27 ELR 10097 | Environmental Law Reporter | copyright © 1997 | All rights reserved

Promise and Reality in the Enforcement of the Amended Clean Air Act Part I: EPA's "Any Credible Evidence" and "Compliance Assurance Monitoring" Rules

Editors' Summary: This Article is the first of a two-part series that examines the promise and reality of Clean Air Act enforcement by reviewing four central enforcement issues: (1) the development of the "any credible evidence" rule; (2) the evolution of the compliance assurance monitoring proposal; (3) the controversy over the requirement of federal enforceability of limitations on a source's potential to emit pollutants for purposes of determining its status as a "major source" under the Act; and (4) the tension between state-law voluntary disclosure, immunity, and environmental audit privileges and EPA's enforcement policy for state Title V operating permit programs.

This Article briefly reviews some of the new enforcement tools created by Congress through the 1990 Clean Air Act Amendments. It then examines one of the most important of these tools—the "any credible evidence" rule—which is central to the prospect of vigorous citizen enforcement of the Act. It next reviews the issues surrounding development of the "compliance assurance monitoring" rule—a rule that is expected to generate detailed compliance information that regulators and citizens can use for enforcement purposes. The Article concludes that while both rules are consistent with Congress' desire to strengthen clean air enforcement, these rules are likely to cause a fundamental shift in the realities of clean-air enforcement for industry, citizens, and regulators alike.

Mr. Van Cleve's firm specializes in environmental law and general litigation in Washington, D.C. He is also a former Deputy Assistant Attorney General, U.S. Department of Justice, Environment and Natural Resources Division. He received his B.A. from the University of Chicago in 1973 and his J.D. from Harvard Law School in 1977. Mr. Holman is an attorney at Jones, Day, Reavis & Pogue in Washington, D.C., specializing in environmental law. He is also a former Assistant Regional Counsel, U.S. Environmental Protection Agency, Region 4. He received his B.A. from the University ofWashington in 1983 and his J.D. from Northwestern School of Law of Lewis & Clark College in 1988. The authors wish to thank Robert Knop, an American University law student, for his invaluable research assistance. The opinions expressed in this Article are solely those of the authors.

[27 ELR 10098]

Introduction

One of Congress' key objectives in amending the Clean Air Act (CAA) in 1990 was to strengthen its enforcement. At this writing, after a process that took much longer than Congress had anticipated, the amended Act's major enforcement provisions are moving toward implementation by the U.S. Environmental Protection Agency (EPA) and the states. This is a good time to consider the size of the gap between the Act's original promise of strengthened enforcement and the reality of the new enforcement framework that has taken shape.

At the time of the 1990 CAA Amendments, EPA envisioned that two new features of the Act would dramatically increase clean air enforcement: first, the Title V program for federal operating permits, requiring "major sources" of air pollution to periodically certify their compliance with all applicable Act requirements;¹ and second, the enhanced-monitoring program, requiring most "major sources" to perform continuous or nearly-continuous monitoring of their air emissions. The compliance certifications and the results of enhanced monitoring would also be available for review by the public, strengthening citizen enforcement of the Act.²

In some respects, however, the amended Act's enforcement framework has not developed precisely as EPA expected. To begin with, the states took longer than anticipated to adopt the Title V program, delaying EPA's timetable for using Title V permits as the vehicle for increased CAA enforcement. In addition, the enhanced-monitoring program originally proposed by EPA was widely criticized as being unworkable. EPA eventually withdrew the proposal and replaced it with the compliance assurance monitoring (CAM) proposal, which represents a different, but perhaps farther-reaching, approach to monitoring. EPA has also delayed promulgation of its "any credible evidence" rule, which EPA viewed as a key tool for strengthened enforcement, for more than three years.

Congressional Efforts to Strengthen Enforcement Under the Amended CAA

The Adoption of the Individual Permit Enforcement Model

The CAA Amendments of 1990 effected sharp changes in the enforcement tools available to the federal and state governments seeking compliance with air quality laws.³ These changes were designed to streamline and strengthen clean air enforcement. Under the CAA as it stood before the 1990 Amendments, sources of air pollution had been regulated at the federal level under a series of plans referred to as state implementation plans (SIPs).⁴ Each SIP functioned as an umbrella, comprehensively describing and limiting air pollution within its geographic area. In order to show violations of the law, enforcement officials were required to demonstrate that the actions of an individual polluter were inconsistent with the requirements for attainment of the pollution goals or restrictions contained within the SIP.In practice, identifying the specific SIP requirements applicable to a particular source and establishing that the source had violated those requirements were often difficult tasks.⁵ In addition, because sources were often not required to submit compliance reports to EPA or the states, identifying instances of noncompliance was also difficult.⁶

In contrast, under the Federal Water Pollution Control Act (FWPCA), individual sources of pollution received national pollutant discharge elimination system (NPDES) permits that described and limited the pollution they are permitted to discharge to waters of the United States.⁷ Under the FWPCA, sources were required periodically to test their discharges and to provide the results to the regulatory authorities in discharge monitoring reports (DMRs) that were publicly available. DMRs could then become the basis of enforcement action either by the regulatory authorities or by citizens acting as private attorneys general. This scheme of periodic public reporting, combined with a public-private enforcement system, has led to a relatively high level of compliance with FWPCA requirements by individual sources.⁸

In large part, the 1990 Amendments to the CAA represented the adaptation of the FWPCA enforcement model to the CAA. The Title V federal operating permit program created by the 1990 amendments provides for a comprehensive source-permitting scheme. Under the Title V system, each covered source would receive an individual source permit, similar in concept to an FWPCA NPDES permit. The source permit would contain all air-quality requirements applicable to the permitted source. As a condition of receiving a Title V permit, each source would be required to certify periodically that it is in compliance with all applicable air-quality requirements, or to present an approvable plan for coming into compliance with those requirements.

Most importantly, however, the creation of a Title V permit system created a separate, entirely independent means of enforcement against violators of air-pollution requirements. A source that violated Title V permit requirements would be subject to the amended Act's heightened civil and criminal penalties, and those penalties could be imposed under simpler, swifter, and less forgiving procedures.⁹ For example, the 1990 Amendments increased [27 ELR 10099] criminal sanctions to make certain CAA violations felonies, and gave EPA the ability to impose penalties of up to $ 200,000 in administrative proceedings for noncompliance with administrative orders.¹⁰

Under the Title V permit scheme, EPA also received separate, unconditional, concurrent enforcement authority for both civil and criminal enforcement of all Title V permits.¹¹ In other words, EPA has the ability to "overfile" with respect to any Title V permit violation, even though a state intends to engage, or is engaged, in enforcement action against the same source under delegated authority.

Finally, Congress clearly expected that the public would play an important role in clean-air enforcement under the 1990 Amendments. Both during and since enactment of the amendments, there has been substantial controversy over the public's precise role in the air permitting and the permit-amendment process; however, it is clear that Congress intended the public to be heavily involved. It is also clear that Congress expected the public to play a substantial role in the air permit enforcement process, because it extended the reporting scheme of the FWPCA to the Title V permit process and broadened the types of remedies available in citizens' suits.¹²

In addition, the CAA provisions for citizens' suits were extended to Title V permits, thus allowing citizens to recover attorneys fees in appropriate cases for their efforts in enforcing such permits.¹³ This decision to extend citizens' suits reinforced the principle that the public's right to know is a "benchmark of United States environmental policy."¹⁴

The New Enhanced-Monitoring Requirement

The 1990 Amendments also added the requirement that major sources of air pollution conduct "enhanced monitoring" of their emissions and their pollution-control equipment.¹⁵ Although the phrase "enhanced monitoring" is not defined in the Act, EPA understood the provision to require facilities to develop monitoring protocols that would be incorporated into their Title V permits.

EPA proposed an enhanced-monitoring rule in October 1993 that applied to major sources subject to Title V.¹⁶ The proposal would have required facilities to submit quarterly enhanced-monitoring reports to their state air agencies and to identify the number and duration of any deviations from their enhanced-monitoring protocols. While the proposed enhanced-monitoring rule did not require continuous emissions monitoring for every covered facility, it did essentially require facilities to provide data indicating that they were in continuous compliance with applicable CAA requirements. Additionally, the enhanced-monitoring rule required facilities to retain records of enhanced-monitoring results, deviations from enhanced-monitoring protocols, and performance of pollution-control equipment. Finally, the proposed enhanced-monitoring rule allowed EPA, a state, or a citizen to use "any credible evidence" in an enforcement action, rather than being forced to use an EPA-prescribed testing protocol to demonstrate that a violation had occurred. EPA expected that CAA enforcement would dramatically increase once enforcement officials and the public were armed with a facility's enhanced-monitoring data and record of deviations, and had the benefit of using any credible evidence to prove a violation of the Act.

EPA's Proposed "Any Credible Evidence" Rule

The Proposed Rule

EPA first proposed its "any credible evidence" (ACE) rule as part of its enhanced-monitoring rule proposal in October 1993.¹⁷ The ACE rule would

amend 40 C.F.R. Parts 51, 52, 60, and 61 to eliminate language that has been read to provide for exclusive reliance on reference test methods as the means of demonstrating compliance with emission limits under the Clean Air Act ("CAA"), and to clarify that credible evidence can be used for compliance determinations.¹⁸

The proposed ACE rule would permit a source to certify compliance by using by a range of methods, including reference test methods and enhanced monitoring.¹⁹ The proposed ACE rule would require that SIPs promulgated under the CAA also allow the use of any credible evidence to establish a sources' violations.²⁰ The proposed ACE rule further provides that any "testing, monitoring, or information-gathering method[s]" that "produced information comparable" to any of the methods approved for compliance certification by the source would be regarded as "presumptively credible" evidence.²¹ Thus, it appears that the drafters of the ACE rule intended that no one type of credible evidence (including reference test methods specified in applicable emissions tests or standards) would be given preferred status under the proposed rule.

[27 ELR 10100]

To understand the issues surrounding the proposed ACE rule, it is necessary to understand the concept of a reference test method and its historic use in EPA and state enforcement under the CAA. Essentially, a reference test method is a specific test method used to determine compliance with a particular EPA emission standard or limitation. In many cases, EPA has specified the use of a particular reference test method in connection with the promulgation of a particular emission standard or limitation. According to EPA, many states and sources believed that the EPA-specified reference test method contained in SIPs was the exclusive method that could be used to determine compliance or noncompliance with the EPA standard.²²

In establishing a reference test method, EPA often specified: (1) training of test personnel; (2) how the test is to be conducted; (3) how test results are to be analyzed; (4) how often tests are to be conducted; and (5) how compliance with the standard is to be determined, using the test results. For example, in the case of certain opacity regulations, EPA has specified: how observers are to be trained to conduct opacity testing; how opacity tests are to be conducted; when opacity tests are to be conducted; and how test results are to be used to determine compliance with opacity regulations.²³

However, it is well known today that there are often alternative methods that could theoretically be used to determine compliance with particular EPA air-quality regulations. For example, in the case of opacity, there are now continuous emission monitors (CEMs) that use lasers to measure the opacity of plumes coming out of a plant stack.²⁴ The readings produced by these monitors are often more frequent, and in many cases more reliable, than the estimates prepared by even the best-trained human observers called for in the EPA opacity reference test.²⁵ In addition to CEMs, there are often physical process parameters related to plant emissions—such as the temperature, pressure, and speed of gas flows—that can be usedto reliably estimate plant emissions of certain pollutants.²⁶

Under historic EPA practice, EPA used and strongly recommended (if it did not actually require) that states use the results of a reference test to determine compliance with the emissions standard for which the reference test was established. The ACE rule would significantly broaden EPA practice, and citizen-suit enforcement, by permitting compliance with EPA emission standards to be determined through use of "comparable" CEMs data and other measurements such as reliable indirect physical measurement or parametric data.

Ironically, it is precisely the increased accuracy of many of the new alternative compliance measurement methods that forms the basis of a substantial part of industry objections to the ACE rule. At the same time, the increased accuracy and data availability provided by alternative compliance test methods, which will undoubtedly lead to increased compliance and enforcement capabilities for governments and citizen enforcers, are the main substantive arguments in favor of the ACE rule.

As previously noted, in the 1990 CAA Amendments, Congress clearly sought to strengthen the enforcement capability of EPA, the states, and citizens.²⁷ Various powerful considerations suggest that EPA should base its broadened regulatory system and a strengthened compliance and enforcement program on the best possible data concerning the actual operations of air-pollution sources. In the circumstances, industry is likely to be forced to shoulder the burden of demonstrating why EPA and citizens' groups should not be able to use the data that would be made available for enforcement under the ACE rule. They will do this by arguing that EPA does not have the legal authority to adopt the ACE rule.

The History of the Legal Debate Over Use of Any Credible Evidence in CAA Enforcement

] [ The Law Before the 1990 CAA Amendments. There has been an extended legal debate over the use of any credible evidence in CAA enforcement. The CAA provides in pertinent part, as it did before the 1990 Amendments:

Whenever, on the basis of any information available to the Administrator, the Administrator finds that any person has violated or is in violation of any requirement or prohibition of an applicable implementation plan,…the Administrator shall notify the person and the State…. At any time after the expiration of 30 days following…such notice…the Administrator may…bring a civil action [to enforce the Act].²⁸

EPA's historic position was that this language, and the general evidence rules of federal law, allowed a court to consider any credible evidence in determining whether CAA violations had occurred. However, in United States v. Kaiser Steel Corp.,²⁹ a case decided before the 1990 Amendments, [27 ELR 10101] a district court disagreed with EPA and held that the EPA reference test method was the exclusive method available to determine compliance with a particular air regulation.

] [ Legislative History of the 1990 Amendments. EPA disagreed with the Kaiser Steel decision. EPA and some distinguished outside commentators believe that the 1990 Amendments overruled the Kaiser Steel decision and clarified that EPA and citizens may employ any credible evidence in determining whether the Act's standards have been met.³⁰ Industry disagrees that the 1990 Amendments permitted this result.

The 1990 Amendments did not add language on the "any credible evidence" issue to CAA § 113(a)(1), entitled "Federal enforcement: In general," even though that was the part of the Act on which EPA had previously relied in asserting its authority to use any credible evidence. Instead, language was added to § 113(e)(1), entitled "Penalty Assessment Criteria." Section 113(e)(1) was amended in pertinent part to read:

In determining the amount of any penalty to be assessed under this section or section 7604(a) of this title, the Administrator or the court, as appropriate, shall take into consideration:…the duration of the violation as established by any credible evidence (including evidence other than the applicable test method).³¹

In industry's view, this language appears to be limited to allowing EPA, the states, or citizens to use any credible evidence to establish how long a violation persisted—a point that had been hotly contested in prior CAA enforcement actions—but not to use the same evidence to establish that the violation had occurred in the first place. The statutory language emphasized above was the same in both the final House and Senate bills.

There is no discussion of the "any credible evidence" language in the Conference Report on the 1990 Amendments, the authoritative statement of the Managers of both Houses on the legislation. There is no discussion of the language in the principal House Committee Report,³² and a scanty discussion in the Senate Committee Report.³³ However, the Senate Committee Report does contain fairly broad language supportive of EPA's position:

This title of the bill enhances the ability of the Environmental Protection Agency to identify facilities that are violating the requirements of the Act by…making clear that the Agency may rely upon any credible evidence of violations in pursuing alleged violations.

Penalty assessment criteria

….

Finally, the amendment clarifies that courts may consider any evidence of violation or compliance admissible under the Federal Rules of Evidence, and that they are not limited to consideration of evidence that is based solely on the applicable test method in the State implementation plan or regulation. For example, courts may consider evidence from continuous emission monitoring systems, expert testimony, and bypassing and control equipment malfunctions, even if these are not the applicable test methods. Thus, this amendment overrules the ruling in United States v. Kaiser Steel Corp., No. 82-2623-IH (C.D. Cal. January 17, 1984), to the extent that the court in that case excluded the consideration of such evidence.³⁴

Industry makes several points in advocating its position in light of the 1990 Amendments' legislative history. First, EPA, like everyone else, is bound by the long-standing judicial principle of statutory interpretation that if statutory language is clear, the courts will not look at legislative history. Industry would assert that the ultimately adopted language on this issue is clear. A reviewing court would therefore be free to—or might even be obligated to—disregard the more expansive description of EPA's authority found in a minor part of the Act's massive legislative history.³⁵

Second, if there were any room for argument about the meaning of this language, industry would argue that the amendment to the Act is not drafted in language that achieves the result EPA sought in a straightforward manner. If Congress wanted to make clear that EPA had the ability to use any credible evidence for any enforcement purposes under the Act, the place to do this was clearly in § 113(a), which contains the general description of EPA's enforcement authority, rather than in a modification to one limited aspect of a court's penalty assessment authority under the Act. While one could view this as an inadvertent draftsman's error, industry would argue that the actual language chosen does not appear to be an inadvertent failure to achieve the intended result, but instead appears to be a "second-best" substitute for broader language that EPA would have preferred but for which EPA could not obtain administration or congressional support. For this same reason, the value of the scanty legislative history is also suspect.

Thus, following traditional canons of statutory interpretation, industry would argue that under the 1990 Amendments, EPA's authority to use any credible evidence must be limited to circumstances where the issue before the court was the duration of the violation as it related to the appropriate penalty, while initial violations—the pure question of liability—would still need to be established using the existing reference test methods.³⁶

[27 ELR 10102]

If the distinction for which industry argues—between establishing an initial violation of the Act, on the one hand, and establishing the duration of the violation, on the other—was an untenable distinction, industry's narrow reading of the 1990 Amendments would be seriously undercut. Yet this distinction was historically a tenable one, because the issues in traditional clean air enforcement cases often broke down into two parts: (1) was there an initial violation? and (2) how long did the violation persist? Indeed, there is another part of the 1990 Amendments that addresses just this distinction in enforcement actions, by providing that once an initial violation and its likely recurrence has been shown, there is a presumption that the violation continued until the violator shows that the violation ceased.³⁷

In response, EPA could argue that industry's reading of the 1990 Amendments is specious because it is too narrow. EPA can point to the undeniable fact that a major purpose of the 1990 Amendments was to strengthen enforcement. EPA could argue that the use of credible evidence to determine the duration of violations necessarily includes the use of credible evidence to determine the fact of an initial violation. Finally, EPA can point to the broad language in the legislative history supporting its view that the 1990 Amendments overruled the Kaiser Steel decision.

In the alternative, EPA could assert a more general ground for upholding its authority to adopt the ACE rule. EPA could argue that the Kaiser Steel decision was incorrect, and that EPA had the power before the 1990 Amendments to use any credible evidence to enforce the Act, regardless of—or at least after changes in—evidentiary limitations contained in state SIPs. EPA could argue that nothing in the 1990 Amendments deprived it of that authority. EPA could argue that the purpose of the 1990 Amendments was to remove any doubt on the any credible evidence issue. Congress did this by explicitly enabling EPA to use any credible evidence to establish the duration of violations. Congress' decision to grant EPA such authority on the very contentious issue of the duration of violations, where most of the monetary penalties under the Act would be imposed in any enforcement proceeding, merely confirms EPA's position that it had the authority to use such evidence to establish initial violations. The basis for EPA's alternative position would be that Congress did not modify the language of § 113(a), on which EPA had previously relied for its general authority, and instead added only a clarifying amendment to make clear that any credible evidence could also be used to establish the duration of violations.

Despite the significance that EPA and industry obviously attributed to this legislative history during the consideration of the ACE rule, there have been only two reported court decisions dealing with this issue since then. Both of these involve citizen enforcement of the Act, and shed only limited light on EPA's authority to promulgate the ACE rule. They do, however, shed considerable light on the nature or direction of what may well be a substantial increase in enforcement activity if the ACE rule is promulgated.

[] Public Service Co. of Colorado. Ironically, Sierra Club v. Public Service Co. of Colorado,³⁸ the first major court decision since the 1990 Amendments to consider the admissibility of nonreference test evidence, was decided on grounds other than the language and history of the Act. Addressing the statutory language and its legislative history almost as an afterthought, the court upheld the Sierra Club's position that credible evidence other than the established EPA reference test method could be used to determine whether violations of the Act had occurred.³⁹

In Public Service Co., the Sierra Club alleged that defendants, owner/operators of the Hayden power plant in Colorado, had violated the CAA more than 19,000 times in five years by emitting pollutants in excess of the 20 percent opacity limitation set forth in the plant's permit. The Sierra Club's sole evidence of these opacity violations was data and reports from the plant's CEMs. Defendants argued that opacity violations could only be established through a reference test method—Method 9—provided for in the Colorado SIP and EPA opacity regulations. In support of their position, defendants cited the fact that the Colorado emissions regulations specifically provided that opacity was to be measured by EPA Method 9. In support of their position that this was the exclusive method of testing compliance, defendants cited an EPA guidance memorandum stating that if CEMs data was not specified as the compliance method, EPA could not refer cases to the U.S. Department of Justice for prosecution on the basis of CEMs evidence alone.⁴⁰

The court disagreed, and admitted the Sierra Club's CEMs data as evidence. The court reasoned that citizen-suit enforcement was an "intricate" part of the Act's enforcement scheme.⁴¹ Because this was so, the court reasoned, the Act gave citizens access to data needed to prove violations of the Act.⁴² The court noted that defendants were required to submit CEMs data under the SIP, and that citizens were given access to this data by the Act. The court held that CEMs data had a "high degree of probative reliability." The court noted that defendants' interpretation of the Act to exclude probative CEMs data would "gut the interstitial remedial functions of the Act's citizen suit provisions."

Extending its "citizens suit" rationale for its decision, the Public Service Co. court further noted that the use of EPA Method 9 required a trained observer, who probably would need access to the premises in order to conduct the [27 ELR 10103] observation properly.⁴³ The court noted that citizens would not be able to insist on access and that requests for access would put a company on notice that violations were suspected, and allow it to conceal evidence of violations.⁴⁴

In conclusion, the court held that nothing in the Act or regulations meant that citizens were bound by the test methods contained in the SIP in a citizen enforcement action. The court then noted, in what appears to be dictum, that its holding was "bolstered" by the 1990 Amendments to § 113(e)(1). Notably, the court gave very short shrift to the defendants' argument that permitting the use of alternative test methods to determine compliance amounted to "judicial amendment" of the applicable emissions standard, an argument repeatedly made since then by industry. The court stated that it was analyzing an evidentiary issue, not one that affected the substance of the basic emissions limitation or standard.

Strictly speaking, the court's holding in Public Service Co. relates only to citizen enforcement of the CAA. Principal parts of the court's reasoning—such as the inability of citizens' groups to conduct reference tests, or the fact that citizens groups were not bound by SIP provisions with respect to reference test methods—would not apply to EPA or the states. EPA and states can require sources to conduct reference tests, and arguably are bound by SIP provisions requiring use of such reference tests unless those provisions are changed.

However, it would be an anomalous to place citizens' groups in a "preferred position" where enforcement is concerned, giving them access to a broader range of evidence than that available to states or EPA, because the CAA contemplates that citizen enforcement—although integral to the Act—will nevertheless be interstitial. Ultimately, if the court's reasoning is sound, it would provide support for EPA's authority to promulgate the ACE rule, and would thus eventually apply to federal and state enforcement as well.

The facts of Public Service Co. make clear the appeal of the position taken by EPA and citizens groups that any credible evidence should be able to be used in enforcement. Defendant Public Service Co. of Colorado had engaged in massive, continuing violations of the CAA over a five-year period. The defendants could not and did not seriously contest the accuracy of their own data, adduced to prove those violations. Indeed, ironically enough, defendants in Public Service Co. had actually asserted in an earlier compliance proceeding that CEMs data was so superior to data obtained using the EPA reference test method, Method 9, that CEMs data should be regarded as "conclusive" evidence in that proceeding. Finally, the court is clearly correct that citizen enforcement—and environmental enforcement in general—will be strengthened if it can be based on data obtained without prior notice to an air-pollution source and without the need to obtain the consent or cooperation of plant officials.

In addition to arguing that the decision is limited to citizens' suits, industry might take some comfort from two limitations on the court's ruling in Public Service Co. First, the court's reasoning appears to suffer from the post hoc fallacy. The court concluded that if certain data was required to be submitted to a state or EPA, it must necessarily be available for use in citizens' enforcement proceedings. However, it is clear that some CEMs data submitted pursuant to SIPs is not used for standards enforcement purposes, but is instead used to monitor compliance with the source's general duty under its permit to maintain and operate its pollution-control equipment in good working order. Second, the court did not closely examine either the language or the legislative history of the 1990 Amendments. The court's conclusory discussion of this history does not consider the apparent conflict between the statutory language and the legislative history.

The Public Service Co. decision seems consistent with EPA's purpose in proposing the ACE rule. The court's decision is also consistent with the broad interpretation of credible evidence adopted in the only other case to address that issue.

[] Hawaiian Cement. In Unitek Environmental Services, Inc. v. Hawaiian Cement,⁴⁵ plaintiff Unitek Environmental Services (Unitek) brought a citizen's enforcement action against defendant Hawaiian Cement (Cement). Unitek alleged that Cement had repeatedly violated the particulate matter limits and related control requirements of the Hawaii SIP. Some of the SIP provisions contained objective requirements, and some did not.⁴⁶

The largest part of the Hawaiian Cement court's opinion dealt with the various evidence plaintiff offered to prove violations of the Act. First, the court permitted the use of the notice of violation (NOV) that EPA issued to Cement as evidence of a violation of the Act. The court noted that the NOV asserted that Cement had been in violation of the Act since at least two years before the action began. It also found that Cement was currently in violation of the SIP requirements.

From the court's discussion, it appears that the NOV was based on information Cement itself provided pursuant to an order under CAA § 114. The court stated that "EPA and Hawaii DOH [Department of Health] approved the method by which [Cement] would obtain the necessary information."⁴⁷ The court stated that pursuant to EPA's order, Cement conducted air monitoring for total suspended particulate and particulate matter with a diameter of less than or equal to ten micrometers (PM[10]). The court found that EPA relied on Cement's monitoring data in issuing the NOV. In short, it appears that EPA's basis for finding violations during the period July-October 1995 was reference test monitoring data.⁴⁸ This is particularly significant in the context of the "any credible evidence" issue, because it appears that the court founded its conclusion that Cement [27 ELR 10104] was in violation of the Act on reference test data, and relied on the other data submitted by Unitek only for its conclusions on the issue of the duration of the violation. However, EPA also found in the NOV that Cement had been in violation since 1993, and the court's opinion does not explain the basis for EPA's position on the duration of the violations.

Cement attacked the monitoring data it had collected at EPA's direction, claiming that the monitoring was "fundamentally flawed," apparently in part because the sites chosen for monitoring were not sites Cement had requested. The court rejected this argument, however, responding that EPA had approved a plan submitted by Cement.⁴⁹

Unitek submitted its own analysis of Cement's monitoring data, which showed continuous violations of the Act during the 1995 monitoring period. The court accepted Unitek's analysis as evidence. In addition, it found that monitoring data collected independently by Unitek could be admitted as evidence of earlier violations and that Unitek's analysis and independent monitoring data constituted "ample evidence" that Cement was causing violations since 1993, "if not longer."⁵⁰

Next, Unitek asserted that Cement's admission that it was not in compliance with the law in its 1994 application for a permit represented an admission that bound Cement in Unitek's later enforcement action. The court agreed that the permit application statements "denoted CAA violations."⁵¹ However, it cannot be determined from the court's opinion whether the court believed that Cement's source permit application, filed in 1994, established more than one day of violation of the Act.⁵²

Unitek also argued that Cement's own modelling data, prepared as part of its 1994 permit application, could be used as evidence of violations. Cement had asserted in its application that the standards "could not be met." In court, Cement attacked the modelling data as hypothetical. The court disagreed, and permitted use of the evidence.

After considering all of the evidence described above, the court found that "the evidence as a whole establishes that [Cement] has been continuously violating the 150 microgram/cubic meter standard since December 1993."⁵³ The court then held that it was permitted to consider the various forms of evidence submitted by Unitek by virtue of the credible evidence language added to the law by the 1990 Amendments.

Hawaiian Cement does not appear to resolve the fundamental question of statutory interpretation posed by the 1990 Amendments (i.e., whether evidence other than the reference test method can establish an initial violation of the Act). But it does clearly show the power and scope of an "any credible evidence" standard in litigating a CAA enforcement action. Unitek was able to use a broad range of evidence other than the reference test method data, much of it generated by defendant Cement, in order to support its position that the Cement had been violating the Act over a period of nearly three years. In particular, the court's willingness to admit modelling data—as opposed to actual monitoring data—may give plaintiffs in some cases fairly broad latitude in proving violations of the Act.⁵⁴

In a sense, Hawaiian Cement was an easy case. The facts clearly indicated that Cement was in violation of the Hawaii SIP. The only real issue was the duration of the defendant's violations, and the evidence on that point, consisting in significant part of defendant's own data, does not appear to have been open to serious question, and in any event much of it appears to have been cumulative. If the very lenient evidentiary standard employed by the court were generally adopted, it would likely support a very sharp increase in the amount of citizen-suit enforcement and in the levels of penalties generated in such actions.

An important question is how compatible the court's approach is with EPA's announced view of the scope of the ACE rule. As discussed above, EPA believes that to be "presumptively credible," evidence must produce results "comparable" to those that would be produced by a reference test method. The Hawaiian Cement court appears to have given the term "credible evidence" the broad meaning it would normally have under the Federal Rules of Evidence, rather than considering whether the evidence should produce "comparable" results before it would be regarded as "credible." This is not to say that EPA would not support the court's approach in Hawaiian Cement. However, the distinction EPA draws in the proposed ACE rule between "presumptively credible" evidence and other evidence would probably increase the burden of using other evidence by imposing an additional burden of proof before it could be used. By contrast, the Hawaiian Cement court does not rely on any similar distinction.

Additional Industry Objections to the ACE Rule

The legal debate between industry and the EPA over the ACE rule has, rather remarkably at this late date, exposed a fundamental difference of view between EPA and industry about the meaning of compliance under the CAA. Under Title V, as a condition of obtaining permits, industry officials are required to certify that their companies are either in compliance with all applicable CAA requirements, or they must submit approvable plans with a schedule for coming into compliance. If there is a fundamental difference about what compliance means, the value of these certifications, either for enforcement or for defensive purposes, is called into serious question. Accordingly, it makes sense [27 ELR 10105] to resolve this difference now, before EPA and the states begin bringing Title V enforcement actions against facilities.

In EPA's view, compliance means that a source is in continuous compliance with the standards set under the Act. Thus, if CEMs data show that continuous compliance is not occurring, a source is considered to be out of compliance with the Act, and the noncompliance must be remedied. EPA asserts that this either has always been the required standard of compliance under the Act, or that if there were any question about this, the 1990 Amendments removed any doubt on this point.⁵⁵

In industry's sharply differing view, at least in the context of EPA emissions standards calling for periodic reference testing, required compliance instead consists of two components: (1) meeting the standards established by EPA using the reference test method when required; (2) proper operation and maintenance of control equipment at other times, a responsibility enforceable under the "general duty" clause of SIPs. In industry's view, if a company can demonstrate it can meet the reference test requirements for a particular emission standard, it is irrelevant for enforcement purposes (as opposed, for example, for the purpose of improving compliance) whether CEMs data show violations of the established standard so long as control equipment is being properly operated and maintained.⁵⁶

EPA's legal argument for continuous compliance is based on the fact that when emissions standards or limitations are adopted, they are stated in the regulations in the form of continuous compliance requirements. Typically, emissions standards provide that after a certain date, particular types of sources must be in compliance with certain specific standards or limitations. There are exceptions to this compliance requirement only for specified technical or operating conditions, such as bypass or upset conditions, and EPA's position is that these are exceptions that prove the rule of continuous compliance. In the past, in at least some cases, courts appear to have held that continuous compliance is a requirement when the permit did not provide a clear exception to such compliance.⁵⁷

Industry's position is that EPA's emissions standards are based in integral part on the use of a reference test to determine compliance with the standard. Industry asserts that EPA is required, both for legal and technical reasons, to establish such reference tests as part of establishing a standard. Therefore, industry argues, EPA cannot change or abandon the reference test without amending the standard.

In industry's view, if EPA or a state were to require a higher frequency of compliance testing, this would substantively modify the established emission standard or limitation. Industry asserts that such a modification cannot be achieved by EPA without amending the individual emission standards or limitations. Perhaps not entirely coincidentally, the individual amendment process would not only be incredibly time and resource intensive, but would probably lead to enormous litigation, and would also have the effect of grandfathering large groups of sources from requirements for continuous compliance or "credible evidence" enforcement until they required new permits.

Industry's legal argument relies on early court decisions requiring EPA to (1) establish upset and bypass conditions as part of its regulatory framework⁵⁸ and (2) establish the statistical variability of its reference tests in setting its emission standards.⁵⁹ From these decisions, industry infers that a reference test is a necessary part of an EPA emission standard. However, industry's reliance on these decisions to support its ultimate position on the exclusivity of reference tests is bootstrapping, for two reasons.

First, if it is necessary as a practical matter to permit exceptions for upset and bypass conditions, which are identifiable physical operating limitations of a facility, EPA regulations should recognize such exceptions as a matter of providing fair notice with respect to enforcement—but this principle has no necessary connection with what evidence constitutes evidence of a violation (or evidence that an upset or bypass condition exists). Indeed, the fact that EPA regulations have for many years provided for enforcement exceptions for bypass and upset conditions, except in rare circumstances, appears to buttress EPA's position that from the beginning of the Act EPA has required continuous compliance except where its regulations explicitly provide otherwise.⁶⁰

Second, industry correctly observes that the courts have held that when EPA is setting a new source performance standard, the CAA requires it to demonstrate that the standard is achievable. Industry is also correct that part of the evidence required to support such a showing is that there is a reliable testing methodology that demonstrates that proposed controls can achieve the standard. It does not follow, however, that because a particular reliable testing methodology is employed or developed to show the achievability of the standard, it is therefore necessarily the only means of determining compliance with the standard. Any equally reliable testing methodology could support a demonstration of achievability. The proposed ACE rule contains a similar, albeit limited, safeguard because it creates the presumption that alternative testing methodologies that produce data "comparable" to reference or other approved test methods can be considered credible evidence, thus rendering other evidence more difficult to use.

Industry representatives have added a twist to their argument, however. They assert that the use of the reference [27 ELR 10106] test developed by EPA in setting a new emissions standard cannot be divorced from the substantive emissions standard because the reference test data used by EPA in standard-setting inevitably show substantial variability in control effectiveness, and therefore influence EPA's decision as to the level at which to set the standard, including fundamental compliance issues such as the "averaging period" for determining compliance with various standards.⁶¹

It may be that the choice of the specific reference test used does influence the EPA standard-setting process, at least to the extent of affecting the level at which the standard is set, or in suggesting the need for an "averaging period" for compliance determinations. However, industry goes a step further, arguing that the use of a particular reference test to set a standard assumes or tolerates a certain level of noncompliance.⁶² Industry argues that as a consequence, abandoning reference tests as exclusive compliance methods is effectively making the standards more stringent.

The industry argument takes the following form. EPA performance standards are set on the basis of limited data, and these limited data exhibit substantial variability. Therefore, even though EPA sets the emission limit above the level of the highest typical emission data it obtains in testing, it may well be that further testing would disclose emission levels even higher than the levels set by EPA. EPA is aware of this possibility, so it adjusts the frequency of required testing in such a way that exceedances of its standards will occur rarely.⁶³

There are at least two potential problems with this argument. First, the administrative records of most EPA performance standards contain little or no direct support for industry's position that EPA actually believed that sources would exceed the performance standards it set a measurable part of the time. This is clear from the fact that industry witnesses who were personally involved in the standard-setting process themselves have relied on their recollections of informal, undocumented conversations with industry representatives wholly outside the record as a basis for establishing EPA's intent.⁶⁴ However, even if an informal EPA-industry "compact" or "understanding" existed, it could not legally form the basis of a limitation on the enforcement of the existing new source performance standards, because suchan agreement would unlawfully have excluded citizens from the standard-setting and enforcement process.⁶⁵

Moreover, other cases have held that regulated sources cannot rely on any such informal assurances regarding relaxation or alteration of permit requirements from regulators in the face of permits containing explicit regulatory requirements. For example, in one citizens' action alleging numerous water quality permit limit violations, the court held that the defendant could not rely on assurances it might have received from regulators that its permit limits were "goals" rather than binding limits, since it had not challenged the limits in its permit.⁶⁶

In addition, the statistical methodology on which industry witnesses have relied may well be open to serious question because of the small size of the samples involved and because of assumptions about the distribution of the data on which their arguments appear to depend. However, whether or not classes of well-controlled sources can meet existing EPA performance standards is an issue that could be resolved on the basis of further statistical review of existing data and, perhaps, some further empirical testing. Indeed, EPA has asked industry to provide data concerning situations in which they are aware that significant numbers of sources cannot meet existing performance standards, and (for perhaps understandable reasons concerning their perceived potential enforcement vulnerability) industry has not provided such data for the record to date.

Notably, the industry "stringency" argument does not suggest that there is anything wrong with the use of data that is comparable to reference test methods.⁶⁷ Rather, industry argues that particular existing EPA performance standards were incorrectly set because they could not actually be met on a continuous basis. If EPA limits the use of comparable data to those situations that are consistent with the limitations of the existing standard, this will minimize conflicts. For example, the creation of an averaging period in an EPA standard assumes that a source may be out of compliance at various points within the averaging period, so long as at the end of that period, it is in compliance on an average basis. Any enforcement use of credible evidence must likewise account for this averaging period, even though data may be collected more frequently than the averaging period would suggest. Also, some EPA standards contain statements concerning operating limits within which the standard will apply, such as facility loading or capacity conditions when testing occurs. Any use of credible evidence in enforcement must take account of these operating limitations or cannot really be regarded as comparable.

Although EPA has, for example, permitted semiannual use of a reference test for compliance with a standard that includes an averaging period or certain load limitations, and semiannual testing undoubtedly would not detect various periods of noncompliance with such a standard, it does not necessarily follow that EPA intended to permit noncompliance with the standard.⁶⁸ This has been expressed in competing analogies during the debate. EPA has argued that it is not changing "speed limits" with the ACE rule, but rather allowing enforcers to use "radar guns" to detect [27 ELR 10107] speeding. EPA acknowledges that the ACE rule will increase instances of detected noncompliance, but argues that this does not increase the stringency of the actual standards EPA is enforcing, and that it does not intend to increase their stringency.⁶⁹

Industry has argued that the ACE rule is analogous to a change from one method of determining average fuel economy to another. Under existing law, industry argues, fuel economy (air emissions standard compliance) is measured by filling up a tank, driving the car until the tank needs a refill, and averaging the mileage over the gasoline consumption. Under the ACE rule, industry argues, EPA is insisting on measuring fuel economy instantaneously during the entire trip, which would produce radically different results for any given period of driving, given different automobile operating conditions.

However, in context, the industry analogy demonstrates the inherent difficulties of industry's position on the use of credible evidence.⁷⁰ If an emissions source is analogized to a car, then operating conditions outside the normal operating range, such as bypass or upset conditions, would be excluded for the purposes of determining "fuel economy" (emissions), no matter what testing methodology is employed. In addition, if an emissions standard or limitation itself contained specifications concerning operating conditions used to test compliance, such as the percentage of capacity at which a source was operating when tested, then data collected at times when the system was operating outside the limiting conditions would also be excluded in determining compliance under any testing method. If the emissions standard contained an "averaging" requirement, such averaging would be included in the calculations, no matter what test data were being used to determine compliance.

When all is said and done, the "driving conditions" for the "car" (source) in determining compliance under the ACE rule would reflect the same "driving conditions" the "car" faces currently, and the results of the data collection under the test reference method or under some alternative comparable form of data collected pursuant to the ACE rule should be the same. This is expected, just as instantaneous measurements of fuel economy, when properly weighted and averaged over the same period of miles as were driven by the car, should yield the same average fuel economy as would be yielded by the average created using the total gas consumed and total miles driven.

Industry has argued that the adoption of the ACE rule would violate the procedural requirements of the Administrative Procedure Act (APA) and/or CAA § 307. However, because EPA made the proposed rule available for notice and comment in substantially the form in which EPA proposes to adopt it, and has held a public meeting on the rule, industry faces some obstacles in arguing that EPA has not met the APA's or the CAA's notice-and-comment requirements.

Instead, industry representatives have asserted that the ACE rule modifies each emissions standard now in effect that contains a reference test. They argue that EPA cannot legally modify individual emissions standards and limitations except on an individual basis, making each proposed new standard available for separate notice and comment, and public hearing. Assuming, for the purposes of this section, that industry is correct that the ACE rule substantively modifies existing emissions standards or limitations, rather than simply modifies a rule of evidence,⁷¹ EPA's current proposal to issue a direct final rule to modify such standards across-the-board without further rulemaking action is open to question.

The APA and its CAA counterpart do not require EPA to choose any particular method of amending existing new source performance standards. There is nothing in the law, for example, to prevent EPA from amending all new source performance standards at one time.⁷² EPA's decision to proceed on what is in effect a generic basis to amend existing emissions standards and limitations, rather than on a rule-by-rule basis, poses no potential harm to the integrity of the decisionmaking process.

EPA is required, however, to provide a reasoned justification for its proposed course of action, and to provide for public comment the factual information and legal justification supporting its proposed rule. Industry will argue that this has not yet occurred, though EPAwould probably disagree. Thus, if EPA intends to proceed with a generic amendment to existing performance standards in the face of industry concerns about the ability of well-controlled sources to comply with those standards, EPA would be well advised to seek specific formal comment from industry identifying those standards with which industry believes it cannot continuously comply, and then determine whether these arguments are factually correct. If EPA determined that certain existing performance standards were in fact not consistently achievable, it would not be required to amend these standards. Nevertheless, EPA might decide to modify the averaging periods used to assess compliance with those standards or make other similar adjustments. If the "amended" standards are still not achievable by existing sources subject to those standards, EPA will have to decide whether to attempt to require sources to upgrade controls, or to wait to apply these requirements to sources receiving new permits in the normal course.

It would utterly frustrate EPA's ability to modernize and strengthen its enforcement program if a court held that EPA lacked the authority to adopt the ACE rule until EPA proposed each emissions standard or limitation now in existence for amendment and separate notice, comment, and eventual judicial review. This would be akin to telling the Food and Drug Administration that it could not adopt new drug testing methodologies with respect to the safety [27 ELR 10108] of existing drugs on the market without proposing individual amendments of each safety determination it had made with respect to all existing drugs. The APA's and the CAA's procedural rules were never intended to hobble EPA in making well-reasoned decisions, even if such decisions are of broad general applicability.

EPA's Move From Enhanced Monitoring to Compliance Assurance Monitoring

Background

Before the 1990 CAA Amendments, emissions monitoring at many facilities occurred infrequently, or was a relatively crude indicator of overall annual emissions. Even when they were used, stack tests and other common monitoring methods often provided only a snapshot of a facility's emissions. Because a facility could prepare for a stack test and be at its operational best, these infrequent tests often provided a poor indicator of the facility's long-term compliance. Accordingly, a heavy enforcement burden was placed on state and EPA inspectors, who had to observe an emission violation firsthand, apply the appropriate reference test method, and develop sufficient data to be able to prove the violation in court. Moreover, while most states required some form of report when a facility's emissions exceeded its permit limits, there was no uniform requirement that facilities report and certify to their compliance status.

Congress responded to these perceived deficiencies in 1990 by adding the following new language to CAA § 114(a), which governs the monitoring and recordkeeping required of facilities:

The Administrator shall in the case of any person which is the owner or operator of a major stationary source, and may, in the case of any other person, require enhanced monitoring and submission of compliance certifications. Compliance certifications shall include (A) identification of the applicable requirement that is the basis of the certification, (B) the method used for determining the compliance status of the source, (C) the compliance status,(D) whether compliance is continuous or intermittent, (E) such other facts as the Administrator may require.⁷³

Additionally, in Title V of the 1990 Amendments, Congress specified that Title V operating permits issued by states or EPA must include:

[A] requirement that the permittee submit to the permitting authority, no less often than every 6 months, the results of any required monitoring….

The Administrator may by rule prescribe procedures and methods for determining compliance and for monitoring and analysis of pollutants regulated under [the Act], but continuous emissions monitoring need not be required if alternative methods are available that provide sufficiently reliable and timely information for determining compliance.⁷⁴

Congress did not define the term "enhanced monitoring," and the legislative history of the 1990 Amendments gives limited guidance as to what kind of monitoring will satisfy the enhanced-monitoring requirement. In describing the Act's new monitoring and certification requirements, the December 1989 Senate Committee Report stated:

Under the current implementation of the Act, EPA typically determines the compliance status of sources by conducting on-site inspections, or by issuing source-specific investigatory letters requiring the collection and submission of emission data. The purpose of this amendment is to clarify and confirm that the Administrator has the authority under section 114(a) to require enhanced monitoring and submission of compliance certifications from major stationary sources. In appropriate cases, enhanced monitoring may include the use of continuous emission monitoring systems. In addition, this amendment clarifies that EPA may require groups or classes of sources to collect emissions data and to certify their compliance to EPA on an intermittent, periodic, or ongoing basis.

…Similar to the reporting requirements of the Clean Water Act, 33 U.S.C. 1342, compliance certifications and emission data submitted pursuant to this authority will facilitate enforcement, due in part to the fact that such data and certifications can be used as evidence.⁷⁵

The combination of the amended Act's language, the placement of enhanced-monitoring requirements in the Title V permitting program, and the available legislative history indicates that Congress intended enhanced monitoring to: (a) apply to major sources, such as Title V facilities,⁷⁶ (b) require facilities to submit monitoring results to permitting authorities at least once every six months, (c) require facilities to operate continuous emissions monitoring systems "in appropriate cases," and (d) at a minimum, indicate whether a facility has been in continuous or intermittent compliance with Act requirements.

EPA's 1993 Proposed Enhanced-Monitoring Rule

Given the task by Congress of writing regulations to implement the enhanced-monitoring requirements of CAA §§ 114 and 504, EPA proposed a comprehensive enhanced-monitoring rule on October 22, 1993.⁷⁷ The enhanced-monitoring rule was designed to be implemented through the Title V and prevention of significant deterioration/new source review (PSD/NSR) preconstruction permitting programs; any emission unit at a Title V facility with potential emissions of at least 30 percent of the major source threshold was subject to enhanced monitoring. Facilities subject to the CAA § 112 hazardous air pollutant (HAP) program were exempted from the enhanced-monitoring rule for their HAPs, since the maximum achievable control technology (MACT) standards designed to control HAPs contain their own monitoring/recordkeeping requirements.

[27 ELR 10109]

The proposed enhanced-monitoring rule required the owner or operator of a facility to propose an enhanced-monitoring protocol (EMP) in the facility's Title V or PSD/NSR permit application, subject to review and approval by the permitting authority. The facility had the burden of demonstrating that the proposed EMP would meet all of the enhanced-monitoring criteria contained in the rule. Facilities could choose between using "established" monitoring methods previously approved by EPA and proposing alternative monitoring methods that the permitting authority would evaluate for their accuracy and reliability. The EMP was required to consider unit-specific factors and objectively be the "best" monitoring system for the unit. An acceptable EMP had to include an explanation of the monitoring method, an installation and testing plan, performance specifications, and a method for quality-control checks. The EMP was to ensure that monitoring was sufficiently representative, accurate, precise, reliable, frequent, and timely, so as to determine whether a unit had been in continuous compliance with its emission requirements.⁷⁸ The proposed enhanced-monitoring rule did not require all facilities to employ CEMs, but the rule clearly favored monitoring methods that sampled or calculated emissions with great frequency. The rule generally required facilities to provide monitoring data for every averaging period, and contained a strong presumption toward relatively short averaging periods (e.g., 24 hours or less).⁷⁹ Facilities were also allowed to use parameter monitoring (e.g., material use records, fuel sampling and analysis, etc.) to demonstrate compliance, but only after the parameters had been proven to be accurate indicators of emissions. Additionally, all units operating under the enhanced-monitoring rule would be required to conduct their monitoring within certain performance specifications forrelative accuracy, calibration error, measurement span, and response time.

With respect to recordkeeping and reporting, the proposed enhanced-monitoring rule required facilities to keep all documentation for five years, including monitoring equipment installation and maintenance information, enhanced-monitoring data, records of deviations (exceptions from compliance), and unit-specific operating records.⁸⁰ The proposed rule required companies to submit quarterly enhanced-monitoring reports, certified by a responsible official, to the permitting authority. The quarterly reports had to summarize the monitoring results for the preceding quarter, identify the number and duration of all deviations from emission limits, provide a record of conformity with the EMP parameters, and state whether compliance had been continuous or intermittent during the quarter.⁸¹

The proposed rule provided that enhanced-monitoring requirements must be included in Title V permits or preconstruction permits as permit conditions, and that failure to comply with an enhanced-monitoring condition was a violation.⁸² Similarly, EPA or a state could determine that an identified deviation was a violation, and initiate an enforcement action. EPA noted in the proposed rule's preamble that:

It is the intent of these proposed regulations that where EPA or a permitting authority determines that reported deviations constitute noncompliance, the owner or operator of an emissions unit may be subject to enforcement under sections 113 or 304 of the Act and any similar State enforcement authority.⁸³

The proposed rule therefore gave EPA and the states fairly broad discretion in deciding how to respond to reported deviations. This broad enforcement discretion, coupled with the uncertainty surrounding the relationship between "intermittent" compliance and noncompliance, created significant potential for enforcement ambiguity. The proposed rule's language on "any credible evidence" only served to deepen this ambiguity; monitoring methods considered presumptively "credible" could establish evidence of deviations that might or might not lead to enforcement by a state or EPA or be used by the public in a citizen's suit.

Industry's Response to the Enhanced-Monitoring Proposal

EPA's proposed enhanced-monitoring rule was immediately unpopular with industry. The agency received approximately 2,000 comment letters on the proposal, the majority of which were from industry commenters. Wholly apart from concerns about the "any credible evidence" provision, industry commenters felt that the proposal was far too detailed, complex, prescriptive, and expensive to comply with. The rule's preference for CEMs and other very frequent testing requirements was regarded as prohibitively expensive for the minimal environmental benefits they would yield in terms of improved emission information. Similarly, the rule's complex requirements for developing EMPs, conducting monitoring, and maintaining records were criticized for their inflexibility and their burdensomeness.

In the wake of the adverse industry reaction to the proposal, EPA held a series of stakeholder meetings in the fall of 1994. The agency continued to receive severe criticism of the proposed rule, to the point that EPA was forced in December 1994 to officially reopen the public-comment period on several issues related to the rule.⁸⁴ Among other things, these issues included the question whether EPA should consider phasing in the implementation of the enhanced-monitoring rule over a five-year period, and how EPA or a state could take costs into account in selecting and approving EMPs.

In April of 1995, however, in response to continuing criticism, EPA decided to abandon its enhanced-monitoring rule, with the exception of the ACE proposal. EPA issued a notice in May 1995 that the agency intended to hold a public meeting to discuss the possible redesign of the rule.⁸⁵ EPA held several meetings on the rule's redesign through the summer of 1995, and a draft preamble of the new CAM rule was circulated for comment in [27 ELR 10110] September 1995. Following nearly a year of additional comment and discussion, the proposed CAM rule was published on August 13, 1996.⁸⁶

The Proposed CAM Rule

In contrast to the proposed enhanced-monitoring rule, which was designed to use monitoring as a method for directly determining continuous compliance with applicable requirements, the CAM rule takes the indirect approach of considering the operation of emission-control measures at a facility. These control measures include air-pollution control devices, process modifications, and operating limitations. The CAM rule essentially requires a facility that employs control measures: to evaluate and document that the control measures are continuously operating within specific performance ranges that have been designed to assure compliance with applicable requirements; to indicate any excursions from these ranges; and to adequately respond to the data so that excursions are corrected.⁸⁷

The rule creates a distinction between units at a Title V facility that do and do not have an active control device (equipment that removes or destroys emissions). Units that use an active control device are subject to Part B of the CAM rule if their potential emissions without controls equal or exceed the major-source threshold.⁸⁸ For these units, the proposed CAM rule requires the owner/operator to develop a CAM plan that describes the monitoring to be performed, selects indicators of control performance, sets indicator ranges, and establishes a corrective action process for excursions from the indicator range.⁸⁹ The Part B CAM plan must subsequently be approved by the permitting authority and incorporated into the facility's Title V permit.

For units at a Title V facility that do not have an active control device, Part C of the CAM rule requires "monitoring sufficient to provide a reasonable assurance of compliance over the anticipated range of operating conditions."⁹⁰ For units that have existing monitoring requirements, the owner/operator may propose in a permit application that the existing monitoring is adequate to reasonably assure compliance. For Part C units that do not have an existing monitoring requirement, the owner/operator may make a showing that no monitoring is necessary, or that record-keeping is sufficient to assure compliance. A state or EPA would have authority under the rule to establish operating and performance requirements and to require Part C CAM units to perform additional monitoring if appropriate.

The proposed CAM rule also requires a unit that experiences prolonged excursions from its established indicator ranges to undergo corrective action through a quality improvement plan (QIP).⁹¹ The QIP contains two elements: an evaluation procedure to determine the cause of the excursions or exceedances,⁹² and a list of the steps the facility will take, on a schedule, to improve the quality of control performance.⁹³ These steps include improved preventive maintenance practices, process operation changes, improvements to control methods, and more frequent or improved monitoring.⁹⁴

The threshold for triggering a QIP should theoretically be set "at a level where the cumulative duration of excursions (or exceedances) is unacceptable and improvements are necessary to assure ongoing compliance."⁹⁵ The exact percentage of excursions or exceedances during monitoring periods over which a QIP will be triggered is to be established by the permitting authority, but may not be greater than 5 percent.⁹⁶ In EPA's view, "the QIP is necessary to assure that the owner or operator pays attention to the data and, if necessary, improves performance to the point where ongoing compliance with applicable requirements is reasonably assured."⁹⁷ The QIP must be implemented as soon as practicable, and must be completed within 180 days from the date that notice of the QIP is given.⁹⁸ To address concerns that facilities might have repeated excursions from their established indicator ranges and operate under continuous QIPs, the proposed CAM rule provides that the implementation of a second QIP for a specific unit during a permit term (typically, five years for a Title V permit) would constitute a permit violation, and be subject to an enforcement action.⁹⁹

With respect to compliance certifications, the proposed CAM rule would require Title V facilities to identify the permit conditions being certified, as well as the methods used to determine whether or not the facility was in compliance with each condition.¹⁰⁰ At a minimum, these methods must include all testing and monitoring that was conducted during the time period. Other material compliance information known to the owner/operator must also be identified in the compliance certification. The facility must report any excursions or exceedances as deviations. Importantly, however, while all deviations are exceptions from compliance, not all deviations are violations that trigger enforcement action.

EPA is under a court order to finalize the CAM rule by July 7, 1997. Once the rule is final and becomes effective, the CAM requirements must be incorporated within 180 days into issued Title V permits or added to Title V applications.¹⁰¹

[27 ELR 10111]

Analysis of the CAM Rule

Taken as a whole, the CAM rule is a relaxation of EPA's approach in the enhanced-monitoring rule. By focusing primarily on monitoring the operating performance of control measures, rather than requiring extensive direct monitoring of emissions, the proposed CAM rule will be less burdensome to comply with than its predecessor. Industry commenters have been critical of the CAM rule proposal, however, for a number of reasons.

To begin with, the CAM rule, as proposed, may actually apply to more emission units than the original enhanced-monitoring rule would have. While the enhanced-monitoring rule would have applied to units with potential emissions equal to 30 percent of the major-source threshold (e.g., 30 percent of the 100-ton Title V major-source threshold for sulfur dioxide, or 30 tons), the proposed CAM rule would apply to units at Title V facilities whose uncontrolled potential emissions meet or exceed the major-source threshold. Because there are many units with control equipment that functions in the 90-95 percent capture-efficiency range, ignoring controls when determining potential emissions will include many otherwise well-controlled units under the CAM requirements. EPA has not provided a persuasive rationale for its proposed method of calculating the potential emissions of units at Title V facilities for CAM purposes. Given the likelihood that EPA will relax its method of calculating a facility's potential to emit in every other context, it is anomalous for the agency to ignore the effect of control equipment in making CAM evaluations.¹⁰²

Industry commenters have also objected to the prescriptiveness of the CAM rule. While the proposal allows more flexibility to facilities in developing CAM plans and in establishing adequate monitoring approaches, as compared to the enhanced-monitoringrule, the CAM rule limits the discretion of state and local permitting authorities to modify unworkable CAM plans, to allow a rate of excursions/exceedances greater than 5 percent, or to allow multiple QIPs during the term of a permit. It seems reasonable to give state and local permitting authorities this discretion.

Several industry commenters also believe that EPA should allow CAM rule requirements to be phased in over a longer period of time than the 180-day period discussed in the proposal. Because of the difficulty involved in incorporating CAM requirements into the first wave of Title V permits, together with the possibility of redundancies between the CAM rule's monitoring requirements and the monitoring requirements of the § 112 MACT standards to be promulgated in 2000, it makes sense to delay CAM applicability for five years, until Title V permits come up for renewal and CAM monitoring can be coordinated with monitoring required under the MACT rules.

Industry and state commenters have pointed out that if CAM requirements are imposed on facilities during the first cycle of Title V permits, thereby compelling states to commit heavy resources to CAM implementation and enforcement, EPA runs a substantial risk of overtaxing and irreversibly damaging the fledgling Title V programs in the states. Because the CAM program will ultimately succeed only if it is accepted by industry, these commenters argue that forcing the immediate integration of CAM into Title V will make implementation of both programs more difficult, and may seriously damage the credibility of EPA and state permitting authorities.¹⁰³

An additional basis for criticism by industry is the perception that the CAM rule imposes significant costs and burdens on companies that have recently expended much on the Title V compliance effort, with very little additional environmental benefits to be gained from CAM. To the extent that the CAM rule would identify emissions sources that are out of compliance or operating poorly, industry would argue that most of those sources have already been identified through the Title V emissions inventory and application certification process. While CAM might be a tool for ensuring long-term compliance with the Act, the short-term environmental benefits of the rule may be far out-weighed by its implementation costs.

Finally, industry continues to be concerned about the enforcement implications of the CAM rule, and its relationship to the ACE rule.¹⁰⁴ In many situations, the CAM rule will give regulators and the public much more detailed information about a facility's operations and deviations than was available in the past, increasing the possibility of enforcement actions and citizen suits. Also, the subtle distinction between CAM deviations and true noncompliance may well be lost on the public, placing companies in an unfairly negative light.

EPA seems to recognize that the requirement to report deviations as exceptions from compliance (but not necessarily as noncompliance) creates some confusion.

Since the CAM data will not necessarily always provide unequivocal proof of compliance or noncompliance (as a compliance test method would), there will be deviations identified through CAM which raise questions about compliance status but may not confirm conclusively that a source is in noncompliance. The Agency emphasizes that a certification which includes exceptions for deviations is not an admission of noncompliance. The existence of deviations only indicates the need to review the compliance information provided in order to determine what, if any, compliance or enforcement actions may be warranted.¹⁰⁵

In spite of EPA's clarifications and assurances that deviations will not always result in enforcement actions, companies will continue to be uncomfortable with the CAM rule until bright lines between "acceptable" deviations and "violation" deviations are established and are widely understood by the public. Even if industry cannot have the desired degree of leniency or flexibility in enforcement, it [27 ELR 10112] must have some assurance of evenhandedness and predictability in enforcement.

While the CAM rule is generally a less burdensome approach to the CAA's enhanced-monitoring and compliance-certification requirements, there are still several aspects of the CAM rule that will be difficult for industry to integrate into the Title V permitting scheme immediately. The CAM approach will have a better chance of being fully accepted by industry and fulfilling its role within the Act if EPA responds to industry's reasonable CAM implementation concerns by: redefining CAM potential-to-emit to consider the effect of control equipment; allowing state and local permitting authorities to have more discretion in administering CAM plans and QIPs; phasing in CAM requirements over a five-year period to coordinate with Title V permit renewals; and better clarifying how CAM and the ACE rule are to mesh together for enforcement purposes.

Conclusion

On the merits, EPA's decision to adopt the ACE rule seems consistent with Congress' desire to strengthen clean-air enforcement. There is support in the Public Service Co. and Hawaiian Cement decisions, and in the history and purpose of the 1990 Amendments, for EPA's authority to adopt the ACE rule. However, an open question remains about the actual effect that the ACE rule will have on the enforcement of existing new source performance standards. Similarly, while EPA has the authority to promulgate the CAM rule, the rule will impose new operating restrictions and costs on industry.

The combination of the ACE rule, with its imposition of a continuous compliance enforcement regime, and the CAM rule, with its requirements for frequent, detailed compliance reporting, will cause profound changes in clean-air enforcement. At the same time, as will be discussed in Part II of this Article, there is growing resistance on the parts of the courts and the states to centralized federal air-quality enforcement based on principles of traditional enforcement policy. This resistance is vividly expressed in recurring conflicts over the "federal enforceability" and "audit privilege and immunity" issues discussed in Part II of this Article.

If EPA does not introduce and use these powerful enforcement tools judiciously, there is great potential for concerted industry resistance. If industry perceives that the rules are applied overbroadly or inconsistently by EPA, the states, and citizens, industry opposition may become entrenched, thereby delaying achievement of the Act's compliance objectives.

At the same time, it behooves industry to appreciate that in adopting the FWPCA model for clean-air enforcement, and by broadening the role of citizens in enforcement, Congress made a clear and decisive choice in favor of replacing paper compliance with measurable actual compliance. In the long run, industry stands to gain little, and lose much, by fighting costly battles whose sole purpose is to delay or procedurally hobble such reforms.

1. 42 U.S.C. §§ 7661-7661f, ELR STAT. CAA §§ 501-507. See 40 C.F.R § 70.6(c) (1996).

2. 42 U.S.C. §§ 7414, 7661c, ELR STAT. CAA §§ 114, 504.

3. Pub. L. No. 101-549, 101 Stat. 2399 (1990).

4. The CAA as it stood before the 1990 Amendments is codified at 42 U.S.C.A. §§ 7401-7642 (1988). The Act as amended is codified at 42 U.S.C. §§ 7401-7671q, ELR STAT. CAA §§ 101-618.

5. SENATE COMM. ON ENVTL. AND PUB. WORKS, CAA AMENDMENTS OF 1989, S. REP. NO. 228, 101st Cong., 1st Sess., at 347 (1989), reprinted in 1990 U.S.C.C.A.N. 3385.

6. Id.

7. 33 U.S.C. §§ 1251-1387, 1342, ELR STAT. FWPCA §§ 101-607.

8. See generally David T. Buente, Citizen Suits and the Clean Air Act Amendments of 1990: Closing the Enforcement Loop, 21 ENVTL. L. 2233 (1991).

9. Former EPA Administrator William Reilly stated, "The permit program is not only the most important procedural reform in the new Clean Air Act, but in many ways the key to effective enforcement and implementation of the law." EPA Press Release, Apr. 24, 1991. More recently, EPA's Office of Enforcement Compliance and Assurance (OECA) has stated that Title V permit terms and incomplete/inaccurate Title V permit applications are primary enforcement targets for the agency. EPA OECA, Draft 1998/99 National Enforcement and Compliance Assurance Priorities (Oct. 17, 1996) (available from the ELR Document Service, ELR Order No. AD-3111).

10. 42 U.S.C. § 7413(c)(1), ELR STAT. CAA § 113(c)(1); id. § 7413(d)(1), ELR STAT. CAA § 113(d)(1).

11. Id. § 7413(a)(3), ELR STAT. CAA § 113(a)(3).

12. Id. § 7661c, ELR STAT. CAA § 504; Buente, supra note 8.

13. 42 U.S.C. § 7604, ELR STAT. CAA § 304.

14. Letter from Robert W. Shavelson, Director, Atlantic States Legal Foundation, Inc. to Carol Browner, EPA Administrator (Feb. 6, 1995) (on file with authors).

15. 42 U.S.C. § 7414(a), ELR STAT. CAA § 114.

16. 58 Fed. Reg. 54648 (Oct. 22, 1993).

17. Id. at 54648. The description of the ACE rule in this Article is based on the form of the rule as it stood on August 30, 1996. EPA has stated that it intends to issue the ACE rule as a final rule, without reproposing it, and it is thus reasonable to anticipate that the rule will not be substantially modified before its issuance. Of course, EPA is not required to proceed in this manner, and could decide to amend the original proposal and provide additional public comment on it before it becomes final. At presstime, the rule was undergoing review in the Office of Management and Budget and was expected to be issued in early 1997.

18. EPA, The Use of Information Other Than Reference Test Results for Determining Compliance With the Clean Air Act, undated position paper (prepared for Apr. 2, 1996 public hearing) (available from the ELR Document Service, ELR Order No. AD-3099) [hereinafter EPA ACE Position Paper].

19. 58 Fed. Reg. at 54648 (Oct. 22, 1993). Because EPA has decided to replace its proposed enhanced-monitoring rule with a separate CAM rule, it is anticipated that the final ACE rule will be modified to delete references to enhanced monitoring.

20. EPA issued a SIP call at the same time it proposed the ACE rule, requiring the states to amend their SIPs to conform them to the ACE rule. Id. at 54648, 54677. Four states have reportedly since done so. Other states have informally indicated to EPA they believe their SIPs already allow use of any credible evidence. Still others have indicated that they preferred to wait to amend their SIPs when the ACE rule becomes final.

21. Id. at 54648, 54676.

22. Id. at 54675. Technically, of course, EPA regulations provided that states were free to petition the Administrator for the use of alternative test methods, but this rarely, if ever, occurred. See 40 C.F.R. § 52.12 (1996). As a practical matter, before 1994, EPA exerted significant pressure on the states to incorporate the reference test method into their SIPs as the exclusive means of determining noncompliance.

23. 40 C.F.R. pt. 60, App. A, Method 9 (1996).

24. Some air-pollution sources are required either by state or by federal air regulations to employ such CEMs, which in some cases, at least, are to be calibrated daily in order to maintain their accuracy. Such sources include the Hayden power plant, whose emissions were the subject of Sierra Club v. Public Serv. Co. of Colo., 894 F. Supp. 1455, 25 ELR 21461 (D. Colo. 1995), discussed below.

25. See id. at 1459-60, 25 ELR at 21464 (noting that defendant opposing use of CEMs data in instant proceeding had argued in an earlier proceeding that CEMs data were far more reliable than EPA reference test data).

26. EPA ACE Position Paper, supra note 18. The paper states: "Today, as a practical matter many sources determine compliance with permitted limits either through the use of continuous emission monitors or by monitoring key parameters of their production processes and pollution control devices." Id. at 2. Industry representatives have argued that such parametric monitoring is neither necessarily inexpensive nor necessarily reliable. See EPA, Credible Evidence Rule Stakeholders' Meeting, Apr. 2, 1996 at 29-30 [hereinafter Apr. 2, 1996, Stakeholders' Meeting].

27. Strengthened enforcement appears to have been the primary purpose behind the creation of the Title V permit program, as well as for the substantial enhancement of the enforcement provisions of the Act. See generally Michael S. Alushin, Enformcement of the Clean Air Amendments of 1990, 21 ENVTL. L. 2217 (1991). See also Reilly comments, supra note 9.

28. 42 U.S.C. § 7413(a)(1), ELR STAT. CAA § 113(a)(1) (emphasis added).

29. United States v. Kaiser Steel Corp., No. 82-2623-IH (C.D. Cal. Feb. 8, 1984). The published materials from this case, available on LEXIS at (GENFED Library, DIST File), include only the court's order. The court's opinion denying EPA's position is unpublished, but the case has been widely cited for the proposition that EPA was limited to the use of reference test methods in enforcement.

30. 54 Fed. Reg. 54648, 54649 (Oct. 22, 1993); EPA ACE Position Paper, supra note 18, at 3; Buente, supra note 8 (§ 113(e) amendment allows citizens to present evidence through general "expert opinion" that sources have violated emissions limitations, even in absence of any actual test data).

31. 42 U.S.C. § 7413(e)(1), ELR STAT. CAA § 113(e)(1) (emphasis added).

32. HOUSE COMM. ON ENERGY AND COMMERCE, REPORT ON THE CLEAN AIR AMENDMENTS OF 1990, H.R. REP. NO. 490, 101st Cong., 2d Sess. (1990).

33. SENATE COMM. ON ENVTL. AND PUB. WORKS, CAA AMENDMENTS OF 1989, S. REP. NO. 228, 101st Cong., 1st Sess. (1989), reprinted in 1990 U.S.C.C.A.N. 3385.

34. Id. at 358, 366.

35. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 8 ELR 20288 (1978), reh'g denied sub nom. Natural Resources Defense Council v. U.S. Nuclear Regulatory Comm'n, 685 F.2d 459, 12 ELR 20465 (D.C. Cir. 1982), rev'd subnom. Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, 462 U.S. 87, 13 ELR 20544 (1983); Estate of Cowart v. Nicklos Drilling Co., 112 S. Ct. 2589, 2594 (1992); West Va. Univ. Hosps. v. Casey, 111 S. Ct. 1138, 1147 (1991) (courts will not look behind the plain meaning of a statute at legislative history or administrative interpretation).

36. Industry further argues that under the CAA and the Administrative Procedure Act, 5 U.S.C. §§ 500-596, ELR STAT. ADMIN. PROC. §§ 500-596, this situation cannot be changed by EPA unless and until EPA amends the individual new source performance standards or other standards that employ the reference test at issue. This concedes that EPA has the power to adopt an ACE rule for specific emission standards.

37. The Act provides that where EPA or a state has given notice of an air-pollution violation, if plaintiffs make a "prima facie showing that the conduct or events giving rise to the violation are likely to have continued or recurred past the date of notice," then the court is to presume that the days of violation include all days from the date of notice until the violator shows it has continuously complied. The violator may rebut this presumption. 42 U.S.C. § 7413(e), ELR STAT. CAA § 113(e). This language was included to overrule United States v. SCM Corp., 667 F. Supp. 1110, 18 ELR 20073 (D. Md. 1987), in which the court refused to shift to the source the burden of proving compliance after EPA established that the source was in violation of the CAA. SENATE COMM. ON ENVTL. AND PUB. WORKS, CAA AMENDMENTS OF 1989, S. REP. NO. 228, 101st Cong., 1st Sess. (1989), reprinted in 1990 U.S.C.C.A.N. 3385. In SCM Corp., the government's evidence of violations was limited to reference test data collected on infrequent occasions over a period of time. The court refused to infer that violations had occurred during the intervening time periods.

38. Sierra Club v. Public Serv. Co. of Colorado, 894 F. Supp. 1455, 25 ELR 21461 (D. Colo. 1995).

39. Id. at 1462, 25 ELR at 21465.

40. Id. at 1459, 25 ELR at 21463.

41. Id.

42. 42 U.S.C. § 7414(a), ELR STAT. CAA § 114(a).

43. Public Serv. Co. at 1457, 25 ELR at 21464.

44. Id.

45. 27 ELR 20483 (D. Haw. Aug. 27, 1996). The action also included state-law claims for nuisance and trespass.

46. Unitek alleged the following violations of the Hawaii SIP: (1) the Hawaii SIP contains a 150 microgram per cubic meter limit on ground-level dust emissions above upwind concentrations for any 12-hour period, unless a source obtains a determination that it is employing "best practical operation and treatment," and Cement had not obtained such a determination; (2) the Hawaii SIP requires sources to take "reasonable precautions" to prevent particular matter from becoming airborne; and (3) the SIP prohibits discharge of visible emissions beyond its property line without implementing "best practical operation or treatment" as determined by the Department of Health. Id.

47. 27 ELR at 20484.

48. However, the opinion is not clear on this point.

49. 27 ELR at 20484.

50. Id. at 20484.

51. Id. at 20485.

52. This is a very significant issue from the point of view of sources that are required to certify compliance under Title V of the CAA. Clearly, there is a realistic possibility that an admission of noncompliance in a Title V application will bind a company in a subsequent proceeding. If the court concludes that in addition to establishing a violation, such an admission, coupled with a compliance plan, also establishes continuing violations, then the source will have the burden of demonstrating that noncompliance was not continuous, so that the effect of admitting noncompliance on a company's penalty liability in a subsequent citizens' action, or enforcement action, could be very substantial.

53. 27 ELR at 20485.

54. On a related matter of particular interest to citizen enforcement, the court rejected the argument that the SIP fugitive dust standard was unenforceable because its interpretation rested in part on a determination whether the defendant had taken "reasonable precautions" to control fugitive dust, or whether defendant had adopted "best practical operation or treatment." Apparently reasoning that these standards were tied to an objective standard, the court held that the fugitive dust provision as a whole was enforceable. Id. at 20486.

55. EPA ACE Position Paper, supra note 18.

56. See Apr. 2, 1996, Stakeholders' Meeting, supra note 26 (Testimony of Clean Air Coalition).

57. Ironically, this appears to be the fundamental lesson of one of the formative cases relied on by industry counsel to attack the ACE rule. Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 3 ELR 20642 (D.C. Cir. 1973), cert. denied sub nom. Portland Cement Corp. v. EPA Adm'r, 417 U.S. 921 (1974), appeal after remand sub nom. Portland Cement Ass'n v. Train, 513 F.2d 506, 5 ELR 20341 (D.C. Cir.), cert. denied, 423 U.S. 1025 (1975), reh'g denied, 423 U.S. 1092 (1976). In Portland Cement, the court required EPA to include upset and bypass conditions in proposed permits as opposed to relying on "enforcement discretion" in situations where a source could not continually meet a permit requirement. Id. at 399, 3 ELR at 20654-55. The court's logic was that if EPA could not demonstrate that sources could continually comply when establishing new source performance standards, it had to allow sources permit exceptions covering such anticipated noncompliance. Id. Of course, it follows that except where the permit specifically allows noncompliance, continuous compliance would be required.

58. See id. at 396, 3 ELR at 20652-53; see also Apr. 2, 1996, Stakeholders' Meeting, supra note 26, at 60.

59. International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 649 (D.C. Cir. 1973). See Apr. 2, 1996, Stakeholders' Meeting, supra note 26, at 60.

60. See supra note 57.

61. See, e.g., Statement of Robert Ajax, P.E. (on behalf of Coalition for Clean Air Implementation), Mar. 9, 1995 (Enhanced Monitoring Rule, Docket No. A-91-52) (on file with authors).

62. This argument is made explicitly in the presentation by the Coalition for Clean Air Implementation at the Apr. 2, 1996, Stakeholders' Meeting, supra note 26 (Tr. at 70).

63. See statistical data in Ajax presentation, supra note 61, at 4-8.

64. See Ajax presentation, supra note 61.

65. See Proffitt v. Rohm & Haas, 850 F.2d 1007, 18 ELR 21165 (3d Cir. 1988); Student Pub. Interest Research Group v. Fritzsche, Dodge & Olcott, Inc., 579 F. Supp. 1528, 14 ELR 20540 (D.N.J. 1984), aff'd, 759 F.2d 1131, 15 ELR 20450 (3d Cir. 1985); Proffitt v. Lower Bucks County Joint Mun. Auth., No. 86-7220, 1987 WL 28350 (E.D. Pa. Dec. 16, 1987), rev'd on other grounds, 877 F.2d 57 (3d Cir. 1989). See also Chemical Mfrs. Ass'n v. EPA, 870 F.2d 177, 196 (5th Cir. 1989); National Audubon Soc'y v. Department of Water, 869 F.2d 1196, 19 ELR 20198 (9th Cir. 1988); Natural Resources Defense Council v. EPA, 859 F.2d 156, 19 ELR 20116 (D.C. Cir. 1988); Glazer v. American Ecology Envtl. Servs. Corp., 894 F. Supp. 1029, 1034 (E.D. Tex. 1995) (discussing the difficult balancing act Congress faced in providing for citizen suits while at the same time preserving limited judicial resources).

66. Connecticut Fund for the Env't, Inc. v. Upjohn Co., 660 F. Supp. 1397, 17 ELR 21137 (D. Conn. 1987).

67. Of course, industry also argues that there are proof problems in determining whether other data produces results "comparable" to reference test methods, but these arguments seem overstated in most circumstances.

68. But see supra text accompanying notes 63-64 for the contrary view.

69. EPA ACE Position Paper, supra note 18 at 9.

70. Not the least of the difficulties with the industry "fuel economy" compliance analogy is that in the case of car fuel economy, gasoline effectively has no place to go except out the tailpipe, so emissions can be measured. In contrast, if air emissions are well above emission standards limits while no testing occurs, there is no way to determine this when testing does occur.

71. At least one court has held such a rule only modifies the rules of evidence. Sierra Club v. Public Serv. Co. of Colorado, 894 F. Supp. at 1461, 25 ELR at 21464 (D. Colo. 1995) (discussing relationship between the Clean Air Act Amendments and Federal Rules of Evidence).

72. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 8 ELR 20288 (1978), reh'g denied sub nom. Natural Resources Defense Council v. U.S. Nuclear Regulatory Comm'n, 685 F.2d 459, 12 ELR 20465 (D.C. Cir. 1982), rev'd sub nom. Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, 462 U.S. 87, 13 ELR 20544 (1983).

73. 42 U.S.C. § 7414(a)(3), ELR STAT. CAA § 114(a)(3) (emphasis added).

74. Id. § 7661c(a), (b), ELR STAT. CAA § 504(a) (emphasis added).

75. SENATE COMM. ON ENVTL. AND PUB. WORKS, CAA AMENDMENTS OF 1989, S. REP. NO. 228, 101st Cong., 1st Sess. 266, 368 (1989), reprinted in 1990 U.S.C.C.A.N. 3385, 3751 (emphasis added).

76. Title V permits are required to include all existing monitoring and testing requirements set forth in applicable Act requirements. 40 C.F.R. § 70.6(a)(3)(i) (1996). Additionally, if a facility's particular applicable requirements do not include some sort of "periodic" testing or monitoring, the Title V permit must contain a requirement for "periodic monitoring" to fill the monitoring gap. Id. § 70.6(a)(3)(i)(B). EPA never actually defined "periodic monitoring," or how it differed in practice from "enhanced monitoring." Ultimately, the "periodic monitoring" concept seems to have been merged into the compliance assurance monitoring approach.

77. 58 Fed. Reg. 54648 (Oct. 22, 1993).

78. See id. at 54686.

79. See id. at 54691. EPA observed that "although the term 'continuous' means 'at all times,' the Agency has determined that less frequent measurements can be used to determine continuous compliance."

80. See id. at 54689-90.

81. See id. at 54688-89.

82. See id. at 54682-84, 54691.

83. Id. at 54674.

84. 59 Fed. Reg. 66844 (Dec. 28, 1994).

85. 60 Fed. Reg. 27943 (May 26, 1995).

86. 61 Fed. Reg. 41991 (Aug. 13, 1996).

87. EPA, Compliance Assurance Monitoring (CAM) Rule Discussion and Rulemaking (Aug. 2, 1996) [hereinafter Draft CAM Rule] (available from the ELR Document Service, ELR Order No. AD-3106); see also Notice of Availability, 61 Fed. Reg. 41991 (Aug. 13, 1996).

88. Draft CAM Rule, supra note 86, at 25-32.

89. Id. at 59-63.

90. Id. at 8.

91. Id. at 69-72.

92. An "exceedance" occurs where emissions measured directly through a CEMs are above an emissions limit. An "excursion" is a departure from the established CAM indicator range. A "deviation" is a failure to meet an established permit term or condition. For compliance certification purposes, an exceedance or excursion is a deviation, but it may or may not be an actual violation of the Act. Id. at 25.

93. Id. at 70.

94. Id.

95. Id.

96. Id. at 69.

97. Id. at 9-10.

98. Id. at 71-72.

99. Id. at 70-71.

100. Id. at 74.

101. Id. at 40-42.

102. EPA attempted in the late 1970s to promulgate a potential-to-emit rule that similarly did not consider the effect of emission controls; that definition was invalidated by the U.S. Court of Appeals in 1979. Alabama Power Co. v. Costle, 636 F.2d 323, 10 ELR 20001 (D.C. Cir. 1979).

103. To relieve the potential burdens of CAM implementation on the Title V program, the state of Texas, for example, has advocated an approach to CAM implementation that would avoid permit-by-permit reviews of CAM plans for facilities that have monitoring requirements in their underlying applicable requirements, and focus instead on those existing requirements. Texas Natural Resource Conservation Commission, Texas Programmatic Approach (Sept. 30, 1996) (available from the ELR Document Service, ELR Order No. AD-3110).

104. See supra text accompanying notes 55-72.

105. Draft CAM rule, supra note 87, at 75-76.

27 ELR 10097 | Environmental Law Reporter | copyright © 1997 | All rights reserved