Air Quality Issues and Animal Agriculture:
EPA’s Air Compliance Agreement
Claudia Copeland
Specialist in Resources and Environmental Policy
July 20, 2012
Congressional Research Service
7-5700
www.crs.gov
RL32947
CRS Report for Congress
Pr
epared for Members and Committees of Congress
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
Summary
From an environmental quality standpoint, much of the interest in animal agriculture has focused
on impacts on water resources, because animal waste, if not properly managed, can harm water
quality through surface runoff, direct discharges, spills, and leaching into soil and groundwater. A
more recent issue is the contribution of emissions from animal feeding operations (AFO),
enterprises where animals are raised in confinement, to air pollution. AFOs can affect air quality
through emissions of gases such as ammonia and hydrogen sulfide, particulate matter, volatile
organic compounds, hazardous air pollutants, and odor. These pollutants and compounds have a
number of environmental and human health effects.
Agricultural operations that emit large quantities of air pollutants may be subject to Clean Air Act
(CAA) regulation and permits. Further, some livestock operations also may be regulated under
the release reporting requirements of the Comprehensive Environmental Response,
Compensation, and Liability Act (Superfund, or CERCLA) and the Emergency Planning and
Community Right-to-Know Act (EPCRA). Questions about the applicability of these laws to
livestock and poultry operations have been controversial and have drawn congressional attention.
Enforcement of these federal environmental laws requires accurate measurement of emissions to
determine whether regulated pollutants are emitted in quantities that exceed specified thresholds.
Yet experts believe that existing data provide a poor basis for regulating and managing air
emissions from AFOs. In an effort to collect scientifically credible data, in 2005 the
Environmental Protection Agency (EPA) announced a plan that had been negotiated with
segments of the animal agriculture industry. Called the Air Compliance Agreement, it is intended
to produce air quality monitoring data on AFO emissions during a two-year study, while at the
same time protecting participants through a “safe harbor” from liability under certain provisions
of federal environmental laws. Many producer groups supported the agreement as essential to
gathering valid data that are needed for decision making. However, critics, including
environmentalists and state and local air quality officials, said that the agreement would grant all
participating producers a sweeping liability shield for violations of environmental laws, yet
because fewer than 30 farms would be monitored, it was too limited in scope to yield
scientifically credible estimates of AFO emissions. Some industry groups had their own questions
and reservations. In August 2006, EPA approved agreements with 2,568 AFOs, representing
nearly 14,000 farms. Monitoring of 25 farms in nine states occurred from mid-2007 to the end of
2009. In January 2011, EPA released the data from the individual monitored sites and began
developing improved emissions estimating methodologies (EEMs) based on the data. Draft EEMs
for some animal sectors were released for review and public comment in February 2012 and have
been widely critiqued, including by EPA’s science advisers.
Separately from the monitoring study, in December 2008, EPA issued a rule to exempt animal
waste emissions to the air from most CERCLA and EPCRA reporting requirements. Legal
challenges to the rule followed. In October 2010, a federal court approved the government’s
request to remand the rule to EPA for reconsideration and possible modification.
This report reviews key issues associated with the Air Compliance Agreement. Background
information on air emissions from poultry and livestock operations, relevant federal
environmental laws and regulations, congressional interest, state activities, and research needs are
discussed in CRS Report RL32948, Air Quality Issues and Animal Agriculture: A Primer, by
Claudia Copeland.
Congressional Research Service
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
Contents
Introduction ...................................................................................................................................... 1
Background ...................................................................................................................................... 1
EPA’s Air Compliance Agreement with Industry ............................................................................. 3
Critiques of the Air Compliance Agreement ............................................................................. 5
Environmental Advocates and Air Program Administrators ............................................... 6
Other Animal Producers ...................................................................................................... 8
Status and Current Developments .................................................................................................. 10
CERCLA/EPCRA Reporting Exemption ...................................................................................... 12
Congressional Interest ............................................................................................................. 14
Contacts
Author Contact Information........................................................................................................... 15
Congressional Research Service
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
Introduction
From an environmental quality standpoint, much of the public and policy interest in animal
agriculture has focused on impacts on water resources, because animal waste, if not properly
managed, can adversely impact water quality through surface runoff and erosion, direct
discharges to surface waters, spills and other dry-weather discharges, and leaching into soil and
groundwater. However, animal feeding operations (AFO), enterprises where animals are kept and
raised in confinement, can also result in emissions to the air of particles and gases such as
ammonia, hydrogen sulfide, and volatile organic chemicals. At issue are questions about the
contribution of AFOs to total air pollution and corresponding ecological and possible public
health effects.
The Environmental Protection Agency (EPA) has authority to address AFO air emissions under
several laws—the Clean Air Act, Comprehensive Environmental Response, Compensation, and
Liability Act, and the Emergency Planning and Community Right-to-Know Act. Implementation
and enforcement of these laws requires scientifically credible data on air emissions and accurate
measurement of emissions to determine whether regulated pollutants are emitted in quantities that
exceed specified thresholds.
This report discusses an EPA plan called the Air Compliance Agreement intended to produce air
quality monitoring data on animal agriculture emissions from a small number of farms, while at
the same time protecting all participants (including farms where no monitoring takes place)
through a “safe harbor” from liability under certain provisions of federal environmental laws.
Some industry sectors involved in negotiating the agreement, which was announced in January
2005—notably many pork and egg producers—strongly supported it, but other industry groups
that were not involved in the discussions had concerns and reservations. State and local air quality
officials and environmental groups opposed the agreement. The monitoring phase of the plan is
complete, and EPA is now using the data to determine emissions estimating methodologies, as
discussed below.
Background1
AFOs2 can affect air quality through emissions of gases (ammonia and hydrogen sulfide),
particulate matter, volatile organic compounds, hazardous air pollutants, microorganisms, and
odor. AFOs also produce gases (carbon dioxide and methane) that are associated with climate
change. The generation rates of odor, manure, gases, particulates and other constituents vary with
weather, time, animal species, type of housing, manure handling system, feed type, and
management system (storage, handling, and stabilization).
1 For more extensive discussion, see CRS Report RL32948, Air Quality Issues and Animal Agriculture: A Primer, by
Claudia Copeland. That report provides general background information on air emissions from poultry and livestock
operations, their sources and health and environmental effects, relevant federal environmental statutes and regulations,
congressional interest in these issues, state activities, and research needs.
2 Under EPA regulations, an AFO is a facility in which livestock or poultry are raised or housed in confinement, and
where the following conditions are met: (1) animals are confined or maintained for a total of 45 days or more in any 12-
month period, and (2) crops are not sustained in a normal growing season over any portion of the lot or facility (i.e.,
animals are not maintained in a pasture or on rangeland). 40 CFR §122.23(b).
Congressional Research Service
1
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
Emission sources include barns, feedlot surfaces, manure storage and treatment units, silage piles,
animal composting structures, and other smaller sources, but air emissions come mostly from the
microbial breakdown of manure stored in pits or lagoons and spread on fields. Pollutants
associated with AFOs have a number of environmental and human health impacts. Most of the
concern with possible health effects focuses on ammonia, hydrogen sulfide, and particulate
matter, while major ecological effects are associated with ammonia, particulates, methane, and
oxides of nitrogen.
The animal sector of agriculture has undergone major changes in the last several decades, a fact
that has drawn the attention of policymakers and the public. In the United States there are an
estimated 238,000 animal feeding operations where livestock and poultry are confined, reared,
and fed, according to the U.S. Department of Agriculture’s 1997 Census of Agriculture.
Organizational changes within the industry to enhance economic efficiency have resulted in larger
confined production facilities that often are geographically concentrated.3 The driving forces
behind structural change in livestock and poultry production are no different than those that affect
many other industries: technological innovation and economies of scale.4 From 1982 to 1997, the
total number of U.S. operations with confined livestock fell by 27%. At the same time, the
number of animals raised at large feedlots (generally confining 300 animals or more) increased by
88%, and the number of large feedlots increased by more than 50%.5 The traditional image of
small farms, located in isolated, rural locales, has given way to very large farming operations,
some on the scale of industrial activities. Increased facility size and regional concentration of
livestock and poultry operations have, in turn, given rise to concerns over the management of
animal wastes from these facilities and potential impacts on environmental quality.
Agricultural operations often have been treated differently from other types of businesses under
numerous federal and state laws. Some laws specifically exempt agriculture from regulatory
provisions, and some are structured in such a way that farms escape most, if not all, of the
regulatory impact. Moreover, in implementing environmental laws, federal and state regulators
have traditionally focused most effort on controlling the largest and most visible sources of
pollution to the water, air, and land—factories, waste treatment plants, motor vehicles—rather
than smaller and more dispersed sources such as farms.
Nevertheless, certain large animal feeding operations are subject to environmental regulation. The
primary regulatory focus has been on protecting water resources and has occurred under the
Clean Water Act. While air emissions from farms typically do not exceed thresholds specified in
the Clean Air Act (CAA) and thus generally escape most CAA regulatory programs, facilities that
emit large quantities of air pollutants may be regulated under the act and state programs which
implement the CAA. A number of state air quality programs supplement federal CAA
requirements with facility construction and operation permits, air quality standards for odor and
certain AFO pollutants, monitoring, inspection, and testing. Some observers believe that
3 For additional information, see CRS Report RL33325, Livestock Marketing and Competition Issues, by Renée
Johnson and Geoffrey S. Becker.
4 Marc Ribaudo et al., U.S. Department of Agriculture, Economic Research Service, Manure Management for Water
Quality: Costs to Animal Feeding Operations of Applying Manure Nutrients to Land, June 2003, Agricultural
Economic Report 824, 87 pp.
5 U.S. Department of Agriculture, Natural Resources Conservation Service, Manure Nutrients Relative to the Capacity
of Cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal Trends for the United States, Publication
no. nps00-0579, December 2000, p. 18.
Congressional Research Service
2
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
increased federal and state attention to air emissions from AFOs, precipitated in part by structural
changes in animal production and public concern, will likely lead to stricter federal regulation.6
Some livestock operations may also be subject to the release reporting requirements of the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA, the
Superfund law) and the Emergency Planning and Community Right-to-Know Act (EPCRA). The
reporting requirements of these laws are triggered when large quantities of certain substances are
released to the environment, including ambient air. Livestock facilities emit hydrogen sulfide and
ammonia, which are reportable substances under these laws. There has been little enforcement of
these provisions against livestock operations, but in lawsuits brought by citizen groups, federal
courts in two circuits found AFOs in violation of the reporting requirement provisions of the laws.
Applicability of other provisions of CERCLA to agriculture (provisions concerning liability for
costs of cleanup of hazardous substance releases and recovery for damages of releases to natural
resources) also have drawn attention. The net result has been concern by the agriculture
community that other legal actions will be brought, thus potentially exposing more of these
operations to enforcement under federal law.
EPA’s Air Compliance Agreement with Industry
Enforcement of applicable provisions of federal environmental laws such as CERCLA, EPRCRA,
and the Clean Air Act (CAA) requires accurate measurement of emissions to determine whether
facilities and operations emit regulated pollutants in quantities that exceed specified thresholds.
Monitoring air emissions from feedlots, waste lagoons, animal confinement buildings and other
components of livestock facilities is complex and has been controversial. Resolving questions
about AFOs’ contribution to total air pollution and corresponding ecological and possible public
health effects is hindered by a lack of adequate, accurate, scientifically credible data on air
emissions. At the same time, increasing public concern about AFO emissions and the possibility
of enforcement actions brought against large AFOs seeking compliance with environmental laws,
have led to efforts to gather more and better data.
Early in 2002, representatives of some agriculture industry groups—especially pork and egg
producers—approached EPA officials with a proposal to negotiate a voluntary agreement that
would produce air quality monitoring data on emissions from animal feedlot operations.
Discussions between EPA and the industry groups continued for more than two years and
eventually led to a plan, called the Air Compliance Agreement, that EPA announced in January
2005. It was published in the Federal Register on January 31, 2005, thus triggering a 90-day
period during which AFOs could sign up to participate in the agreement.7 The signup period was
subsequently extended to August 12, 2005, in order to provide more time for AFO operators to
make decisions about participation.
The agreement was intended to enable scientists to collect and analyze emissions data and create
tools that AFOs could use to estimate their emissions, for purposes of regulatory compliance,
while at the same time protecting participating AFOs under a “safe harbor” in which EPA granted
6 Jody M. Endres and Margaret Rosso Grossman, “Air Emissions from Animal Feeding Operations: Can State Rules
Help?” Pennsylvania State Environmental Law Review, vol. 13, Fall 2004, p. 5.
7 U.S. Environmental Protection Agency, “Animal Feeding Operations Consent Agreement and Final Order,” 70
Federal Register 4958, January 31, 2005.
Congressional Research Service
3
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
covenants not to sue and released participants from EPA liability for failing to comply with
certain provisions of the CAA, CERCLA, and EPCRA. EPA retained the authority to respond to
an imminent and substantial endangerment to public health or the environment, and participants
were not protected against liability for criminal violations of environmental laws.
The agreement applied to AFOs in the egg, broiler chicken, turkey, dairy cattle, and swine
industries. (It did not address AFOs that only have open-air feedlots, such as cattle feedlots.)
Those that signed up to participate paid a civil penalty ranging from $200 to $1,000, depending
on the number of animals at the AFO, and contributed $2,500 per farm to implement a nationwide
air monitoring program for AFOs. EPA estimated that as many as 4,000 AFOs might sign up to
participate in the agreement. Of those that signed agreements with EPA, a small number—
perhaps no more than three dozen representative farms nationwide—would be selected to
participate in on-farm monitoring, but all who signed up were protected by EPA’s covenant not to
sue. EPA reserved the right to decide not to go forward with the agreement and monitoring study
if, for example, an insufficient number of AFOs signs up to generate the $10 million to $12
million estimated to be needed for the study, or if some individual animal groups were under-
represented. EPA also could decline to enter into agreement with an individual AFO if, for
example, it was the subject of ongoing federal, state or local environmental enforcement.
EPA expected that within 30 days after the end of the sign-up period (August 12, 2005), agency
officials would decide whether to proceed with all, part, or none of the monitoring study and sign
the Air Compliance Agreements submitted by industry participants. (As described below, this
process took longer than was anticipated.) Signed agreements would then be forwarded to the
agency’s Environmental Appeals Board (EAB) for final approval. Unlike civil enforcement
actions that are resolved by judicially approved consent decrees, the Air Compliance Agreements
were administrative agreements. Among other responsibilities, the EAB is the final EPA
decisionmaker on administrative appeals under all major environmental statutes that the agency
administers.
Monies collected from participants were to go to a nonprofit organization (NPO) set up by the
AFOs, called the Agricultural Air Research Council.8 The NPO, in turn, was to subcontract with a
science advisor and independent monitoring contractor to run the monitoring study, including
recommending facilities to be monitored. EPA’s role was to review and approve the contractor’s
study plan and, later, to use and analyze the data generated by the study. EPA also collaborated
with industry and other stakeholders to develop protocols for the study, which were published
with the January 2005 Notice of the agreement. Monitoring was to continue for two years.
EPA will use the data and other relevant, available data to develop methodologies for estimating
annual emissions, which are commonly used where site-specific monitoring data are not
available. With use of the methodologies, AFOs will be able to determine whether they are
complying with applicable notification provisions of federal law, as well as applicable
requirements of the CAA. EPA expected that within 18 months after the conclusion of the
nationwide monitoring study, it would publish emission-estimating methodologies for AFOs in
the eligible animal groups. Once the methodologies are final and published, an AFO will have
120 days to apply the methodologies to its facilities, apply for all applicable air permits and
comply with permit conditions, and report any qualifying releases of ammonia and hydrogen
8 Ibid., p. 4970.
Congressional Research Service
4
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
sulfide as required by CERCLA and EPCRA.9 Under the agreement, the EPA’s covenant not to
sue and waiver from liability covered an AFO’s liability for failing to comply with certain
provisions of CERCLA, EPCRA, and the CAA retroactively and from the start of the agreement
up to the time it reports releases and applies for and receives CAA permits (i.e., 120 days after
publication of estimating methodologies) or December 31, 2011, whichever is earlier. This time
period can be extended by mutual agreement of EPA and participants, without limit to how long
such an extension might last.10
Critiques of the Air Compliance Agreement
In comments submitted to EPA, many livestock and poultry groups and individual producers
supported the Air Compliance Agreement—especially those that expected to participate in it. In
their view, comprehensive, valid data are needed to develop appropriate public policy regarding
emissions from animal agriculture operations. The air monitoring study linked to the agreement is
an important effort to establish the criteria that farmers and regulators need to correctly interpret
agricultural compliance requirements. Supporters believe that data from the study will enable
EPA to produce charts that livestock and poultry producers can use to know whether their farms
are subject to federal environmental laws.
Additionally, supporters said that producers need the protection provided by the agreement in
order to volunteer their farms for participation in the study. Without this protection, there was no
incentive for producers to participate in the research, because the potential penalties for alleged
past violations are so great. Many among those who supported the agreement believed that
livestock operations should be entirely exempt from CERCLA and EPCRA reporting
requirements because, in their view, Congress did not intend for these laws to apply to animal
agriculture. Several groups, including cattle feedlots (even though they are not included in the
compliance agreement) and chicken and turkey producers, have for some time requested that EPA
resolve the issue for producers through a finding or guidance to clarify that animal agriculture
facilities are not subject to CERCLA and EPCRA. They fear that, barring statutory change or
some clarification from EPA, the courts will continue to rule that the laws do apply to animal
agriculture. Thus, they view the monitoring study, and the legal protection provided under it, as
an incentive to participants that will provide the data needed to determine on a national scale
which farms are subject to compliance with regulatory requirements.
State and local air quality officials and members of the environmental advocacy community
strongly objected to the agreement, which some characterized as a grant of “retrospective and
prospective immunity from liability” for every AFO in the United States, a sweeping liability
shield to the entire industry.11 Environmental groups and air program administrators were not
included in EPA-industry negotiations on the agreement, but several draft versions of the
agreement document were publicly circulated throughout the period of its development. Letters to
9 As described below (“CERCLA/EPCRA Reporting Exemption”), in 2008 EPA finalized a rule that exempts
hazardous substance releases that are emitted to the air from animal waste at farms from the notification requirements
of CERCLA and, with the exception of large concentrated animal feeding operations (CAFOs), the notification
requirements of EPCRA.
10 70 Federal Register 4964, January 31, 2005.
11 Brent Newell et al. (representatives of six environmental organizations), letter to Christine Todd Whitman (EPA
Administrator), May 5, 2003, pp. 4-5.
Congressional Research Service
5
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
EPA objecting to the proposal were sent by both,12 and environmental groups unsuccessfully
attempted to halt the plan with a September 2003 lawsuit alleging that EPA had violated the
Freedom of Information Act by failing to disclose documents about the proposed agreement. A
legal challenge to the Air Compliance Agreement was brought by several environmental
advocacy groups. The lawsuit was dismissed in July 2007, when the court held that the
agreements constitute an action that is not reviewable by the court, because the agreements fall
within the enforcement discretion of EPA. In a dissenting opinion, one judge said that the
agreement is broader than a discretionary enforcement action, because it could be in force for
years while EPA formulates an emissions regulatory program tailored to livestock operations
(Association of Irritated Residents v. EPA, No. 05-1177, D.C. Cir., July 17, 2007).
Not all industry groups were fully supportive of the agreement, for a number of reasons. Some
agriculture industry groups that did not participate in negotiating the compliance agreement had a
number of their own concerns. Issues presented in critical comments submitted on the January
2005 publication of the agreement addressed a number of points.13
Environmental Advocates and Air Program Administrators
Environmental critics argued that the agreement unlawfully exempts AFOs from requirements of
the Clean Air Act, CERCLA, and EPCRA. They argued that EPA has no authority to defer a
major stationary source’s or a facility’s compliance with these laws, through permit deferrals or
requirements. These opponents argued that the broad liability shield provided by the agreement is
not justified by contending that there is a lack of data. They pointed to research that has been
conducted for quite some time by academic and government researchers (including USDA) that
has documented emissions and adverse health and environmental effects from AFO emissions.
Further, they argued that EPA has authority under CAA Section 114 to require that AFOs provide
emission monitoring data, without the need to provide an industry-wide exemption. In the view of
environmentalists, the penalties required under the agreement (averaging $500 per farm) were a
“payment to pollute,” especially compared with penalties available to EPA under those laws
($27,500 for each civil violation). EPA’s position was that the agreement is the quickest and most
effective way to address the current uncertainties regarding air emissions and to bring the entire
AFO industry into compliance with the CAA, CERCLA, and EPCRA, in contrast to lengthy
litigation and case-by-case enforcement of the laws.14
Environmental critics also were concerned that the agreement did not require AFOs to reduce
pollution. EPA’s publication of emission-estimating methodologies will trigger the obligation of
participating AFOs to determine their emissions and to comply with all applicable CAA
requirements (including permits) and CERCLA and EPCRA reporting requirements. Critics said,
however, that it did not guarantee air pollution controls at any AFO or even require participants to
test technologies or management practices to reduce their emissions, although all AFOs were
12 See, for example, Lloyd L. Eagan (President of State and Territorial Air Pollution Program Administrators) and Ellen
Garvey (President of Association of Local Air Pollution Control Officials), letter to Christine Todd Whitman (EPA
Administrator), April 7, 2003; Shelley Kaderly, STAPPA Agriculture Committee Chair) and Doug Quetin (ALAPCO
Agriculture Committee Chair), letter to Robert Kaplan (EPA Office of Enforcement and Compliance Assistance),
February 18, 2004; and Brent Newell et al. (representatives of six environmental organizations), letter to Christine
Todd Whitman, May 5, 2003.
13 Materials in the EPA docket for the Air Compliance Agreement, EPA-HQ-OAR-2004-0237, can be found at
http://www.regulations.gov/.
14 70 Federal Register 4958, January 31, 2005.
Congressional Research Service
6
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
eligible to secure a lengthy, perhaps indefinite CAA amnesty. At the end of the study EPA could
make regulatory or policy decisions that would leave AFO emissions unregulated, they said, even
if monitoring indicates there are emissions in amounts that would be of concern. In addition, they
were critical of the open-ended timelines in the agreement (especially the 18 months after
monitoring when EPA expects to publish emission-estimating methodologies): if EPA fails to
issue the methodologies, the waiver could last indefinitely, they said.
A number of commenters criticized the small number of sites that EPA expected would be
monitored: the final study plan included 25 farms in 10 states. Such a small number, critics said,
would be insufficient to develop emission-estimating methodologies for all of the covered animal
sectors and possible farm configurations and geographic locations. In response, EPA said that its
technical experts believed that the monitoring protocol will provide sufficient data to get a valid
representative sample. Moreover, significantly increasing the number of farms to be monitored
would be prohibitively expensive and would not add substantially to the value of the data
collected, according to EPA.15
Critics also said that the small sample size for monitoring was inconsistent with recommendations
made by the National Research Council (NRC) of the National Academy of Sciences calling for a
process-based rather than a model farm approach to estimating emissions.16 EPA’s current
approach is to develop emission factors based on representative AFOs, estimating emissions as
the product of the specific factor for a particular animal production sector and the number of
animals associated with the farm or geographic region. To determine accurate emission factors
using this approach requires considerable amounts of data that explain variations in emissions.
The NRC’s preferred approach is to develop process-based models that analyze the component
parts of a farm enterprise, using research studies and mass balance equations to simulate
conversion (changes in form of relevant compounds) and transfers (changes in location of
compounds), as well as emissions. Estimates would be based on understanding of processes
inherent in animal production. EPA said that developing a process-based model of emissions is
part of the agency’s long-term strategy but will take a period of years.
Other critics said that the monitoring protocol under the agreement lacked adequate peer review
and involvement of qualified, independent scientists who were not involved in its formulation. To
assure the scientific rigor of the monitoring program, some commenters recommended an
independent peer review process using reviewers with no active ties to the livestock industry. In
June 2007, when EPA announced that the study was ready to proceed, agency officials said that
Purdue and the other participating universities developed 2,000 pages of protocol for the design
of the study and that it was peer-reviewed by EPA and its contractors, as well as by outside
research groups. Still, some critics complained that the public had not been notified or involved in
the external review process.
State and local air quality officials said that the agreement interferes with their ability to attain air
quality standards and enforce air pollution control laws. In their view, several of the agreement’s
provisions were unclear and could be interpreted to limit the ability of states and localities to
enforce air laws. These groups, along with environmentalists, were greatly concerned that the
15 70 Federal Register 4960, January 31, 2005.
16 In 2001 EPA asked the National Research Council of the National Academy of Sciences for a report evaluating the
current scientific knowledge base and approaches for estimating air emissions from AFOs. Two NRC reports prepared
in response to this request are discussed in CRS Report RL32948, Air Quality Issues and Animal Agriculture: A
Primer, by Claudia Copeland.
Congressional Research Service
7
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
broad waiver of liability would curtail state or local and citizen enforcement, or, at the very least,
create a very high hurdle for enforcement. The agreement said that it was not intended to affect
the ability of states or citizens to enforce applicable state laws. However, these critics contended
that, by saying that the agreement resolved an AFO’s civil liability for certain potential violations,
it seriously raised the bar for state or citizen enforcement, since a participating AFO might claim
in an enforcement action that the agreement provided immunity from state laws or local
ordinances. EPA’s position was that the agreement did not undermine state or local enforcement
authorities and had no impact on the most important state enforcement tools, including zoning
classification, state permits, nuisance actions, workplace regulations, and health and safety laws.
Further, the agreement did not affect the ability of regulators to bring an action under emergency
provisions of the Clean Air Act and other statutes in order to prevent an imminent and substantial
endangerment to public health, welfare, or the environment.
EPA also was criticized for failing to resolve two important definitional issues. In the Notice
announcing the agreement, EPA said that after the monitoring study is complete, it would issue
guidance or a rule on whether to treat emissions from different areas at AFOs as fugitive or
nonfugitive emissions. Fugitive emissions are not counted for purposes of determining whether
under the Clean Air Act a source is major or minor and, thus, subject to pollution controls. Critics
said that EPA should clarify this important issue quickly, should do so in consultation with states
and localities, and should take any action through a formal rulemaking, not a guidance document.
Similarly, EPA said that at the end of the monitoring study, it would issue guidance on the scope
of the term “source” as it relates to animal agriculture and farm activities.17 State and local air
quality officials were concerned that, like the fugitive emissions issue, EPA could define “source”
in such a way that emissions from AFOs do not rise to a threshold of regulatory concern. In their
view, this would be contrary to federal court rulings in cases concerning applicability of
CERCLA and EPCRA reporting requirements to AFOs. States and localities believe that the laws
should be interpreted liberally to accomplish goals of cleaning up and maintaining clean air.
Other Animal Producers
Critical comments on the agreement also were submitted by some industry groups that did not
participate in negotiations with EPA to develop the program, but might be expected to participate
in the agreement. A number of commenters from the dairy farming and broiler and turkey
producer industries noted confusion about many details of the agreement, especially for small
farmers, resulting in uncertainty about implications and costs to them of participating in it (actual
costs and transaction costs). Several asked EPA to review public comments on the agreement,
make suggested changes where appropriate, and allow producers and processors additional time
to sign up, once a final agreement was published. Extending the signup period would allow
groups that are less familiar with the agreement the time that they need to assess it, they said.
Based partly on requests for additional time, EPA did extend the signup deadline until August 12,
2005, but the agreement remained unchanged from what was published in January 2005.
A number of industry commenters objected that the agreement required an admission of liability
and that the term “civil penalty,” which participants were required to pay in order to participate,
carries negative connotations that imply guilt. Some companies objected to having to pay to
resolve unproven violations. EPA responded that, by voluntarily signing the agreement, farmers
17 70 Federal Register 4959, January 31, 2005.
Congressional Research Service
8
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
were not admitting any liability or any sort of wrongdoing. Payment of a penalty was part of the
process to obtain a release from liability for possible violations, according to EPA, and was not
intended to be used for any purposes other than this agreement. In EPA’s view, signing the
agreement was not an admission that participating agricultural operations have been operated
negligently or improperly or in violation of any federal, state, or local law or rule.18
Some dairy farmers also raised concerns that the agreement could jeopardize their role in farm
programs, bank loans, and insurance policies. In response, the Secretary of Agriculture informed
members of Congress that the department had concluded that “voluntary participation in the Air
Compliance Agreement by a producer or processor will not cause the producer or processor to be
ineligible for USDA programs.”19
Both poultry producer groups (sometimes called the meat-bird sector, in contrast to the egg-
laying segment of poultry) and dairy groups said that they would prefer to work with a nonprofit
organization of their own choosing to manage their participation (handling funding, monitoring
facilities, presenting the data), rather than a single organization selected to represent all of the
industry. Purdue University was selected to manage the study. The dairy industry preferred to
work with its own Dairy Environmental Task Force, which already was addressing dairy air
quality issues, and poultry and egg producers preferred to work with researchers that they believe
are more familiar with their operations, such as scientists from the University of Georgia.
Producers in the poultry and dairy sectors also objected to the small number of sites that EPA
planned to monitor (for example, the protocol called for monitoring only four dairy farms and two
broiler operations across the country), saying that the proposed monitoring program was too
limited and that the data would not accurately reflect the variation or range of climatic,
geographic, and operational factors that influence emissions from facilities. Whereas the
environmentalists’ concerns about the small number of sites to be monitored was that the majority
of producers would benefit from the safe harbor without having to do anything, industry groups
had different concerns. They feared that EPA would impose future requirements that will be both
costly and scientifically inappropriate, because the limited monitoring under the protocol would
not adequately reflect different types of operations within specific sectors or for all segments of
animal agriculture. One commenter noted as follows:
[A]n insufficient number of farms are included in the monitoring to allow for the
development of models to estimate emissions from individual AFOs.... It is unclear how the
very limited number of representative farms selected, and the resulting emission estimating
methodologies, will result in data capable of accounting for the various differences in
management styles, feed regimes, water control and numerous other factors that can affect
emissions.20
Comments from poultry and dairy groups raised other concerns, including financial obstacles to
participating in the agreement. Dairy farmers noted that while some animal producers were able
18 U.S. Environmental Protection Agency, “Response to Public Comments on the Animal Feeding Operation Air
Agreement,” June 23, 2005. Available at http://www.epa.gov/compliance/resources/agreements/caa/cafo-agr-response-
com.html.
19 Mike Johanns, Secretary, USDA, Letter to the Honorable Robin Hayes, August 11, 2005.
20 Comments of C. M. Williams, F. J. Humenik, directors of the Animal and Poultry Waste Management Center and the
National Center for Manure and Animal Waste Management, on EPA Docket ID: OAR-2004-0237, March 2, 2005, p.
11.
Congressional Research Service
9
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
to use funds from national check-off programs to pay for the study so that individual producers
did not have to pay the costs out-of-pocket (e.g., the National Pork Board committed $6 million
of check-off funds for pork producers’ participation), the national dairy check-off program might
not be used to fund production-oriented research at the farm level.21 Thus, there was no central
mechanism to fund dairy farmers’ participation in the monitoring study.
A group of pork producers who operate small farms, called the Campaign for Family Farms, and
several individual hog farmers objected to use of mandatory pork check-off funds to support
producers’ participation in the Air Compliance Agreement. In May 2005 they petitioned the
Secretary of Agriculture to halt pork check-off commitments for expenses related to the
agreement. In their view, the EPA study was beyond the type of research and promotion that is
permissible under the Pork Promotion, Research, and Consumer Information Act, which
authorizes the check-off. According to the petitioners, the proposed use of pork check-off funds
was a means for large concentrated animal feeding operations (CAFOs) to buy legal immunity
from environmental laws that would not benefit those producers who are too small to be subject
to the CAA, CERCLA, or EPCRA.22 Despite this challenge, USDA approved use of the pork
check-off funds for the study.
Status and Current Developments
The signup period for participating in the agreement closed on August 12, 2005, and EPA then
began compiling and evaluating responses. Ultimately, 2,681 AFOS, representing more than
6,700 farms, signed up to participate. In November 2005, an initial group of agreements was
forwarded to the Agency’s Environmental Appeals Board for approval. By August 2006, the EAB
ratified a total of 2,568 agreements, representing approximately 13,900 farms in 42 states.
According to EPA, these farms comprise more than 90% of the largest animal feeding operations
in the United States. The total consists of 1,856 swine, 468 dairy, 204 egg-laying, and 40 broiler
operations. According to EPA, the EAB’s determination that the agreements are consistent with
applicable statutes and CAA regulations allowed the monitoring study to officially begin
developing quality assurance and site-specific monitoring plans for those livestock sectors.
In June 2007, EPA announced that the two-year air monitoring study was ready to proceed. The
study involved 25 swine, dairy, and poultry farms located in nine states and included monitoring
of both open source sites and barn sites.23 Farms chosen for monitoring were selected based on
location (relative to climate and typical practice), method of manure collection, and manure
storage structures and buildings relative to the surrounding terrain. Monitoring was managed by
Purdue University, designated by EPA as the Independent Research Contractor (IRC). The IRC
selected equipment and methods in consultation with EPA, and researchers from Purdue and
seven other universities carried out the actual monitoring, with EPA oversight. Monitored
pollutants included pollutants commonly emitted by AFOs: particulate matter, ammonia,
hydrogen sulfide, and volatile organic compounds. Greenhouse gas emissions (e.g., methane)
21 For background information on national check-off programs for promotion and research of crop and livestock
commodities, see CRS Report 95-353, Federal Farm Promotion (“Check-Off”) Programs, by Geoffrey S. Becker.
22 Mark McDowell et al., and the Campaign for Family Farms, “Petition before the Secretary of Agriculture,” AMA
PPRCIA Docket No. 05-0001, May 5, 2005, p. 8.
23 California, Indiana, Iowa, New York, North Carolina, Oklahoma, Texas, Washington, and Wisconsin. In addition,
monitoring of a single Kentucky site was done by Tyson Foods.
Congressional Research Service
10
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
were not measured as part of the study, although carbon dioxide was measured for quality
assurance purposes.
The monitoring study had three primary goals: (1) to quantify aerial pollutant emissions from
dairy, pork, egg, and broiler production facilities; (2) to provide reliable data for developing and
validating emissions models for livestock and poultry production and for comparison with
government regulatory thresholds; and (3) to promote a national consensus on methods and
procedures for measuring emissions from livestock operations.
As contemplated in the agreement, the monitoring was carried out from mid-2007 through the
end of 2009. Purdue University researchers then conducted final processing and reviews of the
data and prepared reports on the individual sites. On January 13, 2011, EPA released the data and
reports on the monitored AFOs.24 The agency has not yet issued a final summary report to
interpret all of the data, but an analysis was prepared by the Environmental Integrity Project
(EIP), a nonprofit organization that focuses on environmental enforcement issues. EIP’s analysis
found that, despite the small number of monitored sites, measured levels of several pollutants—
particles, ammonia, and hydrogen sulfide—exceeded CAA health-based standards, worker
protection standards, and federal emission reporting limits at some of the study sites. EIP was
critical of aspects of the study design (e.g., failure to measure short-term emissions at all sites,
and inclusion of “negative” values that could represent erroneous samples and thus may
underestimate pollution) and recommended that the data be thoroughly peer reviewed.25
EPA acknowledged the need for peer review of both the study data and methodologies for
estimating AFO emissions, which the agency began to develop based on the monitoring data. In
September 2011, EPA asked the agency’s Science Advisory Board (SAB) to provide peer review
on the methodologies.26
At the same time that it released the data from the study, EPA agreed to an AFO industry request
to issue a “Call for Information,” seeking any additional peer-reviewed monitoring data on AFO
emissions, along with information about how animals and waste are managed at specific sites.
EPA expected that the information will help ensure that the agency has the best available
information as it develops the improved emissions estimating methodologies. One reason for
seeking additional data, EPA said, was because the monitoring data alone are inadequate to
develop a process-based modeling approach that incorporates “mass balance” constraints to
determine emissions from AFOs, which was a key recommendation made in NRC reports in 2002
and 2003. Consequently, EPA’s current focus is on developing emissions-estimating
methodologies (EEMs), and the agency expects to develop a process-based modeling approach at
a later unspecified date.
In February 2012, EPA released draft EEMs for some animal sectors—one for broiler AFO
emissions of ammonia, hydrogen sulfide, particles, and VOCs, and one for ammonia emissions
from lagoons and basins at swine and dairy AFOs. EPA made the drafts available for public
24 See http://www.epa.gov/airquality/agmonitoring/data.html.
25 Environmental Integrity Project, “Hazardous Pollution from Factory Farms: An Analysis of EPA’s National Air
Emissions Monitoring Study Data,” March 2011, http://www.environmentalintegrity.org/documents/
HazardousPollutantsfromFactoryFarms.pdf.
26 The SAB was established pursuant to the Environmental Research, Development and Demonstration Authorization
Act (42 U.S.C. §4365) to provide independent scientific and technical peer review, advice, and recommendations to the
EPA Administrator.
Congressional Research Service
11
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
review and comment, and review by the SAB also has begun. Subsequently, EPA will develop
draft EEMs for emissions from egg-layers, swine, and dairy confinement houses and other
pollutants from swine and dairy lagoons and basins. The agency’s initial efforts to develop EEMs
based on the national monitoring study data have been widely criticized by stakeholder groups,
members of the public, and the agency’s science advisers.
• Much of the criticism so far focuses on significant limitations of the study data,
with critics observing that limited data from a small number of sites do not
adequately reflect large differences in animal types, farm operations, geography,
and other factors. Data limitations make it extremely difficult to effectively
predict emissions, critics say. This criticism was predictable from the time when
EPA announced that the monitoring study would only consider 25 farms
nationwide. For example, many commenters object to EPA’s development of a
single EEM that combines swine and dairy farms (different animal types) using
different waste management systems (i.e., lagoons and basins). EPA’s explanation
is that the monitoring study did not produce sufficiently robust data for
developing separate EEMs.
• The quality of the data also has been challenged. Key data are missing or
confounded, critics point out, with significant instances of negative and zero
emissions values. Commenters and EPA’s science advisers do not agree on
whether negative and zero emissions data should be used or not. Further, critics
say that EPA’s criteria for including certain data and excluding other data are
unclear or unsupported. Also, EPA rejected almost all data that were submitted in
response to the agency’s call for additional data to supplement the study, for
reasons such as use of protocols that differed from the EPA study or lack of peer
review. Critics contend that the submitted data could have increased robustness
of the monitored data.
• Many critics continue to fault EPA for not developing a process-based, whole-
farm modeling approach for estimating AFO emissions, as recommended by the
National Research Council (discussed above).
Some observers, especially among industry groups, argue that the available data are too flawed
and limited to support EPA’s current approach, while others, including states and environmental
groups, urge the agency to move forward, even with the imperfect data in hand, in order to
develop EEMs for air regulatory purposes.
CERCLA/EPCRA Reporting Exemption
In 2005, a group of poultry producers petitioned EPA for an exemption from EPCRA and
CERCLA emergency release reporting requirements, arguing that releases from poultry-growing
operations pose little or no risk to public health, while reporting imposes an undue burden on the
regulated community and government responders. In December 2007, EPA responded to the
poultry industry petition with a proposal to exempt releases of ammonia, hydrogen sulfide, and
other hazardous substances to the air from animal waste at farms from the notification
requirements of CERCLA and EPCRA.27 The exemption would apply to releases to the air from
27 U.S. Environmental Protection Agency, “CERCLA/EPCRA Administration Reporting Exemption for Air Releases
(continued...)
Congressional Research Service
12
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
manure, digestive emissions, and urea, including animal waste mixed with bedding, compost, and
other specified materials. EPA explained that the rule was justified because of the resource burden
to industry of complying with reporting requirements, since the agency cannot foresee a situation
where a response action would be taken as a result of notification of releases of hazardous
substances from animal waste at farms.
The proposal drew significant public response during the public comment period, including
criticism from environmental groups and some states. Some argued that an exemption was
premature, since EPA was moving forward with research on emissions levels under the Air
Compliance Agreement, and that the agreement could be undermined by a regulatory exemption.
Industry groups support the proposed exemption.
In September 2008, the Government Accountability Office (GAO) issued a report evaluating
EPA’s activities to regulate air emissions and water discharges from animal feeding operations.
GAO noted the criticism of EPA’s air emissions monitoring activities and concern that the
research may not produce sufficient information to shape future regulation. Moreover, GAO
questioned the basis for the proposed CERCLA/EPCRA exemption. “It is unclear how EPA made
this determination when it has not yet completed its data collection effort and does not yet know
the extent to which animal feeding operations are emitting these pollutants.”28
In December 2008, EPA finalized the CERCLA/EPCRA administrative reporting exemption with
some modifications to the original proposal.29 The final rule exempts hazardous substance
releases that are emitted to the air from animal waste at farms from the notification requirement
of CERCLA. As proposed, the final rule relieves all livestock operations of all size, not just
poultry farms, from CERCLA’s requirement to report hazardous substance releases to the air to
federal officials. In addition, the final rule provides a partial exemption for such releases from
EPCRA’s requirement to report releases to state and local emergency officials. Partially
responding to public comments, the final rule continues to apply EPCRA’s reporting requirement
to large animal feeding operations (those that are subject to permitting requirements under the
Clean Water Act), but exempts smaller facilities. A number of groups criticized the final rule,
again raising concern about the toxicity of chemicals such as ammonia and hydrogen sulfide that
are emitted from animal waste facilities and arguing the CERCLA and EPCRA do not authorize
administrative exemptions for specific industries.
A coalition of environmental advocates challenged the rule in federal court, as did the National
Pork Producers Council (Waterkeeper Alliance v. EPA, D.C. Cir., No. 09-1017). Environmental
advocates continue to argue that the entire rule is deficient, while the pork producers group
objects to the fact that the rule only partially exempted releases from EPCRA. Other industry
groups such as the National Chicken Council intervened in the litigation in support of the final
rule. Parties to the litigation entered into talks to mediate the issues, but in June 2010, the federal
government asked to remand the final rule for EPA to reconsider and possibly modify the rule.
(...continued)
of Hazardous Substances from Animal Waste,” 72 Federal Register 73700 (December 28, 2007).
28 U.S. Government Accountability Office, “Concentrated Animal Feeding Operations, EPA Needs More Information
and a Clearly Defined Strategy to Protect Air and Water Quality from Pollutants of Concern,” September 2008, GAO-
08-944, p. 7.
29 U.S. Environmental Protection Agency, “CERCLA/EPCRA Administrative Reporting Exemption for Air Releases,”
73 Federal Register 76948-76960, December 18, 2008.
Congressional Research Service
13
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
The court approved the government’s request for a remand in October 2010. EPA anticipates
proposing a new or revised rule in 2012. In the meantime, the rule remains in effect.
Congressional Interest
Congressional attention to the issues discussed in this report has been limited, with the result that
developments have proceeded largely by administrative and some judicial actions, not through
legislative policymaking.
Prior to release of the Air Compliance Agreement in January 2005, some individual members
wrote letters to EPA objecting to the pending plan. Some members also were critical of EPA’s
proposal to exempt routine animal waste air releases from CERCLA and EPCRA’s reporting
requirements, questioning the potential for harmful environmental and enforcement impacts of
the proposal.30 At a September 24, 2008, hearing where GAO’s report was discussed, several
House Energy and Commerce subcommittee members said that they were skeptical of the EPA’s
authority for a blanket exemption. Others suggested that an exemption for small farms, whose
emissions are unlikely to cause environmental harm, would make sense. EPA and USDA
witnesses supported the proposal, saying that the air release waiver would only affect reporting
meant for emergency response situations, but would not affect requirements to report emissions
of hazardous substances from other farm sources, or releases of hazardous substances from
manure into soil, ground water, or surface water.
No legislation regarding the Air Compliance Agreement has been introduced. However, in several
Congresses since 2005, legislation has been introduced that would exclude “manure” from the
definition of hazardous substances under CERCLA and remove reporting liability under
CERCLA and EPCRA for releases of manure. Proponents have argued that Congress did not
intend that either of these laws apply to agriculture and that enforcement and regulatory
mechanisms under other laws are adequate to address environmental releases from animal
agriculture. Opponents responded that enacting a statutory exemption would severely hamper the
ability of government and citizens to know about and respond to releases of hazardous substances
caused by an animal agriculture operation.31
In the 112th Congress, bills on this topic are H.R. 2997 and S. 1729. Both are intended to clarify
that manure is not a “hazardous substance” or “pollutant or contaminant” under CERCLA and to
remove emissions reporting liability under CERCLA and EPCRA. Supporters of these bills seek
to block EPA from revising the 2008 exemption rule so as to require reporting of releases. A
House Energy and Commerce subcommittee held a hearing on H.R. 2997 on June 27, 2012. At
that hearing, an EPA witness said that the agency “has concerns with the broad impacts of H.R.
2997,” which the witness described.
The effect of the bill would be to prevent the EPA from using CERCLA response authorities
to respond to releases to the environment when manure is the source of those hazardous
substances, even if the release, for instances such as the failure of a large manure waste
lagoon, presented a substantial danger to the public health and the environment. It would also
30 Letter from Reps. John Dingell, Albert Wynn, Hilda Solis to Stephen L. Johnson, EPA Administrator, March 18,
2008.
31 For additional information, see CRS Report RL33691, Animal Waste and Hazardous Substances: Current Laws and
Legislative Issues, by Claudia Copeland.
Congressional Research Service
14
Air Quality Issues and Animal Agriculture: EPA’s Air Compliance Agreement
prevent the Agency from issuing CERCLA abatement orders to require response to
damaging releases.32
Author Contact Information
Claudia Copeland
Specialist in Resources and Environmental Policy
ccopeland@crs.loc.gov, 7-7227
32 Testimony of Mathy Stanislaus, Assistant Administrator, Office of Solid Waste and Emergency Response, U.S. EPA,
before the Subcommittee on Environment and the Economy, House Committee on Energy and Commerce, June 27,
2012.
Congressional Research Service
15