this time; instead, the EPA intends to take further action on the proposed emission guidelines at a later date.

DATES:

This final rule is effective on July 8, 2024. The incorporation by reference of certain publications listed in the rules is approved by the Director of the Federal Register as of July 8, 2024. The incorporation by reference of certain other materials listed in the rule was approved by the Director of the Federal Register as of October 23, 2015.

ADDRESSES:

The EPA has established a docket for these actions under Docket ID No. EPA-HQ-OAR-2023-0072. All documents in the docket are listed on the https://www.regulations.gov website. Although listed, some information is not publicly available, e.g., Confidential Business Information (CBI) or other information whose disclosure is restricted by statute. Certain other material, such as copyrighted material, is not placed on the internet and will be publicly available only in hard copy form. Publicly available docket materials are available electronically through https://www.regulations.gov .

FOR FURTHER INFORMATION CONTACT:

Lisa Thompson (she/her), Sector Policies and Programs Division (D243-02), Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, 109 T.W. Alexander Drive, P.O. Box 12055, Research Triangle Park, North Carolina 27711; telephone number: (919) 541-5158; and email address: thompson.lisa@epa.gov.

SUPPLEMENTARY INFORMATION:

Preamble acronyms and abbreviations. Throughout this document the use of “we,” “us,” or “our” is intended to refer to the EPA. The EPA uses multiple acronyms and terms in this preamble. While this list may not be exhaustive, to ease the reading of this preamble and for reference purposes, the EPA defines the following terms and acronyms here:

ACE Affordable Clean Energy rule

BSER best system of emissions reduction

Btu British thermal unit

CAA Clean Air Act

CBI Confidential Business Information

CCS carbon capture and sequestration/storage

CCUS carbon capture, utilization, and sequestration/storage

CO₂  carbon dioxide

DER distributed energy resources

DOE Department of Energy

EEA energy emergency alert

EGU electric generating unit

EIA Energy Information Administration

EJ environmental justice

E.O. Executive Order

EPA Environmental Protection Agency

FEED front-end engineering and design

FGD flue gas desulfurization

FR Federal Register

GHG greenhouse gas

GW gigawatt

GWh gigawatt-hour

HAP hazardous air pollutant

HRSG heat recovery steam generator

IIJA Infrastructure Investment and Jobs Act

IRC Internal Revenue Code

kg kilogram

kWh kilowatt-hour

LCOE levelized cost of electricity

LNG liquefied natural gas

MATS Mercury and Air Toxics Standards

MMBtu/h million British thermal units per hour

MMT CO₂ e million metric tons of carbon dioxide equivalent

MW megawatt

MWh megawatt-hour

NAAQS National Ambient Air Quality Standards

NESHAP National Emission Standards for Hazardous Air Pollutants

NGCC natural gas combined cycle

NO_X  nitrogen oxides

NSPS new source performance standards

NSR New Source Review

PM particulate matter

PM_2.5  fine particulate matter

RIA regulatory impact analysis

TSD technical support document

U.S. United States

Organization of this document. The information in this preamble is organized as follows:

I. Executive Summary

A. Climate Change and Fossil Fuel-Fired EGUs

B. Recent Developments in Emissions Controls and the Electric Power Sector

C. Summary of the Principal Provisions of These Regulatory Actions

D. Grid Reliability Considerations

E. Environmental Justice Considerations

F. Energy Workers and Communities

G. Key Changes From Proposal

II. General Information

A. Action Applicability

B. Where To Get a Copy of This Document and Other Related Information

III. Climate Change Impacts

IV. Recent Developments in Emissions Controls and the Electric Power Sector

A. Background

B. GHG Emissions From Fossil Fuel-Fired EGUs

C. Recent Developments in Emissions Control

D. The Electric Power Sector: Trends and Current Structure

E. The Legislative, Market, and State Law Context

F. Future Projections of Power Sector Trends

V. Statutory Background and Regulatory History for CAA Section 111

A. Statutory Authority To Regulate GHGs From EGUs Under CAA Section 111

B. History of EPA Regulation of Greenhouse Gases From Electricity Generating Units Under CAA Section 111 and Caselaw

C. Detailed Discussion of CAA Section 111 Requirements

VI. ACE Rule Repeal

A. Summary of Selected Features of the ACE Rule

B. Developments Undermining ACE Rule's Projected Emission Reductions

C. Developments Showing That Other Technologies Are the BSER for This Source Category

D. Insufficiently Precise Degree of Emission Limitation Achievable From Application of the BSER

E. Withdrawal of Proposed NSR Revisions

VII. Regulatory Approach for Existing Fossil Fuel-Fired Steam Generating Units

A. Overview

B. Applicability Requirements and Fossil Fuel-Type Definitions for Subcategories of Steam Generating Units

C. Rationale for the BSER for Coal-Fired Steam Generating Units

D. Rationale for the BSER for Natural Gas-Fired and Oil-Fired Steam Generating Units

E. Additional Comments Received on the Emission Guidelines for Existing Steam Generating Units and Responses

F. Regulatory Requirement To Review Emission Guidelines for Coal-Fired Units

VIII. Requirements for New and Reconstructed Stationary Combustion Turbine EGUs and Rationale for Requirements

A. Overview

B. Combustion Turbine Technology

C. Overview of Regulation of Stationary Combustion Turbines for GHGs

D. Eight-Year Review of NSPS

E. Applicability Requirements and Subcategorization

F. Determination of the Best System of Emission Reduction (BSER) for New and Reconstructed Stationary Combustion Turbines

G. Standards of Performance

H. Reconstructed Stationary Combustion Turbines

I. Modified Stationary Combustion Turbines

J. Startup, Shutdown, and Malfunction

K. Testing and Monitoring Requirements

L. Recordkeeping and Reporting Requirements

M. Compliance Dates

N. Compliance Date Extension

IX. Requirements for New, Modified, and Reconstructed Fossil Fuel-Fired Steam Generating Units

A. 2018 NSPS Proposal Withdrawal

B. Additional Amendments

C. Eight-Year Review of NSPS for Fossil Fuel-Fired Steam Generating Units

D. Projects Under Development

X. State Plans for Emission Guidelines for Existing Fossil Fuel-Fired EGUs

A. Overview

B. Requirement for State Plans To Maintain Stringency of the EPA's BSER Determination

C. Establishing Standards of Performance

D. Compliance Flexibilities

E. State Plan Components and Submission

XI. Implications for Other CAA Programs

A. New Source Review Program

B. Title V Program

XII. Summary of Cost, Environmental, and Economic Impacts

A. Air Quality Impacts

B. Compliance Cost Impacts

C. Economic and Energy Impacts

D. Benefits

E. Net Benefits

F. Environmental Justice Analytical Considerations and Stakeholder Outreach and Engagement

G. Grid Reliability Considerations and Reliability-Related Mechanisms

XIII. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review and Executive Order 14094: Modernizing Regulatory Review

B. Paperwork Reduction Act (PRA)

C. Regulatory Flexibility Act (RFA)

D. Unfunded Mandates Reform Act of 1995 (UMRA)

E. Executive Order 13132: Federalism

F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

G. Executive Order 13045: Protection of Children From Environmental Health Risks and Safety Risks Populations and Low-Income Populations

H. Executive Order 13211: Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer and Advancement Act (NTTAA) and 1 CFR Part 51

J. Executive Order 12898: Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations and Executive Order 14096: Revitalizing Our Nation's Commitment to Environmental Justice for All

K. Congressional Review Act (CRA)

XIV. Statutory Authority

I. Executive Summary

In 2009, the EPA concluded that GHG emissions endanger our nation's public health and welfare. Since that time, the evidence of the harms posed by GHG emissions has only grown, and Americans experience the destructive and worsening effects of climate change every day. Fossil fuel-fired EGUs are the nation's largest stationary source of GHG emissions, representing 25 percent of the United States' total GHG emissions in 2021. At the same time, a range of cost-effective technologies and approaches to reduce GHG emissions from these sources is available to the power sector—including carbon capture and sequestration/storage (CCS), co-firing with less GHG-intensive fuels, and more efficient generation. Congress has also acted to provide funding and other incentives to encourage the deployment of various technologies, including CCS, to achieve reductions in GHG emissions from the power sector.

In this notice, the EPA is finalizing several actions under section 111 of the Clean Air Act (CAA) to reduce the significant quantity of GHG emissions from fossil fuel-fired EGUs by establishing emission guidelines and new source performance standards (NSPS) that are based on available and cost-effective technologies that directly reduce GHG emissions from these sources. Consistent with the statutory command of CAA section 111, the final NSPS and emission guidelines reflect the application of the best system of emission reduction (BSER) that, taking into account costs, energy requirements, and other statutory factors, is adequately demonstrated.

Specifically, the EPA is first finalizing the repeal of the Affordable Clean Energy (ACE) Rule. Second, the EPA is finalizing emission guidelines for GHG emissions from existing fossil fuel-fired steam generating EGUs, which include both coal-fired and oil/gas-fired steam generating EGUs. Third, the EPA is finalizing revisions to the NSPS for GHG emissions from new and reconstructed fossil fuel-fired stationary combustion turbine EGUs. Fourth, the EPA is finalizing revisions to the NSPS for GHG emissions from fossil fuel-fired steam generating units that undertake a large modification, based upon the 8-year review required by the CAA. The EPA is not finalizing emission guidelines for GHG emissions from existing fossil fuel-fired combustion turbines at this time and plans to expeditiously issue an additional proposal that more comprehensively addresses GHG emissions from this portion of the fleet. The EPA acknowledges that the share of GHG emissions from existing fossil fuel-fired combustion turbines has been growing and is projected to continue to do so, particularly as emissions from other portions of the fleet decline, and that it is vital to regulate the GHG emissions from these sources consistent with CAA section 111.

These final actions ensure that the new and existing fossil fuel-fired EGUs that are subject to these rules reduce their GHG emissions in a manner that is cost-effective and improves the emissions performance of the sources, consistent with the applicable CAA requirements and caselaw. These standards and emission guidelines will significantly decrease GHG emissions from fossil fuel-fired EGUs and the associated harms to human health and welfare. Further, the EPA has designed these standards and emission guidelines in a way that is compatible with the nation's overall need for a reliable supply of affordable electricity.

A. Climate Change and Fossil Fuel-Fired EGUs

These final actions reduce the emissions of GHGs from new and existing fossil fuel-fired EGUs. The increasing concentrations of GHGs in the atmosphere are, and have been, warming the planet, resulting in serious and life-threatening environmental and human health impacts. The increased concentrations of GHGs in the atmosphere and the resulting warming have led to more frequent and more intense heat waves and extreme weather events, rising sea levels, and retreating snow and ice, all of which are occurring at a pace and scale that threaten human health and welfare.

Fossil fuel-fired EGUs that are uncontrolled for GHGs are one of the biggest domestic sources of GHG emissions. At the same time, there are technologies available (including technologies that can be applied to fossil fuel-fired power plants) to significantly reduce emissions of GHGs from the power sector. Low- and zero-GHG electricity are also key enabling technologies to significantly reduce GHG emissions in almost every other sector of the economy.

In 2021, the power sector was the largest stationary source of GHGs in the United States, emitting 25 percent of overall domestic emissions. In 2021, existing fossil fuel-fired steam generating units accounted for 65 percent of the GHG emissions from the sector, but only accounted for 23 percent of the total electricity generation.

Because of its outsized contributions to overall emissions, reducing emissions from the power sector is essential to addressing the challenge of climate change—and sources in the power sector also have many available options for reducing their climate-destabilizing emissions. Particularly relevant to these actions are several key technologies (CCS and co-firing of lower-GHG fuels) that allow fossil fuel-fired steam generating EGUs and stationary combustion turbines to provide power while emitting significantly lower GHG emissions. Moreover, with the increased electrification of other GHG-emitting sectors of the economy, such as personal vehicles, heavy-duty trucks, and the heating and cooling of buildings, reducing GHG emissions from these affected sources can also help reduce power sector pollution that might otherwise result from the electrification of other sectors of the economy.

B. Recent Developments in Emissions Controls and the Electric Power Sector

Several recent developments concerning emissions controls are relevant for the EPA's determination of the BSER for existing coal-fired steam generating EGUs and new natural gas-fired stationary combustion turbines. These include lower costs and continued improvements in CCS technology, alongside Federal tax incentives that allow companies to largely offset the cost of CCS. Well-established trends in the sector further inform where using such technologies is cost effective and feasible, and form part of the basis for the EPA's determination of the BSER.

In recent years, the cost of CCS has declined in part because of process improvements learned from earlier deployments and other advances in the technology. In addition, the Inflation Reduction Act (IRA), enacted in 2022, extended and significantly increased the tax credit for carbon dioxide (CO₂) sequestration under Internal Revenue Code (IRC) section 45Q. The provision of tax credits in the IRA, combined with the funding included in the Infrastructure Investment and Jobs Act (IIJA), enacted in 2021, incentivize and facilitate the deployment of CCS and other GHG emission control technologies. As explained later in this preamble, these developments support the EPA's conclusion that CCS is the BSER for certain subcategories of new and existing EGUs because it is an adequately demonstrated and available control technology that significantly reduces emissions of dangerous pollution and because the costs of its installation and operation are reasonable. Some companies have already made plans to install CCS on their units independent of the EPA's regulations.

Well documented trends in the power sector also influence the EPA's determination of the BSER. In particular, CCS entails significant capital expenditures and is only cost-reasonable for units that will operate enough to defray those capital costs. At the same time, many utilities and power generating companies have recently announced plans to accelerate changing the mix of their generating assets. The IIJA and IRA, state legislation, technology advancements, market forces, consumer demand, and the advanced age of much of the existing fossil fuel-fired generating fleet are collectively leading to, in most cases, decreased use of the fossil fuel-fired units that are the subjects of these final actions. From 2010 through 2022, fossil fuel-fired generation declined from approximately 72 percent of total net generation to approximately 60 percent, with generation from coal-fired sources dropping from 49 percent to 20 percent of net generation during this period. These trends are expected to continue and are relevant to determining where capital-intensive technologies, like CCS, may be feasibly and cost-reasonably deployed to reduce emissions.

Congress has taken other recent actions to drive the reduction of GHG emissions from the power sector. As noted earlier, Congress enacted IRC section 45Q in section 115 of the Energy Improvement and Extension Act of 2008 to provide a tax credit for the sequestration of CO₂. Congress significantly amended IRC section 45Q in the Bipartisan Budget Act of 2018, and more recently in the IRA, to make this tax incentive more generous and effective in spurring long-term deployment of CCS. In addition, the IIJA provided more than $65 billion for infrastructure investments and upgrades for transmission capacity, pipelines, and low-carbon fuels. Further, the Creating Helpful Incentives to Produce Semiconductors and Science Act (CHIPS Act) authorized billions more in funding for development of low- and non-GHG emitting energy technologies that could provide additional low-cost options for power companies to reduce overall GHG emissions. As discussed in greater detail in section IV.E.1 of this preamble, the IRA, the IIJA, and CHIPS contain numerous other provisions encouraging companies to reduce their GHGs.

C. Summary of the Principal Provisions of These Regulatory Actions

These final actions include the repeal of the ACE Rule, BSER determinations and emission guidelines for existing fossil fuel-fired steam generating units, and BSER determinations and accompanying standards of performance for GHG emissions from new and reconstructed fossil fuel-fired stationary combustion turbines and modified fossil fuel-fired steam generating units.

The EPA is taking these actions consistent with its authority under CAA section 111. Under CAA section 111, once the EPA has identified a source category that contributes significantly to dangerous air pollution, it proceeds to regulate new sources and, for GHGs and certain other air pollutants, existing sources. The central requirement is that the EPA must determine the “best system of emission reduction . . . adequately demonstrated,” taking into account the cost of the reductions, non-air quality health and environmental impacts, and energy requirements. The EPA may determine that different sets of sources have different characteristics relevant for determining the BSER and may subcategorize sources accordingly.

Once it identifies the BSER, the EPA must determine the “degree of emission limitation” achievable by application of the BSER. For new sources, the EPA establishes the standard of performance with which the sources must comply, which is a standard for emissions that reflects the degree of emission limitation. For existing sources, the EPA includes the information it has developed concerning the BSER and associated degree of emission limitation in emission guidelines and directs the states to adopt state plans that contain standards of performance that are consistent with the emission guidelines.

Since the early 1970s, the EPA has promulgated regulations under CAA section 111 for more than 60 source categories, which has established a robust set of regulatory precedents that has informed the development of these final actions. During this period, the courts, primarily the U.S. Court of Appeals for the D.C. Circuit and the Supreme Court, have developed a body of caselaw interpreting CAA section 111. As the Supreme Court has recognized, the EPA has typically (and does so in these actions) determined the BSER to be “measures that improve the pollution performance of individual sources,” such as add-on controls and clean fuels. West Virginia v. EPA, 597 U.S. 697, 734 (2022). For present purposes, several of a BSER's key features include that it must reduce emissions, be based on “adequately demonstrated” technology, and have a reasonable cost of control. The case law interpreting section 111 has also recognized that the BSER can be forward-looking in nature and take into account anticipated improvements in control technologies. For example, the EPA may determine a control to be “adequately demonstrated” even if it is new and not yet in widespread commercial use, and, further, that the EPA may reasonably project the development of a control system at a future time and establish requirements that take effect at that time. Further, the most relevant costs under CAA section 111 are the costs to the regulated facility. The actions that the EPA is finalizing are consistent with the requirements of CAA section 111 and its regulatory history and caselaw, which is discussed in further detail in section V of this preamble.

1. Repeal of ACE Rule

The EPA is finalizing its proposed repeal of the existing ACE Rule emission guidelines. First, as a policy matter, the EPA concludes that the suite of heat rate improvements (HRI) that was identified in the ACE Rule as the BSER is not an appropriate BSER for existing coal-fired EGUs. Second, the ACE Rule rejected CCS and natural gas co-firing as the BSER for reasons that no longer apply. Third, the EPA concludes that the ACE Rule conflicted with CAA section 111 and the EPA's implementing regulations because it did not provide sufficient specificity as to the BSER the EPA had identified or the “degree of emission limitation achievable though application of the [BSER].”

Also, the EPA is withdrawing the proposed revisions to the New Source Review (NSR) regulations that were included the ACE Rule proposal (83 FR 44773-83; August 31, 2018).

2. Emission Guidelines for Existing Fossil Fuel-Fired Steam Generating Units

The EPA is finalizing CCS with 90 percent capture as BSER for existing coal-fired steam generating units. These units have a presumptive standard of an 88.4 percent reduction in annual emission rate, with a compliance deadline of January 1, 2032. As explained in detail below, CCS is an adequately demonstrated technology that achieves significant emissions reduction and is cost-reasonable, taking into account the declining costs of the technology and a substantial tax credit available to sources. In recognition of the significant capital expenditures involved in deploying CCS technology and the fact that 45 percent of regulated units already have announced retirement dates, the EPA is finalizing a separate subcategory for existing coal-fired steam generating units that demonstrate that they plan to permanently cease operation before January 1, 2039. The BSER for this subcategory is co-firing with natural gas, at a level of 40 percent of the unit's annual heat input. These units have a presumptive standard of 16 percent reduction in annual emission rate corresponding to this BSER, with a compliance deadline of January 1, 2030.

The EPA is finalizing an applicability exemption for existing coal-fired steam EGUs demonstrating that they plan to permanently cease operation prior to January 1, 2032, based on the Agency's determination that units retiring before this date generally do not have cost-reasonable options for improving their GHG emissions performance. Sources that demonstrate they will permanently cease operation before this applicability deadline will not be subject to these emission guidelines. Further, the EPA is not finalizing the proposed imminent-term or near-term subcategories.

The EPA is finalizing the proposed structure of the subcategory definitions for natural gas- and oil-fired steam generating units. The EPA is also finalizing routine methods of operation and maintenance as the BSER for intermediate load and base load natural gas- and oil-fired steam generating units. Furthermore, the EPA is finalizing presumptive standards for natural gas- and oil-fired steam generating units that are slightly higher than at proposal: base load sources (those with annual capacity factors greater than 45 percent) have a presumptive standard of 1,400 lb CO₂ /MWh-gross, and intermediate load sources (those with annual capacity factors greater than 8 percent and less than or equal to 45 percent) have a presumptive standard of 1,600 lb CO₂ /MWh-gross. For low load (those with annual capacity factors less than 8 percent), the EPA is finalizing a uniform fuels BSER and a presumptive input-based standard of 170 lb CO₂ /MMBtu for oil-fired sources and a presumptive standard of 130 lb CO₂ /MMBtu for natural gas-fired sources.

3. Standards of Performance for New and Reconstructed Fossil Fuel-Fired Combustion Turbines

The EPA is finalizing emission standards for three subcategories of combustion turbines—base load, intermediate load, and low load. The BSER for base load combustion turbines includes two components to be implemented initially in two phases. The first component of the BSER for base load combustion turbines is highly efficient generation (based on the emission rates that the best performing units are achieving) and the second component for base load combustion turbines is utilization of CCS with 90 percent capture. Recognizing the lead time that is necessary for new base load combustion turbines to plan for and install the second component of the BSER ( i.e., 90 percent CCS), including the time that is needed to deploy the associated infrastructure (CO₂ pipelines, storage sites, etc.), the EPA is finalizing a second phase compliance deadline of January 1, 2032, for this second component of the standard.

The EPA has identified highly efficient simple cycle generation as the BSER for intermediate load combustion turbines. For low load combustion turbines, the EPA is finalizing its proposed determination that the BSER is the use of lower-emitting fuels.

4. New, Modified, and Reconstructed Fossil Fuel-Fired Steam Generating Units

The EPA is finalizing revisions of the standards of performance for coal-fired steam generating units that undertake a large modification ( i.e., a modification that increases its hourly emission rate by more than 10 percent) to mirror the emission guidelines for existing coal-fired steam generators. This reflects the EPA's determination that such modified sources are capable of meeting the same presumptive standards that the EPA is finalizing for existing steam EGUs. Further, this revised standard for modified coal-fired steam EGUs will avoid creating an unjustified disparity between emission control obligations for modified and existing coal-fired steam EGUs.

The EPA did not propose, and we are not finalizing, any review or revision of the 2015 standard for large modifications of oil- or gas-fired steam generating units because we are not aware of any existing oil- or gas-fired steam generating EGUs that have undertaken such modifications or have plans to do so, and, unlike an existing coal-fired steam generating EGUs, existing oil- or gas-fired steam units have no incentive to undertake such a modification to avoid the requirements we are including in this final rule for existing oil- or gas-fired steam generating units.

As discussed in the proposal preamble, the EPA is not revising the NSPS for newly constructed or reconstructed fossil fuel-fired steam electric generating units (EGU) at this time because the EPA anticipates that few, if any, such units will be constructed or reconstructed in the foreseeable future. However, the EPA has recently become aware that a new coal-fired power plant is under consideration in Alaska. Accordingly, the EPA is not, at this time, finalizing its proposal not to review the 2015 NSPS, and, instead, will continue to consider whether to review the 2015 NSPS. As developments warrant, the EPA will determine either to conduct a review, and propose revised standards of performance, or not conduct a review.

Also, in this final action, the EPA is withdrawing the 2018 proposed amendments to the NSPS for GHG emissions from coal-fired EGUs.

5. Severability

This final action is composed of four independent rules: the repeal of the ACE rule; GHG emission guidelines for existing fossil fuel-fired steam generating units; NSPS for GHG emissions from new and reconstructed fossil fuel-fired combustion turbines; and revisions to the standards of performance for new, modified, and reconstructed fossil fuel-fired steam generating units. The EPA could have finalized each of these rules in separate Federal Register notices as separate final actions. The Agency decided to include these four independent rules in a single Federal Register notice for administrative ease because they all relate to climate pollution from the fossil fuel-fired electric generating units source category. Accordingly, despite grouping these rules into one single Federal Register notice, the EPA intends that each of these rules described in sections I.C.1 through I.C.4 is severable from the other.

In addition, each rule is severable as a practical matter. For example, the EPA would repeal the ACE Rule separate and apart from finalizing new standards for these sources as explained herein. Moreover, the BSER and associated emission guidelines for existing fossil fuel-fired steam generating units are independent of and would have been the same regardless of whether the EPA finalized the other parts of this rule. In determining the BSER for existing fossil fuel-fired steam generating units, the EPA considered only the technologies available to reduce GHG emissions at those sources and did not take into consideration the technologies or standards of performance for new fossil fuel-fired combustion turbines. The same is true for the Agency's evaluation and determination of the BSER and associated standards of performance for new fossil fuel-fired combustion turbines. The EPA identified the BSER and established the standards of performance by examining the controls that were available for these units. That analysis can stand alone and apart from the EPA's separate analysis for existing fossil fuel-fired steam generating units. Though the record evidence (including, for example, modeling results) often addresses the availability, performance, and expected implementation of the technologies at both existing fossil fuel-fired steam generating units and new fossil fuel-fired combustion turbines in the same record documents, the evidence for each evaluation stands on its own, and is independently sufficient to support each of the final BSERs.

In addition, within section I.C.1, the final action to repeal the ACE Rule is severable from the withdrawal of the NSR revisions that were proposed in parallel with the ACE Rule proposal. Within the group of actions for existing fossil fuel-fired steam generating units in section I.C.2, the requirements for each subcategory of existing sources are severable from the requirements for each other subcategory of existing sources. For example, if a court were to invalidate the BSER and associated emission standard for units in the medium-term subcategory, the BSER and associated emission standard for units in the long-term subcategory could function sensibly because the effectiveness of the BSER for each subcategory is not dependent on the effectiveness of the BSER for other subcategories. Within the group of actions for new and reconstructed fossil fuel-fired combustion turbines in section I.C.3, the following actions are severable: the requirements for each subcategory of new and reconstructed turbines are severable from the requirements for each other subcategory; and within the subcategory for base load turbines, the requirements for each of the two components are severable from the requirements for the other component. Each of these standards can function sensibly without the others. For example, the BSER for low load, intermediate load, and base load subcategories is based on the technologies the EPA determined met the statutory standards for those subcategories and are independent from each other. And in the base load subcategory units may practically be constructed using the most efficient technology without then installing CCS and likewise may install CCS on a turbine system that was not constructed with the most efficient technology. Within the group of actions for new, modified, and reconstructed fossil fuel-fired steam generating units in section I.C.4, the revisions of the standards of performance for coal-fired steam generators that undertake a large modification are severable from the withdrawal of the 2018 proposal to revise the NSPS for emissions of GHG from EGUs. Each of the actions in these final rules that the EPA has identified as severable is functionally independent— i.e., may operate in practice independently of the other actions.

In addition, while the EPA is finalizing this rule at the same time as other final rules regulating different types of pollution from EGUs—specifically the Supplemental Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating Point Source Category (FR 2024-09815, EPA-HQ-OW-2009-0819; FRL-8794-02-OW); National Emission Standards for Hazardous Air Pollutants: Coal and Oil-Fired Electric Utility Steam Generating Units Review of the Residual Risk and Technology Review (FR 2024-09148, EPA-HQ-OAR-2018-0794; FRL-6716.3-02-OAR); Hazardous and Solid Waste Management System: Disposal of Coal Combustion Residuals From Electric Utilities; Legacy CCR Surface Impoundments (FR 2024-09157, EPA-HQ-OLEM-2020-0107; FRL-7814-04-OLEM)—and has considered the interactions between and cumulative effects of these rules, each rule is based on different statutory authority, a different record, and is completely independent of the other rules.

D. Grid Reliability Considerations

The EPA is finalizing multiple adjustments to the proposed rules that ensure the requirements in these final actions can be implemented without compromising the ability of power companies, grid operators, and state and Federal energy regulators to maintain resource adequacy and grid reliability. In response to the May 2023 proposed rule, the EPA received extensive comments from balancing authorities, independent system operators and regional transmission organizations, state regulators, power companies, and other stakeholders on the need for the final rule to accommodate resource adequacy and grid reliability needs. The EPA also engaged with the balancing authorities that submitted comments to the docket, the staff and Commissioners of the Federal Energy Regulatory Commission (FERC), the Department of Energy (DOE), the North American Electric Reliability Corporation (NERC), and other expert entities during the course of this rulemaking. Finally, at the invitation of FERC, the EPA participated in FERC's Annual Reliability Technical Conference on November 9, 2023.

These final actions respond to this input and feedback in multiple ways, including through changes to the universe of affected sources, longer compliance timeframes for CCS implementation, and other compliance flexibilities, as well as articulation of the appropriate use of RULOF to address reliability issues during state plan development and in subsequent state plan revisions. In addition to these adjustments, the EPA is finalizing several programmatic mechanisms specifically designed to address reliability concerns raised by commenters. For existing fossil fuel-fired EGUs, a short-term reliability emergency mechanism is available for states to provide more flexibility by using an alternative emission limitation during acute operational emergencies when the grid might be temporarily under heavy strain. A similar short-term reliability emergency mechanism is also available to new sources. In addition, the EPA is creating an option for states to provide for a compliance date extension for existing sources of up to 1 year under certain circumstances for sources that are installing control technologies to comply with their standards of performance. Lastly, states may also provide, by inclusion in their state plans, a reliability assurance mechanism of up to 1 year that under limited circumstances would allow existing units that had planned to cease operating by a certain date to temporarily remain available to support reliability. Any extensions exceeding 1 year must be addressed through a state plan revision. In order to utilize this reliability pathway, there must be an adequate demonstration of need and certification by a reliability authority, and approval by the appropriate EPA Regional Administrator. The EPA plans to seek the advice of FERC for extension requests exceeding 6 months. Similarly, for new fossil fuel-fired combustion turbines, the EPA is creating a mechanism whereby baseload units may request a 1-year extension of their CCS compliance deadline under certain circumstances.

The EPA has evaluated the resource adequacy implications of these actions in the final technical support document (TSD), Resource Adequacy Analysis, and conducted capacity expansion modeling of the final rules in a manner that takes into account resource adequacy needs. The EPA finds that resource adequacy can be maintained with the final rules. The EPA modeled a scenario that complies with the final rules and that meets resource adequacy needs. The EPA also performed a variety of other sensitivity analyses looking at higher electricity demand (load growth) and impact of the EPA's additional regulatory actions affecting the power sector. These sensitivity analyses indicate that, in the context of higher demand and other pending power sector rules, the industry has available pathways to comply with this rule that respect NERC reliability considerations and constraints.

In addition, the EPA notes that significant planning and regulatory mechanisms exist to ensure that sufficient generation resources are available to maintain reliability. The EPA's consideration of reliability in this rulemaking has also been informed by consultation with the DOE under the auspices of the March 9, 2023, memorandum of understanding (MOU) signed by the EPA Administrator and the Secretary of Energy, as well as by consultation with FERC expert staff. In these final actions, the EPA has included various flexibilities that allow power companies and grid operators to plan for achieving feasible and necessary reductions of GHGs from affected sources consistent with the EPA's statutory charge while ensuring that the rule will not interfere with systems operators' ability to ensure grid reliability.

A thorough description of how adjustments in the final rules address reliability issues, the EPA's outreach to balancing authorities, EPA's supplemental notice, as well as the introduction of mechanisms to address short- and long-term reliability needs is presented in section XII.F of this preamble.

E. Environmental Justice Considerations

Consistent with Executive Order (E.O.) 14096, and the EPA's commitment to upholding environmental justice (EJ) across its policies and programs, the EPA carefully considered the impacts of these actions on communities with environmental justice concerns. As part of the regulatory development process for these rulemakings, and consistent with directives set forth in multiple Executive Orders, the EPA conducted extensive outreach with interested parties including Tribal nations and communities with environmental justice concerns. These opportunities gave the EPA a chance to hear directly from the public, including from communities potentially impacted by these final actions. The EPA took this feedback into account in its development of these final actions. The EPA's analysis of environmental justice in these final actions is briefly summarized here and discussed in further detail in sections XII.E and XIII.J of the preamble and section 6 of the regulatory impact analysis (RIA).

Several environmental justice organizations and community representatives raised significant concerns about the potential health, environmental, and safety impacts of CCS. The EPA takes these concerns seriously, agrees that any impacts to historically disadvantaged and overburdened communities are important to consider, and has carefully considered these concerns as it finalized its determinations of the BSERs for these rules. The Agency acknowledges that while these final actions will result in large reductions of both GHGs and other emissions that will have significant positive benefits, there is the potential for localized increases in emissions, particularly if units installing CCS operate for more hours during the year and/or for more years than they would have otherwise. However, as discussed in section VII.C.1.a.iii(B), a robust regulatory framework exists to reduce the risks of localized emissions increases in a manner that is protective of public health, safety, and the environment. The Council on Environmental Quality's (CEQ) February 2022 Carbon Capture, Utilization, and Sequestration Guidance and the EPA's evaluation of BSER recognize that multiple Federal agencies have responsibility for regulating and permitting CCS projects, along with state and tribal governments. As the CEQ has noted, Federal agencies have “taken actions in the past decade to develop a robust carbon capture, utilization, and sequestration/storage (CCUS) regulatory framework to protect the environment and public health across multiple statutes.” Furthermore, the EPA plans to review and update as needed its guidance on NSR permitting, specifically with respect to BACT determinations for GHG emissions and consideration of co-pollutant increases from sources installing CCS. For the reasons explained in section VII.C, the EPA is finalizing the determination that CCS is the BSER for certain subcategories of new and existing EGUs based on its consideration of all of the statutory criteria for BSER, including emission reductions, cost, energy requirements, and non-air health and environmental considerations. At the same time, the EPA recognizes the critical importance of ensuring that the regulatory framework performs as intended to protect communities.

These actions are focused on establishing NSPS and emission guidelines for GHGs that states will implement to significantly reduce GHGs and move us a step closer to avoiding the worst impacts of climate change, which is already having a disproportionate impact on communities with environmental justice concerns. The EPA analyzed several illustrative scenarios representing potential compliance outcomes and evaluated the potential impacts that these actions may have on emissions of GHG and other health-harming air pollutants from fossil fuel-fired EGUs, as well as how these changes in emissions might affect air quality and public health, particularly for communities with EJ concerns.

The EPA's national-level analysis of emission reduction and public health impacts, which is documented in section 6 of the RIA and summarized in greater detail in section XII.A and XII.D of this preamble, finds that these actions achieve nationwide reductions in EGU emissions of multiple health-harming air pollutants including nitrogen oxides (NO_X), sulfur dioxide (SO₂), and fine particulate matter (PM_2.5), resulting in public health benefits. The EPA also evaluated how the air quality impacts associated with these final actions are distributed, with particular focus on communities with EJ concerns. As discussed in the RIA, our analysis indicates that baseline ozone and PM_2.5 concentration will decline substantially relative to today's levels. Relative to these low baseline levels, ozone and PM_2.5 concentrations will decrease further in virtually all areas of the country, although some areas of the country may experience slower or faster rates of decline in ozone and PM_2.5 pollution over time due to the changes in generation and utilization resulting from these rules. Additionally, our comparison of future air quality conditions with and without these rules suggests that while these actions are anticipated to lead to modest but widespread reductions in ambient levels of PM_2.5 and ozone for a large majority of the nation's population, there is potential for some geographic areas and demographic groups to experience small increases in ozone concentrations relative to the baseline levels which are projected to be substantially lower than today's levels.

It is important to recognize that while these projections of emissions changes and resulting air quality changes under various illustrative compliance scenarios are based upon the best information available to the EPA at this time, with regard to existing sources, each state will ultimately be responsible for determining the future operation of fossil fuel-fired steam generating units located within its jurisdiction. The EPA expects that, in making these determinations, states will consider a number of factors and weigh input from the wide range of potentially affected stakeholders. The meaningful engagement requirements discussed in section X.E.1.b.i of this preamble will ensure that all interested stakeholders—including community members adversely impacted by pollution, energy workers affected by construction and/or other changes in operation at fossil-fuel-fired power plants, consumers and other interested parties—will have an opportunity to have their concerns heard as states make decisions balancing a multitude of factors including appropriate standards of performance, compliance strategies, and compliance flexibilities for existing EGUs, as well as public health and environmental considerations. The EPA believes that these provisions, together with the protections referenced above, can reduce the risks of localized emissions increases in a manner that is protective of public health, safety, and the environment.

F. Energy Workers and Communities

These final actions include requirements for meaningful engagement in development of state plans, including with energy workers and communities. These communities, including energy workers employed at affected EGUs, workers who may construct and install pollution control technology, workers employed by fuel extraction and delivery, organizations representing these workers, and communities living near affected EGUs, are impacted by power sector trends on an ongoing basis and by these final actions, and the EPA expects that states will include these stakeholders as part of their constructive engagement under the requirements in this rule.

The EPA consulted with the Federal Interagency Working Group on Coal and Power Plant Communities and Economic Revitalization (Energy Communities IWG) in development of these rules and the meaningful engagement requirements. The EPA notes that the Energy Communities IWG has provided resources to help energy communities access the expanded federal resources made available by the Bipartisan Infrastructure Law, CHIPS and Science Act, and Inflation Reduction Act, many of which are relevant to the development of state plans.

G. Key Changes From Proposal

The key changes from proposal in these final actions are: (1) the reduction in number of subcategories for existing coal-fired steam generating units, (2) the extension of the compliance date for existing coal-fired steam generating units to meet a standard of performance based on implementation of CCS, (3) the removal of low-GHG hydrogen co-firing as a BSER pathway, and (4) the addition of two reliability-related instruments. In addition, (5), the EPA is not finalizing proposed requirements for existing fossil fuel-fired stationary combustion turbines at this time.

The reduction in number of subcategories for existing coal-fired steam generating units: The EPA proposed four subcategories for existing coal-fired steam generating units, which would have distinguished these units by operating horizon and by load level. These included subcategories for existing coal-fired EGUs planning to cease operations in the imminent-term ( i.e., prior to January 1, 2032) and those planning to cease operations in the near-term ( i.e., prior to January 1, 2035). While commenters were generally supportive of the proposed subcategorization approach, some requested that the cease-operation-by date for the imminent-term subcategory be extended and the utilization limit for the near-term subcategory be relaxed. The EPA is not finalizing the imminent-term and near-term subcategories of coal-fired steam generating units. Rather, the EPA is finalizing an applicability exemption for coal-fired steam generating units demonstrating that they plan to permanently cease operation before January 1, 2032. See section VII.B of this preamble for further discussion.

The extension of the compliance date for existing coal-fired steam generating units to meet a standard of performance based on implementation of CCS. The EPA proposed a compliance date for implementation of CCS for long-term coal-fired steam generating units of January 1, 2030. The EPA received comments asserting that this deadline did not provide adequate lead time. In consideration of those comments, and the record as a whole, the EPA is finalizing a CCS compliance date of January 1, 2032 for these sources.

The removal of low-GHG hydrogen co-firing as a BSER pathway and only use of low-GHG hydrogen as a compliance option: The EPA is not finalizing its proposed BSER pathway of low-GHG hydrogen co-firing for new and reconstructed base load and intermediate load combustion turbines in accordance with CAA section 111(a)(1). The EPA is also not finalizing its proposed requirement that only low-GHG hydrogen may be co-fired in a combustion turbine for the purpose of compliance with the standards of performance. These decisions are based on uncertainties identified for specific criteria used to evaluate low-GHG hydrogen co-firing as a potential BSER, and after further analysis in response to public comments, the EPA has determined that these uncertainties prevent the EPA from concluding that low-GHG hydrogen co-firing is a component of the “best” system of emission reduction at this time. Under CAA section 111, the EPA establishes standards of performance but does not mandate use of any particular technology to meet those standards. Therefore, certain sources may elect to co-fire hydrogen for compliance with the final standards of performance, even absent the technology being a BSER pathway. See section VIII.F.5 of this preamble for further discussion.

The addition of two reliability-related instruments: Commenters expressed concerns that these rules, in combination with other factors, may affect the reliability of the bulk power system. In response to these comments the EPA engaged extensively with balancing authorities, power companies, reliability experts, and regulatory authorities responsible for reliability to inform its decisions in these final rules. As described later in this preamble, the EPA has made adjustments in these final rules that will support power companies, grid operators, and states in maintaining the reliability of the electric grid during the implementation of these final rules. In addition, the EPA has undertaken an analysis of the reliability and resource adequacy implications of these final rules that supports the Agency's conclusion that these final rules can be implemented without adverse consequences for grid reliability. Further, the EPA is finalizing two reliability-related instruments as an additional layer of safeguards for reliability. These instruments include a reliability mechanism for short-term emergency issues, and a reliability assurance mechanism, or compliance flexibility, for units that have chosen compliance pathways with enforceable retirement dates, provided there is a documented and verified reliability concern. In addition, the EPA is finalizing compliance extensions for unanticipated delays with control technology implementation. Specifically, as described in greater detail in section XII.F of this preamble, the EPA is finalizing the following features and changes from the proposal that will provide even greater certainty that these final rules are sensitive to reliability-related issues and constructed in a manner that does not interfere with grid operators' responsibility to deliver reliable power:

(1) longer compliance timelines for existing coal-fired steam generating units;

(2) a mechanism to extend compliance timelines by up to 1 year in the case of unforeseen circumstances, outside of an owner/operator's control, that delay the ability to apply controls ( e.g., supply chain challenges or permitting delays);

(3) transparent unit-specific compliance information for EGUs that will allow grid operators to plan for system changes with greater certainty and precision;

(4) a short-term reliability mechanism to allow affected EGUs to operate at baseline emission rates during documented reliability emergencies; and

(5) a reliability assurance mechanism to allow states to delay cease operation dates by up to 1 year in cases where the planned cease operation date is forecast to disrupt system reliability.

Not finalizing proposed requirements for existing fossil fuel-fired stationary combustion turbines at this time: The EPA proposed emission guidelines for large ( i.e., greater than 300 MW), frequently operated ( i.e., with an annual capacity factor of greater than 50 percent), existing fossil fuel-fired stationary combustion turbines. The EPA received a wide range of comments on the proposed guidelines. Multiple commenters suggested that the proposed provisions would largely result in shifting of generation away from the most efficient natural gas-fired turbines to less efficient natural gas-fired turbines. Commenters stated that, as emissions from coal-fired steam generating units decreased, existing natural gas-fired EGUs were poised to become the largest source of GHG emissions in the power sector. Commenters noted that these units play an important role in grid reliability, particularly as aging coal-fired EGUs retire. Commenters further noted that the existing fossil fuel-fired stationary combustion turbines that were not covered by the proposal ( i.e., the smaller and less frequently operating units) are often less efficient, less well controlled for other pollutants such as NO_X, and are more likely to be located near population centers and communities with environmental justice concerns.

The EPA agrees with commenters who observed that GHG emissions from existing natural gas-fired stationary combustion turbines are a growing portion of the emissions from the power sector. This is consistent with EPA modeling that shows that by 2030 these units will represent the largest portion of GHG emissions from the power sector. The EPA agrees that it is vital to promulgate emission guidelines to address GHG emissions from these sources, and that the EPA has a responsibility to do so under section 111(d) of the Clean Air Act. The EPA also agrees with commenters who noted that focusing only on the largest and most frequently operating units, without also addressing emissions from other units, as the May 2023 proposed rule provided, may not be the most effective way to address emissions from this sector. The EPA's modeling shows that over time as the power sector comes closer to reaching the phase-out threshold of the clean electricity incentives in the Inflation Reduction Act (IRA) ( i.e., a 75 percent reduction in emissions from the power sector from 2022 levels), the average capacity factor for existing natural gas-fired stationary combustion turbines decreases. Therefore, the EPA's proposal to focus only on the largest units with the highest capacity factors may not be the most effective policy design for reducing GHG emissions from these sources.

Recognizing the importance of reducing emissions from all fossil fuel-fired EGUs, the EPA is not finalizing the proposed emission guidelines for certain existing fossil fuel-fired stationary combustion turbines at this time. Instead, the EPA intends to issue a new, more comprehensive proposal to regulate GHGs from existing sources. The new proposal will focus on achieving greater emission reductions from existing stationary combustion turbines—which will soon be the largest stationary sources of GHG emissions—while taking into account other factors including the local non-GHG impacts of gas turbine generation and the need for reliable, affordable electricity.

II. General Information

A. Action Applicability

The source category that is the subject of these actions is composed of fossil fuel-fired electric utility generating units. The North American Industry Classification System (NAICS) codes for the source category are 221112 and 921150. The list of categories and NAICS codes is not intended to be exhaustive, but rather provides a guide for readers regarding the entities that these final actions are likely to affect.

Final amendments to 40 CFR part 60, subpart TTTT, are directly applicable to affected facilities that began construction after January 8, 2014, but before May 23, 2023, and affected facilities that began reconstruction or modification after June 18, 2014, but before May 23, 2023. The NSPS codified in 40 CFR part 60, subpart TTTTa, is directly applicable to affected facilities that begin construction, reconstruction, or modification on or after May 23, 2023. Federal, state, local, and tribal government entities that own and/or operate EGUs subject to 40 CFR part 60, subpart TTTT or TTTTa, are affected by these amendments and standards.

The emission guidelines codified in 40 CFR part 60, subpart UUUUb, are for states to follow in developing, submitting, and implementing state plans to establish performance standards to reduce emissions of GHGs from designated facilities that are existing sources. Section 111(a)(6) of the CAA defines an “existing source” as “any stationary source other than a new source.” Therefore, the emission guidelines would not apply to any EGUs that are new after January 8, 2014, or reconstructed after June 18, 2014, the applicability dates of 40 CFR part 60, subpart TTTT. Under the Tribal Authority Rule (TAR), eligible tribes may seek approval to implement a plan under CAA section 111(d) in a manner similar to a state. See 40 CFR part 49, subpart A. Tribes may, but are not required to, seek approval for treatment in a manner similar to a state for purposes of developing a tribal implementation plan (TIP) implementing the emission guidelines codified in 40 CFR part 60, subpart UUUUb. The TAR authorizes tribes to develop and implement their own air quality programs, or portions thereof, under the CAA. However, it does not require tribes to develop a CAA program. Tribes may implement programs that are most relevant to their air quality needs. If a tribe does not seek and obtain the authority from the EPA to establish a TIP, the EPA has the authority to establish a Federal CAA section 111(d) plan for designated facilities that are located in areas of Indian country. A Federal plan would apply to all designated facilities located in the areas of Indian country covered by the Federal plan unless and until the EPA approves a TIP applicable to those facilities.

B. Where To Get a Copy of This Document and Other Related Information

In addition to being available in the docket, an electronic copy of these final rulemakings is available on the internet at https://www.epa.gov/stationary-sources-air-pollution/greenhouse-gas-standards-and-guidelines-fossil-fuel-fired-power . Following signature by the EPA Administrator, the EPA will post a copy of these final rulemakings at this same website. Following publication in the Federal Register , the EPA will post the Federal Register version of the final rules and key technical documents at this same website.

C. Judicial Review and Administrative Review

Under CAA section 307(b)(1), judicial review of these final actions is available only by filing a petition for review in the United States Court of Appeals for the District of Columbia Circuit by July 8, 2024. These final actions are “standard[s] of performance or requirement[s] under section 111,” and, in addition, are “nationally applicable regulations promulgated, or final action taken, by the Administrator under [the CAA],” CAA section 307(b)(1). Under CAA section 307(b)(2), the requirements established by this final rule may not be challenged separately in any civil or criminal proceedings brought by the EPA to enforce the requirements.

Section 307(d)(7)(B) of the CAA further provides that “[o]nly an objection to a rule or procedure which was raised with reasonable specificity during the period for public comment (including any public hearing) may be raised during judicial review.” This section also provides a mechanism for the EPA to convene a proceeding for reconsideration, “[i]f the person raising an objection can demonstrate to the EPA that it was impracticable to raise such objection within [the period for public comment] or if the grounds for such objection arose after the period for public comment, (but within the time specified for judicial review) and if such objection is of central relevance to the outcome of the rule.” Any person seeking to make such a demonstration to us should submit a Petition for Reconsideration to the Office of the Administrator, U.S. Environmental Protection Agency, Room 3000, WJC West Building, 1200 Pennsylvania Ave. NW, Washington, DC 20460, with a copy to both the person(s) listed in the preceding FOR FURTHER INFORMATION CONTACT section, and the Associate General Counsel for the Air and Radiation Law Office, Office of General Counsel (Mail Code 2344A), U.S. Environmental Protection Agency, 1200 Pennsylvania Ave. NW, Washington, DC 20460.

III. Climate Change Impacts

Elevated concentrations of GHGs have been warming the planet, leading to changes in the Earth's climate that are occurring at a pace and in a way that threatens human health, society, and the natural environment. While the EPA is not making any new scientific or factual findings with regard to the well-documented impact of GHG emissions on public health and welfare in support of these rules, the EPA is providing in this section a brief scientific background on climate change to offer additional context for these rulemakings and to help the public understand the environmental impacts of GHGs.

Extensive information on climate change is available in the scientific assessments and the EPA documents that are briefly described in this section, as well as in the technical and scientific information supporting them. One of those documents is the EPA's 2009 “Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the CAA” (74 FR 66496, December 15, 2009) (“2009 Endangerment Finding”). In the 2009 Endangerment Finding, the Administrator found under section 202(a) of the CAA that elevated atmospheric concentrations of six key well-mixed GHGs—CO₂, methane (CH₄), nitrous oxide (N₂ O), HFCs, perfluorocarbons (PFCs), and sulfur hexafluoride (SF₆)—“may reasonably be anticipated to endanger the public health and welfare of current and future generations” (74 FR 66523, December 15, 2009). The 2009 Endangerment Finding, together with the extensive scientific and technical evidence in the supporting record, documented that climate change caused by human emissions of GHGs threatens the public health of the U.S. population. It explained that by raising average temperatures, climate change increases the likelihood of heat waves, which are associated with increased deaths and illnesses (74 FR 66497, December 15, 2009). While climate change also increases the likelihood of reductions in cold-related mortality, evidence indicates that the increases in heat mortality will be larger than the decreases in cold mortality in the U.S. (74 FR 66525, December 15, 2009). The 2009 Endangerment Finding further explained that compared with a future without climate change, climate change is expected to increase tropospheric ozone pollution over broad areas of the U.S., including in the largest metropolitan areas with the worst tropospheric ozone problems, and thereby increase the risk of adverse effects on public health (74 FR 66525, December 15, 2009). Climate change is also expected to cause more intense hurricanes and more frequent and intense storms of other types and heavy precipitation, with impacts on other areas of public health, such as the potential for increased deaths, injuries, infectious and waterborne diseases, and stress-related disorders (74 FR 66525 December 15, 2009). Children, the elderly, and the poor are among the most vulnerable to these climate-related health effects (74 FR 66498, December 15, 2009).

The 2009 Endangerment Finding also documented, together with the extensive scientific and technical evidence in the supporting record, that climate change touches nearly every aspect of public welfare in the U.S., including the following: changes in water supply and quality due to changes in drought and extreme rainfall events; increased risk of storm surge and flooding in coastal areas and land loss due to inundation; increases in peak electricity demand and risks to electricity infrastructure; and the potential for significant agricultural disruptions and crop failures (though offset to some extent by carbon fertilization). These impacts are also global and may exacerbate problems outside the U.S. that raise humanitarian, trade, and national security issues for the U.S. (74 FR 66530, December 15, 2009).

In 2016, the Administrator issued a similar finding for GHG emissions from aircraft under section 231(a)(2)(A) of the CAA. In the 2016 Endangerment Finding, the Administrator found that the body of scientific evidence amassed in the record for the 2009 Endangerment Finding compellingly supported a similar endangerment finding under CAA section 231(a)(2)(A) and also found that the science assessments released between the 2009 and 2016 Findings “strengthen and further support the judgment that GHGs in the atmosphere may reasonably be anticipated to endanger the public health and welfare of current and future generations” (81 FR 54424, August 15, 2016).

Since the 2016 Endangerment Finding, the climate has continued to change, with new observational records being set for several climate indicators such as global average surface temperatures, GHG concentrations, and sea level rise. Additionally, major scientific assessments continue to be released that further advance our understanding of the climate system and the impacts that GHGs have on public health and welfare for both current and future generations. These updated observations and projections document the rapid rate of current and future climate change both globally and in the U.S.

The most recent information demonstrates that the climate is continuing to change in response to the human-induced buildup of GHGs in the atmosphere. These recent assessments show that atmospheric concentrations of GHGs have risen to a level that has no precedent in human history and that they continue to climb, primarily because of both historical and current anthropogenic emissions, and that these elevated concentrations endanger our health by affecting our food and water sources, the air we breathe, the weather we experience, and our interactions with the natural and built environments. For example, atmospheric concentrations of one of these GHGs, CO₂ , measured at Mauna Loa in Hawaii and at other sites around the world reached 419 parts per million (ppm) in 2022 (nearly 50 percent higher than preindustrial levels) and have continued to rise at a rapid rate. Global average temperature has increased by about 1.1 °C (2.0 °F) in the 2011-2020 decade relative to 1850-1900. The years 2015-2021 were the warmest 7 years in the 1880-2021 record, contributing to the warmest decade on record with a decadal temperature of 0.82 °C (1.48 °F) above the 20th century. The Intergovernmental Panel on Climate Change (IPCC) determined (with medium confidence) that this past decade was warmer than any multi-century period in at least the past 100,000 years. Global average sea level has risen by about 8 inches (about 21 centimeters (cm)) from 1901 to 2018, with the rate from 2006 to 2018 (0.15 inches/year or 3.7 millimeters (mm)/year) almost twice the rate over the 1971 to 2006 period, and three times the rate of the 1901 to 2018 period. The rate of sea level rise over the 20th century was higher than in any other century in at least the last 2,800 years. Higher CO₂ concentrations have led to acidification of the surface ocean in recent decades to an extent unusual in the past 65 million years, with negative impacts on marine organisms that use calcium carbonate to build shells or skeletons. Arctic sea ice extent continues to decline in all months of the year; the most rapid reductions occur in September (very likely almost a 13 percent decrease per decade between 1979 and 2018) and are unprecedented in at least 1,000 years. Human-induced climate change has led to heatwaves and heavy precipitation becoming more frequent and more intense, along with increases in agricultural and ecological droughts in many regions.

The assessment literature demonstrates that modest additional amounts of warming may lead to a climate different from anything humans have ever experienced. The 2022 CO₂ concentration of 419 ppm is already higher than at any time in the last 2 million years. If concentrations exceed 450 ppm, they would likely be higher than any time in the past 23 million years: at the current rate of increase of more than 2 ppm per year, this would occur in about 15 years. While GHGs are not the only factor that controls climate, it is illustrative that 3 million years ago (the last time CO₂ concentrations were above 400 ppm) Greenland was not yet completely covered by ice and still supported forests, while 23 million years ago (the last time concentrations were above 450 ppm) the West Antarctic ice sheet was not yet developed, indicating the possibility that high GHG concentrations could lead to a world that looks very different from today and from the conditions in which human civilization has developed. If the Greenland and Antarctic ice sheets were to melt substantially, sea levels would rise dramatically.

The NCA4 found that it is very likely (greater than 90 percent likelihood) that by mid-century, the Arctic Ocean will be almost entirely free of sea ice by late summer for the first time in about 2 million years. Coral reefs will be at risk for almost complete (99 percent) losses with 1 °C (1.8 °F) of additional warming from today (2 °C or 3.6 °F since preindustrial). At this temperature, between 8 and 18 percent of animal, plant, and insect species could lose over half of the geographic area with suitable climate for their survival, and 7 to 10 percent of rangeland livestock would be projected to be lost. The IPCC similarly found that climate change has caused substantial damages and increasingly irreversible losses in terrestrial, freshwater, and coastal and open ocean marine ecosystems.

Every additional increment of temperature comes with consequences. For example, the half degree of warming from 1.5 to 2 °C (0.9 °F of warming from 2.7 °F to 3.6 °F) above preindustrial temperatures is projected on a global scale to expose 420 million more people to frequent extreme heatwaves at least every five years, and 62 million more people to frequent exceptional heatwaves at least every five years (where heatwaves are defined based on a heat wave magnitude index which takes into account duration and intensity—using this index, the 2003 French heat wave that led to almost 15,000 deaths would be classified as an “extreme heatwave” and the 2010 Russian heatwave which led to thousands of deaths and extensive wildfires would be classified as “exceptional”). It would increase the frequency of sea-ice-free Arctic summers from once in 100 years to once in a decade. It could lead to 4 inches of additional sea level rise by the end of the century, exposing an additional 10 million people to risks of inundation as well as increasing the probability of triggering instabilities in either the Greenland or Antarctic ice sheets. Between half a million and a million additional square miles of permafrost would thaw over several centuries. Risks to food security would increase from medium to high for several lower-income regions in the Sahel, southern Africa, the Mediterranean, central Europe, and the Amazon. In addition to food security issues, this temperature increase would have implications for human health in terms of increasing ozone concentrations, heatwaves, and vector-borne diseases (for example, expanding the range of the mosquitoes which carry dengue fever, chikungunya, yellow fever, and the Zika virus or the ticks which carry Lyme, babesiosis, or Rocky Mountain Spotted Fever). Moreover, every additional increment in warming leads to larger changes in extremes, including the potential for events unprecedented in the observational record. Every additional degree will intensify extreme precipitation events by about 7 percent. The peak winds of the most intense tropical cyclones (hurricanes) are projected to increase with warming. In addition to a higher intensity, the IPCC found that precipitation and frequency of rapid intensification of these storms has already increased, the movement speed has decreased, and elevated sea levels have increased coastal flooding, all of which make these tropical cyclones more damaging.

The NCA4 also evaluated a number of impacts specific to the U.S. Severe drought and outbreaks of insects like the mountain pine beetle have killed hundreds of millions of trees in the western U.S. Wildfires have burned more than 3.7 million acres in 14 of the 17 years between 2000 and 2016, and Federal wildfire suppression costs were about a billion dollars annually. The National Interagency Fire Center has documented U.S. wildfires since 1983, and the 10 years with the largest acreage burned have all occurred since 2004. Wildfire smoke degrades air quality, increasing health risks, and more frequent and severe wildfires due to climate change would further diminish air quality, increase incidences of respiratory illness, impair visibility, and disrupt outdoor activities, sometimes thousands of miles from the location of the fire. Meanwhile, sea level rise has amplified coastal flooding and erosion impacts, requiring the installation of costly pump stations, flooding streets, and increasing storm surge damages. Tens of billions of dollars of U.S. real estate could be below sea level by 2050 under some scenarios. Increased frequency and duration of drought will reduce agricultural productivity in some regions, accelerate depletion of water supplies for irrigation, and expand the distribution and incidence of pests and diseases for crops and livestock. The NCA4 also recognized that climate change can increase risks to national security, both through direct impacts on military infrastructure and by affecting factors such as food and water availability that can exacerbate conflict outside U.S. borders. Droughts, floods, storm surges, wildfires, and other extreme events stress nations and people through loss of life, displacement of populations, and impacts on livelihoods. The NCA5 further reinforces the science showing that climate change will have many impacts on the U.S., as described above in the preamble. Particularly relevant for these rules, the NCA5 states that climate change affects all aspects of the energy system-supply, delivery, and demand-through the increased frequency, intensity, and duration of extreme events and through changing climate trends.”

EPA modeling efforts can further illustrate how these impacts from climate change may be experienced across the U.S. EPA's Framework for Evaluating Damages and Impacts (FrEDI) uses information from over 30 peer-reviewed climate change impact studies to project the physical and economic impacts of climate change to the U.S. resulting from future temperature changes. These impacts are projected for specific regions within the U.S. and for more than 20 impact categories, which span a large number of sectors of the U.S. economy. Using this framework, the EPA estimates that global emission projections, with no additional mitigation, will result in significant climate-related damages to the U.S. These damages to the U.S. would mainly be from increases in lives lost due to increases in temperatures, as well as impacts to human health from increases in climate-driven changes in air quality, dust and wildfire smoke exposure, and incidence of suicide. Additional major climate-related damages would occur to U.S. infrastructure such as roads and rail, as well as transportation impacts and coastal flooding from sea level rise, increases in property damage from tropical cyclones, and reductions in labor hours worked in outdoor settings and buildings without air conditioning. These impacts are also projected to vary from region to region with the Southeast, for example, projected to see some of the largest damages from sea level rise, the West Coast projected to experience damages from wildfire smoke more than other parts of the country, and the Northern Plains states projected to see a higher proportion of damages to rail and road infrastructure. While information on the distribution of climate impacts helps to better understand the ways in which climate change may impact the U.S., recent analyses are still only a partial assessment of climate impacts relevant to U.S. interests and in addition do not reflect increased damages that occur due to interactions between different sectors impacted by climate change or all the ways in which physical impacts of climate change occurring abroad have spillover effects in different regions of the U.S.

Some GHGs also have impacts beyond those mediated through climate change. For example, elevated concentrations of CO₂ stimulate plant growth (which can be positive in the case of beneficial species, but negative in terms of weeds and invasive species, and can also lead to a reduction in plant micronutrients ) and cause ocean acidification. Nitrous oxide depletes the levels of protective stratospheric ozone. Methane reacts to form tropospheric ozone.

Section XII.E of this preamble discusses the impacts of GHG emissions on individuals living in socially and economically vulnerable communities. While the EPA did not conduct modeling to specifically quantify changes in climate impacts resulting from these rules in terms of avoided temperature change or sea-level rise, the Agency did quantify climate benefits by monetizing the emission reductions through the application of the social cost of greenhouse gases (SC-GHGs), as described in section XII.D of this preamble.

These scientific assessments, the EPA analyses, and documented observed changes in the climate of the planet and of the U.S. present clear support regarding the current and future dangers of climate change and the importance of GHG emissions mitigation.

IV. Recent Developments in Emissions Controls and the Electric Power Sector

In this section, we discuss background information about the electric power sector and controls available to limit GHG pollution from the fossil fuel-fired power plants regulated by these final rules, and then discuss several recent developments that are relevant for determining the BSER for these sources. After giving some general background, we first discuss CCS and explain that its costs have fallen significantly. Lower costs are central for the EPA's determination that CCS is the BSER for certain existing coal-fired steam generating units and certain new natural gas-fired combustion turbines. Second, we discuss natural gas co-firing for coal-fired steam generating units and explain recent reductions in cost for this approach as well as its widespread availability and current and potential deployment within this subcategory. Third, we discuss highly efficient generation as a BSER technology for new and reconstructed simple cycle and combined cycle combustion turbine EGUs. The emission reductions achieved by highly efficient turbines are well demonstrated in the power sector, and along with operational and maintenance best practices, represent a cost-effective technology that reduces fuel consumption. Finally, we discuss key developments in the electric power sector that influence which units can feasibly and cost-effectively deploy these technologies.

A. Background

1. Electric Power Sector

Electricity in the U.S. is generated by a range of technologies, and different EGUs play different roles in providing reliable and affordable electricity. For example, certain EGUs generate base load power, which is the portion of electricity loads that are continually present and typically operate throughout all hours of the year. Intermediate EGUs often provide complementary generation to balance variable supply and demand resources. Low load “peaking units” provide capacity during hours of the highest daily, weekly, or seasonal net demand, and while these resources have low levels of utilization on an annual basis, they play important roles in providing generation to meet short-term demand and often must be available to quickly increase or decrease their output. Furthermore, many of these EGUs also play important roles ensuring the reliability of the electric grid, including facilitating the regulation of frequency and voltage, providing “black start” capability in the event the grid must be repowered after a widespread outage, and providing reserve generating capacity in the event of unexpected changes in the availability of other generators.

In general, the EGUs with the lowest operating costs are dispatched first, and, as a result, an inefficient EGU with high fuel costs will typically only operate if other lower-cost plants are unavailable or are insufficient to meet demand. Units are also unavailable during both routine and unanticipated outages, which typically become more frequent as power plants age. These factors result in the mix of available generating capacity types ( e.g., the share of capacity of each type of generating source) being substantially different than the mix of the share of total electricity produced by each type of generating source in a given season or year.

Generated electricity must be transmitted over networks of high voltage lines to substations where power is stepped down to a lower voltage for local distribution. Within each of these transmission networks, there are multiple areas where the operation of power plants is monitored and controlled by regional organizations to ensure that electricity generation and load are kept in balance. In some areas, the operation of the transmission system is under the control of a single regional operator; in others, individual utilities coordinate the operations of their generation and transmission to balance the system across their respective service territories.

2. Types of EGUs

There are many types of EGUs including fossil fuel-fired power plants ( i.e., those using coal, oil, and natural gas), nuclear power plants, renewable generating sources (such as wind and solar) and others. This rule focuses on the fossil fuel-fired portion of the generating fleet that is responsible for the vast majority of GHG emissions from the power sector. The definition of fossil fuel-fired electric utility steam generating units includes utility boilers as well as those that use gasification technology ( i.e., integrated gasification combined cycle (IGCC) units). While coal is the most common fuel for fossil fuel-fired utility boilers, natural gas can also be used as a fuel in these EGUs and many existing coal- and oil-fired utility boilers have refueled as natural gas-fired utility boilers. An IGCC unit gasifies fuel—typically coal or petroleum coke—to form a synthetic gas (or syngas) composed of carbon monoxide (CO) and hydrogen (H₂), which can be combusted in a combined cycle system to generate power. The heat created by these technologies produces high-pressure steam that is released to rotate turbines, which, in turn, spin an electric generator.

Stationary combustion turbine EGUs (most commonly natural gas-fired) use one of two configurations: combined cycle or simple cycle turbines. Combined cycle units have two generating components ( i.e., two cycles) operating from a single source of heat. Combined cycle units first generate power from a combustion turbine ( i.e., the combustion cycle) directly from the heat of burning natural gas or other fuel. The second cycle reuses the waste heat from the combustion turbine engine, which is routed to a heat recovery steam generator (HRSG) that generates steam, which is then used to produce additional power using a steam turbine ( i.e., the steam cycle). Combining these generation cycles increases the overall efficiency of the system. Combined cycle units that fire mostly natural gas are commonly referred to as natural gas combined cycle (NGCC) units, and, with greater efficiency, are utilized at higher capacity factors to provide base load or intermediate load power. An EGU's capacity factor indicates a power plant's electricity output as a percentage of its total generation capacity. Simple cycle turbines only use a combustion turbine to produce electricity ( i.e., there is no heat recovery or steam cycle). These less-efficient combustion turbines are generally utilized at non-base load capacity factors and contribute to reliable operations of the grid during periods of peak demand or provide flexibility to support increased generation from variable energy sources.

Other generating sources produce electricity by harnessing kinetic energy from flowing water, wind, or tides, thermal energy from geothermal wells, or solar energy primarily through photovoltaic solar arrays. Spurred by a combination of declining costs, consumer preferences, and government policies, the capacity of these renewable technologies is growing, and when considered with existing nuclear energy, accounted for 40 percent of the overall net electricity supply in 2022. Many projections show this share growing over time. For example, the EPA's Power Sector Platform 2023 using IPM ( i.e., the EPA's baseline projections of the power sector) projects zero-emitting sources reaching 76 percent of electricity generation by 2040. This shift is driven by multiple factors. These factors include changes in the relative economics of generating technologies, the efforts by states to reduce GHG emissions, utility and other corporate commitments, and customer preference. The shift is further promoted by provisions of Federal legislation, most notably the Clean Electricity Investment and Production tax credits included in IRC sections 48E and 45Y of the IRA, which do not begin to phase out until the later of 2032 or when power sector GHG emissions are 75 percent less than 2022 levels. (See section IV.F of this preamble and the accompanying RIA for additional discussion of projections for the power sector.) These projections are consistent with power company announcements. For example, as the Edison Electric Institute (EEI) stated in pre-proposal public comments submitted to the regulatory docket: “Fifty EEI members have announced forward-looking carbon reduction goals, two-thirds of which include a net-zero by 2050 or earlier equivalent goal, and members are routinely increasing the ambition or speed of their goals or altogether transforming them into net-zero goals . . . . EEI's member companies see a clear path to continued emissions reductions over the next decade using current technologies, including nuclear power, natural gas-based generation, energy demand efficiency, energy storage, and deployment of new renewable energy—especially wind and solar—as older coal-based and less-efficient natural gas-based generating units retire.” The Energy Strategy Coalition similarly said in public comments that “[a]s major electrical utilities and power producers, our top priority is providing clean, affordable, and reliable energy to our customers” and are “seeking to advance” technologies “such as a carbon capture and storage, which can significantly reduce carbon dioxide emissions from fossil fuel-fired EGUs.”

B. GHG Emissions From Fossil Fuel-Fired EGUs

The principal GHGs that accumulate in the Earth's atmosphere above pre-industrial levels because of human activity are CO₂, CH₄, N₂ O, HFCs, PFCs, and SF₆ . Of these, CO₂ is the most abundant, accounting for 80 percent of all GHGs present in the atmosphere. This abundance of CO₂ is largely due to the combustion of fossil fuels by the transportation, electricity, and industrial sectors.

The amount of CO₂ produced when a fossil fuel is burned in an EGU is a function of the carbon content of the fuel relative to the size and efficiency of the EGU. Different fuels emit different amounts of CO₂ in relation to the energy they produce when combusted. The heat content, or the amount of energy produced when a fuel is burned, is mainly determined by the carbon and hydrogen content of the fuel. For example, in terms of pounds of CO₂ emitted per million British thermal units of energy produced when combusted, natural gas is the lowest compared to other fossil fuels at 117 lb CO₂ /MMBtu. The average for coal is 216 lb CO₂ /MMBtu, but varies between 206 to 229 lb CO₂ /MMBtu by type ( e.g., anthracite, lignite, subbituminous, and bituminous). The value for petroleum products such as diesel fuel and heating oil is 161 lb CO₂ /MMBtu.

The EPA prepares the official U.S. Inventory of Greenhouse Gas Emissions and Sinks (the U.S. GHG Inventory) to comply with commitments under the United Nations Framework Convention on Climate Change (UNFCCC). This inventory, which includes recent trends, is organized by industrial sectors. It presents total U.S. anthropogenic emissions and sinks of GHGs, including CO₂ emissions since 1990. According to the latest inventory of all sectors, in 2021, total U.S. GHG emissions were 6,340 million metric tons of CO₂ equivalent (MMT CO₂ e). The transportation sector (28.5 percent), which includes approximately 300 million vehicles, was the largest contributor to total U.S. GHG emissions with 1,804 MMT CO₂ e followed by the power sector (25.0 percent) with 1,584 MMT CO₂ e. In fact, GHG emissions from the power sector were higher than the GHG emissions from all other industrial sectors combined (1,487 MMT CO₂ e). Specifically, the power sector's emissions were far more than petroleum and natural gas systems at 301 MMT CO₂ e; chemicals (71 MMT CO₂ e); minerals (64 MMT CO₂ e); coal mining (53 MMT CO₂ e); and metals (48 MMT CO₂ e). The agriculture (636 MMT CO₂ e), commercial (439 MMT CO₂ e), and residential (366 MMT CO₂ e) sectors combined to emit 1,441 MMT CO₂ e.

Fossil fuel-fired EGUs are by far the largest stationary source emitters of GHGs in the nation. For example, according to the EPA's Greenhouse Gas Reporting Program (GHGRP), of the top 100 large facilities that reported facility-level GHGs in 2022, 85 were fossil fuel-fired power plants while 10 were refineries and/or chemical plants, four were metals facilities, and one was a petroleum and natural gas systems facility. Of the 85 fossil fuel-fired power plants, 81 were primarily coal-fired, including the top 41 emitters of CO₂ . In addition, of the 81 coal-fired plants, 43 have no retirement planned prior to 2039. The top 10 of these plants combined to emit more than 135 MMT of CO₂ e, with the top emitter (James H. Miller power plant in Alabama) reporting approximately 22 MMT of CO₂ e with each of its four EGUs emitting between 5 MMT and 6 MMT CO₂ e that year. The combined capacity of these 10 plants is more than 23 gigawatts (GW), and all except for the Monroe (Michigan) plant operated at annual capacity factors of 50 percent or higher. For comparison, the largest GHG emitter in the U.S. that is not a fossil fuel-fired power plant is the ExxonMobil refinery and chemical plant in Baytown, Texas, which reported 12.6 MMT CO₂ e (No. 6 overall in the nation) to the GHGRP in 2022. The largest metals facility in terms of GHG emissions was the U.S. Steel facility in Gary, Indiana, with 10.4 MMT CO₂ e (No. 16 overall in the nation).

Overall, CO₂ emissions from the power sector have declined by 36 percent since 2005 (when the power sector reached annual emissions of 2,400 MMT CO₂ , its historical peak to date). The reduction in CO₂ emissions can be attributed to the power sector's ongoing trend away from carbon-intensive coal-fired generation and toward more natural gas-fired and renewable sources. In 2005, CO₂ emissions from coal-fired EGUs alone measured 1,983 MMT. This total dropped to 1,351 MMT in 2015 and reached 974 MMT in 2019, the first time since 1978 that CO₂ emissions from coal-fired EGUs were below 1,000 MMT. In 2020, emissions of CO₂ from coal-fired EGUs measured 788 MMT as the result of pandemic-related closures and reduced utilization before rebounding in 2021 to 909 MMT. By contrast, CO₂ emissions from natural gas-fired generation have almost doubled since 2005, increasing from 319 MMT to 613 MMT in 2021, and CO₂ emissions from petroleum products ( i.e., distillate fuel oil, petroleum coke, and residual fuel oil) declined from 98 MMT in 2005 to 18 MMT in 2021.

When the EPA finalized the Clean Power Plan (CPP) in October 2015, the Agency projected that, as a result of the CPP, the power sector would reduce its annual CO₂ emissions to 1,632 MMT by 2030, or 32 percent below 2005 levels (2,400 MMT). Instead, even in the absence of Federal regulations for existing EGUs, annual CO₂ emissions from sources covered by the CPP had fallen to 1,540 MMT by the end of 2021, a nearly 36 percent reduction below 2005 levels. The power sector achieved a deeper level of reductions than forecast under the CPP and approximately a decade ahead of time. By the end of 2015, several months after the CPP was finalized, those sources already had achieved CO₂ emission levels of 1,900 MMT, or approximately 21 percent below 2005 levels. However, progress in emission reductions is not uniform across all states and is not guaranteed to continue, therefore Federal policies play an essential role. As discussed earlier in this section, the power sector remains a leading emitter of CO₂ in the U.S., and, despite the emission reductions since 2005, current CO₂ levels continue to endanger human health and welfare. Further, as sources in other sectors of the economy turn to electrification to decarbonize, future CO₂ reductions from fossil fuel-fired EGUs have the potential to take on added significance and increased benefits.

C. Recent Developments in Emissions Control

This section of the preamble describes recent developments in GHG emissions control in general. Details of those controls in the context of BSER determination are provided in section VII.C.1.a for CCS on coal-fired steam generating units, section VII.C.2.a for natural gas co-firing on coal-fired steam generating units, section VIII.F.2.b for efficient generation on natural gas-fired combustion turbines, and section VIII.F.4.c.iv for CCS on natural gas-fired combustion turbines. Further details of the control technologies are available in the final TSDs, GHG Mitigation Measures for Steam Generating Units and GHG Mitigation Measures—CCS for Combustion Turbines, available in the docket for these actions.

1. CCS

One of the key GHG reduction technologies upon which the BSER determinations are founded in these final rules is CCS—a technology that can capture and permanently store CO₂ from fossil fuel-fired EGUs. CCS has three major components: CO₂ capture, transportation, and sequestration/storage. Solvent-based CO₂ capture was patented nearly 100 years ago in the 1930s and has been used in a variety of industrial applications for decades. Thousands of miles of CO₂ pipelines have been constructed and securely operated in the U.S. for decades. And tens of millions of tons of CO₂ have been permanently stored deep underground either for geologic sequestration or in association with enhanced oil recovery (EOR). The American Petroleum Institute (API) explains that “CCS is a proven technology” and that “[t]he methods that apply to [the] carbon sequestration process are not novel. The U.S. has more than 40 years of CO₂ gas injection and storage experience. During the last 40 years the U.S. gas and oil industry's (EOR) enhanced oil recovery operations) have injected more than 1 billion tonnes of CO₂ .”

In 2009, Mike Morris, then-CEO of American Electric Power (AEP), was interviewed by Reuters and the article noted that Morris's “companies' work in West Virginia on [CCS] gave [Morris] more insight than skeptics who doubt the technology.” In that interview, Morris explained, “I'm convinced it will be primetime ready by 2015 and deployable.” In 2011, Alstom Power, the company that developed the 30 MW pilot project upon which Morris had based his conclusions, reiterated the claim that CCS would be commercially available in 2015. A press release from Alstom Power stated that, based on the results of Alstom's “13 pilot and demonstration projects and validated by independent experts . . . we can now be confident that CCS works and is cost effective . . . and will be available at a commercial scale in 2015 and will allow [plants] to capture 90% of the emitted CO₂ .” The press release went on to note that “the same conclusion applies for a gas plant using CCS.”

In 2011, however, AEP determined that the economic and regulatory environment at the time did not support further development of the technology. After canceling a large-scale commercial project, Morris explained, “as a regulated utility, it is impossible to gain regulatory approval to cover our share of the costs for validating and deploying the technology without federal requirements to reduce greenhouse gas emissions already in place.”

Thirteen years later, the situation is fundamentally different. Since 2011, the technological advances from full-scale deployments ( e.g., the Petra Nova and Boundary Dam projects discussed later in this preamble) combined with supportive policies in multiple states and the financial incentives included in the IRA, mean that CCS can be deployed at scale today. In addition to applications at fossil fuel-fired EGUs, installation of CCS is poised to dramatically increase across a range of industries in the coming years, including ethanol production, natural gas processing, and steam methane reformers. Many of the CCS projects across these industries, including capture systems, pipelines, and sequestration, are already in operation or are in advanced stages of deployment. There are currently at least 15 operating CCS projects in the U.S., and another 121 that are under construction or in advanced stages of development.

Process improvements learned from earlier deployments of CCS, the availability of better solvents, and other advances have decreased the costs of CCS in recent years. As a result, the cost of CO₂ capture, excluding any tax credits, from coal-fired power generation is projected to fall by 50 percent by 2025 compared to 2010. The IRA makes additional and significant reductions in the cost of implementing CCS by extending and increasing the tax credit for CO₂ sequestration under IRC section 45Q.

With this combination of polices, and the advances related to CO₂ capture, multiple projects consistent with the emission reduction requirements of a 90 percent capture amine based BSER are in advanced stages of development. These projects use a wider range of technologies, and some of them are being developed as first-of-a-kind projects and offer significant advantages over the amine-based CCS technology that the EPA is finalizing as BSER.

For instance, in North Dakota, Governor Doug Burgum announced a goal of becoming carbon neutral by 2030 while retaining the core position of its fossil fuel industries, and to do so by significant CCS implementation. Gov. Burgum explained, “This may seem like a moonshot goal, but it's actually not. It's actually completely doable, even with the technologies that we have today.” Companies in the state are backing up this claim with projects in multiple industries in various stages of operation and development. In the power sector, two of the biggest projects under development are Project Tundra and Coal Creek. Project Tundra is a carbon capture project on Minnkota Power's 705 MW Milton R Young Power Plant in Oliver County, North Dakota. Mitsubishi Heavy Industries will be providing an advanced version of its carbon capture equipment that builds upon the lessons learned from the Petra Nova project. Rainbow Energy is developing the project at the Coal Creek Station, located in McLean, North Dakota. Notably, Rainbow Energy purchased the 1,150 MW Coal Creek Station with a business model of installing CCS based on the IRC section 45Q tax credit of $50/ton that existed at the time (the IRA has since increased the amount to $85/ton). Rainbow Energy explains, “CCUS technology has been proven and is an economical option for a facility like Coal Creek Station. We see CCUS as the best way to manage emissions at our facility.”

While North Dakota has encouraged CCS on coal-fired power plants without specific mandates, Wyoming is taking a different approach. Senate Bill 42, enacted in 2024, requires utilities to generate a specified percentage of their electricity using coal-fired power plants with CCS. SB 42 updates HB 200, enacted in 2020, which required the CCS to be installed by 2030, which SB 42 extends to 2033. To comply with those requirements, PacificCorp has stated in its 2023 IRP that it intends to install CCS on two coal-fired units by 2028. Rocky Mountain Power has also announced that it will explore a new carbon capture technology at either its David Johnston plant or its Wyodak plant. Another CCS project is also under development at the Dry Fork Power Plant in Wyoming. Currently, a pilot project that will capture 150 tons of CO₂ per day is under construction and is scheduled to be completed in late 2024. Work has also begun on a full-scale front end engineering design (FEED) study.

Like North Dakota, West Virginia does not have a carbon capture mandate, but there are several carbon capture projects under development in the state. One is a new, 2,000 MW natural gas combined cycle plant being developed by Competitive Power Ventures that will capture 90-95 percent of the CO₂ using GE turbine and carbon capture technology. A second is an Omnis Fuel Technologies project to convert the coal-fired Pleasants Power Station to run on hydrogen. Omnis intends to use a pyrolysis-based process to convert coal into hydrogen and graphite. Because the graphite is a usable, solid form of carbon, no CO₂ sequestration will be required. Therefore, unlike more traditional amine-based approaches, instead of the captured CO₂ being a cost, the graphite product will provide a revenue stream. Omnis states that the Pleasants Power Project broke ground in August 2023 and will be online by 2025.

It should be noted that Wyoming, West Virginia, and North Dakota represented the first-, second-, and seventh-largest coal producers, respectively, in the U.S. in 2022.

In addition to the coal-based CCS projects mentioned above, multiple other projects are in advanced stages of development and/or have completed FEED studies. For instance, Linde/BASF is installing a 10 MW pilot project on the Dallman Power Plant in Illinois. Based on results from small scale pilot studies, techno economic analysis indicates that the Linde/BASF process can provide a significant reduction in capital costs compared to the NETL base case for a supercritical pulverized coal plant with carbon capture.” Multiple other FEED studies are either completed or under development, putting those projects on a path to being able to be built and to commence operation well before January 1, 2032.

In addition to the Competitive Power Partners project, there are multiple post-combustion CCS retrofit projects in various stages of development. In particular, NET Power is in advanced stages of development on a 300 MW project in west Texas using the Allam-Fetvedt cycle, which is being designed to achieve greater than 97 percent CO₂ capture. In addition to working on this first project, NET Power has indicated that it has an additional project under development and is working with suppliers to support additional future projects.

In developing these final rules, the EPA reviewed the current state and cost of CCS technology for use with both steam generating units and stationary combustion turbines. This review is reflected in the respective BSER discussions later in this preamble and is further detailed in the accompanying RIA and final TSDs, GHG Mitigation Measures for Steam Generating Units and GHG Mitigation Measures—Carbon Capture and Storage for Combustion Turbines. These documents are included in the rulemaking docket.

2. Natural Gas Co-Firing

For a coal-fired steam generating unit, the substitution of natural gas for some of the coal so that the unit fires a combination of coal and natural gas is known as “natural gas co-firing.” Existing coal-fired steam generating units can be modified to co-fire natural gas in any desired proportion with coal. Generally, the modification of existing boilers to enable or increase natural gas firing involves the installation of new gas burners and related boiler modifications and may involve the construction of a natural gas supply pipeline if one does not already exist. In recent years, the cost of natural gas co-firing has declined because the expected difference between coal and gas prices has decreased and analysis supports lower capital costs for modifying existing boilers to co-fire with natural gas, as discussed in section VII.C.2.a of this preamble.

It is common practice for steam generating units to have the capability to burn multiple fuels onsite, and of the 565 coal-fired steam generating units operating at the end of 2021, 249 of them reported use of natural gas as a primary fuel or for startup. Based on hourly reported CO₂ emission rates from the start of 2015 through the end of 2020, 29 coal-fired steam generating units co-fired with natural gas at rates at or above 60 percent of capacity on an hourly basis. The capability of those units on an hourly basis is indicative of the extent of boiler burner modifications and sizing and capacity of natural gas pipelines to those units, and it implies that those units are technically capable of co-firing at least 60 percent natural gas on a heat input basis on average over the course of an extended period ( e.g., a year). Additionally, many coal-fired steam generating EGUs have also opted to switch entirely to providing generation from the firing of natural gas. Since 2011, more than 80 coal-fired utility boilers have been converted to natural gas-fired utility boilers.

In developing these final actions, the EPA reviewed in detail the current state of natural gas co-firing technology and costs. This review is reflected in the BSER discussions later in this preamble and is further detailed in the accompanying RIA and final TSD, GHG Mitigation Measures for Steam Generating Units. Both documents are included in the rulemaking docket.

3. Efficient Generation

Highly efficient generation is the BSER technology upon which the first phase standards of performance are based for certain new and reconstructed stationary combustion turbine EGUs. This technology is available for both simple cycle and combined cycle combustion turbines and has been demonstrated—along with best operating and maintenance practices—to reduce emissions. Generally, as the thermal efficiency of a combustion turbine increases, less fuel is burned per gross MWh of electricity produced and there is a corresponding decrease in CO₂ and other air emissions.

For simple cycle turbines, manufacturers continue to improve the efficiency by increasing firing temperature, increasing pressure ratios, using intercooling on the air compressor, and adopting other measures. Best operating practices for simple cycle turbines include proper maintenance of the combustion turbine flow path components and the use of inlet air cooling to reduce efficiency losses during periods of high ambient temperatures. For combined cycle turbines, a highly efficient combustion turbine engine is matched with a high-efficiency HRSG. High efficiency also includes, but is not limited to, the use of the most efficient steam turbine and minimizing energy losses using insulation and blowdown heat recovery. Best operating and maintenance practices include, but are not limited to, minimizing steam leaks, minimizing air infiltration, and cleaning and maintaining heat transfer surfaces.

As discussed in section VIII.F.2.b of this preamble, efficient generation technologies have been in use at facilities in the power sector for decades and the levels of efficiency that the EPA is finalizing in this rule have been achieved by many recently constructed turbines. The efficiency improvements are incremental in nature and do not change how the combustion turbine is operated or maintained and present little incremental capital or compliance costs compared to other types of technologies that may be considered for new and reconstructed sources. In addition, more efficient designs have lower fuel costs, which offset at least a portion of the increase in capital costs. For additional discussion of this BSER technology, see the final TSD, Efficient Generation in Combustion Turbines in the docket for this rulemaking.

Efficiency improvements are also available for fossil fuel-fired steam generating units, and as discussed further in section VII.D.4.a, the more efficiently an EGU operates the less fuel it consumes, thereby emitting lower amounts of CO₂ and other air pollutants per MWh generated. Efficiency improvements for steam generating EGUs include a variety of technology upgrades and operating practices that may achieve CO₂ emission rate reductions of 0.1 to 5 percent for individual EGUs. These reductions are small relative to the reductions that are achievable from natural gas co-firing and from CCS. Also, as efficiency increases, some facilities could increase their utilization and therefore increase their CO₂ emissions (as well as emissions of other air pollutants). This phenomenon is known as the “rebound effect.” Because of this potential for perverse GHG emission outcomes resulting from deployment of efficiency measures at certain steam generating units, coupled with the relatively minor overall GHG emission reductions that would be expected, the EPA is not finalizing efficiency improvements as the BSER for any subcategory of existing coal-fired steam generating units. Specific details of efficiency measures are described in the final TSD, GHG Mitigation Measures for Steam Generating Units, and an updated 2023 Sargent and Lundy HRI report ( Heat Rate Improvement Method Costs and Limitations Memo), available in the docket.

D. The Electric Power Sector: Trends and Current Structure

1. Overview

The electric power sector is experiencing a prolonged period of transition and structural change. Since the generation of electricity from coal-fired power plants peaked nearly two decades ago, the power sector has changed at a rapid pace. Today, natural gas-fired power plants provide the largest share of net generation, coal-fired power plants provide a significantly smaller share than in the recent past, renewable energy provides a steadily increasing share, and as new technologies enter the marketplace, power producers continue to replace aging assets—especially coal-fired power plants—with more efficient and lower-cost alternatives.

These developments have significant implications for the types of controls that the EPA determined to qualify as the BSER for different types of fossil fuel-fired EGUs. For example, power plant owners and operators retired an average annual coal-fired EGU capacity of 10 GW from 2015 to 2023, and coal-fired EGUs comprised 58 percent of all retired capacity in 2023. While use of CCS promises significant emissions reduction from fossil fuel-fired sources, it requires substantial up-front capital expenditure. Therefore, it is not a feasible or cost-reasonable emission reduction technology for units that intend to cease operation before they would be able to amortize its costs. Industry stakeholders requested that the EPA structure these rules to avoid imposing costly control obligations on coal-fired power plants that have announced plans to voluntarily cease operations, and the EPA has determined the BSER in accordance with its understanding of which coal-fired units will be able to feasibly and cost-effectively deploy the BSER technologies. In addition, the EPA recognizes that utilities and power plant operators are building new natural gas-fired combustion turbines with plans to operate them at varying levels of utilization, in coordination with other existing and expected new energy sources. These patterns of operation are important for the type of controls that the EPA is finalizing as the BSER for these turbines.

2. Broad Trends Within the Power Sector

For more than a decade, the power sector has been experiencing substantial transition and structural change, both in terms of the mix of generating capacity and in the share of electricity generation supplied by different types of EGUs. These changes are the result of multiple factors, including normal replacements of older EGUs; technological improvements in electricity generation from both existing and new EGUs; changes in the prices and availability of different fuels; state and Federal policy; the preferences and purchasing behaviors of end-use electricity consumers; and substantial growth in electricity generation from renewable sources.

One of the most important developments of this transition has been the evolving economics of the power sector. Specifically, as discussed in section IV.D.3.b of this preamble and in the final TSD, Power Sector Trends, the existing fleet of coal-fired EGUs continues to age and become more costly to maintain and operate. At the same time, natural gas prices have held relatively low due to increased supply, and renewable costs have fallen rapidly with technological improvement and growing scale. Natural gas surpassed coal in monthly net electricity generation for the first time in April 2015, and since that time natural gas has maintained its position as the primary fuel for base load electricity generation, for peaking applications, and for balancing renewable generation. In 2023, generation from natural gas was more than 2.5 times as much as generation from coal. Additionally, there has been increased generation from investments in zero- and low-GHG emission energy technologies spurred by technological advancements, declining costs, state and Federal policies, and most recently, the IIJA and the IRA. For example, the IIJA provides investments and other policies to help commercialize, demonstrate, and deploy technologies such as small modular nuclear reactors, long-duration energy storage, regional clean hydrogen hubs, CCS and associated infrastructure, advanced geothermal systems, and advanced distributed energy resources (DER) as well as more traditional wind, solar, and battery energy storage resources. The IRA provides numerous tax and other incentives to directly spur deployment of clean energy technologies. Particularly relevant to these final actions, the incentives in the IRA, which are discussed in detail later in this section of the preamble, support the expansion of technologies, such as CCS, that reduce GHG emissions from fossil-fired EGUs.

The ongoing transition of the power sector is illustrated by a comparison of data between 2007 and 2022. In 2007, the year of peak coal generation, approximately 72 percent of the electricity provided to the U.S. grid was produced through the combustion of fossil fuels, primarily coal and natural gas, with coal accounting for the largest single share. By 2022, fossil fuel net generation was approximately 60 percent, less than the share in 2007 despite electricity demand remaining relatively flat over this same period. Moreover, the share of generation supplied by coal-fired EGUs fell from 49 percent in 2007 to 19 percent in 2022 while the share supplied by natural gas-fired EGUs rose from 22 to 39 percent during the same period. In absolute terms, coal-fired generation declined by 59 percent while natural gas-fired generation increased by 88 percent. This reflects both the increase in natural gas capacity as well as an increase in the utilization of new and existing natural gas-fired EGUs. The combination of wind and solar generation also grew from 1 percent of the electric power sector mix in 2007 to 15 percent in 2022.

Additional analysis of the utility power sector, including projections of future power sector behavior and the impacts of these final rules, is discussed in more detail in section XII of this preamble, in the accompanying RIA, and in the final TSD, Power Sector Trends. The latter two documents are available in the rulemaking docket. Consistent with analyses done by other energy modelers, the information provided in the RIA and TSD demonstrates that the sector trend of moving away from coal-fired generation is likely to continue, the share from natural gas-fired generation is projected to decline eventually, and the share of generation from non-emitting technologies is likely to continue increasing. For instance, according to the Energy Information Administration (EIA), the net change in solar capacity has been larger than the net change in capacity for any other source of electricity for every year since 2020. In 2024, EIA projects that the actual increase in generation from solar will exceed every other source of generating capacity. This is in part because of the large amounts of new solar coming online in 2024 but is also due to the large amount of energy storage coming online, which will help reduce renewable curtailments. EIA also projects that in 2024, the U.S. will see its largest year for installation of both solar and battery storage. Specifically, EIA projects that 36.4 GW of solar will be added, nearly doubling last year's record of 18.4 GW. Similarly, EIA projects 14.3 GW of new energy storage. This would more than double last year's record installation of 6.4 GW and nearly double the existing total capacity of 15.5 GW. This compares to only 2.5 GW of new natural gas turbine capacity. The only year since 2013 when renewable generation did not make up the majority of new generation capacity in the U.S. was 2018.

3. Coal-Fired Generation: Historical Trends and Current Structure

a. Historical Trends in Coal-Fired Generation

Coal-fired steam generating units have historically been the nation's foremost source of electricity, but coal-fired generation has declined steadily since its peak approximately 20 years ago. Construction of new coal-fired steam generating units was at its highest between 1967 and 1986, with approximately 188 GW (or 9.4 GW per year) of capacity added to the grid during that 20-year period. The peak annual capacity addition was 14 GW, which was added in 1980. These coal-fired steam generating units operated as base load units for decades. However, beginning in 2005, the U.S. power sector—and especially the coal-fired fleet—began experiencing a period of transition that continues today. Many of the older coal-fired steam generating units built in the 1960s, 1970s, and 1980s have retired or have experienced significant reductions in net generation due to cost pressures and other factors. Some of these coal-fired steam generating units repowered with combustion turbines and natural gas. With no new coal-fired steam generating units larger than 25 MW commencing construction in the past decade—and with the EPA unaware of any plans being approved to construct a new coal-fired EGU—much of the fleet that remains is aging, expensive to operate and maintain, and increasingly uncompetitive relative to other sources of generation in many parts of the country.

Since 2007, the power sector's total installed net summer capacity has increased by 167 GW (17 percent) while coal-fired steam generating unit capacity has declined by 123 GW. This reduction in coal-fired steam generating unit capacity was offset by a net increase in total installed wind capacity of 125 GW, net natural gas capacity of 110 GW, and a net increase in utility-scale solar capacity of 71 GW during the same period. Additionally, significant amounts (40 GW) of DER solar were also added. At least half of these changes were in the most recent 7 years of this period. From 2015 to 2022, coal capacity was reduced by 90 GW and this reduction in capacity was offset by a net increase of 69 GW of wind capacity, 63 GW of natural gas capacity, and 59 GW of utility-scale solar capacity. Additionally, a net summer capacity of 30 GW of DER solar were added from 2015 to 2022.

b. Current Structure of Coal-Fired Generation

Although much of the fleet of coal-fired steam generating units has historically operated as base load, there can be notable differences in design and operation across various facilities. For example, coal-fired steam generating units smaller than 100 MW comprise 18 percent of the total number of coal-fired units, but only 2 percent of total coal-fired capacity. Moreover, average annual capacity factors for coal-fired steam generating units have declined from 74 to 50 percent since 2007. These declining capacity factors indicate that a larger share of units are operating in non-base load fashion largely because they are no longer cost-competitive in many hours of the year.

Older power plants also tend to become uneconomic over time as they become more costly to maintain and operate, especially when competing for dispatch against newer and more efficient generating technologies that have lower operating costs. The average coal-fired power plant that retired between 2015 and 2022 was more than 50 years old, and 65 percent of the remaining fleet of coal-fired steam generating units will be 50 years old or more within a decade. To further illustrate this trend, the existing coal-fired steam generating units older than 40 years represent 71 percent (129 GW) of the total remaining capacity. In fact, more than half (100 GW) of the coal-fired steam generating units still operating have already announced retirement dates prior to 2039 or conversion to gas-fired units by the same year. As discussed later in this section, projections anticipate that this trend will continue.

The reduction in coal-fired generation by electric utilities is also evident in data for annual U.S. coal production, which reflects reductions in international demand as well. In 2008, annual coal production peaked at nearly 1,172 million short tons (MMst) followed by sharp declines in 2015 and 2020. In 2015, less than 900 MMst were produced, and in 2020, the total dropped to 535 MMst, the lowest output since 1965. Following the pandemic, in 2022, annual coal production had increased to 594 MMst. For additional analysis of the coal-fired steam generation fleet, see the final TSD, Power Sector Trends included in the docket for this rulemaking.

Notwithstanding these trends, in 2022, coal-fired energy sources were still responsible for 50 percent of CO₂ emissions from the electric power sector.

4. Natural Gas-Fired Generation: Historical Trends and Current Structure

a. Historical Trends in Natural Gas-Fired Generation

There has been significant expansion of the natural gas-fired EGU fleet since 2000, coinciding with efficiency improvements of combustion turbine technologies, increased availability of natural gas, increased demand for flexible generation to support the expanding capacity of variable energy resources, and declining costs for all three elements. According to data from EIA, annual capacity additions for natural gas-fired EGUs peaked between 2000 and 2006, with more than 212 GW added to the grid during this period (about 35 GW per year). Of this total, approximately 147 GW (70 percent) were combined cycle capacity and 65 GW were simple cycle capacity. From 2007 to 2022, more than 132 GW of capacity were constructed and approximately 77 percent of that total were combined cycle EGUs. This figure represents an average of almost 8.8 GW of new combustion turbine generation capacity per year. In 2022, the net summer capacity of combustion turbine EGUs totaled 419 GW, with 289 GW being combined cycle generation and 130 GW being simple cycle generation.

This trend away from electricity generation using coal-fired EGUs to natural gas-fired turbine EGUs is also reflected in comparisons of annual capacity factors, sizes, and ages of affected EGUs. For example, the average annual capacity factors for natural gas-fired units increased from 28 to 38 percent between 2010 and 2022. And compared with the fleet of coal-fired steam generating units, the natural gas fleet is generally smaller and newer. While 67 percent of the coal-fired steam generating unit fleet capacity is over 500 MW per unit, 75 percent of the gas fleet is between 50 and 500 MW per unit. In terms of the age of the generating units, nearly 50 percent of the natural gas capacity has been in service less than 15 years.

b. Current Structure of Natural Gas-Fired Generation

In the lower 48 states, most combustion turbine EGUs burn natural gas, and some have the capability to fire distillate oil as backup for periods when natural gas is not available, such as when residential demand for natural gas is high during the winter. Areas of the country without access to natural gas often use distillate oil or some other locally available fuel. Combustion turbines have the capability to burn either gaseous or liquid fossil fuels, including but not limited to kerosene, naphtha, synthetic gas, biogases, liquified natural gas (LNG), and hydrogen.

Over the past 20 years, advances in hydraulic fracturing ( i.e., fracking) and horizontal drilling techniques have opened new regions of the U.S. to gas exploration. As the production of natural gas has increased, the annual average price has declined during the same period, leading to more natural gas-fired combustion turbines. Natural gas net generation increased 181 percent in the past two decades, from 601 thousand gigawatt-hours (GWh) in 2000 to 1,687 thousand GWh in 2022. For additional analysis of natural gas-fired generation, see the final TSD, Power Sector Trends included in the docket for this rulemaking.

E. The Legislative, Market, and State Law Context

1. Recent Legislation Impacting the Power Sector

On November 15, 2021, President Biden signed the IIJA (also known as the Bipartisan Infrastructure Law), which allocated more than $65 billion in funding via grant programs, contracts, cooperative agreements, credit allocations, and other mechanisms to develop and upgrade infrastructure and expand access to clean energy technologies. Specific objectives of the legislation are to improve the nation's electricity transmission capacity, pipeline infrastructure, and increase the availability of low-GHG fuels. Some of the IIJA programs that will impact the utility power sector include more than $20 billion to build and upgrade the nation's electric grid, up to $6 billion in financial support for existing nuclear reactors that are at risk of closing, and more than $700 million for upgrades to the existing hydroelectric fleet. The IIJA established the Carbon Dioxide Transportation Infrastructure Finance and Innovation Program to provide flexible Federal loans and grants for building CO₂ pipelines designed with excess capacity, enabling integrated carbon capture and geologic storage. The IIJA also allocated $21.5 billion to fund new programs to support the development, demonstration, and deployment of clean energy technologies, such as $8 billion for the development of regional clean hydrogen hubs and $7 billion for the development of carbon management technologies, including regional direct air capture hubs, carbon capture large-scale pilot projects for development of transformational technologies, and carbon capture commercial-scale demonstration projects to improve efficiency and effectiveness. Other clean energy technologies with IIJA and IRA funding include industrial demonstrations, geologic sequestration, grid-scale energy storage, and advanced nuclear reactors.

The IRA, which President Biden signed on August 16, 2022, has the potential for even greater impacts on the electric power sector. Energy Security and Climate Change programs in the IRA covering grant funding and tax incentives provide significant investments in low and non GHG-emitting generation. For example, one of the conditions set by Congress for the expiration of the Clean Electricity Production Tax Credits of the IRA, found in section 13701, is a 75 percent reduction in GHG emissions from the power sector below 2022 levels. The IRA also contains the Low Emission Electricity Program (LEEP) with funding provided to the EPA with the objective to reduce GHG emissions from domestic electricity generation and use through promotion of incentives, tools to facilitate action, and use of CAA regulatory authority. In particular, CAA section 135, added by IRA section 60107, requires the EPA to conduct an assessment of the GHG emission reductions expected to occur from changes in domestic electricity generation and use through fiscal year 2031 and, further, provides the EPA $18 million “to ensure that reductions in [GHG] emissions are achieved through use of the existing authorities of [the Clean Air Act], incorporating the assessment. . . .” CAA section 135(a)(6).

The IRA's provisions also demonstrate an intent to support development and deployment of low-GHG emitting technologies in the power sector through a broad array of additional tax credits, loan guarantees, and public investment programs. Particularly relevant for these final actions, these provisions are aimed at reducing emissions of GHGs from new and existing generating assets, with tax credits for CCUS and clean hydrogen production, providing a pathway for the use of coal and natural gas as part of a low-GHG electricity grid.

To assist states and utilities in their decarbonizing efforts, and most germane to these final actions, the IRA increased the tax credit incentives for capturing and storing CO₂, including from industrial sources, coal-fired steam generating units, and natural gas-fired stationary combustion turbines. The increase in credit values, found in section 13104 (which revises IRC section 45Q), is 70 percent, equaling $85/metric ton for CO₂ captured and securely stored in geologic formations and $60/metric ton for CO₂ captured and utilized or securely stored incidentally in conjunction with EOR. The CCUS incentives include 12 years of credits that can be claimed at the higher credit value beginning in 2023 for qualifying projects. These incentives will significantly cut costs and are expected to accelerate the adoption of CCS in the utility power and other industrial sectors. Specifically for the power sector, the IRA requires that a qualifying carbon capture facility have a CO₂ capture design capacity of not less than 75 percent of the baseline CO₂ production of the unit and that construction must begin before January 1, 2033. Tax credits under IRC section 45Q can be combined with some other tax credits, in some circumstances, and with state-level incentives, including California's low carbon fuel standard, which is a market-based program with fuel-specific carbon intensity benchmarks. The magnitude of this incentive is driving investment and announcements, evidenced by the increased number of permit applications for geologic sequestration.

The new provisions in section 13204 (IRC section 45V) codify production tax credits for `clean hydrogen' as defined in the provision. The value of the credits earned by a project is tiered (four different tiers) and depends on the estimated GHG emissions of the hydrogen production process as defined in the statute. The credits range from $3/kg H₂ for less than 0.45 kilograms of CO₂ -equivalent emitted per kilogram of low-GHG hydrogen produced (kg CO₂ e/kg H₂) down to $0.6/kg H₂ for 2.5 to 4.0 kg CO₂ e/kg H₂ (assuming wage and apprenticeship requirements are met). Projects with production related GHG emissions greater than 4.0 kg CO₂ e/kg H₂ are not eligible. Future costs for clean hydrogen produced using renewable energy are anticipated to through 2030 due to these tax incentives and concurrent scaling up of manufacturing and deployment of clean hydrogen production facilities.

Both IRC section 45Q and IRC section 45V are eligible for additional provisions that increase the value and usability of the credits. Certain tax-exempt entities, such as electric co-operatives, may elect direct payment for the full 12- or 10-year lifetime of the credits to monetize the credits directly as cash refunds rather than through tax equity transactions. Tax-paying entities may elect to have direct payment of IRC section 45Q or 45V credits for 5 consecutive years. Tax-paying entities may also elect to transfer credits to unrelated taxpayers, enabling direct monetization of the credits again without relying on tax equity transactions.

In addition to provisions such as 45Q that allow for the use of fossil-generating assets in a low-GHG future, the IRA also includes significant incentives to deploy clean energy generation. For instance, the IRA provides an additional 10 percent in production tax credit (PTC) and investment tax credit (ITC) bonuses for clean energy projects located in energy communities with historic employment and tax bases related to fossil fuels. The IRA's Energy Infrastructure Reinvestment Program also provides $250 billion for the DOE to finance loan guarantees that can be used to reduce both the cost of retiring existing fossil assets and of replacement generation for those assets, including updating operating energy infrastructure with emissions control technologies. As a further example, the Empowering Rural America (New ERA) Program provides rural electric cooperatives with funds that can be used for a variety of purposes, including “funding for renewable and zero emissions energy systems that eliminate aging, obsolete or expensive infrastructure” or that allow rural cooperatives to “change [their] purchased-power mixes to support cleaner portfolios, manage stranded assets and boost [the] transition to clean energy.” The $9.7 billion New ERA program represents the single largest investment in rural energy systems since the Rural Electrification Act of 1936.

On September 12, 2023, the EPA released a report assessing the impact of the IRA on the power sector. Modeling results showed that economy-wide CO₂ emissions are lower under the IRA. The results from the EPA's analysis of an array of multi-sector and electric sector modeling efforts show that a wide range of emissions reductions are possible. The IRA spurs CO₂ emissions reductions from the electric power sector of 49 to 83 percent below 2005 levels in 2030. This finding reflects diversity in how the models represent the IRA, the assumptions the models use, and fundamental differences in model structures.

In determining the CAA section 111 emission limitations that are included in these final actions, the EPA did not consider many of the technologies that receive investment under recent Federal legislation. The EPA's determination of the BSER focused on “measures that improve the pollution performance of individual sources,” not generation technologies that entities could employ as alternatives to fossil fuel-fired EGUs. However, these overarching incentives and policies are important context for this rulemaking and influence where control technologies can be feasibly and cost-reasonably deployed, as well as how owners and operators of EGUs may respond to the requirements of these final actions.

2. Commitments by Utilities To Reduce GHG Emissions

Integrated resource plans (IRPs) are filed by public utilities and demonstrate how utilities plan to meet future forecasted energy demand while ensuring reliable and cost-effective service. In developing these rules, the EPA reviewed filed IRPs of companies that have publicly committed to reducing their GHGs. These IRPs demonstrate a range of strategies that public utilities are planning to adopt to reduce their GHGs, independent of these final actions. These strategies include retiring aging coal-fired steam generating EGUs and replacing them with a combination of renewable resources, energy storage, other non-emitting technologies, and natural gas-fired combustion turbines, and reducing GHGs from their natural gas-fired assets through a combination of CCS and reduced utilization. To affirm these findings, according to EIA, as of 2022 there are no new coal-fired EGUs in development. This section highlights recent actions and announced plans of many utilities across the industry to reduce GHGs from their fleets. Indeed, 50 power producers that are members of the Edison Electric Institute (EEI) have announced CO₂ reduction goals, two-thirds of which include net-zero carbon emissions by 2050. The members of the Energy Strategies Coalition, a group of companies that operate and manage electricity generation facilities, as well as electricity and natural gas transmission and distribution systems, likewise are focused on investments to reduce carbon dioxide emissions from the electricity sector. This trend is not unique. Smaller utilities, rural electric cooperatives, and municipal entities are also contributing to these changes.

Many electric utilities have publicly announced near- and long-term emission reduction commitments independent of these final actions. The Smart Electric Power Alliance demonstrates that the geographic footprint of commitments for 100 percent renewable, net-zero, or other carbon emission reductions by 2050 made by utilities, their parent companies, or in response to a state clean energy requirement, covers portions of 47 states and includes 80 percent of U.S. customer accounts. According to this same source, 341 utilities in 26 states have similar commitments by 2040. Additional detail about emission reduction commitments from major utilities is provided in section 2.2 of the RIA and in the final TSD, Power Sector Trends.

3. State Actions To Reduce Power Sector GHG Emissions

States across the country have taken the lead in efforts to reduce GHG emissions from the power sector. As of mid-2023, 25 states had made commitments to reduce economy-wide GHG emissions consistent with the goals of the Paris Agreement, including reducing GHG emissions by 50 to 52 percent by 2030. These actions include legislation to decarbonize state power systems as well as commitments that require utilities to expand renewable and clean energy production through the adoption of renewable portfolio standards (RPS) and clean energy standards (CES).

Several states have enacted binding economy-wide emission reduction targets that will require significant decarbonization from state power sectors, including California, Colorado, Maine, Maryland, Massachusetts, New Jersey, New York, Rhode Island, Vermont, and Washington. These commitments are statutory emission reduction targets accompanied by mandatory agency directives to develop comprehensive implementing regulations to achieve the necessary reductions. Some of these states, along with other neighboring states, also participate in the Regional Greenhouse Gas Initiative (RGGI), a carbon market limiting pollution from power plants throughout New England. The pollution limit combined with carbon price and allowance market has led member states to reduce power sector CO₂ emissions by nearly 50 percent since the start of the program in 2009. This is 10 percent more than all non-RGGI states.

Other states dependent on coal-fired power generation or coal production also have significant, albeit non- binding, commitments that signal broad public support for policy with emissions-based metrics and public affirmation that climate change is fundamentally linked to fossil-intensive energy sources. These states include Illinois, Michigan, Minnesota, New Mexico, North Carolina, Pennsylvania, and Virginia. States like Wyoming, the top coal producing state in the U.S., have promulgated sector-specific regulations requiring their public service commissions to implement low-carbon energy standards for public utilities. Specific standards are further detailed in the sections that follow and in the final TSD, Power Sector Trends.

Technologies like CCS provide a means to achieve significant emission reduction targets. For example, to achieve GHG emission reduction goals legislatively enacted in 2016, California Senate Bill 100, passed in 2018, requires the state to procure 60 percent of all electricity from renewable sources by 2030 and plan for 100 percent from carbon-free sources by 2045. Achieving California's established goal of carbon-free electricity by 2045 requires emissions to be balanced by carbon sequestration, capture, or other technologies. Therefore, California Senate Bill 905, passed in 2022, requires the California Air Resources Board (CARB) to establish programs for permitting CCS projects while preventing the use of captured CO₂ for EOR within the state. As mentioned previously, as the top coal producing state, Wyoming has been exceptionally persistent on the implementation of CCS by incentivizing the national testing of CCS at Basin Electric's coal-fired Dry Fork Station and by requiring the consideration of CCS as an alternative to coal plant retirement. At least five other states, including Montana and North Dakota, also have tax incentives and regulations for CCS. In the case of Montana, the acquisition of an equity interest or lease of coal-fired EGUs is prohibited unless it captures and stores at least 50 percent of its CO₂ emissions. These state policies have coincided with the planning and development of large CCS projects.

Other states have broad decarbonization laws that will drive significant decrease in power sector GHG emissions. In New York, The Climate Leadership and Community Protection Act, passed in 2019, sets several climate targets. The most important goals include an 85 percent reduction in GHG emissions by 2050, 100 percent zero-emission electricity by 2040, and 70 percent renewable energy by 2030. Other targets include 9,000 MW of offshore wind by 2035, 3,000 MW of energy storage by 2030, and 6,000 MW of solar by 2025. Washington State's Climate Commitment Act sets a target of reducing GHG emissions by 95 percent by 2050. The state is required to reduce emissions to 1990 levels by 2020, 45 percent below 1990 levels by 2030, 70 percent below 1990 levels by 2040, and 95 percent below 1990 levels by 2050. This also includes achieving net-zero emissions by 2050. Illinois' Climate and Equitable Jobs Act, enacted in September 2021, requires all private coal-fired or oil-fired power plants to reach zero carbon emissions by 2030, municipal coal-fired plants to reach zero carbon emissions by 2045, and natural gas-fired plants to reach zero carbon emissions by 2045. In October 2021, North Carolina passed House Bill 951 that required the North Carolina Utilities Commission to “take all reasonable steps to achieve a seventy percent (70 percent) reduction in emissions of carbon dioxide (CO₂ ) emitted in the state from electric generating facilities owned or operated by electric public utilities from 2005 levels by the year 2030 and carbon neutrality by the year 2050.”

The ambition and scope of these state power sector polices will impact the electric generation fleet for decades. Seven states with 100-percent power sector decarbonization polices include a total of 20 coal-fired EGUs with slightly less than 10 GW total capacity and without announced retirement dates before 2039. Virginia, which has three coal-steam units with no announced retirement dates and one with a 2045 retirement date, enacted the Clean Economy Act in 2020 to impose a 100 percent RPS requirement by 2050. The combined capacity of all four of these units in Virginia totals nearly 1.5 GW. North Carolina, which has one coal-fired unit without an announced retirement date and one with a planned 2048 retirement, as previously mentioned, enacted a state law in 2021 requiring the state's utilities commission to achieve carbon neutrality by 2050. The combined capacity of both units totals approximately 1.4 GW of capacity. Nebraska, where three public utility boards serving a large portion of the state have adopted net-zero electricity emission goals by 2040 or 2050, includes six coal-fired units with a combined capacity of 2.9 GW. The remaining eight units are in states with long-term decarbonization goals (Illinois, Louisiana, Maryland, and Wisconsin). All four of these states have set 100 percent clean energy goals by 2050.

Twenty-nine states and the District of Columbia have enforceable RPS that require a percentage of electricity that utilities sell to come from eligible renewable sources like wind and solar rather than from fossil fuel-based sources like coal and natural gas. Furthermore, 20 states have adopted a CES that includes some form of clean energy requirement or goal with a 100 percent or net-zero target. A CES shifts generating fleets away from fossil fuel resources by requiring a percentage of retail electricity to come from sources that are defined as clean. Unlike an RPS, which defines eligible generation in terms of the renewable attributes of its energy source, CES eligibility is based on the GHG emission attributes of the generation itself, typically with a zero or net-zero carbon emissions requirement. Additional discussion of state actions and legislation to reduce GHG emissions from the power sector is provided in the final TSD, Power Sector Trends.

F. Future Projections of Power Sector Trends

Projections for the U.S. power sector—based on the landscape of market forces in addition to the known actions of Congress, utilities, and states—have indicated that the ongoing transition will continue for specific fuel types and EGUs. The EPA's Power Sector Platform 2023 using IPM reference case ( i.e., the EPA's projections of the power sector, which includes representation of the IRA absent further regulation), provides projections out to 2050 on future outcomes of the electric power sector. For more information on the details of this modeling, see the model documentation.

Since the passage of the IRA in August 2022, the EPA has engaged with many external partners, including other governmental entities, academia, non-governmental organizations (NGOs), and industry, to understand the impacts that the IRA will have on power sector GHG emissions. In addition to engaging in several workgroups, the EPA has contributed to two separate journal articles that include multi-model comparisons of IRA impacts across several state-of-the-art models of the U.S. energy system and electricity sector and participated in public events exploring modeling assumptions for the IRA. The EPA plans to continue collaborating with stakeholders, conducting external engagements, and using information gathered to refine modeling of the IRA.

While much of the discussion below focuses on the EPA's Power Sector Platform 2023 using IPM reference case, many other analyses show similar trends, and these trends are consistent with utility IRPs and public GHG reduction commitments, as well as state actions, both of which were described in the previous sections.

1. Future Projections for Coal-Fired Generation

As described in the EPA's baseline modeling, coal-fired steam generating unit capacity is projected to fall from 181 GW in 2023 to 52 GW in 2035, of which 11 GW includes retrofit CCS. Generation from coal-fired steam generating units is projected to also fall from 898 thousand GWh in 2021 to 236 thousand GWh by 2035. This change in generation reflects the anticipated continued decline in projected coal-fired steam generating unit capacity as well as a steady decline in annual operation of those EGUs that remain online, with capacity factors falling from approximately 48 percent in 2022 to 45 percent in 2035 at facilities that do not install CCS. By 2050, coal-fired steam generating unit capacity is projected to diminish further, with only 28 GW, or less than 16 percent of 2023 capacity (and approximately 9 percent of the 2010 capacity), still in operation across the continental U.S.

These projections are driven by the eroding economic opportunities for coal-fired steam generating units to operate, the continued aging of the fleet of coal-fired steam generating units, and the continued availability and expansion of low-cost alternatives, like natural gas, renewable technologies, and energy storage. The projected retirements continue the trend of coal plant retirements in recent decades that is described in section IV.D.3. of this preamble (and further in the Power Sector Trends technical support document). The decline in coal generation capacity has generally resulted from a more competitive economic environment and increasing coal plant age. Most notably, declines in natural gas prices associated with the rise of hydraulic fracturing and horizontal drilling lowered the cost of natural gas-fired generation. Lower gas generation costs reduced coal plant capacity factors and revenues. Rapid declines in the costs of renewables and battery storage have put further price pressure on coal plants, given the zero marginal cost operation of solar and wind. In addition, most operational coal plants today were built before 2000, and many are reaching or have surpassed their expected useful lives. Retiring coal plants tend to be old. As plants age, their efficiency tends to decline and operations and maintenance costs increase. Older coal plant operational parameters are less aligned with current electric grid needs. Coal plants historically were used as base load power sources and can be slow (or expensive) to increase or decrease generation output throughout a typical day. That has put greater economic pressure on older coal plants, which are forced to either incur the costs of adjusting their generation or operate during less profitable hours when loads are lower or renewable generation is more plentiful. All of these factors have contributed to retirements over the past 15 years, and similar underlying factors are projected to continue the trend of coal retirements in the coming years.

In 2020, there was a total of 1,439 million metric tons of CO₂ emissions from the power sector with coal-fired sources contributing to more than half of those emissions. In the EPA's Power Sector Platform 2023 using IPM reference case, power sector related CO₂ emission are projected to fall to 724 million metric tons by 2035, of which 23 percent is projected to come from coal-fired sources in 2035.

2. Future Projections for Natural Gas-Fired Generation

As described in the EPA's Power Sector Platform 2023 using IPM reference case, natural gas-fired capacity is expected to continue to build out during the next decade with 34 GW of new capacity projected to come online by 2035 and 261 GW of new capacity by 2050. By 2035, the new natural gas capacity is comprised of 14 GW of simple cycle turbines and 20 GW of combined cycle turbines. By 2050, most of the incremental new capacity is projected to come just from simple cycle turbines. This also represents a higher rate of new simple cycle turbine builds compared to the reference periods ( i.e., 2000-2006 and 2007-2021) discussed previously in this section.

It should be noted that despite this increase in capacity, both overall generation and emissions from the natural gas-fired capacity are projected to decline. Generation from natural gas units is projected to fall from 1,579 thousand GWh in 2021 to 1,344 thousand GWh by 2035. Power sector related CO₂ emissions from natural gas-fired EGUs were 615 million metric tons in 2021. By 2035, emission levels are projected to reach 521 million metric tons, 96 percent of which comes from NGCC sources.

The decline in generation and emissions is driven by a projected decline in NGCC capacity factors. In model projections, NGCC units have a capacity factor early in the projection period of 59 percent, but by 2035, capacity factor projections fall to 48 percent as many of these units switch from base load operation to more intermediate load operation to support the integration of variable renewable energy resources. Natural gas-fired simple cycle turbine capacity factors also fall, although since they are used primarily as a peaking resource and their capacity factors are already below 10 percent annually, their impact on generation and emissions changes are less notable.

Some of the reasons for this anticipated continued growth in natural gas-fired capacity, coupled with a decline in generation and emissions, include the anticipated growth in peak load, retirement of older fossil generators, and growth in renewable energy coupled with the greater flexibility offered by combustion turbines. Simple cycle turbines operate at lower efficiencies than NGCC units but offer fast startup times to meet peaking load demands. In addition, combustion turbines, along with energy storage technologies and demand response strategies, support the expansion of renewable electricity by meeting demand during peak periods and providing flexibility around the variability of renewable generation and electricity demand. In the longer term, as renewables and battery storage grow, they are anticipated to outcompete the need for some natural gas-fired generation and the overall utilization of natural gas-fired capacity is expected to decline. For additional discussion and analysis of projections of future coal- and natural gas-fired generation, see the final TSD, Power Sector Trends in the docket for this rulemaking.

As explained in greater detail later in this preamble and in the accompanying RIA, future generation projections for natural gas-fired combustion turbines differ from those highlighted in recent historical trends. The largest source of new generation is from renewable energy, and projections show that total natural gas-fired combined cycle capacity is likely to decline after 2030 in response to increased generation from renewables, deployment of energy storage, and other technologies. Approximately 95 percent of capacity additions in 2024 are expected to be from non-emitting generation resources including solar, battery storage, wind, and nuclear. The IRA is likely to influence this trend, which is also expected to impact the operation of certain combustion turbines. For example, as the electric output from additional variable renewable generating sources fluctuates daily and seasonally, flexible low and intermediate load combustion turbines will be needed to support these variable sources and provide reliability to the grid. This requires the ability to start and stop quickly and change load more frequently. Today's system includes 212 GW of intermediate and low load combustion turbines. These operational changes, alongside other tools like demand response, energy storage, and expanded transmission, will maintain reliability of the grid.

V. Statutory Background and Regulatory History for CAA Section 111

A. Statutory Authority To Regulate GHGs From EGUs Under CAA Section 111

The EPA's authority for and obligation to issue these final rules is CAA section 111, which establishes mechanisms for controlling emissions of air pollutants from new and existing stationary sources. CAA section 111(b)(1)(A) requires the EPA Administrator to promulgate a list of categories of stationary sources that the Administrator, in his or her judgment, finds “causes, or contributes significantly to, air pollution which may reasonably be anticipated to endanger public health or welfare.” The EPA has the authority to define the scope of the source categories, determine the pollutants for which standards should be developed, and distinguish among classes, types, and sizes within categories in establishing the standards.

1. Regulation of Emissions From New Sources

Once the EPA lists a source category, the EPA must, under CAA section 111(b)(1)(B), establish “standards of performance” for “new sources” in the source category. These standards are referred to as new source performance standards, or NSPS. The NSPS are national requirements that apply directly to the sources subject to them.

Under CAA section 111(a)(1), a “standard of performance” is defined, in the singular, as “a standard for emissions of air pollutants” that is determined in a specified manner, as noted in this section, below.

Under CAA section 111(a)(2), a “new source” is defined, in the singular, as “any stationary source, the construction or modification of which is commenced after the publication of regulations (or, if earlier, proposed regulations) prescribing a standard of performance under this section, which will be applicable to such source.” Under CAA section 111(a)(3), a “stationary source” is defined as “any building, structure, facility, or installation which emits or may emit any air pollutant.” Under CAA section 111(a)(4), “modification” means any physical change in, or change in the method of operation of, a stationary source which increases the amount of any air pollutant emitted by such source or which results in the emission of any air pollutant not previously emitted. While this provision treats modified sources as new sources, EPA regulations also treat a source that undergoes “reconstruction” as a new source. Under the provisions in 40 CFR 60.15, “reconstruction” means the replacement of components of an existing facility such that: (1) The fixed capital cost of the new components exceeds 50 percent of the fixed capital cost that would be required to construct a comparable entirely new facility; and (2) it is technologically and economically feasible to meet the applicable standards. Pursuant to CAA section 111(b)(1)(B), the standards of performance or revisions thereof shall become effective upon promulgation.

In setting or revising a performance standard, CAA section 111(a)(1) provides that performance standards are to reflect “the degree of emission limitation achievable through the application of the best system of emission reduction which (taking into account the cost of achieving such reduction and any non-air quality health and environmental impact and energy requirements) the Administrator determines has been adequately demonstrated.” The term “standard of performance” in CAA 111(a)(1) makes clear that the EPA is to determine both the “best system of emission reduction . . . adequately demonstrated” (BSER) for the regulated sources in the source category and the “degree of emission limitation achievable through the application of the [BSER].” West Virginia v. EPA, 597 U.S. 697, 709 (2022). To determine the BSER, the EPA first identifies the “system[s] of emission reduction” that are “adequately demonstrated,” and then determines the “best” of those systems, “taking into account” factors including “cost,” “nonair quality health and environmental impact,” and “energy requirements.” The EPA then derives from that system an “achievable” “degree of emission limitation.” The EPA must then, under CAA section 111(b)(1)(B), promulgate “standard[s] for emissions”—the NSPS—that reflect that level of stringency.

2. Regulation of Emissions From Existing Sources

When the EPA establishes a standard for emissions of an air pollutant from new sources within a category, it must also, under CAA section 111(d), regulate emissions of that pollutant from existing sources within the same category, unless the pollutant is regulated under the National Ambient Air Quality Standards (NAAQS) program, under CAA sections 108-110, or the National Emission Standards for Hazardous Air Pollutants (NESHAP) program, under CAA section 112. See CAA section 111(d)(1)(A)(i) and (ii); West Virginia, 597 U.S. at 710.

CAA section 111(d) establishes a framework of “cooperative federalism for the regulation of existing sources.” American Lung Ass'n, 985 F.3d at 931. CAA sections 111(d)(1)(A)-(B) require “[t]he Administrator . . . to prescribe regulations” that require “[e]ach state . . . to submit to [EPA] a plan . . . which establishes standards of performance for any existing stationary source for” the air pollutant at issue, and which “provides for the implementation and enforcement of such standards of performance.” CAA section 111(a)(6) defines an “existing source” as “any stationary source other than a new source.”

To meet these requirements, the EPA promulgates “emission guidelines” that identify the BSER and the degree of emission limitation achievable through the application of the BSER. Each state must then establish standards of performance for its sources that reflect that level of stringency. However, the states need not compel regulated sources to adopt the particular components of the BSER itself. The EPA's emission guidelines must also permit a state, “in applying a standard of performance to any particular source,” to “take into consideration, among other factors, the remaining useful life of the existing source to which such standard applies.” 42 U.S.C. 7411(d)(1). Once a state receives the EPA's approval of its plan, the provisions in the plan become federally enforceable against the source, in the same manner as the provisions of an approved State Implementation Plan (SIP) under the Act. CAA section 111(d)(2)(B). If a state elects not to submit a plan or submits a plan that the EPA does not find “satisfactory,” the EPA must promulgate a plan that establishes Federal standards of performance for the state's existing sources. CAA section 111(d)(2)(A).

3. EPA Review of Requirements

CAA section 111(b)(1)(B) requires the EPA to “at least every 8 years, review and, if appropriate, revise” new source performance standards. However, the Administrator need not review any such standard if the “Administrator determines that such review is not appropriate in light of readily available information on the efficacy” of the standard. Id. When conducting a review of an NSPS, the EPA has the discretion and authority to add emission limits for pollutants or emission sources not currently regulated for that source category. CAA section 111 does not by its terms require the EPA to review emission guidelines for existing sources, but the EPA retains the authority to do so. See 81 FR 59277 (August 29, 2016) (explaining legal authority to review emission guidelines for municipal solid waste landfills).

B. History of EPA Regulation of Greenhouse Gases From Electricity Generating Units Under CAA Section 111 and Caselaw

The EPA has listed more than 60 stationary source categories under CAA section 111(b)(1)(A). See 40 CFR part 60, subparts Cb-OOOO. In 1971, the EPA listed fossil fuel-fired EGUs (which includes natural gas, petroleum, and coal) that use steam-generating boilers in a category under CAA section 111(b)(1)(A). See 36 FR 5931 (March 31, 1971) (listing “fossil fuel-fired steam generators of more than 250 million Btu per hour heat input”). In 1977, the EPA listed fossil fuel-fired combustion turbines, which can be used in EGUs, in a category under CAA section 111(b)(1)(A). See 42 FR 53657 (October 3, 1977) (listing “stationary gas turbines”).

Beginning in 2007, several decisions by the U.S. Supreme Court and the D.C. Circuit have made clear that under CAA section 111, the EPA has authority to regulate GHG emissions from listed source categories. The U.S. Supreme Court ruled in Massachusetts v. EPA that GHGs meet the definition of “air pollutant” in the CAA, and subsequently premised its decision in AEP v. Connecticut —that the CAA displaced any Federal common law right to compel reductions in CO₂ emissions from fossil fuel-fired power plants—on its view that CAA section 111 applies to GHG emissions. The D.C. Circuit confirmed in American Lung Ass'n v. EPA, 985 F.3d 914, 977 (D.C. Cir. 2021), discussed in section V.B.5, that the EPA is authorized to promulgate requirements under CAA section 111 for GHG from the fossil fuel-fired EGU source category notwithstanding that the source category is regulated under CAA section 112. As discussed in section V.B.6, the U.S. Supreme Court did not accept certiorari on the question whether the EPA could regulate GHGs from fossil-fuel fired EGUs under CAA section 111(d) when other pollutants from fossil-fuel fired EGUs are regulated under CAA section 112 in West Virginia v. EPA, 597 U.S. 697 (2022), and so the D.C. Circuit's holding on this issue remains good law.

In 2015, the EPA promulgated two rules that addressed CO₂ emissions from fossil fuel-fired EGUs. The first promulgated standards of performance for new fossil fuel-fired EGUs. “Standards of Performance for Greenhouse Gas Emissions From New, Modified, and Reconstructed Stationary Sources: Electric Utility Generating Units; Final Rule,” (80 FR 64510; October 23, 2015) (2015 NSPS). The second promulgated emission guidelines for existing sources. “Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units; Final Rule,” (80 FR 64662; October 23, 2015) (Clean Power Plan, or CPP).

1. 2015 NSPS

In 2015, the EPA promulgated an NSPS to limit emissions of GHGs, manifested as CO₂, from newly constructed, modified, and reconstructed fossil fuel-fired electric utility steam generating units, i.e., utility boilers and IGCC EGUs, and newly constructed and reconstructed stationary combustion turbine EGUs. These final standards are codified in 40 CFR part 60, subpart TTTT. In promulgating the NSPS for newly constructed fossil fuel-fired steam generating units, the EPA determined the BSER to be a new, highly efficient, supercritical pulverized coal (SCPC) EGU that implements post-combustion partial CCS technology. The EPA concluded that CCS was adequately demonstrated (including being technically feasible) and widely available and could be implemented at reasonable cost. The EPA identified natural gas co-firing and IGCC technology (either with natural gas co-firing or implementing partial CCS) as alternative methods of compliance.

The 2015 NSPS included standards of performance for steam generating units that undergo a “reconstruction” as well as units that implement “large modifications,” (i.e., modifications resulting in an increase in hourly CO₂ emissions of more than 10 percent). The 2015 NSPS did not establish standards of performance for steam generating units that undertake “small modifications” ( i.e., modifications resulting in an increase in hourly CO₂ emissions of less than or equal to 10 percent), due to the limited information available to inform the analysis of a BSER and corresponding standard of performance.

The 2015 NSPS also finalized standards of performance for newly constructed and reconstructed stationary combustion turbine EGUs. For newly constructed and reconstructed base load natural gas-fired stationary combustion turbines, the EPA finalized a standard based on efficient NGCC technology as the BSER. For newly constructed and reconstructed non-base load natural gas-fired stationary combustion turbines and for both base load and non-base load multi-fuel-fired stationary combustion turbines, the EPA finalized a heat input-based standard based on the use of lower-emitting fuels (referred to as clean fuels in the 2015 NSPS). The EPA did not promulgate final standards of performance for modified stationary combustion turbines due to lack of information. The 2015 NSPS remains in effect today.

The EPA received six petitions for reconsideration of the 2015 NSPS. On May 6, 2016 (81 FR 27442), the EPA denied five of the petitions on the basis that they did not satisfy the statutory conditions for reconsideration under CAA section 307(d)(7)(B) and deferred action on one petition that raised the issue of the treatment of biomass. Apart from these petitions, the EPA proposed to revise the 2015 NSPS in 2018, as discussed in section V.B.2.

Multiple parties also filed petitions for judicial review of the 2015 NSPS in the D.C. Circuit. These cases have been briefed and, on the EPA's motion, are being held in abeyance pending EPA action concerning the 2018 proposal to revise the 2015 NSPS.

In the 2015 NSPS, the EPA noted that it was authorized to regulate GHGs from the fossil fuel-fired EGU source categories because it had listed those source categories under CAA section 111(b)(1)(A). The EPA added that CAA section 111 did not require it to make a determination that GHGs from EGUs contribute significantly to dangerous air pollution (a pollutant-specific significant contribution finding), but in the alternative, the EPA did make that finding. It explained that “[greenhouse gas] air pollution may reasonably be anticipated to endanger public health or welfare,” 80 FR 64530 (October 23, 2015) and emphasized that power plants are “by far the largest emitters” of greenhouse gases among stationary sources in the U.S. Id. at 64522. In American Lung Ass'n v. EPA, 985 F.3d 977 (D.C. Cir. 2021), the court held that even if the EPA were required to determine that CO₂ from fossil fuel-fired EGUs contributes significantly to dangerous air pollution—and the court emphasized that it was not deciding that the EPA was required to make such a pollutant-specific determination—the determination in the alternative that the EPA made in the 2015 NSPS was not arbitrary and capricious and, accordingly, the EPA had a sufficient basis to regulate greenhouse gases from EGUs under CAA section 111(d) in the ACE Rule. This aspect of the decision remains good law. The EPA is not reopening and did not solicit comment on any of those determinations in the 2015 NSPS concerning its rational basis to regulate GHG emissions from EGUs or its alternative finding that GHG emissions from EGUs contribute significantly to dangerous air pollution.

2. 2018 NSPS Proposal To Revise the 2015 NSPS

In 2018, the EPA proposed to revise the NSPS for new, modified, and reconstructed fossil fuel-fired steam generating units and IGCC units, in the Review of Standards of Performance for Greenhouse Gas Emissions From New, Modified, and Reconstructed Stationary Sources: Electric Utility Generating Units; Proposed Rule (83 FR 65424; December 20, 2018) (2018 NSPS Proposal). The EPA proposed to revise the NSPS for newly constructed units, based on a revised BSER of a highly efficient SCPC, without partial CCS. The EPA also proposed to revise the NSPS for modified and reconstructed units. As discussed in IX.A, in the present action, the EPA is withdrawing this proposed rule.

3. Clean Power Plan

With the promulgation of the 2015 NSPS, the EPA also incurred a statutory obligation under CAA section 111(d) to issue emission guidelines for GHG emissions from existing fossil fuel-fired steam generating EGUs and stationary combustion turbine EGUs, which the EPA initially fulfilled with the promulgation of the CPP. See 80 FR 64662 (October 23, 2015). The EPA first determined that the BSER included three types of measures: (1) improving heat rate ( i.e., the amount of fuel that must be burned to generate a unit of electricity) at coal-fired steam plants; (2) substituting increased generation from lower-emitting NGCC plants for generation from higher-emitting steam plants (which are primarily coal-fired); and (3) substituting increased generation from new renewable energy sources for generation from fossil fuel-fired steam plants and combustion turbines. See 80 FR 64667 (October 23, 2015). The latter two measures are known as “generation shifting” because they involve shifting electricity generation from higher-emitting sources to lower-emitting ones. See 80 FR 64728-29 (October 23, 2015).

The EPA based this BSER determination on a technical record that evaluated generation shifting, including its cost-effectiveness, against the relevant statutory criteria for BSER and on a legal interpretation that the term “system” in CAA section 111(a)(1) is sufficiently broad to encompass shifting of generation from higher-emitting to lower-emitting sources. See 80 FR 64720 (October 23, 2015). The EPA then determined the “degree of emission limitation achievable through the application of the [BSER],” CAA section 111(a)(1), expressed as emission performance rates. See 80 FR 64667 (October 23, 2015). The EPA explained that a state would “have to ensure, through its plan, that the emission standards it establishes for its sources individually, in the aggregate, or in combination with other measures undertaken by the state, represent the equivalent of” those performance rates (80 FR 64667; October 23, 2015). Neither states nor sources were required to apply the specific measures identified in the BSER (80 FR 64667; October 23, 2015), and states could include trading or averaging programs in their state plans for compliance. See 80 FR 64840 (October 23, 2015).

Numerous states and private parties petitioned for review of the CPP before the D.C. Circuit. On February 9, 2016, the U.S. Supreme Court stayed the rule pending review, West Virginia v. EPA, 577 U.S. 1126 (2016). The D.C. Circuit held the litigation in abeyance, and ultimately dismissed it at the petitioners' request. American Lung Ass'n, 985 F.3d at 937.

4. The CPP Repeal and ACE Rule

In 2019, the EPA repealed the CPP and replaced it with the ACE Rule. In contrast to its interpretation of CAA section 111 in the CPP, in the ACE Rule the EPA determined that the statutory “text and reasonable inferences from it” make “clear” that a “system” of emission reduction under CAA section 111(a)(1) “is limited to measures that can be applied to and at the level of the individual source,” (84 FR 32529; July 8, 2019); that is, the system must be limited to control measures that could be applied at and to each source to reduce emissions at each source. See 84 FR 32523-24 (July 8, 2019). Specifically, the ACE Rule argued that the requirements in CAA sections 111(d)(1), (a)(3), and (a)(6), that each state establish a standard of performance “for” “any existing source,” defined, in general, as any “building . . . [or] facility,” and the requirement in CAA section 111(a)(1) that the degree of emission limitation must be “achievable” through the “application” of the BSER, by their terms, impose this limitation. The EPA concluded that generation shifting is not such a control measure. See 84 FR 32546 (July 8, 2019). Based on its view that the CPP was a “major rule,” the EPA further determined that, absent “a clear statement from Congress,” the term “ `system of emission reduction' ” should not be read to encompass “generation-shifting measures.” See 84 FR 32529 (July 8, 2019). The EPA acknowledged, however, that “[m]arket-based forces ha[d] already led to significant generation shifting in the power sector,” (84 FR 32532; July 8, 2019), and that there was “likely to be no difference between a world where the CPP is implemented and one where it is not.” See 84 FR 32561 (July 8, 2019); the Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility Generating Units, 2-1 to 2-5.

In addition, the EPA promulgated in the ACE Rule a new set of emission guidelines for existing coal-fired steam-generating EGUs. See 84 FR 32532 (July 8, 2019). In light of “the legal interpretation adopted in the repeal of the CPP,” (84 FR 32532; July 8, 2019)—which “limit[ed] `standards of performance' to systems that can be applied at and to a stationary source,” (84 FR 32534; July 8, 2019)—the EPA found the BSER to be heat rate improvements alone. See 84 FR 32535 (July 8, 2019). The EPA listed various technologies that could improve heat rate (84 FR 32536; July 8, 2019), and identified the “degree of emission limitation achievable” by “providing ranges of expected [emission] reductions associated with each of the technologies.” See 84 FR 32537-38 (July 8, 2019).

5. D.C. Circuit Decision in American Lung Association v. EPA Concerning the CPP Repeal and ACE Rule

Numerous states and private parties petitioned for review of the CPP Repeal and ACE Rule. In 2021, the D.C. Circuit vacated the ACE Rule, including the CPP Repeal. American Lung Ass'n v. EPA, 985 F.3d 914 (D.C. Cir. 2021). The court held, among other things, that CAA section 111(d) does not limit the EPA, in determining the BSER, to measures applied at and to an individual source. The court noted that “the sole ground on which the EPA defends its abandonment of the [CPP] in favor of the ACE Rule is that the text of [CAA section 111] is clear and unambiguous in constraining the EPA to use only improvements at and to existing sources in its [BSER].” 985 F.3d at 944. The court found “nothing in the text, structure, history, or purpose of [CAA section 111] that compels the reading the EPA adopted.” 985 F.3d at 957. The court likewise rejected the view that the CPP's use of generation-shifting implicated a “major question” requiring unambiguous authorization by Congress. 985 F.3d at 958-68.

The D.C. Circuit concluded that, because the EPA had relied on an “erroneous legal premise,” both the CPP Repeal Rule and the ACE Rule should be vacated. 985 F.3d at 995. The court did not decide, however, “whether the approach of the ACE Rule is a permissible reading of the statute as a matter of agency discretion,” 985 F.3d at 944, and instead “remanded to the EPA so that the Agency may `consider the question afresh,' ” 985 F.3d at 995 (citations omitted).

The court also rejected the arguments that the EPA cannot regulate CO₂ emissions from coal-fired power plants under CAA section 111(d) at all because it had already regulated mercury emissions from coal-fired power plants under CAA section 112. 985 F.3d at 988. In addition, the court held that that the 2015 NSPS included a valid determination that greenhouse gases from the EGU source category contributed significantly to dangerous air pollution, which provided a sufficient basis for a CAA section 111(d) rule regulating greenhouse gases from existing fossil fuel-fired EGUs. Id. at 977.

Because the D.C. Circuit vacated the ACE Rule on the grounds noted above, it did not address the other challenges to the ACE Rule, including the arguments by Petitioners that the heat rate improvement BSER was inadequate because of the limited number of reductions it achieved and because the ACE Rule failed to include an appropriately specific degree of emission limitation.

Upon a motion from the EPA, the D.C. Circuit agreed to stay its mandate with respect to vacatur of the CPP Repeal, American Lung Assn v. EPA, No. 19-1140, Order (February 22, 2021), so that the CPP remained repealed. Therefore, following the D.C. Circuit's decision, no EPA rule under CAA section 111 to reduce GHGs from existing fossil fuel-fired EGUs remained in place.

6. U.S. Supreme Court Decision in West Virginia v. EPA Concerning the CPP

The Supreme Court granted petitions for certiorari from the D.C. Circuit's American Lung Association decision, limited to the question of whether CAA section 111 authorized the EPA to determine that “generation shifting” was the best system of emission reduction for fossil-fuel fired EGUs. The Supreme Court did not grant certiorari on the question of whether the EPA was authorized to regulate GHG emissions from fossil-fuel fired power plants under CAA section 111, when fossil-fuel fired power plants are regulated for other pollutants under CAA section 112. In 2022, the U.S. Supreme Court reversed the D.C. Circuit's vacatur of the ACE Rule's embedded repeal of the CPP. West Virginia v. EPA, 597 U.S. 697 (2022). The Supreme Court stated that CAA section 111 authorizes the EPA to determine the BSER and the degree of emission limitation that state plans must achieve. Id. at 2601-02. The Supreme Court concluded, however, that the CPP's BSER of “generation-shifting” raised a “major question,” and was not clearly authorized by section 111. The Court characterized the generation-shifting BSER as “restructuring the Nation's overall mix of electricity generation,” and stated that the EPA's claim that CAA section 111 authorized it to promulgate generation shifting as the BSER was “not only unprecedented; it also effected a fundamental revision of the statute, changing it from one sort of scheme of regulation into an entirely different kind.” Id. at 2612 (internal quotation marks, brackets, and citation omitted). The Court explained that the EPA, in prior rules under CAA section 111, had set emissions limits based on “measures that would reduce pollution by causing the regulated source to operate more cleanly.” Id. at 2610. The Court noted with approval those “more traditional air pollution control measures,” and gave as examples “fuel-switching” and “add-on controls,” which, the Court observed, the EPA had considered in the CPP. Id. at 2611 (internal quotations marks and citation omitted). In contrast, the Court continued, generation shifting was “unprecedented” because “[r]ather than focus on improving the performance of individual sources, it would improve the overall power system by lowering the carbon intensity of power generation. And it would do that by forcing a shift throughout the power grid from one type of energy source to another.” Id. at 2611-12 (internal quotation marks, emphasis, and citation omitted).

The Court recognized that a rule based on traditional measures “may end up causing an incidental loss of coal's market share,” but emphasized that the CPP was “obvious[ly] differen[t]” because, with its generation-shifting BSER, it “simply announc[ed] what the market share of coal, natural gas, wind, and solar must be, and then require[ed] plants to reduce operations or subsidize their competitors to get there.” Id. at 2613 n.4. The Court also emphasized “the magnitude and consequence” of the CPP. Id. at 2616. It noted “the magnitude of this unprecedented power over American industry,” id. at 2612 (internal quotation marks and citation omitted), and added that the EPA's adoption of generation shifting “represent[ed] a transformative expansion in its regulatory authority.” Id. at 2610 (internal quotation marks and citation omitted). The Court also viewed the CPP as promulgating “a program that . . . Congress had considered and rejected multiple times.” Id. at 2614 (internal quotation marks and citation omitted). For these and related reasons, the Court viewed the CPP as raising a major question, and therefore, requiring “clear congressional authorization” as a basis. Id. (internal quotation marks and citation omitted).

The Court declined to address the D.C. Circuit's conclusion that the text of CAA section 111 did not limit the type of “system” the EPA could consider as the BSER to measures applied at and to an individual source. See id. at 2615. Nor did the Court address the scope of the states' compliance flexibilities.

7. D.C. Circuit Order Reinstating the ACE Rule

On October 27, 2022, the D.C. Circuit responded to the U.S. Supreme Court's reversal by recalling its mandate for the vacatur of the ACE Rule. American Lung Ass'n v. EPA, No. 19-1140, Order (October 27, 2022). Accordingly, at that time, the ACE Rule came back into effect. The court also revised its judgment to deny petitions for review challenging the CPP Repeal Rule, consistent with the judgment in West Virginia, so that the CPP remains repealed. The court took further action denying several of the petitions for review unaffected by the Supreme Court's decision in West Virginia, which means that certain parts of its 2021 decision in American Lung Association remain in effect. These parts include the holding that the EPA's prior regulation of mercury emissions from coal-fired electric power plants under CAA section 112 does not preclude the Agency from regulating CO₂ from coal-fired electric power plants under CAA section 111, and the holding, discussed above, that the 2015 NSPS included a valid significant contribution determination and therefore provided a sufficient basis for a CAA section 111(d) rule regulating greenhouse gases from existing fossil fuel-fired EGUs. The court's holding to invalidate amendments to the implementing regulations applicable to emission guidelines under CAA section 111(d) that extended the preexisting schedules for state and Federal actions and sources' compliance, also remains in force. Based on the EPA's stated intention to replace the ACE Rule, the court stayed further proceedings with respect to the ACE Rule, including the various challenges that its BSER was flawed because it did not achieve sufficient emission reductions and failed to specify an appropriately specific degree of emission limitation.

C. Detailed Discussion of CAA Section 111 Requirements

This section discusses in more detail the key requirements of CAA section 111 for both new and existing sources that are relevant for these rulemakings.

1. Approach to the Source Category and Subcategorizing

CAA section 111 requires the EPA first to list stationary source categories that cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare and then to regulate new sources within each such source category. CAA section 111(b)(2) grants the EPA discretion whether to “distinguish among classes, types, and sizes within categories of new sources for the purpose of establishing [new source] standards,” which we refer to as “subcategorizing.” Whether and how to subcategorize is a decision for which the EPA is entitled to a “high degree of deference” because it entails “scientific judgment.” Lignite Energy Council v. EPA, 198 F.3d 930, 933 (D.C. Cir. 1999).

Although CAA section 111(d)(1) does not explicitly address subcategorization, since its first regulations implementing the CAA, the EPA has interpreted it to authorize the Agency to exercise discretion as to whether and, if so, how to subcategorize, for the following reasons. CAA section 111(d)(1) grants the EPA authority to “prescribe regulations which shall establish a procedure . . . under which each State shall submit to the Administrator a plan [with standards of performance for existing sources.]” The EPA promulgates emission guidelines under this provision directing the states to regulate existing sources. The Supreme Court has recognized that, under CAA section 111(d), the “Agency, not the States, decides the amount of pollution reduction that must ultimately be achieved. It does so by again determining, as when setting the new source rules, `the best system of emission reduction . . . that has been adequately demonstrated for [existing covered] facilities.' West Virginia, 597 U.S. at 710 (citations omitted).

The EPA's authority to determine the BSER includes the authority to create subcategories that tailor the BSER for differently situated sets of sources. Again, for new sources, CAA section 111(b)(2) confers authority for the EPA to “distinguish among classes, types, and sizes within categories.” Though CAA section 111(d) does not speak specifically to the creation of subcategories for a category of existing sources, the authority to identify the “best” system of emission reduction for existing sources includes the discretion to differentiate between differently situated sources in the category, and group those sources into subcategories in appropriate circumstances. The size, type, class, and other characteristics can make different emission controls more appropriate for different sources. A system of emission reduction that is “best” for some sources may not be “best” for others with different characteristics. For more than four decades, the EPA has interpreted CAA section 111(d) to confer authority on the Agency to create subcategories. The EPA's implementing regulations under CAA section 111(d), promulgated in 1975, 40 FR 53340 (November 17, 1975), provide that the Administrator will specify different emission guidelines or compliance times or both “for different sizes, types, and classes of designated facilities when [based on] costs of control, physical limitations, geographical location, or [based on] similar factors.” This regulation governs the EPA's general authority to subcategorize under CAA section 111(d), and the EPA is not reopening that issue here. At the time of promulgation, the EPA explained that subcategorization allows the EPA to take into account “differences in sizes and types of facilities and similar considerations, including differences in control costs that may be involved for sources located in different parts of the country” so that the “EPA's emission guidelines will in effect be tailored to what is reasonably achievable by particular classes of existing sources. . . .” Id. at 53343. The EPA's authority to “distinguish among classes, types, and sizes within categories,” as provided under CAA section 111(b)(2), generally allows the Agency to place types of sources into subcategories. This is consistent with the commonly understood meaning of the term “type” in CAA section 111(b)(2): “a particular kind, class, or group,” or “qualities common to a number of individuals that distinguish them as an identifiable class.” See https://www.merriam-webster.com/dictionary/type .

The EPA has developed subcategories in many rulemakings under CAA section 111 since the 1970s. These rulemakings have included subcategories on the basis of the size of the sources, see 40 CFR 60.40b(b)(1)-(2) (subcategorizing certain coal-fired steam generating units on the basis of heat input capacity); the types of fuel combusted, see Sierra Club, v. EPA, 657 F.2d 298, 318-19 (D.C. Cir. 1981) (upholding a rulemaking that established different NSPS “for utility plants that burn coal of varying sulfur content”), 2015 NSPS, 80 FR 64510, 64602 (table 15) (October 23, 2015) (subdividing new combustion turbines on the basis of type of fuel combusted); the types of equipment used to produce products, see 81 FR 35824 (June 3, 2016) (promulgating separate NSPS for many types of oil and gas sources, such as centrifugal compressors, pneumatic controllers, and well sites); types of manufacturing processes used to produce product, see 42 FR 12022 (March 1, 1977) (announcing availability of final guideline document for control of atmospheric fluoride emissions from existing phosphate fertilizer plants) and “Final Guideline Document: Control of Fluoride Emissions From Existing Phosphate Fertilizer Plants,” EPA-450/2-77-005 1-7 to 1-9, including table 1-2 (applying different control requirements for different manufacturing operations for phosphate fertilizer); levels of utilization of the sources, see 2015 NSPS, 80 FR 64510, 64602 (table 15) (October 23, 2015) (dividing new natural gas-fired combustion turbines into the subcategories of base load and non-base load); the activity level of the sources, see 81 FR 59276, 59278-79 (August 29, 2016) (dividing municipal solid waste landfills into the subcategories of active and closed landfills); and geographic location of the sources, see 71 FR 38482 (July 6, 2006) (SO₂ NSPS for stationary combustion turbines subcategorizing turbines on the basis of whether they are located in, for example, a continental area, a non-continental area, the part of Alaska north of the Arctic Circle, and the rest of Alaska). Thus, the EPA has subcategorized many times in rulemaking under CAA sections 111(b) and 111(d) and based on a wide variety of physical, locational, and operational characteristics.

Regardless of whether the EPA subcategorizes within a source category for purposes of determining the BSER and the degree of emission limitation achievable, a state retains certain flexibility in assigning standards of performance to its affected EGUs. The statutory framework for CAA section 111(d) emission guidelines, and the flexibilities available to states within that framework, are discussed below.

2. Key Elements of Determining a Standard of Performance

Congress first included the definition of “standard of performance” when enacting CAA section 111 in the 1970 Clean Air Act Amendments (CAAA), amended it in the 1977 CAAA, and then amended it again in the 1990 CAAA to largely restore the definition as it read in the 1970 CAAA. The current text of CAA section 111(a)(1) reads: “The term `standard of performance' means a standard for emission of air pollutants which reflects the degree of emission limitation achievable through the application of the best system of emission reduction which (taking into account the cost of achieving such reduction and any non-air quality health and environmental impact and energy requirements) the Administrator determines has been adequately demonstrated.” The D.C. Circuit has reviewed CAA section 111 rulemakings on numerous occasions since 1973, ¹⁸⁸ and has developed a body of caselaw that interprets the term “standard of performance,” as discussed throughout this preamble.

The basis for standards of performance, whether promulgated by the EPA under CAA section 111(b) or established by the states under CAA section 111(d), is that the EPA determines the “degree of emission limitation” that is “achievable” by the sources by application of a “system of emission reduction” that the EPA determines is “adequately demonstrated,” “taking into account” the factors of “cost . . . and any nonair quality health and environmental impact and energy requirements,” and that the EPA determines to be the “best.” The D.C. Circuit has stated that in determining the “best” system, the EPA must also take into account “the amount of air pollution” reduced and the role of “technological innovation.” The D.C. Circuit has also stated that to determine the “best” system, the EPA may weigh the various factors identified in the statute and caselaw against each other, and has emphasized that the EPA has discretion in weighing the factors.

The EPA's overall approach to determining the BSER and degree of emission limitation achievable, which incorporates the various elements, is as follows: The EPA identifies “system[s] of emission reduction” that have been “adequately demonstrated” for a particular source category and determines the “best” of these systems after evaluating the amount of emission reductions, costs, any non-air health and environmental impacts, and energy requirements. As discussed below, for each of numerous subcategories, the EPA followed this approach to determine the BSER on the basis that the identified costs are reasonable and that the BSER is rational in light of the statutory factors, including the amount of emission reductions, that the EPA examined in its BSER analysis, consistent with governing precedent.

After determining the BSER, the EPA determines an achievable emission limit based on application of the BSER. For a CAA section 111(b) rule, the EPA determines the standard of performance that reflects the achievable emission limit. For a CAA section 111(d) rule, the states have the obligation of establishing standards of performance for the affected sources that reflect the degree of emission limitation that the EPA has determined. As discussed below, the EPA is finalizing these determinations in association with each of the BSER determinations.

The remainder of this subsection discusses each element in our general analytical approach.

a. System of Emission Reduction

The CAA does not define the phrase “system of emission reduction.” In West Virginia v. EPA, the Supreme Court recognized that historically, the EPA had looked to “measures that improve the pollution performance of individual sources and followed a “technology-based approach” in identifying systems of emission reduction. In particular, the Court identified “the sort of `systems of emission reduction' [the EPA] had always before selected,” which included “ ‘efficiency improvements, fuel-switching,' and `add-on controls'.” 597 U.S. at 727 (quoting the Clean Power Plan). Section 111 itself recognizes that such systems may include off-site activities that may reduce a source's pollution contribution, identifying “precombustion cleaning or treatment of fuels” as a “system” of “emission reduction.” 42 U.S.C. 7411(a)(7)(B). A “system of emission reduction” thus, at a minimum, includes measures that an individual source applies that improve the emissions performance of that source. Measures are fairly characterized as improving the pollution performance of a source where they reduce the individual source's overall contribution to pollution.

In West Virginia, the Supreme Court did not define the term “system of emissions reduction,” and so did not rule on whether “system of emission reduction” is limited to those measures that the EPA has historically relied upon. It did go on to apply the major questions doctrine to hold that the term “system” does not provide the requisite clear authorization to support the Clean Power Plan's BSER, which the Court described as “carbon emissions caps based on a generation shifting approach.” Id. at 2614. While the Court did not define the outer bounds of the meaning of “system,” systems of emissions reduction like fuel switching, add-on controls, and efficiency improvements fall comfortably within the scope of prior practice as recognized by the Supreme Court.

b. “Adequately Demonstrated”

Under CAA section 111(a)(1), an essential, although not sufficient, condition for a “system of emission reduction” to serve as the basis for an “achievable” emission standard is that the Administrator must determine that the system is “adequately demonstrated.” The concepts of adequate demonstration and achievability are closely related: as the D.C. Circuit has stated, “[i]t is the system which must be adequately demonstrated and the standard which must be achievable,” through application of the system. An achievable standard means a standard based on the EPA's record-based finding that sufficient evidence exists to reasonably determine that the affected sources in the source category can adopt a specific system of emission reduction to achieve the specified degree of emission limitation. As discussed below, consistent with Congress's use of the word “demonstrated,” the caselaw has approved the EPA's “adequately demonstrated” determinations concerning systems utilized at test sources or other individual sources operating at commercial scale. The case law also authorizes the EPA to set an emissions standard at levels more stringent than has regularly been achieved, based on the understanding that sources will be able to adopt specific technological improvements to the system in question that will enable them to achieve the lower standard. Importantly, and contrary to some comments received on the proposed rule, CAA section 111(a)(1) does not require that a system of emission reduction exist in widespread commercial use in order to satisfy the “adequately demonstrated” requirement. Instead, CAA section 111(a)(1) authorizes the EPA to establish standards which encourage the deployment of more effective systems of emission reduction that have been adequately demonstrated but that are not yet in widespread use. This aligns with Congress's purpose in enacting the CAA, in particular its recognition that polluting sources were not widely adopting emission control technology on a voluntary basis and that Federal regulation was necessary to spur the development and deployment of those technologies.

i. Plain Text, Statutory Context, and Legislative History of the “Adequately Demonstrated” Provision in CAA Section 111(a)(1)

Analysis of the plain text, statutory context, and legislative history of CAA section 111(a)(1) establishes two primary themes. First, Congress assigned the task of determining the appropriate BSER to the Administrator, based on a reasonable review of available evidence. Second, Congress authorized the EPA to set a standard, based on the evidence, that encourages broader adoption of an emissions-reducing technological approach that may not yet be in widespread use.

The plain text of CAA section 111(a)(1), and in particular the phrase “the Administrator determines” and the term “adequately,” confer discretion to the EPA in identifying the appropriate system. Rather than providing specific criteria for determining what constitutes appropriate evidence, Congress directed the Administrator to “determine[ ]” that the demonstration is “adequate[ ].” Courts have typically deferred to the EPA's scientific and technological judgments in making such determinations. Further, use of the term “adequate” in provisions throughout the CAA highlights EPA flexibility and discretion in setting standards and in analyzing data that forms the basis for standard setting.

In setting NAAQS under CAA section 109, for example, the EPA is directed to determine, according to “the judgment of the Administrator,” an “adequate margin of safety.” The D.C. Circuit has held that the use of the term “adequate” confers significant deference to the Administrator's scientific and technological judgment. In Mississippi v. EPA, for example, the D.C. Circuit in 2013 upheld the EPA's choice to set the NAAQS for ozone below 0.08 ppm, and noted that any disagreements with the EPA's interpretations of the scientific evidence that underlay this decision “must come from those who are qualified to evaluate the science, not [the court].” This Mississippi v. EPA precedent aligns with the general standard for judicial review of the EPA's understanding of the evidence under CAA section 307(d)(9)(A) (“arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law”).

The plain language of the phrase “has been adequately demonstrated,” in context, and in light of the legislative history, further strongly indicates that the system in question need not be in widespread use at the time the EPA's rule is published. To the contrary, CAA section 111(a)(1) authorizes technology forcing, in the sense that the EPA is authorized to promote a system which is not yet in widespread use; provided the technology is in existence and the EPA has adequate evidence to extrapolate.

Some commenters argued that use of the phrase “has been” in “has been adequately demonstrated” means that the system must be in widespread commercial use at the time of rule promulgation. We disagree. Considering the plain text, the use of the past tense, “ has been adequately demonstrated” indicates a requirement that the technology currently be demonstrated. However, “demonstrated” in common usage at the time of enactment meant to “explain or make clear by using examples, experiments, etc. ” As a general matter, and as this definition indicates, the term “to demonstrate” suggests the need for a test or study—as in, for example, a “demonstration project” or “demonstration plant”—that is, examples of technological feasibility.

The statutory context is also useful in establishing that where Congress wanted to specify the availability of the control system, it did so. The only other use of the exact term “adequately demonstrated” occurs in CAA section 119, which establishes that, in order for the EPA to require a particular “means of emission limitation” for smelters, the Agency must establish that such means “has been adequately demonstrated to be reasonably available. . . .” The lack of the phrase “reasonably available” in CAA section 111(a)(1) is notable, and suggests that a system may be “adequately demonstrated” under CAA section 111 even if it is not “reasonably available” for every single source.

The term “demonstration” also appears in CAA section 103 in an instructive context. CAA section 103, which establishes a “national research and development program for the prevention and control of air pollution” directs that as part of this program, the EPA shall “conduct, and promote the coordination and acceleration of, research, investigations, experiments, demonstrations, surveys, and studies relating to” the issue of air pollution. According to the canon of noscitur a sociis, associated words in a list bear on one another's meaning. In CAA section 103, the word “demonstrations” appears alongside “research,” “investigations,” “experiments,” and “studies”—all words suggesting the development of new and emerging technology. This supports interpreting CAA section 111(a)(1) to authorize the EPA to determine a system of emission reduction to be “adequately demonstrated” based on demonstration projects, testing, examples, or comparable evidence.

Finally, the legislative history of the CAA in general, and section 111 in particular, strongly supports the point that BSER technology need not be in widespread use at the time of rule enactment. The final language of CAA section 111(a)(1), requiring that systems of emission reduction be “adequately demonstrated,” was the result of compromise in the Conference Committee between the House and Senate bill language. The House bill would have required that the EPA give “appropriate consideration to technological and economic feasibility” when establishing standards. The Senate bill would have required that standards “reflect the greatest degree of emission control which the Secretary determines to be achievable through application of the latest available control technology, processes, operating methods, or other alternatives.” Although the exact language of neither the House nor Senate bill was adopted in the final bill, both reports made clear their intent that CAA section 111 would be significantly technology-forcing. In particular, the Senate Report referred to “available control technology”—a phrase that, as just noted, the Senate bill included—but clarified that the technology need not “be in actual, routine use somewhere.” The House Report explained that EPA regulations would “prevent and control such emissions to the fullest extent compatible with the available technology and economic feasibility as determined by [the EPA],” and “[i]n order to be considered `available' the technology may not be one which constitutes a purely theoretical or experimental means of preventing or controlling air pollution.” This last statement implies that the House Report anticipated that the EPA's determination may be technology forcing. Nothing in the legislative history suggests that Congress intended that the technology already be in widespread commercial use.

ii. Caselaw

In a series of cases reviewing standards for new sources, the D.C. Circuit has held that an adequately demonstrated standard of performance may reflect the EPA's reasonable projection of what that particular system may be expected to achieve going forward, extrapolating from available data from pilot projects or individual commercial-scale sources. A standard may be considered achievable even if the system upon which the standard is based has not regularly achieved the standard in testing. See, e.g., Essex Chem. Corp. v. Ruckelshaus (upholding a standard of 4.0 lbs per ton based on a system whose average control rate was 4.6 lbs per ton, and which had achieved 4.0 lbs per ton on only three occasions and “`nearly equaled' [the standard] on the average of nineteen different readings.”) The Ruckelshaus court concluded that the EPA's extrapolation from available data was “the result of the exercise of reasoned discretion by the Administrator” and therefore “[could not] be upset by [the] court.” The court also emphasized that in order to be considered achievable, the standard set by the EPA need not be regularly or even specifically achieved at the time of rule promulgation. Instead, according to the court, “[a]n achievable standard is one which is within the realm of the adequately demonstrated system's efficiency and which, while not at a level that is purely theoretical or experimental, need not necessarily be routinely achieved within the industry prior to its adoption.”

Case law also establishes that the EPA may set a standard more stringent than has regularly been achieved based on its identification of specific available technological improvements to the system. See Sierra Club v. Costle (upholding a 90 percent standard for SO₂ emissions from coal-fired steam generators despite the fact that not all plants had previously achieved this standard, based on the EPA's expectations for improved performance with specific technological fixes and the use of “coal washing” going forward). Further, the EPA may extrapolate based on testing at a particular kind of source to conclude that the technology at issue will also be effective at a different, related, source. See Lignite Energy Council v. EPA (holding it permissible to base a standard for industrial boilers on application of SCR based on extrapolated information about the application of SCR on utility boilers). The Lignite court clarified that “where data are unavailable, EPA may not base its determination that a technology is adequately demonstrated or that a standard is achievable on mere speculation or conjecture,” but the “EPA may compensate for a shortage of data through the use of other qualitative methods, including the reasonable extrapolation of a technology's performance in other industries.”

As a general matter, the case law is clear that at the time of Rule promulgation, the system which the EPA establishes as BSER need not be in widespread use. See, e.g., Ruckelshaus (upholding a standard based on a relatively new system which was in use at only one United States plant at the time of rule promulgation. Although the system was in use more extensively in Europe at the time of rule promulgation, the EPA based its analysis on test results from the lone U.S. plant only.) This makes good sense, because, as discussed above, CAA section 111(a)(1) authorizes a technology-forcing standard that encourages broader adoption of an emissions-reducing technological approach that is not yet broadly used. It follows that at the time of promulgation, not every source will be prepared to adopt the BSER at once. Instead, as discussed next, the EPA's responsibility is to determine that the technology can be adopted in a reasonable period of time, and to base its requirements on this understanding.

iii. Compliance Timeframe

The preceding subsections have shown various circumstances under which the EPA may determine that a system of emission reduction is “adequately demonstrated.” In order to establish that a system is appropriate for the source category as a whole, the EPA must also demonstrate that the industry can deploy the technology at scale in the compliance timeframe. The D.C. Circuit has stated that the EPA may determine a “system of emission reduction” to be “adequately demonstrated” if the EPA reasonably projects that it may be more broadly deployed with adequate lead time. This view is well-grounded in the purposes of CAA section 111(a)(1), discussed above, which aim to control dangerous air pollution by allowing for standards which encourage more widespread adoption of a technology demonstrated at individual plants.

As a practical matter, CAA section 111's allowance for lead time recognizes that existing pollution control systems may be complex and may require a predictable amount of time for sources across the source category to be able to design, acquire, install, test, and begin to operate them. Time may also be required to allow for the development of skilled labor, and materials like steel, concrete, and speciality parts. Accordingly, in setting 111 standards for both new and existing sources, the EPA has typically allowed for some amount of time before sources must demonstrate compliance with the standards. For instance, in the 2015 NSPS for residential wood heaters, the EPA established a “stepped compliance approach” which phased in requirements over 5 years to “allow manufacturers lead time to develop, test, field evaluate and certify current technologies” across their model lines. The EPA also allowed for a series of phase-ins of various requirements in the 2023 oil and gas NSPS. For example: the EPA finalized a compliance deadline for process controllers allowing for 1 year from the effective date of the final rule, to allow for delays in equipment availability; the EPA established a 1-year lead time period for pumps, also in response to possible equipment and labor shortages; and the EPA built in 24 months between publication in the Federal Register and the commencement of a requirement to end routine flaring and route associated gas to a sales line.

Finally, the EPA's longstanding regulations for new source performance standards under CAA section 111 specifically authorize a minimum period for lead time. Pursuant to 40 CFR 60.11, compliance with CAA section 111 standards is generally determined in accordance with performance tests conducted under 40 CFR 60.8. Both of these regulatory provisions were adopted in 1971. Under 40 CFR 60.8, source performance is generally measured via performance tests, which must typically be carried out “within 60 days after achieving the maximum production rate at which the affected facility will be operated, but not later than 180 days after initial startup of such facility, or at such other times specified by this part, and at such other times as may be required by the Administrator under section 114 of the Act. . . .” The fact that this provision has been in place for over 50 years indicates that the EPA has long recognized the need for lead time for at least one component of control development.

c. Costs

Under CAA section 111(a)(1), in determining whether a particular emission control is the “best system of emission reduction . . . adequately demonstrated,” the EPA is required to take into account “the cost of achieving [the emission] reduction.” Although the CAA does not describe how the EPA is to account for costs to affected sources, the D.C. Circuit has formulated the cost standard in various ways, including stating that the EPA may not adopt a standard the cost of which would be “excessive” or “unreasonable.”

The EPA has discretion in its consideration of cost under section 111(a), both in determining the appropriate level of costs and in balancing costs with other BSER factors. To determine the BSER, the EPA must weigh the relevant factors, including the cost of controls and the amount of emission reductions, as well as other factors.

The D.C. Circuit has repeatedly upheld the EPA's consideration of cost in reviewing standards of performance. In several cases, the court upheld standards that entailed significant costs, consistent with Congress's view that “the costs of applying best practicable control technology be considered by the owner of a large new source of pollution as a normal and proper expense of doing business.” See Essex Chemical Corp. v. Ruckelshaus, 486 F.2d 427, 440 (D.C. Cir. 1973); Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 387-88 (D.C. Cir. 1973); Sierra Club v. Costle, 657 F.2d 298, 313 (D.C. Cir. 1981) (upholding NSPS imposing controls on SO₂ emissions from coal-fired power plants when the “cost of the new controls . . . is substantial. The EPA estimates that utilities will have to spend tens of billions of dollars by 1995 on pollution control under the new NSPS.”).

In its CAA section 111 rulemakings, the EPA has frequently used a cost-effectiveness metric, which determines the cost in dollars for each ton or other quantity of the regulated air pollutant removed through the system of emission reduction. See, e.g.,81 FR 35824 (June 3, 2016) (NSPS for GHG and VOC emissions for the oil and natural gas source category); 71 FR 9866, 9870 (February 27, 2006) (NSPS for NO_X, SO₂, and PM emissions from fossil fuel-fired electric utility steam generating units); 61 FR 9905, 9910 (March 12, 1996) (NSPS and emission guidelines for nonmethane organic compounds and landfill gas from new and existing municipal solid waste landfills); 50 FR 40158 (October 1, 1985) (NSPS for SO₂ emissions from sweetening and sulfur recovery units in natural gas processing plants). This metric allows the EPA to compare the amount a regulation would require sources to pay to reduce a particular pollutant across regulations and industries. In rules for the electric power sector, the EPA has also looked at a metric that determines the dollar increase in the cost of a MWh of electricity generated by the affected sources due to the emission controls, which shows the cost of controls relative to the output of electricity. See section VII.C.1.a.ii of this preamble, which discusses $/MWh costs of the Good Neighbor Plan for the 2015 Ozone NAAQS (88 FR 36654; June 5, 2023) and the Cross-State Air Pollution Rule (CSAPR) (76 FR 48208; August 8, 2011). This metric facilitates comparing costs across regulations and pollutants. In these final actions, as explained herein, the EPA looks at both of these metrics, in addition to other cost evaluations, to assess the cost reasonableness of the final requirements. The EPA's consideration of cost reasonableness in this way meets the statutory requirement that the EPA take into account “the cost of achieving [the emission] reduction” under section 111(a)(1).

d. Non-Air Quality Health and Environmental Impact and Energy Requirements

Under CAA section 111(a)(1), the EPA is required to take into account “any nonair quality health and environmental impact and energy requirements” in determining the BSER. Non-air quality health and environmental impacts may include the impacts of the disposal of byproducts of the air pollution controls, or requirements of the air pollution control equipment for water. Portland Cement Ass'n v. Ruckelshaus, 465 F.2d 375, 387-88 (D.C. Cir. 1973), cert. denied, 417 U.S. 921 (1974). Energy requirements may include the impact, if any, of the air pollution controls on the source's own energy needs.

e. Sector or Nationwide Component of Factors in Determining the BSER

Another component of the D.C. Circuit's interpretations of CAA section 111 is that the EPA may consider the various factors it is required to consider on a national or regional level and over time, and not only on a plant-specific level at the time of the rulemaking. The D.C. Circuit based this interpretation—which it made in the 1981 Sierra Club v. Costle case regarding the NSPS for new power plants—on a review of the legislative history, stating,

[T]he Reports from both Houses on the Senate and House bills illustrate very clearly that Congress itself was using a long-term lens with a broad focus on future costs, environmental and energy effects of different technological systems when it discussed section 111.

The court has upheld EPA rules that the EPA “justified . . . in terms of the policies of the Act,” including balancing long-term national and regional impacts. For example, the court upheld a standard of performance for SO₂ emissions from new coal-fired power plants on grounds that it—

reflects a balance in environmental, economic, and energy consideration by being sufficiently stringent to bring about substantial reductions in SO₂ emissions (3 million tons in 1995) yet does so at reasonable costs without significant energy penalties. . . .

The EPA interprets this caselaw to authorize it to assess the impacts of the controls it is considering as the BSER, including their costs and implications for the energy system, on a sector-wide, regional, or national basis, as appropriate. For example, the EPA may assess whether controls it is considering would create risks to the reliability of the electricity system in a particular area or nationwide and, if they would, to reject those controls as the BSER.

f. “Best”

In determining which adequately demonstrated system of emission reduction is the “best,” the EPA has broad discretion. In AEP v. Connecticut, 564 U.S. 410, 427 (2011), the Supreme Court explained that under CAA section 111, “[t]he appropriate amount of regulation in any particular greenhouse gas-producing sector cannot be prescribed in a vacuum: . . . informed assessment of competing interests is required. Along with the environmental benefit potentially achievable, our Nation's energy needs and the possibility of economic disruption must weigh in the balance. The Clean Air Act entrusts such complex balancing to the EPA in the first instance, in combination with state regulators. Each “standard of performance” the EPA sets must “tak[e] into account the cost of achieving [emissions] reduction and any nonair quality health and environmental impact and energy requirements.” (paragraphing revised; citations omitted)).

Likewise, in Sierra Club v. Costle, 657 F.2d 298 (D.C. Cir. 1981), the court explained that “section 111(a) explicitly instructs the EPA to balance multiple concerns when promulgating a NSPS,” and emphasized that “[t]he text gives the EPA broad discretion to weigh different factors in setting the standard,” including the amount of emission reductions, the cost of the controls, and the non-air quality environmental impacts and energy requirements. And in Lignite Energy Council v. EPA, 198 F.3d 930 (D.C. Cir. 1999), the court reiterated:

Because section 111 does not set forth the weight that should be assigned to each of these factors, we have granted the agency a great degree of discretion in balancing them . . . . EPA's choice [of the `best system'] will be sustained unless the environmental or economic costs of using the technology are exorbitant . . . . EPA [has] considerable discretion under section 111.

Importantly, the courts recognize that the EPA must consider several factors and that determining what is “best” depends on how much weight to give the factors. In promulgating certain standards of performance, the EPA may give greater weight to particular factors than it does in promulgating other standards of performance. Thus, the determination of what is “best” is complex and necessarily requires an exercise of judgment. By analogy, the question of who is the “best” sprinter in the 100-meter dash primarily depends on only one criterion—speed—and therefore is relatively straightforward, whereas the question of who is the “best” baseball player depends on a more complex weighing of multiple criteria and therefore requires a greater exercise of judgment.

The term “best” also authorizes the EPA to consider factors in addition to the ones enumerated in CAA section 111(a)(1), that further the purpose of the statute. In Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375 (D.C. Cir. 1973), the D.C. Circuit held that under CAA section 111(a)(1) as it read prior to the enactment of the 1977 CAA Amendments that added a requirement that the EPA take account of non-air quality environmental impacts, the EPA must consider “counter-productive environmental effects” in Determining the BSER. Id. at 385. The court elaborated: “The standard of the `best system' is comprehensive, and we cannot imagine that Congress intended that `best' could apply to a system which did more damage to water than it prevented to air.” Id., n.42. In Sierra Club v. Costle, 657 F.2d at 326, 346-47, the court added that the EPA must consider the amount of emission reductions and technology advancement in determining BSER, as discussed in section V.C.2.g of this preamble.

The court's view that “best” includes additional factors that further the purpose of CAA section 111 is a reasonable interpretation of that term in its statutory context. The purpose of CAA section 111 is to reduce emissions of air pollutants that endanger public health or welfare. CAA section 111(b)(1)(A). The court reasonably surmised that the EPA's determination of whether a system of emission reduction that reduced certain air pollutants is “best” should be informed by impacts that the system may have on other pollutants that affect public or welfare. Portland Cement Ass'n, 486 F.2d at 385. The Supreme Court confirmed the D.C. Circuit's approach in Michigan v. EPA, 576 U.S. 743 (2015), explaining that administrative agencies must engage in “reasoned decisionmaking” that, in the case of pollution control, cannot be based on technologies that “do even more damage to human health” than the emissions they eliminate. Id. at 751-52. After Portland Cement Ass'n, Congress revised CAA section 111(a)(1) to make explicit that in determining whether a system of emission reduction is the “best,” the EPA should account for non-air quality health and environmental impacts. By the same token, the EPA takes the position that in determining whether a system of emission reduction is the “best,” the EPA may account for the impacts of the system on air pollutants other than the ones that are the subject of the CAA section 111 regulation. We discuss immediately below other factors that the D.C. Circuit has held the EPA should account for in determining what system is the “best.”

g. Amount of Emissions Reductions

Consideration of the amount of emissions from the category of sources or the amount of emission reductions achieved as factors the EPA must consider in determining the “best system of emission reduction” is implicit in the plain language of CAA section 111(a)(1)—the EPA must choose the best system of emission reduction. Indeed, consistent with this plain language and the purpose of CAA section 111, the EPA must consider the quantity of emissions at issue. See Sierra Club v. Costle, 657 F.2d 298, 326 (D.C. Cir. 1981) (“we can think of no sensible interpretation of the statutory words “best . . . system” which would not incorporate the amount of air pollution as a relevant factor to be weighed when determining the optimal standard for controlling . . . emissions”). The fact that the purpose of a “system of emission reduction” is to reduce emissions, and that the term itself explicitly incorporates the concept of reducing emissions, supports the court's view that in determining whether a “system of emission reduction” is the “best,” the EPA must consider the amount of emission reductions that the system would yield. Even if the EPA were not required to consider the amount of emission reductions, the EPA has the discretion to do so, on grounds that either the term “system of emission reduction” or the term “best” may reasonably be read to allow that discretion.

h. Expanded Use and Development of Technology

The D.C. Circuit has long held that Congress intended for CAA section 111 to create incentives for new technology and therefore that the EPA is required to consider technological innovation as one of the factors in determining the “best system of emission reduction.” See Sierra Club v. Costle, 657 F.2d at 346-47. The court has grounded its reading in the statutory text of CAA 111(a)(1), defining the term “standard of performance.” In addition, the court's interpretation finds support in the legislative history. The legislative history identifies three different ways that Congress designed CAA section 111 to authorize standards of performance that promote technological improvement: (1) The development of technology that may be treated as the “best system of emission reduction . . . adequately demonstrated;” under CAA section 111(a)(1); (2) the expanded use of the best demonstrated technology; and (3) the development of emerging technology. Even if the EPA were not required to consider technological innovation as part of its determination of the BSER, it would be reasonable for the EPA to consider it because technological innovation may be considered an element of the term “best,” particularly in light of Congress's emphasis on technological innovation.

i. Achievability of the Degree of Emission Limitation

For new sources, CAA section 111(b)(1)(B) and (a)(1) provides that the EPA must establish “standards of performance,” which are standards for emissions that reflect the degree of emission limitation that is “achievable” through the application of the BSER. A standard of performance is “achievable” if a technology can reasonably be projected to be available to an individual source at the time it is constructed that will allow it to meet the standard. Moreover, according to the court, “[a]n achievable standard is one which is within the realm of the adequately demonstrated system's efficiency and which, while not at a level that is purely theoretical or experimental, need not necessarily be routinely achieved within the industry prior to its adoption.” To be achievable, a standard “must be capable of being met under most adverse conditions which can reasonably be expected to recur and which are not or cannot be taken into account in determining the ‘costs’ of compliance.” To show a standard is achievable, the EPA must “(1) identify variable conditions that might contribute to the amount of expected emissions, and (2) establish that the test data relied on by the agency are representative of potential industry-wide performance, given the range of variables that affect the achievability of the standard.”

Although the courts have established these standards for achievability in cases concerning CAA section 111(b) new source standards of performance, generally comparable standards for achievability should apply under CAA section 111(d), although the BSER may differ in some cases as between new and existing sources due to, for example, higher costs of retrofit. 40 FR 53340 (November 17, 1975). For existing sources, CAA section 111(d)(1) requires the EPA to establish requirements for state plans that, in turn, must include “standards of performance.” As the Supreme Court has recognized, this provision requires the EPA to promulgate emission guidelines that determine the BSER for a source category and then identify the degree of emission limitation achievable by application of the BSER. See West Virginia v. EPA, 597 U.S. at 710.

The EPA has promulgated emission guidelines on the basis that the existing sources can achieve the degree of emission limitation described therein, even though under the RULOF provision of CAA section 111(d)(1), the state retains discretion to apply standards of performance to individual sources that are less stringent, which indicates that Congress recognized that the EPA may promulgate emission guidelines that are consistent with CAA section 111(d) even though certain individual sources may not be able to achieve the degree of emission limitation identified therein by applying the controls that the EPA determined to be the BSER. Note further that this requirement that the emission limitation be “achievable” based on the “best system of emission reduction . . . adequately demonstrated” indicates that the technology or other measures that the EPA identifies as the BSER must be technically feasible.

3. EPA Promulgation of Emission Guidelines for States To Establish Standards of Performance

CAA section 111(d)(1) directs the EPA to promulgate regulations establishing a procedure similar to that provided by CAA section 110 under which states submit state plans that establish “standards of performance” for emissions of certain air pollutants from sources which, if they were new sources, would be regulated under CAA section 111(b), and that provide for the implementation and enforcement of such standards of performance. The term “standard of performance” is defined under CAA section 111(a)(1), quoted above. Thus, CAA sections 111(a)(1) and (d)(1) collectively require the EPA to determine the degree of emission limitation achievable through application of the BSER to existing sources and to establish regulations under which states establish standards of performance reflecting that degree of emission limitation. The EPA addresses both responsibilities through its emission guidelines, as well as through its general implementing regulations for CAA section 111(d). Consistent with the statutory requirements, the general implementing regulations require that the EPA's emission guidelines reflect—

the degree of emission limitation achievable through the application of the best system of emission reduction which (taking into account the cost of such reduction and any non-air quality health and environmental impact and energy requirements) the Administrator has determined has been adequately demonstrated from designated facilities.

Following the EPA's promulgation of emission guidelines, each state must establish standards of performance for its existing sources, which the EPA's regulations call “designated facilities.” Such standards of performance must reflect the degree of emission limitation achievable through application of the best system of emission reduction as determined by the EPA, which the Agency may express as a presumptive standard of performance in the applicable emission guidelines.

While the standards of performance that states establish in their plans must generally be no less stringent than the degree of emission limitation determined by the EPA, CAA section 111(d)(1) also requires that the EPA's regulations “permit the State in applying a standard of performance to any particular source . . . to take into consideration, among other factors, the remaining useful life of the existing source to which such standard applies.” Consistent with this statutory direction, the EPA's general implementing regulations for CAA section 111(d) provide a framework for states' consideration of remaining useful life and other factors (referred to as “RULOF”) when applying a standard of performance to a particular source. In November 2023, the EPA finalized clarifications to its regulations governing states' consideration of RULOF to apply less stringent standards of performance to particular existing sources. As amended, these regulations provide that states may apply a standard of performance to a particular designated facility that is less stringent than, or has a longer compliance schedule than, otherwise required by the applicable emission guideline taking into consideration that facility's remaining useful life and other factors. To apply a less stringent standard of performance or longer compliance schedule, the state must demonstrate with respect to each facility (or class of such facilities), that the facility cannot reasonably achieve the degree of emission limitation determined by the EPA based on unreasonable cost of control resulting from plant age, location, or basic process design; physical impossibility or technical infeasibility of installing necessary control equipment; or other circumstances specific to the facility. In doing so, the state must demonstrate that there are fundamental differences between the information specific to a facility (or class of such facilities) and the information the EPA considered in determining the degree of emission limitation achievable through application of the BSER or the compliance schedule that make achieving such degree of emission reduction or meeting such compliance schedule unreasonable for that facility.

In addition, under CAA section 116, states may establish standard of performances that are more stringent than the presumptive standards of performance contained in the EPA's emission guidelines. The state must include the standards of performance in their state plans and submit the plans to the EPA for review according to the procedures established in the Agency's general implementing regulations for CAA section 111(d). Under CAA section 111(d)(2)(A), the EPA approves state plans that are determined to be “satisfactory.” CAA section 111(d)(2)(A) also gives the Agency “the same authority” as under CAA section 110(c) to promulgate a Federal plan in cases where a state fails to submit a satisfactory state plan.

VI. ACE Rule Repeal

The EPA is finalizing repeal of the ACE Rule. The EPA proposed to repeal the ACE Rule and did not receive significant comments objecting to the proposal. The EPA is finalizing the proposal largely as proposed. A general summary of the ACE Rule, including its regulatory and judicial history, is included in section V.B.4 of this preamble. The EPA repeals the ACE Rule on three grounds that each independently justify the rule's repeal.

First, as a policy matter, the EPA concludes that the suite of heat rate improvements (HRI) the ACE Rule selected as the BSER is not an appropriate BSER for existing coal-fired EGUs. In the EPA's technical judgment, the suite of HRI set forth in the ACE Rule provide negligible CO₂ reductions at best and, in many cases, may increase CO₂ emissions because of the “rebound effect,” as explained in section VII.D.4.a.iii of this preamble. These concerns, along with the EPA's experience in implementing the ACE Rule, cast doubt that the ACE Rule would achieve emission reductions and increase the likelihood that the ACE Rule could make CO₂ pollution worse. As a result, the EPA has determined it is appropriate to repeal the rule, and to reevaluate whether other technologies constitute the BSER.

Second, even assuming the ACE Rule's rejection of CCS and natural gas co-firing was supported at the time, the ACE Rule's rationale for rejecting CCS and natural gas co-firing as the BSER no longer applies because of new factual developments. Since the ACE Rule was promulgated, changes in the power industry, developments in the costs of controls, and new federal subsidies have made other controls more broadly available and less expensive. Considering these developments, the EPA has determined that co-firing with natural gas and CCS are the BSER for certain subcategories of sources as described in section VII.C of this preamble, and that the HRI technologies adopted by the ACE Rule are not the BSER. Thus, repeal of the ACE Rule is proper on this ground as well.

Third, the EPA concludes that the ACE Rule conflicted with CAA section 111 and the EPA's implementing regulations because it did not specifically identify the BSER or the “degree of emission limitation achievable though application of the [BSER].” Instead, the ACE Rule described only a broad range of values as the “degree of emission limitation achievable.” In doing so, the rule did not provide the states with adequate guidance on the degree of emission limitation that must be reflected in the standards of performance so that a state plan would be approvable by the EPA. The ACE Rule is repealed for this reason also.

A. Summary of Selected Features of the ACE Rule

The ACE Rule determined that the BSER for coal-fired EGUs was a “list of `candidate technologies,' ” consisting of seven types of the “most impactful HRI technologies, equipment upgrades, and best operating and maintenance practices,” (84 FR 32536; July 8, 2019), including, among others, “Boiler Feed Pumps” and “Redesign/Replace Economizer.” Id. at 32537 (table 1). The rule provided a range of improvements in heat rate that each of the seven “candidate technologies” could achieve if applied to coal-fired EGUs of different capacities. For six of the technologies, the expected level of improvement in heat rate ranged from 0.1-0.4 percent to 1.0-2.9 percent, and for the seventh technology, “Improved Operating and Maintenance (O&M) Practices,” the range was “0 to >2%.” Id. The ACE Rule explained that states must review each of their designated facilities, on either a source-by-source or group-of-sources basis, and “evaluate the applicability of each of the candidate technologies.” Id. at 32550. States were to use the list of HRI technologies “as guidance but will be expected to conduct unit-specific evaluations of HRI potential, technical feasibility, and applicability for each of the BSER candidate technologies.” Id. at 32538.

The ACE Rule emphasized that states had “inherent flexibility” in evaluating candidate technologies with “a wide range of potential outcomes.” Id. at 32542. The ACE Rule provided that states could conclude that it was not appropriate to apply some technologies. Id. at 32550. Moreover, if a state decided to apply a particular technology to a particular source, the state could determine the level of heat rate improvement from the technology could be anywhere within the range that the EPA had identified for that technology, or even outside that range. Id. at 32551. The ACE Rule stated that after the state evaluated the technologies and calculated the amount of HRI in this way, it should determine the standard of performance 0that the source could achieve, Id. at 32550, and then adjust that standard further based on the application of source-specific factors such as remaining useful life. Id. at 32551.

The ACE Rule then identified the process by which states had to take these actions. States must “evaluat[e] each” of the seven candidate technologies and provide a summary, which “include[s] an evaluation of the . . . degree of emission limitation achievable through application of the technologies.” Id. at 32580. Then, the state must provide a variety of information about each power plant, including, the plant's “annual generation,” “CO₂ emissions,” “[f]uel use, fuel price, and carbon content,” “operation and maintenance costs,” “[h]eat rates,” “[e]lectric generating capacity,” and the “timeline for implementation,” among other information. Id. at 32581. The EPA explained that the purpose of this data was to allow the Agency to “adequately and appropriately review the plan to determine whether it is satisfactory.” Id. at 32558.

The ACE Rule projected a very low level of overall emission reduction if states generally applied the set of candidate technologies to their sources. The rule was projected to achieve a less-than-1-percent reduction in power-sector CO₂ emissions by 2030. Further, the EPA also projected that it would increase CO₂ emissions from power plants in 15 states and the District of Columbia because of the “rebound effect” as coal-fired sources implemented HRI measures and became more efficient. This phenomenon is explained in more detail in section VII.D.4.a.iii of this document.

The ACE Rule considered several other control measures as the BSER, including co-firing with natural gas and CCS, but rejected them. The ACE Rule rejected co-firing with natural gas primarily on grounds that it was too costly in general. 84 FR 32545 (July 8, 2019). The rule also concluded that generating electricity by co-firing natural gas in a utility boiler would be an inefficient use of the gas when compared to combusting it in a combustion turbine. Id. The ACE Rule rejected CCS on grounds that it was too costly. Id. at 32548. The rule identified the high capital and operating costs of CCS and noted the fact that the IRC section 45Q tax credit, as it then applied, would provide only limited benefit to sources. Id. at 32548-49.

B. Developments Undermining ACE Rule's Projected Emission Reductions

The EPA's first basis for repealing the ACE Rule is that it is unlikely that—if implemented—the rule would reduce emissions, and implementation could increase CO₂ emissions instead. Thus, the EPA concludes that as a matter of policy it is appropriate to repeal the rule and evaluate anew whether other technologies qualify as the BSER.

Two factors, taken together, undermine the ACE Rule's projected emission reductions and create the risk that implementation of the ACE Rule could increase—rather than reduce—CO₂ emissions from coal-fired EGUs. First, HRI technologies achieve only limited GHG emission reductions. The ACE Rule projected that if states generally applied the set of candidate technologies to their sources, the rule would achieve a less-than-1-percent reduction in power-sector CO₂ emissions by 2030. The EPA now doubts that even these minimal reductions would be achieved. The ACE Rule's projected benefits were premised in part on a 2009 technical report by Sargent & Lundy that evaluated the effects of HRI technologies. In 2023, Sargent & Lundy issued an updated report which details that the HRI selected as the BSER in the ACE Rule would bring fewer emissions reductions than estimated in 2009. The 2023 report concludes that, with few exceptions, HRI technologies are less effective at reducing CO₂ emissions than assumed in 2009. Further reinforcing the conclusion that HRIs would bring few reductions, the 2023 report also concluded that most sources had already optimized application of HRIs, and so there are fewer opportunities to reduce emissions than previously anticipated.

Second, for a subset of sources, HRI are likely to cause a “rebound effect” leading to an increase in GHG emissions for those sources. The rebound effect is explained in detail in section VII.D.4.a.iii of this preamble. The ACE Rule's analysis projected that the rule would increase CO₂ emissions from power plants in 15 states and the District of Columbia. The EPA's modeling projections assumed that, consistent with the rule, some sources would impose a small degree of efficiency improvements. The modeling showed that, as a consequence of these improvements, the rule would increase absolute emissions at some coal-fired sources as these sources became more efficient and displaced lower emitting sources like natural gas-fired EGUs.

Even though the ACE Rule was projected to increase emissions in many states, these states were nevertheless obligated under the rule to assemble detailed state plans that evaluated available technologies and the performance of each existing coal-fired power plant, as described in section IX.A of this preamble. For example, the state was required to analyze the plant's “annual generation,” “CO₂ emissions,” “[f]uel use, fuel price, and carbon content,” “operation and maintenance costs,” “[h]eat rates,” “[e]lectric generating capacity,” and the “timeline for implementation,” among other information. 84 FR 32581 (July 8, 2019). The risk of an increase in emissions raises doubts that the HRI for coal-fired sources satisfies the statutory criteria to constitute the BSER for this category of sources. The core element of the BSER analysis is whether the emission reduction technology selected reduces emissions. See Essex Chem. Corp. v. Ruckelshaus, 486 F.2d 427, 441 (D.C. Cir. 1973) (noting “counter productive environmental effects” raises questions as to whether the BSER selected was in fact the “best”). Moreover, this evaluation and the imposition of standards of performance was mandated even though the state plan would lead to an increase rather than decrease CO₂ emissions. Imposing such an obligation on states under these circumstances was arbitrary.

The EPA's experience in implementing the ACE Rule reinforces these concerns. After the ACE Rule was promulgated, one state drafted a state plan that set forth a standard of performance that allowed the affected source to increase its emission rate. The draft partial plan would have applied to one source, the Longview Power, LLC facility, and would have established a standard of performance, based on the state's consideration of the “candidate technologies,” that was higher ( i.e., less stringent) than the source's historical emission rate. Thus, the draft plan would not have achieved any emission reductions from the source, and instead would have allowed the source to increase its emissions, if it had been finalized.

Because there is doubt that the minimal reductions projected by the ACE Rule would be achieved, and because the rebound effect could lead to an increase in emissions for many sources in many states, the EPA concludes that it is appropriate to repeal the ACE Rule and reevaluate the BSER for this category of sources.

C. Developments Showing That Other Technologies Are the BSER for This Source Category

Since the promulgation of the ACE Rule in 2019, the factual underpinnings of the rule have changed in several ways and lead the EPA to determine that HRI are not the BSER for coal-fired power plants. This reevaluation is consistent with FCC v. Fox Television Stations, Inc., 556 U.S. 502 (2009). There, the Supreme Court explained that an agency issuing a new policy “need not demonstrate to a court's satisfaction that the reasons for the new policy are better than the reasons for the old one.” Instead, “it suffices that the new policy is permissible under the statute, that there are good reasons for it, and that the agency believes it to be better, which the conscious change of course adequately indicates.” Id. at 514-16 (emphasis in original; citation omitted).

Along with changes in the anticipated reductions from HRI, it makes sense for the EPA to reexamine the BSER because the costs of two control measures, co-firing with natural gas and CCS, have fallen for sources with longer-term operating horizons. As noted, the ACE Rule rejected natural gas co-firing as the BSER on grounds that it was too costly and would lead to inefficient use of natural gas. But as discussed in section VII.C.2.b of this preamble, the costs of natural gas co-firing are presently reasonable, and the EPA concludes that the costs of co-firing 40 percent by volume natural gas are cost-effective for existing coal-fired EGUs that intend to operate after January 1, 2032, and cease operation before January 1, 2039. In addition, changed circumstances—including that natural gas is available in greater amounts, that many coal-fired EGUs have begun co-firing with natural gas or converted wholly to natural-gas, and that there are fewer coal-fired EGUs in operation—mitigate the concerns the ACE Rule identified about inefficient use of natural gas.

Similarly, the ACE Rule rejected CCS as the BSER on grounds that it was too costly. But the costs of CCS have substantially declined, as discussed in section VII.C.1.a.ii of the preamble, partly because of developments in the technology that have lowered capital costs, and partly because the IRA extended and increased the IRS section 45Q tax credit so that it defrays a higher portion of the costs of CCS. Accordingly, for coal-fired EGUs that will continue to operate past 2039, the EPA concludes that the costs of CCS are reasonable, as described in section VII.C.1.a.ii of the preamble.

The emission reductions from these two technologies are substantial. For long-term coal-fired steam generating units, the BSER of 90 percent capture CCS results in substantial CO₂ emissions reductions amounting to emission rates that are 88.4 percent lower on a lb/MWh-gross basis and 87.1 percent lower on a lb/MWh-net basis compared to units without capture, as described in section VII.C.2.b.iv of this preamble. For medium term units, the BSER of 40 percent natural gas co-firing achieves CO₂ stack emissions reductions of 16 percent, as described in section VII.C.2.b.iv of this preamble. Given the availability of more effective, cost-reasonable technology, the EPA concludes that HRIs are not the BSER for all coal-fired EGUs.

The EPA is thus finalizing a new policy for coal-fired power plants. This rule applies to those sources that intend to operate past January 1, 2032. For sources that intend to cease operations after January 1, 2032, but before January 1, 2039, the EPA concludes that the BSER is co-firing 40 percent by volume natural gas. The EPA concludes this control measure is appropriate because it achieves substantial reductions at reasonable cost. In addition, the EPA believes that because a large supply of natural gas is available, devoting part of this supply for fuel for a coal-fired steam generating unit in place of a percentage of the coal burned at the unit is an appropriate use of natural gas and will not adversely impact the energy system, as described in section VII.C.2.b.iii(B) of this preamble. For sources that intend to operate past January 1, 2039, the EPA concludes that the BSER is CCS with 90 percent capture of CO₂. The EPA believes that this control measure is appropriate because it achieves substantial reductions at reasonable cost, as described in section VII.C.1 of this preamble.

The EPA is not concluding that HRI is the BSER for any coal-fired EGUs. As discussed in section VII.D.4.a, the EPA does not consider HRIs an appropriate BSER for coal-fired EGUs because these technologies would achieve few, if any, emissions reductions and may increase emissions due to the rebound effect. Most importantly, changed circumstances show that co-firing natural gas and CCS are available at reasonable cost, and will achieve more GHG emissions reductions. Accordingly, the EPA believes that HRI do not qualify as the BSER for any coal-fired EGUs, and that other approaches meet the statutory standard. On this basis, the EPA repeals the ACE Rule.

D. Insufficiently Precise Degree of Emission Limitation Achievable From Application of the BSER

The third independent reason why the EPA is repealing the ACE Rule is that the rule did not identify with sufficient specificity the BSER or the degree of emission limitation achievable through the application of the BSER. Thus, states lacked adequate guidance on the BSER they should consider and level of emission reduction that the standards of performance must achieve. The ACE Rule determined the BSER to be a suite of HRI “candidate technologies,” but did not identify with specificity the degree of emission limitation states should apply in developing standards of performance for their sources. As a result, the ACE Rule conflicted with CAA section 111 and the implementing regulations, and thus failed to provide states adequate guidance so that they could ensure that their state plans were satisfactory and approvable by the EPA.

CAA section 111 and the EPA's longstanding implementing regulations establish a clear process for the EPA and states to regulate emissions of certain air pollutants from existing sources. “The statute directs the EPA to (1) ‘determine[ ],’ taking into account various factors, the ‘best system of emission reduction which . . . has been adequately demonstrated,’ (2) ascertain the ‘degree of emission limitation achievable through the application’ of that system, and (3) impose an emissions limit on new stationary sources that `reflects' that amount.” West Virginia v. EPA, 597 U.S. at 709 (quoting 42 U.S.C. 7411(d)). Further, “[a]lthough the States set the actual rules governing existing power plants, EPA itself still retains the primary regulatory role in Section 111(d) . . . [and] decides the amount of pollution reduction that must ultimately be achieved.” Id. at 2602.

Once the EPA makes these determinations, the state must establish “standards of performance” for its sources that are based on the degree of emission limitation that the EPA determines in the emission guidelines. CAA section 111(a)(1) makes this clear through its definition of “standard of performance” as “a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of the [BSER].” After the EPA determines the BSER, 40 CFR 60.22(b)(5), and the degree of emission limitation achievable from application of the BSER, “the States then submit plans containing the emissions restrictions that they intend to adopt and enforce in order not to exceed the permissible level of pollution established by EPA.” 597 U.S. at 710 (citing 40 CFR 60.23, 60.24; 42 U.S.C. 7411(d)(1)).

The EPA then reviews the plan and approves it if the standards of performance are “satisfactory,” under CAA section 111(d)(2)(A). The EPA's longstanding implementing regulations make clear that the EPA's basis for determining whether the plan is “satisfactory” includes that the plan must contain “emission standards . . . no less stringent than the corresponding emission guideline(s).” 40 CFR 60.24(c), 40 CFR 60.24a(c). In addition, under CAA section 111(d)(1), in “applying a standard of performance to any particular source” a state may consider, “among other factors, the remaining useful life of the existing source to which such standard applies.” This is also known as the RULOF provision and is discussed in section X.C.2 of this preamble.

In the ACE Rule, the EPA recognized that the CAA required it to determine the BSER and identify the degree of emission limitation achievable through application of the BSER. 84 FR 32537 (July 8, 2019). But the rule did not make those determinations. Rather, the ACE Rule described the BSER as a list of “candidate technologies.” And the rule described the degree of emission limitation achievable by application of the BSER as ranges of reductions from the HRI technologies. The rule thus shifted the responsibility for determining the BSER and degree of emission limitation achievable from the EPA to the states. Accordingly, the ACE Rule did not meet the CAA section 111 requirement that the EPA determine the BSER or the degree of emission limitation from application of the BSER.

As described above, the ACE Rule identified the HRI in the form of a list of seven “candidate technologies,” accompanied by a wide range of percentage improvements to heat rate that these technologies could provide. Indeed, for one of them, improved “O&M” practices (that is, operation and management practices), the range was “0 to >2%,” which is effectively unbounded. 84 FR 32537 (table 1) (July 8, 2019). The ACE Rule was clear that this list was simply the starting point for a state to calculate the standards of performance for its sources. That is, the seven sets of technologies were “candidate[s]” that the state could apply to determine the standard of performance for a source, and if the state did choose to apply one or more of them, the state could do so in a manner that yielded any percentage of heat rate improvement within the range that the EPA identified, or even outside that range. Thus, as a practical matter, the ACE Rule did not determine the BSER or any degree of emission limitation from application of the BSER, and so states had no guidance on how to craft approvable state plans. In this way, the ACE Rule did not adhere to the applicable statutory obligations. See 84 FR 32537-38 (July 8, 2019).

The only constraints that the ACE Rule imposed on the states were procedural ones, and those did not give the EPA any benchmark to determine whether a plan could be approved or give the states any certainty on whether their plan would be approved. As noted above, when a state submitted its plan, it needed to show that it evaluated each candidate technology for each source or group of sources, explain how it determined the degree of emission limitation achievable, and include data about the sources. But because the ACE Rule did not identify a BSER or include a degree of emission limitation that the standards must reflect, the states lacked specific guidance on how to craft adequate standards of performance, and the EPA had no benchmark against which to evaluate whether a state's submission was “satisfactory” under CAA section 111(d)(2)(A). Thus, the EPA's review of state plans would be essentially a standardless exercise, notwithstanding the Agency's longstanding view that it was “essential” that “EPA review . . . [state] plans for their substantive adequacy.” 40 FR 53342-43 (November 17, 1975). In 1975, the EPA explained that it was not appropriate to limit its review based “solely on procedural criteria” because otherwise “states could set extremely lenient standards . . . so long as EPA's procedural requirements were met.” Id. at 53343.

Finally, the ACE Rule's approach to determining the BSER and degree of emission limitation departed from prior emission guidelines under CAA section 111(d), in which the EPA included a numeric degree of emission limitation. See, e.g.,42 FR 55796, 55797 (October 18, 1977) (limiting emission rate of acid mist from sulfuric acid plants to 0.25 grams per kilogram of acid); 44 FR 29829 (May 22, 1979) (limiting concentrations of total reduced sulfur from most of the subcategories of kraft pulp mills, such as digester systems and lime kilns, to 5, 20, or 25 ppm over 12-hour averages); 61 FR 9919 (March 12, 1996) (limiting concentration of non-methane organic compounds from solid waste landfills to 20 parts per million by volume or a 98 percent reduction). The ACE Rule did not grapple with this change in position as required by FCC v. Fox Television Stations, Inc., 556 U.S. 502 (2009), or explain why it was appropriate to provide a boundless degree of emission limitation achievable in this context.

The EPA is finalizing the repeal the ACE Rule on this ground as well. The ACE Rule's failure to determine the BSER and the associated degree of emission limitation achievable from application of the BSER deviated from CAA section 111 and the implementing regulations. Without these determinations, the ACE Rule lacked any benchmark that would guide the states in developing their state plans, and by which the EPA could determine whether those state plans were satisfactory.

For each of these three, independent reasons, repeal of the ACE Rule is proper.

E. Withdrawal of Proposed NSR Revisions

In addition to repealing the ACE Rule, the Agency is withdrawing the proposed revisions to the NSR applicability provisions that were included the ACE Rule proposal (83 FR 44756, 44773-83; August 31, 2018). These proposed revisions would have included an hourly emissions rate test to determine NSR applicability for a modified EGU, with the expressed purpose of alleviating permitting burdens for sources undertaking HRI projects pursuant to the ACE Rule emission guidelines. The ACE Rule final action did not include the NSR revisions, and the EPA indicated in that preamble that it intended to take final action on the NSR proposal in a separate action at a later date. However, the EPA did not take a final action on the NSR revisions, and the EPA has decided to no longer pursue them and to withdraw the proposed revisions.

Withdrawal of the proposal to establish an hourly emissions test for NSR applicability for EGUs is appropriate because of the repeal of the ACE rule and the EPA's conclusion that HRI is not the BSER for coal-fired EGUs. The EPA's basis for proposing the NSR revisions was to ease permitting burdens for state agencies and sources that may result from implementing the ACE Rule. There was concern that, for sources that modified their EGU to improve the heat rate, if a source were to be dispatched more frequently because of improved efficiency (the “rebound effect”), the source could experience an increase in absolute emissions for one or more pollutants and potentially trigger major NSR requirements. The hourly emissions rate test was proposed to relieve such sources that were undertaking HRI projects to comply with their state plans from the burdens of NSR permitting, particularly in cases in which a source has an increase in annual emissions of a pollutant. However, given that this final rule BSER is not based on HRIs for coal-fired EGUs, the NSR revisions proposed as part of the ACE Rule would no longer serve the purpose that the EPA expressed in that proposal preamble.

Furthermore, in the event that any sources are increasing their absolute emissions after modifying an EGU, applicability of the NSR program is beneficial as a backstop that provides review of those situations to determine if additional controls or other emission limitations are necessary on a case-by-case basis to protect air quality. In addition, given that considerable time has passed since these EGU-specific NSR applicability revisions were proposed in 2018, should the EPA decide to pursue them at a later time, it is prudent for the Agency to propose them again at that time, accompanied with the EPA's updated context and justification to support re-proposing the NSR revisions, rather than relying on the proposal from 2018. Therefore, the EPA is withdrawing these proposed NSR revisions.

VII. Regulatory Approach for Existing Fossil Fuel-Fired Steam Generating Units

Existing fossil fuel-fired steam generation units are the largest stationary source of CO₂ emissions, emitting 909 MMT CO₂ e in 2021. Recent developments in control technologies offer opportunities to reduce CO₂ emissions from these sources. The EPA's regulatory approach for these units is to require emissions reduction consistent with these technologies, where their use is cost-reasonable.

A. Overview

In this section of the preamble, the EPA identifies the BSER and degree of emission limitation achievable for the regulation of GHG emissions from existing fossil fuel-fired steam generating units. As detailed in section V of this preamble, to meet the requirements of CAA section 111(d), the EPA promulgates “emission guidelines” that identify the BSER and the degree of emission limitation achievable through the application of the BSER, and states then establish standards of performance for affected sources that reflect that level of stringency. To determine the BSER for a source category, the EPA identifies systems of emission reduction ( e.g., control technologies) that have been adequately demonstrated and evaluates the potential emissions reduction, costs, any non-air health and environmental impacts, and energy requirements. As described in section V.C.1 of this preamble, the EPA has broad authority to create subcategories under CAA section 111(d). Therefore, where the sources in a category differ from each other by some characteristic that is relevant for the suitability of the emission controls, the EPA may create separate subcategories and make separate BSER determinations for those subcategories.

The EPA considered the characteristics of fossil fuel-fired steam generating units that may impact the suitability of different control measures. First, the EPA observed that the type and amounts of fossil fuels—coal, oil, and natural gas—fired in the steam generating unit affect the performance and emissions reductions achievable by different control technologies, in part due to the differences in the carbon content of those fuels. The EPA recognized that many sources fire multiple types of fossil fuel. Therefore, the EPA is finalizing subcategories of coal-fired, oil-fired, and natural gas-fired steam generating units. The EPA is basing these subcategories, in part, on the amount of fuel combusted by the steam generating unit.

The EPA then considered the BSER that may be suitable for each of those subcategories of fuel type. For coal-fired steam generating units, of the available control technologies, the EPA is determining that CCS with 90 percent capture of CO₂ meets the requirements for BSER, including being adequately demonstrated and achieving significant emission reductions at reasonable cost for units operating in the long-term, as detailed in section VII.C.1.a of this preamble. Application of this BSER results in a degree of emission limitation equivalent to an 88.4 percent reduction in emission rate (lb CO₂ /MWh-gross). The compliance date for these sources is January 1, 2032.

Typically, the EPA assumes that sources subject to controls operate in the long-term. See, for example, the 2015 NSPS (80 FR 64509; October 23, 2015) or the 2011 CSAPR (76 FR 48208; August 8, 2011). Under that assumption, fleet average costs for CCS are comparable to the cost metrics the EPA has previously considered to be reasonable. However, the EPA observes that about half of the capacity (87 GW out of 181 GW) of existing coal-fired steam generating units have announced plans to permanently cease operation prior to 2039, as detailed in section IV.D.3.b of this preamble, affecting the period available for those sources to amortize the capital costs of CCS. Accordingly, the EPA evaluated the costs of CCS for different amortization periods. For an amortization period of more than 7 years—such that sources operate after January 1, 2039—annualized fleet average costs are comparable to or less than the metrics of costs for controls that the EPA has previously found to be reasonable. However, the group of sources ceasing operation prior to January 1, 2039, have less time available to amortize the capital costs of CCS, resulting in higher annualized costs.

Because the costs of CCS depend on the available amortization period, the EPA is creating a subcategory for sources demonstrating that they plan to permanently cease operation prior to January 1, 2039. Instead, for this subcategory of sources, the EPA is determining that natural gas co-firing at 40 percent of annual heat input meets the requirements of BSER. Application of the natural gas co-firing BSER results in a degree of emission limitation equivalent to a 16 percent reduction in emission rate (lb CO₂ /MWh-gross). Co-firing at 40 percent entails significantly less control equipment and infrastructure than CCS, and as a result, the EPA has determined that affected sources are able to implement it more quickly than CCS, by January 1, 2030. Importantly, co-firing at 40 percent also entails significantly less capital cost than CCS, and as a result, the costs of co-firing are comparable to or less than the metrics for cost reasonableness with an amortization period that is significantly shorter than the period for CCS. The EPA has determined that the costs of co-firing meet the metrics for cost reasonableness for the majority of the capacity that permanently cease operation more than 2 years after the January 1, 2030, implementation date, or after January 1, 2032 (and up to December 31, 2038), and that therefore have an amortization period of more than 2 years (and up to 9 years).

The EPA is also determining that sources demonstrating that they plan to permanently cease operation before January 1, 2032, are not subject to the 40 percent co-firing requirement. This is because their amortization period would be so short—2 years or less—that the costs of co-firing would, in general, be less comparable to the cost metrics for reasonableness for that group of sources. Accordingly, the EPA is defining the medium-term subcategory to include those sources demonstrating that they plan to permanently cease operating after December 31, 2031, and before January 1, 2039.

Considering the limited emission reductions available in light of the cost reasonableness of controls with short amortization periods, the EPA is finalizing an applicability exemption for coal-fired steam generating units demonstrating that they plan to permanently cease operation before January 1, 2032.

For natural gas- and oil-fired steam generating units, the EPA is finalizing subcategories based on capacity factor. Because natural gas- and oil-fired steam generating units with similar annual capacity factors perform similarly to one another, the EPA is finalizing a BSER of routine methods of operation and maintenance and a degree of emission limitation of no increase in emission rate for intermediate and base load subcategories. For low load natural gas- and oil-fired steam generating units, the EPA is finalizing a BSER of uniform fuels and respective degrees of emission limitation defined on a heat input basis (130 lb CO₂ /MMBtu and 170 lb CO₂ /MMBtu). Furthermore, the EPA is finalizing presumptive standards for natural gas- and oil-fired steam generating units as follows: base load sources (those with annual capacity factors greater than 45 percent) have a presumptive standard of 1,400 lb CO₂ /MWh-gross, intermediate load sources (those with annual capacity factors greater than 8 percent and or less than or equal to 45 percent) have a presumptive standard of 1,600 lb CO₂ /MWh-gross. For low load oil-fired sources, the EPA is finalizing a presumptive standard of 170 lb CO₂ /MMBtu, while for low load natural gas-fired sources the EPA is finalizing a presumptive standard of 130 lb CO₂ /MMBtu. A compliance date of January 1, 2030, applies for all natural gas- and oil-fired steam generating units.

The final subcategories and BSER are summarized in table 1 of this document.

B. Applicability Requirements and Fossil Fuel-Type Definitions for Subcategories of Steam Generating Units

In this section of the preamble, the EPA describes the rationale for the final applicability requirements for existing fossil fuel-fired steam generating units. The EPA also describes the rationale for the fuel type definitions and associated subcategories.

1. Applicability Requirements

For the emission guidelines, the EPA is finalizing that a designated facility is any fossil fuel-fired electric utility steam generating unit ( i.e., utility boiler or IGCC unit) that: (1) was in operation or had commenced construction on or before January 8, 2014; (2) serves a generator capable of selling greater than 25 MW to a utility power distribution system; and (3) has a base load rating greater than 260 GJ/h (250 million British thermal units per hour (MMBtu/h)) heat input of fossil fuel (either alone or in combination with any other fuel). Consistent with the implementing regulations, the term “designated facility” is used throughout this preamble to refer to the sources affected by these emission guidelines. For the emission guidelines, consistent with prior CAA section 111 rulemakings concerning EGUs, the term “designated facility” refers to a single EGU that is affected by these emission guidelines. The rationale for the final applicability requirements is the same as that for 40 CFR part 60, subpart TTTT (80 FR 64543-44; October 23, 2015). The EPA includes that discussion by reference here.

Section 111(a)(6) of the CAA defines an “existing source” as “any stationary source other than a new source.” Therefore, the emission guidelines do not apply to any steam generating units that are new after January 8, 2014, or reconstructed after June 18, 2014, the applicability dates of 40 CFR part 60, subpart TTTT. Moreover, because the EPA is now finalizing revised standards of performance for coal-fired steam generating units that undertake a modification, a modified coal-fired steam generating unit would be considered “new,” and therefore not subject to these emission guidelines, if the modification occurs after the date the proposal was published in the Federal Register (May 23, 2023). Any coal-fired steam generating unit that has modified prior to that date would be considered an existing source that is subject to these emission guidelines.

In addition, the EPA is finalizing in the applicability requirements of the emission guidelines many of the same exemptions as discussed for 40 CFR part 60, subpart TTTT, in section VIII.E.1 of this preamble. EGUs that may be excluded from the requirement to establish standards under a state plan are: (1) units that are subject to 40 CFR part 60, subpart TTTT, as a result of commencing a qualifying modification or reconstruction; (2) steam generating units subject to a federally enforceable permit limiting net-electric sales to one-third or less of their potential electric output or 219,000 MWh or less on an annual basis and annual net-electric sales have never exceeded one-third or less of their potential electric output or 219,000 MWh; (3) non-fossil fuel units ( i.e., units that are capable of deriving at least 50 percent of heat input from non-fossil fuel at the base load rating) that are subject to a federally enforceable permit limiting fossil fuel use to 10 percent or less of the annual capacity factor; (4) combined heat and power (CHP) units that are subject to a federally enforceable permit limiting annual net-electric sales to no more than either 219,000 MWh or the product of the design efficiency and the potential electric output, whichever is greater; (5) units that serve a generator along with other affected EGU(s), where the effective generation capacity (determined based on a prorated output of the base load rating of EGU) is 25 MW or less; (6) municipal waste combustor units subject to 40 CFR part 60, subpart Eb; (7) commercial or industrial solid waste incineration units that are subject to 40 CFR part 60, subpart CCCC; (8) EGUs that derive greater than 50 percent of the heat input from an industrial process that does not produce any electrical or mechanical output or useful thermal output that is used outside the affected EGU; or (9) coal-fired steam generating units that have elected to permanently cease operation prior to January 1, 2032.

The exemptions listed above at (4), (5), (6), and (7) are among the current exemptions at 40 CFR 60.5509(b), as discussed in section VIII.E.1 of this preamble. The exemptions listed above at (2), (3), and (8) are exemptions the EPA is finalizing revisions for 40 CFR part 60, subpart TTTT, and the rationale for the exemptions is in section VIII.E.1 of this preamble. For consistency with the applicability requirements in 40 CFR part 60, subpart TTTT, and 40 CFR part 60, subpart TTTTa, the Agency is finalizing these same exemptions for the applicability of the emission guidelines.

2. Coal-Fired Units Permanently Ceasing Operation Before January 1, 2032

The EPA is not addressing existing coal-fired steam generating units demonstrating that they plan to permanently cease operating before January 1, 2032, in these emission guidelines. Sources ceasing operation before that date have far less emission reduction potential than sources that will be operating longer, because there are unlikely to be appreciable, cost-reasonable emission reductions available on average for the group of sources operating in that timeframe. This is because controls that entail capital expenditures are unlikely to be of reasonable cost for these sources due to the relatively short period over which they could amortize the capital costs of controls.

In particular, in developing the emission guidelines, the EPA evaluated two systems of emission reduction that achieve substantial emission reductions for coal-fired steam generating units: CCS with 90 percent capture; and natural gas co-firing at 40 percent of heat input. For CCS, the EPA has determined that controls can be installed and fully operational by the compliance date of January 1, 2032, as detailed in section VII.C.1.a.i(E) of this preamble. CCS would therefore, in most cases, be unavailable to coal-fired steam generating units planning to cease operation prior to that date. Furthermore, the EPA evaluated the costs of CCS for different amortization periods. For an amortization period of more than 7 years—such that sources operate after January 1, 2039—annualized fleet average costs are comparable to or less than the costs of controls the EPA has previously determined to be reasonable ($18.50/MWh of generation and $98/ton of CO₂ reduced), as detailed in section VII.C.1.a.ii of this preamble. However, the costs for shorter amortization periods are higher. For sources ceasing operation by January 1, 2032, it would be unlikely that the annualized costs of CCS would be reasonable even were CCS installed at an earlier date ( e.g., by January 1, 2030) due to the shorter amortization period available.

Because the costs of CCS would be higher for shorter amortization periods, the EPA is finalizing a separate subcategory for sources demonstrating that they plan to permanently cease operating by January 1, 2039, with a BSER of 40 percent natural gas co-firing, as detailed in section VII.C.2.b.ii of this preamble. For natural gas co-firing, the EPA is finalizing a compliance date of January 1, 2030, as detailed in section VII.C.2.b.i(C) of this preamble. Therefore, the EPA assumes sources subject to a natural gas co-firing BSER can amortize costs for a period of up to 9 years. The EPA has determined that the costs of natural gas co-firing at 40 percent meet the metrics for cost reasonableness for the majority of the capacity that operate more than 2 years after the January 1, 2030, implementation date, i.e., that operate after January 1, 2032 (and up to December 31, 2038), and that therefore have an amortization period of more than 2 years (and up to 9 years).

However, for sources ceasing operation prior to January 1, 2032, the EPA believes that establishing a best system of emission reduction corresponding to a substantial level of natural gas co-firing would broadly entail costs of control that are above those that the EPA is generally considering reasonable. Sources permanently ceasing operation before January 1, 2032 would have less than 2 years to amortize the capital costs, as detailed in section VII.C.2.a of this preamble. Compared to the metrics for cost reasonableness that EPA has previously deemed reasonable ($18.50/MWh of generation and $98/ton of CO₂ reduced), very few sources can co-fire 40 percent natural gas at costs comparable to these metrics with an amortization period of only one year; only 1 percent of units have costs that are below both $18.50/MWh of generation and $98/ton of CO₂ reduced. The number of sources that can co-fire lower amounts of natural gas at costs comparable to these metrics is likewise limited—only approximately 34 percent of units can co-fire with 20 percent natural gas at costs lower than both cost metrics. Furthermore, the period that these sources would operate with co-firing for would be short, so that the emission reductions from that group of sources would be limited.

By contrast, assuming a two-year amortization period, many more units can co-fire with meaningful amounts of natural gas at a cost that is consistent with the metrics EPA has previously used: 18 percent of units can co-fire with 40 percent natural gas at costs less than $98/ton and $18.50/MWh, and 50 percent of units can co-fire with 20 percent natural gas at costs lower than both metrics. Because a substantial number of sources can implement 40-percent co-firing with natural gas with an amortization period of two years or longer with reasonable costs, and even more can co-fire with lesser amounts with reasonable costs with amortization periods longer than two years, the EPA determined that a technology-based BSER was available for coal-fired units operating past January 1, 2032.

Sources that retire before that date, however, are differently situated as described above. In light of the small number of sources that are planning to retire before January 1, 2032 that could cost-effectively co-fire with natural gas, coupled with the small amount of emissions reductions that can be achieved from co-firing in such a short time span, the EPA is choosing not to establish a BSER for these sources.

Because, at this time, the EPA has determined that CCS and natural gas co-firing are not available at reasonable cost for sources ceasing operation before January 1, 2032, the EPA is not finalizing a BSER for such sources. Not finalizing a BSER for these sources is consistent with the Agency's discretion to take incremental steps to address CO₂ from sources in the category, and to direct the EPA's limited resources at regulation of those sources that can achieve the most emission reductions. The EPA is therefore providing that existing coal-fired steam generating EGUs that have elected to cease operating before January 1, 2032, are not regulated by these emission guidelines. This exemption applies to a source until the earlier of December 31, 2031, or the date it demonstrates in the state plan that it plans to cease operation. If a source continues to operate past this date, it is no longer exempt from these emission guidelines. See section X.E.1 of this preamble for discussion of how state plans should address sources subject to exemption (9).

3. Sources Outside of the Contiguous U.S.

The EPA proposed the same emission guidelines for fossil fuel-fired steam generating units in non-continental areas ( i.e., Hawaii, the U.S. Virgin Islands, Guam, American Samoa, the Commonwealth of Puerto Rico, and the Northern Mariana Islands) and non-contiguous areas (non-continental areas and Alaska) as the EPA proposed for comparable units in the contiguous 48 states. The EPA notes that the modeling that supports the final emission guidelines focus on sources in the contiguous U.S. Further, the EPA notes that few, if any, coal-fired steam generating units operate outside of the contiguous 48 states and meet the applicability criteria. Finally, the EPA notes that the proposed BSER and degree of emissions limitation for non-continental oil-fired steam generating units would have achieved few emission reductions. Therefore, the EPA is not finalizing emission guidelines for existing steam generating units in states and territories (including Alaska, Hawaii, Guam, Puerto Rico, and the U.S. Virgin Islands) that are outside of the contiguous U.S. at this time.

4. IGCC Units

The EPA notes that existing IGCC units were included in the proposed applicability requirements and that, in section VII.B of this preamble, the EPA is finalizing inclusion of those units in the subcategory of coal-fired steam generating units. IGCC units gasify coal or solid fossil fuel ( e.g., pet coke) to produce syngas (a mixture of carbon monoxide and hydrogen), and either burn the syngas directly in a combined cycle unit or use a catalyst for water-gas shift (WGS) to produce a pre-combustion gas stream with a higher concentration of CO₂ and hydrogen, which can be burned in a hydrogen turbine combined cycle unit. As described in section VII.C of this preamble, the final BSER for coal-fired steam generating units includes co-firing natural gas and CCS. The few IGCC units that now operate in the U.S. either burn natural gas exclusively—and as such operate as natural gas combined cycle units—or in amounts near to the 40 percent level of the natural gas co-firing BSER. Additionally, IGCC units may be suitable for pre-combustion CO₂ capture. Because the CO₂ concentration in the pre-combustion gas, after WGS, is high relative to coal-combustion flue gas, pre-combustion CO₂ capture for IGCC units can be performed using either an amine-based (or other solvent-based) capture process or a physical absorption capture process. Alternatively, post-combustion CO₂ capture can be applied to the source. The one existing IGCC unit that still uses coal was recently awarded funding from DOE for a front-end engineering design (FEED) study for CCS targeting a capture efficiency of more than 95 percent. For these reasons, the EPA is not distinguishing IGCC units from other coal-fired steam generating EGUs, so that the BSER of co-firing for medium-term coal-fired units and CCS for long-term coal-fired units apply to IGCC units.

5. Fossil Fuel-Type Definitions for Subcategories of Steam Generating Units

In this action, the EPA is finalizing definitions for subcategories of existing fossil fuel-fired steam generating units based on the type and amount of fossil fuel used in the unit. The EPA is finalizing separate subcategories based on fuel type because the carbon content of the fuel combusted affects the output emission rate ( i.e., lb CO₂ /MWh). Fuels with a higher carbon content produce a greater amount of CO₂ emissions per unit of fuel combusted (on a heat input basis, MMBtu) and per unit of electricity generated ( i.e., MWh).

The EPA proposed fossil fuel type subcategory definitions based on the definitions in 40 CFR part 63, subpart UUUUU, and the fossil fuel definitions in 40 CFR part 60, subpart TTTT. Those proposed definitions were determined by the relative heat input contribution of the different fuels combusted in a unit during the 3 years prior to the proposed compliance date of January 1, 2030. Further, to be considered an oil-fired or natural gas-fired unit for purposes of this emission guideline, a source would no longer retain the capability to fire coal after December 31, 2029.

The EPA proposed a 3-year lookback period, so that the proposed fuel-type subcategorization would have been based, in part, on the fuel type fired between January 1, 2027, and January 1, 2030. However, the intent of the proposed fuel type subcategorization was to base the fuel type definition on the state of the source on January 1, 2030. Therefore, the EPA is finalizing the following fuel type subcategory definitions:

• Acoal-fired steam generating unit is an electric utility steam generating unit or IGCC unit that meets the definition of “fossil fuel-fired” and that burns coal for more than 10.0 percent of the average annual heat input during any continuous 3-calendar-year period after December 31, 2029, or for more than 15.0 percent of the annual heat input during any one calendar year after December 31, 2029, or that retains the capability to fire coal after December 31, 2029.
• Anoil-fired steam generating unit is an electric utility steam generating unit meeting the definition of “fossil fuel-fired” that is not a coal-fired steam generating unit, that no longer retains the capability to fire coal after December 31, 2029, and that burns oil for more than 10.0 percent of the average annual heat input during any continuous 3-calendar-year period after December 31, 2029, or for more than 15.0 percent of the annual heat input during any one calendar year after December 31, 2029.
• Anatural gas-fired steam generating unit is an electric utility steam generating unit meeting the definition of “fossil fuel-fired,” that is not a coal-fired or oil-fired steam generating unit, that no longer retains the capability to fire coal after December 31, 2029, and that burns natural gas for more than 10.0 percent of the average annual heat input during any continuous 3-calendar-year period after December 31, 2029, or for more than 15.0 percent of the annual heat input during any one calendar year after December 31, 2029.

The EPA received some comments on the fuel type definitions. Those comments and responses are as follows.

Comment: Some industry stakeholders suggested changes to the proposed definitions for fossil fuel type. Specifically, some commenters requested that the reference to the initial compliance date be removed and that the fuel type determination should instead be rolling and continually update after the initial compliance date. Those commenters suggested this would, for example, allow sources in the coal-fired subcategory that begin natural gas co-firing in 2030 to convert to the natural-gas fired subcategory prior to the proposed date of January 1, 2040, instead of ceasing operation.

Other industry commenters suggested that to be a natural gas-fired steam generating unit, a source could either meet the heat input requirements during the 3 years prior to the compliance date or (emphasis added) no longer retain the capability to fire coal after December 31, 2029. Those commenters noted that, as proposed, a source that had planned to convert to 100 percent natural gas-firing would essentially have to do so prior to January 1, 2027, to meet the proposed heat input-based definition, in addition to removing the capability to fire coal by the compliance date.

Response: Although full natural gas conversions are not a measure that the EPA considered as a potential BSER, the emission guidelines do not prohibit such conversions should a state elect to require or accommodate them. As noted above, the EPA recognizes that many steam EGUs that formerly utilized coal as a primary fuel have fully or partially converted to natural gas, and that additional steam EGUs may elect to do so during the implementation period for these emission guidelines. However, these emission guidelines place reasonable constraints on the timing of such a conversion in situations where a source seeks to be regulated as a natural gas-fired steam EGU rather than as a coal-fired steam EGU. The EPA believes that such constraints are necessary in order to avoid creating a perverse incentive for EGUs to defer conversions in a way that could undermine the emission reduction purpose of the rule. Therefore, the EPA disagrees with those commenters that suggest the EPA should, in general, allow EGUs to be regulated as natural gas-fired steam EGUs when they undertake such conversions past January 1, 2030.

However, the EPA acknowledges that the proposed subcategorization would have essentially required a unit to convert to natural gas by January 1, 2027 in order to be regulated as a natural gas-fired steam EGU. The EPA is finalizing fuel type subcategorization based on the state of the source on the compliance date of January 1, 2030, and during any period thereafter, as detailed in section VII.B of this preamble. Should a source not be able to fully convert to natural gas by this date, it would be treated as a coal-fired steam generating EGU; however, the state may be able to use the RULOF provisions, as discussed in section X.C.2 of this preamble, to particularize a standard of performance for the unit. Note that if a state relies on operating conditions within the control of the source as the basis of providing a less stringent standard of performance or longer compliance schedule, it must include those operating conditions as an enforceable requirement in the state plan. 40 CFR 60.24a(g).

C. Rationale for the BSER for Coal-Fired Steam Generating Units

This section of the preamble describes the rationale for the final BSERs for existing coal-fired steam generating units based on the criteria described in section V.C of this preamble.

At proposal, the EPA evaluated two primary control technologies as potentially representing the BSER for existing coal-fired steam generating units: CCS and natural gas co-firing. For sources operating in the long-term, the EPA proposed CCS with 90 percent capture as BSER. For sources operating in the medium-term ( i.e., those demonstrating that they plan to permanently cease operation by January 1, 2040), the EPA proposed 40 percent natural gas co-firing as BSER. For imminent-term and near-term sources ceasing operation earlier, the EPA proposed BSERs of routine methods of operation and maintenance.

The EPA is finalizing CCS with 90 percent capture as BSER for coal-fired steam generating units because CCS can achieve a substantial amount of emission reductions and satisfies the other BSER criteria. CCS has been adequately demonstrated and results in by far the largest emissions reductions of the available control technologies. As noted below, the EPA has also determined that the compliance date for CCS is January 1, 2032. CCS, however, entails significant up-front capital expenditures that are amortized over a period of years. The EPA evaluated the cost for different amortization periods, and the EPA has concluded that CCS is cost-reasonable for units that operate past January 1, 2039. As noted in section IV.D.3.b of this preamble, about half (87 GW out of 181 GW) of all coal-fired capacity currently in existence has announced plans to permanently cease operations by January 1, 2039, and additional sources are likely to do so because they will be older than the age at which sources generally have permanently ceased operations since 2000. The EPA has determined that the remaining sources that may operate after January 1, 2039, can, on average, install CCS at a cost that is consistent with the EPA's metrics for cost reasonableness, accounting for an amortization period for the capital costs of more than 7 years, as detailed in section VII.C.1.a.ii of this preamble. If a particular source has costs of CCS that are fundamentally different from those amounts, the state may consider it to be a candidate for a different control requirement under the RULOF provision, as detailed in section X.C.2 of this preamble. For the group of sources that permanently cease operation before January 1, 2039, the EPA has concluded that CCS would in general be of higher cost, and therefore is finalizing a subcategory for these units, termed medium-term units, and finalizing 40 percent natural gas co-firing on a heat input basis as the BSER.

These final subcategories and BSERs are largely consistent with the proposal, which included a long-term subcategory for sources that did not plan to permanently cease operations by January 1, 2040, with 90 percent capture CCS as the BSER; and a medium-term subcategory for sources that permanently cease operations by that date and were not in any of the other proposed subcategories, discussed next, with 40 percent co-firing as the BSER. For both subcategories, the compliance date was January 1, 2030. The EPA also proposed an imminent-term subcategory, for sources that planned to permanently cease operations by January 1, 2032; and a near-term subcategory, for sources that planned to permanently case operations by January 1, 2035, and that limited their annual capacity utilization to 20 percent. The EPA proposed a BSER of routine methods of operation and maintenance for these two subcategories.

The EPA is not finalizing these imminent-term and near-term subcategories. In addition, after considering the comments, the EPA acknowledges that some additional time from what was proposed may be beneficial for the planning and installation of CCS. Therefore, the EPA is finalizing a January 1, 2032, compliance date for long-term existing coal-fired steam generating units. As noted above, the EPA's analysis of the costs of CCS also indicates that CCS is cost-reasonable with a minimum amortization period of seven years; as a result, the final emission guidelines would apply a CCS-based standard only to those units that plan to operate for at least seven years after the compliance deadline ( i.e., units that plan to remain in operation after January 1, 2039). For medium-term sources subject to a natural gas co-firing BSER, the EPA is finalizing a January 1, 2030, compliance date because the EPA has concluded that this provides a reasonable amount of time to begin co-firing, a technology that entails substantially less up-front infrastructure and, relatedly, capital expenditure than CCS.

1. Long-Term Coal-Fired Steam Generating Units

The EPA is finalizing CCS with 90 percent capture of CO₂ at the stack as BSER for long-term coal-fired steam generating units. Coal-fired steam generating units are the largest stationary source of CO₂ in the United States. Coal-fired steam generating units have higher emission rates than other generating technologies, about twice the emission rate of a natural gas combined cycle unit. Typically, even newer, more efficient coal-fired steam generating units emit over 1,800 lb CO₂ /MWh-gross, while many existing coal-fired steam generating units have emission rates of 2,200 lb CO₂ /MWh-gross or higher. As noted in section IV.B of this preamble, coal-fired sources emitted 909 MMT CO₂ e in 2021, 59 percent of the GHG emissions from the power sector and 14 percent of the total U.S. GHG emissions—contributing more to U.S. GHG emissions than any other sector, aside from transportation road sources. Furthermore, considering the sources in the long-term subcategory will operate longer than sources with shorter operating horizons, long-term coal-fired units have the potential to emit more total CO₂.

CCS is a control technology that can be applied at the stack of a steam generating unit, achieves substantial reductions in emissions and can capture and permanently sequester more than 90 percent of CO₂ emitted by coal-fired steam generating units. The technology is adequately demonstrated, given that it has been operated at scale and is widely applicable to these sources, and there are vast sequestration opportunities across the continental U.S. Additionally, the costs for CCS are reasonable, in light of recent technology cost declines and policies including the tax credit under IRC section 45Q. Moreover, the non-air quality health and environmental impacts of CCS can be mitigated and the energy requirements of CCS are not unreasonably adverse. The EPA's weighing of these factors together provides the basis for finalizing CCS as BSER for these sources. In addition, this BSER determination aligns with the caselaw, discussed in section V.C.2.h of the preamble, stating that CAA section 111 encourages continued advancement in pollution control technology.

At proposal, the EPA also evaluated natural gas co-firing at 40 percent of heat input as a potential BSER for long-term coal-fired steam generating units. While the unit level emission rate reductions of 16 percent achieved by 40 percent natural gas co-firing are appreciable, those reductions are substantially less than CCS with 90 percent capture of CO₂. Therefore, because CCS achieves more reductions at the unit level and is cost-reasonable, the EPA is not finalizing natural gas co-firing as the BSER for these units. Further, the EPA is not finalizing partial-CCS at lower capture rates ( e.g., 30 percent) because it achieves substantially fewer unit-level reductions at greater cost, and because CCS at 90 percent is achievable. Notably, the IRC section 45Q tax credit may not be available to defray the costs of partial CCS and the emission reductions would be limited. And the EPA is not finalizing HRI as the BSER for these units because of the limited reductions and potential rebound effect.

a. Rationale for CCS as the BSER for Long-Term Coal-Fired Steam Generating Units

In this section of the preamble, the EPA explains the rationale for CCS as the BSER for existing long-term coal-fired steam generating units. This section discusses the aspects of CCS that are relevant for existing coal-fired steam generating units and, in particular, long-term units. As noted in section VIII.F.4.c.iv of this preamble, much of this discussion is also relevant for the EPA's determination that CCS is the BSER for new base load combustion turbines.

In general, CCS has three major components: CO₂ capture, transportation, and sequestration/storage. Detailed descriptions of these components are provided in section VII.C.1.a.i of this preamble. As an overview, post-combustion capture processes remove CO₂ from the exhaust gas of a combustion system, such as a utility boiler or combustion turbine. This technology is referred to as “post-combustion capture” because CO₂ is a product of the combustion of the primary fuel and the capture takes place after the combustion of that fuel. The exhaust gases from most combustion processes are at atmospheric pressure, contain somewhat dilute concentrations of CO_2, and are moved through the flue gas duct system by fans. To separate the CO₂ contained in the flue gas, most current post-combustion capture systems utilize liquid solvents—commonly amine-based solvents—in CO₂ scrubber systems using chemical absorption (or chemisorption). In a chemisorption-based separation process, the flue gas is processed through the CO₂ scrubber and the CO₂ is absorbed by the liquid solvent. The CO₂ -rich solvent is then regenerated by heating the solvent to release the captured CO₂.

The high purity CO₂ is then compressed and transported, generally through pipelines, to a site for geologic sequestration ( i.e., the long-term containment of CO₂ in subsurface geologic formations). Pipelines are subject to Federal safety regulations administered by PHMSA. Furthermore, sequestration sites are widely available across the nation, and the EPA has developed a comprehensive regulatory structure to oversee geologic sequestration projects and assure their safety and effectiveness.

i. Adequately Demonstrated

In this section of the preamble, the EPA explains the rationale for finalizing its determination that 90 percent capture applied to long-term coal-fired steam generating units is adequately demonstrated. In this section, the EPA first describes how simultaneous operation of all components of CCS functioning in concert with one another has been demonstrated, including a commercial scale application on a coal-fired steam generating unit. The demonstration of the individual components of CO₂ capture, transport, and sequestration further support that CCS is adequately demonstrated. The EPA describes how demonstrations of CO₂ capture support that 90 percent capture rates are adequately demonstrated. The EPA further describes how transport and geologic sequestration are adequately demonstrated, including the feasibility of transport infrastructure and the broad availability of geologic sequestration reservoirs in the U.S.

(A) Simultaneous Demonstration of CO₂ Capture, Transport, and Sequestration

The EPA proposed that CCS was adequately demonstrated for applications on combustion turbines and existing coal-fired steam generating units.

On reviewing the available information, all components of CCS—CO₂ capture, CO₂ transport, and CO₂ sequestration—have been demonstrated concurrently, with each component operating simultaneously and in concert with the other components.

(1) Industrial Applications of CCS

Solvent-based CO₂ capture was patented nearly 100 years ago in the 1930s and has been used in a variety of industrial applications for decades. For example, since 1978, an amine-based system has been used to capture approximately 270,000 metric tons of CO₂ per year from the flue gas of the bituminous coal-fired steam generating units at the 63 MW Argus Cogeneration Plant at Searles Valley Minerals (Trona, California). Furthermore, thousands of miles of CO₂ pipelines have been constructed and securely operated in the U.S. for decades. And tens of millions of tons of CO₂ have been permanently stored deep underground either for geologic sequestration or in association with EOR. There are currently at least 15 operating CCS projects in the U.S., and another 121 that are under construction or in advanced stages of development. This broad application of CCS demonstrates that the components of CCS have been successfully operated simultaneously. The Shute Creek Facility has a capture capacity of 7 million metric tons per year and has been in operation since 1986. The facility uses a solvent-based process to remove CO₂ from natural gas, and the captured CO₂ is stored in association with EOR. Another example of CCS in industrial applications is the Great Plains Synfuels Plant has a capture capacity of 3 million metric tons per year and has been in operation since 2000. The Great Plains Synfuels Plant (Beulah, North Dakota) uses a solvent-based process to remove CO₂ from lignite-derived syngas, the CO₂ is transported by the Souris Valley pipeline, and stored underground in association with EOR in the Weyburn and Midale Oil Units in Saskatchewan, Canada. Over 39 million metric tons of CO₂ has been captured since 2000.

(2) Various CO₂ capture methods are used in industrial applications and are tailored to the flue gas conditions of a particular industry (see the TSD GHG Mitigation Measures for Steam Generating Units for details). Of those capture technologies, amine solvent-based capture has been demonstrated for removal of CO₂ from the post-combustion flue gas of fossil fuel-fired EGUs. The Quest CO₂ capture facility in Alberta, Canada, uses amine-based CO₂ capture retrofitted to three existing steam methane reformers at the Scotford Upgrader facility (operated by Shell Canada Energy) to capture and sequester approximately 80 percent of the CO₂ in the produced syngas. Amine-solvents are also applied for post-combustion capture from fossil fuel fired EGUs. The Quest facility has been operating since 2015 and captures approximately 1 million metric tons of CO₂ per year.

Applications of CCS at Coal-Fired Steam Generating Units

For electricity generation applications, this includes operation of CCS at Boundary Dam Unit 3 in Saskatchewan, Canada. CCS at Boundary Dam Unit 3 includes capture of the CO₂ from the flue-gas of the fossil fuel-fired EGU, compression of the CO₂ onsite and transport via pipeline offsite, and storage of the captured CO₂ underground. Storage of the CO₂ captured at Boundary Dam primarily occurs via EOR. Moreover, CO₂ captured from Boundary Dam Unit 3 is also stored in a deep saline aquifer at the Aquistore Deep Saline CO₂ Storage Project, which has permanently stored over 550,000 tons of CO₂ to date. Other demonstrations of CCS include the 240 MWe Petra Nova CCS project at the subbituminous coal-fired W.A. Parish plant in Texas, which, because it was EPAct05-assisted, we cite as useful in section VII.C.1.a.i(B)(2) of this preamble, but not essential, corroboration. See section VII.C.1.a.i(H)(1) for a detailed description of how the EPA considers information from EPAct05-assisted projects.

Commenters stated that that all constituent components of CCS—carbon capture, transportation, and sequestration—have not been adequately demonstrated in integrated, simultaneous operation. We disagree with this comment. The record described in the preceding shows that all components have been demonstrated simultaneously. Even if the record only included demonstration of the individual components of CCS, the EPA would still determine that CCS is adequately demonstrated as it would be reasonable on a technical basis that the individual components are capable of functioning together—they have been engineered and designed to do so, and the record for the demonstration of the individual components is based on decades of direct data and experience.

(B) CO₂ Capture Technology at Coal-Fired Steam Generating Units

The EPA is finalizing the determination that the CO₂ capture component of CCS has been adequately demonstrated at a capture efficiency of 90 percent, is technically feasible, and is achievable over long periods ( e.g., a year) for the reasons summarized here and detailed in the following subsections of this preamble. This determination is based, in part, on the demonstration of the technology at existing coal-fired steam generating units, including the commercial-scale installation at Boundary Dam Unit 3. The application of CCS at Boundary Dam follows decades of development of CO₂ capture for coal-fired steam generating units, as well as numerous smaller-scale demonstrations that have successfully implemented this technology. Review of the available information has also identified specific, currently available, minor technological improvements that can be applied today to better the performance of new capture plant retrofits, and which can assure that the capture plants achieve 90 percent capture. The EPA's determination that 90 percent capture of CO₂ is adequately demonstrated is further corroborated by EPAct05-assisted projects, including the Petra Nova project.

Moreover, several CCS retrofit projects on coal-fired steam generating units are in progress that apply the lessons from the prior projects and use solvents that achieve higher capture rates. Technology providers that supply those solvents and the associated process technologies have made statements concluding that the technology is commercially proven and available today and have further stated that those solvents achieve capture rates of 95 percent or greater. Technology providers have decades of experience and have done the work to responsibly scale up the technology over that time across a range of flue gas compositions. Taking all of those factors into consideration, and accounting for the operation and flue gas conditions of the affected sources, solvent-based capture will consistently achieve capture rates of 90 percent or greater for the fleet of long-term coal-fired steam generating units.

Various technologies may be used to capture CO₂, the details of which are described generally in section IV.C.1 of this preamble and in more detail in the final TSD, GHG Mitigation Measures for Steam Generating Units, which is available in the rulemaking docket. For post-combustion capture, these technologies include solvent-based methods ( e.g., amines, chilled ammonia), solid sorbent-based methods, membrane filtration, pressure-swing adsorption, and cryogenic methods. Lastly, oxy-combustion uses a purified oxygen stream from an air separation unit (often diluted with recycled CO₂ to control the flame temperature) to combust the fuel and produce a higher concentration of CO₂ in the flue gas, as opposed to combustion with oxygen in air which contains 80 percent nitrogen. The CO₂ can then be separated by the aforementioned CO₂ capture methods. Of the available capture technologies, solvent-based processes have been the most widely demonstrated at commercial scale for post-combustion capture and are applicable to use with either combustion turbines or steam generating units.

The EPA's identification of CCS with 90 percent capture as the BSER is premised, in part, on an amine solvent-based CO₂ system. Amine solvents used for carbon capture are typically proprietary, although non-proprietary solvents ( e.g., monoethanolamine, MEA) may be used. Carbon capture occurs by reactive absorption of the CO₂ from the flue gas into the amine solution in an absorption column. The amine reacts with the CO₂ but will also react with impurities in the flue gas, including SO₂ . PM will also affect the capture system. Adequate removal of SO₂ and PM prior to the CO₂ capture system is therefore necessary. After pretreatment of the flue gas with conventional SO₂ and PM controls, the flue gas goes through a quencher to cool the flue gas and remove further impurities before the CO₂ absorption column. After absorption, the CO₂ -rich amine solution passes to the solvent regeneration column, while the treated gas passes through a water and/or acid wash column to limit emission of amines or other byproducts. In the solvent regeneration column, the solution is heated (using steam) to release the absorbed CO₂ . The released CO₂ is then compressed and transported offsite, usually by pipeline. The amine solution from the regenerating column is then cooled, a portion of the lean solvent is treated in a solvent reclaiming process to mitigate degradation of the solvent, and the lean solvent streams are recombined and sent back to the absorption column.

(1) Capture Demonstrations at Coal-Fired Steam Generating Units

(a) SaskPower's Boundary Dam Unit 3

SaskPower's Boundary Dam Unit 3, a 110 MW lignite-fired unit in Saskatchewan, Canada, was designed to achieve CO₂ capture rates of 90 percent using an amine-based post-combustion capture system retrofitted to the existing steam generating unit. The capture plant, which began operation in 2014, is the first full-scale CO₂ capture system retrofit on an existing coal-fired power plant. It uses the amine-based Shell CANSOLV® process, which includes an amine-based SO₂ scrubbing process and a separate amine-based CO₂ capture process, with integrated heat and power from the steam generating unit.

After undergoing maintenance and design improvements in September and October of 2015 to address technical and mechanical challenges faced in its first year of operation, Boundary Dam Unit 3 completed a 72-hour test of its design capture rate (3,240 metric tons/day), and captured 9,695 metric tons of CO₂ or 99.7 percent of the design capacity (approximately 89.7 percent capture) with a peak rate of 3,341 metric tons/day. However, the capture plant has not consistently operated at this total capture efficiency. In general, the capture plant ran less than 100 percent of the flue gas through the capture equipment and the coal-fired steam generating unit also operates when the capture plant is offline for maintenance. As a result, although the capture plant has consistently achieved 90 percent capture rates of the CO₂ in the processed slipstream, the amount of CO₂ captured was less than 90 percent of the total amount of CO₂ in the flue gas of the steam generating unit. Some of the reasons for this operation were due to the economic incentives and regulatory requirements of the project, while other reasons were due to technical challenges. The EPA has reviewed the record of CO₂ capture at Boundary Dam Unit 3. While Boundary Dam is in Canada and therefore not subject to this action, these technical challenges have been sufficiently overcome or are actively mitigated so that Boundary Dam has more recently been capable of achieving capture rates of 83 percent when the capture plant is online.²⁹² Furthermore, the improvements already employed and identified at Boundary Dam can be readily applied during the initial construction of a new CO₂ capture plant today.

The CO₂ captured at Boundary Dam is mostly used for EOR and CO₂ is also stored geologically in a deep saline reservoir at the Aquistore site. The amount of flue gas captured is based in part on economic reasons ( i.e., to meet related contract requirements). The incentives for CO₂ capture at Boundary Dam beyond revenue from EOR have been limited to date, and there have been limited regulatory requirements for CO₂ capture at the facility. As a result, a portion (about 25 percent on average) of the flue gas bypasses the capture plant and is emitted untreated. However, because of increasing requirements to capture CO₂ in Canada, Boundary Dam Unit 3 has more recently pursued further process optimization.

Total capture efficiencies at the plant have also been affected by technical issues, particularly with the SO₂ removal system that is upstream of the CO₂ capture system. Operation of the SO₂ removal system affects downstream CO₂ capture and the amount of flue gas that can be processed. Specifically, fly ash (PM) in the flue gas at Boundary Dam Unit 3 contributed to fouling of SO₂ system components, particularly in the SO₂ reboiler and the demisters of the SO₂ absorber column. Buildup of scale in the SO₂ reboiler limited heat transfer and regeneration of the SO₂ scrubbing amine, and high pressure drop affected the flowrate of the SO₂ lean-solvent back to the SO₂ absorber. Likewise, fouling of the demisters in the SO₂ absorber column caused high pressure drop and restricted the flow of flue gas through the system, limiting the amount of flue gas that could be processed by the downstream CO₂ capture system. To address these technical issues, additional wash systems were added, including “demister wash systems, a pre-scrubber flue gas inlet curtain spray wash system, flue gas cooler throat sprays, and a booster fan wash system.”

Such issues will definitively not occur in a different type of SO₂ removal system ( e.g., wet lime scrubber flue gas desulfurization, wet-FGD). SO₂ scrubbers have been successfully operated for decades across a large number of U.S. coal-fired sources. Of the coal-fired sources with planned operation after 2039, 60 percent have wet FGD and 23 percent have a dry FGD. In section VII.C.1.a.ii of this preamble, the EPA accounts for the cost of adding a wet-FGD for those sources that do not have an FGD.

To further mitigate fouling due to fly ash, the PM controls (electrostatic precipitators) at Boundary Dam Unit 3 were upgraded in 2015/2016 by adding switch integrated rectifiers. Of the coal-fired sources with planned operation after 2039, 31 percent have baghouses and 67 percent have electrostatic precipitators. Sources with baghouses have greater or more consistent degrees of emission control, and wet FGD also provides additional PM control.

Fouling at Boundary Dam Unit 3 also affected the heat exchangers in both the SO₂ removal system and the CO₂ capture system. Additional redundancies and isolations to those key components were added in 2017 to allow for online maintenance. Damage to the capture plant's CO₂ compressor resulted in an unplanned outage in 2021, and the issue was corrected. The facility reported 98.3 percent capture system availability in the third quarter of 2023.

Regular maintenance further mitigates fouling in the SO₂ and CO₂ absorbers, and other challenges ( e.g., foaming, biological fouling) typical of gas-liquid absorbers can be mitigated by standard procedures. According to the 2022 paper co-authored by the International CCS Knowledge Centre and SaskPower, “[a] number of initiatives are ongoing or planned with the goal of eliminating flue gas bypass as follows: Since 2016, online cleaning of demisters has been effective at controlling demister pressure; Chemical cleans and replacement of fouled packing in the absorber towers to reduce pressure losses; Optimization of antifoam injection and other aspects of amine health, to minimize foaming potential; [and] Optimization of Liquid-to-Gas (L/G) ratio in the absorber and other process parameters,” as well as other optimization procedures. While foaming is mitigated by an antifoam injection regimen, the EPA further notes that the extent of foaming that could occur may be specific to the chemistry of the solvent and the source's flue gas conditions—foaming was not reported for MHI's KS-1 solvent when treating bituminous coal post-combustion flue gas at Petra Nova. Lastly, while biological fouling in the CO₂ absorber wash water and the SO₂ absorber caustic polisher has been observed, “the current mitigation plan is to perform chemical shocking to remove this particular buildup.”

Based on the experiences of Boundary Dam Unit 3, key improvements can be implemented in future CCS deployments during initial design and construction. Improvements to PM and SO₂ controls can be made prior to operation of the CO₂ capture system. Where fly ash is present in the flue gas, wash systems can be installed to limit associated fouling. Additional redundancies and isolations of key heat-exchangers can be made to allow for in-line cleaning during operation. Redundancy of key equipment ( e.g., utilizing two CO₂ compressor trains instead of one) will further improve operational availability. A feasibility study for the Shand power plant, which is also operated by SaskPower, includes many such design improvements, at an overall cost that was less than the cost for Boundary Dam.

(b) Other Coal-Fired Demonstrations

Several other projects have successfully demonstrated the capture component of CCS at electricity generating plants and other industrial facilities, some of which were previously noted in the discussion in the 2015 NSPS. Since 1978, an amine-based system has been used to capture approximately 270,000 metric tons of CO₂ per year from the flue gas of the bituminous coal-fired steam generating units at the 63 MW Argus Cogeneration Plant (Trona, California). Amine-based carbon capture has further been demonstrated at AES's Warrior Run (Cumberland, Maryland) and Shady Point (Panama, Oklahoma) coal-fired power plants, with the captured CO₂ being sold for use in the food processing industry. At the 180 MW bituminous coal-fired Warrior Run plant, approximately 10 percent of the plant's CO₂ emissions (about 110,000 metric tons of CO₂ per year) has been captured since 2000 and sold to the food and beverage industry. AES's 320 MW Shady Point plant fires subbituminous and bituminous coal, and captured CO₂ from an approximate 5 percent slipstream (about 66,000 metric tons of CO₂ per year) from 2001 through around 2019. These facilities, which have operated for multiple years, clearly show the technical feasibility of post-combustion carbon capture.

(2) EPAct05-Assisted CO₂ Capture Projects at Coal-Fired Steam Generating Units

(a) Petra Nova

Petra Nova is a 240 MW-equivalent capture facility that is the first at-scale application of carbon capture at a coal-fired power plant in the U.S. The system is located at the subbituminous coal- fired W.A. Parish Generating Station in Thompsons, Texas, and began operation in 2017, successfully capturing and sequestering CO₂ for several years. The system was put into reserve shutdown ( i.e., idled) in May 2020, citing the poor economics of utilizing captured CO₂ for EOR at that time. On September 13, 2023, JX Nippon announced that the carbon capture facility at Petra Nova had been restarted. A final report from the National Energy Technology Laboratory (NETL) details the success of the project and what was learned from this first-of-a-kind demonstration at scale. The project used Mitsubishi Heavy Industry's proprietary KM-CDR Process®, a process that is similar to an amine-based solvent process but that uses a proprietary solvent. During its operation, the project successfully captured 92.4 percent of the CO₂ from the slip stream of flue gas processed with 99.08 percent of the captured CO₂ sequestered by EOR.

The amount of flue gas treated at Petra Nova was consistent with a 240 MW size coal-fired steam EGU. The properties of the flue gas—composition, temperature, pressure, density, flowrate, etc. —are the same as would occur for a similarly sized coal-firing unit. Therefore, Petra Nova corroborates that the capture equipment—including the CO₂ absorption column, solvent regeneration column, balance of plant equipment, and the solvent itself—work at commercial scale and can achieve capture rates of 90 percent.

The Petra Nova project did experience periodic outages that were unrelated to the CO₂ capture facility and do not implicate the basis for the EPA's BSER determination. These include outages at either the coal-fired steam generating unit (W.A. Parish Unit 8) or the auxiliary combined cycle facility, extreme weather events (Hurricane Harvey), and the operation of the EOR site and downstream oil recovery and processing. Outages at the coal-fired steam generating unit itself do not compromise the reliability of the CO₂ capture plant or the plant's ability to achieve a standard of performance based on CCS, as there would be no CO₂ to capture. Outages at the auxiliary combined cycle facility are also not relevant to the EPA's BSER determination, because the final BSER is not premised on the CO₂ capture plant using an auxiliary combined cycle plant for steam and power. Rather, the final BSER assumes the steam and power come directly from the associated steam generating unit. Extreme weather events can affect the operation of any facility. Furthermore, the BSER is not premised on EOR, and it is not dependent on downstream oil recovery or processing. Outages attributable to the CO₂ capture facility were 41 days in 2017, 34 days in 2018, and 29 days in 2019—outages decreased year-on-year and were on average less than 10 percent of the year. Planned and unplanned outages are normal for industrial processes, including steam generating units.

Petra Nova experienced some technical challenges that were addressed during its first 3 years of operation. One of these issues was leaks from heat exchangers due to the properties of the gasket materials—replacement of the gaskets addressed the issue. Another issue was vibration of the flue gas blower due to build-up of slurry and solids carryover. W.A. Parish Unit 8 uses a wet limestone FGD scrubber to remove SO₂, and the flue gas connection to the capture plant is located at the bottom of the duct running from the wet-FGD to the original stack. A diversion wall and collection drains were installed to mitigate solids and slurry carryover. Regular maintenance is required to clean affected components and reduce the amount of slurry carryover to the quencher. Solids and slurry carryover also resulted in calcium scale buildup on the flue gas blower. Although calcium concentrations were observed to increase in the solvent, impacts of calcium on the quencher and capture plant chemistry were not observed. Some scaling may have been occurring in the cooling section of the quencher and would have been addressed during a planned outage in 2020. Another issue encountered was scaling related to the CO₂ compressor intercoolers, compressor dehydration system, and an associated heat exchanger. The issue was determined to be due to a material incompatibility of the CO₂ compressor intercooler, and the components were replaced during a 2018 planned outage. To mitigate the scaling prior to the replacement of those components, the compressor drain was also rerouted to the reclaimer and a backup filtering system was also installed and used, both of which proved to be effective. Some decrease in performance was also observed in heat exchangers. The presence of cooling tower fill (a solid medium used to increase surface area in cooling towers) in the cooling water system exchangers may have impacted performance. It is also possible that there could have been some fouling in heat exchangers. Fill was planned to be removed and fouling checked for during regular maintenance. Petra Nova did not observe fouling of the CO₂ absorber packing or high pressure drops across the CO₂ absorber bed, and Petra Nova also did not report any foaming of the solvent. Even with the challenges that were faced, Petra Nova was never restricted in reaching its maximum capture rate of 5,200 tons of CO₂ per day, a scale that was substantially greater than Boundary Dam Unit 3 (approximately 3,600 tons of CO₂ per day).

(b) Plant Barry

Plant Barry, a bituminous coal-fired steam generating unit in Mobile, Alabama, began using the KM-CDR Process® in 2011 for a fully integrated 25 MWe CCS project with a capture rate of 90 percent. The CCS project at Plant Barry captured approximately 165,000 tons of CO₂ annually, which was then transported via pipeline and sequestered underground in geologic formations.

(c) Project Tundra

Project Tundra is a carbon capture project in North Dakota at the Milton R. Young Station lignite coal-fired power plant. Project Tundra will capture up to 4 million metric tons of CO₂ per year for permanent geologic storage. One planned storage site is collocated with the power plant and is already fully permitted, while permitting for a second nearby storage site is in progress. An air permit for the capture facility has also been issued by North Dakota Department of Environmental Quality. The project is designed to capture CO₂ at a rate of about 95 percent of the treated flue gas. The capture plant will treat the flue gas from the 455 MW Unit 2 and additional flue gas from the 250 MW Unit 1, and will treat an equivalent capacity of 530 MW. The project began a final FEED study in February 2023 with planned completion in April 2024, and, prior to selection by DOE for funding award negotiation, the project was scheduled to begin construction in 2024. The project will use MHI's KS-21 solvent and the Advanced KM-CDR process. The MHI solvent KS-1 and an advanced MHI solvent (likely KS-21) were previously tested on the lignite post-combustion flue gas from the Milton R. Young Station. To provide additional conditioning of the flue gas, the project is utilizing a wet electrostatic precipitator (WESP). A draft Environmental Assessment summarizing the project and potential environmental impacts was released by DOE. Finally, Project Tundra was selected for award negotiation for funding from DOE.

That this project has funding through the Bipartisan Infrastructure Law, and that this funding is facilitated through DOE's Office of Clean Energy Demonstration's (OCED) Carbon Capture Demonstration Projects Program, does not detract from the adequate demonstration of CCS. Rather, the goal of that program is, “to accelerate the implementation of integrated carbon capture and storage technologies and catalyze significant follow-on investments from the private sector to mitigate carbon emissions sources in industries across America.” For the commercial scale projects, the stated requirement of the funding opportunity announcement (FOA) is not that projects demonstrate CCS in general, but that they “demonstrate significant improvements in the efficiency, effectiveness, cost, operational and environmental performance of existing carbon capture technologies.” This implies that the basic technology already exists and is already demonstrated. The FOA further notes that the technologies used by the projects receiving funding should be proven such that, “the technologies funded can be readily replicated and deployed into commercial practice.” The EPA also notes that this and other on-going projects were announced well in advance of the FOA. Considering these factors, Project Tundra and other similarly funded projects are supportive of the determination that CCS is adequately demonstrated.

(d) Project Diamond Vault

Project Diamond Vault will capture up to 95 percent of CO₂ emissions from the 600 MW Madison Unit 3 at Brame Energy Center in Lena, Louisiana. Madison Unit 3 fires approximately 70 percent petroleum coke and 30 percent bituminous (Illinois Basin) coal in a circulating fluidized bed. The FEED study for the project is targeted for completion on September 9, 2024. Construction is planned to begin by the end of 2025 with commercial operation starting in 2028. From the utility: “Government Inflation Reduction Act (IRA) funding through 45Q tax credits makes the project financially viable. With these government tax credits, the company does not expect a rate increase as a result of this project.”

(e) Other Projects

Other projects have completed or are in the process of completing feasibility work or FEED studies, or are taking other steps towards installing CCS on coal-fired steam generating units. These projects are summarized in the final TSD, GHG Mitigation Measures for Steam Generating Units, available in the docket. In general, these projects target capture rates of 90 percent or above and provide evidence that sources are actively pursuing the installation of CCS.

(3) CO₂ Capture Technology Vendor Statements

CO₂ capture technology providers have issued statements supportive of the application of systems and solvents for CO₂ capture at fossil fuel-fired EGUs. These statements speak to the decades of experience that technology providers have and as noted below, vendors attest, and offer guarantees that 90 percent capture rates are achievable. Generally, while there are many CO₂ capture methods available, solvent-based CO₂ capture from post-combustion flue gas is particularly applicable to fossil fuel-fired EGUs. Solvent-based CO₂ capture systems are commercially available from technology providers including Shell, Mitsubishi Heavy Industries (MHI), Linde/BASF, Fluor and ION Clean Energy.

Technology providers have made statements asserting extensive experience in CO₂ capture and the commercial availability of CO₂ capture technologies. Solvent-based CO₂ capture was first patented in the 1930s. Since then, commercial solvent-based capture systems have been developed that are focused on applications to post-combustion flue gas. Several technology providers have over 30 years of experience applying solvent-based CO₂ capture to the post-combustion flue gas of fossil fuel-fired EGUs. In general, technology providers describe the technologies for CO₂ capture from post-combustion flue gas as “proven” or “commercially available” or “commercially proven” or “available now” and describe their experience with CO₂ capture from post-combustion flue gas as “extensive.” CO₂ capture rates of 90 percent or higher from post-combustion flue gas have been proven by CO₂ capture technology providers using several commercially available solvents. Many of the available solvent technologies have over 50,000 hours of operation, equivalent to over 5 years of operation.

Shell has decades of experience in CO₂ capture systems. Shell notes that “[c]apturing and safely storing carbon is an option that's available now.” Shell has developed the CANSOLV® CO₂ capture system for CO₂ capture from post-combustion flue gas, a regenerable amine that the company claims has multiple advantages including “low parasitic energy consumption, fast kinetics and extremely low volatility.” Shell further notes, “Moreover, the technology has been designed for reliability through its highly flexible turn-up and turndown capacity.” The company has stated that “Over 90% of the CO₂ in exhaust gases can be effectively and economically removed through the implementation of Shell's carbon capture technology.” Shell also notes, “Systems can be guaranteed for bulk CO₂ removal of over 90%.”

MHI in collaboration with Kansai Electric Power Co., Inc. began developing a solvent-based capture process (the KM CDR Process^TM) using the KS-1^TM solvent in 1990. MHI describes the extensive experience of commercial application of the solvent, “KS-1^TM —a solvent whose high reliability has been confirmed by a track record of deliveries to 15 commercial plants worldwide.” Notable applications of KS-1^TM and the KM-CDR Process^TM include applications at Plant Barry and Petra Nova. Previously, MHI has achieved capture rates of greater than 90 percent over long periods and at full scale at the Petra Nova project where the KS-1^TM solvent was used. MHI has further improved on the original process and solvent by making available the Advanced KM CDR Process^TM using the KS-21^TM solvent. From MHI, “Commercialization of KS-21^TM solvent was completed following demonstration testing in 2021 at the Technology Centre Mongstad in Norway, one of the world's largest carbon capture demonstration facilities.” MHI has achieved CO₂ capture rates of 95 to 98 percent using both the KS-1^TM and KS-21^TM solvent at the Technology Centre Mongstad (TCM). Higher capture rates under modified conditions were also measured, “In addition, in testing conducted under modified operating conditions, the KS-21^TM solvent delivered an industry-leading carbon capture rate was 99.8% and demonstrated the successful recovery of CO₂ from flue gas of lower concentration than the CO₂ contained in the atmosphere.”

Linde engineering in partnership with BASF has made available BASF's OASE® blue amine solvent technology for post-combustion CO₂ capture. Linde notes their experience: “We have longstanding experience in the design and construction of chemical wash processes, providing the necessary amine-based solvent systems and the CO₂ compression, drying and purification system.” Linde also notes that “[t]he BASF OASE® process is used successfully in more than 400 plants worldwide to scrub natural, synthesis and other industrial gases.” The OASE® blue technology has been successfully piloted at RWE Power, Niederaussem, Germany (from 2009 through 2017; 55,000 operating hours) and the National Center for Carbon Capture in Wilsonville, Alabama (January 2015 through January 2016; 3,200 operating hours). Based on the demonstrated performance, Linde concludes that “PCC plants combining Linde's engineering skills and BASF's OASE® blue solvent technology are now commercially available for a wide range of applications.” Linde and BASF have demonstrated capture rates over 90 percent and operating availability rates of more than 97 percent during 55,000 hours of operation.

Fluor provides a solvent technology (Econamine FG Plus) and EPC services for CO₂ capture. Fluor describes their technology as “proven,” noting that, “Proven technology. Fluor Econamine FG Plus technology is a propriety carbon capture solution with more than 30 licensed plants and more than 30 years of operation.” Fluor further notes, “The technology builds on Fluor's more than 400 CO₂ removal units in natural gas and synthesis gas processing.” Fluor further states, “Fluor is a global leader in CO₂ capture [. . .] with long-term commercial operating experience in CO₂ recovery from flue gas.” On the status of Econamine FG Plus, Fluor notes that the “[the] Technology [is] commercially proven on natural gas, coal, and fuel oil flue gases,” and further note that “[o]perating experience includes using steam reformers, gas turbines, gas engines, and coal/natural gas boilers.”

ION Clean Energy is a company focused on post-combustion carbon capture founded in 2008. ION's ICE-21 solvent has been used at NCCC and TCM Norway. ION has achieved capture rates of 98 percent using the ICE-31 solvent.

(4) CCS User Statements on CCS

A number of the companies who have either completed large scale pilot projects or who are currently developing full scale projects have also indicated that CCS technology is currently a viable technology for large coal-fired power plants. In 2011, announcing a decision not to move forward with the first full scale commercial CCS installation of a carbon capture system on a coal plant, AEP did not cite any technology concerns, but rather indicated that “it is impossible to gain regulatory approval to recover our share of the costs for validating and deploying the technology without federal requirements to reduce greenhouse gas emissions already in place.” Enchant Energy, a company developing CCS for coal-fired power plants explained that its FEED study for the San Juan Generating Station, “shows that the technical and business case for adding carbon capture to existing coal-fired power plants is strong.” Rainbow Energy, who is developing a carbon capture project at the Coal Creek Power Station in North Dakota explains, “CCUS technology has been proven and is an economical option for a facility like Coal Creek Station. We see CCUS as the best option to manage CO₂ emissions at our facility.”

(5) State CCS Requirements

Several states encourage or even require sources to install CCS. These state requirements further indicate that CCS is well-established and effective. These state laws include the Illinois 2021 Climate and Equitable Jobs Act, which requires privately owned coal- fired units to reduce emissions to zero by 2030 and requires publicly owned coal-fired units to reduce emissions to zero by 2045. Illinois has also imposed CCS-based CO₂ emission standards on new coal-fired power plants since 2009 when the state adopted its Clean Coal Portfolio Standard law. The statute required an initial capture rate of 50 percent when enacted but steadily increased the capture rate requirement to 90 percent in 2017, where it remains.

Michigan in 2023 established a 100 percent clean energy requirement by 2040 with a nearer term 80 percent clean energy by 2035 requirement. The statute encourages the application of CCS by defining “clean energy” to include generation resources that achieve 90 percent carbon capture.

California identifies carbon capture and sequestration as a necessary tool to reduce GHG emissions within its 2022 scoping plan update and, that same year, enacted a statutory requirement through Assembly Bill 1279 requiring the state to plan and implement policies that enable carbon capture and storage technologies.

Several states in different parts of the country have adopted strategic and planning frameworks that also encourage CCS. Louisiana, which in 2020 set an economy-wide net-zero goal by 2050, has explored policies that encourage CCS deployment in the power sector. The state's 2022 Climate Action Plan proposes a Renewable and Clean Portfolio Standard requiring 100 percent renewable or clean energy by 2035. That proposal defines power plants achieving 90 percent carbon capture as a qualifying clean energy resource that can be used to meet the standard.

Pennsylvania's 2021 Climate Action Plan notes that the state is well positioned to install CCS to transition the state's electric fleet to a zero-carbon economy. The state also established an interagency workgroup in 2019 to identify ways to speed the deployment of CCS.

The Governor of North Dakota announced in 2021 an economy-wide carbon neutral goal by 2030. The announcement singled out the Project Tundra Initiative, which is working to apply CCS technology to the state's Milton R. Young Power Station.

The Governor of Wyoming has broadly promoted a Decarbonizing the West initiative that includes the study of CCS technologies to reduce carbon emissions from the region. A 2024 Wyoming law also requires utilities in the state to install CCS technologies on a portion of their existing coal-fired power plants by 2033.

(6) Variable Load and Startups and Shutdowns

In this section of the preamble, the EPA considers the effects of variable load and startups and shutdowns on the achievability of 90 percent capture. First, the coal-fired steam generating unit can itself turndown to only about 40 percent of its maximum design capacity. Due to this, coal-fired EGUs have relatively high duty cycles —that is, they do not cycle as frequently as other sources and typically have high average loads when operating. In 2021, coal-fired steam generating units had an average duty cycle of 70 percent, and more than 75 percent of units had duty cycles greater than 60 percent. Prior demonstrations of CO₂ capture plants on coal-fired steam generating units have had turndown limits of approximately 60 percent of throughput for Boundary Dam Unit 3 and about 70 percent throughput for Petra Nova. Based on the technology currently available, turndown to throughputs of 50 percent are achievable for a single capture train. Considering that coal units can typically only turndown to 40 percent, a 50 percent turndown ratio for the CO₂ capture plant is likely sufficient for most sources, although utilizing two CO₂ capture trains would allow for turndown to as low as 25 percent of throughput. When operating at less than maximum throughputs, the CO₂ capture facility actually achieves higher capture efficiencies, as evidenced by the data collected at Boundary Dam Unit 3. Data from the Shand Feasibility Report suggests that, for a solvent and design achieving 90 percent capture at 100 percent of net load, 97.5 percent capture is achievable at 62.5 percent of net load. Considering these factors, CO₂ capture is, in general, able to meet the variable load of coal-fired steam generating units without any adverse impact on the CO₂ capture rate. In fact, operation at lower loads may lead to higher achievable capture rates over long periods of time.

Coal-fired steam generating units also typically have few startups and shutdowns per year, and CO₂ emissions during those periods are low. Although capacity factor has declined in recent years, as noted in section IV.D.3 of the preamble, the number of startups per year has been relatively stable. In 2011, coal-fired sources had about 10 startups on average. In 2021, coal-fired steam generating units had only 12 startups on average, see the final TSD, GHG Mitigation Measures for Steam Generating Units, available in the docket. Prior to generation of electricity, coal-fired steam generating units use natural gas or distillate oil—which have a lower carbon content than coal—because of their ignition stability and low ignition temperature. Heat input rates during startup are relatively low, to slowly raise the temperature of the boiler. Existing natural gas- or oil-fired ignitors designed for startup purposes are generally sized for up to 15 percent of the maximum heat-input. Considering the low heat input rate, use of fuel with a lower carbon content, and the relatively few startups per year, the contribution of startup to total GHG emissions is relatively low. Shutdowns are relatively short events, so that the contribution to total emissions are also low. The emissions during startup and shutdown are therefore small relative to emissions during normal operation, so that any impact is averaged out over the course of a year.

Furthermore, the IRC section 45Q tax credit provides incentive for units to operate more. Sources operating at higher capacity factors are likely to have fewer startups and shutdowns and spend less time at low loads, so that their average load would be higher. This would further minimize the insubstantial contribution of startups and shutdowns to total emissions. Additionally, as noted in the preceding sections of the preamble, new solvents achieve capture rates of 95 percent at full load, and ongoing projects are targeting capture rates of 95 percent. Considering all of these factors, startup and shutdown, in general, do not affect the achievability of 90 percent capture over long periods ( i.e., a year).

(7) Coal Rank

CO₂ capture at coal-fired steam generating units achieves 90 percent capture, for the reasons detailed in sections VII.C.1.a.i(B)(1) through (6) of this preamble. Moreover, 90 percent capture is achievable for all coal types because amine solvents have been used to remove CO₂ from a variety of flue gas compositions including a broad range of different coal ranks, differences in CO₂ concentration are slight and the capture process can be designed to the appropriate scale, amine solvents have been used to capture CO₂ from flue gas with much lower CO₂ concentrations, and differences in flue gas impurities due to different coal compositions can be managed or mitigated by controls.

As detailed in the preceding sections, CO₂ capture has been operated on flue gas from the combustion of a broad range of coal ranks including lignite, bituminous, subbituminous, and anthracite coals. Post-combustion CO₂ capture from the flue gas of an EGU firing lignite has been demonstrated at the Boundary Dam Unit 3 EGU (Saskatchewan, Canada). Most lignites have a higher ash and moisture content than other coal types and, in that respect, the flue gas can be more challenging to manage for CO₂ capture. Amine CO₂ capture has also been used to treat lignite post-combustion flue gas in pilot studies at the Milton R. Young station (North Dakota). CO₂ capture solvents have been used to treat subbituminous post-combustion flue gas from W.A. Parish Generating Station (Texas), and the bituminous post-combustion flue gas from Plant Barry (Mobile, Alabama), Warrior Run (Maryland), and Argus Cogeneration Plant (California). Amine solvents have also been used to remove CO₂ from the flue gas of the bituminous- and subbituminous-fired Shady Point plant. CO₂ capture solvents have been used to treat anthracite post-combustion flue gas at the Wilhelmshaven power plant (Germany). There are also ongoing projects that will apply CCS to the flue gas of coal-fired steam generating units. The EPA considers these ongoing projects to be indicative of the confidence that industry stakeholders have in CCS. These include Project Tundra at the lignite-fired Milton R. Young station (North Dakota), Project Diamond Vault at the petroleum coke- and subbituminous-fired Brame Energy Center Madison Unit 3 (Louisiana) and two units at the Jim Bridger Plant (Wyoming).

Different coal ranks have different carbon contents, affecting the concentration of CO₂ in flue gas. In general, however, CO₂ concentration of coal combustion flue gas varies only between 13 and 15 percent. Differences in CO₂ concentration can be accounted for by appropriately designing the capture equipment, including sizing the absorber columns. As detailed in section VIII.F.4.c.iv of the preamble, CO₂ has been captured from the post-combustion flue gas of NGCCs, which typically have a CO₂ concentration of 4 percent.

Prior to emission controls and pre-conditioning, characteristics of different coal ranks and boiler design result in other differences in the flue gas composition, including in the concentration of SO₂, NO_X, PM, and trace impurities. Such impurities in the flue gas can react with the solvent or cause fouling of downstream processes. However, in general, most existing coal-fired steam generating units in the U.S. have controls that are necessary for the pre-conditioning of flue gas prior to the CO₂ capture plant, including PM and SO₂ controls. For those sources without an FGD for SO₂ control, the EPA included the costs of adding an FGD in its cost analysis. Other marginal differences in flue gas impurities can be managed by appropriately designing the polishing column (direct contact cooler) for the individual source's flue gas. Trace impurities can be mitigated using conventional controls in the solvent reclaiming process ( e.g., an activated carbon bed).

Considering the broad range of coal post-combustion flue gases amine solvents have been operated with, that solvents capture CO₂ from flue gases with lower CO₂ concentrations, that the capture process can be designed for different CO₂ concentrations, and that flue gas impurities that may differ by coal rank can be managed by controls, the EPA therefore concludes that 90 percent capture is achievable across all coal ranks, including waste coal.

(8) Natural Gas-Fired Combustion Turbines

Additional information supporting the EPA's determination that 90 percent capture of CO₂ from steam generating units is adequately demonstrated is the experience from CO₂ capture from natural gas-fired combustion turbines. The EPA describes this information in section VIII.F.4.c.iv(B)(1), including explaining how information about CO₂ capture from coal-fired steam generating units also applies to natural gas-fired combustion turbines. The reverse is true as well; information about CO₂ capture from natural gas-fired turbines can be applied to coal fired-units, for much the same reasons.

(9) Summary of Evidence Supporting BSER Determination Without EPAct05-Assisted Projects

As noted above, under the EPA's interpretation of the EPAct05 provisions, the EPA may not rely on capture projects that received assistance under EPAct05 as the sole basis for a determination of adequate demonstration, but the EPA may rely on those projects to support or corroborate other information that supports such a determination. The information described above that supports the EPA's determination that 90 percent CO₂ capture from coal-fired steam generating units is adequately demonstrated, without consideration of the EPAct05-assisted projects, includes (i) the information concerning Boundary Dam, coupled with engineering analysis concerning key improvements that can be implemented in future CCS deployments during initial design and construction ( i.e., all the information in section VII.C.1.a.i.(B)(1)(a) and the information concerning Boundary Dam in section VII.C.1.a.i.(B)(1)(b)); (ii) the information concerning other coal-fired demonstrations, including the Argus Cogeneration Plant and AES's Warrior Run ( i.e., all the information concerning those sources in section VII.C.1.a.i.(B)(1)(a)); (iii) the information concerning industrial applications of CCS ( i.e., all the information in section VII.C.1.a.i.(A)(1); (iv) the information concerning CO₂ capture technology vendor statements ( i.e., all the information in section VII.C.1.a.i.(B)(3)); (v) information concerning carbon capture at natural gas-fired combustion turbines other than EPAct05-assisted projects ( i.e., all the information other than information about EPAct05-assisted projects in section VIII.F.4.c.iv.(B)(1)). All this information by itself is sufficient to support the EPA's determination that 90 percent CO₂ capture from coal-fired steam generating units is adequately demonstrated. Substantial additional information from EPAct05-assisted projects, as described in section VII.C.1.a.i.(B), provides additional support and confirms that 90 percent CO₂ capture from coal-fired steam generating units is adequately demonstrated.

(C) CO₂ Transport

The EPA is finalizing its determination that CO₂ transport by pipelines as a component of CCS is adequately demonstrated. The EPA anticipates that in the coming years, a large-scale interstate pipeline network may develop to transport CO₂ . Indeed, PHMSA is currently engaged in a rulemaking to update and strengthen its safety regulations for CO₂ pipelines, which assumes that such a pipeline network will develop. For purposes of determining the CCS BSER in this final action, however, the EPA did not base its analysis of the availability of CCS on the projected existence of a large-scale interstate pipeline network. Instead, the EPA adopted a more conservative approach. The BSER is premised on the construction of relatively short lateral pipelines that extend from the source to the nearest geologic storage reservoir. While the EPA anticipates that sources would likely avail themselves of an existing interstate pipeline network if one were constructed and that using an existing network would reduce costs, the EPA's analysis focuses on steps that an individual source could take to access CO₂ storage independently.

EGUs that do not currently capture and transport CO₂ will need to construct new CO₂ pipelines to access CO₂ storage sites, or make arrangements with pipeline owners and operators who can do so. Most coal-fired steam EGUs, however, are located in relatively close proximity to deep saline formations that have the potential to be used as long-term CO₂ storage sites. Of existing coal-fired steam generating capacity with planned operation during or after 2039, more than 50 percent is located less than 32 km (20 miles) from potential deep saline sequestration sites, 73 percent is located within 50 km (31 miles), 80 percent is located within 100 km (62 miles), and 91 percent is within 160 km (100 miles). While the EPA's analysis focuses on the geographic availability of deep saline formations, unmineable coal seams and depleted oil and gas reservoirs could also potentially serve as storage formations depending on site-specific characteristics. Thus, for the majority of sources, only relatively short pipelines would be needed for transporting CO₂ from the source to the sequestration site. For the reasons described below, the EPA believes that both new and existing EGUs are capable of constructing CO₂ pipelines as needed. New EGUs may also be planned to be co-located with a storage site so that minimal transport of the CO₂ is required. The EPA has assurance that the necessary pipelines will be safe because the safety of existing and new supercritical CO₂ pipelines is comprehensively regulated by PHMSA.

(1) CO₂ Transport Demonstrations

The majority of CO₂ transported in the United States is moved through pipelines. CO₂ pipelines have been in use across the country for nearly 60 years. Operation of this pipeline infrastructure for this period of time establishes that the design, construction, and operational requirements for CO₂ pipelines have been adequately demonstrated. PHMSA reported that 8,666 km (5,385 miles) of CO₂ pipelines were in operation in 2022, a 14 percent increase in CO₂ pipeline miles since 2011. This pipeline infrastructure continues to expand with a number of anticipated projects underway.

The U.S. CO₂ pipeline network includes major trunkline ( i.e., large capacity) pipelines as well as shorter, smaller capacity lateral pipelines connecting a CO₂ source to a larger trunkline or connecting a CO₂ source to a nearby CO₂ end use. While CO₂ pipelines are generally more economical, other methods of CO₂ transport may also be used in certain circumstances and are detailed in the final TSD, GHG Mitigation Measures for Steam Generating Units.

(a) Distance of CO₂ Transport for Coal-Fired Power Plants

An important factor in the consideration of the feasibility of CO₂ transport from existing coal-fired steam generating units to sequestration sites is the distance the CO₂ must be transported. As discussed in section VII.C.1.a.i(D), potential sequestration formations include deep saline formations, unmineable coal seams, and oil and gas reservoirs. Based on data from DOE/NETL studies of storage resources, of existing coal-fired steam generating capacity with planned operation during or after 2039, 80 percent is within 100 km (62 miles) of potential deep saline sequestration sites, and another 11 percent is within 160 km (100 miles). In other words, 91 percent of this capacity is within 160 km (100 miles) of potential deep saline sequestration sites. In gigawatts, of the 81 GW of coal-fired steam generation capacity with planned operation during or after 2039, only 16 GW is not within 100 km (62 miles) of a potential saline sequestration site, and only 7 GW is not within 160 km (100 mi). The vast majority of these units (on the order of 80 percent) can reach these deep saline sequestration sites by building an intrastate pipeline. This distance is consistent with the distances referenced in studies that form the basis for transport cost estimates for this final rule. While the EPA's analysis focuses on the geographic availability of deep saline formations, unmineable coal seams and depleted oil and gas reservoirs could also potentially serve as storage formations depending on site-specific characteristics.

Of the 9 percent of existing coal-fired steam generating capacity with planned operation during or after 2039 that is not within 160 km (100 miles) of a potential deep saline sequestration site, 5 percent is within 241 km (150 miles) of potential saline sequestration sites, an additional 3 percent is within 322 km (200 miles) of potential saline sequestration sites, and another 1 percent is within 402 km (250 miles) of potential sequestration sites. In total, assuming all existing coal-fired steam generating capacity with planned operation during or after 2039 adopts CCS, the EPA analysis shows that approximately 8,000 km (5,000 miles) of CO₂ pipelines would be constructed by 2032. This includes units located at any distance from sequestration. Note that this value is not optimized for the least total pipeline length, but rather represents the approximate total pipeline length that would be required if each power plant constructed a lateral pipeline connecting their power plant to the nearest potential saline sequestration site.

Additionally, the EPA's compliance modeling projects 3,300 miles of CO₂ pipeline buildout in the baseline and 4,700 miles of pipeline buildout in the policy scenario. This is comparable to the 4,700 to 6,000 miles of CO₂ pipeline buildout estimated by other simulations examining similar scenarios of coal CCS deployment. Over 5 years, this total projected CO₂ pipeline capacity would amount to about 660 to 940 miles per year on average. This projected pipeline mileage is comparable to other types of pipelines that are regularly constructed in the United States each year. For example, based on data collected by EIA, the total annual mileage of natural gas pipelines constructed over the 2017-2021 period ranged from approximately 1,000 to 2,500 miles per year. The projected annual average CO₂ pipeline mileage is less than each year in this historical natural gas pipeline range, and significantly less than the upper end of this range.

The EPA also notes that the pipeline construction estimates presented in this section are not additive with the natural gas co-firing pipeline construction estimates presented below because individual sources will not elect to utilize both compliance methods. In other words, more pipeline buildout for one compliance method necessarily means less pipeline buildout for the other method. Therefore, there is no compliance scenario in which the total pipeline construction is equal to the sum of the CCS and natural gas co-firing pipeline estimates presented in this preamble.

While natural gas line construction may be easier in some circumstances given the uniform federal regulation that governs those such construction, the historical trends support the EPA's conclusion that constructing less CO₂ pipeline length over a several year period is feasible.

(b) CO₂ Pipeline Examples

PHMSA reported that 8,666 km (5,385 miles) of CO₂ pipelines were in operation in 2022. Due to the unique nature of each project, CO₂ pipelines vary widely in length and capacity. Examples of projects that have utilized CO₂ pipelines include the following: Beaver Creek (76 km), Monell (52.6 km), Bairoil (258 km), Salt Creek (201 km), Sheep Mountain (656 km), Slaughter (56 km), Cortez (808 km), Central Basin (231 km), Canyon Reef Carriers (354 km), and Choctaw (294 km). These pipelines range in capacity from 1.6 million tons per year to 27 million tons per year, and transported CO₂ for uses such as EOR.

Most sources deploying CCS are anticipated to construct pipelines that run from the source to the sequestration site. Similar CO₂ pipelines have been successfully constructed and operated in the past. For example, a 109 km (68 mile) CO₂ pipeline was constructed from a fertilizer plant in Coffeyville, Kansas, to the North Burbank Unit, an EOR operation in Oklahoma. Chaparral Energy entered a long-term CO₂ purchase and sale agreement with a subsidiary of CVR Energy for the capture of CO₂ from CVR's nitrogen fertilizer plant in 2011. The pipeline was then constructed, and operations started in 2013. Furthermore, a 132 km (82 mile) pipeline was constructed from the Terrell Gas facility (formerly Val Verde) in Texas to supply CO₂ for EOR projects in the Permian Basin. Additionally, the Kemper Country CCS project in Mississippi, was designed to capture CO₂ from an integrated gasification combined cycle power plant, and transport CO₂ via a 96 km (60 mile) pipeline to be used in EOR. Construction for this facility commenced in 2010 and was completed in 2014. Furthermore, the Citronelle Project in Alabama, which was the largest demonstration of a fully integrated, pulverized coal-fired CCS project in the United States as of 2016, utilized a dedicated 19 km (12 mile) pipeline constructed by Denbury Resources in 2011 to transport CO₂ to a saline storage site.

(c) EPAct05-Assisted CO₂ Pipelines for CCS

Consistent with the EPA's legal interpretation that the Agency can rely on experience from EPAct05 funded facilities in conjunction with other information, this section provides additional examples of CO₂ pipelines with EPAct05 funding. CCS projects with EPAct05 funding have built pipelines to connect the captured CO₂ source with sequestration sites, including Illinois Industrial Carbon Capture and Storage in Illinois, Petra Nova in Texas, and Red Trail Energy in North Dakota. The Petra Nova project, which restarted operations in September 2023, transports CO₂ via a 131 km (81 mile) pipeline to the injection site, while the Illinois Industrial Carbon Capture project and Red Trail Energy transport CO₂ using pipelines under 8 km (5 miles) long. Additionally, Project Tundra, a saline sequestration project planned at the lignite-fired Milton R. Young Station in North Dakota will transport CO₂ via a 0.4 km (0.25 mile) pipeline.

(d) Existing and Planned CO₂ Trunklines

Although the BSER is premised on the construction of pipelines that connect the CO₂ source to the sequestration site, in practice some sources may construct short laterals to existing CO₂ trunklines, which can reduce the number of miles of pipeline that may need to be constructed. A map displaying both existing and planned CO₂ pipelines, overlayed on potential geologic sequestration sites, is available in the final TSD, GHG Mitigation Measures for Steam Generating Units. Pipelines connect natural CO₂ sources in south central Colorado, northeast New Mexico, and Mississippi to oil fields in Texas, Oklahoma, New Mexico, Utah, and Louisiana. The Cortez pipeline is the longest CO₂ pipeline, and it traverses over 800 km (500) miles from southwest Colorado to Denver City, Texas CO₂ Hub, where it connects with several other CO₂ pipelines. Many existing CO₂ pipelines in the U.S. are located in the Permian Basin region of west Texas and eastern New Mexico. CO₂ pipelines in Wyoming, Texas, and Louisiana also carry CO₂ captured from natural gas processing plants and refineries to EOR projects. Additional pipelines have been constructed to meet the demand for CO₂ transportation. A 170 km (105 mile) CO₂ pipeline owned by Denbury connecting oil fields in the Cedar Creek Anticline (located along the Montana-North Dakota border) to CO₂ produced in Wyoming was completed in 2021, and a 30 km (18 mile) pipeline also owned by Denbury connects to the same oil field and was completed in 2022. These pipelines form a network with existing pipelines in the region—including the Denbury Greencore pipeline, which was completed in 2012 and is 232 miles long, running from the Lost Cabin gas plant in Wyoming to Bell Creek Field in Montana.

In addition to the existing pipeline network, there are a number of large CO₂ trunklines that are planned or in progress, which could further reduce the number of miles of pipeline that a source may need to construct. Several major projects have recently been announced to expand the CO₂ pipeline network across the United States. For example, the Summit Carbon Solutions Midwest Carbon Express project has proposed to add more than 3,200 km (2,000) miles of dedicated CO₂ pipeline in Iowa, Nebraska, North Dakota, South Dakota, and Minnesota. The Midwest Carbon Express is projected to begin operations in 2026. Further, Wolf Carbon Solutions has recently announced that it plans to refile permit applications for the Mt. Simon Hub, which will expand the CO₂ pipeline by 450 km (280 miles) in the Midwest. Tallgrass announced in 2022 a plan to convert an existing 630 km (392 mile) natural gas pipeline to carry CO₂ from an ADM ethanol production facility in Nebraska to a planned commercial-scale CO₂ sequestration hub in Wyoming aimed for completion in 2024. Recently, as part of agreeing to a communities benefits plan, a number of community groups have agreed that they will support construction of the Tallgrass pipeline in Nebraska. While the construction of larger networks of trunklines could facilitate CCS for power plants, the BSER is not predicated on the buildout of a trunkline network and the existence of future trunklines was not assumed in the EPA's feasibility or costing analysis. The EPA's analysis is conservative in that it does not presume the buildout of trunkline networks. The development of more robust and interconnected pipeline systems over the next several years would merely lower the EPA's cost projections and create additional CO₂ transport options for power plants that do CCS.

Moreover, pipeline projects have received funding under the IIJA to conduct front-end engineering and design (FEED) studies. Carbon Solutions LLC received funding to conduct a FEED study for a commercial-scale pipeline to transport CO₂ in support of the Wyoming Trails Carbon Hub as part of a statewide pipeline system that would be capable of transporting up to 45 million metric tons of CO₂ per year from multiple sources. In addition, Howard Midstream Energy Partners LLC received funding to conduct a FEED study for a 965 km (600 mi) CO₂ pipeline system on the Gulf Coast that would be capable of moving at least 250 million metric tons of CO₂ annually and connecting carbon sources within 30 mi of the trunkline.

Other programs were created by the IIJA to facilitate the buildout of large pipelines to carry carbon dioxide from multiple sources. For example, the Carbon Dioxide Transportation Infrastructure Finance and Innovation Act (CIFIA) was incorporated into the IIJA and provided $2.1 billion to DOE to finance projects that build shared ( i.e., common carrier) transport infrastructure to move CO₂ from points of capture to conversion facilities and/or storage wells. The program offers direct loans, loan guarantees, and “future growth grants” to provide cash payments to specifically for eligible costs to build additional capacity for potential future demand.

(2) Permitting and Rights of Way

The permitting process for CO₂ pipelines often involves a number of private, local, state, tribal, and/or Federal agencies. States and local governments are directly involved in siting and permitting proposed CO₂ pipeline projects. CO₂ pipeline siting and permitting authorities, landowner rights, and eminent domain laws are governed by the states and vary by state.

State laws determine pipeline siting and the process for developers to acquire rights-of-way needed to build. Pipeline developers may secure rights-of-way for proposed projects through voluntary agreements with landowners; pipeline developers may also secure rights-of-way through eminent domain authority, which typically accompanies siting permits from state utility regulators with jurisdiction over CO₂ pipeline siting. The permitting process for interstate pipelines may take longer than for intrastate pipelines. Whereas multiple state regulatory agencies would be involved in the permitting process for an interstate pipeline, only one primary state regulatory agency would be involved in the permitting process for an intrastate pipeline.

Most regulation of CO₂ pipeline siting and development is conducted at the state level, and under state specific regulatory regimes. As the interest in CO₂ pipelines has grown, states have taken steps to facilitate pipeline siting and construction. State level regulation related to CO₂ sequestration and transport is an very active area of legislation across states in all parts of the country, with many states seeking to facilitate pipeline siting and construction. Many states, including Kentucky, Michigan, Montana, Arkansas, and Rhode Island, treat CO₂ pipeline operators as common carriers or public utilities. This is an important classification in some jurisdictions where it may be required for pipelines seeking to exercise eminent domain. Currently, 17 states explicitly allow CO₂ pipeline operators to exercise eminent domain authority for acquisition of CO₂ pipeline rights-of-way, should developers not secure them through negotiation with landowners. Some states have recognized the need for a streamlined CO₂ pipeline permitting process when there are multiple layers of regulation and developed joint permit applications. Illinois, Louisiana, New York, and Pennsylvania have created a joint permitting form that allows applicants to file a single application for pipeline projects covering both state and federal permitting requirements. Even in states without this streamlined process, pipeline developers can pursue required state permits concurrently with federal permits, NEPA review (as applicable), and the acquisition of rights-of-way.

Pipeline developers have been able to successfully secure the necessary rights-of way for CO₂ pipeline projects. For example, Summit Carbon Solutions, which has proposed to add more than 3,200 km (2,000 mi) of dedicated CO₂ pipeline in Iowa, Nebraska, North Dakota, South Dakota, and Minnesota, has stated that as of November 7, 2023, it had reached easement agreements with 2,100 landowners along the route. As of February 23, 2024, Summit Carbon Solutions stated that it had acquired about 75 percent of the rights of way needed in Iowa, about 80 percent in North Dakota, about 75 percent in South Dakota, and about 89 percent in Minnesota. The company has successfully navigated hurdles, such as rerouting the pipelines in certain counties where necessary. The EPA notes that this successful acquisition of right-of-way easements for thousands of miles of pipeline across five states has taken place in just the three years since the project launched in 2021. In addition, the Citronelle Project, which was constructed in Alabama in 2011, successfully acquired rights-of-way through 9 miles of forested and commercial timber land and 3 miles of emergent shrub and forested wetlands. The Citronelle Project was able to attain rights-of-way through the habitat of an endangered species by mitigating potential environmental impacts. Even projects that require rights-of-way across multiple ownership regimes including state, private, and federally owned land have been successfully developed. The 170 km (105 mile) Cedar Creek Anticline CO₂ pipeline owned by Denbury required easements for approximately 10 km (6.2 mi) to cross state school trust lands in Montana, 27 km (17 mi) across Federal land and the remaining miles across private lands. The pipeline was completed in 2021.

Federal actions ( e.g., funding a CCS project) must generally comply with NEPA, which often requires that an environmental assessment (EA) or environmental impact statement (EIS) be conducted to consider environmental impacts of the proposed action, including consideration of reasonable alternatives. An EA determines whether or not a Federal action has the potential to cause significant environmental effects. Each Federal agency has adopted its own NEPA procedures for the preparation of EAs. If the agency determines that the action will not have significant environmental impacts, the agency will issue a Finding of No Significant Impact (FONSI). Some projects may also be “categorically excluded” from a detailed environmental analysis when the Federal action normally does not have a significant effect on the human environment. Federal agencies prepare an EIS if a proposed Federal action is determined to significantly affect the quality of the human environment. The regulatory requirements for an EIS are more detailed and rigorous than the requirements for an EA. The determination of the level of NEPA review depends on the potential for significant environmental impacts considering the whole project ( e.g., crossings of sensitive habitats, cultural resources, wetlands, public safety concerns). Consequently, whether a pipeline project is covered by NEPA and the associated permitting timelines may vary depending on site characteristics ( e.g., pipeline length, whether a project crosses a water of the U.S.) and funding source. Pipelines through Bureau of Land Management (BLM) land, U.S. Forest Service (USFS) land, or other Federal land would be subject to NEPA. To ensure that agencies conduct NEPA reviews as efficiently and expeditiously as practicable, the Fiscal Responsibility Act amendments to NEPA established deadlines for the preparation of environmental assessments and environmental impact statements. Environmental assessments must be completed within 1 year and environmental impact statements must be completed within 2 years A lead agency that determines it is not able to meet the deadline may extend the deadline, in consultation with the applicant, to establish a new deadline that provides only so much additional time as is necessary to complete such environmental impact statement or environmental assessment.

As discussed above, it is anticipated that most EGUs would need shorter, intrastate pipeline segments. For example, ADM's Decatur, Illinois, pipeline, which spans 1.9 km (1.18 miles), was constructed after Decatur was selected for the DOE Phase 1 research and development grants in October 2009. Construction of the CO₂ compression, dehydration, and pipeline facilities began in July 2011 and was completed in June 2013. The ADM project required only an EA. Additionally, Air Products operates a large-scale system to capture CO₂ from two steam methane reformers located within the Valero Refinery in Port Arthur, Texas. The recovered and purified CO₂ is delivered by pipeline for use in enhanced oil recovery operations. This 12-mile pipeline required only an EA. Conversely, the Petra Nova project in Texas required an EIS to evaluate the potential environmental impacts associated with DOE's proposed action of providing financial assistance for the project. This EIS addressed potential impacts from both the associated 131 km (81 mile) pipeline and other aspects of the larger CCS system, including the post-combustion CO₂ . For Petra Nova, a notice of intent to issue an EIS was published on November 14, 2011, and the record of decision was issued less than 2 years later, on May 23, 2013. Construction of the CO₂ pipeline for Petra Nova from the W.A. Parish Power Plant to the West Ranch Oilfield in Jackson County, TX began in July 2014 and was completed in July 2016.

Compliance with section 7 of the Endangered Species Act related to Federal agency consultation and biological assessment is also required for projects on Federal lands. Specifically, the Endangered Species Act requires consultation with the Department of Interior's Fish and Wildlife Service and Department of Commerce's NOAA Fisheries, in order to avoid or mitigate impacts to any threatened or endangered species and their habitats. This agency consultation process and biological assessment are generally conducted during preparation of the NEPA documentation (EIS or EA) for the Federal project and generally within the regulatory timeframes for environmental assessment or environmental impact statement preparation. Consequently, the EPA does not anticipate that compliance with the Endangered Species Act will change the anticipated timeline for most projects.

The EPA notes that the Fixing America's Surface Transportation Act (FAST Act) is also relevant to CCS projects and pipelines. Title 41 of this Act (42 U.S.C. 4370m et seq.), referred to as “FAST-41,” created a new governance structure, set of procedures, and funding authorities to improve the Federal environmental review and authorization process for covered infrastructure projects. The Utilizing Significant Emissions with Innovative Technologies (USE IT) Act, among other actions, clarified that CCS projects and CO₂ pipelines are eligible for this more predictable and transparent review process. FAST-41 created the Federal Permitting Improvement Steering Council (Permitting Council), composed of agency Deputy Secretary-level members and chaired by an Executive Director appointed by the President. FAST-41 establishes procedures that standardize interagency consultation and coordination practices. FAST-41 codifies into law the use of the Permitting Dashboard to track project timelines, including qualifying actions that must be taken by the EPA and other Federal agencies. Project sponsor participation in FAST-41 is voluntary.

Community engagement also plays a role in the safe operation and construction of CO₂ pipelines. These efforts can be supported using the CCS Pipeline Route Planning Database that was developed by NETL, a public resource designed to support pipeline routing decisions and increase transportation safety. The database includes state-specific regulations and restrictions, energy and social justice factors, land use requirements, existing infrastructure, and areas of potential risk. The database produces weighted values ranging from zero to one, where zero represents acceptable areas for pipeline placement and one represents areas that should be avoided. The database will be a key input for the CCS Pipeline Route Planning Tool under development by NETL. The purpose of the siting tool is to aid pipeline routing decisions and facilitate avoidance of areas that would pose permitting challenges.

In sum, the permitting process for CO₂ pipelines often involves private, local, state, tribal, and/or Federal agencies, and permitting timelines may vary depending on site characteristics. Projects that opt in to the FAST-41 process are eligible for a more transparent and predictable review process. EGUs can generally proceed to obtain permits and rights-of-way simultaneously, and the EPA anticipates that, in total, the permitting process would only take around 2.5 years for pipelines that only need an EA, with a possible additional year if the project requires an EIS (see the final TSD, GHG Mitigation Measures for Steam Generating Units for additional information). This is consistent with the anticipated timelines for CCS discussed in section VII.C.1.a.i(E). Furthermore, the EPA notes that there is over 60 years of experience in the CO₂ pipeline industry designing, permitting, building and operating CO₂ pipelines, and that this expertise can be applied to the CO₂ pipelines that would be constructed to connect to sequestration sites and units.

As discussed above in section VII.C.1.a.i.(C)(1)(a), the core of the EPA's analysis of pipeline feasibility focuses on units located within 100 km (62 miles) of potential deep saline sequestration formations. The EPA notes that the majority (80 percent) of the coal-fired steam generating capacity with planned operation during or after 2039 is located within 100 km (62 miles) of the nearest potential deep saline sequestration site. For these sources, as explained, units would be required only to build relatively short pipelines, and such buildout would be feasible within the required timeframe. For the capacity that is more than 100 km (62 miles) away from sequestration, building a pipeline may become more complex. Almost all (98 percent) of this capacity's closest sequestration site is located outside state boundaries, and access to the nearest sequestration site would require building an interstate pipeline and coordinating with multiple state authorities for permitting purposes. Conversely, for capacity where the distance to the nearest potential sequestration site is less than 100 km (62 miles), only about 19 percent would require the associated pipeline to cross state boundaries. Therefore, the EPA believes that distance to the nearest sequestration site is a useful proxy for considerations related to the complexity of pipeline construction and how long it will take to build a pipeline.

A unit that is located more than 100 km away from sequestration may face complexities in pipeline construction, including additional permitting hurdles, difficulties in obtaining the necessary rights of way over such a distance, or other considerations, that may make it unreasonable for that unit to meet the compliance schedule that is generally reasonable for sources in the subcategory as a whole. Pursuant to the RULOF provisions of 40 CFR 60.2a(e)-(h), if a state can demonstrate that there is a fundamental difference between the information relevant to a particular affected EGU and the information the EPA considered in determining the compliance deadline for sources in the long-term subcategory, and that this difference makes it unreasonable for the EGU to meet the compliance deadline, a longer compliance schedule may be warranted. The EPA does not believe that the fact that a pipeline crosses state boundaries standing alone is sufficient to show that an extended timeframe would be appropriate—many such pipelines could be reasonably accomplished in the required timeframe. Rather, it is the confluence of factors, including that a pipeline crosses state boundaries, along with others that may make RULOF appropriate.

(3) Security of CO₂ Transport

As part of its analysis, the EPA also considered the safety of CO₂ pipelines. The safety of existing and new CO₂ pipelines that transport CO₂ in a supercritical state is regulated by PHMSA. These regulations include standards related to pipeline design, pipeline construction and testing, pipeline operations and maintenance, operator reporting requirements, operator qualifications, corrosion control and pipeline integrity management, incident reporting and response, and public awareness and communications. PHMSA has regulatory authority to conduct inspections of supercritical CO₂ pipeline operations and issue notices to operators in the event of operator noncompliance with regulatory requirements.

CO₂ pipelines have been operating safely for more than 60 years. In the past 20 years, 500 million metric tons of CO₂ moved through over 5,000 miles of CO₂ pipelines with zero incidents involving fatalities. PHMSA reported a total of 102 CO₂ pipeline incidents between 2003 and 2022, with one injury (requiring in-patient hospitalization) and zero fatalities.

As noted previously in this preamble, a significant CO₂ pipeline rupture occurred in 2020 in Satartia, Mississippi, following heavy rains that resulted in a landslide. Although no one required in-patient hospitalization as a result of this incident, 45 people received treatment at local emergency rooms after the incident and 200 hundred residents were evacuated. Typically, when CO₂ is released into the open air, it vaporizes into a heavier-than-air gas and dissipates. During the Satartia incident, however, unique atmospheric conditions and the topographical features of the area delayed this dissipation. As a result, residents were exposed to high concentrations of CO₂ in the air after the rupture. Furthermore, local emergency responders were not informed by the operator of the rupture and the nature of the unique safety risks of the CO₂ pipeline.

PHMSA initiated a rulemaking in 2022 to develop and implement new measures to strengthen its safety oversight of supercritical CO₂ pipelines following the investigation into the CO₂ pipeline failure in Satartia. PHMSA submitted the associated Notice of Proposed Rulemaking to the White House Office of Management and Budget on February 1, 2024 for pre-publication review. Following the Satartia incident, PHMSA also issued a Notice of Probable Violation, Proposed Civil Penalty, and Proposed Compliance Order (Notice) to the operator related to probable violations of Federal pipeline safety regulations. The Notice was ultimately resolved through a Consent Agreement between PHMSA and the operator that includes the assessment of civil penalties and identifies actions for the operator to take to address the alleged violations and risk conditions. PHMSA has further issued an updated nationwide advisory bulletin to all pipeline operators and solicited research proposals to strengthen CO₂ pipeline safety. Given the Federal and state regulation of CO₂ pipelines and the steps that PHMSA is taking to further improve pipeline safety, the EPA believes CO₂ can be safely transported by pipeline.

Certain states have authority delegated from the U.S. Department of Transportation to conduct safety inspections and enforce state and Federal pipeline safety regulations for intrastate CO₂ pipelines. PHMSA's state partners employ about 70 percent of all pipeline inspectors, which covers more than 80 percent of regulated pipelines. Federal law requires certified state authorities to adopt safety standards at least as stringent as the Federal standards. Further, there are required steps that CO₂ pipeline operators must take to ensure pipelines are operated safely under PHMSA standards and related state standards, such as the use of pressure monitors to detect leaks or initiate shut-off valves, and annual reporting on operations, structural integrity assessments, and inspections. These CO₂ pipeline controls and PHMSA standards are designed to ensure that captured CO₂ will be securely conveyed to a sequestration site.

(4) Comments Received on CO₂ Transport and Responses

The EPA received comments on CO₂ transport, including CO₂ pipelines. Those comments, and the EPA's responses, are as follows.

Comment: Some commenters identified challenges to the deployment of a national, interstate CO₂ pipeline network. In particular, those commenters discussed the experience faced by long ( e.g., over 1,000 miles) CO₂ pipelines seeking permitting and right-of-way access in Midwest states including Iowa and North Dakota. Commenters claimed those challenges make CCS as BSER infeasible. Some commenters argued that the existing CO₂ pipeline capacity is not adequate to meet potential demand caused by this rule and that the ability of the network to grow and meet future potential demand is hindered by significant public opposition.

Response: The EPA acknowledges the challenges that some large multi-state pipeline projects have faced, but does not agree that those experiences show that the BSER is not adequately demonstrated or that the standards finalized in these actions are not achievable. As detailed in the preceding subsections of the preamble, the BSER is not premised on the buildout of a national, trunkline CO₂ pipeline network. Most coal-fired steam generating units are in relatively close proximity to geologic storage, and those shorter pipelines would not likely be as challenging to permit and build as demonstrated by the examples of smaller pipeline discussed above.

The EPA acknowledges that some larger trunkline CO₂ pipeline projects, specifically the Heartland Greenway project, have recently been delayed or canceled. However, many projects are still moving forward and several major projects have recently been announced to expand the CO₂ pipeline network across the United States. The EPA notes that there are often opportunities to reroute pipelines to minimize permitting challenges and landowner concerns. For example, Summit Carbon Solutions changed their planned pipeline route in North Dakota after their initial permit was denied, leading to successful acquisition of rights of way. Additionally, Tallgrass, which is planning to convert a 630 km (392 mile) natural gas pipeline to carry CO₂ , announced that they had reach a community benefits agreement, in which certain organizations have agreed not to oppose the pipeline project while Tallgrass has agreed to terms such as contributing funds to first responders along the pipeline route and providing royalty checks to landowners. See section VII.C.1.a.i(C)(1)(d) for additional discussion of planned CO₂ pipelines. While access to larger trunkline projects would not be required for most EGUs, at least some larger trunkline projects are likely to be constructed, which would increase opportunities for connecting to pipeline networks.

Comment: Some commenters disagreed with the modeling assumption that 100 km is a typical pipeline distance. The commenters asserted that there is data showing the actual locations of the power plants affected by the rule, and the required pipeline distance is not always 100 km.

Response: The EPA acknowledges that the physical locations of EGUs and the physical locations of carbon sequestration capacity and corresponding pipeline distance will not be 100 km in all cases. As discussed previously in section VII.C.1.a.i(C)(1)(a), the EPA modeled the unique approximate distance from each existing coal-fired steam generating capacity with planned operation during or after 2039 to the nearest potential saline sequestration site, and found that the majority (80 percent) is within 100 km (62 miles) of potential saline sequestration sites, and another 11 percent is within 160 km (100 miles). Furthermore, the EPA disagrees with the comments suggesting that the use of 100 km is an inappropriate economic modeling assumption. The 100 km assumption was not meant to encompass the physical location of every potentially affected EGU. The 100 km assumption is intended as an economic modeling assumption and is based on similar assumptions applied in NETL studies used to estimate CO₂ transport costs. The EPA carefully reviewed the assumptions on which the NETL transport cost estimates are based and continues to find them reasonable. The NETL studies referenced in section VII.C.1.a.ii based transport costs on a generic 100 km (62 mile) pipeline and a generic 80 km pipeline. For most EGUs, the necessary pipeline distance is anticipated to be less than 100 km and therefore the associated costs could also be lower than these assumptions. Other published economic models applying different assumptions have also reached the conclusion that CO₂ transport and sequestration are adequately demonstrated.

Comment: Commenters also stated that the permitting and construction processes can be time-consuming.

Response: The EPA acknowledges building CO₂ pipelines requires capital expenditure and acknowledges that the timeline for siting, engineering design, permitting, and construction of CO₂ pipelines depends on factors including the pipeline capacity and pipeline length, whether the pipeline route is intrastate or interstate, and the specifics of the state pipeline regulator's regulatory requirements. In the BSER analysis, individual EGUs that are subject to carbon capture requirements are assumed to take a point-to-point approach to CO₂ transport and sequestration. These smaller-scale projects require less capital and may present less complexity than larger projects. The EPA considers the timeline to permit and install such pipelines in section VII.C.1.a.i(E) of the preamble, and has determined that a compliance date of January 1, 2032 allows for a sufficient amount of time.

Comment: Some commenters expressed significant concerns about the safety of CO₂ pipelines following the CO₂ pipeline failure in Satartia, Mississippi in 2020.

Response: For a discussion of the safety of CO₂ pipelines and the Satartia pipeline failure, see section VII.C.1.a.i(C)(3). The EPA believes that the framework of Federal and state regulation of CO₂ pipelines and the steps that PHMSA is taking to further improve pipeline safety, is sufficient to ensure CO₂ can be safely transported by pipeline.

(D) Geologic Sequestration of CO₂

The EPA is finalizing its determination that geologic sequestration ( i.e., the long-term containment of a CO₂ stream in subsurface geologic formations) is adequately demonstrated. In this section, we provide an overview of the availability of sequestration sites in the U.S., discuss how geologic sequestration of CO₂ is well proven and broadly available throughout the U.S, explain the effectiveness of sequestration, discuss the regulatory framework for UIC wells, and discuss the timing of permitting for sequestration sites. We then provide a summary of key comments received concerning geologic sequestration and our responses to those comments.

(1) Sequestration Sites for Coal-Fired Power Plants Subject to CCS Requirements

(a) Broad Availability of Sequestration

Sequestration is broadly available in the United States, which makes clear that it is adequately demonstrated. By far the most widely available and well understood type of sequestration is that in deep saline formations. These formations are common in the U.S. These formations are numerous and only a small subset of the existing saline storage capacity would be required to store the CO₂ from EGUs. Many projects are in the process of completing thorough subsurface studies of these deep saline formations to determine their suitability for regional-scale storage. Furthermore, sequestration formations could also include unmineable coal seams and oil and gas reservoirs. CO₂ may be stored in oil and gas reservoirs in association with EOR and enhanced gas recovery (EGR) technologies, collectively referred to as enhanced recovery (ER), which include the injection of CO₂ in oil and gas reservoirs to increase production. ER is a technology that has been used for decades in states across the U.S.

Geologic sequestration is based on a demonstrated understanding of the trapping and containment processes that retain CO₂ in the subsurface. The presence of a low permeability seal is an important component of demonstrating secure geologic sequestration. Analyses of the potential availability of geologic sequestration capacity in the United States have been conducted by DOE, and the U.S. Geological Survey (USGS) has also undertaken a comprehensive assessment of geologic sequestration resources in the United States. Geologic sequestration potential for CO₂ is widespread and available throughout the United States. Nearly every state in the United States has or is in close proximity to formations with geologic sequestration potential, including areas offshore. There have been numerous efforts demonstrating successful geologic sequestration projects in the United States and overseas, and the United States has developed a detailed set of regulatory requirements to ensure the security of sequestered CO₂ . Moreover, the amount of storage potential can readily accommodate the amount of CO₂ for which sequestration could be expected under this final rule.

The EPA has performed a geographic availability analysis in which the Agency examined areas of the U.S. with sequestration potential in deep saline formations, unmineable coal seams, and oil and gas reservoirs; information on existing and probable, planned or under study CO₂ pipelines; and areas within a 100 km (62-mile) area of potential sequestration sites. This availability analysis is based on resources from the DOE, the USGS, and the EPA. The distance of 100 km is consistent with the assumptions underlying the NETL cost estimates for transporting CO₂ by pipeline. The scoping assessment by the EPA found that at least 37 states have geologic characteristics that are amenable to deep saline sequestration, and an additional 6 states are within 100 kilometers of potentially amenable deep saline formations in either onshore or offshore locations. Of the 7 states that are further than 100 km (62 mi) of onshore or offshore storage potential in deep saline formations, only New Hampshire has coal EGUs that were assumed to be in operation after 2039, with a total capacity of 534 MW. However, the EPA notes that as of March 27, 2024, the last coal-fired steam EGUs in New Hampshire announced that they would cease operation by 2028. Therefore, the EPA anticipates that there will no existing coal-fired steam EGUs located in states that are further than 100 km (62 mi) of potential geologic sequestration sites. Furthermore, as described in section VII.C.1.a.i(C), new EGUs would have the ability to consider proximity and access to geologic sequestration sites or CO₂ pipelines in the siting process.

The DOE and the United States Geological Survey (USGS) have independently conducted preliminary analyses of the availability and potential CO₂ sequestration resources in the United States. The DOE estimates are compiled in the DOE's National Carbon Sequestration Database and Geographic Information System (NATCARB) using volumetric models and are published in its Carbon Utilization and Sequestration Atlas (NETL Atlas). The DOE estimates that areas of the United States with appropriate geology have a sequestration potential of at least 2,400 billion to over 21,000 billion metric tons of CO₂ in deep saline formations, unmineable coal seams, and oil and gas reservoirs. The USGS assessment estimates a mean of 3,000 billion metric tons of subsurface CO₂ sequestration potential across the United States. With respect to deep saline formations, the DOE estimates a sequestration potential of at least 2,200 billion metric tons of CO₂ in these formations in the United States. The EPA estimates that the CO₂ emissions reductions for this rule (which is similar to the amount of CO₂ may be sequestered under this rule) are estimated in the range of 1.3 to 1.4 billion metric tons over the 2028 to 2047 timeframe. This volume of sequestered CO₂ is less than a tenth of a percent of the storage capacity in deep saline formations estimated to be available by DOE.

Unmineable coal seams offer another potential option for geologic sequestration of CO₂ . Enhanced coalbed methane recovery is the process of injecting and storing CO₂ in unmineable coal seams to enhance methane recovery. These operations take advantage of the preferential chemical affinity of coal for CO₂ relative to the methane that is naturally found on the surfaces of coal. When CO₂ is injected, it is adsorbed to the coal surface and releases methane that can then be captured and produced. This process effectively “locks” the CO₂ to the coal, where it remains stored. States with the potential for sequestration in unmineable coal seams include Iowa and Missouri, which have little to no saline sequestration potential and have existing coal-fired EGUs. Unmineable coal seams have a sequestration potential of at least 54 billion metric tons of CO₂, or 2 percent of total potential in the United States, and are located in 22 states.

The potential for CO₂ sequestration in unmineable coal seams has been demonstrated in small-scale demonstration projects, including the Allison Unit pilot project in New Mexico, which injected a total of 270,000 tons of CO₂ over a 6-year period (1995-2001). Further, DOE Regional Carbon Sequestration Partnership projects have injected CO₂ volumes in unmineable coal seams ranging from 90 tons to 16,700 tons, and completed site characterization, injection, and post-injection monitoring for sites. DOE has included unmineable coal seams in the NETL Atlas. One study estimated that in the United States, 86.16 billion tons of CO₂ could be permanently stored in unmineable coal seams. Although the large-scale injection of CO₂ in coal seams can lead to swelling of coal, the literature also suggests that there are available technologies and techniques to compensate for the resulting reduction in injectivity. Further, the reduced injectivity can be anticipated and accommodated in sizing and characterizing prospective sequestration sites.

Depleted oil and gas reservoirs present additional potential for geologic sequestration. The reservoir characteristics of developed fields are well known as a result of exploration and many years of hydrocarbon production and, in many areas, infrastructure already exists which could be evaluated for conversion to CO₂ transportation and sequestration service. Other types of geologic formations such as organic rich shale and basalt may also have the ability to store CO₂, and DOE is continuing to evaluate their potential sequestration capacity and efficacy.

(b) Inventory of Coal-Fired Power Plants That Are Candidates for CCS

Sequestration potential as it relates to distance from existing coal-fired steam generating units is a key part of the EPA's regular power sector modeling, using data from DOE/NETL studies. As discussed in section VII.C.1.a.i(D)(1)(a), the availability analysis shows that of the coal-fired steam generating capacity with planned operation during or after 2039, more than 50 percent is less than 32 km (20 miles) from potential deep saline sequestration sites, 73 percent is located within 50 km (31 miles), 80 percent is located within 100 km (62 miles), and 91 percent is within 160 km (100 miles).

(2) Geologic Sequestration of CO₂ Is Adequately Demonstrated

Geologic sequestration is based on a demonstrated understanding of the processes that affect the fate of CO₂ in the subsurface. Existing project and regulatory experience, along with other information, indicate that geologic sequestration is a viable long-term CO₂ sequestration option. As discussed in this section, there are many examples of projects successfully injecting and containing CO₂ in the subsurface.

Research conducted through the Department of Energy's Regional Carbon Sequestration Partnerships has demonstrated geologic sequestration through a series of field research projects that increased in scale over time, injecting more than 12 million tons of CO₂ with no indications of negative impacts to either human health or the environment. Building on this experience, DOE launched the Carbon Storage Assurance Facility Enterprise (CarbonSAFE) Initiative in 2016 to demonstrate how knowledge from the Regional Carbon Sequestration Partnerships can be applied to commercial-scale safe storage. This initiative is furthering the development and refinement of technologies and techniques critical to the characterization of sites with the potential to sequester greater than 50 million tons of CO₂ . In Phase I of CarbonSAFE, thirteen projects conducted economic feasibility analyses, collected, analyzed, and modeled extensive regional data, evaluated multiple storage sites and infrastructure, and evaluated business plans. Six projects were funded for Phase II which involves storage complex feasibility studies. These projects evaluate initial reservoir characteristics to determine if the reservoir is suitable for geologic sequestration sites of more than 50 million tons of CO₂, address technical and non-technical challenges that may arise, develop a risk assessment and CO₂ management strategy for the project; and assist with the validation of existing tools. Five projects have been funded for CarbonSAFE Phase III and are currently performing site characterization and permitting.

The EPA notes that, while only sequestration facilities with Federal funding are currently operational in the United States, multiple commercial sequestration facilities, other than those funded under EPAct05, are in construction or advanced development, with some scheduled to open for operation as early as 2025. These facilities have proposed sequestration capacities ranging from 0.03 to 6 million tons of CO₂ per year. The Great Plains Synfuel Plant currently captures 2 million metric tons of CO₂ per year, which is exported to Canada for use in EOR; a planned addition of sequestration in a saline formation for this facility is expected to increase the amount of CO₂ captured and sequestered (through both geologic sequestration and EOR) to 3.5 million metric tons of CO₂ per year. The EPA and states with approved UIC Class VI programs (including Wyoming, North Dakota, and Louisiana) are currently reviewing UIC Class VI geologic sequestration well permit applications for proposed sequestration sites in fourteen states. As of March 15, 2024, 44 projects with 130 injection wells are under review by the EPA.

Currently, there are planned geologic sequestration facilities across the United States in various phases of development, construction, and operation. The Wyoming Department of Environmental Quality issued three UIC Class VI permits in December 2023 to Frontier Carbon Solutions. The Frontier Carbon Solutions project will sequester 5 million metric tons of CO₂ /year. Additionally, UIC Class VI permit applications have been submitted to the Wyoming Department of Environmental Quality for a proposed Eastern Wyoming Sequestration Hub project that would sequester up to 3 million metric tons of CO₂ /year. The North Dakota Oil and Gas Division has issued UIC Class VI permits to 6 sequestration projects that collectively will sequester 18 million metric tons of CO₂ /year. Since 2014, the EPA has issued two UIC Class VI permits to Archer Daniels Midland (ADM) in Decatur, Illinois, which authorize the injection of up to 7 million metric tons of CO₂ . One of the AMD wells is in the injection phase while the other is in the post-injection phase. In January 2024, the EPA issued two UIC Class VI permits to Wabash Carbon Services LLC for a project that will sequester up to 1.67 million metric tons of CO₂ /year over an injection period of 12 years. In December 2023, the EPA released for public comment four UIC Class VI draft permits for the Carbon TerraVault projects, to be located in California. These projects propose to sequester CO₂ captured from multiple different sources in California including a hydrogen plant, direct air capture, and pre-combustion gas treatment. TerraVault plans to inject 1.46 million metric tons of CO₂ annually into the four proposed wells over a 26-year injection period with a total potential capacity of 191 million metric tons. One of the proposed wells is an existing UIC Class II well that would be converted to a UIC Class VI well for the TerraVault project.

Geologic sequestration has been proven to be successful and safe in projects internationally. In Norway, facilities conduct offshore sequestration under the Norwegian continental shelf. In addition, the Sleipner CO₂ Storage facility in the North Sea, which began operations in 1996, injects around 1 million metric tons of CO₂ per year from natural gas processing. The Snohvit CO₂ Storage facility in the Barents Sea, which began operations in 2008, injects around 0.7 million metric tons of CO₂ per year from natural gas processing. The SaskPower carbon capture and sequestration facility at Boundary Dam Power Station in Saskatchewan, Canada had, as of the end of 2023, captured 5.6 million metric tons of CO₂ since it began operating in 2014. Other international sequestration facilities in operation include Glacier Gas Plant MCCS (Canada), Quest (Canada), and Qatar LNG CCS (Qatar). The CarbFix project in Iceland injects CO₂ into a geologic formation in which the CO₂ reacts with basalt rock formations to form stone. The CarbFix project has injected approximately 100,000 metric tons of CO₂ into geologic formations since 2014.

EOR, the process of injecting CO₂ into oil and gas formations to extract additional oil and gas, has been successfully used for decades at numerous production fields throughout the United States to increase oil and gas recovery. The oil and gas industry in the United States has nearly 60 years of experience with EOR. This experience provides a strong foundation for demonstrating successful CO₂ injection and monitoring technologies, which are needed for safe and secure geologic sequestration that can be used for deployment of CCS across geographically diverse areas. The amount of CO₂ that can be injected for an EOR project and the duration of operations are of similar magnitude to the duration and volume of CO₂ that is expected to be captured from fossil fuel-fired EGUs. The Farnsworth Unit, the Camrick Unit, the Shute Creek Facility, and the Core Energy CO₂ -EOR facility are all examples of operations that store anthropogenic CO₂ as a part of EOR operations. Currently, 13 states have active EOR operations, and these states also have areas that are amenable to deep saline sequestration in either onshore or offshore locations.

(3) EPAct05-Assisted Geologic Sequestration Projects

Consistent with the EPA's legal interpretation that the Agency can rely on experience from EPAct05 funded facilities in conjunction with other information, this section provides examples of EPAct05-assisted geologic sequestration projects. While the EPA has determined that the sequestration component of CCS is adequately demonstrated based on the non-EPAct05 examples discussed above, adequate demonstration of geologic sequestration is further corroborated by planned and operational geologic sequestration projects assisted by grants, loan guarantees, and the IRC section 48A federal tax credit for “clean coal technology” authorized by the EPAct05.

At present, there are 13 operational and one post-injection phase commercial carbon sequestration facilities in the United States. Red Trail Energy CCS Project in North Dakota and Illinois Industrial Carbon Capture and Storage in Illinois are dedicated saline sequestration facilities, while the other facilities, including Petra Nova in Texas, are sequestration via EOR. Several other facilities are under development. The Red Trail Energy CCS facility in North Dakota began injecting CO₂ captured from ethanol production plants in 2022. This project is expected to inject 180,000 tons of CO₂ per year. The Illinois Industrial Carbon Capture and Storage Project began injecting CO₂ from ethanol production into the Mount Simon Sandstone in April 2017. According to the facility's report to the EPA's Greenhouse Gas Reporting Program (GHGRP), as of 2022, 2.9 million metric tons of CO₂ had been injected into the saline reservoir. CO₂ injection for one of the two permitted Class VI wells ceased in 2021 and this well is now in the post-operation data collection phase.

There are additional planned geologic sequestration projects under review by the EPA and across the United States. Project Tundra, a saline sequestration project planned at the lignite-fired Milton R. Young Station in North Dakota is projected to capture 4 million metric tons of CO₂ annually. In Wyoming, Class VI permit applications have been issued by the Wyoming Department of Environmental Quality for the proposed Eastern Wyoming Sequestration Hub project, a saline sequestration facility proposed to be located in Southwestern Wyoming. At full capacity, the facility would permanently store up to 5 million metric tons of CO₂ captured from industrial facilities annually in the Nugget saline sandstone reservoir. In Texas, three NGCCs plan to add carbon capture equipment. Deer Park NGCC plans to capture 5 million tons per year, Quail Run NGCC plans to capture 1.5 million tons of CO₂ per year, and Baytown NGCC plans to capture up to 2 million tons of CO₂ per year.

(4) Security of Geologic Sequestration and Related Regulatory Requirements

As discussed in section VII.C.1.a.i(D)(2) of this preamble, there have been numerous instances of geologic sequestration in the U.S. and overseas, and the U.S. has developed a detailed set of regulatory requirements to ensure the security of sequestered CO₂. This regulatory framework includes the UIC well regulations pursuant to SDWA authority, and the GHGRP pursuant to CAA authority.

Regulatory oversight of geologic sequestration is built upon an understanding of the proven mechanisms by which CO₂ is retained in geologic formations. These mechanisms include (1) Structural and stratigraphic trapping (generally trapping below a low permeability confining layer); (2) residual CO₂ trapping (retention as an immobile phase trapped in the pore spaces of the geologic formation); (3) solubility trapping (dissolution in the in situ formation fluids); (4) mineral trapping (reaction with the minerals in the geologic formation and confining layer to produce carbonate minerals); and (5) preferential adsorption trapping (adsorption onto organic matter in coal and shale).

(a) Overview of Legal and Regulatory Framework

For the reasons detailed below, the UIC Program, the GHGRP, and other regulatory requirements comprise a detailed regulatory framework for geologic sequestration in the United States. This framework is analyzed in a 2021 report from the Council on Environmental Quality (CEQ), and statutory and regulatory frameworks that may be applicable for CCS are summarized in the EPA CCS Regulations Table. This regulatory framework includes the UIC regulations, promulgated by the EPA under the authority of the Safe Drinking Water Act (SDWA); and the GHGRP, promulgated by the EPA under the authority of the CAA. The requirements of the UIC and GHGRP programs work together to ensure that sequestered CO₂ will remain securely stored underground. Furthermore, geologic sequestration efforts on Federal lands as well as those efforts that are directly supported with Federal funds would need to comply with the NEPA and other Federal laws and regulations, depending on the nature of the project. In cases where sequestration is conducted offshore, the SDWA, the Marine Protection, Research, and Sanctuaries Act (MPRSA) or the Outer Continental Shelf Lands Act (OCSLA) may apply. The Department of Interior Bureau of Safety and Environmental Enforcement and Bureau of Ocean Energy Management are developing new regulations and creating a program for oversight of carbon sequestration activities on the outer continental shelf. Furthermore, Title V of the Federal Land Policy and Management Act of 1976 (FLPMA) and its implementing regulations, 43 CFR part 2800, authorize the Bureau of Land Management (BLM) to issue rights-of-way (ROWs) to geologically sequester CO₂ in Federal pore space, including BLM ROWs for the necessary physical infrastructure and for the use and occupancy of the pore space itself. The BLM has published a policy defining access to pore space on BLM lands, including clarification of Federal policy for situations where the surface and pore space are under the control of different Federal agencies.

(b) Underground Injection Control (UIC) Program

The UIC regulations, including the Class VI program, authorize the injection of CO₂ for geologic sequestration while protecting human health by ensuring the protection of underground sources of drinking water (USDW). These regulations are built upon nearly a half-century of Federal experience regulating underground injection wells, and many additional years of state UIC program expertise. The IIJA established a $50 million grant program to assist states and tribal regulatory authorities in developing and implementing UIC Class VI programs. Major components included in UIC Class VI permits are site characterization, area of review, corrective action, well construction and operation, testing and monitoring, financial responsibility, post-injection site care, well plugging, emergency and remedial response, and site closure. The EPA's UIC regulations are included in 40 CFR parts 144-147. The UIC regulations ensure that injected CO₂ does not migrate out of the authorized injection zone, which in turn ensures that CO₂ is securely stored underground.

Review of a UIC permit application by the permitting authority, including for Class VI geologic sequestration, entails a multidisciplinary evaluation to determine whether the application includes the required information, is technically accurate, and supports a determination that USDWs will not be endangered by the proposed injection activity. The EPA promulgated UIC regulations to ensure underground injection wells are constructed, operated, and closed in a manner that is protective of USDWs and to address potential risks to USDWs associated with injection activities. The UIC regulations address the major pathways by which injected fluids can migrate into USDWs, including along the injection well bore, via improperly completed or plugged wells in the area near the injection well, direct injection into a USDW, faults or fractures in the confining strata, or lateral displacement into hydraulically connected USDWs. States may apply to the EPA to be the UIC permitting authority in the state and receive primary enforcement authority (primacy). Where a state has not obtained primacy, the EPA is the UIC permitting authority.

Recognizing that CO₂ injection, for the purpose of geologic sequestration, poses unique risks relative to other injection activities, the EPA promulgated Federal Requirements Under the UIC Program for Carbon Dioxide GS Wells, known as the Class VI Rule, in December 2010. The Class VI Rule created and set requirements for a new class of injection wells, Class VI. The Class VI Rule builds upon the long-standing protective framework of the UIC Program, with requirements that are tailored to address issues unique to large-scale geologic sequestration, including large injection volumes, higher reservoir pressures relative to other injection formations, the relative buoyancy of CO₂, the potential presence of impurities in captured CO₂, the corrosivity of CO₂ in the presence of water, and the mobility of CO₂ within subsurface geologic formations. These additional protective requirements include more extensive geologic testing, detailed computational modeling of the project area and periodic re-evaluations, detailed requirements for monitoring and tracking the CO₂ plume and pressure in the injection zone, unique financial responsibility requirements, and extended post-injection monitoring and site care.

UIC Class VI permits are designed to ensure that geologic sequestration does not cause the movement of injected CO₂ or formation fluids outside the authorized injection zone; if monitoring indicates leakage of injected CO₂ from the injection zone, the leakage may trigger a response per the permittee's Class VI Emergency and Remedial Response Plan including halting injection, and the permitting authority may prescribe additional permit requirements necessary to prevent such movement to ensure USDWs are protected or take appropriate enforcement action if the permit has been violated. Class II EOR permits are also designed to ensure the protection of USDWs with requirements appropriate for the risks of the enhanced recovery operation. In general, the EPA believes that the protection of USDWs by preventing leakage of injected CO₂ out of the injection zone will also ensure that CO₂ is sufficiently sequestered in the subsurface, and therefore will not leak from the subsurface to the atmosphere.

The UIC program works with injection well operators throughout the life of the well to confirm practices do not pose a risk to USDWs. The program conducts inspections to verify compliance with the UIC permit, including checking for leaks. Inspections are only one way that programs deter noncompliance. Programs also evaluate periodic monitoring reports submitted by operators and discuss potential issues with operators. If a well is found to be out of compliance with applicable requirements in its permit or UIC regulations, the program will identify specific actions that an operator must take to address the issues. The UIC program may assist the operator in returning the well to compliance or use administrative or judicial enforcement to return a well to compliance.

UIC program requirements address potential safety concerns with induced seismicity. More specifically, through the UIC Class VI program, the EPA has put in place mechanisms to identify, monitor, and reduce risks associated with induced seismicity in any areas within or surrounding a sequestration site through permit and program requirements such as site characterization and monitoring, and the requirement for applicants to demonstrate that induced seismic activity will not endanger USDWs. The National Academy of Sciences released a report in 2012 on induced seismicity from CCS and determined that with appropriate site selection, a monitoring program, a regulatory system, and the appropriate use of remediation methods, the induced seismicity risks of geologic sequestration could be mitigated. Furthermore, the Ground Water Protection Council and Interstate Oil and Gas Compact Commission have published a “Potential Induced Seismicity Guide.” This report found that the strategies for avoiding, mitigating, and responding to potential risks of induced seismicity should be determined based on site-specific characteristics ( i.e., local geology). These strategies could include supplemental seismic monitoring, altering operational parameters (such as rates and pressures) to reduce the ground motion hazard and risk, permit modification, partial plug back of the well, controlled restart (if feasible), suspending or revoking injection authorization, or stopping injection and shutting in a well. The EPA's UIC National Technical Workgroup released technical recommendations in 2015 to address induced seismicity concerns in Class II wells and elements of these recommendations have been utilized in developing Class VI emergency and remedial response plans for Class VI permits. For example, as identified by the EPA's UIC National Technical Workgroup, sufficient pressure buildup from disposal activities, the presence of Faults of Concern ( i.e., a fault optimally oriented for movement and located in a critically stressed region), and the existence of a pathway for allowing the increased pressure to communicate with the fault contribute to the risk of injection-induced seismicity. The UIC requirements, including site characterization ( e.g., ensuring the confining zone is free of faults of concern) and operating requirements ( e.g., ensuring injection pressure in the injection zone is below the fracture pressure), work together to address these components and reduce the risk of injection-induced seismicity, particularly any injection-induced seismicity that could be felt by people at the surface. Additionally, the EPA recommends that Class VI permits include an approach for monitoring for seismicity near the site, including seismicity that cannot be felt at the surface, and that injection activities be stopped or reduced in certain situations if seismic activity is detected to ensure that no seismic activity will endanger USDWs. This also reduces the likelihood of any future injection-induced seismic activity that will be felt at the surface.

Furthermore, during site characterization, if any of the geologic or seismic data obtained indicate a substantial likelihood of seismic activity, the EPA may require further analyses, potential planned operational changes, and additional monitoring. The EPA has the authority to require seismic monitoring as a condition of the UIC permit if appropriate, or to deny the permit if the injection-induced seismicity risk could endanger USDWs.

The EPA believes that meaningful engagement with local communities is an important step in the development of geologic sequestration projects and has programs and public participation requirements in place to support this process. The EPA is committed to advancing EJ for overburdened communities in all its programs, including the UIC Class VI program. The EPA is also committed to supporting states' and tribes' efforts to obtain UIC Class VI primacy and strongly encourages such states and tribes to incorporate environmental justice principles and equity into proposed UIC Class VI programs. The EPA is taking steps to address EJ in accordance with Presidential Executive Order 14096, Revitalizing Our Nation's Commitment to Environmental Justice for All (88 FR 25251, April 26, 2023). In 2023, the EPA released Environmental Justice Guidance for UIC Class VI Permitting and Primacy that builds on the 2011 UIC Quick Reference Guide: Additional Tools for UIC Program Directors Incorporating Environmental Justice Considerations into the Class VI Injection Well Permitting Process. The 2023 guidance serves as an operating framework for identifying, analyzing, and addressing EJ concerns in the context of implementing and overseeing UIC permitting and primacy programs, including primacy approvals. The EPA notes that while this guidance is focused on the UIC Class VI program, EPA Regions should apply them to the other five injection well classes wherever possible, including class II. The guidance includes recommended actions across five themes to address various aspects of EJ in UIC Class VI permitting including: (1) identify communities with potential EJ concerns, (2) enhance public involvement, (3) conduct appropriately scoped EJ assessments, (4) enhance transparency throughout the permitting process, and (5) minimize adverse effects to USDWs and the communities they may serve.

As a part of the UIC Class VI permit application process, applicants and the EPA Regions should complete an EJ review using the EPA's EJScreen Tool, an online mapping tool that integrates numerous demographic, socioeconomic, and environmental data sets that are overlain on an applicant's UIC Area of Review to identify whether any disadvantaged communities are encompassed. If the results indicate a potential EJ impact, applicants and the EPA Regions should consider potential measures to mitigate the impacts of the UIC Class VI project on identified vulnerable communities and enhance the public participation process to be inclusive of all potentially affected communities (e.g., conduct early targeted outreach to communities and identify and mitigate any communication obstacles such as language barriers or lack of technology resources).

ER technologies are used in oil and gas reservoirs to increase production. Injection wells used for ER are regulated through the UIC Class II program. Injection of CO₂ is one of several techniques used in ER. Sometimes ER uses CO₂ from anthropogenic sources such as natural gas processing, ammonia and fertilizer production, and coal gasification facilities. Through the ER process, much of the injected CO₂ is recovered from production wells and can be separated and reinjected into the subsurface formation, resulting in the storage of CO₂ underground. The EPA's Class II regulations were designed to regulate ER injection wells, among other injection wells associated with oil and natural gas production. See e.g., 40 CFR 144.6(b)(2). The EPA's Class II program is designed to prevent Class II injection activities from endangering USDWs. The Class II programs of states and tribes must be approved by the EPA and must meet the EPA regulatory requirements for Class II programs, 42 U.S.C. 300h-1, or otherwise represent an effective program to prevent endangerment of USDWs. 42 U.S.C 300h-4 .

In promulgating the Class VI regulations, the EPA recognized that if the business model for ER shifts to focus on maximizing CO₂ injection volumes and permanent storage, then the risk of endangerment to USDWs is likely to increase. As an ER project shifts away from oil and/or gas production, injection zone pressure and carbon dioxide volumes will likely increase if carbon dioxide injection rates increase, and the dissipation of reservoir pressure will decrease if fluid production from the reservoir decreases. Therefore, the EPA's regulations require the operator of a Class II well to obtain a Class VI permit when there is an increased risk to USDWs. 40 CFR 144.19. While the EPA's regulations require the Class II well operator to assess whether there is an increased risk to USDWs (considering factors identified in the EPA's regulations), the permitting authority can also make this assessment and, in the event that an operator makes changes to Class II operations such that the increased risk to USDWs warrants transition to Class VI and the operator does not notify the permitting authority, the operator may be subject to SDWA enforcement and compliance actions to protect USDWs, including cessation of injection. The determination of whether there is an increased risk to USDWs would be based on factors specified in 40 CFR 144.19(b), including increase in reservoir pressure within the injection zone; increase in CO₂ injection rates; and suitability of the Class II Area of Review (AoR) delineation.

(c) Greenhouse Gas Reporting Program (GHGRP)

The GHGRP requires reporting of greenhouse gas (GHG) data and other relevant information from large GHG emission sources, fuel and industrial gas suppliers, and CO₂ injection sites in the United States. Approximately 8,000 facilities are required to report their emissions, injection, and/or supply activity annually, and the non-confidential reported data are made available to the public around October of each year. To complement the UIC regulations, the EPA included in the GHGRP air-side monitoring and reporting requirements for CO₂ capture, underground injection, and geologic sequestration. These requirements are included in 40 CFR part 98, subpart RR and subpart VV, also referred to as “GHGRP subpart RR” and “GHGRP subpart VV.”

GHGRP subpart RR applies to “any well or group of wells that inject a CO₂ stream for long-term containment in subsurface geologic formations” and provides the monitoring and reporting mechanisms to quantify CO₂ storage and to identify, quantify, and address potential leakage. The EPA designed GHGRP subpart RR to complement the UIC monitoring and testing requirements. See e.g., 40 CFR 146.90-91. Reporting under GHGRP subpart RR is required for, but not limited to, all facilities that have received a UIC Class VI permit for injection of CO₂ . Under existing GHGRP regulations, facilities that conduct ER in Class II wells are not subject to reporting data under GHGRP subpart RR unless they have chosen to submit a proposed monitoring, reporting, and verification (MRV) plan to the EPA and received an approved plan from the EPA. Facilities conducting ER and who do not choose to submit a subpart RR MRV plan to the EPA would otherwise be required to report CO₂ data under subpart UU. GHGRP subpart RR requires facilities meeting the source category definition (40 CFR 98.440) for any well or group of wells to report basic information on the mass of CO₂ received for injection; develop and implement an EPA-approved monitoring, reporting, and verification (MRV) plan; report the mass of CO₂ sequestered using a mass balance approach; and report annual monitoring activities. Extensive subsurface monitoring is required for UIC Class VI wells at 40 CFR 146.90 and is the primary means of determining if the injected CO₂ remains in the authorized injection zone and otherwise does not endanger any USDW, and monitoring under a GHGRP subpart RR MRV Plan complements these requirements. The MRV plan includes five major components: a delineation of monitoring areas based on the CO₂ plume location; an identification and evaluation of the potential surface leakage pathways and an assessment of the likelihood, magnitude, and timing, of surface leakage of CO₂ through these pathways; a strategy for detecting and quantifying any surface leakage of CO₂ in the event leakage occurs; an approach for establishing the expected baselines for monitoring CO₂ surface leakage; and, a summary of considerations made to calculate site-specific variables for the mass balance equation.

In April 2024, the EPA finalized a new GHGRP subpart, “Geologic Sequestration of Carbon Dioxide with Enhanced Oil Recovery (EOR) Using ISO 27916” (or GHGRP subpart VV). GHGRP subpart VV applies to facilities that quantify the geologic sequestration of CO₂ in association with EOR operations in conformance with the ISO standard designated as CSA/ANSI ISO 27916:2019, Carbon Dioxide Capture, Transportation and Geological Storage—Carbon Dioxide Storage Using Enhanced Oil Recovery. Facilities that have chosen to submit an MRV plan and report under GHGRP subpart RR must not report data under GHGRP subpart VV. GHGRP subpart VV is largely modeled after the requirements in this ISO standard and focuses on quantifying storage of CO₂ . Facilities subject to GHGRP subpart VV must include in their GHGRP annual report a copy of their EOR Operations Management Plan (EOR OMP). The EOR OMP includes a description of the EOR complex and engineered system, establishes that the EOR complex is adequate to provide safe, long-term containment of CO₂ , and includes site-specific and other information including a geologic characterization of the EOR complex, a description of the facilities within the EOR project, a description of all wells and other engineered features in the EOR project, and the operations history of the project reservoir.

Based on the understanding developed from existing projects, the security of sequestered CO₂ is expected to increase over time after injection ceases. This is due to trapping mechanisms that reduce CO₂ mobility over time ( e.g., physical CO₂ trapping by a low-permeability geologic seal or chemical trapping by conversion or adsorption). The EPA acknowledges the potential for some leakage of CO₂ to the atmosphere at sequestration sites, primarily while injection operations are active. For example, small quantities of the CO₂ that were sent to the sequestration site may be emitted from leaks in pipes and valves that are traversed before the CO₂ actually reaches the sequestration formation. However, the EPA's robust UIC regulatory protections protect against leakage out of the injection zone. Relative to the 46.75 million metric tons of CO₂ reported as sequestered under subpart RR of the GHGRP between 2016 to 2022, only 196,060 metric tons were reported as leakage/emissions to the atmosphere in the same time period (representing less than 0.5% of the sequestration amount). Of these emissions, most were from equipment leaks and vented emissions of CO₂ from equipment located on the surface rather than leakage from the subsurface. Furthermore, any leakage of CO₂ at a sequestration facility would be required to be quantified and reported under the GHGRP subpart RR or subpart VV, and such data are made publicly available on the EPA's website.

(5) Timing of Permitting for Sequestration Sites

As previously discussed, the EPA is the Class VI permitting authority for states, tribes, and territories that have not obtained primacy over their Class VI programs. The EPA is committed to reviewing UIC Class VI permits as expeditiously as possible when the agency is the permitting authority. The EPA has the experience to properly regulate and review permits for UIC Class VI injection wells, and technical experts of multiple disciplines to review permit applications submitted to the EPA.

The EPA has seen a considerable uptick in Class VI permit applications over the past few years. The 2018 passage of revisions and enhancements to the IRC section 45Q tax credit that provides tax credits for carbon oxide (including CO₂) sequestration has led to an increase in Class VI permit applications submitted to the EPA. The 2022 IRA further expanded the IRC section 45Q tax credit and the 2021 IIJA established a $50 million program for grants to help states and tribes in developing and implementing a UIC Class VI primacy program, leading to even more interest in this area. Between 2011, when the Class VI rule went into effect, and 2020, the EPA received a total of 8 permit applications for Class VI wells. The EPA then received 12 Class VI permit applications in 2021, 44 in 2022, and 123 in 2023. As of March 2024, the EPA has 130 Class VI permit applications under review (56 permit applications were transferred to Louisiana in February 2024 when the EPA rule granting Class VI primacy to the state became effective). The majority of those 130 permit applications (63%) were submitted to the EPA within the past 12 months. Also, as of March 2024, the EPA has issued eight Class VI permits, including six for projects in Illinois and two for projects in Indiana, and has released for public comment four additional draft permits for proposed projects in California. Two of the permits are in the pre-operation phase, one is in the injection phase, and one is in the post-injection monitoring phase.

In light of the recent flurry of interest in this area, the EPA is devoting increased resources to the Class VI program, including through increased staffing levels in order to meet the increased demand for action on Class VI permit applications. Reviewing a Class VI permit application entails a multidisciplinary evaluation to determine whether the application includes the required information, is technically accurate, and supports a risk-based determination that underground sources of drinking water will not be endangered by the proposed injection activity. A wide variety of technical experts—from geologists to engineers to physical scientists—review permit applications submitted to the EPA. The EPA has been working to develop staff expertise and increase capacity in the UIC program, and the agency has effectively deployed appropriated resources over the last five years to scale UIC program staff from a few employees to the equivalent of more than 25 full-time employees across the agency's headquarters and regional offices. We expect that the additional resources and staff capacity for the Class VI program will lead to increased efficiencies in the Class VI permitting process.

In addition to increased staffing resources, the EPA has made considerable improvements to the Class VI permitting process to reduce the time needed to make final permitting decisions for Class VI wells while maintaining a robust and thorough review process that ensures USDWs are protected. The EPA has created additional resources for applicants including upgrading the Geologic Sequestration Data Tool (GSDT) to guide applicants through the application process. The EPA has also created resources for permit writers including training series and guidance documents to build capacity for Class VI permitting. Additionally, the EPA issued internal guidelines to streamline and create uniformity and consistency in the Class VI permitting process, which should help to reduce permitting timeframes. These internal guidelines include the expectation that EPA Regions will classify all Class VI well applications received on or after December 12, 2023, as applications for major new UIC injection wells, which requires the Regions to develop project decision schedules for reviewing Class VI permit applications. The guidelines also set target timeframes for components of the permitting process, such as the number of days EPA Regions should set for public comment periods and for developing responses to comments and final permit decisions. The EPA will continue to evaluate its internal UIC permitting processes to identify potential opportunities for streamlining and other improvements over time. Although the available data for Class VI wells is limited, the timeframe for processing Class I wells, which follows a similar regulatory structure, is typically less than 2 years.

The EPA notes that a Class VI permit tracker is available on its website. This tracker shows information for the 44 projects (representing 130 wells) that have submitted Class VI applications to the EPA, including details such as the current permit review stage, whether a project has been sent a Notice of Deficiency (NOD) or Request for Additional Information (RAI), and the applicant's response time to any NODs or RAIs. As mentioned above, most of the permits submitted to the EPA have been submitted within the past 12 months. The EPA aims to review complete Class VI applications and issue permits when appropriate within approximately 24 months. This timeframe is dependent on several factors, including the complexity of the project and the quality and completeness of the submitted application. It is important for the applicant to submit a complete application and provide any information requested by the permitting agency in a timely manner so as not to extend the overall time for the review.

States may apply to the EPA for primacy to administer the Class VI programs within their states. The primacy application process has four phases: (1) pre-application activities, (2) completeness review and determination, (3) application evaluation, and (4) rulemaking and codification. To date, three states have been granted primacy for Class VI wells, including North Dakota, Wyoming, and most recently Louisiana. As discussed above, North Dakota has issued 6 Class VI permits since receiving Class VI primacy in 2018, and Wyoming issued its first three Class VI permits in December 2023. The EPA finalized a rule granting Louisiana Class VI primacy in January 2024 and the state's program became effective in February 2024. At that time, EPA Region 6 transferred 56 Class VI permit applications for projects in Louisiana to the state for continued review and permit issuance if appropriate. Prior to receiving primacy, the state worked with the EPA in understanding where each application was in the evaluation process. Currently, the EPA is working with the states of Texas, Arizona, and West Virginia as they are developing their UIC primacy applications. Arizona submitted a primacy application to the EPA on February 13, 2024. Texas and West Virginia are engaging with the EPA to complete pre-application activities. If more states apply for and receive Class VI primacy, the number of permits in EPA review is expected to be reduced. The EPA has also created resources for regulators including training series and guidance documents to build capacity for Class VI permitting within UIC programs across the U.S. Through state primacy for Class VI programs, state expertise and capacity can be leveraged to support effective and efficient permit application reviews. The IIJA established a $50 million grant program to support states, Tribes, and territories in developing and implementing UIC Class VI programs. The EPA has allocated $1,930,000 to each state, tribe, and territory that submitted letters of intent.

(6) Comments Received on Geologic Sequestration and Responses

The EPA received comments on geologic sequestration. Those comments, and the EPA's responses, are as follows.

Comment: Some commenters expressed concerns that the EPA has not demonstrated the adequacy of carbon sequestration at a commercial scale.

Response: The EPA disagrees that commercial carbon sequestration capacity will be inadequate to support this rule. As detailed in section VII.C.1.a.i(D)(1), commercial geologic sequestration capacity is growing in the United States. Multiple commercial sequestration facilities, other than those funded under EPAct05, are in construction or advanced development, with some scheduled to open for operation as early as 2025. These facilities have proposed sequestration capacities ranging from 0.03 to 6 million tons of CO₂ per year. The EPA and states with approved UIC Class VI programs (including Wyoming, North Dakota, and Louisiana) are currently reviewing UIC Class VI geologic sequestration well permit applications for proposed sequestration sites in fourteen states. As of March 2024, there are 44 projects with 130 injection wells are under review by the EPA. Furthermore, the EPA anticipates that as the demand for commercial sequestration grows, more commercial sites will be developed in response to financial incentives.

Comment: Some commenters expressed concern about leakage of CO₂ from sequestration sites.

Response: The EPA acknowledges the potential for some leakage of CO₂ to the atmosphere at sequestration sites (such as leaks through valves before the CO₂ reaches the injection formation). However, as detailed in the preceding sections of preamble, the EPA's robust UIC permitting process is adequate to protect against CO₂ escaping the authorized injection zone (and then entering the atmosphere). As discussed in the preceding section, leakage out of the injection zone could trigger emergency and remedial response action including ceasing injection, possible permit modification, and possible enforcement action. Furthermore, the GHGRP subpart RR and subpart VV regulations prescribe accounting methodologies for facilities to quantify and report any potential leakage at the surface, and the EPA makes sequestration data and related monitoring plans publicly available on its website. The reported emissions/leakage from sequestration sites under subpart RR is a comparatively small fraction (less than 0.5 percent) of the associated sequestration volumes, with most of these reported emissions attributable to leaks or vents from surface equipment.

Comment: Some commenters expressed concern over safety due to induced seismicity.

Response: The EPA believes that the UIC program requirements adequately address potential safety concerns with induced seismicity at site-adjacent communities. More specifically, through the UIC Class VI program the EPA has put in place mechanisms to identify, monitor, and mitigate risks associated with induced seismicity in any areas within or surrounding a sequestration site through permit and program requirements, such as site characterization and monitoring, and the requirement for applicants to demonstrate that induced seismic activity will not endanger USDWs. See section VII.C.1.a.i(D)(4)(b) for further discussion of mitigating induced seismicity risk. Although the UIC Class II program does not have specific requirements regarding seismicity, it includes discretionary authority to add additional conditions to a UIC permit on a case-by-case basis. The EPA created a document outlining practical approaches for UIC Directors to use to minimize and manage injection-induced seismicity in Class II wells. Furthermore, during site characterization, if any of the geologic or seismic data obtained indicate a substantial likelihood of seismic activity, further analyses, potential planned operational changes, and additional monitoring may be required. The EPA has the authority to require seismic monitoring as a condition of the UIC permit if appropriate, or to deny the permit if the injection-induced seismicity risk could endanger USDWs.

Comment: Some commenters have expressed concern that the EPA has not meaningfully engaged with historically disadvantaged and overburdened communities who may be impacted by environmental changes due to geologic sequestration.

Response: The EPA acknowledges that meaningful engagement with local communities is an important step in the development of geologic sequestration projects and has programs and public participation requirements in place to support this process. The EPA is committed to advancing environmental justice for overburdened communities in all its programs, including the UIC Class VI program. The EPA's environmental justice guidance for Class VI permitting and primacy states that many of the expectations are broadly applicable, and EPA Regions should apply them to the other five injection well classes, including Class II, wherever possible. See section VII.C.1.a.i(D)(4) for a detailed discussion of environmental justice requirements and guidance.

Comment: Commenters expressed concern that companies are not always in compliance with reporting requirements for subpart RR when required for other Federal programs.

Response: The EPA recognizes the need for geologic sequestration facilities to comply with the reporting requirements of the GHGRP, and acknowledges that there have been instances of entities claiming geologic sequestration under non-EPA programs ( e.g., to qualify for IRC section 45Q tax credits) while not having an EPA-approved MRV plan or reporting data under subpart RR. The EPA does not implement the IRC section 45Q tax credit program, and it is not privy to taxpayer information. Thus, the EPA has no role in implementing or enforcing these tax credit claims, and it is unclear, for example, whether these companies would have been required by GHGRP regulations to report data under subpart RR, or if they would have been required only by the IRC section 45Q rules to opt-in to reporting under subpart RR. The EPA disagrees that compliance with the GHGRP would be a problem for this rule because the rule requires any affected unit that employs CCS technology that captures enough CO₂ to meet the proposed standard and injects the captured CO₂ underground to report under GHGRP subpart RR or GHGRP subpart VV. Unlike the IRC section 45Q tax credit program, which is implemented by the Internal Revenue Service (IRS), the EPA will have the information necessary to discern whether a facility is in compliance with any applicable GHGRP requirements. If the emitting EGU sends the captured CO₂ offsite, it must transfer the CO₂ to a facility that reports in accordance with GHGRP subpart RR or GHGRP subpart VV. For more information on the relationship to GHGRP requirements, see section X.C.5 of this preamble.

Comment: Commenters expressed concerns that UIC regulations allow Class II wells to be used for long-term CO₂ storage if the operator assesses that a Class VI permit is not required and asserted that Class II regulations are less protective than Class VI regulations.

Response: The EPA acknowledges that Class II wells for EOR may be used to inject CO₂ including CO₂ captured from an EGU. However, the EPA disagrees that the use of Class II wells for ER will be less protective of human health than the use of Class VI wells for geologic sequestration. Class II wells are used only to inject fluids associated with oil and natural gas production, and Class II ER wells are used specifically for the injection of fluids, including CO_2, for the purpose of enhanced recovery of oil or natural gas. The EPA's UIC Class II program is designed to prevent Class II injection activities from endangering USDWs. Any leakage out of the designated injection zone could pose a risk to USDWs and therefore could be subject to enforcement action or permit modification. Therefore, the EPA believes that UIC protections for USDWs would also ensure that the injected CO₂ is contained in the subsurface formations. The Class II programs of states and tribes must be approved by the EPA and must meet EPA regulatory requirements for Class II programs, 42 U.S.C. 300h-1, or otherwise represent an effective program to prevent endangerment of USDWs. 42 U.S.C 300h-4 . The EPA's regulations require the operator of a Class II well to obtain a Class VI permit when operations shift to geologic sequestration and there is consequently an increased risk to USDWs. 40 CFR 144.19. UIC Class VI regulations require that owners or operators must show that the injection zone has sufficient volume to contain the injected carbon dioxide stream and report any fluid migration out of the injection zone and into or between USDWs. 40 CFR 146.83 and 40 CFR 146.91. The EPA emphasizes that while CO₂ captured from an EGU can be injected into a Class II ER injection well, it cannot be injected into the other two types of Class II wells, which are Class II disposal wells and Class II wells for the storage of hydrocarbons. 40 CFR 144.6(b).

Comment: Some commenters expressed concern that because few Class VI permits have been issued, the EPA's current level of experience in properly regulating and reviewing permits for these wells is limited.

Response: The EPA disagrees that the Agency lacks experience to properly regulate, and review permits for Class VI injection wells. We expect that the additional resources that have been allocated for the Class VI program will lead to increased efficiencies in the Class VI permitting process and timeframes. For a more detailed discussion of Class VI permitting and timeframes, see sections VII.C.1.a.i(D)(4)(b) and VII.C.1.a.i(D)(5) of this preamble. The EPA emphasizes that incomplete or insufficient application materials can result in substantially delayed permitting decisions. When the EPA receives incomplete or insufficient permit applications, the EPA communicates the deficiencies, waits to receive additional materials from the applicant, and then reviews any new data. This back and forth can result in longer permitting timeframes. The EPA therefore encourages applicants to contact their permitting authority early on so applicants can gain a thorough understanding of the Class VI permitting process and the permitting authority's expectations. To assist potential permit applicants, the EPA maintains a list of UIC contacts within each EPA Regional Office on the Agency's website. The EPA has met with more than 100 companies and other interested parties.

Comment: Some commenters claimed that various legal uncertainties preclude a finding that geologic sequestration of CO₂ has been adequately demonstrated. This concern has been raised in particular with issues of pore space ownership and the lack of long-term liability insurance and noted uncertainties regarding long-term liability generally.

Response: The EPA disagrees that these uncertainties are sufficient to prohibit the development of geologic sequestration projects. An interagency CCS task force examined sequestration-related legal issues thoroughly and concluded that early CCS projects could proceed under the existing legal framework with respect to issues such as property rights and liability. The development of CCS projects may be more complex in certain regions, due to distinct pore space ownership regulatory regimes at the state level, except on Federal lands.

As discussed in section VII.C.1.a.i.(D)(4) of this preamble, Title V of the FLPMA and its implementing regulations, 43 CFR part 2800, authorize the BLM to issue ROWs to geologically sequester CO₂ in Federal pore space, including BLM ROWs for the necessary physical infrastructure and for the use and occupancy of the pore space itself. The BLM has published a policy defining access to pore space on BLM lands, including clarification of Federal policy for situations where the surface and pore space are under the control of different Federal agencies.

States have established legislation and regulations defining pore space ownership and providing clarification to prospective users of surface pore space. For example, in North Dakota, the surface owner also owns the pore space underlying their surface estate. North Dakota state courts have determined that in situations where the surface ownership and mineral ownership have been legally severed the mineral estate is the dominant estate and has the right to use as much of the surface estate as reasonably necessary. The North Dakota legislature codified this interpretation in 2019. Summit Carbon Solutions, which is developing a carbon storage hub in North Dakota to store an estimated one billion tons of CO₂ , indicated that they had secured the majority of the pore space needed through long term leases with landowners. Wyoming defines ownership of pore space underlying surfaces within the state. Other states have also established laws, implementing regulations and guidance defining ownership and access to pore space. The EPA notes that many states are actively enacting legislation addressing pore space ownership. See e.g., Wyoming H.B. No. 89 (2008) (Wyo. Stat. § 34-1-152); Montana S.B. No. 498 (2009) (Mont. Code Ann. 82-11-180); North Dakota S.B. No. 2139 (2009) (N.D. Cent. Code § 47-31-03); Kentucky H.B. 259 (2011) (Ky. Rev. Stat. Ann. § 353.800); West Virginia H.B. 4491 (2022) (W. Va. Code § 22-11B-18); California S.B. No. 905 (2022) (Cal. Pub. Res. Code § 71462); Indiana Public Law 163 (2022) (Ind. Code § 14-39-2-3); Utah H.B. 244 (2022) (Utah Code § 40-6-20.5).

Liability during operation is usually assumed by the project operator, so liability concerns primarily arise after the period of operations. Research has previously shown that the environmental risk is greatest before injection stops. In terms of long-term liability and permittee obligations under the SDWA, the EPA's Class VI regulations impose various requirements on permittees even after injection ceases, including regarding injection well plugging (40 CFR 146.92), post-injection site care (PISC), and site closure (40 CFR 146.93). The default time period for post-injection site care is 50 years, during which the permittee must monitor the position of the CO₂ plume and pressure front and demonstrate that USDWs are not being endangered. 40 CFR 146.93. The permittee must also generally maintain financial responsibility sufficient to cover injection well plugging, corrective action, emergency and remedial response, PISC, and site closure until the permitting authority approves site closure. 40 CFR 146.85(a)&(b). Even after the former permittee has fulfilled all its UIC regulatory obligations, it may still be held liable for previous regulatory noncompliance, such as where the permittee provided erroneous data to support approval of site closure. A former permittee may always be subject to an order that the EPA Administrator deems necessary to protect public health if there is fluid migration that causes or threatens imminent and substantial endangerment to a USDW. 42 U.S.C. 300i; 40 CFR 144.12(e).

The EPA notes that many states are enacting legislation addressing long term liability. See e.g., Montana S.B. No. 498 (2009) (Mont. Code Ann. 82-11-183); Texas H.B. 1796 (2009) (Tex. Health & Safety Code Ann. § 382.508); North Dakota S.B. No. 2095 (2009) (N.D. Cent. Code § 38-22-17); Kansas H.B. 2418 (2010) (Kan. Stat. Ann. § 55-1637(h)); Wyoming S.F. No. 47 (2022) (Wyo. Stat. §§ 35-11-319); Louisiana H.B. 661 (2009) & H.B. 571 (2023) (La. Stat. Ann. § 30:1109). Because states are actively working to address pore space and liability uncertainties, the EPA does not believe these to be issues that would delay project implementation beyond the timelines discussed in this preamble.

(E) Compliance Date for Long-Term Coal-Fired Steam Generating Units

The EPA proposed a January 1, 2030 compliance date for long-term coal fired steam generating units subject to a CCS BSER. That compliance date assumed installation of CCS was concurrent with development of state plans. While several commenters were supportive of the proposed compliance date, the EPA also received comments on the proposed rule that stated that the proposed compliance date was not achievable. Commenters referenced longer project timelines for CO₂ capture. Commenters also requested that the EPA should account for the state plan process in determining the appropriate compliance date.

The EPA has considered the comments and information available and is finalizing a compliance date of January 1, 2032, for long-term coal-fired steam generating units. The EPA is also finalizing a mechanism for a 1-year compliance date extension in cases where a source faces delays outside its control, as detailed in section X.C.1.d of this preamble. The justification for the January 1, 2032 compliance date does not require substantial work to be done during the state planning process. Rather, the justification for the compliance date reflects the assumption that only the initial feasibility work which is necessary to inform the state planning process would occur during state plan development, with the start of more substantial work beginning after the due date for state plan submission, and a longer timeline for installation of CCS than at proposal. In total, this allows for 6 years and 7 months for both initial feasibility and more substantial work to occur after issuance of this rule. This is consistent with the approximately 6 years from start to finish for Boundary Dam Unit 3 and Petra Nova.

The timing for installation of CCS on existing coal-fired steam generating units is based on the baseline project schedule for the CO₂ capture plant developed by Sargent and Lundy (S&L and a review of the available information for installation of CO₂ pipelines and sequestration sites. Additional details on the timeline are in the TSD GHG Mitigation Measures for Steam Generating Units, available in the docket. The dates for intermediate steps are for reference. The specific sequencing of steps may differ slightly, and, for some sources, the duration of one step may be shorter while another may be longer, however the total duration is expected to be the same. The resulting timeline is therefore an accurate representation of the time necessary to install CCS in general.

The EPA assumes that feasibility work, amounting to less than 1 year (June 2024 through June 2025) for each component of CCS (capture, transport, and storage) occurs during the state plan development period (June 2024 through June 2026). This feasibility work is limited to initial conceptual design and other preliminary tasks, and the costs of the feasibility work in general are substantially less than other components of the project schedule. The EPA determined that it was appropriate to assume that this work would take place during the state plan development period because it is necessary for evaluating the controls that the state may determine to be appropriate for a source and is necessary for determining the resulting standard of performance that the state may apply to the source on the basis of those controls. In other words, without such feasibility and design work, it would be very difficult for a state to determine whether CCS is appropriate for a given source or the resulting standard of performance. While the EPA accounts for up to 1 year for feasibility for the capture plant, the S&L baseline schedule estimates this initial design activity can be completed in 6 months. For the capture plant, feasibility includes a preliminary technical evaluation to review the available utilities and siting footprint for the capture plant, as well as screening of the available capture technologies and vendors for the project, with an associated initial economic estimate. For sequestration, in many cases, general geologic characterization of regional areas has already been conducted by U.S. DOE and regional initiatives; however, the EPA assumes an up to 1 year period for a storage complex feasibility study. For the pipeline, the feasibility includes the initial pipeline routing analysis, taking less than 1 year. This exercise involves using software to review existing right-of-way and other considerations to develop an optimized pipeline route. Inputs to that analysis have been made publicly available by DOE in NETL's Pipeline Route Planning Database.

When state plans are submitted 24 months after publication of the final rule, requirements included within those state plans should be effective at the state level. On that basis, the EPA assumes that sources installing CCS are fully committed, and more substantial work ( e.g., FEED study for the capture plant, permitting, land use and right-of-way acquisition) resumes in June 2026. The EPA notes, however, that it would be possible that a source installing CCS would choose to continue these activities as soon as the initial feasibility work is completed even if not yet required to do so, rather than wait for state plan submission to occur for the reasons explained in full below.

Of the components of CCS, the CO₂ capture plant is the more technically involved and time consuming, and therefore is the primary driver for determining the compliance date. The EPA assumes substantial work commences only after submission due date for state plans. The S&L baseline timeline accounts for 5.78 years (301 weeks) for final design, permitting, and installation of the CO₂ capture plant. First, the EPA describes the timeline that is consistent with the S&L baseline for substantial work. Subsequently, the EPA describes the rationale for slight adjustments that can be made to that timeline based upon an examination of actual project timelines.

In the S&L baseline, substantial work on the CO₂ capture plant begins with a 1-year FEED study (June 2026 to June 2027). The information developed in the FEED study is necessary for finalizing commercial arrangements. In the S&L baseline, the commercial arrangements can take up to 9 months (June 2027 to March 2028). Commercial arrangements include finalizing funding as well as finalizing contracts with a CO₂ capture technology provider and engineering, procurement, and construction companies. The S&L baseline accounts for 1 year for permitting, beginning when commercial arrangements are nearly complete (December 2027 to December 2028). After commercial arrangements are complete, a 2-year period for engineering and procurement begins (March 2028 to March 2030). Detailed engineering starts after commercial arrangements are complete because engineers must consider details regarding the selected CO₂ capture technology, equipment providers, and coordination with construction. Shortly after permitting is complete, 6 months of sitework (March 2029 to September 2029) occur. Sitework is followed by 2 years of construction (July 2029 to July 2031). Approximately 8 months prior to the completion of construction, a roughly 14 month (60 weeks) period for startup and commissioning begins (January 2031 to March 2032).

In many cases, the EPA believes that sources are positioned to install CO₂ capture on a slightly faster timeline than the baseline S&L timeline detailed in the prior paragraph, because CCS projects have been developed in a shorter timeframe. Including these minor adjustments, the total time for detailed engineering, procurement, construction, startup and commissioning is 4 years, which is consistent with completed projects (Boundary Dam Unit 3 and Petra Nova) and project schedules developed in completed FEED studies, see the final TSD, GHG Mitigation Measures for Steam Generating Units for additional details. In addition, the IRC tax credits incentivize sources to begin complying earlier to reap economic benefits earlier. Sources that have already completed feasibility or FEED studies, or that have FEED studies ongoing are likely to be able to have CCS fully operational well in advance of January 1, 2032. Ongoing projects have planned dates for commercial operation that are much earlier. For example, Project Diamond Vault has plans to be fully operational in 2028. While the EPA assumes FEED studies start after the date for state plan submission, in practice sources are likely to install CO₂ capture as expeditiously as practicable. Moreover, the preceding timeline is derived from project schedules developed in the absence of any regulatory impetus. Considering these factors, sources have opportunities to slightly condense the duration, overlap, or sequencing of steps so that the total duration for completing substantial work on the capture plant is reduced by 2 months. For example, by expediting the duration for commercial arrangements from 9 months to 7 months, reasonably assuming sources immediately begin sitework as soon as permitting is complete, and accounting for 13 months (rather than 14) for startup and testing, the CO₂ capture plant will be fully operational by January 2032. Therefore, the EPA concludes that CO₂ capture can be fully operational by January 1, 2032. To the extent additional time is needed to take into account the particular circumstances of a particular source, the state may take those circumstances into account to provide a different compliance schedule, as detailed in section X.C.2 of this preamble.

The EPA also notes that there is additional time for permitting than described in the S&L baseline. The key permitting that affects the timeline are air permits because of the permits' impact on the ability to construct and operate the CCS capture equipment, in which the EPA is the expert in. The S&L baseline assumes permitting starts after the FEED study is complete while commercial arrangements are ongoing, however permitting can begin earlier allowing a more extended period for permitting. Examples of CCS permitting being completed while FEED studies are on-going include the air permits for Project Tundra, Baytown Energy Center, and Deer Park Energy Center. Therefore, while the FEED study is on-going, the EPA assumes that a 2-year process for permitting can begin.

The EPA's compliance deadline assumes that storage and pipelines for the captured CO₂ can be installed concurrently with deployment of the capture system. Substantial work on the storage site starts with 3 years (June 2026 to June 2029) for final site characterization, pore-space acquisition, and permitting, including at least 2 years for permitting of Class VI wells during that period. Lastly, construction for sequestration takes 1 year (June 2029 to June 2030). While the EPA assumes that storage can be permitted and constructed in 4 years, the EPA notes that there is at least an additional 12 months of time available to complete construction of the sequestration site without impacting progress of the other components.

The EPA assumes the substantial work on the pipeline lags the start of substantial work on the storage site by 6 months. After the 1 year of feasibility work prior to state plan submission, the general timeline for the CO₂ pipeline assumes up to 3 years for final routing, permitting activities, and right-of-way acquisition (December 2026 to December 2029). Lastly, there are 1.5 years for pipeline construction (December 2029 to June 2031).

The EPA does not assume that CCS projects are, in general, subject to NEPA. NEPA review is required for reasons including sources receiving federal funding ( e.g., through USDA or DOE) or projects on federal lands. NEPA may also be triggered for a CCS project if NEPA compliance is necessary for construction of the pipeline, such as where necessary because of a Clean Water Act section 404 permit, or for sequestration. Generally, if one aspect of a project is subject to NEPA, then the other project components could be as well. In cases where a project is subject to NEPA, an environmental assessment (EA) that takes 1 year, can be finalized concurrently during the permitting periods of each component of CCS (capture, pipeline, and sequestration). However, the EPA notes that the final timeline can also accommodate a concurrent 2-year period if an EIS were required under NEPA across all components of the project. The EPA also notes that, in some circumstances, NEPA review may begin prior to completion of a FEED study. For Petra Nova, a notice of intent to issue an EIS was published on November 14, 2011, and the record of decision was issued less than 2 years later, on May 23, 2013, while the FEED study was completed in 2014.

Based on this detailed analysis, the EPA has concluded that January 1, 2032, is an achievable compliance date for CCS on existing coal-fired steam generating units that takes into account the state plan development period, as well as the technical and bureaucratic steps necessary to install and implement CCS and is consistent with other expert estimates and real-world experience.

(F) Long-Term Coal-Fired Steam Generating Units Potentially Subject to This Rule

In this section of the preamble, the EPA estimates the size of the inventory of coal-fired power plants in the long-term subcategory likely subject to CCS as the BSER. Considering that capacity, the EPA also describes the distance to storage for those sources.

(1) Capacity of Units Potentially Subject to This Rule

First, the EPA estimates the total capacity of units that are currently operating and that have not announced plans to retire by 2039, or to cease firing coal by 2030. Starting from that first estimate, the EPA then estimates the capacity of units that would likely be subject to the CCS requirement, based on unit age, industry trends, and economic factors.

Currently, there are 181 GW of coal-fired steam generating units. About half of that capacity, totaling 87 GW, have announced plans to retire before 2039, and an additional 13 GW have announced plans to cease firing coal by that time. The remaining amount, 81 GW, are likely to be the most that could potentially be subject to requirements based on CCS.

However, the capacity of affected coal-fired steam generating units that would ultimately be subject to a CCS BSER is likely approximately 40 GW. This determination is supported by several lines of analysis of the historical data on the size of the fleet over the past several years. Historical trends in the coal-fired generation fleet are detailed in section IV.D.3 of this preamble. As coal-fired units age, they become less efficient and therefore the costs of their electricity go up, rendering them even more competitively disadvantaged. Further, older sources require additional investment to replace worn parts. Those circumstances are likely to continue through the 2030s and beyond and become more pronounced. These factors contribute to the historical changes in the size of the fleet.

One way to analyze historical changes in the size of the fleet is based on unit age. As the average age of the coal-fired fleet has increased, many sources have ceased operation. From 2000 to 2022, the average age of a unit that retired was 53 years. At present, the average age of the operating fleet is 45 years. Of the 81 GW that are presently operating and that have not announced plans to retire or convert to gas prior to 2039, 56 GW will be 53 years or older by 2039.

Another line of analysis is based on the rate of change of the size of the fleet. The final TSD, Power Sector Trends, available in the rulemaking docket, includes analysis showing sharp and steady decline in the total capacity of the coal-fired steam generating fleet. Over the last 15 years (2009-2023), average annual coal retirements have been 8 GW/year. Projecting that retirements will continue at approximately the same pace from now until 2039 is reasonable because the same circumstances will likely continue or accelerate further given the incentives under the IRA. Applying this level of annual retirement would result in 45 GW of coal capacity continuing to operate by 2039. Alternatively, the TSD also includes a graph that shows what the fleet would look like assuming that coal units without an announced retirement date retire at age 53 (the average retirement age of units over the 2000-2022 period). It shows that the amount of coal-fired capacity that remains in operation by 2039 is 38 GW.

The EPA also notes that it is often the case that coal-fired units announce that they plan to retire only a few years in advance of the retirement date. For instance, of the 15 GW of coal-fired EGUs that reported a 2022 retirement year in DOE's EIA Form 860, only 0.5 GW of that capacity had announced its retirements plans when reporting in to the same EIA-860 survey 5 years earlier, in 2017. Thus, although many coal-fired units have already announced plans to retire before 2039, it is likely that many others may anticipate retiring by that date but have not yet announced it.

Finally, the EPA observes that modeling the baseline circumstances, absent this final rule, shows additional retirements of coal-fired steam generating units. At the end of 2022, there were 189 GW of coal active in the U.S. By 2039, the IPM baseline projects that there will be 42 GW of operating coal-fired capacity (not including coal-to-gas conversions). Between 2023-2039, 95 GW of coal capacity have announced retirement and an additional 13 have announced they will cease firing coal. Thus, of the 81 GW that have not announced retirement or conversion to gas by 2039, the IPM baseline projects 39 GW will retire by 2039 due to economic reasons.

For all these reasons, the EPA considers that it is realistic to expect that 42 GW of coal-fired generating will be operating by 2039—based on announced retirements, historical trends, and model projections—and therefore constitutes the affected sources in the long-term subcategory that would be subject to requirements based on CCS. It should be noted that the EPA does not consider the above analysis to predict with precision which units will remain in operation by 2039. Rather, the two sets of sources should be considered to be reasonably representative of the inventory of sources that are likely to remain in operation by 2039, which is sufficient for purposes of the BSER analysis that follows.

(2) Distance to Storage for Units Potentially Subject to This Rule

The EPA believes that it is conservative to assume that all 81 GW of capacity with planned operation during or after 2039 would need to construct pipelines to connect to sequestration sites. As detailed in section VII.B.2 of this preamble, the EPA is finalizing an exemption for coal-fired sources permanently ceasing operation by January 1, 2032. About 42 percent (34 GW) of the existing coal-fired steam generation capacity that is currently in operation and has not announced plans to retire prior to 2039 will be 53 years or older by 2032. As discussed in section VII.C.1.a.i(F), from 2000 to 2022, the average age of a coal unit that retired was 53 years old. Therefore, the EPA anticipates that approximately 34 GW of the total capacity may permanently cease operation by 2032 despite not having yet announced plans to do so. Furthermore, of the coal-fired steam generation capacity that has not announced plans to cease operation before 2039 and is further than 100 km (62 miles) of a potential saline sequestration site, 45 percent (7 GW) will be over 53 years old in 2032. Therefore, it is possible that much of the capacity that is further than 100 km (62 miles) of a saline sequestration site and has not announced plans to retire will permanently cease operation due to age before 2032 and thus the rule would not apply to them. Similarly, of the coal-fired steam generation capacity that has not announced plans to cease operation before 2039 and is further than 160 km (100 miles) of a potential saline sequestration site, 56 percent (4 GW) will be over 53 years old in 2032. Therefore, the EPA notes that it is possible that the majority of capacity that is further than 160 km (100 miles) of a saline sequestration and has not announced plans to retire site will permanently cease operation due to age before 2032 and thus be exempt from the requirements of this rule.

The EPA also notes that a majority (56 GW) of the existing coal-fired steam generation capacity that is currently in operation and has not announced plans to permanently cease operation prior to 2039 will be 53 years or older by 2039. Of the coal-fired steam generation capacity with planned operation during or after 2039 that is not located within 100 km (62 miles) of a potential saline sequestration site, the majority (58 percent or 9 GW) of the units will be 53 years or older in 2039. Consequently, the EPA believes that many of these units may permanently cease operation due to age prior to 2039 despite not at this point having announced specific plans to do so, and thereby would likely not be subject to a CCS BSER.

(G) Resources and Workforce To Install CCS

Sufficient resources and an available workforce are required for installation and operation of CCS. Raw materials necessary for CCS are generally available and include common commodities such as steel and concrete for construction of the capture plant, pipelines, and storage wells.

Drawing on data from recently published studies, the DOE completed an order-of-magnitude assessment of the potential requirements for specialized equipment and commodity materials for retrofitting existing U.S. coal-fueled EGUs with CCS. Specialized equipment analyzed included absorbers, strippers, heat exchangers, and compressors. Commodity materials analyzed included monoethanolamine (MEA) solvent for carbon capture, triethylene glycol (TEG) for carbon dioxide drying, and steel and cement for construction of certain aspects of the CCS value chain. The DOE analyzed one scenario in which 42 GW of coal-fueled EGUs are retrofitted with CCS and a second scenario in which 73 GW of coal-fueled EGUs are retrofitted with CCS. The analysis determined that in both scenarios, the maximum annual commodity requirements to construct and operate the CCS systems are likely to be much less than their respective global production rates. The maximum requirements are expected to be at least one order of magnitude lower than global annual production for all of the commodities considered except MEA, which was estimated to be approximately 14 percent of global annual production in the 42 GW scenario and approximately 24 percent of global annual production in the 73 GW scenario. For steel and cement, the maximum annual requirements are also expected to be at least one order of magnitude lower than U.S. annual production rates. Finally, the DOE analysis determined that it is unlikely that the deployment scenarios would encounter any bottlenecks in the supplies of specialized equipment (absorbers, strippers, heat exchangers, and compressors) because of the large pool of potential suppliers.

The workforce necessary for installing and operating CCS is readily available. The required workforce includes construction, engineering, manufacturing, and other skilled labor ( e.g., electrical, plumbing, and mechanical trades). The existing workforce is well positioned to meet the demand for installation and operation of CCS. Many of the skills needed to build and operate carbon capture plants are similar to those used by workers in existing industries, and this experience can be leveraged to support the workforce needed to deploy CCS. In addition, government programs, industry workforce investments, and IRC section 45Q prevailing wage and apprenticeship provisions provide additional significant support to workforce development and demonstrate that the CCS industry likely has the capacity to train and expand the available workforce to meet future needs.

Overall, quantitative estimates of workforce needs indicates that the total number of jobs needed for deploying CCS on coal power plants is significantly less than the size of the existing workforce in adjacent occupations with transferrable skills in the electricity generation and fuels industries. The majority of direct jobs, approximately 90 percent, are expected to be in the construction of facilities, which tend to be project-based. The remaining 10 percent of jobs are expected to be tied to ongoing facility operations and maintenance. Recent project-level estimates bear this out. The Boundary Dam CCS facility in Canada employed 1,700 people at peak construction. A recent workforce projection estimates average annual jobs related to investment in carbon capture retrofits at coal power plants could range from 1,070 to 1,600 jobs per plant. A DOE memorandum estimates that 71,400 to 107,100 average annual jobs resulting from CCS project investments—across construction, project management, machinery installers, sales representatives, freight, and engineering occupations—would likely be needed over a five-year construction period to deploy CCS at a subset of coal power plants. The memorandum further estimates that 116,200 to 174,300 average annual jobs would likely be needed if CCS were deployed at all coal-fired EGUs that currently have no firm commitment to retire or convert to natural gas by 2040. For comparison, the DOE memorandum further categorizes potential workforce needs by occupation, and estimates 11,420 to 27,890 annual jobs for construction trade workers, while the U.S. Energy and Employment Report estimates that electric power generation and fuels accounted for more than 292,000 construction jobs in 2022, which is an order of magnitude greater than the potential workforce needs for CCS deployment under this rule. Overall energy-related construction activities across the entire energy industry accounted for nearly 2 million jobs, or 25 percent of all construction jobs in 2022, indicating that there is a very large pool of workers potentially available.

As noted in section VII.C.1.a.i(F), the EPA determined that the population of sources without announced plans to cease operation or discontinue coal-firing by 2039, and that is therefore potentially subject to a CCS BSER, is not more than 81 GW, as indicated in the final TSD, Power Sector Trends. The DOE CCS Commodity Materials and Workforce Memos evaluated material resource and workforce needs for a similar capacity (about 73 GW), and determined that the resources and workforce available are more than sufficient, in most cases by an order of magnitude. Considering these factors, and the similar scale of the population of sources considered, the EPA therefore concludes that the workforce and resources available are more than sufficient to meet the demands of coal-fired steam generating units potentially subject to a CCS BSER.

(H) Determination That CCS Is “Adequately Demonstrated”

As discussed in detail in section V.C.2.b, pursuant to the text, context, legislative history, and judicial precedent interpreting CAA section 111(a)(1), a technology is “adequately demonstrated” if there is sufficient evidence that the EPA may reasonably conclude that a source that applies the technology will be able to achieve the associated standard of performance under the reasonably expected operating circumstances. Specifically, an adequately demonstrated standard of performance may reflect the EPA's reasonable expectation of what that particular system will achieve, based on analysis of available data from individual commercial scale sources, and, if necessary, identifying specific available technological improvements that are expected to improve performance. The law is clear in establishing that at the time a section 111 rule is promulgated, the system that the EPA establishes as BSER need not be in widespread use. Instead, the EPA's responsibility is to determine that the demonstrated technology can be implemented at the necessary scale in a reasonable period of time, and to base its requirements on this understanding.

In this case, the EPA acknowledged in the proposed rule, and reaffirms now, that sources will require some amount of time to install CCS. Installing CCS requires the building of capture facilities and pipelines to transport captured CO₂ to sequestration sites, and the development of sequestration sites. This is true for both existing coal plants, which will need to retrofit CCS, and new gas plants, which must incorporate CCS into their construction planning. As the EPA explained at proposal, D.C. Circuit caselaw supports this approach. Moreover, the EPA has determined that there will be sufficient resources for all coal-fired power plants that are reasonably expected to be operating as of January 1, 2039, to install CCS. Nothing in the comments alters the EPA's view of the relevant legal requirements related to the EPA's determination of time necessary to allow for adoption of the system.

With all of the above in mind, the preceding sections show that CCS technology with 90 percent capture is clearly adequately demonstrated for coal-fired steam generating units, that the 90 percent standard is achievable, and that it is reasonable for the EPA to determine that CCS can be deployed at the necessary scale in the compliance timeframe.

(1) EPAct05

In the proposal, the EPA noted that in the 2015 NSPS, the EPA had considered coal-fired industrial projects that had installed at least some components of CCS technology. In doing so, the EPA recognized that some of those projects had received assistance in the form of grants, loan guarantees, and Federal tax credits for investment in “clean coal technology,” under provisions of the Energy Policy Act of 2005 (“EPAct05”). See 80 FR 64541-42 (October 23, 2015). (The EPA refers to projects that received assistance under that legislation as “EPAct05-assisted projects.”) The EPA further recognized that the EPAct05 included provisions that constrained how the EPA could rely on EPAct05-assisted projects in determining whether technology is adequately demonstrated for the purposes of CAA section 111. In the 2015 NSPS, the EPA went on to provide a legal interpretation of those constraints. Under that legal interpretation, “these provisions [in the EPAct05] . . . preclude the EPA from relying solely on the experience of facilities that received [EPAct05] assistance, but [do] not . . . preclude the EPA from relying on the experience of such facilities in conjunction with other information.” Id. at 64541-42. In this action, the EPA is adhering to the interpretation of these provisions that it announced in the 2015 NSPS.

Some commenters criticized the legal interpretation that the EPA advanced in the 2015 NSPS, and others supported the interpretation. The EPA has responded to these comments in the Response to Comments Document, available in the docket for this rulemaking.

ii. Costs

The EPA has analyzed the costs of CCS for existing coal-fired long-term steam generating units, including costs for CO₂ capture, transport, and sequestration. The EPA has determined costs of CCS for these sources are reasonable. The EPA also evaluated costs assuming shorter amortization periods. As elsewhere in this section of the preamble, costs are presented in 2019 dollars. In sum, the costs of CCS are reasonable under a variety of metrics. The costs of CCS are reasonable as compared to the costs of other controls that the EPA has required for these sources. And the costs of CCS are reasonable when looking to the dollars per ton of CO₂ reduced. The reasonableness of CCS as an emission control is further reinforced by the fact that some sources are projected to install CCS even in the absence of any EPA rule addressing CO₂ emissions—11 GW of coal-fired EGUs install CCS in the modeling base case.

Specifically, the EPA assessed the average cost of CCS for the fleet of coal-fired steam generating units with no announced retirement or gas conversion prior to 2039. In evaluating costs, the EPA accounts for the IRC section 45Q tax credit of $85/metric ton (assumes prevailing wage and apprenticeship requirements are met), a detailed discussion of which is provided in section VII.C.1.a.ii(C) of this preamble. The EPA also accounts for increases in utilization that will occur for units that apply CCS due to the incentives provided by the IRC section 45Q tax credit. In other words, because the IRC section 45Q tax credit provides a significant economic benefit, sources that apply CCS will have a strong economic incentive to increase utilization and run at higher capacity factors than occurred historically. This assumption is confirmed by the modeling, which projects that sources that install CCS run at a high capacity factor—generally, about 80 percent or even higher. The EPA notes that the NETL Baseline study assumes 85 percent as the default capacity factor assumption for coal CCS retrofits, noting that coal plants in market conditions supporting baseload operation have demonstrated the ability to operate at annual capacity factors of 85 percent or higher. This assumption is also supported by observations of wind generators who receive the IRC section 45 production tax credit who continue to operate even during periods of negative power prices. Therefore, the EPA assessed the costs for CCS retrofitted to existing coal-fired steam generating units assuming an 80 percent annual capacity factor. Assuming an 80 percent capacity factor and 12-year amortization period, the average costs of CCS for the fleet are −$5/ton of CO₂ reduced or −$4/MWh of generation. Assuming at least a 12-year amortization period is reasonable because any unit that installs CCS and seeks to maximize its profitability will be incentivized to recoup the full value of the 12-year tax credit.

Therefore for long-term coal-fired steam generating units—ones that operate after January 1, 2039—the costs of CCS are similar or better than the representative costs of controls detailed in section VII.C.1.a.ii(D) of this preamble ( i.e., costs for SCRs and FGDs on EGUs of $10.60 to $18.50/MWh and the costs in the 2016 NSPS regulating GHGs for the Crude Oil and Natural Gas source category of $98/ton of CO₂ e reduced (80 FR 56627; September 18, 2015)).

The EPA also evaluated the costs for shorter amortization periods, considering the $/MWh and $/ton metrics, as well as other cost indicators, as described in section VII.C.1.a.ii.(D). Specifically, with an initial compliance date of January 1, 2032, sources operating through the end of 2039 have at least 8 years to amortize costs. For an 80 percent capacity factor and an 8-year amortization period, the average costs of CCS for the fleet are $19/ton of CO₂ reduced or $18/MWh of generation; these costs are comparable to those costs that the EPA has previously determined to be reasonable. Sources operating through the end of 2040, 2041, and beyond ( i.e., sources with 9, 10, or more years to amortize the costs of CCS) have even more favorable average costs per MWh and per ton of CO₂ reduced. Sources ceasing operation by January 1, 2039, have 7 years to amortize costs. For an 80 percent capacity factor and a 7-year amortization period, the fleet average costs are $29/ton of CO₂ reduced or $28/MWh of generation; these average costs are less comparable on a $/MWh of generation basis to those costs the EPA has previously determined to be reasonable, but substantially lower than costs the EPA has previously determined to be reasonable on a $/ton of CO₂ reduced basis. The EPA further notes that the costs presented are average costs for the fleet. For a substantial amount of capacity, costs assuming a 7-year amortization period are comparable to those costs the EPA has previously determined to be reasonable on both a $/MWh basis ( i.e., less than $18.50/MWh) and a $/ton basis ( i.e. less than $98/ton CO2e), and the EPA concludes that a substantial amount of capacity can install CCS at reasonable cost with a 7-year amortization period. Considering that a significant number of sources can cost reasonably install CCS even assuming a 7-year amortization period, the EPA concludes that sources operating in 2039 should be subject to a CCS BSER, and for this reason, is finalizing the date of January 1, 2039 as the dividing line between the medium-term and long-term subcategories. Moreover, the EPA underscores that given the strong economic incentives of the IRC section 45Q tax credit, sources that install CCS will have strong economic incentives to operate at high capacity for the full 12 years that the tax credit is available.

As discussed in the RTC section 2.16, the EPA has also examined the reasonableness of the costs of this rule in additional ways: considering the total annual costs of the rule as compared to past CAA rules for the electricity sector and as compared to the industry's annual revenues and annual capital expenditures, and considering the effects of this rule on electricity prices. Taking all of these into consideration, in addition to the cost metrics just discussed, the EPA concludes that, in general, the costs of CCS are reasonable for sources operating after January 1, 2039.

(A) Capture Costs

The EPA developed an independent engineering cost assessment for CCS retrofits, with support from Sargent and Lundy. The EPA cost analysis assumes installation of one CO₂ capture plant for each coal-fired EGU, and that sources without SO₂ controls (FGD) or NO_X controls (specifically, selective catalytic reduction—SCR; or selective non-catalytic reduction—SNCR) add a wet FGD and/or SCR.

(B) CO₂ Transport and Sequestration Costs

To calculate the costs of CCS for coal-fired steam generating units for purposes of determining BSER as well as for EPA modeling, the EPA relied on transportation and storage costs consistent with the cost of transporting and storing CO₂ from each power plant to the nearest saline reservoir. For a power plant composed of multiple coal-fired EGUs, the EPA's cost analysis assumes installation and operation of a single, common CO₂ pipeline.

The EPA notes that NETL has also developed costs for transport and storage. NETL's “Quality Guidelines for Energy System Studies; Carbon Dioxide Transport and Sequestration Costs in NETL Studies” provides an estimation of transport costs based on the CO₂ Transport Cost Model. The CO₂ Transport Cost Model estimates costs for a single point-to-point pipeline. Estimated costs reflect pipeline capital costs, related capital expenditures, and operations and maintenance costs.

NETL's Quality Guidelines also provide an estimate of sequestration costs. These costs reflect the cost of site screening and evaluation, permitting and construction costs, the cost of injection wells, the cost of injection equipment, operation and maintenance costs, pore volume acquisition expense, and long-term liability protection. Permitting and construction costs also reflect the regulatory requirements of the UIC Class VI program and GHGRP subpart RR for geologic sequestration of CO₂ in deep saline formations. NETL calculates these sequestration costs on the basis of generic plant locations in the Midwest, Texas, North Dakota, and Montana, as described in the NETL energy system studies that utilize the coal found in Illinois, East Texas, Williston, and Powder River basins.

There are two primary cost drivers for a CO₂ sequestration project: the rate of injection of the CO₂ into the reservoir and the areal extent of the CO₂ plume in the reservoir. The rate of injection depends, in part, on the thickness of the reservoir and its permeability. Thick, permeable reservoirs provide for better injection and fewer injection wells. The areal extent of the CO₂ plume depends on the sequestration capacity of the reservoir. Thick, porous reservoirs with a good sequestration coefficient will present a small areal extent for the CO₂ plume and have a smaller monitoring footprint, resulting in lower monitoring costs. NETL's Quality Guidelines model costs for a given cumulative sequestration potential.

In addition, provisions in the IIJA and IRA are expected to significantly increase the CO₂ pipeline infrastructure and development of sequestration sites, which, in turn, are expected to result in further cost reductions for the application of CCS at new combined cycle EGUs. The IIJA establishes a new Carbon Dioxide Transportation Infrastructure Finance and Innovation program to provide direct loans, loan guarantees, and grants to CO₂ infrastructure projects, such as pipelines, rail transport, ships and barges. The IIJA also establishes a new Regional Direct Air Capture Hubs program that includes funds to support four large-scale, regional direct air capture hubs and more broadly support projects that could be developed into a regional or inter-regional network to facilitate sequestration or utilization. DOE is additionally implementing IIJA section 40305 (Carbon Storage Validation and Testing) through its CarbonSAFE initiative, which aims to further develop geographically widespread, commercial-scale, safe sequestration. The IRA increases and extends the IRC section 45Q tax credit, discussed next.

(C) IRC Section 45Q Tax Credit

In determining the cost of CCS, the EPA is taking into account the tax credit provided under IRC section 45Q, as revised by the IRA. The tax credit is available at $85/metric ton ($77/ton) and offsets a significant portion of the capture, transport, and sequestration costs noted above.

Several other aspects of the tax credit should be noted. A tax credit offsets tax liability dollar for dollar up to the amount of the taxpayer's tax liability. Any credits in excess of the taxpayer's liability are eligible to be carried back (3 years in the case of IRC section 45Q) and then carried forward up to 20 years. As noted above, the IRA also enabled additional methods to monetize tax credits in the event the taxpayer does not have sufficient tax liability, such as through credit transfer.

The EPA has determined that it is likely that EGUs installing CCS will meet the 45Q prevailing wage and apprenticeship requirements. First, the requirements provide a significant economic incentive, increasing the value of the 45Q credit by five times over the base value of the credit available if the prevailing wage and apprenticeship requirements are not met. This provides a significant incentive to meet the requirements. Second, the increased cost of meeting the requirements is likely significantly less than the increase in credit value. A recent EPRI assessment found meeting the requirements for other types of power generation projects resulted in significant savings across projects, and other studies indicate prevailing wage laws and requirements for construction projects in general do not significantly affect overall construction costs. The EPA expects a similar dynamic for 45Q projects. Third, the use of registered apprenticeship programs for training new employees is generally well-established in the electric power generation sector, and apprenticeship programs are widely available to generate additional trained workers in this field. The overall U.S. apprentice market has more than doubled between 2014 and 2023, growing at an average annual rate of more than 7 percent. Additional programs support the skilled construction trade workforce required for CCS implementation and maintenance.

As discussed in section V.C.2.c of this preamble, CAA section 111(a)(1) is clear that the cost that the Administrator must take into account in determining the BSER is the cost of the controls to the source. It is reasonable to take the tax credit into account because it reduces the cost of the controls to the source, which has a significant effect on the actual cost of installing and operating CCS. In addition, all sources that install CCS to meet the requirements of these final actions are eligible for the tax credit. The legislative history of the IRA makes clear that Congress was well aware that the EPA may promulgate rulemaking under CAA section 111 based on CCS and the utility of the tax credit in reducing the costs of CCUS ( i.e., CCS). Rep. Frank Pallone, the chair of the House Energy & Commerce Committee, included a statement in the Congressional Record when the House adopted the IRA in which he explained: “The tax credit[ ] for CCUS . . . included in this Act may also figure into CAA Section 111 GHG regulations for new and existing industrial sources[.] . . . Congress anticipates that EPA may consider CCUS . . . as [a] candidate[ ] for BSER for electric generating plants . . . . Further, Congress anticipates that EPA may consider the impact of the CCUS . . . tax credit[ ] in lowering the costs of [that] measure[ ].” 168 Cong. Rec. E879 (August 26, 2022) (statement of Rep. Frank Pallone).

In the 2015 NSPS, in which the EPA determined partial CCS to be the BSER for GHGs from new coal-fired steam generating EGUs, the EPA recognized that the IRC section 45Q tax credit or other tax incentives could factor into the cost of the controls to the sources. Specifically, the EPA calculated the cost of partial CCS on the basis of cost calculations from NETL, which included “a range of assumptions including the projected capital costs, the cost of financing the project, the fixed and variable O&M costs, the projected fuel costs, and incorporation of any incentives such as tax credits or favorable financing that may be available to the project developer.” 80 FR 64570 (October 23, 2015).

Similarly, in the 2015 NSPS, the EPA also recognized that revenues from utilizing captured CO₂ for EOR would reduce the cost of CCS to the sources, although the EPA did not account for potential EOR revenues for purposes of determining the BSER. Id. At 64563-64. In other rules, the EPA has considered revenues from sale of the by-products of emission controls to affect the costs of the emission controls. For example, in the 2016 Oil and Gas Methane Rule, the EPA determined that certain control requirements would reduce natural gas leaks and therefore result in the collection of recovered natural gas that could be sold; and the EPA further determined that revenues from the sale of the recovered natural gas reduces the cost of controls. See 81 FR 35824 (June 3, 2016). The EPA made the same determination in the 2024 Oil and Gas Methane Rule. See 89 FR 16820, 16865 (May 7, 2024). In a 2011 action concerning a regional haze SIP, the EPA recognized that a NO_X control would alter the chemical composition of fly ash that the source had previously sold, so that it could no longer be sold; and as a result, the EPA further determined that the cost of the NO_X control should include the foregone revenues from the fly ash sales. 76 FR 58570, 58603 (September 21, 2011). In the 2016 emission guidelines for landfill gas from municipal solid waste landfills, the EPA reduced the costs of controls by accounting for revenue from the sale of electricity produced from the landfill gas collected through the controls. 81 FR 59276, 19679 (August 29, 2016).

The amount of the IRC section 45Q tax credit that the EPA is taking into account is $85/metric ton for CO₂ that is captured and geologically stored. This amount is available to the affected source as long as it meets the prevailing wage and apprenticeship requirements of IRC section 45Q(h)(3)-(4). The legislative history to the IRA specifically stated that when the EPA considers CCS as the BSER for GHG emissions from industrial sources in CAA section 111 rulemaking, the EPA should determine the cost of CCS by assuming that the sources would meet those prevailing wage and apprenticeship requirements. 168 Cong. Rec. E879 (August 26, 2022) (statement of Rep. Frank Pallone). If prevailing wage and apprenticeship requirements are not met, the value of the IRC section 45Q tax credit falls to $17/metric ton. The substantially higher credit available provides a considerable incentive to meeting the prevailing wage and apprenticeship requirements. Therefore, the EPA assumes that investors maximize the value of the IRC section 45Q tax credit at $85/metric ton by meeting those requirements.

(D) Comparison to Other Costs of Controls and Other Measures of Cost Reasonableness

In assessing cost reasonableness for the BSER determination for this rule, the EPA looks at a range of cost information. As discussed in Chapter 2 of the RTC, the EPA considered the total annual costs of the rule as compared to past CAA rules for the electricity sector and as compared to the industry's annual revenues and annual capital expenditures, and considered the effects of this rule on electricity prices.

For each of the BSER determinations, the EPA also considers cost metrics that it has historically considered in assessing costs to compare the costs of GHG control measures to control costs that the EPA has previously determined to be reasonable. This includes comparison to the costs of controls at EGUs for other air pollutants, such as SO₂ and NO_X, and costs of controls for GHGs in other industries. Based on these costs, the EPA has developed two metrics for assessing the cost reasonableness of controls: the increase in cost of electricity due to controls, measured in $/MWh, and the control costs of removing a ton of pollutant, measured in $/ton CO₂ e. The costs presented in this section of the preamble are in 2019 dollars.

In different rulemakings, the EPA has required many coal-fired steam generating units to install and operate flue gas desulfurization (FGD) equipment—that is, wet or dry scrubbers—to reduce their SO₂ emissions or SCR to reduce their NO_X emissions. The EPA compares these control costs across technologies—steam generating units and combustion turbines—because these costs are indicative of what is reasonable for the power sector in general. The facts that the EPA required these controls in prior rules, and that many EGUs subsequently installed and operated these controls, provide evidence that these costs are reasonable, and as a result, the cost of these controls provides a benchmark to assess the reasonableness of the costs of the controls in this preamble. In the 2011 CSAPR (76 FR 48208; August 8, 2011), the EPA estimated the annualized costs to install and operate wet FGD retrofits on existing coal-fired steam generating units. Using those same cost equations and assumptions ( i.e., a 63 percent annual capacity factor—the average value in 2011) for retrofitting wet FGD on a representative 700 to 300 MW coal-fired steam generating unit results in annualized costs of $14.80 to $18.50/MWh of generation, respectively. In the Good Neighbor Plan for the 2015 Ozone NAAQS (2023 GNP), 88 FR 36654 (June 5, 2023), the EPA estimated the annualized costs to install and operate SCR retrofits on existing coal-fired steam generating units. Using those same cost equations and assumptions (including a 56 percent annual capacity factor—a representative value in that rulemaking) to retrofit SCR on a representative 700 to 300 MW coal-fired steam generating unit results in annualized costs of $10.60 to $11.80/MWh of generation, respectively.

The EPA also compares costs to the costs for GHG controls in rulemakings for other industries. In the 2016 NSPS regulating GHGs for the Crude Oil and Natural Gas source category, the EPA found the costs of reducing methane emissions of $2,447/ton to be reasonable (80 FR 56627; September 18, 2015). Converted to a ton of CO₂ e reduced basis, those costs are expressed as $98/ton of CO₂ e reduced.

The EPA does not consider either of these metrics, $18.50/MWh and $98/ton of CO₂ e, to be bright line standards that distinguish between levels of control costs that are reasonable and levels that are unreasonable. But they do usefully indicate that control costs that are generally consistent with those levels of control costs should be considered reasonable. The EPA has required controls with comparable costs in prior rules for the electric power industry and the industry has successfully complied with those rules by installing and operating the applicable controls. In the case of the $/ton metric, the EPA has required other industries—specifically, the oil and gas industry—to reduce their climate pollution at this level of cost-effectiveness. In this rulemaking, the costs of the controls that the EPA identifies as the BSER generally match up well against both of these $/MWh and $/ton metrics for the affected subcategories of sources. And looking broadly at the range of cost information and these cost metrics, the EPA concludes that the costs of these rules are reasonable.

(E) Comparison to Costs for CCS in Prior Rulemakings

In the CPP and ACE Rule, the EPA determined that CCS did not qualify as the BSER due to cost considerations. Two key developments have led the EPA to reevaluate this conclusion: the costs of CCS technology have fallen and the extension and increase in the IRC section 45Q tax credit, as included in the IRA, in effect provide a significant stream of revenue for sequestered CO₂ emissions. The CPP and ACE Rule relied on a 2015 NETL report estimating the cost of CCS. NETL has issued updated reports to incorporate the latest information available, most recently in 2022, which show significant cost reductions. The 2015 report estimated incremental levelized cost of CCS at a new pulverized coal facility relative to a new facility without CCS at $74/MWh (2022$), while the 2022 report estimated incremental levelized cost at $44/MWh (2022$). Additionally, the IRA increased the IRC section 45Q tax credit from $50/metric ton to $85/metric ton (and, in the case of EOR or certain industrial uses, from $35/metric ton to $60/metric ton), assuming prevailing wage and apprenticeship conditions are met. The IRA also enhanced the realized value of the tax credit through the elective pay (informally known as direct pay) and transferability monetization options described in section IV.E.1. The combination of lower costs and higher tax credits significantly improves the cost reasonableness of CCS for purposes of determining whether it qualifies as the BSER.

iii. Non-Air Quality Health and Environmental Impact and Energy Requirements

The EPA considered non-GHG emissions impacts, the water use impacts, the transport and sequestration of captured CO₂, and energy requirements resulting from CCS for steam generating units. As discussed below, where the EPA has found potential for localized adverse consequences related to non-air quality health and environmental impacts or energy requirements, the EPA also finds that protections are in place to mitigate those risks. Because the non-air quality health and environmental impacts are closely related to the energy requirements, we discuss the latter first.

(A) Energy Requirements

For a steam generating unit with 90 percent amine-based CO₂ capture, parasitic/auxiliary energy demand increases and the net power output decreases. In particular, the solvent regeneration process requires heat in the form of steam and CO₂ compression requires a large amount of electricity. Heat and power for the CO₂ capture equipment can be provided either by using the steam and electricity produced by the steam generating unit or by an auxiliary cogeneration unit. However, any auxiliary source of heat and power is part of the “designated facility,” along with the steam generating unit. The standards of performance apply to the designated facility. Thus, any CO₂ emissions from the connected auxiliary equipment need to be captured or they will increase the facility's emission rate.

Using integrated heat and power can reduce the capacity ( i.e., the amount of electricity that a unit can distribute to the grid) of an approximately 474 MW-net (501 MW-gross) coal-fired steam generating unit without CCS to approximately 425 MW-net with CCS and contributes to a reduction in net efficiency of 23 percent. For retrofits of CCS on existing sources, the ductwork for flue gas and piping for heat integration to overcome potential spatial constraints are a component of efficiency reduction. The EPA notes that slightly greater efficiency reductions than in the 2016 NETL retrofit report are assumed for the BSER cost analyses, as detailed in the final TSD, GHG Mitigation Measures for Steam Generating Units, available in the docket. Despite decreases in efficiency, IRC section 45Q tax credit provides an incentive for increased generation with full operation of CCS because the amount of revenue from the tax credit is based on the amount of captured and sequestered CO₂ emissions and not the amount of electricity generated. In this final action, the Agency considers the energy penalty to not be unreasonable and to be relatively minor compared to the benefits in GHG reduction of CCS.

(B) Non-GHG Emissions

As a part of considering the non-air quality health and environmental impacts of CCS, the EPA considered the potential non-GHG emission impacts of CO₂ capture. The EPA recognizes that amine-based CO₂ capture can, under some circumstances, result in the increase in emission of certain co-pollutants at a coal-fired steam generating unit. However, there are protections in place that can mitigate these impacts. For example, as discussed below, CCS retrofit projects with co-pollutant increases may be subject to preconstruction permitting under the New Source Review (NSR) program, which could require the source to adopt emission limitations based on applicable NSR requirements. Sources obtaining major NSR permits would be required to either apply Lowest Achievable Emission Rate (LAER) and fully offset any anticipated increases in criteria pollutant emissions (for their nonattainment pollutants) or apply Best Available Control Technology (BACT) and demonstrate that its emissions of criteria pollutants will not cause or contribute to a violation of applicable National Ambient Air Quality Standards (for their attainment pollutants). The EPA expects facility owners, states, permitting authorities, and other responsible parties will use these protections to address co-pollutant impacts in situations where individual units use CCS to comply with these emission guidelines.

The EPA also expects that the meaningful engagement requirements discussed in section X.E.1.b.i of this preamble will ensure that all interested stakeholders, including community members who might be adversely impacted by non-GHG pollutants, will have an opportunity to raise this concern with states and permitting authorities. Additionally, state permitting authorities are, in general, required to provide notice and an opportunity for public comment on construction projects that require NSR permits. This provides additional opportunities for affected stakeholders to engage in that process, and it is the EPA's expectation that the responsible authorities will consider these concerns and take full advantage of existing protections. Moreover, the EPA through its regional offices is committed to thoroughly review draft NSR permits associated with CO₂ capture projects and provide comments as necessary to state permitting authorities to address any concerns or questions with regard to the draft permit's consideration and treatment of non-GHG pollutants.

In the following discussion, the EPA describes the potential emissions of non-GHG pollutants resulting from installation and operation of CO₂ capture plants, the protections in place such as the controls and processes for mitigating those emissions, as well as regulations and permitting that may require review and implementation of those controls. The EPA first discusses these issues in relation to criteria air pollutants and precursor pollutants (SO₂, NO_X, and PM), and subsequently provides details regarding hazardous air pollutants (HAPs) and volatile organic compounds (VOCs).

Operation of an amine-based CO₂ capture plant on a coal-fired steam generating unit can impact the emission of criteria pollutants from the facility, including SO₂ and PM, as well as precursor pollutants, like NO_X. Sources installing CCS may operate more due to the incentives provided by the IRC section 45Q tax credit, and increased utilization would—all else being equal—result in increases in SO₂, PM, and NO_X . However, certain impacts are mitigated by the flue gas conditioning required by the CO₂ capture process and by other control equipment that the units already have or may need to install to meet other CAA requirements. Substantial flue gas conditioning, particularly to remove SO₂ and PM, is critical to limiting solvent degradation and maintaining reliable operation of the capture plant. To achieve the necessary limits on SO₂ levels in the flue gas for the capture process, steam generating units will need to add an FGD scrubber, if they do not already have one, and will usually need an additional polishing column ( i.e., quencher), thereby further reducing the emission of SO_2. A wet FGD column and a polishing column will also reduce the emission rate of PM. Additional improvements in PM removal may also be necessary to reduce the fouling of other components ( e.g., heat exchangers) of the capture process, including upgrades to existing PM controls or, where appropriate, the inclusion of various wash stages to limit fly ash carry-over to the CO₂ removal system. Although PM emissions from the steam generating unit may be reduced, PM emissions may occur from cooling towers for those sources using wet cooling for the capture process. For some sources, a WESP may be necessary to limit the amount of aerosols in the flue gas prior to the CO₂ capture process. Reducing the amount of aerosols to the CO₂ absorber will also reduce emissions of the solvent out of the top of the absorber. Controls to limit emission of aerosols installed at the outlet of the absorber could be considered, but could lead to higher pressure drops. Thus, emission increases of SO₂ and PM would be reduced through flue gas conditioning and other system requirements of the CO₂ capture process, and NSR permitting would serve as an added backstop to review remaining SO₂ and PM increases for mitigation.

NO_X emissions can cause solvent degradation and nitrosamine formation, depending on the chemical structure of the solvent. Limits on NO_X levels of the flue gas required to avoid solvent degradation and nitrosamine formation in the CO₂ scrubber vary. For most units, the requisite limits on NO_X levels to assure that the CO₂ capture process functions properly may be met by the existing NO_X combustion controls. Other units may need to install SCR to achieve the required NOx level. Most existing coal-fired steam generating units either already have SCR or will be covered by final Federal Implementation Plan (FIP) requirements regulating interstate transport of NO_X (as ozone precursors) from EGUs. See 88 FR 36654 (June 5, 2023). For units not otherwise required to have SCR, an increase in utilization from a CO₂ capture retrofit could result in increased NO_X emissions at the source that, depending on the quantity of the emissions increase, may trigger major NSR permitting requirements. Under this scenario, the permitting authority may determine that the NSR permit requires the installation of SCR for those units, based on applying the control technology requirements of major NSR. Alternatively, a state could, as part of its state plan, develop enforceable conditions for a source expected to trigger major NSR that would effectively limit the unit's ability to increase its emissions in amounts that would trigger major NSR. Under this scenario, with no major NSR requirements applying due to the limit on the emissions increase, the permitting authority may conclude for the minor NSR permit that installation of SCR is not required for the units and the source is to minimize its NO_X emission increases using other techniques. Finally, a source with some lesser increase in NO_X emissions may not trigger major NSR to begin with and, as with the previous scenario, the permitting authority would determine the NO_X control requirements pursuant to its minor NSR program requirements.

Recognizing that potential emission increases of SO₂, PM, and NO_X from operating a CO₂ capture process are an area of concern for stakeholders, the EPA plans to review and update as needed its guidance on NSR permitting, specifically with respect to BACT determinations for GHG emissions and consideration of co-pollutant increases from sources installing CCS. In its analysis to support this final action, the EPA accounted for controlling these co-pollutant increases by assuming that coal-fired units that install CCS would be required to install SCR and/or FGD if they do not already have those controls installed. The costs of these controls are included in the total program compliance cost estimates through IPM modeling, as well as in the BSER cost calculations.

An amine-based CO₂ capture plant can also impact emissions of HAP and VOC (as an ozone precursor) from the coal-fired steam generating unit. Degradation of the solvent can produce HAP, and organic HAP and amine solvent emissions from the absorber would contribute to VOC emissions out of the top of the CO₂ absorber. A conventional multistage water or acid wash and mist eliminator (demister) at the exit of the CO₂ scrubber is effective at removal of gaseous amine and amine degradation products ( e.g., nitrosamine) emissions. The DOE's Carbon Management Pathway report notes that monitoring and emission controls for such degradation products are currently part of standard operating procedures for amine-based CO₂ capture systems. Depending on the solvent properties, different amounts of aldehydes including acetaldehyde and formaldehyde may form through oxidative processes, contributing to total HAP and VOC emissions. While a water wash or acid wash can be effective at limiting emission of amines, a separate system of controls would be required to reduce aldehyde emissions; however, the low temperature and likely high water vapor content of the gas emitted out of absorber may limit the applicability of catalytic or thermal oxidation. Other controls ( e.g., electrochemical, ultraviolet) common to water treatment could be considered to reduce the loading of copollutants in the water wash section, although their efficacy is still in development and it is possible that partial treatment could result in the formation of additional degradation products. Apart from these potential controls, any increase in VOC emissions from a CCS retrofit project would be mitigated through NSR permitting. As such VOC increases are not expected to be large enough to trigger major NSR requirements, they would likely be reviewed and addressed under a state's minor NSR program.

There is one nitrosamine that is a listed HAP regulated under CAA section 112. Carbon capture systems that are themselves a major source of HAP should evaluate the applicability of CAA section 112(g) and conduct a case-by-case MACT analysis if required, to establish MACT for any listed HAP, including listed nitrosamines, formaldehyde, and acetaldehyde. Because of the differences in the formation and effectiveness of controls, such a case-by-case MACT analysis should evaluate the performance of controls for nitrosamines and aldehydes separately, as formaldehyde or acetaldehyde may not be a suitable surrogate for amine and nitrosamine emissions. However, measurement of nitrosamine emissions may be challenging when the concentration is low ( e.g., less than 1 part per billion, dry basis).

HAP emissions from the CO₂ capture plant will depend on the flue gas conditions, solvent, size of the source, and process design. The air permit application for Project Tundra includes potential-to-emit (PTE) values for CAA section 112 listed HAP specific to the 530 MW-equivalent CO₂ capture plant, including emissions of 1.75 tons per year (TPY) of formaldehyde (CASRN 50-00-0), 32.9 TPY of acetaldehyde (CASRN 75-07-0), 0.54 TPY of acetamide (CASRN 60-35-5), 0.018 TPY of ethylenimine (CASRN 151-56-4), 0.044 TPY of N-nitrosodimethylamine (CASRN 62-75-9), and 0.018 TPY of N-nitrosomorpholine (CASRN 59-89-2). Additional PTE other species that are not CAA section 112 listed HAP were also included, including 0.022 TPY of N-nitrosodiethylamine (CASRN 55-18-5). PTE values for other CO₂ capture plants may differ. To comply with North Dakota Department of Environmental Quality (ND-DEQ) Air Toxics Policy, an air toxics assessment was included in the permit application. According to that assessment, the total maximum individual carcinogenic risk was 1.02E-6 (approximately 1-in-1 million, below the ND-DEQ threshold of 1E-5) primarily driven by N-nitrosodiethylamine and N-nitrosodimethylamine. The hazard index value was 0.022 (below the ND-DEQ threshold of 1), with formaldehyde being the primary driver. Results of air toxics risk assessments for other facilities would depend on the emissions from the facility, controls in place, stack height and flue gas conditions, local ambient conditions, and the relative location of the exposed population.

Emissions of amines and nitrosamines at Project Tundra are controlled by the water wash section of the absorber column. According to the permit to construct issued by ND-DEQ, limits for formaldehyde and acetaldehyde will be established based on testing after initial operation of the CO₂ capture plant. The permit does not include a mechanism for establishing limits for nitrosamine emissions, as they may be below the limit of detection (less than 1 part per billion, dry basis).

The EPA received several comments related to the potential for non-GHG emissions associated with CCS. Those comments and the EPA's responses are as follows.

Comment: Some commenters noted that there is a potential for increases in co-pollutants when operating amine-based CO₂ capture systems. One commenter requested that the EPA proactively regulate potential nitrosamine emissions.

Response: The EPA carefully considered these concerns as it finalized its determination of the BSERs for these rules. The EPA takes these concerns seriously, agrees that any impacts to local and downwind communities are important to consider and has done so as part of its analysis discussed at section XII.E. While the EPA acknowledges that, in some circumstances, there is potential for some non-GHG emissions to increase, there are several protections in place to help mitigate these impacts. The EPA believes that these protections, along with the meaningful engagement of potentially affected communities, can facilitate a responsible deployment of this technology that mitigates the risk of any adverse impacts.

There is one nitrosamine that is a listed HAP under CAA section 112 (N-Nitrosodimethylamine; CASRN 62-75-9). Other nitrosamines would have to be listed before the EPA could establish regulations limiting their emission. Furthermore, carbon capture systems are themselves not a listed source category of HAP, and the listing of a source category under CAA section 112 would first require some number of the sources to exist for the EPA to develop MACT standards. However, if a new CO₂ capture facility were to be permitted as a separate entity (rather than as part of the EGU) then it may be subject to case-by-case MACT under section 112(g), as detailed in the preceding section of this preamble.

Comment: Commenters noted that a source could attempt to permit CO₂ facilities as separate entities to avoid triggering NSR for the EGU.

Response: For the CO₂ capture plant to be permitted as a separate entity, the source would have to demonstrate to the state permitting authority that the EGU and CO₂ capture plant are not a single stationary source under the NSR program. In determining what constitutes a stationary source, the EPA's NSR regulations set forth criteria that are to be used when determining the scope of a “stationary source.” These criteria require the aggregation of different pollutant-emitting activities if they (1) belong to the same industrial grouping as defined by SIC codes, (2) are located on contiguous or adjacent properties, and (3) are under common control. In the case of an EGU and CO₂ capture plant that are collocated, to permit them as separate sources they should not be under common control or not be defined by the same industrial grouping.

The EPA would anticipate that, in most cases, the operation of the EGU and the CO₂ capture plant will intrinsically affect one another—typically steam, electricity, and the flue gas of the EGU will be provided to the CO₂ capture plant. Conditions of the flue gas will affect the operation of the CO₂ capture plant, including its emissions, and the steam and electrical load will affect the operation of the EGU. Moreover, the emissions from the EGU will be routed through the CO₂ capture system and emitted out of the top of the CO₂ absorber. Even if the EGU and CO₂ capture plant are owned by separate entities, the CO₂ capture plant is likely to be on or directly adjacent to land owned by the owners of the EGU and contractual obligations are likely to exist between the two owners. While each of these individual factors may not ultimately determine the outcome of whether two nominally-separate facilities should be treated as a single stationary source for permitting purposes, the EPA expects that in most cases an EGU and its collocated CO₂ capture plant would meet each of the aforementioned NSR regulatory criteria necessary to make such a determination. Thus, the EPA generally would not expect an EGU and its CO₂ capture plant to be permitted as separate stationary sources.

(C) Water Use

Water consumption at the plant increases when applying carbon capture, due to solvent water makeup and cooling demand. Water consumption can increase by 36 percent on a gross basis. A separate cooling water system dedicated to a CO₂ capture plant may be necessary. However, the amount of water consumption depends on the design of the cooling system. For example, the cooling system cited in the CCS feasibility study for SaskPower's Shand Power station would rely entirely on water condensed from the flue gas and thus would not require any increase in external water consumption—all while achieving higher capture rates at lower cost than Boundary Dam Unit 3. Regions with limited water supply may therefore rely on dry or hybrid cooling systems. Therefore, the EPA considers the water use requirements to be manageable and does not expect this consideration to preclude coal-fired power plants generally from being able to install and operate CCS.

(D) CO₂ Capture Plant Siting

With respect to siting considerations, CO₂ capture systems have a sizeable physical footprint and a consequent land-use requirement. One commenter cited their analysis showing that, for a subset of coal-fired sources greater than 300 MW, 98 percent (154 GW of the existing fleet) have adjacent land available within 1 mile of the facility, and 83 percent have adjacent land available within 100 meters of the facility. Furthermore, the cited analysis did not include land available onsite, and it is therefore possible there is even greater land availability for siting capture equipment. Qualitatively, some commenters claimed there is limited land available for siting CO₂ capture plants adjacent to coal-fired steam generating units. However, those commenters provided no data or analysis to support their assertion. The EPA has reviewed the analysis provided by the first commenter, and the approach, methods, and assumptions are logical. Further, the EPA has reviewed the available information, including the location of coal-fired steam generating units and visual inspection of the associated maps and plots. Although in some cases longer duct runs may be required, this would not preclude coal-fired power plants generally from being able to install and operate CCS. Therefore, the EPA has concluded that siting and land-use requirements for CO₂ capture are not unreasonable.

(E) Transport and Geologic Sequestration

As noted in section VII.C.1.a.i(C) of this preamble, PHMSA oversight of supercritical CO₂ pipeline safety protects against environmental release during transport. The vast majority of CO₂ pipelines have been operating safely for more than 60 years. PHMSA reported a total of 102 CO₂ pipeline incidents between 2003 and 2022, with one injury (requiring in-patient hospitalization) and zero fatalities. In the past 20 years, 500 million metric tons of CO₂ moved through over 5,000 miles of CO₂ pipelines with zero incidents involving fatalities. PHMSA initiated a rulemaking in 2022 to develop and implement new measures to strengthen its safety oversight of supercritical CO₂ pipelines. Furthermore, UIC Class VI and Class II regulations under the SDWA, in tandem with GHGRP subpart RR and subpart VV requirements, ensure the protection of USDWs and the security of geologic sequestration. The EPA believes these protections constitute an effective framework for addressing potential health and environmental concerns related to CO₂ transportation and sequestration, and the EPA has taken this regulatory framework into consideration in determining that CCS represents the BSER for long-term steam EGUs.

(F) Impacts on the Energy Sector

Additionally, the EPA considered the impacts on the power sector, on a nationwide and long-term basis, of determining CCS to be the BSER for long-term coal-fired steam generating units. In this final action, the EPA considers that designating CCS as the BSER for these units would have limited and non-adverse impacts on the long-term structure of the power sector or on the reliability of the power sector. Absent the requirements defined in this action, the EPA projects that 11 GW of coal-fired steam generating units would apply CCS by 2035 and an additional 30 GW of coal-fired steam generating units, without controls, would remain in operation in 2040. Designating CCS to be the BSER for existing long-term coal-fired steam generating units may result in more of the coal-fired steam generating unit capacity applying CCS. The time available before the compliance deadline of January 1, 2032, provides for adequate resource planning, including accounting for the downtime necessary to install the CO₂ capture equipment at long-term coal-fired steam generating units. For the 12-year duration that eligible EGUs earn the IRC section 45Q tax credit, long-term coal-fired steam generating units are anticipated to run at or near base load conditions in order to maximize the amount of tax credit earned through IRC section 45Q. Total generation from coal-fired steam generating units in the medium-term subcategory would gradually decrease over an extended period of time through 2039, subject to the commitments those units have chosen to adopt. Additionally, for the long-term units applying CCS, the EPA has determined that the increase in the annualized cost of generation is reasonable. Therefore, the EPA concludes that these elements of BSER can be implemented while maintaining a reliable electric grid. A broader discussion of reliability impacts of these final rules is available in section XII.F of this preamble.

iv. Extent of Reductions in CO₂ Emissions

CCS is an extremely effective technology for reducing CO₂ emissions. As of 2021, coal-fired power plants are the largest stationary source of GHG emissions by sector. Furthermore, emission rates (lb CO₂ /MWh-gross) from coal-fired sources are almost twice those of natural gas-fired combined cycle units, and sources operating in the long-term have the more substantial emissions potential. CCS can be applied to coal-fired steam generating units at the source to reduce the mass of CO₂ emissions by 90 percent or more. Increased steam and power demand have a small impact on the reduction in emission rate ( i.e., lb CO₂ /MWh-gross) that occurs with 90 percent capture. According to the 2016 NETL Retrofit report, 90 percent capture will result in emission rates that are 88.4 percent lower on a lb/MWh-gross basis and 87.1 percent lower on a lb/MWh-net basis compared to units without capture. After capture, CO₂ can be transported and securely sequestered. Although steam generating units with CO₂ capture will have an incentive to operate at higher utilization because the cost to install the CCS system is largely fixed and the IRC section 45Q tax credit increases based on the amount of CO₂ captured and sequestered, any increase in utilization will be far outweighed by the substantial reductions in emission rate.

v. Promotion of the Development and Implementation of Technology

The EPA considered the potential impact on technology advancement of designating CCS as the BSER for long-term coal-fired steam generating units, and in this final rule, the EPA considers that designating CCS as the BSER will provide for meaningful advancement of CCS technology. As indicated above, the EPA's IPM modeling indicates that 11 GW of coal-fired power plants install CCS and generate 76 terawatt-hours (TWh) per year in the base case, and that another 8 GW of plants install CCS and generate another 57 TWh per year in the policy case. In this manner, this rule advances CCS technology more widely throughout the coal-fired power sector. As discussed in section VIII.F.4.c.iv(G) of this preamble, this rule advances CCS technology for new combined cycle base load combustion turbines, as well. It is also likely that this rule supports advances in the technology in other industries.

vi. Comparison With 2015 NSPS For Newly Constructed Coal-Fired EGUs

In the 2015 NSPS, the EPA determined that the BSER for newly constructed coal-fired EGUs was based on CCS with 16 to 23 percent capture, based on the type of coal combusted, and consequently, the EPA promulgated standards of performance of 1,400 lb CO₂ /MWh-g. 80 FR 64512 (table 1), 64513 (October 23, 2015). The EPA made those determinations based on the costs of CCS at the time of that rulemaking. In general, those costs were significantly higher than at present, due to recent technology cost declines as well as related policies, including the IRC section 45Q tax credit for CCS, which were not available at that time for purposes of consideration during the development of the NSPS. Id. at 64562 (table 8). Based on of these higher costs, the EPA determined that 16-23 percent capture qualified as the BSER, rather than a significantly higher percentage of capture. Given the substantial differences in the cost of CCS during the time of the 2015 NSPS and the present time, the capture percentage of the 2015 NSPS necessarily differed from the capture percentage in this final action, and, by the same token, the associated degree of emission limitation and resulting standards of performance necessarily differ as well. If the EPA had strong evidence to indicate that new coal-fired EGUs would be built, it would propose to revise the 2015 NSPS to align the BSER and emissions standards to reflect the new information regarding the costs of CCS. Because there is no evidence to suggest that there are any firm plans to build new coal-fired EGUs in the future, however, it is not at present a good use of the EPA's limited resources to propose to update the new source standard to align with the existing source standard finalized today. While the EPA is not revising the new source standard for new coal-fired EGUs in this action, the EPA is retaining the ability to propose review in the future.

vii. Requirement That Source Must Transfer CO₂ to an Entity That Reports Under the Greenhouse Gas Reporting Program

The final rule requires that EGUs that capture CO₂ in order to meet the applicable emission standard report in accordance with the GHGRP requirements of 40 CFR part 98, including subpart PP. GHGRP subpart RR and subpart VV requirements provide the monitoring and reporting mechanisms to quantify CO₂ storage and to identify, quantify, and address potential leakage. Under existing GHGRP regulations, sequestration wells permitted as Class VI under the UIC program are required to report under subpart RR. Facilities with UIC Class II wells that inject CO₂ to enhance the recovery of oil or natural gas can opt-in to reporting under subpart RR by submitting and receiving approval for a monitoring, reporting, and verification (MRV) plan. Subpart VV applies to facilities that conduct enhanced recovery using ISO 27916 to quantify geologic storage unless they have opted to report under subpart RR. For this rule, if injection occurs on site, the EGU must report data accordingly under 40 CFR part 98 subpart RR or subpart VV. If the CO₂ is injected off site, the EGU must transfer the captured CO₂ to a facility that reports in accordance with the requirements of 40 CFR part 98, subpart RR or subpart VV. They may also transfer the captured CO₂ to a facility that has received an innovative technology waiver from the EPA.

b. Options Not Determined To Be the BSER for Long-Term Coal-Fired Steam Generating Units

In this section, we explain why CCS at 90 percent capture best balances the BSER factors and therefore why the EPA has determined it to be the best of the possible options for the BSER.

i. Partial Capture CCS

Partial capture for CCS was not determined to be BSER because the emission reductions are lower and the costs would, in general, be higher. As discussed in section IV.B of this preamble, individual coal-fired power plants are by far the highest-emitting plants in the nation, and the coal-fired power plant sector is higher-emitting than any other stationary source sector. CCS at 90 percent capture removes very high absolute amounts of emissions. Partial capture CCS would fail to capture large quantities of emissions. With respect to costs, designs for 90 percent capture in general take greater advantage of economies of scale. Eligibility for the IRC section 45Q tax credit for existing EGUs requires design capture rates equivalent to 75 percent of a baseline emission rate by mass. Even assuming partial capture rates meet that definition, lower capture rates would receive fewer returns from the IRC section 45Q tax credit (since these are tied to the amount of carbon sequestered, and all else being equal lower capture rates would result in lower amounts of sequestered carbon) and costs would thereby be higher.

ii. Natural Gas Co-Firing

(A) Reasons Why Not Selected as BSER

As discussed in section VII.C.2, the EPA is determining 40 percent natural gas co-firing to qualify as the BSER for the medium-term subcategory of coal-fired steam generating units. This subcategory consists of units that will permanently cease operation by January 1, 2039. In making this BSER determination, the EPA analyzed the ability of all existing coal-fired units—not only medium-term units—to install and operate 40 percent co-firing. As a result, all of the determinations concerning the criteria for BSER that the EPA made for 40 percent co-firing apply to all existing coal-fired units, including the units in the long-term subcategory. For example, 40 percent co-firing is adequately demonstrated for the long-term subcategory, and has reasonable energy requirements and reasonable non-air quality environmental impacts. It would also be of reasonable cost for the long-term subcategory. Although the capital expenditure for natural gas co-firing is lower than CCS, the variable costs are higher. As a result, the total costs of natural gas co-firing, in general, are higher on a $/ton basis and not substantially lower on a $/MWh basis, than for CCS. Were co-firing the BSER for long-term units, the cost that industry would bear might then be considered similar to the cost for CCS. In addition, the GHG Mitigation Measures TSD shows that all coal-fired units would be able to achieve the requisite infrastructure build-out and obtain sufficient quantities of natural gas to comply with standards of performance based on 40 percent co-firing by January 1, 2030.

The EPA is not selecting 40 percent natural gas co-firing as the BSER for the long-term subcategory, however, because it requires substantially less emission reductions at the unit-level than 90 percent capture CCS. Natural gas co-firing at 40 percent of the heat input to the steam generating unit achieves 16 percent reductions in emission rate at the stack, while CCS achieves an 88.4 percent reduction in emission rate. As discussed in section IV.B of this preamble, individual coal-fired power plants are by far the highest-emitting plants in the nation, and the coal-fired power plant sector is higher-emitting than any other stationary source sector. Because the unit-level emission reductions achievable by CCS are substantially greater, and because CCS is of reasonable cost and matches up well against the other BSER criteria, the EPA did not determine natural gas co-firing to be BSER for the long-term subcategory although, under other circumstances, it could be. Determining BSER requires the EPA to select the “best” of the systems of emission reduction that are adequately demonstrated, as described in section V.C.2; in this case, there are two systems of emission reduction that match up well against the BSER criteria, but based on weighing the criteria together, and in light of the substantially greater unit-level emission reductions from CCS, the EPA has determined that CCS is a better system of emission reduction than co-firing for the long-term subcategory.

The EPA notes that if a state demonstrates that a long-term coal-fired steam generating unit cannot install and operate CCS and cannot otherwise reasonably achieve the degree of emission limitation that the EPA has determined based on CCS, following the process the EPA has specified in its applicable regulations for consideration of RULOF, the state would evaluate natural gas co-firing as a potential basis for establishing a less stringent standard of performance, as detailed in section X.C.2 of this document.

iii. Heat Rate Improvements

Heat rate improvements were not considered to be BSER for long-term steam generating units because the achievable reductions are very low and may result in a rebound effect whereby total emissions from the source increase, as detailed in section VII.D.4.a of this preamble.

Comment: One commenter requested that HRI be considered as BSER in addition to CCS, so that long-term sources would be required to achieve reductions in emission rate consistent with performing HRI and adding CCS with 90 percent capture to the source.

Response: As described in section VII.D.4.a, the reductions from HRI are very low and many sources have already made HRI, so that additional reductions are not available. It is possible that a source installing CO₂ capture will make efficiency improvements as a matter of best practices. For example, Boundary Dam Unit 3 made upgrades to the existing steam generating unit when CCS was installed, including installing a new steam turbine. However, the reductions from efficiency improvements would not be additive to the reductions from CCS because of the impact of the CO₂ capture plant on the efficiency of source due to the required steam and electricity load of the capture plant.

c. Conclusion

Coal-fired EGUs remain the largest stationary source of dangerous CO₂ emissions. The EPA is finalizing CCS at a capture rate of 90 percent as the BSER for long-term coal-fired steam generating units because this system satisfies the criteria for BSER as summarized here. CCS at a capture rate of 90 percent as the BSER for long-term coal-fired steam generating units is adequately demonstrated, as indicated by the facts that it has been operated at scale, is widely applicable to these sources, and that there are vast sequestration opportunities across the continental U.S. Additionally, accounting for recent technology cost declines as well as policies including the tax credit under IRC section 45Q, the costs for CCS are reasonable. Moreover, any adverse non-air quality health and environmental impacts and energy requirements of CCS, including impacts on the power sector on a nationwide basis, are limited and can be effectively avoided or mitigated. In contrast, co-firing 40 percent natural gas would achieve far fewer emission reductions without improving the cost reasonableness of the control strategy.

These considerations provide the basis for finalizing CCS as the best of the systems of emission reduction for long-term coal-fired power plants. In addition, determining CCS as the BSER promotes advancements in control technology for CO₂, which is a relevant consideration when establishing BSER under section 111 of the CAA.

i. Adequately Demonstrated

CCS with 90 percent capture is adequately demonstrated based on the information in section VII.C.1.a.i of this preamble. Solvent-based CO₂ capture was patented nearly 100 years ago in the 1930s and has been used in a variety of industrial applications for decades. Thousands of miles of CO₂ pipelines have been constructed and securely operated in the U.S. for decades. And tens of millions of tons of CO₂ have been permanently stored deep underground either for geologic sequestration or in association with EOR. There are currently at least 15 operating CCS projects in the U.S., and another 121 that are under construction or in advanced stages of development. This broad application of CCS demonstrates the successful operation of all three components of CCS, operating both independently and simultaneously. Various CO₂ capture methods are used in industrial applications and are tailored to the flue gas conditions of a particular industry (see the final TSD, GHG Mitigation Measures for Steam Generating Units for details). Of those capture technologies, amine solvent-based capture has been demonstrated for removal of CO₂ from the post-combustion flue gas of fossil fuel-fired EGUs.

Since 1978, an amine-based system has been used to capture approximately 270,000 metric tons of CO₂ per year from the flue gas of the bituminous coal-fired steam generating units at the 63 MW Argus Cogeneration Plant (Trona, California). Amine solvent capture has been further demonstrated at coal-fired power plants including AES's Warrior Run and Shady Point. And since 2014, CCS has been applied at the commercial scale at Boundary Dam Unit 3, a 110 MW lignite coal-fired steam generating unit in Saskatchewan, Canada.

Impending increases in Canadian regulatory CO₂ emission requirements have prompted optimization of Boundary Dam Unit 3 so that the facility now captures 83 percent of its total CO₂ emissions. Moreover, from the flue gas treated, Boundary Dam Unit 3 consistently captured 90 percent or more of the CO₂ over a 3-year period. The adequate demonstration of CCS is further corroborated by the EPAct05-assisted 240MW-equivalent Petra Nova CCS project at the coal-fired W.A. Parish Unit 8, which achieved over 90 percent capture from the treated flue gas during a 3-year period. Additionally, the technical improvements put in practice at Boundary Dam Unit 3 and Petra Nova can be put in place on new capture facilities during initial construction. This includes redundancies and isolations for key equipment, and spray systems to limit fly ash carryover. Projects that have announced plans to install CO₂ capture directly include these improvements in their design and employ new solvents achieving higher capture rates that are commercially available from technology providers. As a result, these projects target capture efficiencies of at least 95 percent, well above the BSER finalized here.

Precedent, building upon the statutory text and context, has established that the EPA may make a finding of adequate demonstration by drawing upon existing data from individual commercial-scale sources, including testing at these sources, and that the agency may make projections based on existing data to establish a more stringent standard than has been regularly shown, in particular in cases when the agency can specifically identify technological improvements that can be expected to achieve the standard in question. Further, the EPA may extrapolate based on testing at a particular kind of source to conclude that the technology at issue will also be effective at a different, related, source. Following this legal standard, the available data regarding performance and testing at Boundary Dam, a commercial-scale plant, is enough, by itself, to support the EPA's adequate demonstration finding for a 90 percent standard. In addition to this, however, in the 9 years since Boundary Dam began operating, operators and the EPA have developed a clear understanding of specific technological improvements which, if implemented, the EPA can reasonably expect to lead to a 90 percent capture rate on a regular and ongoing basis. The D.C. Circuit has established that this information is more than enough to establish that a 90 percent standard is achievable. And per Lignite Energy Council, the findings from Boundary Dam can be extrapolated to other, similarly operating power plants, including natural gas plants.

Transport of CO₂ and geological storage of CO₂ have also been adequately demonstrated, as detailed in VII.C.1.a.i(B)(7) and VII.C.1.a.i(D)(2). CO₂ has been transported through pipelines for over 60 years, and in the past 20 years, 500 million metric tons of CO₂ moved through over 5,000 miles of CO₂ pipelines. CO₂ pipeline controls and PHMSA standards ensure that captured CO₂ will be securely conveyed to a sequestration site. Due to the proximity of sources to storage, it would be feasible for most sources to build smaller and shorter source-to-sink laterals, rather than rely on a trunkline network buildout. In addition to pipelines, CO₂ can also be transported via vessel, highway, or rail. Geological storage is proven and broadly available, and of the coal-fired steam generating units with planned operation during or after 2030, 77 percent are within 40 miles of the boundary of a saline reservoir.

The EPA also considered the timelines, materials, and workforce necessary for installing CCS, and determined they are sufficient.

ii. Cost

Process improvements have resulted in a decrease in the projected costs to install CCS on existing coal-fired steam generating units. Additionally, the IRC section 45Q tax credit provides $85 per metric ton ($77 per ton) of CO₂. It is reasonable to account for the IRC section 45Q tax credit because the costs that should be accounted for are the costs to the source. For the fleet of coal-fired steam generating units with planned operation during or after 2033, and assuming a 12-year amortization period and 80 percent annual capacity factor and including source specific transport and storage costs, the average total costs of CCS are −$5/ton of CO₂ reduced and −$4/MWh. And even for shorter amortization periods, the $/MWh costs are comparable to or less than the costs for other controls ($10.60-$18.50/MWh) for a substantial number of sources. Notably, the EPA's IPM model projects that even without this final rule—that is, in the base case, without any CAA section 111 requirements—some units would deploy CCS. Similarly, the IPM model projects that even if this rule determined 40 percent co-firing to be the BSER for long-term coal, instead of CCS, some additional units would deploy CCS. Therefore, the costs of CCS with 90 percent capture are reasonable.

iii. Non-Air Quality Health and Environmental Impacts and Energy Requirements

The CO₂ capture plant requires substantial pre-treatment of the flue gas to remove SO₂ and fly ash (PM) while other controls and process designs are necessary to minimize solvent degradation and solvent loss. Although CCS has the potential to result in some increases in non-GHG emissions, a robust regulatory framework, generally implemented at the state level, is in place to mitigate other non-GHG emissions from the CO₂ capture plant. For transport, pipeline safety is regulated by PHMSA, while UIC Class VI regulations under the SDWA, in tandem with GHGRP subpart RR requirements, ensure the protection of USDWs and the security of geologic sequestration. Therefore, the potential non-air quality health and environmental impacts do not militate against designating CCS as the BSER for long-term steam EGUs. The EPA also considered energy requirements. While the CO₂ capture plant requires steam and electricity to operate, the incentives provided by the IRC section 45Q tax credit will likely result in increased total generation from the source. Therefore, the energy requirements are not unreasonable, and there would be limited, non-adverse impacts on the broader energy sector.

2. Medium-Term Coal-Fired Steam Generating Units

The EPA is finalizing its conclusion that 40 percent natural gas co-firing on a heat input basis is the BSER for medium-term coal-fired steam generating units. Co-firing 40 percent natural gas, on an annual average heat input basis, results in a 16 percent reduction in CO₂ emission rate. The technology has been adequately demonstrated, can be implemented at reasonable cost, does not have significant adverse non-air quality health and environmental impacts or energy requirements, including impacts on the energy sector, and achieves meaningful reductions in CO₂ emissions. Co-firing also advances useful control technology, which provides additional, although not essential, support for treating it as the BSER.

a. Rationale for the Medium-Term Coal-Fired Steam Generating Unit Subcategory

For the development of the emission guidelines, the EPA first considered CCS as the BSER for existing coal-fired steam generating units. CCS generally achieves significant emission reductions at reasonable cost. Typically, in setting the BSER, the EPA assumes that regulated units will continue to operate indefinitely. However, that assumption is not appropriate for all coal-fired steam generating units. 62 percent of existing coal-fired steam generating units greater than 25 MW have already announced that they will retire or convert from coal to gas by 2039. CCS is capital cost-intensive, entailing a certain period to amortize the capital costs. Therefore, the EPA evaluated the costs of CCS for different amortization periods, as detailed in section VII.C.1.a.ii of the preamble, and determined that CCS was cost reasonable, on average, for sources operating more than 7 years after the compliance date of January 1, 2032. Accordingly, units that cease operating before January 1, 2039, will generally have less time to amortize the capital costs, and the costs for those sources would be higher and thereby less comparable to those the EPA has previously determined to be reasonable. Considering this, and the other factors evaluated in determining BSER, the EPA is not finalizing CCS as BSER for units demonstrating that they plan to permanently cease operation prior to January 1, 2039.

Instead, the EPA is subcategorizing these units into the medium-term subcategory and finalizing a BSER based on 40 percent natural gas co-firing on a heat input basis for these units. Co-firing natural gas at 40 percent has significantly lower capital costs than CCS and can be implemented by January 1, 2030. For sources that expect to continue in operation until January 1, 2039, and that therefore have a 9-year amortization period, the costs of 40 percent co-firing are $73/ton of CO₂ reduced or $13/MWh of generation, which supports their reasonableness because they are comparable to or less than the costs detailed in section VII.C.1.a.ii(D) of this preamble for other controls on EGUs ($10.60 to $18.50/MWh) and for GHGs for the Crude Oil and Natural Gas source category in the 2016 NSPS of $98/ton of CO_2e reduced (80 FR 56627; September 18, 2015). Co-firing is also cost-reasonable for sources permanently ceasing operations sooner, and that therefore have a shorter amortization period. As discussed in section VII.B.2 of this preamble, with a two-year amortization period, many units can co-fire with meaningful amounts of natural gas at reasonable cost. Of course, even more can co-fire at reasonable costs with amortization periods longer than two years. For example, the EPA has determined that 33 percent of sources with an amortization period of at least three years have costs for 40 percent co-firing below both of the $/ton and $/MWh metrics, and 68 percent of those sources have costs for 20 percent co-firing below both of those metrics. Therefore, recognizing that operating horizon affects the cost reasonableness of controls, the EPA is finalizing a separate subcategory for coal-fired steam generating units operating in the medium-term—those demonstrating that they plan to permanently cease operation after December 31, 2031, and before January 1, 2039—with 40 percent natural gas co-firing as the BSER.

i. Legal Basis for Establishing the Medium-Term Subcategory

As noted in section V.C.1 of this preamble, the EPA has broad authority under CAA section 111(d) to identify subcategories. As also noted in section V.C.1, the EPA's authority to “distinguish among classes, types, and sizes within categories,” as provided under CAA section 111(b)(2) and as we interpret CAA section 111(d) to provide as well, generally allows the Agency to place types of sources into subcategories when they have characteristics that are relevant to the controls that the EPA may determine to be the BSER for those sources. One element of the BSER is cost reasonableness. See CAA section 111(d)(1) (requiring the EPA, in setting the BSER, to “tak[e] into account the cost of achieving such reduction”). As noted in section V, the EPA's longstanding regulations under CAA section 111(d) explicitly recognize that subcategorizing may be appropriate for sources based on the “costs of control.” Subcategorizing on the basis of operating horizon is consistent with a key characteristic of the coal-fired power industry that is relevant for determining the cost reasonableness of control requirements: A large percentage of the sources in the industry have already announced, and more are expected to announce, dates for ceasing operation, and the fact that many coal-fired steam generating units intend to cease operation in the near term affects what controls are “best” for different subcategories. At the outset, installation of emission control technology takes time, sometimes several years. Whether the costs of control are reasonable depends in part on the period of time over which the affected sources can amortize those costs. Sources that have shorter operating horizons will have less time to amortize capital costs. Thus, the annualized cost of controls may thereby be less comparable to the costs the EPA has previously determined to be reasonable.

In addition, subcategorizing by length of period of continued operation is similar to two other bases for subcategorization on which the EPA has relied in prior rules, each of which implicates the cost reasonableness of controls: The first is load level, noted in section V.C.1. of this preamble. For example, in the 2015 NSPS, the EPA divided new natural gas-fired combustion turbines into the subcategories of base load and non-base load. 80 FR 64602 (table 15) (October 23, 2015). The EPA did so because the control technologies that were “best”—including consideration of feasibility and cost reasonableness—depended on how much the unit operated. The load level, which relates to the amount of product produced on a yearly or other basis, bears similarity to a limit on a period of continued operation, which concerns the amount of time remaining to produce the product. In both cases, certain technologies may not be cost-reasonable because of the capacity to produce product— i.e., the costs are spread over less product produced. Subcategorization on this basis is also supported by how utilities manage their assets over the long term, and was widely supported by industry commenters.

The second basis for subcategorization on which EPA has previously relied is fuel type, as also noted in section V.C.1 of this preamble. The 2015 NSPS provides an example of this type of subcategorization as well. There, the EPA divided new combustion turbines into subcategories on the basis of type of fuel combusted. Id. Subcategorizing on the basis of the type of fuel combusted may be appropriate when different controls have different costs, depending on the type of fuel, so that the cost reasonableness of the control depends on the type of fuel. In that way, it is similar to subcategorizing by operating horizon because in both cases, the subcategory is based upon the cost reasonableness of controls. Subcategorizing by operating horizon is also tantamount to the length of time over which the source will continue to combust the fuel. Subcategorizing on this basis may be appropriate when different controls for a particular fuel have different costs, depending on the length of time when the fuel will continue to be combusted, so that the cost reasonableness of controls depends on that timeframe. Some prior EPA rules for coal-fired sources have made explicit the link between length of time for continued operation and type of fuel combusted by codifying federally enforceable retirement dates as the dates by which the source must “cease burning coal.”

As noted above, creating a subcategory on the basis of operating horizon does not preclude a state from considering RULOF in applying a standard of performance to a particular source. The EPA's authority to set BSER for a source category (including subcategories) and a state's authority to invoke RULOF for individual sources within a category or subcategory are distinct. The EPA's statutory obligation is to determine a generally applicable BSER for a source category, and where that source category encompasses different classes, types, or sizes of sources, to set generally applicable BSERs for subcategories accounting for those differences. By contrast, states' authority to invoke RULOF is premised on the state's ability to take into account information relevant to individual units that is fundamentally different than the information the EPA took into account in determining BSER generally. As noted, the EPA may subcategorize on the basis of cost of controls, and operating horizon may factor into the cost of controls. Moreover, through section 111(d)(1), Congress also required the EPA to develop regulations that permit states to consider “among other factors, the remaining useful life” of a particular existing source. The EPA has interpreted these other factors to include costs or technical feasibility specific to a particular source, even though these are factors the EPA itself considers in setting the BSER. In other words, the factors the EPA may consider in setting the BSER and the factors the states may consider in applying standards of performance are not distinct. As noted above, the EPA is finalizing these subcategories in response to requests by power sector representatives that this rule accommodate the fact that there is a class of sources that plan to voluntarily cease operations in the near term. Although the EPA has designed the subcategories to accommodate those requests, a particular source may still present source-specific considerations—whether related to its remaining useful life or other factors—that the state may consider relevant for the application of that particular source's standard of performance, and that the state should address as described in section X.C.2 of this preamble.

ii. Comments Received on Existing Coal-Fired Subcategories

Comment: The EPA received several comments on the proposed subcategories for coal-fired steam generating units. Many commenters, including industry commenters, supported these subcategories. Some commenters opposed these proposed subcategories. They argued that the subcategories were designed to force coal-fired power plants to retire.

Response: We disagree with comments suggesting that the subcategories for existing coal-fired steam EGUs that the EPA has finalized in this rule were designed to force retirements. The subcategories were not designed for that purpose, and the commenters do not explain their allegations to the contrary. The subcategories were designed, at industry's request, to ensure that subcategories of units that can feasibly and cost-reasonably employ emissions reduction technologies—and only those subcategories of units that can do so—are required to reduce their emissions commensurate with those technologies. As explained above, in determining the BSER, the EPA generally assumes that a source will operate indefinitely, and calculates expected control costs on that basis. Under that assumption, the BSER for existing fossil-fuel fired EGUs is CCS. Nevertheless, the EPA recognizes that many fossil-fuel fired EGUs have already announced plans to cease operation. In recognition of this unique, distinguishing factor, the EPA determined whether a different BSER would be appropriate for fossil fuel-fired EGUs that do not intend to operate over the long term, and concluded, for the reasons stated above, that natural gas co-firing was appropriate for these sources that intended to cease operation before 2039. This subcategory is not intended to force retirements, and the EPA is not directing any state or any unit as to the choice of when to cease operation. Rather, the EPA has created this subcategory to accommodate these sources' intended operation plans. In fact, a number of industry commenters specifically requested and supported subcategories based on retirement dates in recognition of the reality that many operators are choosing to retire these units and that whether or not a control technology is feasible and cost-reasonable depends upon how long a unit intends to operate.

Specifically, as noted in section VII.B of this preamble, in this final action, the medium-term subcategory includes a date for permanently ceasing operation, which applies to coal-fired plants demonstrating that they plan to permanently cease operating after December 31, 2031, and before January 1, 2039. The EPA is retaining this subcategory because 55 percent of existing coal-fired steam generating units greater than 25 MW have already announced that they will retire or convert from coal to gas by January 1, 2039. Accordingly, the costs of CCS—the high capital costs of which require a lengthy amortization period from its January 1, 2032, implementation date—are higher than the traditional metric for cost reasonableness for these sources. As discussed in section VII.C.2 of this preamble, the BSER for these sources is co-firing 40 percent natural gas. This is because co-firing, which has an implementation date of January 1, 2030, has lower capital costs and is therefore cost-reasonable for sources continuing to operate on or after January 1, 2032. It is further noted that this subcategory is elective. Furthermore, states also have the authority to establish a less stringent standard through RULOF in the state plan process, as detailed in section X.C.2 of this preamble.

In sum, these emission guidelines do not require any coal-fired steam EGU to retire, nor are they intended to induce retirements. Rather, these emission guidelines simply set forth presumptive standards that are cost-reasonable and achievable for each subcategory of existing coal-fired steam EGUs. See section VII.E.1 of this preamble (responding to comments that this rule violates the major questions doctrine).

Comment: The EPA broadly solicited comment on the dates and values defining the proposed subcategories for coal-fired steam generating units. Regarding the proposed dates for the subcategories, one industry stakeholder commented that the “EPA's proposed retirement dates for applicability of the various subcategories are appropriate and broadly consistent with system reliability needs.” More specifically, industry commenters requested that the cease-operation-by date for the imminent-term subcategory be changed from January 1, 2032, to January 1, 2033. Industry commenters also stated that the 20 percent utilization limit in the definition of the near-term subcategory was overly restrictive and inconsistent with the emissions stringency of either the proposed medium term or imminent term subcategory—commenters requested greater flexibility for the near-term subcategory. Other comments from NGOs and other groups suggested various other changes to the subcategory definitions. One commenter requested moving the cease-operation-by date for the medium-term subcategory up to January 1, 2038, while eliminating the imminent-term subcategory and extending the near-term subcategory to January 1, 2038.

Response: The EPA is not finalizing the proposed imminent-term or near-term subcategories. The EPA is finalizing an applicability exemption for sources demonstrating that they plan to permanently cease operation prior to January 1, 2032, as detailed in section VII.B of this preamble. The EPA is finalizing the cease operating by date of January 1, 2039, for medium-term coal-fired steam generating units. These dates are all based on costs of co-firing and CCS, driven by their amortization periods, as discussed in the preceding sections of this preamble.

b. Rationale for Natural Gas Co-Firing as the BSER for Medium-Term Coal-Fired Steam Generating Units

In this section of the preamble, the EPA describes its rationale for natural gas co-firing as the final BSER for medium-term coal-fired steam generating units.

For a coal-fired steam generating unit, the substitution of natural gas for some of the coal, so that the unit fires a combination of coal and natural gas, is known as “natural gas co-firing.” The EPA is finalizing natural gas co-firing at a level of 40 percent of annual heat input as BSER for medium-term coal-fired steam generating units.

i. Adequately Demonstrated

The EPA is finalizing its determination that natural gas co-firing at the level of 40 percent of annual heat input is adequately demonstrated for coal-fired steam generating units. Many existing coal-fired steam generating units already use some amount of natural gas, and several have co-fired at relatively high levels at or above 40 percent of heat input in recent years.

(A) Boiler Modifications

Existing coal-fired steam generating units can be modified to co-fire natural gas in any desired proportion with coal, up to 100 percent natural gas. Generally, the modification of existing boilers to enable or increase natural gas firing typically involves the installation of new gas burners and related boiler modifications, including, for example, new fuel supply lines and modifications to existing air ducts. The introduction of natural gas as a fuel can reduce boiler efficiency slightly, due in large part to the relatively high hydrogen content of natural gas. However, since the reduction in coal can result in reduced auxiliary power demand, the overall impact on net heat rate can range from a 2 percent increase to a 2 percent decrease.

It is common practice for steam generating units to have the capability to burn multiple fuels onsite, and of the 565 coal-fired steam generating units operating at the end of 2021, 249 of them reported consuming natural gas as a fuel or startup source. Coal-fired steam generating units often use natural gas or oil as a startup fuel, to warm the units up before running them at full capacity with coal. While startup fuels are generally used at low levels (up to roughly 1 percent of capacity on an annual average basis), some coal-fired steam generating units have co-fired natural gas at considerably higher shares. Based on hourly reported CO₂ emission rates from the start of 2015 through the end of 2020, 29 coal-fired steam generating units co-fired with natural gas at rates at or above 60 percent of capacity on an hourly basis. The capability of those units on an hourly basis is indicative of the extent of boiler burner modifications and sizing and capacity of natural gas pipelines to those units, and implies that those units are technically capable of co-firing at least 60 percent natural gas on a heat input basis on average over the course of an extended period ( e.g., a year). Additionally, during that same 2015 through 2020 period, 29 coal-fired steam generating units co-fired natural gas at over 40 percent on an annual heat input basis. Because of the number of units that have demonstrated co-firing above 40 percent of heat input, the EPA is finalizing that co-firing at 40 percent is adequately demonstrated. A more detailed discussion of the record of natural gas co-firing, including current trends, at coal-fired steam generating units is included in the final TSD, GHG Mitigation Measures for Steam Generating Units.

(B) Natural Gas Pipeline Development

In addition to any potential boiler modifications, the supply of natural gas is necessary to enable co-firing at existing coal-fired steam boilers. As discussed in the previous section, many plants already have at least some access to natural gas. In order to increase natural gas access beyond current levels, plants may find it necessary to construct natural gas supply pipelines.

The U.S. natural gas pipeline network consists of approximately 3 million miles of pipelines that connect natural gas production with consumers of natural gas. To increase natural gas consumption at a coal-fired boiler without sufficient existing natural gas access, it is necessary to connect the facility to the natural gas pipeline transmission network via the construction of a lateral pipeline. The cost of doing so is a function of the total necessary pipeline capacity (which is characterized by the length, size, and number of laterals) and the location of the plant relative to the existing pipeline transmission network. The EPA estimated the costs associated with developing new lateral pipeline capacity sufficient to meet 60 percent of the net summer capacity at each coal-fired steam generating unit that could be included in this subcategory. As discussed in the final TSD, GHG Mitigation Measures for Steam Generating Units, the EPA estimates that this lateral capacity would be sufficient to enable each unit to achieve 40 percent natural gas co-firing on an annual average basis.

The EPA considered the availability of the upstream natural gas pipeline capacity to satisfy the assumed co-firing demand implied by these new laterals. This analysis included pipeline development at all EGUs that could be included in this subcategory, including those without announced plans to cease operating before January 1, 2039. The EPA's assessment reviewed the reasonableness of each assumed new lateral by determining whether the peak gas capacity of that lateral could be satisfied without modification of the transmission pipeline systems to which it is assumed to be connected. This analysis found that most, if not all, existing pipeline systems are currently able to meet the peak needs implied by these new laterals in aggregate, assuming that each existing coal-fired unit in the analysis co-fired with natural gas at a level implied by these new laterals, or 60 percent of net summer generating capacity. While this is a reasonable assumption for the analysis to support this mitigation measure in the BSER context, it is also a conservative assumption that overstates the amount of natural gas co-firing expected under the final rule.

Most of these individual laterals are less than 15 miles in length. The maximum aggregate amount of pipeline capacity, if all coal-fired steam capacity that could be included in the medium-term subcategory ( i.e., all capacity that has not announced that it plans to retire by 2032) implemented the final BSER by co-firing 40 percent natural gas, would be comparable to pipeline capacity constructed recently. The EPA estimates that this maximum total capacity would be nearly 14.7 billion cubic feet per day, which would require about 3,500 miles of pipeline costing roughly $11.5 billion. Over 2 years, this maximum total incremental pipeline capacity would amount to less than 1,800 miles per year, with a total annual capacity of roughly 7.35 billion cubic feet per day. This represents an estimated annual investment of approximately $5.75 billion per year in capital expenditures, on average. By comparison, based on data collected by EIA, the total annual mileage of natural gas pipelines constructed over the 2017-2021 period ranged from approximately 1,000 to 2,500 miles per year, with a total annual capacity of 10 to 25 billion cubic feet per day. This represents an estimated annual investment of up to nearly $15 billion. The upper end of these historical annual values is much higher than the maximum annual values that could be expected under this final BSER measure—which, as noted above, represent a conservative estimate that significantly overstates the amount of co-firing that the EPA projects would occur under this final rule.

These conservatively high estimates of pipeline requirements also compare favorably to industry projections of future pipeline capacity additions. Based on a review of a 2018 industry report, titled “North America Midstream Infrastructure through 2035: Significant Development Continues,” investment in midstream infrastructure development is expected to range between $10 to $20 billion per year through 2035. Approximately $5 to $10 billion annually is expected to be invested in natural gas pipelines through 2035. This report also projects that an average of over 1,400 miles of new natural gas pipeline will be built through 2035, which is similar to the approximately 1,670 miles that were built on average from 2013 to 2017. These values are consistent with the average annual expenditure of $5.75 billion on less than 1,800 miles per year of new pipeline construction that would be necessary for the entire operational fleet of existing coal-fired steam generating units to co-fire with natural gas. The actual pipeline investment for this subcategory would be substantially lower.

(C) Compliance Date for Medium-Term Coal-Fired Steam Generating Units

The EPA is finalizing a compliance date for medium-term coal-fired steam generating units of January 1, 2030.

As in the timeline for CCS for the long term coal-fired steam generating units described in section VII.C.1.a.i(E), the EPA assumes here that feasibility work occurs during the state plan development period, and that all subsequent work occurs after the state plan is submitted and thereby effective at the state level. The EPA assumes 12 months of feasibility work for the natural gas pipeline lateral and 6 months of feasibility work for boiler modifications (both to occur over June 2024 to June 2025). As with the feasibility analysis for CCS, the feasibility analysis for co-firing will inform the state plan and therefore it is reasonable to assume units will perform it during the state planning window. Feasibility for the pipeline includes a right-of-way and routing analysis. Feasibility for the boiler modifications includes conceptual studies and design basis.

The timeline for the natural gas pipeline permitting and construction is based on a review of recently completed permitting approvals and construction. The average time to complete permitting and approval is less than 1.5 years, and the average time to complete actual construction is less than 1 year. Of the 31 reviewed pipeline projects, the vast majority (27 projects) took less than a total of 3 years for permitting and construction, and none took more than 3.5 years. Therefore, it is reasonable to assume that permitting and construction would take no more than 3 years for most sources (June 2026 to June 2029), noting that permitting and construction for many sources would be faster.

The timeline for boiler modifications based on the baseline duration co-firing conversion project schedule developed by Sargent and Lundy. The EPA assumes that, with the exception of the feasibility studies discussed above, work on the boiler modifications begins after the state plan submission due date. The EPA also assumes permitting for the boiler modifications is required and takes 12 months (June 2026 to June 2027). In the schedule developed by Sargent and Lundy, commercial arrangements for the boiler modification take about 6 months (June 2026 to December 2026). Detailed engineering and procurement takes about 7 months (December 2026 to July 2027), and begins after commercial arrangements are complete. Site work takes 3 months (July 2027 to October 2027), followed by 4 months of construction (October 2027 to February 2028). Lastly, startup and testing takes about 2 months (June 2029 to August 2029), noting that the EPA assumes this occurs after the natural gas pipeline lateral is constructed. Considering the preceding information, the EPA has determined January 1, 2030 is the compliance date for medium-term coal-fired steam generating units.

ii. Costs

The capital costs associated with the addition of new gas burners and other necessary boiler modifications depend on the extent to which the current boiler is already able to co-fire with some natural gas and on the amount of gas co-firing desired. The EPA estimates that, on average, the total capital cost associated with modifying existing boilers to operate at up to 100 percent of heat input using natural gas is approximately $52/kW. These costs could be higher or lower, depending on the equipment that is already installed and the expected impact on heat rate or steam temperature.

While fixed O&M (FOM) costs can potentially decrease as a result of decreasing the amount of coal consumed, it is common for plants to maintain operation of one coal pulverizer at all times, which is necessary for maintaining several coal burners in continuous service. In this case, coal handling equipment would be required to operate continuously and therefore natural gas co-firing would have limited effect on reducing the coal-related FOM costs. Although, as noted, coal-related FOM costs have the potential to decrease, the EPA does not anticipate a significant increase in impact on FOM costs related to co-firing with natural gas.

In addition to capital and FOM cost impacts, any additional natural gas co-firing would result in incremental costs related to the differential in fuel cost, taking into consideration the difference in delivered coal and gas prices, as well as any potential impact on the overall net heat rate. The EPA's reference case projects that in 2030, the average delivered price of coal will be $1.56/MMBtu and the average delivered price of natural gas will be $2.95/MMBtu. Thus, assuming the same level of generation and no impact on heat rate, the additional fuel cost would be $1.39/MMBtu on average in 2030. The total additional fuel cost could increase or decrease depending on the potential impact on net heat rate. An increase in net heat rate, for example, would result in more fuel required to produce a given amount of generation and thus additional cost. In the final TSD, GHG Mitigation Measures for Steam Generating Units, the EPA's cost estimates assume a 1 percent average increase in net heat rate.

Finally, for plants without sufficient access to natural gas, it is also necessary to construct new natural gas pipelines (“laterals”). Pipeline costs are typically expressed in terms of dollars per inch of pipeline diameter per mile of pipeline distance ( i.e., dollars per inch-mile), reflecting the fact that costs increase with larger diameters and longer pipelines. On average, the cost for lateral development within the contiguous U.S. is approximately $280,000 per inch-mile (2019$), which can vary based on site-specific factors. The total pipeline cost for each coal-fired steam generating unit is a function of this cost, as well as a function of the necessary pipeline capacity and the location of the plant relative to the existing pipeline transmission network. The pipeline capacity required depends on the amount of co-firing desired as well as on the desired level of generation—a higher degree of co-firing while operating at full load would require more pipeline capacity than a lower degree of co-firing while operating at partial load. It is reasonable to assume that most plant owners would develop sufficient pipeline capacity to deliver the maximum amount of desired gas use in any moment, enabling higher levels of co-firing during periods of lower fuel price differentials. Once the necessary pipeline capacity is determined, the total lateral cost can be estimated by considering the location of each plant relative to the existing natural gas transmission pipelines as well as the available excess capacity of each of those existing pipelines.

The EPA determined the costs of 40 percent co-firing based on the fleet of coal-fired steam generating units that existed in 2021 and that do not have known plans to cease operations or convert to gas by 2032, and assuming that each of those units continues to operate at the same level as it operated over 2017-2021. The EPA assessed those costs against the cost reasonableness metrics, as described in section VII.C.1.a.ii(D) of this preamble ( i.e., emission control costs on EGUs of $10.60 to $18.50/MWh and the costs in the 2016 NSPS regulating GHGs for the Crude Oil and Natural Gas source category of $98/ton of CO_2e reduced (80 FR 56627; September 18, 2015)). On average, the EPA estimates that the weighted average cost of co-firing with 40 percent natural gas as the BSER on an annual average basis is approximately $73/ton CO₂ reduced, or $13/MWh. The costs here reflect an amortization period of 9 years. These estimates support a conclusion that co-firing is cost-reasonable for sources that continue to operate up until the January 1, 2039, threshold date for the subcategory. The EPA also evaluated the fleet average costs of natural gas co-firing for shorter amortization periods and has determined that the costs are consistent with the cost reasonableness metrics for the majority of sources that will operate past January 1, 2032, and therefore have an amortization period of at least 2 years and up to 9 years. These estimates and all underlying assumptions are explained in detail in the final TSD, GHG Mitigation Measures for Steam Generating Units. Based on this cost analysis, alongside the EPA's overall assessment of the costs of this rule, the EPA is finalizing that the costs of natural gas co-firing are reasonable for the medium-term coal-fired steam generating unit subcategory. If a particular source has costs of 40 percent co-firing that are fundamentally different from the cost reasonability metrics, the state may consider this fact under the RULOF provisions, as detailed in section X.C.2 of this preamble. The EPA previously estimated the cost of natural gas co-firing in the Clean Power Plan (CPP). 80 FR 64662 (October 23, 2015). The cost-estimates for co-firing presented in this section are lower than in the CPP, for several reasons. Since then, the expected difference between coal and gas prices has decreased significantly, from over $3/MMBtu to less than $1.50/MMBtu in this final rule. Additionally, a recent analysis performed by Sargent and Lundy for the EPA supports a considerably lower capital cost for modifying existing boilers to co-fire with natural gas. The EPA also recently conducted a highly detailed facility-level analysis of natural gas pipeline costs, the median value of which is slightly lower than the value used by the EPA previously to approximate the cost of co-firing at a representative unit.

iii. Non-Air Quality Health and Environmental Impact and Energy Requirements

Natural gas co-firing for steam generating units is not expected to have any significant adverse consequences related to non-air quality health and environmental impacts or energy requirements.

(A) Non-GHG Emissions

Non-GHG emissions are reduced when steam generating units co-fire with natural gas because less coal is combusted. SO₂, PM_2.5, acid gas, mercury and other hazardous air pollutant emissions that result from coal combustion are reduced proportionally to the amount of natural gas consumed, i.e., under this final rule, by 40 percent. Natural gas combustion does produce NO_X emissions, but in lesser amounts than from coal-firing. However, the magnitude of this reduction is dependent on the combustion system modifications that are implemented to facilitate natural gas co-firing.

Sufficient regulations also exist related to natural gas pipelines and transport that assure natural gas can be safely transported with minimal risk of environmental release. PHMSA develops and enforces regulations for the safe, reliable, and environmentally sound operation of the nation's 2.6 million mile pipeline transportation system. Recently, PHMSA finalized a rule that will improve the safety and strengthen the environmental protection of more than 300,000 miles of onshore gas transmission pipelines. PHMSA also recently promulgated a separate rule covering natural gas transmission, as well as a rule that significantly expanded the scope of safety and reporting requirements for more than 400,000 miles of previously unregulated gas gathering lines. FERC is responsible for the regulation of the siting, construction, and/or abandonment of interstate natural gas pipelines, gas storage facilities, and Liquified Natural Gas (LNG) terminals.

(B) Energy Requirements

The introduction of natural gas co-firing will cause steam boilers to be slightly less efficient due to the high hydrogen content of natural gas. Co-firing at levels between 20 percent and 100 percent can be expected to decrease boiler efficiency between 1 percent and 5 percent. However, despite the decrease in boiler efficiency, the overall net output efficiency of a steam generating unit that switches from coal- to natural gas-firing may change only slightly, in either a positive or negative direction. Since co-firing reduces coal consumption, the auxiliary power demand related to coal handling and emissions controls typically decreases as well. While a site-specific analysis would be required to determine the overall net impact of these countervailing factors, generally the effect of co-firing on net unit heat rate can vary within approximately plus or minus 2 percent.

The EPA previously determined in the ACE Rule (84 FR 32545; July 8, 2019) that “co-firing natural gas in coal-fired utility boilers is not the best or most efficient use of natural gas and [. . .] can lead to less efficient operation of utility boilers.” That determination was informed by the more limited supply of natural gas, and the larger amount of coal-fired EGU capacity and generation, in 2019. Since that determination, the expected supply of natural gas has expanded considerably, and the capacity and generation of the existing coal-fired fleet has decreased, reducing the total mass of natural gas that might be required for sources to implement this measure.

Furthermore, regarding the efficient operation of boilers, the ACE determination was based on the observation that “co-firing can negatively impact a unit's heat rate (efficiency) due to the high hydrogen content of natural gas and the resulting production of water as a combustion by-product.” That finding does not consider the fact that the effect of co-firing on net unit heat rate can vary within approximately plus or minus 2 percent, and therefore the net impact on overall utility boiler efficiency for each steam generating unit is uncertain.

For all of these reasons, the EPA is finalizing that natural gas co-firing at medium-term coal-fired steam generating units does not result in any significant adverse consequences related to energy requirements.

Additionally, the EPA considered longer term impacts on the energy sector, and the EPA is finalizing these impacts are reasonable. Designating natural gas co-firing as the BSER for medium-term coal-fired steam generating units would not have significant adverse impacts on the structure of the energy sector. Steam generating units that currently are coal-fired would be able to remain primarily coal-fired. The replacement of some coal with natural gas as fuel in these sources would not have significant adverse effects on the price of natural gas or the price of electricity.

iv. Extent of Reductions in CO₂ Emissions

One of the primary benefits of natural gas co-firing is emission reduction. CO₂ emissions are reduced by approximately 4 percent for every additional 10 percent of co-firing. When moving from 100 percent coal to 60 percent coal and 40 percent natural gas, CO₂ stack emissions are reduced by approximately 16 percent. Non-CO₂ emissions are reduced as well, as noted earlier in this preamble.

v. Technology Advancement

Natural gas co-firing is already well-established and widely used by coal-fired steam boiler generating units. As a result, this final rule is not likely to lead to technological advances or cost reductions in the components of natural gas co-firing, including modifications to boilers and pipeline construction. However, greater use of natural gas co-firing may lead to improvements in the efficiency of conducting natural gas co-firing and operating the associated equipment.

c. Options Not Determined To Be the BSER for Medium-Term Coal-Fired Steam Generating Units

i. CCS

As discussed earlier in this preamble, the compliance date for CCS is January 1, 2032. Accordingly, sources in the medium-term subcategory—which have elected to commit to permanently cease operations prior to 2039—would have less than 7 years to amortize the capital costs of CCS. As a result, for these sources, the overall costs of CCS would exceed the metrics for cost reasonableness that the EPA is using in this rulemaking, which are detailed in section VII.C.1.a.ii(D). For this reason, the EPA is not finalizing CCS as the BSER for the medium-term subcategory.

ii. Heat Rate Improvements

Heat rate improvements were not considered to be BSER for medium-term steam generating units because the achievable reductions are low and may result in rebound effect whereby total emissions from the source increase, as detailed in section VII.D.4.a.

d. Conclusion

The EPA is finalizing that natural gas co-firing at 40 percent of heat input is the BSER for medium-term coal-fired steam generating units because natural gas co-firing is adequately demonstrated, as indicated by the facts that it has been operated at scale and is widely applicable to sources. Additionally, the costs for natural gas co-firing are reasonable. Moreover, natural gas co-firing can be expected to reduce emissions of several other air pollutants in addition to GHGs. Any adverse non-air quality health and environmental impacts and energy requirements of natural gas co-firing are limited. In contrast, CCS, although achieving greater emission reductions, would be of higher cost, in general, for the subcategory of medium-term units, and HRI would achieve few reductions and, in fact, may increase emissions.

3. Degree of Emission Limitation for Final Standards

Under CAA section 111(d), once the EPA determines the BSER, it must determine the “degree of emission limitation” achievable by the application of the BSER. States then determine standards of performance and include them in the state plans, based on the specified degree of emission limitation. Final presumptive standards of performance are detailed in section X.C.1.b of this preamble. There is substantial variation in emission rates among coal-fired steam generating units—the range is, approximately, from 1,700 lb CO₂ /MWh-gross to 2,500 lb CO₂ /MWh-gross—which makes it challenging to determine a single, uniform emission limit. Accordingly, the EPA is finalizing the degrees of emission limitation by a percentage change in emission rate, as follows.

a. Long-Term Coal-Fired Steam Generating Units

As discussed earlier in this preamble, the EPA is finalizing the BSER for long-term coal-fired steam generating units as “full-capture” CCS, defined as 90 percent capture of the CO₂ in the flue gas. The degree of emission limitation achievable by applying this BSER can be determined on a rate basis. A capture rate of 90 percent results in reductions in the emission rate of 88.4 percent on a lb CO₂ /MWh-gross basis, and this reduction in emission rate can be observed over an extended period ( e.g., an annual calendar-year basis). Therefore, the EPA is finalizing that the degree of emission limitation for long-term units is an 88.4 percent reduction in emission rate on a lb CO₂ /MWh-gross basis over an extended period ( e.g., an annual calendar-year basis).

b. Medium-Term Coal-Fired Steam Generating Units

As discussed earlier in this preamble, the BSER for medium-term coal-fired steam generating units is 40 percent natural gas co-firing. The application of 40 percent natural gas co-firing results in reductions in the emission rate of 16 percent. Therefore, the degree of emission limitation for these units is a 16 percent reduction in emission rate on a lb CO₂ /MWh-gross basis over an extended period ( e.g., an annual calendar-year basis).

D. Rationale for the BSER for Natural Gas-Fired And Oil-Fired Steam Generating Units

This section of the preamble describes the rationale for the final BSERs for existing natural gas- and oil-fired steam generating units based on the criteria described in section V.C of this preamble.

1. Subcategorization of Natural Gas- and Oil-Fired Steam Generating Units

The EPA is finalizing subcategories based on load level ( i.e., annual capacity factor), specifically, units that are base load, intermediate load, and low load. The EPA is finalizing routine methods of operation and maintenance as BSER for intermediate and base load units. Applying that BSER would not achieve emission reductions but would prevent increases in emission rates. The EPA is finalizing presumptive standards of performance that differ between intermediate and base load units due to their differences in operation, as detailed in section X.C.1.b.iii of this preamble. The EPA proposed a separate subcategory for non-continental oil-fired steam generating units, which operate differently from continental units; however, the EPA is not finalizing emission guidelines for sources outside of the contiguous U.S., as described in section VII.B. At proposal, the EPA solicited comment on a BSER of “uniform fuels” for low load natural gas- and oil-fired steam generating units, and the EPA is finalizing this approach for those sources.

Natural gas- and oil-fired steam generating units combust natural gas or distillate fuel oil or residual fuel oil in a boiler to produce steam for a turbine that drives a generator to create electricity. In non-continental areas, existing natural gas- and oil-fired steam generating units may provide base load power, but in the continental U.S., most existing units operate in a load-following manner. There are approximately 200 natural gas-fired steam generating units and fewer than 30 oil-fired steam generating units in operation in the continental U.S. Fuel costs and inefficiency relative to other technologies ( e.g., combustion turbines) result in operation at lower annual capacity factors for most units. Based on data reported to EIA and the EPA for the contiguous U.S., for natural gas-fired steam generating units in 2019, the average annual capacity factor was less than 15 percent and 90 percent of units had annual capacity factors less than 35 percent. For oil-fired steam generating units in 2019, no units had annual capacity factors above 8 percent. Additionally, their load-following method of operation results in frequent cycling and a greater proportion of time spent at low hourly capacities, when generation is less efficient. Furthermore, because startup times for most boilers are usually long, natural gas steam generating units may operate in standby mode between periods of peak demand. Operating in standby mode requires combusting fuel to keep the boiler warm, and this further reduces the efficiency of natural gas combustion.

Unlike coal-fired steam generating units, the CO₂ emission rates of oil- and natural gas-fired steam generating units that have similar annual capacity factors do not vary considerably between units. This is partly due to the more uniform qualities ( e.g., carbon content) of the fuel used. However, the emission rates for units that have different annual capacity factors do vary considerably, as detailed in the final TSD, Natural Gas- and Oil-fired Steam Generating Units. Low annual capacity factor units cycle frequently, have a greater proportion of CO₂ emissions that may be attributed to startup, and have a greater proportion of generation at inefficient hourly capacities. Intermediate annual capacity factor units operate more often at higher hourly capacities, where CO₂ emission rates are lower. High annual capacity factor units operate still more at base load conditions, where units are more efficient and CO₂ emission rates are lower.

Based on these performance differences between these load levels, the EPA, in general, proposed subcategories based on dividing natural gas- and oil-fired steam generating units into three groups each—low load, intermediate load, and base load.

The EPA is finalizing subcategories for oil-fired and natural gas-fired steam generating units, based on load levels. The EPA proposed the following load levels: “low” load, defined by annual capacity factors less than 8 percent; “intermediate” load, defined by annual capacity factors greater than or equal to 8 percent and less than 45 percent; and “base” load, defined by annual capacity factors greater than or equal to 45 percent.

The EPA is finalizing January 1, 2030, as the compliance date for natural gas- and oil-fired steam generating units and this date is consistent with the dates in the fuel type definitions.

The EPA received comments that were generally supportive of the proposed subcategory definitions, and the EPA is finalizing the subcategory definitions as proposed.

2. Options Considered for BSER

The EPA has considered various methods for controlling CO₂ emissions from natural gas- and oil-fired steam generating units to determine whether they meet the criteria for BSER. Co-firing natural gas cannot be the BSER for these units because natural gas- and oil-fired steam generating units already fire large proportions of natural gas. Most natural gas-fired steam generating units fire more than 90 percent natural gas on a heat input basis, and any oil-fired steam generating units that would potentially operate above an annual capacity factor of around 15 percent typically combust natural gas as a large proportion of their fuel as well. Nor is CCS a candidate for BSER. The utilization of most gas-fired units, and likely all oil-fired units, is relatively low, and as a result, the amount of CO₂ available to be captured is low. However, the capture equipment would still need to be sized for the nameplate capacity of the unit. Therefore, the capital and operating costs of CCS would be high relative to the amount of CO₂ available to be captured. Additionally, again due to lower utilization, the amount of IRC section 45Q tax credits that owner/operators could claim would be low. Because of the relatively high costs and the relatively low cumulative emission reduction potential for these natural gas- and oil-fired steam generating units, the EPA is not determining CCS as the BSER for them.

The EPA has reviewed other possible controls but is not finalizing any of them as the BSER for natural gas- and oil-fired units either. Co-firing hydrogen in a boiler is technically possible, but there is limited availability of hydrogen now and in the near future and it should be prioritized for more efficient units. Additionally, for natural gas-fired steam generating units, setting a future standard based on hydrogen would likely have limited GHG reduction benefits given the low utilization of natural gas- and oil-fired steam generating units. Lastly, HRI for these types of units would face many of the same issues as for coal-fired steam generating units; in particular, HRI could result in a rebound effect that would increase emissions.

However, the EPA recognizes that natural gas- and oil-fired steam generating units could possibly, over time, operate more, in response to other changes in the power sector. Additionally, some coal-fired steam generating units have converted to 100 percent natural gas-fired, and it is possible that more may do so in the future. The EPA also received several comments from industry stating plans to do so. Moreover, in part because the fleet continues to age, the plants may operate with degrading emission rates. In light of these possibilities, identifying the BSER and degrees of emission limitation for these sources would be useful to provide clarity and prevent backsliding in GHG performance. Therefore, the EPA is finalizing BSER for intermediate and base load natural gas- and oil-fired steam generating units to be routine methods of operation and maintenance, such that the sources could maintain the emission rates (on a lb/MWh-gross basis) currently maintained by the majority of the fleet across discrete ranges of annual capacity factor. The EPA is finalizing this BSER for intermediate load and base load natural gas- and oil-fired steam generating units, regardless of the operating horizon of the unit.

A BSER based on routine methods of operation and maintenance is adequately demonstrated because units already operate with those practices. There are no or negligible additional costs because there is no additional technology that units are required to apply and there is no change in operation or maintenance that units must perform. Similarly, there are no adverse non-air quality health and environmental impacts or adverse impacts on energy requirements. Nor do they have adverse impacts on the energy sector from a nationwide or long-term perspective. The EPA's modeling, which supports this final rule, indicates that by 2040, a number of natural gas-fired steam generating units will have remained in operation since 2030, although at reduced annual capacity factors. There are no CO₂ reductions that may be achieved at the unit level, but applying routine methods of operation and maintenance as the BSER prevents increases in emission rates. Routine methods of operation and maintenance do not advance useful control technology, but this point is not significant enough to offset their benefits.

At proposal, the EPA also took comment on a potential BSER of uniform fuels for low load natural gas- and oil-fired steam generating units. As noted earlier in this preamble, non-coal fossil fuels combusted in utility boilers typically include natural gas, distillate fuel oil ( i.e., fuel oil No. 1 and No. 2), and residual fuel oil ( i.e., fuel oil No. 5 and No. 6). The EPA previously established heat-input based fuel composition as BSER in the 2015 NSPS (termed “clean fuels” in that rulemaking) for new non-base load natural gas- and multi-fuel-fired stationary combustion turbines (80 FR 64615-17; October 23, 2015), and the EPA is similarly finalizing lower-emitting fuels as BSER for new low load combustion turbines as described in section VIII.F of this preamble. For low load natural gas- and oil-fired steam generating units, the high variability in emission rates associated with the variability of load at the lower-load levels limits the benefits of a BSER based on routine maintenance and operation. That is because the high variability in emission rates would make it challenging to determine an emission rate ( i.e., on a lb CO₂ /MWh-gross basis) that could serve as the presumptive standard of performance that would reflect application of a BSER of routine operation and maintenance. On the other hand, for those units, a BSER of “uniform fuels” and an associated presumptive standard of performance based on a heat input basis, as described in section X.C.1.b.iii of this preamble, is reasonable. Therefore, the EPA is finalizing a BSER of uniform fuels for low load natural gas- and oil-fired steam generating units, with presumptive standards depending on fuel type detailed in section X.C.1.b.iii.

3. Degree of Emission Limitation

As discussed above, because the BSER for base load and intermediate load natural gas- and oil-fired steam generating units is routine operation and maintenance, which the units are, by definition, already employing, the degree of emission limitation by application of this BSER is no increase in emission rate on a lb CO₂ /MWh-gross basis over an extended period of time ( e.g., a year).

For low load natural gas- and oil-fired steam generating units, the EPA is finalizing a BSER of uniform fuels, with a degree of emission limitation on a heat input basis consistent with a fixed 130 lb CO₂ /MMBtu for natural gas-fired steam generating units and 170 lb CO₂ /MMBtu for oil-fired steam generating units. The degree of emission limitation for natural gas- and oil-fired steam generating units is higher than the corresponding values under 40 CFR part 60, subpart TTTT, because steam generating units may fire fuels with slightly higher carbon contents.

4. Other Emission Reduction Measures Not Considered BSER

a. Heat Rate Improvements

Heat rate is a measure of efficiency that is commonly used in the power sector. The heat rate is the amount of energy input, measured in Btu, required to generate 1 kilowatt-hour (kWh) of electricity. The lower an EGU's heat rate, the more efficiently it operates. As a result, an EGU with a lower heat rate will consume less fuel and emit lower amounts of CO₂ and other air pollutants per kWh generated as compared to a less efficient unit. HRI measures include a variety of technology upgrades and operating practices that may achieve CO₂ emission rate reductions of 0.1 to 5 percent for individual EGUs. The EPA considered HRI to be part of the BSER in the CPP and to be the BSER in the ACE Rule. However, the reductions that may be achieved by HRI are small relative to the reductions from natural gas co-firing and CCS. Also, some facilities that apply HRI would, as a result of their increased efficiency, increase their utilization and therefore increase their CO₂ emissions (as well as emissions of other air pollutants), a phenomenon that the EPA has termed the “rebound effect.” Therefore, the EPA is not finalizing HRI as a part of BSER.

i. CO₂ Reductions From HRI in Prior Rulemakings

In the CPP, the EPA quantified emission reductions achievable through heat rate improvements on a regional basis by an analysis of historical emission rate data, taking into consideration operating load and ambient temperature. The Agency concluded that EGUs can achieve on average a 4.3 percent improvement in the Eastern Interconnection, a 2.1 percent improvement in the Western Interconnection, and a 2.3 percent improvement in the Texas Interconnection. See 80 FR 64789 (October 23, 2015). The Agency then applied all three of the building blocks to 2012 baseline data and quantified, in the form of CO₂ emission rates, the reductions achievable in Each interconnection in 2030, and then selected the least stringent as a national performance rate. Id. at 64811-19. The EPA noted that building block 1 measures could not by themselves constitute the BSER because the quantity of emission reductions achieved would be too small and because of the potential for an increase in emissions due to increased utilization ( i.e., the “rebound effect”).

ii. Updated CO₂ Reductions From HRI

The HRI measures include improvements to the boiler island ( e.g., neural network system, intelligent sootblower system), improvements to the steam turbine ( e.g., turbine overhaul and upgrade), and other equipment upgrades ( e.g., variable frequency drives). Some regular practices that may recover degradation in heat rate to recent levels—but that do not result in upgrades in heat rate over recent design levels and are therefore not HRI measures—include practices such as in-kind replacements and regular surface cleaning ( e.g., descaling, fouling removal). Specific details of the HRI measures are described in the final TSD, GHG Mitigation Measures for Steam Generating Units and an updated 2023 Sargent and Lundy HRI report ( Heat Rate Improvement Method Costs and Limitations Mem o), available in the docket. Most HRI upgrade measures achieve reductions in heat rate of less than 1 percent. In general, the 2023 Sargent and Lundy HRI report, which updates the 2009 Sargent and Lundy HRI report, shows that HRI achieve less reductions than indicated in the 2009 report, and shows that several HRI either have limited applicability or have already been applied at many units. Steam path overhaul and upgrade may achieve reductions up to 5.15 percent, with the average being around 1.5 percent. Different combinations of HRI measures do not necessarily result in cumulative reductions in emission rate ( e.g., intelligent sootblowing systems combined with neural network systems). Some of the HRI measures ( e.g., variable frequency drives) only impact heat rate on a net generation basis by reducing the parasitic load on the unit and would thereby not be observable for emission rates measured on a gross basis. Assuming many of the HRI measures could be applied to the same unit, adding together the upper range of some of the HRI percentages could yield an emission rate reduction of around 5 percent. However, the reductions that the fleet could achieve on average are likely much smaller. As noted, the 2023 Sargent and Lundy HRI report notes that, in many cases, units have already applied HRI upgrades or that those upgrades would not be applicable to all units. The unit level reductions in emission rate from HRI are small relative to CCS or natural gas co-firing. In the CPP and ACE Rule, the EPA viewed CCS and natural gas co-firing as too costly to qualify as the BSER; those costs have fallen since those rules and, as a result, CCS and natural gas co-firing do qualify as the BSER for the long-term and medium-term subcategories, respectively.

iii. Potential for Rebound in CO₂ Emissions

Reductions achieved on a rate basis from HRI may not result in overall emission reductions and could instead cause a “rebound effect” from increased utilization. A rebound effect would occur where, because of an improvement in its heat rate, a steam generating unit experiences a reduction in variable operating costs that makes the unit more competitive relative to other EGUs and consequently raises the unit's output. The increase in the unit's CO₂ emissions associated with the increase in output would offset the reduction in the unit's CO₂ emissions caused by the decrease in its heat rate and rate of CO₂ emissions per unit of output. The extent of the offset would depend on the extent to which the unit's generation increased. The CPP did not consider HRI to be BSER on its own, in part because of the potential for a rebound effect. Analysis for the ACE Rule, where HRI was the entire BSER, observed a rebound effect for certain sources in some cases. In this action, where different subcategories of units are to be subject to different BSER measures, steam generating units in a hypothetical subcategory with HRI as BSER could experience a rebound effect. Because of this potential for perverse GHG emission outcomes resulting from deployment of HRI at certain steam generating units, coupled with the relatively minor overall GHG emission reductions that would be expected from this measure, the EPA is not finalizing HRI as the BSER for any subcategory of existing coal-fired steam generating units.

E. Additional Comments Received on the Emission Guidelines for Existing Steam Generating Units and Responses

1. Consistency With West Virginia v. EPA and the Major Questions Doctrine

Comment: Some commenters argued that the EPA's determination that CCS is the BSER for existing coal-fired power plants is invalid under West Virginia v. EPA, 597 U.S. 697 (2022), and the major questions doctrine (MQD). Commenters state that for various reasons, coal-fired power plants will not install CCS and instead will be forced to retire their units. They point to the EPA's IPM modeling which, they say, shows that many coal-fired power plants retire rather than install CCS. They add that, in this way, the rule effectively results in the EPA's requiring generation-shifting from coal-fired generation to renewable and other generation, and thus is like the Clean Power Plan (CPP). For those reasons, they state that the rule raises a major question, and further that CAA section 111(d) does not contain a clear authorization for this type of rule.

Response: The EPA discussed West Virginia and its articulation of the MQD in section V.B.6 of this preamble.

The EPA disagrees with these comments. This rule is fully consistent with the Supreme Court's interpretation of the EPA's authority in West Virginia. The EPA's determination that CCS—a traditional, add-on emissions control—is the BSER is consistent with the plain text of section 111. As explained in detail in section VII.C.1.a, for long-term coal-fired steam generating units, CCS meets all of the BSER factors: it is adequately demonstrated, of reasonable cost, and achieves substantial emissions reductions. That some coal-fired power plants will choose not to install emission controls and will instead retire does not raise major questions concerns.

In West Virginia, the U.S. Supreme Court held that “generation-shifting” as the BSER for coal- and gas-fired units “effected a fundamental revision of the statute, changing it from one sort of scheme of regulation into an entirely different kind.” 597 U.S. at 728 (internal quotation marks, brackets, and citation omitted). The Court explained that prior CAA section 111 rules were premised on “more traditional air pollution control measures” that “focus on improving the performance of individual sources.” Id. at 727 (citing “fuel-switching” and “add-on controls”). The Court said that generation-shifting as the BSER was “unprecedented” because it was designed to “improve the overall power system by lowering the carbon intensity of power generation . . . by forcing a shift throughout the power grid from one type of energy source to another.” Id. at 727-28 (internal quotation marks, emphasis, and citation omitted). The Court cited statements by the then-Administrator describing the CPP as “not about pollution control so much as it was an investment opportunity for States, especially investments in renewables and clean energy.” Id. at 728. The Court further concluded that the EPA's view of its authority was virtually unbounded because the “EPA decides, for instance, how much of a switch from coal to natural gas is practically feasible by 2020, 2025, and 2030 before the grid collapses, and how high energy prices can go as a result before they become unreasonably exorbitant.” Id. at 729.

Here, the EPA's determination that CCS is the BSER does not affect a fundamental revision of the statute, nor is it unbounded. CCS is not directed at improvement of the overall power system. Rather, CCS is a traditional “add-on [pollution] control[ ]” akin to measures that the EPA identified as BSER in prior CAA section 111 rules. See id. at 727. It “focus[es] on improving the performance of individual sources”—it reduces CO₂ pollution from each individual source—because each affected source is able to apply it to its own facility to reduce its own emissions. Id. at 727. Further, the EPA determined that CCS qualifies as the BSER by applying the criteria specified in CAA section 111(a)(1)—including adequate demonstration, costs of control, and emissions reductions. See section VII.C.1.a of this preamble. Thus, CCS as the BSER does not “chang[e]” the statute “from one sort of scheme of regulation into an entirely different kind.” Id. at 728 (internal quotation marks, brackets, and citation omitted).

Commenters contend that notwithstanding these distinctions, the choice of CCS as the BSER has the effect of shifting generation because modeling projections for the rule show that coal-fired generation will become less competitive, and gas-fired and renewable-generated electricity will be more competitive and dispatched more frequently. That some coal-fired sources may retire rather than reduce their CO₂ pollution does not mean that the rule “represents a transformative expansion [of EPA's] regulatory authority”. Id. at 724. To be sure, this rule's determination that CCS is the BSER imposes compliance costs on coal-fired power plants. That sources will incur costs to control their emissions of dangerous pollution is an unremarkable consequence of regulation, which, as the Supreme Court recognized, “may end up causing an incidental loss of coal's market share.” Id. at 731 n.4. Indeed, ensuring that sources internalize the full costs of mitigating their impacts on human health and the environment is a central purpose of traditional environmental regulation.

In particular, for the power sector, grid operators constantly shift generation as they dispatch electricity from sources based upon their costs. The EPA's IPM modeling, which is based on the costs of the various types of electricity generation, projects these impacts. Viewed as a whole, these projected impacts show that, collectively, coal-fired power plants will likely produce less electricity, and other sources (like gas-fired units and renewable sources) will likely produce more electricity, but this pattern does not constitute a transformative expansion of statutory authority (EPA's Power Sector Platform 2023 using IPM; final TSD, Power Sector Trends.)

These projected impacts are best understood by comparing the IPM model's “base case,” i.e., the projected electricity generation without any rule in place, to the model's “policy case,” i.e., the projected electricity generation expected to result from this rule. The base case projects that many coal-fired units will retire over the next 20 years (EPA's Power Sector Platform 2023 using IPM; final TSD, Power Sector Trends). Those projected retirements track trends over the past two decades where coal-fired units have retired in high numbers because gas-fired units and renewable sources have become increasingly able to generate lower-cost electricity. As more gas-fired and renewable generation sources deploy in the future, and as coal-fired units continue to age—which results in decreased efficiency and increased costs—the coal-fired units will become increasingly marginal and continue to retire (EPA's Power Sector Platform 2023 using IPM; final TSD, Power Sector Trends.) That is true in the absence of this rule. The EPA's modeling results also project that even if the EPA had determined BSER for long-term sources to be 40 percent co-firing, which requires significantly less capital investment, and not 90 percent capture CCS, a comparable number of sources would retire instead of installing controls. These results confirm that the primary cause for the projected retirements is the marginal profitability of the sources.

Importantly, the base-case projections also show that some coal-fired units install CCS and run at high capacity factors, in fact, higher than they would have had they not installed CCS. This is because the IRC section 45Q tax credit significantly reduces the variable cost of operation for qualifying sources. This incentivizes sources to increase generation to maximize the tons of CO₂ the CCS equipment captures, and thereby increase the amount of the tax credit they receive. In the “policy case,” beginning when the CCS requirement applies in the 2035 model year, some additional coal-fired units will likely install CCS, and also run at high capacity factors, again, significantly higher than they would have without CCS. Other units may retire rather than install emission controls (EPA's Power Sector Platform 2023 using IPM; final TSD, Power Sector Trends). On balance, the coal-fired units that install CCS collectively generate nearly the same amount of electricity in the 2040 model year as do the group of coal-fired units in the base case.

The policy case also shows that in the 2045 model year, by which time the 12-year period for sources to claim the IRC section 45Q tax credit will have expired, most sources that install CCS retire due to the costs of meeting the CCS-based standards without the benefit of the tax credit. However, in fact, these projected outcomes are far from certain as the modeling results generally do not account for numerous potential changes that may occur over the next 20 or more years, any of which may enable these units to continue to operate economically for a longer period. Examples of potential changes include reductions in the operational costs of CCS through technological improvements, or the development of additional potential revenue streams for captured CO₂ as the market for beneficial uses of CO₂ continues to develop, among other possible changed economic circumstances (including the possible extension of the tax credits). In light of these potential significant developments, the EPA is committing to review and, if appropriate, revise the requirements of this rule by January 1, 2041, as described in section VII.F.

In any event, the modeling projections showing that many sources retire instead of installing controls are in line with the trends for these units in the absence of the rule—as the coal-fired fleet ages and lower-cost alternatives become increasingly available, more operators will retire coal-fired units with or without this rule. In 2045, the average age of coal-fired units that have not yet announced retirement dates or coal-to-gas conversion by 2039 will be 61 years old. And, on average, between 2000 and 2022, even in the absence of this rule, coal-fired units generally retired at 53 years old. Thus, taken as a whole, this rule does not dramatically reduce the expected operating horizon of most coal-fired units. Indeed, for units that install CCS, the generous IRC section 45Q tax credit increases the competitiveness of these units, and it allows them to generate more electricity with greater profit than the sources would otherwise generate if they did not install CCS.

The projected effects of the rule do not show the BSER—here, CCS—is akin to generation shifting, or otherwise represents an expansion of EPA authority with vast political or economic significance. As described above at VII.C.1.a.ii, CCS is an affordable emissions control technology. It is also very effective, reducing CO₂ emissions from coal-fired units by 90 percent, as described in section VII.C.1.a.i. Indeed, as noted, the IRA tax credits make CCS so affordable that coal-fired units that install CCS run at higher capacity factors than they would otherwise.

Considered as a whole, and in context with historical retirement trends, the projected impacts of this rule on coal-fired generating units do not raise MQD concerns. The projected impacts are merely incidental to the CCS control itself—the unremarkable consequence of marginally increasing the cost of doing business in a competitive market. Nor is the rule “transformative.” The rule does not “announce what the market share of coal, natural gas, wind, and solar must be, and then requiring plants to reduce operations or subsidize their competitors to get there.” 597 U.S. at 731 n.4. As noted above, coal-fired units that install CCS are projected to generate substantial amounts of electricity. The retirements that are projected to occur are broadly consistent with market trends over the past two decades, which show that coal-fired electricity production is generally less economic and less competitive than other forms of electricity production. That is, the retirements that the model predicts under this rule, and the structure of the industry that results, diverge little from the prior rate of retirements of coal-fired units over the past two decades. They also diverge little from the rate of retirements from sources that have already announced that they will retire, or from the additional retirements that IPM projects will occur in the base case (EPA's Power Sector Platform 2023 using IPM; final TSD, Power Sector Trends).

As discussed above, because much of the coal-fired fleet is operating on the edge of viability, many sources would retire instead of installing any meaningful CO₂ emissions control—whether CCS, natural gas co-firing, or otherwise. Under commenters' view that such retirements create a major question, any form of meaningful regulation of these sources would create a major question and effect a fundamental revision of the statute. That cannot possibly be so. Section 111(d)(1) plainly mandates regulation of these units, which are the biggest stationary source of dangerous CO₂ emissions.

The legislative history for the CAA further makes clear that Congress intended the EPA to promulgate regulations even where emissions controls had economic costs. At the time of the 1970 CAA Amendments, Congress recognized that the threats of air pollution to public health and welfare had grown urgent and severe. Sen. Edmund Muskie (D-ME), manager of the bill and chair of the Public Works Subcommittee on Air and Water Pollution, which drafted the bill, regularly referred to the air pollution problem as a “crisis.” As Sen. Muskie recognized, “Air pollution control will be cheap only in relation to the costs of lack of control.” The Senate Committee Report for the 1970 CAA Amendments specifically discussed the precursor provision to section 111(d) and noted, “there should be no gaps in control activities pertaining to stationary source emissions that pose any significant danger to public health or welfare.” Accordingly, some of the EPA's prior CAA section 111 rulemakings have imposed stringent requirements, at significant cost, in order to achieve significant emission reductions.

Congress's enactment of the IRA and IIJA further shows its view that reducing air pollution—specifically, in those laws, GHG emissions to address climate change—is a high priority. As discussed in section IV.E.1, that law provided funds for DOE grant and loan programs to support CCS, and extended and increased the IRC section 45Q tax credit for carbon capture. It also adopted the Low Emission Electricity Program (LEEP), which allocates funds to the EPA for the express purpose of using CAA regulatory authority to reduce GHG emissions from domestic electricity generation through use of its existing CAA authorities. CAA section 135, added by IRA section 60107. The EPA is promulgating the present rulemaking with those funds. The congressional sponsor of the LEEP made clear that it authorized the type of rulemaking that the EPA is promulgating today: he stated that the EPA may promulgate rulemaking under CAA section 111, based on CCS, to address CO₂ emissions from fossil fuel-fired power plants, which may be “impactful” by having the “incidental effect” of leading some “companies . . . to choose to retire such plants. . . .”

For these reasons, the rule here is consistent with the Supreme Court's decision in West Virginia. The selection of CCS as the BSER for existing coal-fired units is a traditional, add-on control intended to reduce the emissions performance of individual sources. That some sources may retire instead of controlling their emissions does not otherwise show that the rule runs afoul of the MQD. The modeling projections for this rule show that the anticipated retirements are largely consistent with historical trends, and due to many coal-fired units' advanced age and lack of competitiveness with lower cost methods of electricity generation.

2. Redefining the Source

Comment: Some commenters contended that the proposed 40 percent natural gas co-firing performance standard violates legal precedent that bars the EPA from setting technology-based performance standards that would have the effect of “redefining the source.” They stated that this prohibition against the redefinition of the source bars the EPA from adopting the proposed performance standard for medium-term coal-fired EGUs, which requires such units to operate in a manner for which the unit was never designed to do, namely operate as a hybrid coal/natural gas co-firing generating unit and combusting 40 percent of its fuel input as natural gas (instead of coal) on an annual basis.

Commenters argued that co-firing would constitute forcing one type of source to become an entirely different kind of source, and that the Supreme Court precluded such a requirement in West Virginia v. EPA when it stated in footnote 3 of that case that the EPA has “never ordered anything remotely like” a rule that would “simply require coal plants to become natural gas plants” and the Court “doubt[ed that EPA] could.”

Response: The EPA disagrees with these comments.

Standards based on co-firing, as contemplated in this rule, are based on a “traditional pollution control measure,” in particular, “fuel switching,” as the Supreme Court recognized in West Virginia. Rules based on switching to a cleaner fuel are authorized under the CAA, an authorization directly acknowledged by Congress. Specifically, as part of the 1977 CAA Amendments, Congress required that the EPA base its standards regulating certain new sources, including power plants, on “technological” controls, rather than simply the “best system.” Congress understood this to mean that new sources would be required to implement add-on controls, rather than merely relying on fuel switching, and noted that one of the purposes of this amendment was to allow new sources to burn high sulfur coal while still decreasing emissions, and thus to increase the availability of low sulfur coal for existing sources, which were not subject to the “technological” control requirement. In 1990, however, Congress removed the “technological” language, allowing the EPA to set fuel-switching based standards for both new and existing power plants.

The EPA has a tradition of promulgating rules based on fuel switching. For example, the 2006 NSPS for stationary compression ignition internal combustion engines required the use of ultra-low sulfur diesel. Similarly, in the 2015 NSPS for EGUs, the EPA determined that the BSER for peaking plants was to burn primarily natural gas, with distillate oil used only as a backup fuel. Nor is this approach unique to CAA section 111; in the 2016 rule setting section 112 standards for hazardous air pollutant emissions from area sources, for example, the EPA finalized an alternative particulate matter (PM) standard that specified that certain oil-fired boilers would meet the applicable standard if they combusted only ultra-low-sulfur liquid fuel.

Moreover, the West Virginia Court's statements in footnote 3 are irrelevant to the question of the validity of a 40 percent co-firing standard. There, the Court was referring to a complete transformation of the coal-fired unit to a 100 percent gas fired unit—a change that would require entirely repowering the unit. By contrast, increasing co-firing at existing coal-fired units to 40 percent would require only minor changes to the units' boilers. In fact, many coal-fired units are already capable of co-firing some amount of gas without any changes at all, and several have fired at 40 percent and above in recent years. Of the 565 coal-fired EGUs operating at the end of 2021, 249 of them reported consuming natural gas as a fuel or startup source, 162 reported more than one month of consumption of natural gas at their boiler, and 29 co-fired at over 40 percent on an annual heat input basis in at least one year while also operating with annual capacity factors greater than 10 percent. For more on this, see section IV.C.2 of this preamble; see also the final TSD, GHG Mitigation Measures for Steam Generating Units.

F. Commitment To Review and, If Appropriate, Revise Emission Guidelines for Coal-Fired Units

The EPA recognizes that the IRC 45Q tax credit is a key component to the cost of CCS, as discussed in section VII.C.1.a.ii(C) of this preamble. The EPA further recognizes that for any affected source, the tax credit is currently available for a 12-year period and not subsequently. The tax credit is generally sufficient to defray the capital costs of CCS and much, if not all, of the operating costs during that 12-year period. Following the 12-year period, affected sources that continue to operate the CCS equipment would have higher costs of generation, due to the CCS operating costs, including parasitic load. Under certain circumstances, these higher costs could push the affected sources lower on the dispatch curve, and thereby lead to reductions in the amount of their generation, i.e., if affected sources are not able to replace the revenue from the tax credit with revenue from other sources, or if the price of electricity does not reflect any additional costs needed to minimize GHG emissions.

However, the costs of CCS and the overall economic viability of operating CO₂ capture at power plants are improving and can be expected to continue to improve in years to come. CO₂ that is captured from fossil-fuel fired sources is currently beneficially used, including, for example, for enhanced oil recovery and in the food and beverage industry. There is much research into developing beneficial uses for many other industries, including construction, chemical manufacturing, graphite manufacturing. The demand for CO₂ is expected to grow considerably over the next several decades. As a result, in the decades to come, affected sources may well be able to replace at least some of the revenues from the tax credit with revenues from the sale of CO₂. We discuss these potential developments in chapter 2 of the Response to Comments document, available in the rulemaking docket.

In addition, numerous states have imposed requirements to decarbonize generation within their borders. Many utilities have also announced plans to decarbonize their fleet, including building small modular (advanced nuclear) reactors. Given the relatively high capital and fixed costs of small modular reactors, plans for their construction represent an expectation of higher future energy prices. This suggests that, in the decades to come, at least in certain areas of the country, affected sources may be able to maintain a place in the dispatch curve that allows them to continue to generate while they continue to operate CCS, even in the absence of additional revenues for CO₂. We discuss these potential developments in the final TSD, Power Sector Trends, available in the rulemaking docket.

These developments, which may occur by the 2040s—the expiration of the 12-year period for the IRC 45Q tax credit, the potential development of the CO₂ utilization market, and potential market supports for low-GHG generation—may significantly affect the costs to coal-fired steam EGUs of operating their CCS controls. As a result, the EPA will closely monitor these developments. Our efforts will include consulting with other agencies with expertise and information, including DOE, which currently has a program, the Carbon Conversion Program, in the Office of Carbon Management, that funds research into CO₂ utilization. We regularly consult with stakeholders, including industry stakeholders, and will continue to do so.

In light of these potential significant developments and their impacts, potentially positive or negative, on the economics of continued generation by affected sources that have installed CCS, the EPA is committing to review and, if appropriate, revise this rule by January 1, 2041. This commitment is included in the regulations that the EPA is promulgating with this rule. The EPA will conduct this review based on what we learn from monitoring these developments, as noted above. Completing this review and any appropriate revisions by that date will allow time for the states to revise, if necessary, standards applicable to affected sources, and for the EPA to act on those state revisions, by the early to mid-2040s. That is when the 12-year period for the 45Q tax credit is expected to expire for affected sources that comply with the CCS requirement by January 1, 2032, and when other significant developments noted above may be well underway.

VIII. Requirements for New and Reconstructed Stationary Combustion Turbine EGUs and Rationale for Requirements

A. Overview

This section discusses the requirements for stationary combustion turbine EGUs that commence construction or reconstruction after May 23, 2023. The requirements are codified in 40 CFR part 60, subpart TTTTa. The EPA explains in section VIII.B of this document the two basic turbine technologies that are used in the power sector and are covered by 40 CFR part 60, subpart TTTTa. Those are simple cycle combustion turbines and combined cycle combustion turbines. The EPA also explains how these technologies are used in the three subcategories: low load turbines, intermediate load turbines, and base load turbines. Section VIII.C provides an overview of how stationary combustion turbines have been previously regulated. Section VIII.D discusses the EPA's decision to revisit the standards for new and reconstructed turbines as part of the statutorily required 8-year review of the NSPS. Section VIII.E discusses changes that the EPA is finalizing in both applicability and subcategories in the new 40 CFR part 60, subpart TTTTa, as compared to those codified previously in 40 CFR part 60, subpart TTTT. Most notably, for new and reconstructed natural gas-fired combustion turbines, the EPA is finalizing BSER determinations and standards of performance for the three subcategories mentioned above—low load, intermediate load, and base load.

Sections VIII.F and VIII.G of this document discuss the EPA's determination of the BSER for each of the three subcategories of combustion turbines and the applicable standards of performance, respectively. For low load combustion turbines, the EPA is finalizing a determination that the use of lower-emitting fuels is the appropriate BSER. For intermediate load combustion turbines, the EPA is finalizing a determination that highly efficient simple cycle generation is the appropriate BSER. For base load combustion turbines, the EPA is finalizing a determination that the BSER includes two components that correspond initially to a two-phase standard of performance. The first component of the BSER, with an immediate compliance date (phase 1), is highly efficient generation based on the performance of a highly efficient combined cycle turbine and the second component of the BSER, with a compliance date of January 1, 2032 (phase 2), is based on the use of CCS with a 90 percent capture rate, along with continued use of highly efficient generation. For base load turbines, the standards of performance corresponding to both components of the BSER would apply to all new and reconstructed sources that commence construction or reconstruction after May 23, 2023. The EPA occasionally refers to these standards of performance as the phase 1 or phase 2 standards.

B. Combustion Turbine Technology

For purposes of 40 CFR part 60, subparts TTTT and TTTTa, stationary combustion turbines include both simple cycle and combined cycle EGUs. Simple cycle turbines operate in the Brayton thermodynamic cycle and include three primary components: a multi-stage compressor, a combustion chamber ( i.e., combustor), and a turbine. The compressor is used to supply large volumes of high-pressure air to the combustion chamber. The combustion chamber converts fuel to heat and expands the now heated, compressed air through the turbine to create shaft work. The shaft work drives an electric generator to produce electricity. Combustion turbines that recover the energy in the high-temperature exhaust—instead of venting it directly to the atmosphere—are combined cycle EGUs and can obtain additional useful electric output. A combined cycle EGU includes an HRSG operating in the Rankine thermodynamic cycle. The HRSG receives the high-temperature exhaust and converts the heat to mechanical energy by producing steam that is then fed into a steam turbine that, in turn, drives an electric generator. As the thermal efficiency of a stationary combustion turbine EGU is increased, less fuel is burned to produce the same amount of electricity, with a corresponding decrease in fuel costs and lower emissions of CO₂ and, generally, of other air pollutants. The greater the output of electric energy for a given amount of fuel energy input, the higher the efficiency of the electric generation process.

Combustion turbines serve various roles in the power sector. Some combustion turbines operate at low annual capacity factors and are available to provide temporary power during periods of high load demand. These turbines are often referred to as “peaking units.” Some combustion turbines operate at intermediate annual capacity factors and are often referred to as cycling or load-following units. Other combustion turbines operate at high annual capacity factors to serve base load demand and are often referred to as base load units. In this rulemaking, the EPA refers to these types of combustion turbines as low load, intermediate load, and base load, respectively.

Low load combustion turbines provide reserve capacity, support grid reliability, and generally provide power during periods of peak electric demand. As such, the units may operate at or near their full capacity, but only for short periods, as needed. Because these units only operate occasionally, capital expenses are a major factor in the overall cost of electricity, and often, the lowest capital cost (and generally less efficient) simple cycle EGUs are intended for use only during periods of peak electric demand. Due to their low efficiency, these units require more fuel per MWh of electricity produced and their operating costs tend to be higher. Because of the higher operating costs, they are generally some of the last units in the dispatch order. Important characteristics for low load combustion turbines include their low capital costs, their ability to start quickly and to rapidly ramp up to full load, and their ability to operate at partial loads while maintaining acceptable emission rates and efficiencies. The ability to start quickly and rapidly attain full load is important to maximize revenue during periods of peak electric prices and to meet sudden shifts in demand. In contrast, under steady-state conditions, more efficient combined cycle EGUs are dispatched ahead of low load turbines and often operate at higher annual capacity factors.

Highly efficient simple cycle turbines and flexible fast-start combined cycle turbines both offer different advantages and disadvantages when operating at intermediate loads. One of the roles of these intermediate or load following EGUs is to provide dispatchable backup power to support variable renewable generating sources ( e.g., solar and wind). A developer's decision as to whether to build a simple cycle turbine or a combined cycle turbine to serve intermediate load demand is based on several factors related to the intended operation of the unit. These factors would include how frequently the unit is expected to cycle between starts and stops, the predominant load level at which the unit is expected to operate, and whether this level of operation is expected to remain consistent or is expected to vary over the lifetime of the unit. In areas of the U.S. with vertically integrated electricity markets, utilities determine dispatch orders based generally on economic merit of individual units. Meanwhile, in areas of the U.S. inside organized wholesale electricity markets, owner/operators of individual combustion turbines control whether and how units will operate over time, but they do not necessarily control the precise timing of dispatch for units in any given day or hour. Such short-term dispatch decisions are often made by regional grid operators that determine, on a moment-to-moment basis, which available individual units should operate to balance supply and demand and other requirements in an optimal manner, based on operating costs, price bids, and/or operational characteristics. However, operating permits for simple cycle turbines often contain restrictions on the annual hours of operation that owners/operators incorporate into longer-term operating plans and short-term dispatch decisions.

Intermediate load combustion turbines vary their generation, especially during transition periods between low and high electric demand. Both high-efficiency simple cycle turbines and flexible fast-start combined cycle turbines can fill this cycling role. While the ability to start quickly and quickly ramp up is important, efficiency is also an important characteristic. These combustion turbines generally have higher capital costs than low load combustion turbines but are generally less expensive to operate.

Base load combustion turbines are designed to operate for extended periods at high loads with infrequent starts and stops. Quick-start capability and low capital costs are less important than low operating costs. High-efficiency combined cycle turbines typically fill the role of base load combustion turbines.

The increase in generation from variable renewable energy sources during the past decade has impacted the way in which dispatchable generating resources operate. For example, the electric output from wind and solar generating sources fluctuates daily and seasonally due to increases and decreases in the wind speed or solar intensity. Due to this variable nature of wind and solar, dispatchable EGUs, including combustion turbines as well as other technologies like energy storage, are used to ensure the reliability of the electric grid. This requires dispatchable power plants to have the ability to quickly start and stop and to rapidly and frequently change load—much more often than was previously needed. These are important characteristics of the combustion turbines that provide firm backup capacity. Combustion turbines are much more flexible than coal-fired utility boilers in this regard and have played an important role during the past decade in ensuring that electric supply and demand are balanced.

As discussed in section IV.F.2 of this preamble, in the final TSD, Power Sector Trends, and in the accompanying RIA, the EPA's Power Sector Platform 2023 using IPM projects that natural gas-fired combustion turbines will continue to play an important role in meeting electricity demand. However, that role is projected to evolve as additional renewable and non-renewable low-GHG generation and energy storage technologies are added to the grid. Energy storage technologies can store energy during periods when generation from renewable resources is high relative to demand and can provide electricity to the grid during other periods. Energy storage technologies are projected to reduce the need for base load fossil fuel-fired firm dispatchable power plants, and the capacity factors of combined cycle EGUs are forecast to decline by 2040.

C. Overview of Regulation of Stationary Combustion Turbines for GHGs

As explained earlier in this preamble, the EPA originally regulated new and reconstructed stationary combustion turbine EGUs for emissions of GHGs in 2015 under 40 CFR part 60, subpart TTTT. In 40 CFR part 60, subpart TTTT, the EPA created three subcategories: two for natural gas-fired combustion turbines and one for multi-fuel-fired combustion turbines. For natural gas-fired turbines, the EPA created a subcategory for base load turbines and a separate subcategory for non-base load turbines. Base load turbines were defined as combustion turbines with electric sales greater than a site-specific electric sales threshold based on the design efficiency of the combustion turbine. Non-base load turbines were defined as combustion turbines with a capacity factor less than or equal to the site-specific electric sales threshold. For base load turbines, the EPA set a standard of 1,000 lb CO₂ /MWh-gross based on efficient combined cycle turbine technology. For non-base load and multi-fuel-fired turbines, the EPA set a standard based on the use of lower-emitting fuels that varied from 120 lb CO₂ /MMBtu to 160 lb CO₂ /MMBtu, depending upon whether the turbine burned primarily natural gas or other lower-emitting fuels.

D. Eight-Year Review of NSPS

CAA section 111(b)(1)(B) requires the Administrator to “at least every 8 years, review and, if appropriate, revise [the NSPS] . . . .” The provision further provides that “the Administrator need not review any such standard if the Administrator determines that such review is not appropriate in light of readily available information on the efficacy of such [NSPS].”

The EPA promulgated the NSPS for GHG emissions for stationary combustion turbines in 2015. Announcements and modeling projections show that project developers are building new fossil fuel-fired combustion turbines and have plans to continue building additional capacity. Because the emissions from this added capacity have the potential to be large and these units are likely to have long operating lives (25 years or more), it is important to limit emissions from these new units. Accordingly, in this final rule, the EPA is updating the NSPS for newly constructed and reconstructed fossil fuel-fired stationary combustion turbines.

E. Applicability Requirements and Subcategorization

This section describes the amendments to the specific applicability criteria for non-fossil fuel-fired EGUs, industrial EGUs, CHP EGUs, and combustion turbine EGUs not connected to a natural gas pipeline. The EPA is also making certain changes to the applicability requirements for stationary combustion turbines affected by this final rule as compared to those for sources affected by the 2015 NSPS. The amendments are described below and include the elimination of the multi-fuel-fired subcategory, further binning non-base load combustion turbines into low load and intermediate load subcategories and establishing a capacity factor threshold for base load combustion turbines.

1. Applicability Requirements

In general, the EPA refers to fossil fuel-fired EGUs that would be subject to a CAA section 111 NSPS as “affected” EGUs or units. An EGU is any fossil fuel-fired electric utility steam generating unit ( i.e., a utility boiler or IGCC unit) or stationary combustion turbine (in either simple cycle or combined cycle configuration). To be considered an affected EGU under the 2015 NSPS at 40 CFR part 60, subpart TTTT, the unit must meet the following applicability criteria: The unit must: (1) be capable of combusting more than 250 MMBtu/h (260 gigajoules per hour (GJ/h)) of heat input of fossil fuel (either alone or in combination with any other fuel); and (2) serve a generator capable of supplying more than 25 MW net to a utility distribution system ( i.e., for sale to the grid). However, 40 CFR part 60, subpart TTTT, includes applicability exemptions for certain EGUs, including: (1) non-fossil fuel-fired units subject to a federally enforceable permit that limits the use of fossil fuels to 10 percent or less of their heat input capacity on an annual basis; (2) CHP units that are subject to a federally enforceable permit limiting annual net electric sales to no more than either the unit's design efficiency multiplied by its potential electric output, or 219,000 MWh, whichever is greater; (3) stationary combustion turbines that are not physically capable of combusting natural gas ( e.g., those that are not connected to a natural gas pipeline); (4) utility boilers and IGCC units that have always been subject to a federally enforceable permit limiting annual net electric sales to one-third or less of their potential electric output ( e.g., limiting hours of operation to less than 2,920 hours annually) or limiting annual electric sales to 219,000 MWh or less; (5) municipal waste combustors that are subject to 40 CFR part 60, subpart Eb; (6) commercial or industrial solid waste incineration units subject to 40 CFR part 60, subpart CCCC; and (7) certain projects under development, as discussed in the preamble for the 2015 final NSPS.

a. Revisions to 40 CFR Part 60, Subpart TTTT

The EPA is amending 40 CFR 60.5508 and 60.5509 to reflect that stationary combustion turbines that commenced construction after January 8, 2014, or reconstruction after June 18, 2014, and before May 24, 2023, and that meet the relevant applicability criteria are subject to 40 CFR part 60, subpart TTTT. For steam generating EGUs and IGCC units, 40 CFR part 60, subpart TTTT, remains applicable for units constructed after January 8, 2014, or reconstructed after June 18, 2014. The EPA is finalizing 40 CFR part 60, subpart TTTTa, to be applicable to stationary combustion turbines that commence construction or reconstruction after May 23, 2023, and that meet the relevant applicability criteria.

b. Revisions to 40 CFR Part 60, Subpart TTTT, That Are Also Included in 40 CFR Part 60, Subpart TTTTa

The EPA is finalizing that 40 CFR part 60, subpart TTTT, and 40 CFR part 60, subpart TTTTa, use similar regulatory text except where specifically stated. This section describes amendments included in both subparts.

i. Applicability to Non-Fossil Fuel-Fired EGUs

The current non-fossil applicability exemption in 40 CFR part 60, subpart TTTT, is based strictly on the combustion of non-fossil fuels ( e.g., biomass). To be considered a non-fossil fuel-fired EGU, the EGU must be both: (1) Capable of combusting more than 50 percent non-fossil fuel and (2) subject to a federally enforceable permit condition limiting the annual heat input capacity for all fossil fuels combined of 10 percent or less. The current language does not take heat input from non-combustion sources ( e.g., solar thermal) into account. Certain solar thermal installations have natural gas backup burners larger than 250 MMBtu/h. As currently treated in 40 CFR part 60, subpart TTTT, these solar thermal installations are not eligible to be considered non-fossil units because they are not capable of deriving more than 50 percent of their heat input from the combustion of non-fossil fuels. Therefore, solar thermal installations that include backup burners could meet the applicability criteria of 40 CFR part 60, subpart TTTT, even if the burners are limited to an annual capacity factor of 10 percent or less. These EGUs would readily comply with the standard of performance, but the reporting and recordkeeping would increase costs for these EGUs.

The EPA proposed and is finalizing several amendments to align the applicability criteria with the original intent to cover only fossil fuel-fired EGUs. These amendments ensure that solar thermal EGUs with natural gas backup burners, like other types of non-fossil fuel-fired units that derive most of their energy from non-fossil fuel sources, are not subject to the requirements of 40 CFR part 60, subpart TTTT or TTTTa. Amending the applicability language to include heat input derived from non-combustion sources allows these facilities to avoid the requirements of 40 CFR part 60, subpart TTTT or TTTTa, by limiting the use of the natural gas burners to less than 10 percent of the capacity factor of the backup burners. Specifically, the EPA is amending the definition of non-fossil fuel-fired EGUs from EGUs capable of “combusting 50 percent or more non-fossil fuel” to EGUs capable of “ deriving 50 percent or more of the heat input from non-fossil fuel at the base load rating ” (emphasis added). The definition of base load rating is also being amended to include the heat input from non-combustion sources ( e.g., solar thermal).

Revising “combusting” to “deriving” in the amended non-fossil fuel applicability language ensures that 40 CFR part 60, subparts TTTT and TTTTa, cover the fossil fuel-fired EGUs that the original rule was intended to cover, while minimizing unnecessary costs to EGUs fueled primarily by steam generated without combustion ( e.g., thermal energy supplied through the use of solar thermal collectors). The corresponding change in the base load rating to include the heat input from non-combustion sources is necessary to determine the relative heat input from fossil fuel and non-fossil fuel sources.

ii. Industrial EGUs

(A) Applicability to Industrial EGUs

In simple terms, the current applicability provisions in 40 CFR part 60, subpart TTTT, require that an EGU be capable of combusting more than 250 MMBtu/h of fossil fuel and be capable of selling 25 MW to a utility distribution system to be subject to 40 CFR part 60, subpart TTTT. These applicability provisions exclude industrial EGUs. However, the definition of an EGU also includes “integrated equipment that provides electricity or useful thermal output.” This language facilitates the integration of non-emitting generation and avoids energy inputs from non-affected facilities being used in the emission calculation without also considering the emissions of those facilities ( e.g., an auxiliary boiler providing steam to a primary boiler). This language could result in certain large processes being included as part of the EGU and meeting the applicability criteria. For example, the high-temperature exhaust from an industrial process ( e.g., calcining kilns, dryer, metals processing, or carbon black production facilities) that consumes fossil fuel could be sent to a HRSG to produce electricity. If the industrial process uses more than 250 MMBtu/h heat input and the electric sales exceed the applicability criteria, then the unit could be subject to 40 CFR part 60, subpart TTTT or TTTTa. This is potentially problematic for multiple reasons. First, it is difficult to determine the useful output of the EGU ( i.e., HRSG) since part of the useful output is included in the industrial process. In addition, the fossil fuel that is combusted could have a relatively high CO₂ emissions rate on a lb/MMBtu basis, making it potentially problematic to meet the standard of performance using efficient generation. This could result in the owner/operator reducing the electric output of the industrial facility to avoid the applicability criteria. Finally, the compliance costs associated with 40 CFR part 60, subpart TTTT or TTTTa, could discourage the development of environmentally beneficial projects.

To avoid these outcomes, the EPA is, as proposed, amending the applicability provision that exempts EGUs where greater than 50 percent of the heat input is derived from an industrial process that does not produce any electrical or mechanical output or useful thermal output that is used outside the affected EGU. Reducing the output or not developing industrial electric generating projects where the majority of the heat input is derived from the industrial process itself would not necessarily result in reductions in GHG emissions from the industrial facility. However, the electricity that would have been produced from the industrial project could still be needed. Therefore, projects of this type provide significant environmental benefit by providing additional useful output with little if any additional environmental impact. Including these types of projects would result in regulatory burden without any associated environmental benefit and could discourage project development, leading to potential overall increases in GHG emissions.

(B) Industrial EGUs Electric Sales Threshold Permit Requirement

The current electric sales applicability exemption in 40 CFR part 60, subpart TTTT, for non-CHP steam generating units includes the provision that EGUs have “ always been subject to a federally enforceable permit limiting annual net electric sales to one-third or less of their potential electric output ( e.g., limiting hours of operation to less than 2,920 hours annually) or limiting annual electric sales to 219,000 MWh or less” (emphasis added). The justification for this restriction includes that the 40 CFR part 60, subpart Da, applicability language includes “constructed for the purpose of . . .” and the Agency concluded that the intent was defined by permit conditions (80 FR 64544; October 23, 2015). This applicability criterion is important both for determining applicability with the new source CAA section 111(b) requirements and for determining whether existing steam generating units are subject to the existing source CAA section 111(d) requirements. For steam generating units that commenced construction after September 18, 1978, the applicability of 40 CFR part 60, subpart Da, would be relatively clear as to what criteria pollutant NSPS is applicable to the facility. However, for steam generating units that commenced construction prior to September 18, 1978, or where the owner/operator determined that criteria pollutant NSPS applicability was not critical to the project ( e.g., emission controls were sufficient to comply with either the EGU or industrial boiler criteria pollutant NSPS), owners/operators might not have requested that an electric sales permit restriction be included in the operating permit. Under the current applicability language, some onsite EGUs could be covered by the existing source CAA section 111(d) requirements even if they have never sold electricity to the grid. To avoid covering these industrial EGUs, the EPA proposed and is finalizing amendments to the electric sales exemption in 40 CFR part 60, subparts TTTT and TTTTa, to read, “annual net electric sales have never exceeded one-third of its potential electric output or 219,000 MWh, whichever is greater, and is [the “always been” would be deleted] subject to a federally enforceable permit limiting annual net electric sales to one-third or less of their potential electric output ( e.g., limiting hours of operation to less than 2,920 hours annually) or limiting annual electric sales to 219,000 MWh or less” (emphasis added). EGUs that reduce current generation will continue to be covered as long as they sold more than one-third of their potential electric output at some time in the past. The revisions make it possible for an owner/operator of an existing industrial EGU to provide evidence to the Administrator that the facility has never sold electricity in excess of the electricity sales threshold and to modify their permit to limit sales in the future. Without the amendment, owners/operators of any non-CHP industrial EGU capable of selling 25 MW would be subject to the existing source CAA section 111(d) requirements even if they have never sold any electricity. Therefore, the EPA is eliminating the requirement that existing industrial EGUs must have always been subject to a permit restriction limiting net electric sales.

iii. Determination of the Design Efficiency

The design efficiency ( i.e., the efficiency of converting thermal energy to useful energy output) of a combustion turbine is used to determine the electric sales applicability threshold. In 40 CFR part 60, subpart TTTT, the sales criteria are based in part on the individual EGU design efficiency. Three methods for determining the design efficiency are currently provided in 40 CFR part 60, subpart TTTT. Since the 2015 NSPS was finalized, the EPA has become aware that owners/operators of certain existing EGUs do not have records of the original design efficiency. These units would not be able to readily determine whether they meet the applicability criteria (and would therefore be subject to CAA section 111(d) requirements for existing sources) in the same way that 111(b) sources would be able to determine if the facility meets the applicability criteria. Many of these EGUs are CHP units that are unlikely to meet the 111(b) applicability criteria and would therefore not be subject to any future 111(d) requirements. However, the language in the 2015 NSPS would require them to conduct additional testing to demonstrate this. The requirement would result in burden to the regulated community without any environmental benefit. The electricity generating market has changed, in some cases dramatically, during the lifetime of existing EGUs, especially concerning ownership. As a result of acquisitions and mergers, original EGU design efficiency documentation, as well as performance guarantee results that affirmed the design efficiency, may no longer exist. Moreover, such documentation and results may not be relevant for current EGU efficiencies, as changes to original EGU configurations, upon which the original design efficiencies were based, render those original design efficiencies moot, meaning that there would be little reason to maintain former design efficiency documentation since it would not comport with the efficiency associated with current EGU configurations. As the three specified methods would rely on documentation from the original EGU configuration performance guarantee testing, and results from that documentation may no longer exist or be relevant, it is appropriate to allow other means to demonstrate EGU design efficiency. To reduce potential future compliance burden, the EPA proposed and is finalizing in 40 CFR part 60, subparts TTTT and TTTTa, to allow alternative methods as approved by the Administrator on a case-by-case basis. Owners/operators of EGUs can petition the Administrator in writing to use an alternate method to determine the design efficiency. The Administrator's discretion is intentionally left broad and can extend to other American Society of Mechanical Engineers (ASME) or International Organization for Standardization (ISO) methods as well as to operating data to demonstrate the design efficiency of the EGU. The EPA also proposed and is finalizing a change to the applicability of paragraph 60.8(b) in table 3 of 40 CFR part 60, subpart TTTT, from “no” to “yes” and that the applicability of paragraph 60.8(b) in table 3 of 40 CFR part 60, subpart TTTTa, is “yes.” This allows the Administrator to approve alternatives to the test methods specified in 40 CFR part 60, subparts TTTT and TTTTa.

c. Applicability for 40 CFR Part 60, Subpart TTTTa

This section describes applicability criteria that are only incorporated into 40 CFR part 60, subpart TTTTa, and that differ from the requirements in 40 CFR part 60, subpart TTTT.

Section 111 of the CAA defines a new or modified source for purposes of a given NSPS as any stationary source that commences construction or modification after the publication of the proposed regulation. Thus, the standards of performance apply to EGUs that commence construction or reconstruction after the date of proposal of this rule—May 23, 2023. EGUs that commenced construction after the date of the proposal for the 2015 NSPS and by May 23, 2023, will remain subject to the standards of performance promulgated in the 2015 NSPS. A modification is any physical change in, or change in the method of operation of, an existing source that increases the amount of any air pollutant emitted to which a standard applies. The NSPS general provisions (40 CFR part 60, subpart A) provide that an existing source is considered a new source if it undertakes a reconstruction.

The EPA is finalizing the same applicability requirements in 40 CFR part 60, subpart TTTTa, as the applicability requirements in 40 CFR part 60, subpart TTTT. The stationary combustion turbine must meet the following applicability criteria: The stationary combustion turbine must: (1) be capable of combusting more than 250 MMBtu/h (260 gigajoules per hour (GJ/h)) of heat input of fossil fuel (either alone or in combination with any other fuel); and (2) serve a generator capable of supplying more than 25 MW net to a utility distribution system ( i.e., for sale to the grid). In addition, the EPA proposed and is finalizing in 40 CFR part 60, subpart TTTTa, to include applicability exemptions for stationary combustion turbines that are: (1) capable of deriving 50 percent or more of the heat input from non-fossil fuel at the base load rating and subject to a federally enforceable permit condition limiting the annual capacity factor for all fossil fuels combined of 10 percent (0.10) or less; (2) combined heat and power units subject to a federally enforceable permit condition limiting annual net electric sales to no more than 219,000 MWh or the product of the design efficiency and the potential electric output, whichever is greater; (3) serving a generator along with other steam generating unit(s), IGCC, or stationary combustion turbine(s) where the effective generation capacity is 25 MW or less; (4) municipal waste combustors that are subject to 40 CFR part 60, subpart Eb; (5) commercial or industrial solid waste incineration units subject to 40 CFR part 60, subpart CCCC; and (6) deriving greater than 50 percent of heat input from an industrial process that does not produce any electrical or mechanical output that is used outside the affected stationary combustion turbine.

The EPA proposed the same requirements to combustion turbines in non-continental areas ( i.e., Hawaii, the Virgin Islands, Guam, American Samoa, the Commonwealth of Puerto Rico, and the Northern Mariana Islands) and non-contiguous areas (non-continental areas and Alaska) as the EPA did for comparable units in the contiguous 48 states. However, the Agency solicited comment on whether owners/operators of new and reconstructed combustion turbines in non-continental and non-contiguous areas should be subject to different requirements. Commenters generally commented that due to the difference in non-contiguous areas relative to the lower 48 states, the proposed requirements should not apply to owners/operators of new or reconstructed combustion turbines in non-contiguous areas. The Agency has considered these comments and is finalizing that only the initial BSER component will be applicable to owners/operators of combustion turbines located in non-contiguous areas. Therefore, owners/operators of base load combustions turbines would not be subject to the CCS-based numerical standards in 2032 and would continue to comply with the efficiency-based numeric standard. Based on information reported in the 2022 EIA Form EIA-860, there are no planned new combustion turbines in either Alaska or Hawaii. In addition, since 2015 no new combustion turbines have commenced operation in Hawaii. Two new combustion turbine facilities totaling 190 MW have commenced operation in Alaska since 2015. One facility is a combined cycle CHP facility and the other is at an industrial facility and neither facility would likely meet the applicability of 40 CFR part 60, subpart TTTTa. Therefore, not finalizing phase-2 BSER for non-continental and non-contiguous areas will have limited, if any, impacts on emissions or costs. The EPA notes that the Agency has the authority to amend this decision in future rulemakings.

i. Applicability to CHP Units

For 40 CFR part 60, subpart TTTT, owners/operators of CHP units calculate net electric sales and net energy output using an approach that includes “at least 20.0 percent of the total gross or net energy output consists of electric or direct mechanical output.” It is unlikely that a CHP unit with a relatively low electric output ( i.e., less than 20.0 percent) would meet the applicability criteria. However, if a CHP unit with less than 20.0 percent of the total output consisting of electricity were to meet the applicability criteria, the net electric sales and net energy output would be calculated the same as for a traditional non-CHP EGU. Even so, it is not clear that these CHP units would have less environmental benefit per unit of electricity produced than would more traditional CHP units. For 40 CFR part 60, subpart TTTTa, the EPA proposed and is finalizing to eliminate the restriction that CHP units produce at least 20.0 percent electrical or mechanical output to qualify for the CHP-specific method for calculating net electric sales and net energy output.

In the 2015 NSPS, the EPA did not issue standards of performance for certain types of sources—including industrial CHP units and CHPs that are subject to a federally enforceable permit limiting annual net electric sales to no more than the unit's design efficiency multiplied by its potential electric output, or 219,000 MWh or less, whichever is greater. For CHP units, the approach in 40 CFR part 60, subpart TTTT, for determining net electric sales for applicability purposes allows the owner/operator to subtract the purchased power of the thermal host facility. The intent of the approach is to determine applicability similarly for third-party developers and CHP units owned by the thermal host facility. However, as written in 40 CFR part 60, subpart TTTT, each third-party CHP unit would subtract the entire electricity use of the thermal host facility when determining its net electric sales. It is clearly not the intent of the provision to allow multiple third-party developers that serve the same thermal host to all subtract the purchased power of the thermal host facility when determining net electric sales. This would result in counting the purchased power multiple times. In addition, it is not the intent of the provision to allow a CHP developer to provide a trivial amount of useful thermal output to multiple thermal hosts and then subtract all the thermal hosts' purchased power when determining net electric sales for applicability purposes. The EPA proposed and is finalizing in 40 CFR part 60, subpart TTTTa, to limit to the amount of thermal host purchased power that a third-party CHP developer can subtract for electric sales when determining net electric sales equivalent to the percentage of useful thermal output provided to the host facility by the specific CHP unit. This approach eliminates both circumvention of the intended applicability by sales of trivial amounts of useful thermal output and double counting of thermal host-purchased power.

Finally, to avoid potential double counting of electric sales, the EPA proposed and is finalizing that for CHP units determining net electric sales, purchased power of the host facility be determined based on the percentage of thermal power provided to the host facility by the specific CHP facility.

ii. Non-Natural Gas Stationary Combustion Turbines

There is currently an exemption in 40 CFR part 60, subpart TTTT, for stationary combustion turbines that are not physically capable of combusting natural gas ( e.g., those that are not connected to a natural gas pipeline). While combustion turbines not connected to a natural gas pipeline meet the general applicability of 40 CFR part 60, subpart TTTT, these units are not subject to any of the requirements. The EPA is not including in 40 CFR part 60, subpart TTTTa, the exemption for stationary combustion turbines that are not physically capable of combusting natural gas. As described in the standards of performance section, owners/operators of combustion turbines burning fuels with a higher heat input emission rate than natural gas would adjust the natural gas-fired emissions rate by the ratio of the heat input-based emission rates. The overall result is that new stationary combustion turbines combusting fuels with higher GHG emissions rates than natural gas on a lb CO₂ /MMBtu basis must maintain the same efficiency compared to a natural gas-fired combustion turbine and comply with a standard of performance based on the identified BSER.

2. Subcategorization

In this final rule, the EPA is continuing to include both simple and combined cycle turbines in the definition of a stationary combustion turbine, and like in prior rules for this source category, the Agency is finalizing three subcategories—low load, intermediate load, and base load combustion turbines. These subcategories are determined based on electric sales ( i.e., utilization) relative to the combustion turbines' potential electric output to an electric distribution network on both a 12-operating month and 3-year rolling average basis. The applicable subcategory is determined each operating month and a stationary combustion turbine can switch subcategories if the owner/operator changes the way the facility is operated. Subcategorization based on percent electric sales is a proxy for how a combustion turbine operates and for determining the BSER and corresponding emission standards. For example, low load combustion turbines tend to spend a relatively high percentage of operating hours starting and stopping. However, within each subcategory not all combustion turbines operate the same. Some low load combustion turbines operate with less starting and stopping, but in general, combustion turbines tend to operate with fewer starts and stops ( i.e., more steady-state hours of operation) with increasing percentages of electric sales. The BSER for each subcategory is based on representative operation of the combustion turbines in that subcategory and on what is achievable for the subcategory as a whole.

Subcategorization by electric sales is similar, but not identical, to subcategorizing by heat input-based capacity factors or annual hours of operation limits. The EPA has determined that, for NSPS purposes, electric sales is appropriate because it reflects operational limitations inherent in the design of certain units, and also that—given these differences—certain emission reduction technologies are more suitable for some units than for others. This subcategorization approach is also consistent with industry practice. For example, operating permits for simple cycle turbines often include annual operating hour limitations of 1,500 to 4,000 hours annually. When average hourly capacity factors ( i.e., duty cycles) are accounted for, these hourly restrictions are similar to annual capacity factor restrictions of approximately 15 percent and 40 percent, respectively. The owners or operators of these combustion turbines never intend for them to provide base load power. In contrast, operating permits do not typically restrict the number of hours of annual operation for combined cycle turbines, reflecting that these types of combustion turbines are intended to have the ability to provide base load power.

The EPA evaluated the operation of the three general combustion turbine technologies—combined cycle turbines, frame-type simple cycle turbines, and aeroderivative simple cycle turbines—when determining the subcategorization approach in this rulemaking. The EPA found that, at the same capacity factor, aeroderivative simple cycle turbines have more starts (including fewer operating hours per start) than either frame simple cycle turbines or combined cycle turbines. The maximum number of starts for aeroderivative simple cycle turbines occurs at capacity factors of approximately 30 percent and the maximum number of starts for frame simple cycle turbines and combined cycle turbines both occur at capacity factors of approximately 25 percent. In terms of the median hours of operation per start, the hours per starts increases exponentially with capacity factor for each type of combustion turbine. The rate of increase is greatest for combined cycle turbines with the run times per start increasing significantly at capacity factors of 40 and greater. This threshold roughly matches the subcategorization threshold for intermediate load and base load turbines in this final rule. As is discussed later in section VIII.F.3 and VIII.F.4, technology options including those related to efficiency and to post combustion capture are impacted by the way units operate and can be more effective for units with fewer stops and starts.

a. Legal Basis for Subcategorization

As noted in section V.C.1 of this preamble, CAA section 111(b)(2) provides that the EPA “may distinguish among classes, types, and sizes within categories of new sources for the purpose of establishing . . . standards [of performance].” The D.C. Circuit has held that the EPA has broad discretion in determining whether and how to subcategorize under CAA section 111(b)(2). Lignite Energy Council, 198 F.3d at 933. As also noted in section V.C.1 of this preamble, in prior CAA section 111 rules, the EPA has subcategorized on numerous bases, including, among other things, fuel type and load, i.e., utilization. In particular, as noted in section V.C.1 of this preamble, the EPA subcategorized on the basis of utilization—for base load and non-base load subcategories—in the 2015 NSPS for GHG emissions from combustion turbines, Standards of Performance for Greenhouse Gas Emissions From New, Modified, and Reconstructed Stationary Sources: Electric Utility Generating Units, 80 FR 64509 (October 23, 2015), and also in the NESHAP for Reciprocating Internal Combustion Engines; NSPS for Stationary Internal Combustion Engines,79 FR 48072-01 (August 15, 2014).

Subcategorizing combustion turbines based on utilization is appropriate because it recognizes the way differently designed combustion turbines actually operate. Project developers do not construct combined cycle combustion turbine system to start and stop often to serve peak demand. Similarly, project developers do not construct and install simple cycle combustion turbines to operate at higher capacity factors to provide base load demand. And intermediate load demand may be served by higher efficiency simple cycle turbine systems or by “quick start” combined cycle units. Thus, there are distinguishing features ( i.e., different classes, types, and sizes) of turbines that are predominantly used in each of the utilization-based subcategories. Further, the amount of utilization and the mode of operation are relevant for the systems of emission reduction that the EPA may evaluate to be the BSER and therefore for the resulting standards of performance. See section VII.C.2.a.i for more discussion of the legal basis to subcategorize based upon characteristics relevant to the controls the EPA may determine to be the BSER.

As noted in sections VIII.E.2.b and VIII.F of this preamble, combustion turbines that operate at low load have highly variable operation and therefore highly variable emission rates. This variability made it challenging for the EPA to specify a BSER based on efficient design and operation and limits the BSER for purposes of this rulemaking to lower-emitting fuels. The EPA notes that the subcategorization threshold and the standard of performance are related. For example, the Agency could have finalized a lower electric sales threshold for the low load subcategory ( e.g., 15 percent) and evaluated the emission rates at the lower capacity factors. In future rulemaking the Agency could further evaluate the costs and emissions impacts of reducing the threshold for combustion turbines subject to a BSER based on the use of lower emitting fuels.

Intermediate load combustion turbines ( i.e., those that operate at loads that are somewhat higher than the low load peaking units) are most often designed to be simple cycle units rather than combined cycle units. This is because combustion turbines operating in the intermediate load range also start and stop and vary their load frequently (though not as often as low load peaking units). Because of the more frequent starts and stops, simple cycle combustion turbines are more economical for project developers when compared to combined cycle combustion turbines. Utilization of CCS technology is not practicable for those simple cycle units due to the lack of a HRSG. Therefore, the EPA has determined that efficient design and operation is the BSER for intermediate load combustion turbines.

While use of CCS is practicable for combined cycle combustion turbines, it is most appropriate for those units that operate at relatively higher loads ( i.e., as base load units) that do not frequently start, stop, and change load. Moreover, with current technology, CCS works better on units running at base load levels.

b. Electric Sales Subcategorization (Low, Intermediate, and Base Load Combustion Turbines)

As noted earlier, in the 2015 NSPS, the EPA established separate standards of performance for new and reconstructed natural gas-fired base load and non-base load stationary combustion turbines. The electric sales threshold distinguishing the two subcategories is based on the design efficiency of individual combustion turbines. A combustion turbine qualifies as a non-base load turbine—and is thus subject to a less stringent standard of performance—if it has net electric sales equal to or less than the design efficiency of the turbine (not to exceed 50 percent) multiplied by the potential electric output (80 FR 64601; October 23, 2015). If the net electric sales exceed that level on both a 12-operating month and 3-calendar year basis, then the combustion turbine is in the base load subcategory and is subject to a more stringent standard of performance. Subcategory applicability can change on a month-to-month basis since applicability is determined each operating month. For additional discussion on this approach, see the 2015 NSPS (80 FR 64609-12; October 23, 2015). The 2015 NSPS non-base load subcategory is broad and includes combustion turbines that assure grid reliability by providing electricity during periods of peak electric demand. These peaking turbines tend to have low annual capacity factors and sell a small amount of their potential electric output. The non-base load subcategory in the 2015 NSPS also includes combustion turbines that operate at intermediate annual capacity factors and are not considered base load EGUs. These intermediate load EGUs provide a variety of services, including providing dispatchable power to support variable generation from renewable sources of electricity. The need for this service has been expanding as the amount of electricity from wind and solar continues to grow. In the 2015 NSPS, the EPA determined the BSER for the non-base load subcategory to be the use of lower-emitting fuels ( e.g., natural gas and Nos. 1 and 2 fuel oils). In 2015, the EPA explained that efficient generation did not qualify as the BSER due in part to the challenge of determining an achievable output-based CO₂ emissions rate for all combustion turbines in this subcategory.

In this action, the EPA proposed and is finalizing the subcategories in 40 CFR part 60, subpart TTTTa, that will be applicable to sources that commence construction or reconstruction after May 23, 2023. First, the Agency proposed and is finalizing the definition of design efficiency so that the heat input calculation of an EGU is based on the higher heating value (HHV) of the fuel instead of the lower heating value (LHV), as explained immediately below. This has the effect of lowering the calculated potential electric output and the electric sales threshold. In addition, the EPA proposed and is finalizing division of the non-base load subcategory into separate intermediate and low load subcategories.

i. Higher Heating Value as the Basis for Calculation of the Design Efficiency

The heat rate is the amount of energy used by an EGU to generate 1 kWh of electricity and is often provided in units of Btu/kWh. As the thermal efficiency of a combustion turbine EGU is increased, less fuel is burned per kWh generated and there is a corresponding decrease in emissions of CO₂ and other air pollutants. The electric energy output as a fraction of the fuel energy input expressed as a percentage is a common practice for reporting the unit's efficiency. The greater the output of electric energy for a given amount of fuel energy input, the higher the efficiency of the electric generation process. Lower heat rates are associated with more efficient power generating plants.

Efficiency can be calculated using the HHV or the LHV of the fuel. The HHV is the heating value directly determined by calorimetric measurement of the fuel in the laboratory. The LHV is calculated using a formula to account for the moisture in the combustion gas ( i.e., subtracting the energy required to vaporize the water in the flue gas) and is a lower value than the HHV. Consequently, the HHV efficiency for a given EGU is always lower than the corresponding LHV efficiency because the reported heat input for the HHV is larger. For U.S. pipeline natural gas, the HHV heating value is approximately 10 percent higher than the corresponding LHV heating value and varies slightly based on the actual constituent composition of the natural gas. The EPA default is to reference all technologies on a HHV basis, and the Agency is basing the heat input calculation of an EGU on HHV for purposes of the definition of design efficiency. However, it should be recognized that manufacturers of combustion turbines typically use the LHV to express the efficiency of combustion turbines.

Similarly, the electric energy output for an EGU can be expressed as either of two measured values. One value relates to the amount of total electric power generated by the EGU, or gross output. However, a portion of this electricity must be used by the EGU facility to operate the unit, including compressors, pumps, fans, electric motors, and pollution control equipment. This within-facility electrical demand, often referred to as the parasitic load or auxiliary load, reduces the amount of power that can be delivered to the transmission grid for distribution and sale to customers. Consequently, electric energy output may also be expressed in terms of net output, which reflects the EGU gross output minus its parasitic load.

When using efficiency to compare the effectiveness of different combustion turbine EGU configurations and the applicable GHG emissions control technologies, it is important to ensure that all efficiencies are calculated using the same type of heating value ( i.e., HHV or LHV) and the same basis of electric energy output ( i.e., MWh-gross or MWh-net). Most emissions data are available on a gross output basis and the EPA is finalizing output-based standards based on gross output. However, to recognize the superior environmental benefit of minimizing auxiliary/parasitic loads, the Agency is including optional equivalent standards on a net output basis. To convert from gross to net output-based standards, the EPA used a 2 percent auxiliary load for simple and combined cycle turbines and a 7 percent auxiliary load for combined cycle EGUs using 90 percent CCS.

ii. Lowering the Threshold Between the Base Load and Non-Base Load Subcategories

The subpart TTTT distinction between a base load and non-base load combustion turbine is determined by the unit's actual electric sales relative to its potential electric sales, assuming the EGU is operated continuously ( i.e., percent electric sales). Specifically, stationary combustion turbines are categorized as non-base load and are subsequently subject to a less stringent standard of performance if they have net electric sales equal to or less than their design efficiency (not to exceed 50 percent) multiplied by their potential electric output (80 FR 64601; October 23, 2015). Because the electric sales threshold is based in part on the design efficiency of the EGU, more efficient combustion turbine EGUs can sell a higher percentage of their potential electric output while remaining in the non-base load subcategory. This approach recognizes both the environmental benefit of combustion turbines with higher design efficiencies and provides flexibility to the regulated community. In the 2015 NSPS, it was unclear how often high-efficiency simple cycle EGUs would be called upon to support increased generation from variable renewable generating resources. Therefore, the Agency determined it was appropriate to provide maximum flexibility to the regulated community. To do this, the Agency based the numeric value of the design efficiency, which is used to calculate the electric sales threshold, on the LHV efficiency. This had the impact of allowing combustion turbines to sell a greater share of their potential electric output while remaining in the non-base load subcategory.

The EPA proposed and is finalizing that the design efficiency in 40 CFR part 60, subpart TTTTa be based on the HHV efficiency instead of LHV efficiency and to not include the 50 percent maximum and 33 percent minimum restrictions. When determining the potential electric output used in calculating the electric sales threshold in 40 CFR part 60, subpart TTTT, design efficiencies of greater than 50 percent are reduced to 50 percent and design efficiencies of less than 33 percent are increased to 33 percent for determining electric sales threshold subcategorization criteria. The 50 percent criterion was established to limit non-base load EGUs from selling greater than 55 percent of their potential electric sales. The 33 percent criterion was included to be consistent with applicability thresholds in the electric utility criteria pollutant NSPS (40 CFR part 60, subpart Da).

Neither of those criteria are appropriate for 40 CFR part 60, subpart TTTTa, and the EPA proposed and is finalizing a decision that they are not incorporated when determining the electric sales threshold. Instead, as discussed later in the section, the EPA is finalizing a fixed percent electric sales thresholds and the design efficiency does not impact the subcategorization thresholds. However, the design efficiency is still used when determining the potential electric sales and any restriction on using the actual design efficiency of the combustion turbine would have the impact of changing the threshold. If this restriction were maintained, it would reduce the regulatory incentive for manufacturers to invest in programs to develop higher efficiency combustion turbines.

The EPA also proposed and is finalizing a decision to eliminate the 33 percent minimum design efficiency in the calculation of the potential electric output. The EPA is unaware of any new combustion turbines with design efficiencies meeting the general applicability criteria of less than 33 percent; and this will likely have no cost or emissions impact.

The EPA solicited comment on whether the intermediate/base load electric sales threshold should be reduced further to a range that would lower the base load electric sales threshold for simple cycle turbines to between 29 to 35 percent (depending on the design efficiency) and to between 40 to 49 percent for combined cycle turbines (depending on the design efficiency). The specific approach the EPA solicited comment on was reducing the design efficiency by 6 percent ( e.g., multiplying by 0.94) when determining the electric sales threshold. Some commenters supported lowering the proposed electric sales threshold while others supported maintaining the proposed standards.

After considering comments, in 40 CFR part 60, subpart TTTTa, the EPA has determined it is appropriate to eliminate the sliding scale electric sales threshold based on the design efficiency and instead base the subcategorization thresholds on fixed electric sales (also referred to sometimes here as capacity factor). In 40 CFR part 60 subpart TTTTa, the EPA is finalizing that the fixed electric sales threshold between intermediate load combustion turbines and base load combustion turbines is 40 percent. The 40 percent electric sales (capacity factor) threshold reflects the maximum capacity factor for intermediate load simple cycle turbines and the minimum prorated efficiency approach for base load combined cycle turbines that the EPA solicited comment on in proposal.

The base load electric sales threshold is appropriate for new combustion turbines because, as will be discussed later, the first component of BSER for base load turbines is based on highly efficient combined cycle generation. Combined cycle units are significantly more efficient than simple cycle turbines; and therefore, in general, the EPA should be focusing its determination of the BSER for base load units on that more efficient technology. The electric sales thresholds and the emission standards are related because, at lower capacity factors, combustion turbines tend to have more variable operation ( e.g., more starts and stops and operation at part load conditions) that reduces the efficiency of the combustion turbine. This is particularly the case for combined cycle turbines because while the turbine engine can come to full load relatively quickly, the HRSG and steam turbine cannot, and combined cycle turbines responding to highly variable load will have efficiencies similar to simple cycle turbines. This has implications for the appropriate control technologies and corresponding emission reduction potential. The EPA determined the final standard of performance based on review of emissions data for recently installed combined cycle combustion turbines with 12-operating month capacity factors of 40 percent or greater. The EPA considered a capacity factor threshold lower than 40 percent. However, expanding the subcategory to include combustion turbines with a 12-operating month electric sales of less than 40 percent would require the EPA to consider the emissions performance of combined cycle turbines operating at lower capacity factors and, while it would expand the number of sources in the base load subcategory, it would also result in a higher ( i.e., less stringent) numerical emission standard for the sources in the category.

Direct comparison of the costs of combined cycle turbines relative to simple cycle turbines can be challenging because model plant costs are often for combustion turbines of different sizes and do not account for variable operation. For example, combined cycle turbine model plants are generally for an EGU that is several hundred megawatts while simple cycle turbine model plants are generally less than a hundred megawatts. Direct comparison of the LCOE from these model plants is not relevant because the facilities are not comparable. Consider a facility with a block of 10 simple cycle turbines that are each 50 MW (so the overall facility capacity is 500 MW). Each simple cycle turbine operates as an individual unit and provides a different value to the electric grid as compared to a single 500 MW combined cycle turbine. While the minimum load of the combined cycle facility might be 200 MW, the block of 10 simple cycle turbines can provide from approximately 20 MW to 500 MW to the electric grid.

A more accurate cost comparison accounts for economies of scale and estimates the cost of a combined cycle turbine with the same net output as a simple cycle turbine. Comparing the modeled LCOE of these combustion turbines provides a meaningful comparison, at least for base load combustion turbines. Without accounting for economies of scale and variable operation, combined cycle turbines can appear to be more cost effective than simple cycle turbines under almost all conditions. In addition, without accounting for economies of scale, large frame simple cycle turbines can appear to be more cost effective than higher efficiency aeroderivative simple cycle turbines, even if operated at a 100 percent capacity factor. These cost models are not intended to make direct comparisons, and the EPA appropriately accounted for economies of scale when estimating the cost of the BSER. Since base load combustion turbines tend to operate under steady state conditions with few starts and stops, startup and shutdown costs and the efficiency impact of operating at variable loads are not important for determining the compliance costs of base load combustion turbines.

Based on an adjusted model plant comparison, combined cycle EGUs have a lower LCOE at capacity factors above approximately 40 percent compared to simple cycle EGUs operating at the same capacity factors. This supports the final base load fixed electric sales threshold of 40 percent for simple cycle turbines because it would be cost-effective for owners/operators of simple cycle turbines to add heat recovery if they elected to operate at higher capacity factors as a base load unit. Furthermore, based on an analysis of monthly emission rates, recently constructed combined cycle EGUs maintain consistent emission rates at capacity factors of less than 55 percent (which is the base load electric sales threshold in subpart TTTT) relative to operation at higher capacity factors. Therefore, the base load subcategory operating range can be expanded in 40 CFR part 60, subpart TTTTa, without impacting the stringency of the numeric standard. However, at capacity factors of less than approximately 40 percent, emission rates of combined cycle EGUs increase relative to their operation at higher capacity factors. It takes much longer for a HRSG to begin producing steam that can be used to generate additional electricity than it takes a combustion engine to reach full power. Under operating conditions with a significant number of starts and stops, typical of some intermediate and especially low load combustion turbines, there may not be enough time for the HRSG to generate steam that can be used for additional electrical generation. To maximize overall efficiency, combined cycle EGUs often use combustion turbine engines that are less efficient than the most efficient simple cycle turbine engines. Under operating conditions with frequent starts and stops where the HRSG does not have sufficient time to begin generating additional electricity, a combined cycle EGU may be no more efficient than a highly efficient simple cycle EGU. These distinctions in operation are thus meaningful for determining which emissions control technologies are most appropriate for types of units. Once a combustion turbine unit exceeds approximately 40 percent annual capacity factor, it is economical to add a HRSG which results in the unit becoming both more efficient and less likely to cycle its operation. Such units are, therefore, better suited for more stringent emission control technologies including CCS.

After the 2015 NSPS was finalized, some stakeholders expressed concerns about the approach for distinguishing between base load and non-base load turbines. They posited a scenario in which increased utilization of wind and solar resources, combined with low natural gas prices, would create the need for certain types of simple cycle turbines to operate for longer time periods than had been contemplated when the 2015 NSPS was being developed. Specifically, stakeholders have claimed that in some regional electricity markets with large amounts of variable renewable generation, some of the most efficient new simple cycle turbines—aeroderivative turbines—could be called upon to operate at capacity factors greater than their design efficiency. However, if those new simple cycle turbines were to operate at those higher capacity factors, they would become subject to the more stringent standard of performance for base load turbines. As a result, according to these stakeholders, the new aeroderivative turbines would have to curtail their generation and instead, less-efficient existing turbines would be called upon to run by the regional grid operators, which would result in overall higher emissions. The EPA evaluated the operation of simple cycle turbines in areas of the country with relatively large amounts of variable renewable generation and did not find a strong correlation between the percentage of generation from the renewable sources and the 12-operating month capacity factors of simple cycle turbines. In addition, most of the simple cycle turbines that commenced operation between 2010 and 2016 (the most recent simple cycle turbines not subject to 40 CFR part 60, subpart TTTT) have operated well below the base load electric sales threshold in 40 CFR part 60, subpart TTTT. Therefore, the Agency does not believe that the concerns expressed by stakeholders necessitates any revisions to the regulatory scheme. In fact, as noted above, the EPA is finalizing that the electric sales threshold can be lowered without impairing the availability of simple cycle turbines where needed, including to support the integration of variable generation. The EPA believes that the final threshold is not overly restrictive since a simple cycle turbine could operate on average for more than 9 hours a day in the intermediate load subcategory.

iii. Low and Intermediate Load Subcategories

This section discusses the EPA's rationale for subcategorizing non-base load combustion turbines into two subcategories—low load and intermediate load.

(A) Low Load Subcategory

The EPA proposed and is finalizing in 40 CFR part 60, subpart TTTTa, a low load subcategory to includes combustion turbines that operate only during periods of peak electric demand ( i.e., peaking units), which will be separate from the intermediate load subcategory. Low load combustion turbines also provide ramping capability and other ancillary services to support grid reliability. The EPA evaluated the operation of recently constructed simple cycle turbines to understand how they operate and to determine at what electric sales level or capacity factor their emissions rate is relatively steady. (Note that for purposes of this discussion, the terms “electric sales” and “capacity factor” are used interchangeably.) Low load combustion turbines generally only operate for short periods of time and potentially at relatively low duty cycles. This type of operation reduces the efficiency and increases the emissions rate, regardless of the design efficiency of the combustion turbine or how it is maintained. For this reason, it is difficult to establish a reasonable output-based standard of performance for low load combustion turbines.

To determine the electric sales threshold—that is, to distinguish between the intermediate load and low load subcategories—the EPA evaluated capacity factor electric sales thresholds of 10 percent, 15 percent, 20 percent, and 25 percent. The EPA proposed to find and is finalizing a conclusion that the 10 percent threshold is problematic for two reasons. First, simple cycle turbines operating at that level or lower have highly variable emission rates, and therefore it is difficult for the EPA to establish a meaningful output-based standard of performance. In addition, only one-third of simple cycle turbines that have commenced operation since 2015 have maintained 12-operating month capacity factors of less than 10 percent. Therefore, setting the threshold at this level would bring most new simple cycle turbines into the intermediate load subcategory, which would subject them to a more stringent emission rate that is only achievable for simple cycle turbines operating at higher capacity factors. This could create a situation where simple cycle turbines might not be able to comply with the intermediate load standard of performance while operating at the low end of the intermediate load capacity factor subcategorization criteria.

Based on the EPA's review of hourly emissions data, at a capacity factor above 15 percent, GHG emission rates for many simple cycle turbines begin to stabilize. At higher capacity factors, more time is typically spent at steady state operation rather than ramping up and down; and emission rates tend to be lower while in steady-state operation. Of recently constructed simple cycle turbines, half have maintained 12-operating month capacity factors of 15 percent or less, two-thirds have maintained capacity factors of 20 percent or less; and approximately 80 percent have maintained maximum capacity factors of 25 percent or less. The emission rates clearly stabilize for most simple cycle turbines operating at capacity factors of greater than 20 percent. Based on this information, the EPA proposed the low load electric sales threshold—again, the dividing line to distinguish between the intermediate and low load subcategories—to be 20 percent and solicited comment on a range of 15 to 25 percent. The EPA also solicited comment on whether the low load electric sales threshold should be determined by a site-specific threshold based on three-fourths of the design efficiency of the combustion turbine. Under this approach, simple cycle turbines selling less than 18 to 22 percent of their potential electric output (depending on the design efficiency) would still have been considered low load combustion turbines. This “sliding scale” electric sales threshold approach is like the approach the EPA used in the 2015 NSPS to recognize the environmental benefit of installing the most efficient combustion turbines for low load applications. Using this approach, combined cycle EGUs would have been able to sell between 26 to 31 percent of their potential electric output while still being considered low load combustion turbines. Some commenters supported a lower electric sales threshold while others supported a higher threshold. Based on these comments, the EPA is finalizing the proposed low load electric sales threshold of 20 percent of the potential electric sales. The fixed 20 percent capacity factor threshold represents a level of utilization at which most simple cycle combustion turbines perform at a consistent level of efficiency and GHG emission performance, enabling the EPA to establish a standard of performance that reflects a BSER of efficient operation. The 20 percent capacity factor threshold is also more environmentally protective than the higher thresholds the EPA considered, since owners and operators of combustion turbines operating above a 20 percent capacity factor would be subject to an output-based emissions standard instead of a heat input-based emissions standard based on the use of lower-emitting fuels. This ensures that owners/operators of intermediate load combined cycle turbines properly maintain and operate their combustion turbines.

(B) Intermediate Load Subcategory

The proposed sliding scale subcategorization approach essentially included two subcategories within the proposed intermediate load subcategory. As proposed, simple cycle turbines would be classified as intermediate load combustion turbines when operated between capacity factors of 20 percent and approximately 40 percent while combined cycle turbines would be classified as intermediate load combustion turbines when operated between capacity factors of 20 percent to approximately 55 percent. Owners/operators of combined cycle turbines operating at the high end of the intermediate load subcategory would only be subject to an emissions standard based on a BSER of high-efficiency simple cycle turbine technology. The proposed approach provided a regulatory incentive for owners/operators to purchase the most efficient technologies in exchange for additional compliance flexibility. The use of a prorated efficiency the EPA solicited comment on would have lowered the simple cycle and combined cycle turbine thresholds to approximately 35 percent and 50 percent, respectively.

In this final rule, the BSER for the intermediate load subcategory is consistent with the proposal—high-efficiency simple cycle turbine technology. While not specifically identified in the proposal, the BSER for the base load subcategory is also consistent with the proposal—the use of combined cycle technology.

The 12-operating month electric sales ( i.e., capacity factor) thresholds for the stationary combustion turbine subcategories in this final rule are summarized below in Table 2.

iv. Integrated Onsite Generation and Energy Storage

Integrated equipment is currently included as part of the affected facility, and the EPA proposed and is finalizing amended regulatory text to clarify that the output from integrated renewables is included as output when determining the NSPS emissions rate. The EPA also proposed that the output from the integrated renewable generation is not included when determining the net electric sales for applicability purposes ( i.e., generation from integrated renewables would not be considered when determining if a combustion turbine is subcategorized as a low, intermediate, or base load combustion turbine). In the alternative, the EPA solicited comment on whether instead of exempting the generation from the integrated renewables from counting toward electric sales, the potential output from the integrated renewables would be included when determining the design efficiency of the facility. Since the design efficiency is used when determining the electric sales threshold this would increase the allowable electric sales for subcategorization purposes. Including the integrated renewables when determining the design efficiency of the affected facility has the impact of increasing the operational flexibility of owners/operators of combustion turbines. Commenters generally supported maintaining that integrated renewables are part of the affected facility and including the output of the renewables when determining the emissions rate of the affected facility. Therefore, the Agency is finalizing a decision that the rated output of integrated renewables be included when determining the design efficiency of the affected facility, which is used to determine the potential electric output of the affected facility, and that the output of the integrated renewables be included in determining the emissions rate of the affected facility. However, since the design efficiency is not a factor in determining the subcategory thresholds in 40 CFR part 60, subpart TTTTa, the output of the integrated renewables will not be included for determining the applicable subcategory. If the output from the integrated renewable generation were included for subcategorization purposes, this could discourage the use of integrated renewables (or curtailments) because affected facilities could move to a subcategory with a more stringent emissions standard that could cause the owner/operator to be out of compliance. The impact of this approach is that the electric sales threshold of the combustion turbine island itself, not including the integrated renewables, for an owner/operator of a combustion turbine that includes integrated renewables that increase the potential electric output by 1 percent would be 1 or 2 percent higher for the stationary combustion turbine island not considering the integrated renewables, depending on the design efficiency of the combustion turbine itself, than an identical combustion turbine without integrated renewables. In addition, when the output from the integrated renewables is considered, the output from the integrated renewables lowers the emissions rate of the affected facility by approximately 1 percent.

For integrated energy storage technologies, the EPA solicited comment on and is finalizing a decision to include the rated output of the energy storage when determining the design efficiency of the affected facility. Similar to integrated renewables, this increases the flexibility of owner/operators to sell larger amounts of electricity while remaining in the low, variable, and intermediate load subcategories. While energy storage technologies have high capital costs, operating costs are low and would dispatch prior to the combustion turbine the technology is integrated with. Therefore, simple cycle turbines with integrated energy storage would likely operate at lower capacity factors than an identical simple cycle turbine at the same location. However, while the energy storage might be charged with renewables that would otherwise be curtailed, there is no guarantee that low emitting generation would be used to charge the energy storage. Therefore, the output from the energy storage is not considered in either determining the NSPS emissions rate or as net electric sales for subcategorization applicability purposes. In future rulemaking the Agency could further evaluate the impact of integrated energy storage on the operation of simple cycle turbines to determine if the number of starts and stops are reduced and increases the efficiency of simple cycle turbines relative to simple cycle turbines without integrated energy storage. If this is the case, it could be appropriate to lower the threshold for combustion turbines subject to a lower emitting fuels BSER because emission rates would be stable at lower capacity factors.

v. Definition of System Emergency

In 2015, the EPA included a provision that electricity sold during hours of operation when a unit is called upon due to a system emergency is not counted toward the percentage electric sales subcategorization threshold in 40 CFR part 60, subpart TTTT. The Agency concluded that this exclusion is necessary to provide flexibility, maintain system reliability, and minimize overall costs to the sector. The intent is that the local grid operator will determine the EGUs essential to maintaining grid reliability. Subsequent to the 2015 NSPS, members of the regulated community informed the EPA that additional clarification of a system emergency is needed to determine and document generation during system emergencies. The EPA proposed to include the system emergency approach in 40 CFR part 60, subpart TTTTa, and solicited comment on amending the definition of system emergency to clarify in implementation in 40 CFR part 60, subparts TTTT and TTTTa. Commenters generally agreed with the proposal to allow owners/operators of EGUs called upon during a system emergency to operate without impacting the EGUs' subcategorization ( i.e., electric sales during system emergencies would not be considered when determining net electric sales), and that the Agency should clarify how system emergencies are determined and documented.

In terms of the definition of the system emergency provision, commenters stated that “abnormal” be deleted from the definition, and instead of referencing “the Regional Transmission Organizations (RTO), Independent System Operators (ISO) or control area Administrator,” the definition should reference “the balancing authority or reliability coordinator.” This change would align the regulation's definition with the terms used by NERC. Some commenters also stated that the EPA should specify that electric sales during periods the grid operator declares energy emergency alerts (EEA) levels 1 through 3 be included in the definition of system emergency. In addition, some commenters stated that the definition should be expanded to include the concept of energy emergencies. Specifically, the definition should also exempt generation during periods when a load-serving entity or balancing authority has exhausted all other resource options and can no longer meet its expected load obligations. Finally, commenters stated that the definition should apply to all EGUs, regardless of if they are already operating when the system emergency is declared. This would avoid regulatory incentive to come offline prior to a potential system emergency to be eligible for the electric sales exemption and would treat all EGUs similarly during system emergencies ( i.e., not penalize EGUs that are already operating to maintain grid reliability and avoiding the need to declare grid emergencies).

The Agency is including the system emergency concept in 40 CFR part 60, subpart TTTTa, along with a definition that clarifies how to determine generation during periods of system emergencies. The EPA agrees with commenters that the definition of system emergency should be clarified and that it should not be limited to EGUs not operating when the system emergency is declared. Based on information provided by entities with reliability expertise, the EPA has determined that a system emergency should be defined to include EEA levels 2 and 3. These EEA levels generally correspond to time-limited, well-defined, and relatively infrequent situations in which the system is experiencing an energy deficiency. During EEA level 2 and 3 events, all available generation is online and demand-response or other load management procedures are in effect, or firm load interruption is imminent or in progress. The EPA believes it is appropriate to exclude hours of operation during such events in order to ensure that EGUs are not impeded from maintaining or increasing their output as needed to respond to a declared energy emergency. Because these events tend to be short, infrequent, and well-defined, the EPA also believes any incremental GHG emissions associated with operations during these periods would be relatively limited.

The EPA has determined not to include EEA level 1 in the definition of a “system emergency.” The EPA's understanding is that EEA level 1 events often include situations in which an energy deficiency does not yet exist, and in which balancing authorities are preparing to pursue various options for either bringing additional resources online or managing load. The EPA also understands that EEA level 1 events tend to be more frequently declared, and longer in duration, than level 2 or 3 events. Based on this information, the EPA believes that including EEA level 1 events in the definition of a “system emergency” would carry a greater risk of increasing overall GHG emissions without making a meaningful contribution to supporting reliability. This approach balances the need to have operational flexibility when the grid may be strained to help ensure that all available generating sources are available for grid reliability, while balancing with important considerations about potential GHG emission tradeoffs. The EPA is also amending the definition in 40 CFR part 60, subpart TTTT, to be consistent with the definition in 40 CFR part 60, subpart TTTTa.

Commenters also added that operation during system emergencies should be subject to alternate standards of performance ( e.g., owners/operators are not required to use the CCS system during system emergencies to increase power output). The EPA agrees with commenters that since system emergencies are defined and historically rare events, an alternate standard of performance should apply during these periods. Carbon capture systems require significant amounts of energy to operate. Allowing owners/operators of EGUs equipped with CCS systems to temporarily reduce the capture rate or cease capture will increase the electricity available to end users during system emergencies. In place of the applicable output-based emissions standard, the owner/operator of an intermediate or base load combustion turbine would be subject to a BSER based on the combustion of lower-emitting fuels during system emergencies. The emissions and output would not be included when calculating the 12-operating month emissions rate. The EPA considered an alternate emissions standard based on efficient generation but rejected that for multiple reasons. First, since system emergencies are limited in nature the emissions calculation would include a limited number of hours and would not necessarily be representative of an achievable longer-term emissions rate. In addition, EGUs that are designed to operate with CCS will not necessarily operate as efficiently without the CCS system operating compared to a similar EGU without a CCS system. Therefore, the Agency is not able to determine a reasonable efficiency-based alternate emissions standard for periods of system emergencies. Due to both the costs and time associated with starting and stopping the CCS system, the Agency has determined it is unlikely that an owner/operator of an affected facility would use it where it is not needed. System emergencies have historically been relatively brief and any hours of operation outside of the system emergencies are included when determining the output-based emissions standard. During short-duration system emergencies, the costs associated with stopping and starting the CCS system could outweigh the increased revenue from the additional electric sales. In addition, the time associated with starting and stopping a CCS system would likely result in an EGU operating without the CCS system in operation during periods of non-system emergencies. This would require the owner/operator to overcontrol during other periods of operation to maintain emissions below the applicable standard of performance. Therefore, it is likely an owner/operator would unnecessarily adjust the operation of the CCS system during EEA levels 2 and 3.

In addition to these measures, DOE has authority pursuant to section 202(c) of the Federal Power Act to, on its own motion or by request, order, among other things, the temporary generation of electricity from particular sources in certain emergency conditions, including during events that would result in a shortage of electric energy, when the Secretary of Energy determines that doing so will meet the emergency and serve the public interest. An affected source operating pursuant to such an order is deemed not to be operating in violation of its environmental requirements. Such orders may be issued for 90 days and may be extended in 90-day increments after consultation with the EPA. DOE has historically issued section 202(c) orders at the request of electric generators and grid operators such as RTOs in order to enable the supply of additional generation in times of expected emergency-related generation shortfalls.

c. Multi-Fuel-Fired Combustion Turbines

In 40 CFR part 60, subpart TTTT, multi-fuel-fired combustion turbines are subcategorized as EGUs that combust 10 percent or more of fuels not meeting the definition of natural gas on a 12-operating month rolling average basis. The BSER for this subcategory is the use of lower-emitting fuels with a corresponding heat input-based standard of performance of 120 to 160 lb CO₂ /MMBtu, depending on the fuel, for newly constructed and reconstructed multi-fuel-fired stationary combustion turbines. Lower-emitting fuels for these units include natural gas, ethylene, propane, naphtha, jet fuel kerosene, Nos. 1 and 2 fuel oils, biodiesel, and landfill gas. The definition of natural gas in 40 CFR part 60, subpart TTTT, includes fuel that maintains a gaseous state at ISO conditions, is composed of 70 percent by volume or more methane, and has a heating value of between 35 and 41 megajoules (MJ) per dry standard cubic meter (dscm) (950 and 1,100 Btu per dry standard cubic foot). Natural gas typically contains 95 percent methane and has a heating value of 1,050 Btu/lb. A potential issue with the multi-fuel subcategory is that owners/operators of simple cycle turbines can elect to burn 10 percent non-natural gas fuels, such as Nos. 1 or 2 fuel oil, and thereby remain in that subcategory, regardless of their electric sales. As a result, they would remain subject to the less stringent standard that applies to multi-fuel-fired sources, the lower-emitting fuels standard. This could allow less efficient combustion turbine designs to operate as base load units without having to improve efficiency and could allow EGUs to avoid the need for efficient design or best operating and maintenance practices. These potential circumventions would result in higher GHG emissions.

To avoid these outcomes, the EPA proposed and is finalizing a decision not to include the multi-fuel subcategory for low, intermediate, and base load combustion turbines in 40 CFR part 60, subpart TTTTa. This means that new multi-fuel-fired turbines that commence construction or reconstruction after May 23, 2023, will fall within a particular subcategory depending on their level of electric sales. The EPA also proposed and is finalizing a decision that the performance standards for each subcategory be adjusted appropriately for multi-fuel-fired turbines to reflect the application of the BSER for the subcategories to turbines burning fuels with higher GHG emission rates than natural gas. To be consistent with the definition of lower-emitting fuels in the 2015 NSPS, the maximum allowable heat input-based emissions rate is 160 lb CO₂ /MMBtu. For example, a standard of performance based on efficient generation would be 33 percent higher for a fuel oil-fired combustion turbine compared to a natural gas-fired combustion turbine. This assures that the BSER, in this case efficient generation, is applied, while at the same time accounting for the use of multiple fuels.

d. Rural Areas and Small Utility Distribution Systems

As part of the original proposal and during the Small Business Advocacy Review (SBAR) outreach the EPA solicited comment on creating a subcategory for rural electric cooperatives and small utility distribution systems (serving 50,000 customers or less). Commenters expressed concerns that a BSER based on either co-firing hydrogen or CCS may present an additional hardship on economically disadvantaged communities and on small entities, and that the EPA should evaluate potential increased energy costs, transmission upgrade costs, and infrastructure encroachment which may directly affect the disproportionately impacted communities. As described in section VIII.F, the BSER for new stationary combustion turbines does not include hydrogen co-firing and CCS qualifies as the BSER for base load combustion turbines on a nationwide basis. Therefore, the EPA has determined that a subcategory for rural cooperatives and/or small utility distribution systems is not appropriate.

F. Determination of the Best System of Emission Reduction (BSER) for New and Reconstructed Stationary Combustion Turbines

In this section, the EPA describes the technologies it proposed as the BSER for each of the subcategories of new and reconstructed combustion turbines that commence construction after May 23, 2023, as well as topics for which the Agency solicited comment. In the following section, the EPA describes the technologies it is determining are the final BSER for each of the three subcategories of affected combustion turbines and explains its basis for selecting those controls, and not others, as the final BSER. The controls that the EPA evaluated included combusting non-hydrogen lower-emitting fuels ( e.g., natural gas and distillate oil), using highly efficient generation, using CCS, and co-firing with low-GHG hydrogen.

For the low load subcategory, the EPA proposed the use of lower-emitting fuels as the BSER. This was consistent with the BSER and performance standards established in the 2015 NSPS for the non-base load subcategory as discussed earlier in section VIII.C.

For the intermediate load subcategory, the EPA proposed an approach under which the BSER was made up of two components: (1) highly efficient generation; and (2) co-firing 30 percent (by volume) low-GHG hydrogen. Each component of the BSER represented a different set of controls, and those controls formed the basis of corresponding standards of performance that applied in two phases. Specifically, the EPA proposed that affected facilities ( i.e., facilities that commence construction or reconstruction after May 23, 2023) could apply the first component of the BSER ( i.e., highly efficient generation) upon initial startup to meet the first phase of the standard of performance. Then, by 2032, the EPA proposed that affected facilities could apply the second component of the BSER ( i.e., co-firing 30 percent (by volume) low-GHG hydrogen) to meet a second and more stringent standard of performance. The EPA also solicited comment on whether the intermediate load subcategory should apply a third component of the BSER: co-firing 96 percent (by volume) low-GHG hydrogen by 2038. In addition, the EPA solicited comment on whether the low load subcategory should also apply the second component of BSER, co-firing 30 percent (by volume) low-GHG hydrogen, by 2032. The Agency proposed that these latter components of the BSER would continue to include the application of highly efficient generation.

For the base load subcategory, the EPA also proposed a multi-component BSER and multi-phase standard of performance. The EPA proposed that each new base load combustion turbine would be required to meet a phase-1 standard of performance based on the application of the first component of the BSER—highly efficient generation—upon initial startup of the affected source. For the second component of the BSER, the EPA proposed two potential technology pathways for base load combustion turbines with corresponding standards of performance. One proposed technology pathway was 90 percent CCS, which base load combustion turbines would install and begin to operate by 2035 to meet the phase-2 standard of performance. A second proposed technology pathway was co-firing low-GHG hydrogen, which base load combustion turbines would implement in two steps: (1) By co-firing 30 percent (by volume) low-GHG hydrogen to meet the phase-2 standard of performance by 2032, and (2) by co-firing 96 percent (by volume) low-GHG hydrogen to meet a phase 3 standard of performance by 2038. Throughout, the Agency proposed base load turbines, like intermediate load turbines, would remain subject to the first component of the BSER based on highly efficient generation.

The proposed approach reflected the EPA's view that the BSER components for the intermediate load and base load subcategories could achieve deeper reductions in GHG emissions by implementing CCS and co-firing low-GHG hydrogen. This proposed approach also recognized that building the infrastructure required to support widespread use of CCS and low-GHG hydrogen technologies in the power sector will take place on a multi-year time scale. Accordingly, new and reconstructed facilities would be aware of their need to ramp toward more stringent phases of the standards, which would reflect application of the more stringent controls in the BSER. This would occur either by co-firing a lower percentage (by volume) of low-GHG hydrogen by 2032 and a higher percentage (by volume) of low-GHG hydrogen by 2038, or with installation and use of CCS by 2035. The EPA also solicited comment on the potential for an earlier compliance date for the second phase.

For the base load subcategory, the EPA proposed two potential BSER pathways because the Agency believed there was more than one viable technology for these combustion turbines to significantly reduce their CO₂ emissions. The Agency also found value in receiving comments on, and potentially finalizing, both BSER pathways to enable project developers to elect how they would reduce their CO₂ emissions on timeframes that make sense for each BSER pathway. The EPA solicited comment on whether the co-firing of low-GHG hydrogen should be considered a compliance pathway for sources to meet a single standard of performance based on the application of CCS rather than a separate BSER pathway. The EPA proposed that there would be earlier opportunities for units to begin co-firing lower amounts of low-GHG hydrogen than to install and begin operating 90 percent CCS systems. However, the Agency proposed that it would likely take longer for those units to increase their co-firing to significant quantities of low-GHG hydrogen. Therefore, in the proposal, the EPA presented the BSER pathways as separate subcategories and solicited comment on the option of finalizing a single standard of performance based on the application of CCS.

For the low load subcategory, the EPA proposed and is finalizing that the BSER is the use of lower-emitting fuels. For the intermediate load subcategory, the EPA proposed and is finalizing that the BSER is highly efficient generating technology—simple cycle technology as well as operating and maintaining it efficiently. The EPA is not finalizing a second component of the BSER or a phase-2 standard of performance for intermediate load combustion turbines at this time. For the base load subcategory, the EPA proposed and is finalizing that the first component of the BSER is highly efficient generating technology—combined cycle technology as well as operating and maintaining it efficiently. The EPA proposed and is finalizing a second component of the BSER or a phase-2 standard of performance for base load combustion turbines—efficient generation in combination with 90 percent CCS.

The EPA is not finalizing low-GHG hydrogen co-firing as the second component of the BSER for the intermediate load or base load combustion turbines at this time. (See section VIII.F.5.b for the EPA's explanation of this decision.) With respect to the CCS pathway for base load combustion turbines, the EPA is finalizing a second phase of the standards of performance that includes a single CCS BSER pathway, which includes the use of highly efficient generation and 90 percent CCS. Owners/operators of new and reconstructed base load combustion turbines will be required to meet the second phase standards of performance for 12-operating month rolling averages that begin on or after January 2032, that reflect application of both the phase-1 and phase-2 components of the BSER. Table 3 of this document summarizes the final BSER for combustion turbine EGUs that commence construction or reconstruction after May 23, 2023. The EPA is finalizing standards of performance based on those BSER for each subcategory, as discussed in section VIII.G.

1. BSER for Low Load Subcategory

This section describes the BSER for the low load ( i.e., peaking) subcategory at this time, which is the use of lower-emitting fuels. The Agency proposed and is finalizing a determination that the use of lower-emitting fuels, which the EPA determined to be the BSER for the non-base load subcategory in the 2015 NSPS, is the BSER for this low load subcategory. As explained in section VIII.E.2.b, the EPA is narrowing the definition of the low load subcategory by lowering the electric sales threshold (as compared to the electric sales threshold for non-base load combustion turbines in the 2015 NSPS), so that combustion turbines with higher electric sales would be placed in the intermediate load subcategory and therefore be subject to a more stringent standard based on the more stringent BSER.

a. Background: The Non-Base Load Subcategory in the 2015 NSPS

The 2015 NSPS defined non-base load natural gas-fired EGUs as stationary combustion turbines that (1) burn more than 90 percent natural gas and (2) have net electric sales equal to or less than their design efficiency (not to exceed 50 percent) multiplied by their potential electric output (80 FR 64601; October 23, 2015). These are calculated on 12-operating month and 3-calendar year rolling average bases. The EPA also determined in the 2015 NSPS that the BSER for newly constructed and reconstructed non-base load natural gas-fired stationary combustion turbines is the use of lower-emitting fuels. Id. at 64515. These lower-emitting fuels are primarily natural gas with a small allowance for distillate oil ( i.e., Nos. 1 and 2 fuel oils), which have been widely used in stationary combustion turbine EGUs for decades.

The EPA also determined in the 2015 NSPS that the standard of performance for sources in this subcategory is a heat input-based standard of 120 lb CO₂ /MMBtu. The EPA established this clean-fuels BSER for this subcategory because of the variability in the operation in non-base load combustion turbines and the challenges involved in determining a uniform output-based standard that all new and reconstructed non-base load units could achieve.

Specifically, in the 2015 NSPS, the EPA recognized that a BSER for the non-base load subcategory based on the use of lower-emitting fuels results in limited GHG reductions, but further recognized that an output-based standard of performance could not reasonably be applied to the subcategory. The EPA explained that a combustion turbine operating at a low capacity factor could operate with multiple starts and stops, and that its emission rate would be highly dependent on how it was operated and not its design efficiency. Moreover, combustion turbines with low annual capacity factors typically operated differently from each other, and therefore had different emission rates. The EPA recognized that, as a result, at the time it would not be possible to determine a standard of performance that could reasonably apply to all combustion turbines in the subcategory. For that reason, the EPA further recognized, efficient design and operation would not qualify as the BSER; rather, the BSER should be lower-emitting fuels and the associated standard of performance should be based on heat input. Since the 2015 NSPS, all newly constructed simple cycle turbines have been non-base load units and thus have become subject to this standard of performance.

b. BSER

Consistent with the rationale of the 2015 NSPS, the EPA proposed and is finalizing that the use of fuels with an emissions rate of less than 160 lb CO₂ /MMBtu ( i.e., lower-emitting fuels) meets the BSER requirements for the low load subcategory at this time. Use of these fuels is technically feasible for combustion turbines. Natural gas comprises the majority of the heat input for simple cycle turbines and is the lowest cost fossil fuel. In the 2015 NSPS, the EPA determined that natural gas comprised 96 percent of the heat input for simple cycle turbines. See 80 FR 64616 (October 23, 2015). Therefore, a BSER based on the use of natural gas and/or distillate oil would have minimal, if any, costs to regulated entities. The use of lower-emitting fuels would not have any significant adverse energy requirements or non-air quality or environmental impacts, as the EPA determined in the 2015 NSPS. Id. at 64616. In addition, the use of fuels meeting this criterion would result in some emission reductions by limiting the use of fuels with higher carbon content, such as residual oil, as the EPA also explained in the 2015 NSPS. Id. Although the use of fuels meeting this criterion would not advance technology, in light of the other reasons described here, the EPA proposed and is finalizing that the use of natural gas, Nos. 1 and 2 fuel oils, and other fuels currently specified in 40 CFR part 60, subpart TTTT, qualify as the BSER for new and reconstructed combustion turbine EGUs in the low load subcategory at this time. The EPA also proposed including low-GHG hydrogen on the list of fuels meeting the uniform fuels criteria in 40 CFR part 60, subpart TTTTa. The EPA is finalizing the inclusion of hydrogen, regardless of the production pathway, on the list of fuels meeting the uniform fuels criteria in 40 CFR part 60, subpart TTTTa. The addition of hydrogen (and fuels derived from hydrogen) to 40 CFR part 60, subpart TTTTa, simplifies the recordkeeping and reporting requirements for low load combustion turbines that elect to burn hydrogen.

For the reasons discussed in the 2015 NSPS and noted above, the EPA did not propose that efficient design and operation qualify as the BSER for the low load subcategory. The emissions rate of a low load combustion turbine is highly dependent upon the way the specific combustion turbine is operated. For example, a combustion turbine with multiple startups and shutdowns and operation at part loads will have high emissions relative to if it were operated at steady-state high-load conditions. Important characteristics for reducing GHG emissions from low load combustion turbines are the ability to minimize emissions during periods of startup and shutdown and efficient operation at part loads and while changing loads. If the combustion turbine is frequently operated at part-load conditions with frequent starts and stops, a combustion turbine with a high design efficiency, which is determined at full-load steady-state conditions, would not necessarily emit at a lower GHG rate than a combustion turbine with a lower design efficiency. In addition, combustion turbines with higher design efficiencies have higher initial costs compared to combustion turbines with lower design efficiencies. Since the EPA does not have sufficient information at this time to determine emission reduction for the subcategory it is not possible to determine the cost effectiveness of a BSER based on high efficiency simple cycle turbines.

The EPA solicited comment on whether, and the extent to which, high-efficiency designs also operate more efficiently at part loads and can start more quickly and reach the desired load more rapidly than combustion turbines with less efficient design efficiencies. In addition, the EPA solicited comment on the cost premium of high-efficiency simple cycle turbines. To the extent the Agency received additional relevant information, the EPA was considering promulgating design standard requirements pursuant to CAA section 111(h). However, the EPA did not receive comments that changed the proposal conclusions.

The EPA did not propose the use of CCS or hydrogen co-firing as the BSER (or as a component of the BSER) for low load combustion turbines. The EPA did not propose that CCS is the BSER for simple cycle turbines based on the Agency's assessment that currently available post-combustion amine-based carbon capture systems require that the exhaust from a combustion turbine be cooled prior to entering the carbon capture equipment. The most energy efficient way to cool the exhaust gas is to use a HRSG, which is an integral component of a combined cycle turbine system but is not incorporated in a simple cycle unit. For this reason and due to the high costs of CCS for low load combustion turbines, the Agency did not propose and is not finalizing a determination that CCS qualifies as the BSER for this subcategory of sources.

The EPA did not propose low-GHG hydrogen co-firing as the BSER for low load combustion turbines because not all new combustion turbines can necessarily co-fire higher percentages of hydrogen, there are potential infrastructure issues specific to low load combustion turbines, and at the relatively infrequent levels of utilization that characterize the low load subcategory, a low-GHG hydrogen co-firing BSER would not necessarily result in cost-effective GHG reductions for all low load combustion turbines. As discussed later in this section, the Agency is not determining that low-GHG hydrogen co-firing qualifies as the BSER for combustion turbines. In future rulemaking the Agency could further evaluate the costs and emissions performance of other technologies to reduce emissions from low-load units to determine if other technologies qualify as the BSER.

2. BSER for Intermediate Load Subcategory

This section describes the BSER for new and reconstructed combustion turbines in the intermediate load subcategory. For combustion turbines in the intermediate load subcategory, the BSER is the use of high-efficiency simple cycle turbine technology in combination with the best operating and maintenance practices.

a. Lower-Emitting Fuels

The EPA did not propose and is not finalizing lower-emitting fuels as the BSER for intermediate load combustion turbines because, as described earlier in this section, it would achieve few GHG emission reductions compared to highly efficient generation.

b. Highly Efficient Generation

This section includes a discussion of the various highly efficient generation technologies used by owners/operators of combustion turbines. The appropriate technology depends on how the combustion turbine is operated, and the EPA has determined it does not have sufficient information to determine an appropriate output-based emissions standard for low load combustion turbines. At higher capacity factors, emission rates for simple cycle combustion turbines are more consistent, and the EPA has sufficient information to determine a BSER other than lower-emitting fuels.

The use of highly efficient generating technology in combination with the best operating and maintenance practices has been demonstrated by multiple facilities for decades. Notably, over time, as technologies have improved, what is considered highly efficient has changed as well. Highly efficient generating technology is available and offered by multiple vendors for both simple cycle and combined cycle turbines. Both types of combustion turbines can also employ best operating and maintenance practices, which include routine operating and maintenance practices that minimize fuel use.

For simple cycle turbines, manufacturers continue to improve the efficiency by increasing firing temperature, increasing pressure ratios, using intercooling on the air compressor, and adopting other measures. These improved designs allow for improved operating efficiencies and reduced emission rates. Design efficiencies of simple cycle turbines range from 33 to 40 percent. Best operating practices for simple cycle turbines include proper maintenance of the combustion turbine flow path components and the use of inlet air cooling to reduce efficiency losses during periods of high ambient temperatures.

For combined cycle turbines, high-efficiency technology uses a highly efficient combustion turbine engine matched with a high-efficiency HRSG. The most efficient combined cycle EGUs use HRSG with three different steam pressures and incorporate a steam reheat cycle to maximize the efficiency of the Rankine cycle. It is not necessarily practical for owners/operators of combined cycle facilities using a turbine engine with an exhaust temperature below 593 °C or a steam turbine engine smaller than 60 MW to incorporate a steam reheat cycle. Smaller combustion turbine engines, less than those rated at approximately 2,000 MMBtu/h, tend to have lower exhaust temperatures and are paired with steam turbines of 60 MW or less. These smaller combined cycle units are limited to using a HRSG with three different steam pressures, but without a reheat cycle. This increases the heat rate of the combined cycle unit by approximately 2 percent. High efficiency also includes, but is not limited to, the use of the most efficient steam turbine and minimizing energy losses using insulation and blowdown heat recovery. Best operating and maintenance practices include, but are not limited to, minimizing steam leaks, minimizing air infiltration, and cleaning and maintaining heat transfer surfaces.

A potential drawback of combined cycle turbines with the highest design efficiencies is that the facility is relatively complicated and startup times can be relatively long. Combustion turbine manufacturers have invested in fast-start technologies that reduce startup times and improve overall efficiencies. According to the NETL Baseline Flexible Operation Report, while the design efficiencies are the same, the capital costs of fast-start combined cycle turbines are 1.6 percent higher than a comparable conventional start combined cycle facility. The additional costs include design parameters that significantly reduce start times. However, fast-start combined cycle turbines are still significantly less flexible than simple cycle turbines and generally do not serve the same role. The startup time to full load from a hot start takes a simple cycle turbine 5 to 8 minutes, while a combined cycle turbines ranges from 30 minutes for a fast-start combined cycle turbine to 90 minutes for a conventional start combined cycle turbine. The startup time to full load from a cold start takes a simple cycle turbine 10 minutes, while a combined cycle turbines ranges from 120 minutes for a fast-start combined cycle turbine to 250 minutes for a conventional start combined cycle turbine. In addition, fast-start combined cycle turbines require the use of an auxiliary boiler during warm and cold starts. In addition, minimum run times for simple cycle aeroderivative engines and combined cycle EGUs equal one minute and 120 minutes, respectively. Minimum downtime for the same group is five minutes and 60 minutes, respectively. Finally, simple cycle aeroderivative turbines have no limit to the number of starts per year. Combined cycle EGUs are limited in the number of starts, and additional maintenance costs will occur if the hours/start ratio drops below 25. The model combined cycle turbines in the NETL Baseline Flexible Operation Report use a HRSG with three different steam pressures and a reheat cycle. While the use of this type of HRSG increases design efficiencies at steady state conditions, it increases the capital costs and decreases the flexibility ( e.g., longer start times) of the combined cycle turbine. While less common, combined cycle turbines can be designed with a relatively simple HRSG that produces either a single or two pressures of steam without a reheat cycle. While design efficiencies are lower, the combined cycle turbines are more flexible and have the potential to operate similar to at least a portion of the simple cycle turbines in the intermediate load subcategory and provide the same value to the grid.

The EPA solicited comment on whether additional technologies for new simple and combined cycle EGUs that could reduce emissions beyond what is currently being achieved by the best performing EGUs should be included in the BSER. Specifically, the EPA sought comment on whether pressure gain combustion should be incorporated into a standard of performance based on an efficient generation BSER for both simple and combined cycle turbines. In addition, the EPA sought comment on whether the HRSG for combined cycle turbines should be designed to utilize supercritical steam conditions or to utilize supercritical CO₂ as the working fluid instead of water; whether useful thermal output could be recovered from a compressor intercooler and boiler blowdown; and whether fuel preheating should be implemented. Commenters generally noted that these technologies are promising, but that because the EPA did not sufficiently evaluate the BSER criteria in the proposal and none of these technologies should be incorporated as part of the BSER. The EPA continues to believe these technologies are promising, but the Agency is not including them as part of the BSER at this time.

The EPA also solicited comment on whether the use of steam injection is applicable to intermediate load combustion turbines. Steam injection is the use of a relatively simple and low-cost HRSG to produce steam, but instead of recovering the energy by expanding the steam through a steam turbine, the steam is injected into the compressor and/or through the fuel nozzles directly into the combustion chamber and the energy is extracted by the combustion turbine engine. Advantages of steam injection include improved efficiency and increased output of the combustion turbine as well as reduced NO_X emissions. Combustion turbines using steam injection have characteristics in-between simple cycle and combined cycle combustion turbines. They are more efficient, but more complex and have higher capital costs than simple cycle combustion turbines without steam injection. Conversely, compared to combined cycle EGUs, simple cycle combustion turbines using steam injection are simpler, have shorter construction times, and have lower capital costs, but have lower efficiencies. Combustion turbines using steam injection can start quickly, have good part-load performance, and can respond to rapid changes in demand, making the technology a potential solution for reducing GHG emissions from intermediate load combustion turbines. A potential drawback of steam injection is that the additional pressure drop across the HRSG can reduce the efficiency of the combustion turbine when the facility is running without the steam injection operating.

The EPA is aware of a limited number of combustion turbines that are using steam injection that have maintained 12-operating month emission rates of less than 1,000 lb CO₂ /MWh-gross. Commenters stated that steam injection does not qualify as the BSER because it has not been adequately demonstrated and the EPA did not include sufficient analysis of the technology in the proposal to determine it as the BSER for intermediate load combustion turbines. The EPA continues to believe the technology is promising and it may be used to comply with the standard of performance, but the Agency is not determining that it is the BSER for intermediate load combustion turbines at this time. In a potential future rulemaking, the Agency could further evaluate the costs and emissions performance of steam injection to determine if the technology qualifies as the BSER.

i. Adequately Demonstrated

The EPA proposed and is finalizing that highly efficient simple cycle designs are adequately demonstrated because highly efficient simple cycle turbines have been demonstrated by multiple facilities for decades, the efficiency improvements of the most efficient designs are incremental in nature and do not change in any significant way how the combustion turbine is operated or maintained, and the levels of efficiency that the EPA is proposing have been achieved by many recently constructed combustion turbines. Therefore, efficient generation technology described in this BSER is commercially available and the standards of performance are achievable.

ii. Costs

In general, advanced generation technologies enhance operational efficiency compared to lower efficiency designs. Such technologies present little incremental capital cost compared to other types of technologies that may be considered for new and reconstructed sources. In addition, more efficient designs have lower fuel costs, which offsets at least a portion of the increase in capital costs.

For the intermediate load subcategory, the EPA considers that the costs of high-efficiency simple cycle combustion turbines are reasonable. As described in the subcategory section, the cost of combustion turbine engines is dependent upon many factors, but the EPA estimates that that the capital cost of a high-efficiency simple cycle turbine is 10 percent more than a comparable lower efficiency simple cycle turbine. Assuming all other costs are the same and that the high-efficiency simple cycle turbine uses 8 percent less fuel, high-efficiency simple cycle combustion turbines have a lower LCOE compared to standard efficiency simple cycle combustion turbines at a 12-operating month capacity factor of approximately 31 percent. At a 20 percent and 15 percent capacity factors, the compliance costs are $1.5/MWh and $35/metric ton and $3.0/MWh and $69/metric ton, respectively. The EPA has determined that the incremental costs the use of high efficiency simple cycle turbines as the BSER for intermediate load combustion turbines is reasonable. The EPA notes that the approach the Agency used to estimate these costs have a relatively high degree of uncertainty and are likely high given the common use of high efficiency simple cycle turbines without a regulatory driver.

The EPA considered but is not finalizing combined cycle unit design for combustion turbines as the BSER for the intermediate load subcategory because it is unclear if combined cycle turbines could serve the same role as intermediate load simple cycle turbines as a whole. Specifically, the EPA does not have sufficient information to determine that an intermediate load combined cycle turbine can start and stop with enough flexibility to provide the same level of grid support as intermediate load simple cycle turbines as a whole. In addition, the amount of GHG reductions that could be achieved by operating combined cycle EGUs as intermediate load EGUs is unclear. Intermediate load combustion turbines start and stop so frequently that there would often not be sufficient periods of continuous operation where the HRSG would have sufficient time to generate steam to operate the steam turbine enough to significantly lower the emissions rate of the EGU.

Some commenters agreed with the proposed rationale of the EPA, and other commenters disagreed and said that combined cycle turbine technology is cost effective and lower-emitting than simple cycle turbine technology and therefore qualifies as the BSER for intermediate load combustion turbines. Commenters supporting combined cycle technology as the BSER submitted cost information that indicated that combined cycle EGUs have lower capital costs and LCOE than simple cycle turbines. However, the commenters compared capital costs of larger combined cycle turbines to smaller simple cycle turbines and did not account for economies of scale. The EPA has concluded that the appropriate cost comparison is for combustion turbines with the same rated net output. Comparing the costs of different size EGUs is not appropriate because these EGUs provide different grid services. In addition, the commenters did not account for startup costs and the time required for a steam turbine to begin operating when determining the LCOE.

The EPA considered the operation of simple cycle turbine to determine the potential for simple cycle turbine to add a HRSG while continuing to operate in the same manner, providing the same grid services, as current simple cycle turbines. As noted previously, aeroderivative simple cycle turbines have shorter run times per start than frame type simple cycle turbines at the same capacity factor. At an annual capacity factor of 20 percent, the median run time per start for aeroderivative and frame simple cycle turbines is 12 and 16 hours respectively. At an annual capacity factor of 30 percent, the average run times per start increase to 17 and 26 hours for aeroderivative and frame turbines respectively. The higher operating times of frame type simple cycle turbines, along with the larger size of frame type turbines, indicate that combined cycle technology could be applicable to at least a portion of intermediate load combustion turbines. In future rulemakings addressing GHGs from new as well as existing combustion turbines, the EPA intends to further evaluate the costs and potential emission reductions of the use of faster starting and lower cost HRSG technology for intermediate load combustion turbines to determine if the technology does in fact qualify as the BSER.

iii. Non-Air Quality Health and Environmental Impact and Energy Requirements

Use of highly efficient generation reduces all non-air quality health and environmental impacts and energy requirements assuming it displaces less efficient or higher-emitting generation. Even when operating at the same input-based emissions rate, the more efficient a unit is, the less fuel is required to produce the same level of output; and, as a result, emissions are reduced for all pollutants. The use of highly efficient combustion turbines, compared to the use of less efficient combustion turbines, reduces all pollutants. By the same token, because improved efficiency allows for more electricity generation from the same amount of fuel, it will not have any adverse effects on energy requirements.

Designating highly efficient generation as part of the BSER for new and reconstructed intermediate load combustion turbines will not have significant impacts on the nationwide supply of electricity, electricity prices, or the structure of the electric power sector. On a nationwide basis, the additional costs of the use of highly efficient generation will be small because the technology does not add significant costs and at least some of those costs are offset by reduced fuel costs. In addition, at least some of these new combustion turbines would be expected to incorporate highly efficient generation technology in any event.

iv. Extent of Reductions in CO₂ Emissions

The EPA estimated the potential emission reductions associated with a standard that reflects the application of highly efficient generation as BSER for the intermediate load subcategory. As discussed in section VIII.G.1, the EPA determined that the standards of performance reflecting this BSER are 1,170 lb CO₂ /MWh-gross for intermediate load combustion turbines.

Between 2015 and 2022, 113 simple cycle turbines, an average of 16 per year, commenced operation. Of these, 112 reported 12-operating month capacity factors. The EPA estimates that 23 simple cycle turbines operated at 12-operating month capacity factors greater than 20 percent and potentially would be considered intermediate combustion turbines. To estimate reductions, the EPA assumed that the number of simple cycle turbines constructed between 2015 and 2022 and the operation of those combustion turbines would continue on an annual basis. For each simple cycle turbine that operated at a capacity greater than 20 percent, the EPA determined the percent reduction in emissions, based on the maximum 12-operating months intermediate load emission rate, that would be required to comply with the final NSPS for intermediate load turbines. The EPA then applied that same percent reduction in emissions to the average operating capacity factor to determine the emission reductions from the NSPS. Using this approach, the EPA estimates that the intermediate load standard will impact approximately a quarter of new simple cycle turbines. The EPA divided the total amount of calculated reductions for intermediate load simple cycle turbines built between 2015 and 2022 and divided that value by 7 (the number of years evaluated) to get estimated annual reductions. This approach results in annual reductions of 31,000 tons of CO₂ as well as 8 tons of NO_X. The emission reductions are projected to result primarily from building additional higher efficiency aeroderivative simple cycle turbines instead of less efficient frame simple cycle turbines. The reduced emissions come from relatively small reductions in the emission rates of the intermediate load aeroderivative simple cycle turbines. This is a snapshot of projected emission reductions from applying the NSPS retroactively to 2022. If more intermediate load simple cycle turbines are built in the future, the emission reductions would be higher than this estimate. Conversely, if fewer intermediate load simple cycles are built, the emission reductions would be lower than the EPA's estimate.

Importantly, the “highly efficient generation” which the EPA has determined to be the BSER for new and reconstructed intermediate load combustion turbines and to be the first component BSER for base load stationary combustions, is not the same as the “heat rate improvements” (HRI, or “efficiency improvements”) that the EPA determined to be the BSER for existing coal-fired steam generating EGUs in the ACE Rule. As noted earlier in this document, the EPA has concluded that the suite of HRI in the ACE Rule is not an appropriate BSER for existing coal-fired EGUs. In the EPA's technical judgment, the suite of HRI set forth in the ACE Rule would provide negligible CO₂ reductions at best and, in many cases, may increase CO₂ emissions because of the “rebound effect,” which is explained and discussed in section VII.D.4.a.iii of this preamble. Increased CO₂ emissions from the “rebound effect” can occur when a coal-fired EGU improves its efficiency (heat rate), which can move the unit up on the dispatch order—resulting in an EGU operating for more hours during the year than it would have without having done the efficiency improvements. There is also the possibility that a more efficient coal-fired EGU could displace a lower emitting generating source, further exacerbating the problem.

Conversely, including “highly efficient generation” as a component of the BSER for new and reconstructed does not create this risk of displacing a lower-emitting generating source. A new highly efficient stationary combustion turbine may be dispatched more than it would have been if it were not built as a highly efficient turbine, but it is more likely to displace an existing coal-fired EGU or a less efficient existing stationary combustion turbine. It would be unlikely to displace a renewable generating source.

For base load stationary combustion turbines, “highly efficient generation” is the first component of the BSER—with 90 percent capture CCS being the second component of the BSER. This is very similar to the Agency's BSER determination for the NSPS for new fossil fuel-fired steam generating units. In that final rule, the EPA established standards of performance for newly constructed fossil fuel-fired steam generating units based on the performance of a new highly efficient supercritical pulverized coal (SCPC) EGU implementing post-combustion partial CCS technology, which the EPA determined to be the BSER for these sources.

v. Promotion of the Development and Implementation of Technology

The EPA also considered the potential impact of selecting highly efficient simple cycle generation technology as the BSER for the intermediate load subcategory in promoting the development and implementation of improved control technology. New highly efficient simple cycle turbines are more efficient than the average new simple cycle turbine and a standard based on the performance of the most efficient, best performing simple cycle turbine will promote penetration of the most efficient units throughout the industry. Accordingly, consideration of this factor supports the EPA's proposal to determine this technology to be the BSER.

c. Low-GHG Hydrogen and CCS

The EPA did not propose and is not finalizing either CCS or co-firing low-GHG hydrogen as the first component of the BSER for intermediate load combustion turbines, for the reasons given in sections VIII.F.4.c.iii (CCS) and VIII.F.5 (low-GHG hydrogen).

d. Summary of BSER Determinations

The EPA is finalizing that highly efficient generating technology in combination with the best operating and maintenance practices is the BSER for intermediate load combustion turbines. Specifically, the use of highly efficient simple cycle technology in combination with the best operating and maintenance practices is the BSER for intermediate load combustion turbines.

Highly efficient generation qualifies the BSER because it is adequately demonstrated, it can be implemented at reasonable cost, it achieves emission reductions, and it does not have significant adverse non-air quality health or environmental impacts or significant adverse energy requirements. The fact that it promotes greater use of advanced technology provides additional support; however, the EPA considers highly efficient generation to the BSER for intermediate load combustion turbines even without taking this factor into account.

3. BSER for Base Load Subcategory—First Component

This section describes the first component of the BSER for newly constructed and reconstructed combustion turbines in the base load subcategory. For combustion turbines in the base load subcategory, the first component of the BSER is the use of high-efficiency combined cycle technology in combination with the best operating and maintenance practices.

a. Lower-Emitting Fuels

The EPA did not propose and is not finalizing lower-emitting fuels as the BSER for base load combustion turbines because, as described earlier in this section, it would achieve few GHG emission reductions compared to highly efficient generation.

b. Highly Efficient Generation

i. Adequately Demonstrated

The EPA proposed and is finalizing that highly efficient combined cycle designs are adequately demonstrated because highly efficient combined cycle EGUs have been demonstrated by multiple facilities for decades, and the efficiency improvements of the most efficient designs are incremental in nature and do not change in any significant way how the combustion turbine is operated or maintained. Due to the differences in HRSG efficiencies for smaller combined cycle turbines, the EPA proposed and is finalizing less stringent standards of performance for smaller base load turbines with base load ratings of less than 2,000 MMBtu/h relative to those for larger base load turbines. The levels of efficiency that the EPA is proposing have been achieved by many recently constructed combustion turbines. Therefore, efficient generation technology described in this BSER is commercially available and the standards of performance are achievable.

ii. Costs

For the base load subcategory, the EPA considers the cost of high-efficiency combined cycle EGUs to be reasonable. While the capital costs of a higher efficiency combined cycle EGUs are 1.9 percent higher than standard efficiency combined cycle EGUs, fuel use is 2.6 percent lower. The reduction in fuel costs fully offset the capital costs at capacity factors of 40 percent or greater over the expected 30-year life of the facility. Therefore, a BSER based on the use of high-efficiency combined cycle combustion turbines for base load combustion turbines would have minimal, if any, overall compliance costs since the capital costs would be recovered through reduced fuel costs over the expected 30-year life of the facility.

iii. Non-Air Quality Health and Environmental Impact and Energy Requirements

Use of highly efficient generation reduces all non-air quality health and environmental impacts and energy requirements as compared to use of less efficient generation. Even when operating at the same input-based emissions rate, the more efficient a unit is, the less fuel is required to produce the same level of output; and, as a result, emissions are reduced for all pollutants. The use of highly efficient combustion turbines, compared to the use of less efficient combustion turbines, reduces all pollutants. By the same token, because improved efficiency allows for more electricity generation from the same amount of fuel, it will not have any adverse effects on energy requirements.

Designating highly efficient generation as part of the BSER for new and reconstructed base load combustion turbines will not have significant impacts on the nationwide supply of electricity, electricity prices, or the structure of the electric power sector. On a nationwide basis, the additional costs of the use of highly efficient generation will be small because the technology does not add significant costs and at least some of those costs are offset by reduced fuel costs. In addition, at least some of these new combustion turbines would be expected to incorporate highly efficient generation technology in any event.

iv. Extent of Reductions in CO₂ Emissions

The EPA used a similar approach to estimating emission reductions for base load combustion turbines as intermediate load combustion turbines, except the Agency reviewed recently constructed combined cycle EGUs. As discussed in section VIII.G.1, the EPA determined that the standard of performance reflecting this BSER is 800 lb CO₂ /MWh-gross for base load combustion turbines. The Agency assumed all new combined cycle turbines would be impacted by the base load emissions standard. Between the beginning of 2015 and the beginning of 2022, 129 combined cycle turbines, an average of 18 per year, commenced operation. Of those combined cycle turbines, 107 had 12-operating month emissions data. For each of these 107 combined cycle turbines that had a maximum 12-operating month emissions rate greater than 800 lb CO₂ /MWh-gross, the EPA determined the reductions that would occur assuming the combined cycle turbine reduced its emissions rate to 800 lb CO₂ /MWh-gross and continued to operate at its average capacity factor. The EPA summed the results and divided by 8 (the number of years evaluated) to estimate the annual GHG reductions that will result from this final rule. The EPA estimates that the base load standard will result in annual reductions of 313,000 tons of CO₂ as well as 23 tons of NO_X . The reductions increase each year and in year 3 the annual reductions would be 939,000 tons of CO₂ and 69 tons of NO_X.

v. Promotion of the Development and Implementation of Technology

The EPA also considered the potential impact of selecting highly efficient generation technology as the BSER in promoting the development and implementation of improved control technology. The highly efficient combustion turbines are more efficient and lower emitting than the average new combustion turbine generation technology. Determining that highly efficient turbines are a component of the BSER will advance penetration of the best performing combustion turbines throughout the industry—and will incentivize manufacturers to offer improved turbines that meet the final standard of performance associated with application of the BSER. Accordingly, consideration of this factor supports the EPA's proposal to determine this technology to be the BSER.

c. Low-GHG Hydrogen and CCS

The EPA did not propose and is not finalizing either CCS or co-firing low-GHG hydrogen as the first component of the BSER for base load combustion turbines, for the reasons given in sections VIII.F.4.c.iii (CCS) and VIII.F.5 (low-GHG hydrogen).

d. Summary of BSER Determinations

The EPA is finalizing that highly efficient generating technology in combination with the best operating and maintenance practices is the BSER for first component of the BSER for base load combustion turbines. The phase-1 standards of performance are based on the application of that technology. Specifically, the use of highly efficient combined cycle technology in combination with best operating and maintenance practices is the first component of the BSER for base load combustion turbines.

Highly efficient generation qualifies as the BSER because it is adequately demonstrated, it can be implemented at reasonable cost, it achieves emission reductions, and it does not have significant adverse non-air quality health or environmental impacts or significant adverse energy requirements. The fact that it promotes greater use of advanced technology provides additional support; however, the EPA considers highly efficient generation to be a component of the BSER for base load combustion turbines even without taking this factor into account.

4. BSER for Base Load Subcategory—Second Component

a. Authority To Promulgate a Multi-Part BSER and Standard of Performance

The EPA's approach of promulgating standards of performance that apply in multiple phases, based on determining the BSER to be a set of controls with multiple components, is consistent with CAA section 111(b). That provision authorizes the EPA to promulgate “standards of performance,” CAA section 111(b)(1)(B), defined, in the singular, as “a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of the [BSER].” CAA section 111(a)(1). CAA section 111(b)(1)(B) further provides, “[s]tandards of performance . . . shall become effective upon promulgation.” In this rulemaking, the EPA is determining that the BSER is a set of controls that, depending on the subcategory, include highly efficient generation plus use of CCS. The EPA is determining that affected sources can apply the first component of the BSER—highly efficient generation—by the effective date of the final rule and can apply both the first and second components of the BSER—highly efficient generation in combination with 90 percent CCS—in 2032.

Accordingly, the EPA is finalizing standards of performance that reflect the application of this multi-component BSER and that take the form of standards of performance that affected sources must comply with in two phases. This multi-phase standard of performance “become[s] effective upon promulgation.” CAA section 111(b)(1)(B). That is, upon promulgation, affected sources become legally subject to the multi-phase standard of performance and must comply with it by its terms. Specifically, affected sources must comply with the first phase standards, which are based on the application of the first component of the BSER, upon initial startup of the facility. They must comply with the second phase standards, which are based on the application of both the first and second components of the BSER, beginning January 2032.

D.C. Circuit caselaw supports the proposition that CAA section 111 authorizes the EPA to determine that controls qualify as the BSER—including meeting the “adequately demonstrated” criterion—even if the controls require some amount of “lead time,” which the court has defined as “the time in which the technology will have to be available.” The caselaw's interpretation of “adequately demonstrated” to accommodate lead time accords with common sense and the practical experience of certain types of controls, discussed below. Consistent with this caselaw, the phased implementation of the standards of performance in this rule ensures that facilities have sufficient lead time for planning and implementation of the use of CCS-based controls necessary to comply with the second phase of the standards, and thereby ensures that the standards are achievable. For further discussion of this point, see section V.C.2.b.iii.

The EPA has promulgated several prior rulemakings under CAA section 111(b) that have similarly provided the regulated sector with lead time to accommodate the availability of technology, which also serve as precedent for the two-phase implementation approach proposed in this rule. See 81 FR 59332 (August 29, 2016) (establishing standards for municipal solid waste landfills with 30-month compliance timeframe for installation of control device, with interim milestones); 80 FR 13672, 13676 (March 16, 2015) (establishing stepped compliance approach to wood heaters standards to permit manufacturers lead time to develop, test, field evaluate and certify current technologies to meet Step 2 emission limits); 78 FR 58416, 58420 (September 23, 2013) (establishing multi-phased compliance deadlines for revised storage vessel standards to permit sufficient time for production of necessary supply of control devices and for trained personnel to perform installation); 77 FR 56422, 56450 (September 12, 2012) (establishing standards for petroleum refineries, with 3-year compliance timeframe for installation of control devices); 71 FR 39154, 39158 (July 11, 2006) (establishing standards for stationary compression ignition internal combustion engines, with 2- to 3-year compliance timeframe and up to 6 years for certain emergency fire pump engines); 70 FR 28606, 28617 (March 18, 2005) (establishing two-phase caps for mercury standards of performance from new and existing coal-fired electric utility steam generating units based on timeframe when additional control technologies were projected to be adequately demonstrated). Cf.80 FR 64662, 64743 (October 23, 2015) (establishing interim compliance period to phase in final power sector GHG standards to allow time for planning and investment necessary for implementation activities). In each action, the standards and compliance timelines were effective upon the final rule, with affected facilities required to comply consistent with the phased compliance deadline specified in each action.

It should be noted that the multi-phased implementation of the standards of performance that the EPA is finalizing in this rule, like the delayed or multi-phased standards in prior rules just described, is distinct from the promulgation of revised standards of performance under the 8-year review provision of CAA section 111(b)(1)(B). As discussed in section VIII.F, the EPA has determined that the proposed BSER—highly efficient generation and use of CCS—meet all of the statutory criteria and are adequately demonstrated for the compliance timeframes being finalized. Thus, the second phase of the standard of performance applies to affected facilities that commence construction after May 23, 2023 (the date of the proposal). In contrast, when the EPA later reviews and (if appropriate) revises a standard of performance under the 8-year review provision, then affected sources that commence construction after the date of that proposal of the revised standard of performance will be subject to that standard, but not sources that commenced construction earlier.

Similarly, the multi-phased implementation of the standard of performance that the EPA is including in this rule is also distinct from the promulgation of emission guidelines for existing sources under CAA section 111(d). Emission guidelines only apply to existing sources, which are defined in CAA section 111(a)(6) as “any stationary source other than a new source.” Because new sources are defined relative to the proposal of standards pursuant to CAA section 111(b)(1)(B), standards of performance adopted pursuant to emission guidelines will only apply to sources constructed before May 23, 2023, the date of the proposed standards of performance for new sources.

b. BSER for the Intermediate Load Subcategory—Second Component

The EPA proposed that the second component of the BSER for intermediate load combustion turbines was co-firing 30 percent low-GHG hydrogen in 2032. As discussed in section VIII.F.5.b, the EPA is not determining that low-GHG hydrogen qualifies as the BSER at this time. Therefore, the Agency is not finalizing a second component of the BSER for intermediate load combustion turbines.

c. BSER for Base Load Subcategory—Second Component

i. Lower-Emitting Fuels

The EPA did not propose and is not finalizing lower-emitting fuels as the second component of the BSER for intermediate or base load combustion turbines because it would achieve few emission reductions, compared to highly efficient generation without or in combination with the use of CCS.

ii. Highly Efficient Generation

For the reasons described above, the EPA is determining that highly efficient generation in combination with best operating and maintenance practices continues to be a component of the BSER that is reflected in the second phase of the standards of performance for base load combustion turbine EGUs. Highly efficient generation reduces fuel use and, therefore, the amount of CO₂ that must be captured by a CCS system. Since a highly efficient turbine system would produce less flue gas that would need to be treated (compared to a less efficient turbine system), physically smaller carbon capture equipment may be used—potentially reducing capital, fixed, and operating costs.

iii. Hydrogen Co-Firing

The EPA proposed a pathway for the second component of the BSER for base load combustion turbines of co-firing 30 percent low-GHG hydrogen in 2032 increasing to 96 percent low-GHG hydrogen co-firing in 2038. As discussed in section VIII.F.5.b of this preamble, the EPA is not finalizing a determination that low-GHG hydrogen co-firing qualifies as the BSER. Therefore, the Agency is not finalizing a second component low-GHG hydrogen co-firing pathway of the BSER for base load combustion turbines. As the EPA's standard of performance is technology neutral, however, affected sources may comply with it by co-firing hydrogen.

iv. CCS

(A) Overview

In this section of the preamble, the EPA explains its rationale for finalizing that CCS with 90 percent capture is a component of the BSER for new base load combustion turbines. CCS is a control technology that can be applied at the stack of a combustion turbine EGU, achieves substantial reductions in emissions and can capture and permanently sequester at least 90 percent of the CO₂ emitted by combustion turbines. The technology is adequately demonstrated, given that it has been operated on a large scale and is widely applicable to these sources, and there are vast sequestration opportunities across the continental U.S. Additionally, the costs for CCS are reasonable in light of recent technology cost declines and policies including the tax credit under IRC section 45Q. Moreover, the non-air quality health and environmental impacts of CCS can be mitigated, and the energy requirements of CCS are not unreasonably adverse. The EPA's weighing of these factors together provides the basis for finalizing 90 percent capture CCS as a component of BSER for these sources. In addition, this BSER determination aligns with the caselaw, discussed in section V.C.2.h of the preamble, stating that CAA section 111 encourages continued advancement in pollution control technology.

This section incorporates by reference the parts of section VII.C.1.a. of this preamble that discuss the many aspects of CCS that are common to both steam generating units and to new combustion turbines. This includes the discussion of simultaneous demonstration of CO₂ capture, transport, and sequestration discussed at VII.C.1.a.i(A); the discussion of CO₂ capture technology used at coal-fired steam generating units at VII.C.1.a.i(B) (the Agency explains below why that record is also relevant to our BSER analysis for new combustion turbines); the discussion of CO₂ transport at VII.C.1.a.i(C); and the discussion of geologic storage of CO₂ at VII.C.1.a.i(D). And the record supporting that transport and sequestration of CO₂ from coal-fired units is adequately demonstrated and meets the other requirements for BSER applies as well to transport and sequestration of CO₂ from combustion turbines.

The primary differences between using post-combustion capture from a coal combustion flue gas and a natural gas combustion flue gas are associated with the level of CO₂ in the flue gas stream and the levels of other pollutants that must be removed. In coal combustion flue gas, the concentration of CO₂ is typically approximately 13 to 15 volume percent, while the concentration of CO₂ from natural gas-fired combined cycle combustion flue gas is approximately 3 to 4 volume percent. Capture of CO₂ at dilute concentrations is more challenging but there are commercially available amine-based solvents that can be used with dilute CO₂ streams to achieve 90 percent capture. In addition, flue gas from a coal-fired steam EGU contains a variety of non-carbonaceous components that must be removed to meet environmental limits ( e.g., mercury and other metals, particulate matter (fly ash), and acid gases (including sulfur dioxide (SO₂) and hydrogen chloride and hydrogen fluoride). When amine-based post-combustion carbon capture is used with a coal-fired EGU, the flue gas stream must be further cleaned, sometimes beyond required environmental standards, to avoid the fouling of downstream process equipment and to prevent degradation of the amine solvent. Absent pretreatment of the coal combustion flue gas, the amines can absorb SO₂ and other acid gases to form heat stable salts, thereby degrading the performance of the solvent. Amine solvents can also experience catalytic oxidative degradation in the presence of some metal contaminants. Thermal oxidation of the solvent can also occur but can be mitigated by interstage cooling of the absorber column. Natural gas combustion flue gas typically contains very low (if any) levels of SO₂, acid gases, fly ash, and metals. Therefore, fouling and solvent degradation are less of a concern for carbon capture from natural gas-fired EGUs.

New natural gas-fired combustion turbine EGUs also have the option of using oxy-combustion technology—such as that currently being demonstrated and developed by NET Power. As discussed earlier, the NET Power system uses oxy-combustion (combustion in pure oxygen) of natural gas and a high-pressure supercritical CO₂ working fluid (instead of steam) to produce electricity in a combined cycle turbine configuration. The combustion products are water and high-purity, pipeline-ready CO₂ which is available for sequestration or sale to another industry. The NET Power technology does not involve solvent-based CO₂ separation and capture since pure CO₂ is a product of the process. The NET Power technology is not currently applicable to coal-fired steam generating utility boilers—though it could be utilized with combustion of gasified coal or other solid fossil fuels ( e.g., petroleum coke).

For new base load combustion turbines, the EPA proposed that CCS with a 90 percent capture rate, beginning in 2035, meets the BSER criteria. Some commenters agreed with the EPA that CCS for base load combustion turbines satisfies the BSER criteria. Other commenters claimed that CCS is not a suitable BSER for new base load combustion turbines. The EPA disagrees with these commenters.

As with existing coal-fired steam generating units, CCS applied to new combined cycle combustion turbines has three major components: CO₂ capture, transportation, and sequestration/storage. CCS with 90 percent capture has been adequately demonstrated for combined cycle combustion turbines for many of the same reasons described in section VII.C.1.a.i. The Bellingham Energy Center, a natural gas-fired combined cycle combustion turbine in south central Massachusetts, successfully applied post-combustion carbon capture using the Fluor Econamine FG Plus amine-based solvent from 1991-2005 with 85-95 percent CO₂ capture. The plant captured approximately 365 tons of CO₂ per day from a 40 MW slip stream and was ultimately shut down and decommissioned primarily due to rising gas prices.

As discussed in further detail below, additional natural gas-fired combined cycle combustion turbine CCS projects are in the planning stage, which confirms that CCS is becoming accepted across the industry. As discussed above, CCS with 90 percent capture has been demonstrated for coal-fired steam generating units, and that information forms part of the basis for the EPA's determination that CCS with 90 percent capture has been have adequately demonstrated for these combustion turbines. Statements from vendors and the experience of industrial applications of CCS provide further support that post-combustion CCS with 90 percent capture is adequately demonstrated for these combustion turbines.

The EPA's analysis of the transportation and sequestration components of CCS for new base load combustion turbines is similar to its analysis of those components for existing coal-fired steam generating units and, therefore, for much the same reasons, the EPA is determining that each of those components is adequately demonstrated, and that CCS as a whole—including those components when combined with the 90 percent CO₂ capture component—is adequately demonstrated. In addition, new sources may consider access to CO₂ transport and storage sites in determining where to build, and the EPA expects that since this rule was proposed, companies siting new base load combustion turbines have taken into consideration the likelihood of a regulatory regime requiring significant emissions reductions.

The use of CCS at 90 percent capture can be implemented at reasonable cost because it allows affected sources to maximize the benefits of the IRC section 45Q tax credit. Finally, any adverse health and environmental impacts and energy requirements are limited and, in many cases, can be mitigated or avoided. It should also be noted that a determination that CCS is the BSER for these units will promote further use and development of this advanced technology. After balancing these factors, the EPA is determining that utilization of CCS with 90 percent capture for new base load combustion turbine EGUs satisfies the criteria for BSER.

(B) Adequately Demonstrated

The legal test for an adequately demonstrated system, and an achievable standard, has been discussed at length above. (See sections V.C.2.b and VII.C.a.i of this preamble). As previously noted, concepts of adequate demonstration and achievability are closely related: “[i]t is the system which must be adequately demonstrated and the standard which must be achievable,” based on application of the system. An achievable standard means a standard based on the EPA's finding that sufficient evidence exists to reasonably determine that the affected sources in the source category can adopt a specific system of emission reduction to achieve the specified degree of emission limitation. The foregoing sections have shown that CCS, specifically using amine post-combustion CO₂ capture, is adequately demonstrated for existing coal units, and that a 90 percent capture standard is achievable.

Pursuant to Lignite Energy Council v. EPA, the EPA may extrapolate based on data from a particular kind of source to conclude that the technology at issue will also be effective at a similar source. This standard is satisfied in our case, because of the essential ways in which CO₂ capture at coal-fired steam generating units is identical to CO₂ capture at natural gas-fired combined cycle turbines. As detailed in section VII.C.1.a.i(B), amine-based CO₂ capture removes CO₂ from post-combustion flue gas by reaction of the CO₂ with amine solvent. The same technology ( i.e., the same solvents and processes) that is employed on coal-fired steam generating units—and that is employed to capture CO₂ from fossil fuel combustion in other industrial processes—can be applied to remove CO₂ from the post-combustion flue gas of natural gas-fired combined cycle EGUs. In fact, the only differences in application of amine-based CO₂ capture on a natural gas-fired combined cycle unit relative to a coal-fired steam generating unit are related to the differences in composition of the respective post-combustion flue gases, and as explained below, these differences do not preclude achieving 90 percent capture from a gas-fired turbine.

First, while coal flue gas contains impurities including SO₂, PM, and trace minerals that can affect the downstream CO₂ process, and thus coal flue gas requires substantial pre-treatment, the post-combustion flue gas of natural gas-fired combustion turbines has few, if any, impurities that would impact the downstream CO₂ capture plant. Where impurities are present, SO₂ in particular can cause solvent degradation, and coal-fired sources without an FGD would likely need to install one. Filterable PM (fly ash) from coal, if not properly managed, can cause fouling and scale to accumulate on downstream blower fans, heat exchangers, and absorber packing material. Further, additional care in the solvent reclamation is necessary to mitigate solvent degradation that could otherwise occur due to the trace elements that can be present in coal. Because the flue gas from natural gas-fired combustion turbines contains few, if any, impurities that would impact downstream CO₂ capture, the flue gas from natural gas-fired combined cycle EGUs is easier to work with for CO₂ capture, and many of the challenges that were faced by earlier commercial scale demonstrations on coal-fired units can be avoided in the application of CCS at natural gas-fired combustion turbines.

Second, the CO₂ concentration of natural gas-fired combined cycle flue gas is lower than that of coal flue gas (approximately 3-to-4 volume percent for natural gas combined cycle EGUs; 13-to-15 volume percent for coal). For solvent-based CO₂ capture, CO₂ concentration is the driving force for mass transfer and the reaction of CO₂ with the solvent. However, flue gases with lower CO₂ concentrations can be readily addressed by the correct sizing and design of the capture equipment—and such considerations have been made in evaluating the BSER here and are reflected in the cost analysis in VII.C.1.a.ii(A) of this preamble. Moreover, as is detailed in the following sections of the preamble, amine-based CO₂ capture has been shown to be effective at removal of CO₂ from the flue gas of natural gas-fired combined cycle EGUs. In fact, there is not a technical limit to removal of CO₂ from flue gases with low CO₂ concentrations—the EPA notes that amine solvents have been shown to be able to remove CO₂ to concentrations that are less than the concentration of CO₂ in the atmosphere.

Considering these factors, the evidence that underlies the EPA's determination that amine post-combustion CO₂ capture is adequately demonstrated, and that a 90 percent capture standard is achievable, at coal-fired steam generating units, also applies to natural gas-fired combined cycle EGUs. Where differences exist, due to differences in flue gas composition, CCS at natural gas-fired combined cycle combustion turbines will in general face fewer challenges than CCS at coal-fired steam generators. Moreover, in addition to the evidence outlined above, the following sections provide additional information specific to, including examples of, anime-based capture at natural gas-fired combined cycle EGUs. For these reasons, the EPA has determined that CCS at 90 percent capture is adequately demonstrated for natural gas fired combined cycle EGUs.

(1) CO₂ Capture for Combined Cycle Combustion Turbines

As discussed in the preceding, new stationary combustion turbines can use amine-based post-combustion capture. Additionally, new stationary combustion turbines may also utilize oxy-combustion, which uses a purified oxygen stream from an air separation unit (often diluted with recycled CO₂ to control the flame temperature) to combust the fuel and produce a nearly pure stream of CO₂ in the flue gas, as opposed to combustion with oxygen in air which contains 80 percent nitrogen. Currently available post-combustion amine-based CO₂ capture systems require that the flue gas be cooled prior to entering the capture equipment. This holds true for the exhaust from either a coal-fired utility boiler or from a combustion turbine. The most energy efficient way to cool the flue gas stream is to use a HRSG—which, as explained above, is an integral component of a combined cycle turbine system—to generate additional useful output.

CO₂ capture has been successfully applied to an existing combined cycle turbine and several other projects are in development, as discussed immediately below.

(a) CCS on Combined Cycle EGUs

The most prominent example of the use of carbon capture technology on a natural gas-fired combined cycle turbine EGU was at the 386 MW Bellingham Cogeneration Facility in Bellingham, Massachusetts. The plant used Fluor's Econamine FG Plus amine-based CO₂ capture system with a capture capacity of 360 tons of CO₂ per day. The system was used to produce food-grade CO₂ and was in continuous commercial operation from 1991 to 2005 (14 years). The capture system was able to continuously capture 85-95 percent of the CO₂ that would have otherwise been emitted from the flue gas of a 40 MW slip stream. The natural gas combustion flue gas at the facility contained 3.5 volume percent CO₂ and 13-14 volume percent oxygen. As mentioned earlier, the flue gas from a coal combustion flue gas stream has a typical CO₂ concentration of approximately 15 volume percent and more dilute CO₂ stream are more challenging to separate and capture. Just before the CO₂ capture system was shut down in 2005 (due to high natural gas price), the system had logged more than 120,000 hours of CO₂ capture and had a 98.5 percent on-stream (availability) factor.

The Fluor Econamine FG Plus is a propriety carbon capture solution with more than 30 licensed plants and more than 30 years of operation. This technology uses a proprietary solvent to capture CO₂ from post-combustion sources. The process is well suited to capture CO₂ from large, single-point emission sources such as power plants or refineries, including large facilities with CO₂ capture capacities greater than 10,000 tons per day. On February 6, 2024, Fluor Corporation announced that Chevron New Energies plans to use the Econamine FG Plus carbon capture technology to reduce CO₂ emissions at Chevron's Eastridge Cogeneration combustion turbine facility in Kern County, California. When installed, Fluor's carbon capture solution is expected to reduce the Eastridge Cogeneration facility's carbon emissions by approximately 95 percent.

Moreover, recently, CO₂ capture technology has been operated on NGCC post-combustion flue gas at the Technology Centre Mongstad (TCM) in Norway. TCM can treat a 12 MWe flue gas stream from a natural gas combined cycle cogeneration plant at Mongstad power station. Many different solvents have been operated at TCM including MHI's KS-21 solvent, achieving capture rates of over 98 percent.

Additionally, in Scotland, the proposed 900 MW Peterhead Power Station combined cycle EGU with CCS is in the planning stages of development. MHI is developing a FEED for the power plant and capture facility. It is anticipated that the power plant will be operational by the end of the 2020s and will have the potential to capture 90 percent of the CO₂ emitting from the combined cycle facility and sequester up to 1.5 million metric tons of CO₂ annually. A storage site being developed 62 miles off the Scottish North Sea coast will serve as a destination for the captured CO₂ .

Furthermore, the Global CCS Centre is tracking other international CCS on combustion turbine projects that are in on-going stages of development.

(b) NET Power Cycle

In addition, there are several planned projects using NET Power's Allam-Fetvedt Cycle. The Allam-Fetvedt Cycle is a proprietary process for producing electricity that combusts a fuel with purified oxygen (diluted with recycled CO₂ to control flame temperature) and uses supercritical CO₂ as the working fluid instead of water/steam. This cycle is designed to achieve thermal efficiencies of up to 59 percent. Potential advantages of this cycle are that it emits no NO_X and produces a stream of high-purity CO₂ that can be delivered by pipeline to a storage or sequestration site without extensive processing. A 50 MW (thermal) test facility in La Porte, Texas was completed in 2018 and has since accumulated over 1,500 hours of runtime. There are several announced NET Power commercial projects proposing to use the Allam-Fetvedt Cycle. These include the 280 MW Broadwing Clean Energy Complex in Illinois, and several international projects.

In Scotland, the proposed 900 MW Peterhead Power Station combined cycle EGU with CCS is in the planning stages of development. MHI is developing a FEED for the power plant and capture facility.⁷⁷⁷ It is anticipated that the power plant will be operational by the end of the 2020s and will have the potential to capture 90 percent of the CO₂ emitting from the combined cycle facility and sequester up to 1.5 million metric tons of CO₂ annually. A storage site being developed 62 miles off the Scottish North Sea coast will serve as a destination for the captured CO₂ .^{778 779}

(c) Coal-Fired Steam Generating Units

As detailed in section VII.C.1.a, CCS has been demonstrated on coal-fired power plants, which provides further support that CCS on base load combined cycle units is adequately demonstrated. Further, 90 percent capture is expected to be, in some ways, more straightforward to achieve for natural gas-fired combined cycle combustion turbines than for coal-fired steam generators. Many of the challenges faced by Boundary Dam Unit 3—which proved to be solvable—were caused by the impurities, including fly ash, SO₂, and trace contaminants in coal-fired post-combustion flue gas. Such impurities naturally occur in coal (sulfur and trace contaminants) or are a natural result of combusting coal (fly ash), but not in natural gas, and thus they do not appear in the natural gas post-combustion flue gas. As a result, for CO₂ capture for natural gas combustion, flue gas handling is simpler, solvent degradation is easier to prevent, and fewer redundancies may be necessary for various components ( e.g., heat exchangers).

(d) Other Industry

As discussed in section VII.C.1.a.i.(A)(1) of this preamble, CCS installations in other industries support that capture equipment can achieve 90 percent capture of CO₂ from natural gas-fired base load combined cycle combustion turbines.

(e) EPAct05-Assisted CO₂ Capture Projects at Stationary Combustion Turbines

As for steam generating units, EPAct05-assisted CO₂ capture projects on stationary combustion turbines corroborate that CO₂ capture on gas combustion turbines is adequately demonstrated. Several CCS projects with at least 90 percent capture at commercial-scale combined cycle turbines are in the planning stages. These projects support that CCS with at least 90 percent capture for these units is the industry standard and support the EPA's determination that CCS is adequately demonstrated.

CCS is planned for the existing 550 MW natural gas-fired combined cycle (two combustion turbines) at the Sutter Energy Center in Yuba City, California. The Sutter Decarbonization project will use ION Clean Energy's amine-based solvent technology at a capture rate of 95 percent or more. The project expects to complete a FEED study in 2024 and, prior to being selected by DOE for funding award negotiation, planned commercial operation in 2027. Sutter Decarbonization is one of the projects selected by DOE for funding as part of OCED's Carbon Capture Demonstration Projects program.

The CO₂ capture project at the Deer Park Energy Center in Deer Park, Texas will be designed to capture 95 percent or more of the flue gas from the five combustion turbines at the 1,200 MW natural gas-fired combined cycle power plant, using technology from Shell CANSOLV. The CO₂ capture project already has an air permit issued for the project, which includes a reduction in the allowable emission limits for NO_X from four of the combustion turbines. The CO₂ capture facility will include two quencher columns, two absorber columns, and one stripping column.

The Baytown Energy Center in Baytown, Texas is an existing natural gas-fired combined cycle cogeneration facility providing heat and power to a nearby industrial facility, while distributing additional electricity to the grid. CCS using Shell's CANSOLV solvent is planned for the equivalent of two of the three combustion turbines at the 896 MW natural gas-fired combined cycle power plant, with a capture rate of 95 percent. The CO₂ capture facility at Baytown Energy Center also has an air permit in place, and the permit application provides some details on the process design. The CO₂ capture facility will include two quencher columns, two absorber columns, and one stripping column. To mitigate NO_X emissions, the operation of the SCR systems for the combustion turbines will be adjusted to meet lower NO_X allowable limits—adjustments may include increasing ammonia flow, more frequent SCR repacking and head cleaning, and, possibly, optimization of the ammonia distribution system. The Baytown CO₂ capture project is one of the projects selected by DOE for funding as part of OCED's Carbon Capture Demonstration Projects program. Captured CO₂ will be transported and stored at sites along the U.S. Gulf Coast.

An 1,800 MW natural gas-fired combustion turbine that will be constructed in West Virginia and will utilize CCS has been announced. The project is planned to begin operation later this decade.

There are numerous other EPAct05-assisted projects related to natural gas-fired combined cycle turbines including the following. These projects provide corroborating evidence that capture of at least 90 percent is accepted within the industry.

• General Electric (GE) (Bucks, Alabama) was awarded $5,771,670 to retrofit a combined cycle turbine with CCS technology to capture 95 percent of CO₂ and is targeting commercial deployment by 2030.
• Wood Environmental & Infrastructure Solutions (Blue Bell, Pennsylvania) was awarded $4,000,000 to complete an engineering design study for CO₂ capture at the Shell Chemicals Complex. The aim is to reduce CO₂ emissions by 95 percent using post-combustion technology to capture CO₂ from several plants, including an onsite natural gas CHP plant.
• General Electric Company, GE Research (Niskayuna, New York) was awarded $1,499,992 to develop a design to capture 95 percent of CO₂ from combined cycle turbine flue gas with the potential to reduce electricity costs by at least 15 percent.
• SRI International (Menlo Park, California) was awarded $1,499,759 to design, build, and test a technology that can capture at least 95 percent of CO₂ while demonstrating a 20 percent cost reduction compared to existing combined cycle turbine carbon capture.
• CORMETECH, Inc. (Charlotte, North Carolina) was awarded $2,500,000 to further develop, optimize, and test a new, lower-cost technology to capture CO₂ from combined cycle turbine flue gas and improve scalability to large, combined cycle turbines.
• TDA Research, Inc. (Wheat Ridge, Colorado) was awarded $2,500,000 to build and test a post-combustion capture process to improve the performance of combined cycle turbine flue gas CO₂ capture.
• GE Gas Power (Schenectady, New York) was awarded $5,771,670 to perform an engineering design study to incorporate a 95 percent CO₂ capture solution for an existing combined cycle turbine site while providing lower costs and scalability to other sites.
• Electric Power Research Institute (EPRI) (Palo Alto, California) was awarded $5,842,517 to complete a study to retrofit a 700 MWe combined cycle turbine with a carbon capture system to capture 95 percent of CO₂.
• Gas Technology Institute (Des Plaines, Illinois) was awarded $1,000,000 to develop membrane technology capable of capturing more than 97 percent of combined cycle turbine CO₂ flue gas and demonstrate upwards of 40 percent reduction in costs.
• RTI International (Research Triangle Park, North Carolina) was awarded $1,000,000 to test a novel non-aqueous solvent technology aimed at demonstrating 97 percent capture efficiency from simulated combined cycle turbine flue gas.
• Tampa Electric Company (Tampa, Florida) was awarded $5,588,173 to conduct a study retrofitting Polk Power Station with post-combustion CO₂ capture technology aiming to achieve a 95 percent capture rate.

There are also several announced NET Power Allam-Fetvedt Cycle based CO₂ capture projects that are EPAct05-assisted. These include the 280 MW Coyote Clean Power Project on the Southern Ute Indian Reservation in Colorado and a 300 MW project located near Occidental's Permian Basin operations close to Odessa, Texas. Commercial operation of the facility near Odessa, Texas is expected in 2028.

(f) Range of Conditions

The composition of natural gas combined cycle post-combustion flue gas is relatively uniform as the level of impurities is, in general, low. There may be some difference in NO_X emissions, but considering the sources are new, it is likely that they will be installed with SCR, resulting in uniform NO_X concentrations in the flue gas. The EPA notes that some natural gas combined cycle units applying CO₂ capture may use exhaust gas recirculation to increase the concentration of CO₂ in the flue gas—this produces a higher concentration of CO₂ in the flue gas. For those sources that apply that approach, the CO₂ capture system can be scaled smaller, reducing overall costs. Considering these factors, the EPA concludes that there are not substantial differences in flue gas conditions for natural gas combined cycle units, and the small differences that could exist would not adversely impact the operation of the CO₂ capture equipment.

As detailed in section VII.C.1.a.i(B)(7), single trains of CO₂ capture facilities have turndown capabilities of 50 percent. Effective turndown to 25 percent of throughputs can be achieved by using 2 trains of capture equipment. CO₂ capture rates have also been shown to be higher at lower throughputs. Moreover, during off-peak hours when electricity prices are lower, additional lean solvent can be produced and held in reserve, so that during high-demand hours, the auxiliary demands to the capture plant stripping column reboiler be reduced. Considering these factors, the capture rate would not be affected by load following operation, and the operation of the combustion turbine would not be limited by turndown capabilities of the capture equipment. As detailed in preceding sections, simple cycle combustion turbines cycle frequently, and have a number of startups and shutdowns per year. However, combined cycle units cycle less frequently and have fewer startups and shutdowns per year. Startups of combined cycle units are faster than coal-fired steam generating units described in section VII.C.1.a.i(B)(7) of the preamble. Cold startups of combined cycle units typically take not more than 3 hours (hot startups are faster), and shutdown takes less than 1 hour. During startup, heat input to the unit is lower to slowly raise the temperature of the HRSG.

Importantly, natural gas post-combustion flue gas does not require the same pretreatment as coal post-combustion flue gas. Therefore, amine solvents are able to capture CO₂ as soon as the flue gas contacts the lean solvent, and startup does not have to wait for operation of other emission controls. Furthermore, there are several different process strategies that can be employed to enable capture during cold startup. These include using an additional reserve of lean solvent (solvent without absorbed CO₂), dedicated heat storage for reboiler preheating, and fast starting steam cycle technologies or high-pressure bypass extraction. Each of these three options has been modeled to show that 95 percent capture rates can be achieved during startup. The first option simply uses a reserve of lean solvent during startup so that capture can occur without needing to wait for the stripping column reboiler to heat up. For hot starts, the startup time of the NGCC is faster, and since the reboiler is already warm, the capture plant can begin operating faster. Shutdowns are short, and high capture efficiencies can be maintained.

Considering that startup and shutdown for natural gas combined cycle units is fast, startups are relatively few, and simple process strategies can be employed so that high capture efficiencies can be achieved during startup, the EPA has concluded that startup and shutdown do not adversely impact the achievable CO₂ capture rate.

Considering the preceding information, the EPA has determined that 90 percent capture is achievable over long periods ( i.e., 12-month rolling averages) for base load combustion turbines for all relevant flue gas conditions, variable load, and startup and shutdown.

(g) Summary of Evidence Supporting BSER Determination Without EPAct05-Aassisted Projects

As noted above, under the EPA's interpretation of the EPAct05 provisions, the EPA may not rely on capture projects that received assistance under EPAct05 as the sole basis for a determination of adequate demonstration, but the EPA may rely on those projects to support or corroborate other information that supports such a determination. The information described above that supports the EPA's determination that 90 percent CO₂ capture from natural gas-fired combustion turbines is adequately demonstrated, without consideration of the EPAct05-assisted projects, includes (i) the information concerning coal-fired steam generating units listed in VII.C.1.a.i.(B)(9) (other than the information concerning EPAct05-assisted coal-fired unit projects and the information concerning natural gas-fired combustion turbines); (ii) the information that a 90 percent capture standard is achievable at coal-fired steam generating units, also applies to natural gas-fired combined cycle EGUs ( i.e., all the information in VIII.F.4.c.iv.(B) (before (1)) and (1) (before (a)); (iii) the information concerning CCS on combined cycle EGUs ( i.e., all the information in VIII.F.4.c.iv.(B)(1)(a)); and (iv) the information concerning Net Power ( i.e., all the information in VIII.F.4.c.iv.(B)(1)(b)). All this information by itself is sufficient to support the EPA's determination that 90 percent CO₂ capture from coal-fired steam generating units is adequately demonstrated. Substantial additional information from EPAct05-assisted projects, as described in section VIII.F.4.c.iv.(B)(1)(e), provides additional support and confirms that 90 percent CO₂ capture from natural gas-fired combustion turbines is adequately demonstrated.

(2) Transport of CO₂

In section VII.C.1.a.i.(C) of this document, the EPA described its rationale for finalizing a determination that CO₂ transport by pipelines as a component of CCS is adequately demonstrated for use of CCS with existing steam generating EGUs. The Agency's rationale for finalizing the same determination—that CO₂ transport by pipelines as a component of CCS is adequately demonstrated for CCS use with new combustion turbine EGUs—is much the same as that described in section VII.C.1.a.i.(C). As discussed in section VII.C.1.a.i.(C) of this preamble, CO₂ pipelines are available and their network is expanding in the U.S., and the safety of existing and new supercritical CO₂ pipelines is comprehensively regulated by PHMSA. A new combustion turbine may also be co-located with a storage site, so that minimal transport of the CO₂ is required.

Pipeline transport of CO₂ captured from newly constructed or reconstructed natural gas-fired combustion turbine EGUs meets the BSER requirements based on the same evidence, and for the same reasons, as does pipeline transport of CO₂ captured from existing coal-fired steam generating EGUs, as described in section VII.C.1.a.i.(C) of this preamble. This is because the CO₂ that is captured from a natural gas-fired turbine, compressed, and delivered into a pipeline is indistinguishable from the CO₂ that is captured from an existing coal-fired steam generating unit. Accordingly, all the evidence and explanation in section VII.C.1.a.i.(C) of this preamble that it is adequately demonstrated, cost-effective, and consistent with the other BSER factors for an existing coal-fired steam generating unit to construct a lateral pipeline from its facility to a sequestration site applies to new natural gas-fired turbines. This includes the history of CO₂ pipeline build-out (VII.C.1.a.i.(C)(1)), the recent examples of new pipelines (VII.C.1.a.i.(C)(1)(b)), EPAct05-assisted CO₂ pipelines for CCS (VII.C.1.a.i.(C)(1)(c)), the network of existing and planned CO₂ trunklines (VII.C.1.a.i.(C)(1)(d)), permitting and rights of way considerations (VII.C.1.a.i.(C)(2)), and considerations of the security of CO₂ transport, including PHMSA requirements (VII.C.1.a.i.(C)(3)).

The only difference between pipeline transport for the coal-fired steam generation and the gas-fired turbines is that the coal-fired units are already in existence and, as a result, the location and length of their pipelines, as needed to transport their CO₂ to nearby sequestration, is already known, whereas new gas-fired turbines are not yet sited. We discuss the implications for new gas-fired turbines in the next section.

(3) Geologic Sequestration of CO₂

In section VII.C.1.a.i.(D) of this document, the EPA described its rationale for finalizing a determination that geologic sequestration ( i.e., the long-term containment of a CO₂ stream in subsurface geologic formations) is adequately demonstrated as a component of the use of CCS with existing coal-fired steam generating EGUs. Similar to the previous discussion regarding CO₂ transport, the Agency's rationale for finalizing a determination that geologic sequestration is adequately demonstrated as a component of the use of CCS with new combustion turbine EGUs is the same as described in VII.C.1.a.i.(D) for existing coal-fired steam generating EGUs. The storage/sequestration sites used to store captured CO₂ from existing coal-fired EGUs could also be used to store captured CO₂ from newly constructed or reconstructed combustion turbine EGUs. All of the considerations and challenges associated with developing geologic storage sites for existing sources are also considerations and challenges associated with developing such sites for newly constructed or reconstructed sources.

(a) In General

Geologic sequestration ( i.e., the long-term containment of a CO₂ stream in subsurface geologic formations) is well proven. Deep saline formations, which may be evaluated and developed for CO₂ sequestration are broadly available throughout the U.S. Geologic sequestration requires a demonstrated understanding of the processes that affect the fate of CO₂ in the subsurface. As discussed in section VII.C.1.a.i.(D) of this preamble, there have been numerous instances of geologic sequestration in the U.S. and overseas, and the U.S. has developed a detailed set of regulatory requirements to ensure the security of sequestered CO₂. This regulatory framework includes the UIC well regulations, which are under the authority of the SDWA, and the GHGRP, under the authority of the CAA.

Geologic settings which may be suitable for geologic sequestration of CO₂ are widespread and available throughout the U.S. Through an availability analysis of sequestration potential in the U.S. based on resources from the DOE, the USGS, and the EPA, the EPA found that there are 43 states with access to, or are within 100 km from, onshore or offshore storage in deep saline formations, unmineable coal seams, and depleted oil and gas reservoirs.

All of the evidence and explanation that geological sequestration of CO₂ is adequately demonstrated and meets the other BSER factors that the EPA described with respect to sequestration of CO₂ from existing coal-fired steam generating units in section VII.C.1.a.i.(D) of this preamble apply with respect to CO₂ from new natural gas-fired combustion turbines. Sequestration is broadly available (VII.C.1.a.i.(D)(1)(a)). It is adequately demonstrated, with many examples of projects successfully injecting and containing CO₂ in the subsurface (VII.C.1.a.i.(D)(2)). It provides secure storage, with a detailed set of regulatory requirements to ensure the security of sequestered CO₂, including the UIC well regulations pursuant to SDWA authority, and the GHGRP pursuant to CAA authority (VII.C.1.a.i.(D)(4)). The EPA has the experience to properly regulate and review permits for UIC Class VI injection wells, has made considerable improvements to its permitting process to expedite permitting decisions, and has granted several states primacy to issue permits, and is supporting that state permitting (VII.C.1.a.i.(D)(5)).

(b) New Natural Gas-Fired Combustion Turbines

As discussed in section VII.C.1.a.i.(D)(1), deep saline formations that may be considered for use in geologic sequestration (or storage) are common in the continental United States. In addition, there are numerous unmineable coal seams and depleted oil and gas reserves throughout the country that could potentially be utilized as sequestration sites. The DOE estimates that areas of the U.S. with appropriate geology have a sequestration potential of at least 2,400 billion to over 21,000 billion metric tons of CO₂ in deep saline formations, unmineable coal seams, and oil and gas reservoirs. The EPA's scoping assessment found that at least 37 states have geologic characteristics that are amenable to deep saline sequestration and identified an additional 6 states are within 100 kilometers of potentially amenable deep saline formations in either onshore or offshore locations. In terms of land area, 80 percent of the continental U.S. is within 100 km of deep saline formations. While the EPA's geographic availability analyses focus on deep saline formations, other geologic formations such as unmineable coal seams or depleted oil and gas reservoirs represent potential additional CO₂ storage options. Therefore, we expect that the vast majority of new base load combustion turbine EGUs could be sited within 100 km of a sequestration site.

While the potential for some type of sequestration exists in large swaths of the continental U.S., we recognize that there are a few states that do not have geologic conditions suitable for geologic sequestration within or near their borders. If an area does not have a suitable geologic sequestration site, then a utility or project developer seeking to build a new combustion turbine EGU for base load generation has two options—either (1) the new EGU may be located near the electricity demand and the CO₂ transported via a CO₂ pipeline to a geologic sequestration site, or (2) the new EGU may be located closer to a geologic sequestration site and the electricity delivered to customers through transmission lines. Regarding option 1, as discussed in VII.C.1.a.i(C), the EPA believes that both new and existing EGUs are capable of constructing CO₂ pipelines as needed. With regard to option 2, we expect that this option may be preferred for projects where a CO₂ pipeline of substantial length would be required to reach the sequestration site. However, we note that for new base load combustion turbine EGUs, project developers have flexibility with regard to siting such that they can balance whether to site a new unit closer to a potential geologic sequestration site or closer to a load area depending on their specific needs.

Electricity demand in areas that may not have geologic sequestration sites may be served by gas-fired EGUs that are built in areas with geologic sequestration, and the generated electricity can be delivered through transmission lines to the load areas through “gas-by-wire.” An analogous approach, known as “coal-by-wire” has long been used in the electricity sector for coal-fired EGUs because siting a coal-fired EGU near a coal mine and transmitting the generated electricity long distances to the load area is sometimes less expensive than siting the coal EGU near the load area and shipping the coal long distances. The same principle may apply to new base load combustion turbine EGUs such that it may be more practicable for an project developer to site a new base load combustion turbine EGU in a location in close proximity to a geologic sequestration site and to deliver the electricity generated through transmission lines to the load area rather than siting the new gas-fired combustion turbine EGU near the load area and building a lengthy pipeline to the geologic sequestration site.

Gas-by-wire and coal-by-wire are possible due to the electricity grid's extensive high voltage transmission networks that enable electricity to be transmitted over long distances. See the memorandum, Geographic Availability of CCS for New Base Load NGCC Units, which is available in the rulemaking docket for this action. In many of the areas without reasonable access to geologic sequestration, utilities, electric cooperatives, and municipalities have a history of joint ownership of electricity generation outside the region or contracting with electricity generation in outside areas to meet demand. Some of the areas are in Regional Transmission Organizations (RTOs), which engage in planning as well as balancing supply and demand in real time throughout the RTO's territory. Accordingly, generating resources in one part of the RTO can serve load in other parts of the RTO, as well as load outside of the RTO.

In the coal context, there are many examples of where coal-fired power generation in one state has been used to supply electricity in other states. For example, the Prairie State Generating Plant, a 2-unit 1,600 MW coal-fired power plant in Illinois that is currently considering retrofitting with CCS, serves load in eight different states from the Midwest to the mid-Atlantic. The Intermountain Power Project, a coal-fired plant located in Delta, Utah, that is converting to co-fire hydrogen and natural gas, serves customers in both Utah and California. Additionally, historically nearly 40 percent of the power for the City of Los Angeles was provided from two coal-fired power plants located in Arizona and Utah. Further, Idaho Power, which serves customers in Idaho and eastern Oregon has met demand in part from power generating at coal-fired power plants located in Wyoming and Nevada. This same concept of siting generation in one location to serve demand in another area and using existing transmission infrastructure to do so could similarly be applied to gas-fired combustion turbine power plants, and, in fact, there are examples of gas-fired combustion turbine EGUs serving demand more than 100 km away from where they are sited. For example, Portland General Electric's Carty Generating Station, a 436-MW NGCC unit located in Boardman, Oregon serves demand in Portland, Oregon, which is approximately 270 km away from the source.

In the memorandum, Geographic Availability of CCS for New Base Load NGCC Units, we explore in detail the potential for gas-by-wire and the ability of demand in areas without geologic sequestration potential to be served by gas generation located in areas that have access to geologic sequestration. As discussed in the memorandum, the vast majority of the United States is within 100 km of an area with geologic sequestration potential. A review of our scoping assessment indicates that there are limited areas of the country that are not within 100 km of a potential deep saline sequestration formation (although some of these areas may be within 100 km of an unmineable coal seam or depleted oil and gas reservoir that could potentially serve as a sequestration site). In many instances, these areas include areas with low population density, areas that are already served by transmission lines that could deliver gas-by-wire, and/or include areas that have made policy or other decisions not to pursue a resource mix that includes new NGCC due to state renewable portfolio standards or for other reasons.

In many of these areas, utilities, electric cooperatives, and municipalities have a history of obtaining electricity from generation in outside areas to meet demand. Some of the relevant areas are in an RTO or ISO, which operate the transmission system and dispatch generation to balance supply and demand regionwide, as well as engage in regionwide planning and cost allocation to facilitate needed transmission development. Accordingly, generating resources in one part of an RTO/ISO, such as through an NGCC plant, can serve loads in other parts of the RTO/ISO, as well as serving load areas outside of the RTO/ISO. As we consider each of these geographic areas in the memorandum, Geographic Availability of CCS for New Base Load NGCC Units, we make key points as to why this final rule does not negatively impact the ability of these regions to access new NGCC generation to the extent that NGCC generation is needed to supply demand and/or those regions want to include new NGCC generation in their resource mixes.

(C) Costs

The EPA has evaluated the costs of CCS for new combined cycle units, including the cost of installing and operating CO₂ capture equipment as well as the costs of transport and storage. The EPA has also compared the costs of CCS for new combined cycle units to other control costs, in part derived from other rulemakings that the EPA has determined to be cost-reasonable, and the costs are comparable. Based on these analyses, the EPA considers the costs of CCS for new combined cycle units to be reasonable. Certain elements of the transport and storage costs are similar for new combustion turbines and existing steam generating units. In this section, the EPA outlines these costs and identifies the considerations specific to new combustion turbines. These costs are significantly reduced by the IRC section 45Q tax credit.

(1) Capture Costs

According to the NETL Fossil Energy Baseline Report (October 2022 revision), before accounting for the IRC section 45Q tax credit for sequestered CO₂, using a 90 percent capture amine-based post-combustion CO₂ capture system increases the capital costs of a new combined cycle EGU by 115 percent on a $/kW basis, increases the heat rate by 13 percent, increases incremental operating costs by 35 percent, and derates the unit ( i.e., decreases the capacity available to generate useful output) by 11 percent. For a base load combustion turbine, carbon capture increases the LCOE by 62 percent (an increase of 27 $/MWh) and has an estimated cost of $81/ton ($89/metric ton) of onsite CO₂ reduction. The NETL costs are based on the use of a second-generation amine-based capture system without exhaust gas recirculation (EGR) and, as discussed below, do not take into account further cost reductions that can be expected to occur from efficiency improvements as post-combustion capture systems are more widely deployed, as well as potential technological developments.

The flue gas from natural gas-fired combined cycle turbine differs from that of coal-fired EGUs in several ways that impact the cost of CO₂ capture. These include that the CO₂ concentration in the flue gas is approximately one-third of that observed at coal-fired EGUs, the volumetric flow rate on a per MW basis is larger, and the oxygen concentration is approximately 3 times that of a coal-fired EGU. While the higher amount of excess oxygen has the potential to reduce the efficiency of amine-based solvents that are susceptible to oxidation, natural gas post-combustion flue gas does not have other impurities (SO₂, PM, trace metals) that are present and must be managed in coal flue gas. Other important factors include that the lower concentrations of CO₂ reduce the efficiency of the capture process and that the larger volumetric flow rates require a larger CO₂ absorber, which increases the capital cost of the capture process. Exhaust gas recirculation (EGR), also referred to as flue gas recirculation (FGR), is a process that addresses all these issues. EGR diverts some of the combustion turbine exhaust gas back into the inlet stream for the combustion turbine. Doing so increases the CO₂ concentration and decreases the O₂ concentration in the exhaust stream and decreases the flow rate, producing more favorable conditions for CCS. One study found that EGR can decrease the capital costs of a combined cycle EGU with CCS by 6.4 percent, decrease the heat rate by 2.5 percent, decrease the LCOE by 3.4 percent, and decrease the overall CO₂ capture costs by 11 percent relative to a combined cycle EGU without EGR. The EPA notes that the NETL costs on which the EPA bases its cost calculations for combined cycle CCS do not assume the use of EGR, but as discussed below, EGR use is plausible and would reduce those costs.

While the costs considered in the preceding are based on the current costs of CCS, the EPA notes that the costs of capture systems can be expected to decrease over the rest of this decade and continue to decrease afterwards. As part of the plan to reduce the costs of CO₂ capture, the DOE is funding multiple projects to further advance CCS technology from various point sources, including combined cycle turbines, cement manufacturing plants, and iron and steel plants. It should be noted that some of these projects may be EPAct05-assisted. The general aim is to lower the costs of the technologies, and to increase investor confidence in the commercial scale applications, particularly for newer technologies or proven technologies applied under unique circumstances. In particular, OCED's Carbon Capture Demonstration Projects are targeted to accelerate continued power sector carbon capture commercialization through reducing costs and reducing uncertainties to project development. These cost and uncertainty reductions arise from reductions in cost of capital, increases in system scale, standardization and reduction in non-recurring engineering costs, maturation of supply chain ecosystem, and improvements in engineering design and materials over time.

Although current post-combustion CO₂ capture projects have primarily been based on amine capture systems, there are multiple alternate capture technologies in development—many of which are funded through industry research programs—that could yield reductions in capital, operating, and auxiliary power requirements and could reduce the cost of capture significantly or improve performance. More specifically, post combustion carbon capture systems generally fall into one of several categories: solvents, sorbents, membranes, cryogenic, and molten carbonate fuel cells systems. It is expected that as CCS infrastructure increases, technologies from each of these categories will become more economically competitive. For example, advancements in solvents that are potentially direct substitutes for current amine-solvents will reduce auxiliary energy requirements and reduce both operating and capital costs, and thereby, increase the economic competitiveness of CCS. Planned large-scale projects, pilot plants, and research initiatives will also decrease the capital and operating costs of future CCS technologies.

In general, CCS costs have been declining as carbon capture technology advances. While the cost of capture has been largely dependent on the concentration of CO₂ in the gas stream, advancements in varying individual CCS technologies tend to drive down the cost of capture for other CCS technologies. The increase in CCS investment is already driving down the costs of near-future CCS technologies. The Global CCS Institute has tracked publicly available information on previously studied, executed, and proposed CO₂ capture projects. The cost of CO₂ capture from low-to-medium partial pressure sources such as coal-fired power generation has been trending downward over the past decade, and is projected to fall by 50 percent by 2025 compared to 2010. This is driven by the familiar learning-processes that accompany the deployment of any industrial technology. A review of learning rates (the reduction in cost for a doubling of production or capacity) for various energy related technologies similar to carbon capture (flue gas desulfurization, selective catalytic reduction, combined cycle turbines, pulverized coal boilers, LNG production, oxygen production, and hydrogen production via steam methane reforming) demonstrated learning rates of 5 percent to 27 percent for both capital expenditures and operations and maintenance costs. Studies of the cost of capture and compression of CO₂ from power stations completed 10 years ago averaged around $95/metric ton ($2020). Comparable studies completed in 2018/2019 estimated capture and compression costs could fall to approximately $50/metric ton CO₂ by 2025. Current target pricing for announced projects at coal-fired steam generating units is approximately $40/metric ton on average, compared to Boundary Dam whose actual costs were reported to be $105/metric ton, noting that these estimates do not include the impact of the 45Q tax credit as enhanced by the IRA. Additionally, IEA suggests this trend will continue in the future as technology advancements “spill over” into other projects to reduce costs. Similarly, EIA incorporates a minimum 20 percent reduction in carbon capture and sequestration costs by 2035 in their Annual Energy Outlook 2023 modeling in part to account for the impact of spillover and international learning. The Annual Technology Baseline published by NREL with input from NETL projects a 10 percent reduction in capital expenditures from 2021 through 2032 in the “Conservative Technology Innovation Scenario” for natural gas carbon capture retrofit projects, under the assumption that only learning processes lead to future cost reductions and that there are no additional improvements from investments in targeted technology research and development. In a recent case study of the cost and performance of carbon capture retrofits on existing natural gas combined cycle units, based on discussions with external technology providers, engineering consultants, asset developers, and applicants for DOE awards, DOE used a 25 percent capital cost reduction estimate to illustrate the potential future capital costs of an Nth-of-a-Kind facility, as well as “conservatively model[ing]” operating expense reductions at 1 percent, for a combined overall decrease in the levelized cost of energy of about 10 percent for the Nth-of-a-Kind facility compared to a First-of-a-Kind facility. DOE further found this illustrative cost reduction estimate from learning through doing to be consistent with other studies that use hybrid engineering-economic and experience-curve approaches to estimate potential decreases in the levelized cost of energy of 10-11 percent for Nth-of-a-Kind plants compared with First-of-a-Kind plants. Policies in the IIJA and IRA are further increasing investment in CCS technology that can accelerate the pace of innovation and deployment.

(2) CO₂ Transport and Sequestration Costs

NETL's “Quality Guidelines for Energy System Studies; Carbon Dioxide Transport and Sequestration Costs in NETL Studies” provides an estimation of transport costs based on the CO₂ Transport Cost Model. The CO₂ Transport Cost Model estimates costs for a single point-to-point pipeline. Estimated costs reflect pipeline capital costs, related capital expenditures, and operations and maintenance costs.

NETL's Quality Guidelines also provide an estimate of sequestration costs. These costs reflect the cost of site screening and evaluation, permitting and construction costs, the cost of injection wells, the cost of injection equipment, operation and maintenance costs, pore volume acquisition expense, and long-term liability protection. Permitting and construction costs also reflect the regulatory requirements of the UIC Class VI program and GHGRP subpart RR for geologic sequestration of CO₂ in deep saline formations. NETL calculates these sequestration costs on the basis of generic plant locations in the Midwest, Texas, North Dakota, and Montana, as described in the NETL energy system studies.

There are two primary cost drivers for a CO₂ sequestration project: the rate of injection of the CO₂ into the reservoir and the areal extent of the CO₂ plume in the reservoir. The rate of injection depends, in part, on the thickness of the reservoir and its permeability. Thick, permeable reservoirs provide for better injection and fewer injection wells. The areal extent of the CO₂ plume depends on the sequestration capacity of the reservoir. Thick, porous reservoirs with a good sequestration coefficient will present a small areal extent for the CO₂ plume and have lower testing and monitoring costs. NETL's Quality Guidelines model costs for a given cumulative storage potential.

In addition, provisions in the IIJA and IRA are expected to significantly increase the CO₂ pipeline infrastructure and development of sequestration sites, which, in turn, are expected to result in further cost reductions for the application of CCS at a new combined cycle EGUs. The IIJA establishes a new Carbon Dioxide Transportation Infrastructure Finance and Innovation program to provide direct loans, loan guarantees, and grants to CO₂ infrastructure projects, such as pipelines, rail transport, ships and barges. The IIJA also establishes a new Regional Direct Air Capture Hubs program which includes funds to support four large-scale, regional direct air capture hubs and more broadly support projects that could be developed into a regional or inter-regional network to facilitate sequestration or utilization. DOE is additionally implementing IIJA section 40305 (Carbon Storage Validation and Testing) through its CarbonSAFE initiative, which aims to further development of geographically widespread, commercial-scale, safe storage. The IRA increases and extends the IRC section 45Q tax credit, discussed next.