13 ELR 10036 | Environmental Law Reporter | copyright © 1983 | All rights reserved

Court Upholds States' Relaxation of SO₂ Controls: Interstate Impacts, Sulfate Pollution Allowable

Editors' Summary: The long-range transport of sulfur dioxide emissions, their transformation into sulfates in the atmosphere, and their eventual return to earth through the phenomena popularly lumped together under the name "acid rain" is a serious environmental problem that many argue has been given inadequate attention in the Clean Air Act. In three decisions issued on December 1, 1982, the Second Circuit Court of Appeals analyzed carefully the Clean Air Act's response to the interstate transport of SO₂ and its transformation into sulfates. The court upheld EPA approval of revisions to the New York and Connecticut state implementation plans, including one utilizing the "bubble policy, which allow increases in SO₂ emissions from existing facilities. Phillip D. Reed reviews the decisions and argues that while they illustrate the limitations of the Act's response to interstate SO₂ pollution and the atmospheric formation of sulfates, they also establish a basis on which those provisions could be given greater effect in later cases.

[13 ELR 10036]

On December 1, 1982 the Second Circuit Court of Appeals decided three Clean Air Act¹ cases² upholding Environmental Protection Agency (EPA) approval of New York and Connecticut state implementation plan (SIP) revisions relaxing limits on sulfur dioxide (SO₂) emissions. The decisions address a wide range of issues concerning SIP revisions and shed some new light on the limits of EPA's "bubble policy," but their main thrust involves the Act's response to interstate air pollution and the transformation of SO₂ into sulfate, which is widely believed to be a critical link in the formation of acid rain. The three opinions, referred to by the court as Connecticut Fund I, II, and III³ will be closely read by lawyers fighting other pending legal battles over the long range implications of SO₂ relaxations.⁴ The lengthy Second Circuit opinions offer something to both sides. The court struggled to reconcile the fact that control of interstate pollution is vital to achievement of the Clean Air Act's objectives with the fact that the Act's interstate pollution provisions are both limited and ambiguous. Thus, it held that EPA properly rejected petitioners' allegations that the SIP revisions allowed illegal interstate pollution, but it also warned EPA that it would apply the provisions more vigorously under other facts.

Background

The three cases involve different reactions to the same problem. In the early 1970's New York and Connecticut responded to the Clean Air Act mandate to reduce SO₂ emissions in order to attain the national ambient air quality standards (NAAQS) for that gaseous pollutant by limiting the sulfur content of the fuel burned at large oil and coal fired industrial facilities such as power plants. Connecticut promulgated a 0.5 percent sulfur rule while New York limited similar facilities to 1.0 percent sulfur fuel. As a result of these measures, both states attained the NAAQS for SO₂. However, the steep climb of the price of oil that began with the Arab oil embargo of 1973 made this pollution control strategy a costly one indeed.⁵

As fuel prices and utility rates climbed it also became clear that both states' sulfur-in-fuel rules were more stringent than necessary to attain and maintain the NAAQS. Both sought to relax the rules. New York granted utilities variances from the rule for individual power plants. Connecticut [13 ELR 10037] proposed an across-the-board relaxation in its sulfur-in-fuel rule to 1.0 percent and established an Energy Trade Program⁶ designed to allow facilities that saved fuel through energy conservation to burn fuel higher in sulfur.

Each of the disputed actions required state promulgation and EPA approval as a SIP revision. Each was challenged as a threat to attainment or maintenance of the NAAQS for SO₂ or total suspended particulate (TSP) in the state originating the change and/or in its downwind neighbors. The challenges touched on the nonattainment provisions of Part D,⁷ the prevention of significant deterioration (PSD) provisions of Part C,⁸ and the validity of the bubble policy⁹ as applied in the Connecticut Energy Trade Program, but the key issues were the extent to which interstate pollution and the transformation of SO₂ into TSP must be considered in evaluating a SIP revision for the former.

The Clean Air Act reflects the view that air pollution is primarily a problem in the vicinity of sources¹⁰ and that each pollutant is a separate problem. However, in the three Connecticut Fund cases the primary concern was not SO₂ pollution near the regulated facilities, but the effect of relaxations of SO₂ rules on particulate pollution miles away. Such particulates were an unwelcome addition in Connecticut and were of special concern to the extent they were sulfates. Sulfates constitute a direct health threat and can form acids in the atmosphere or after any deposition on the ground. Thus, concern over acid rain was a factor in the litigation, although the court, like the Clean Air Act itself, did not mention the subject.

While silent on the subject of acid rain, the Act does address interstate pollution. In 1977 Congress revised § 110(a)(2)(E)¹¹ and added a new § 126¹² to ensure that states would address interstate pollution in the SIP process. The former sets the standard governing when a SIP must be tightened to cut off interstate pollution, while the latter requires notification of the interstate impacts of major new sources and establishes a procedure for forcing consideration of interstate pollution.

The Decisions

The New York Sulfur-in-Fuel Variances

In Connecticut v. Environmental Protection Agency¹³ (Connecticut Fund I) the state and the Connecticut Fund for the Environment sought review of an EPA rule approving a New York SIP variance allowing two Long Island Lighting Company (LILCO) power plants in Suffolk County, Long Island to continue burning 2.8 percent sulfur fuel for an additional three years, renewing a variance that had lapsed earlier. Petitioners argued that the variance from the otherwise applicable 1.0 percent sulfur rule would violate § 110(a)(2)(E)¹⁴ of the Act by preventing attainment of the NAAQS for TSP, preventing maintenance of the standard for SO₂, and interfering with prevention of significant deterioration (PSD) measures across Long Island Sound in Connecticut. Connecticut Fund I was the first decision to squarely address all three of § 110(a)(2)(E)'s protective provisions. The court did not resolve all uncertainty about their scope but it did take a significant first try at defining them in meaningful and practical terms.

The court first addressed the question of whether EPA properly interpreted § 110(a)(2)(E)(i)(I), which prohibits EPA from approving a SIP revision unless it ensures that no source will emit an air pollutant in amounts that will "prevent attainment or maintenance … of any such" NAAQS in another state. The court divided this question into two parts, one concerning maintenance of the standard for SO₂, which Connecticut has attained, and the other concerning attainment of the standard for TSP, which it has not.

* Prevent Maintenance. The key to the first part of the question was the meaning of "prevent maintenance." The court accepted EPA's interpretation. The Agency had concluded that to prevent maintenance of its neighbor's NAAQS, a state had to export enough pollution to raise the level of pollution in the other's air over the standard, taking into account the controls the receiving state had already required. The receiving state is under no obligation to impose stricter controls on its own sources to make room for the transboundary pollution. The court found that this view was consistent with the legislative history, which indicated that § 110(a)(2)(E) was designed to prevent one state from imposing extra pollution control requirements on another.

In upholding EPA's interpretation, the court rejected two alternatives. One, which it deemed a literal reading of the provision, would have allowed pollution exports as long as the receiving state could counter them by increasing controls in its SIP. This interpretation was inconsistent with the legislative history. The other, put forward by petitioners, was that a state could prevent maintenance of standards in another by exporting "substantial" pollution. The Second Circuit found this interpretation inconsistent with the statutory language and ruled that exports of "substantial" pollution were governed by § 110(a)(2)(E)(i)(II), which bans interstate pollution [13 ELR 10038] "interfering with" PSD measures required under the Act.

Having settled on the meaning of "prevent maintenance," the court held that EPA applied the term properly. It considered and rejected four technical challenges raised by petitioners. Initially it noted that the type of technical issues in dispute are particularly within the realm of EPA expertise. It then upheld EPA's choice of a model for estimating the SO₂ transport from Long Island to Connecticut. The Agency had considered problems with this model raised by petitioners and had responded to them.

The court also upheld EPA's assumptions concerning the heights of the smokestacks from which LILCO's emissions emanate. EPA had followed § 123¹⁵ of the Act and relied on heights dictated by "good engineering practice," rather than the taller actual heights. However, it also had supplemented this analysis with additional calculations using the stacks' true height, which showed that the variances would not cause NAAQS violations in Connecticut. The court faulted EPA's reliance on § 123,¹⁶ but given the Agency's consideration of the effect of the actual heights, declined to find that the Agency had abused its discretion.

It also rejected petitioners' argument that EPA abused its discretion by failing to consider a proposed relaxation in Connecticut's SIP for SO₂ before approving the variance in New York's. It declined to rule on whether EPA was required to make such an analysis because the Agency had in fact considered the combined effects of the two changes before approving the Connecticut SIP revision.¹⁷

* Prevent Attainment. Turning next to the question of whether the LILCO SO₂ variances would "prevent attainment" of the NAAQS for particulates in Connecticut, the court articulated two ways in which this might happen.Use of high sulfur fuel in New York could increase direct emissions of particulate matter which would then be transported to Connecticut. In addition, the SO₂ emitted to the atmosphere could be transformed into sulfate particles.

Initially, the court had to deal with an obstacle of its own making to consideration of the TSP effects of the New York SIP revision. In Connecticut Fund II¹⁸ it held that the Clean Air Act calls for separate, pollutant-specific SIPs and that in reviewing changes in Connecticut's plan for SO₂, EPA need not consider the effects on particulate pollution in that state. EPA argued that the same rule applied in the interstate context. The court noted its obligation to defer to a reasonable EPA interpretation of the Act, but balked at doing so on this issue. It pointed out that the wording of § 110(a)(2)(E) differs from that of § 110(a)(1)(A) and (a)(2)(A), (B), and (H)¹⁹ which it had construed in Connecticut Fund II. In addition, it noted that EPA's interpretation would limit Connecticut's flexibility in selecting control strategies. Having mounted a strong attack on EPA's statutory analysis, the court stopped short of ruling on the issue, however, finding that in fact EPA had adequately considered the effects of the New York SIP revision on TSP levels in Connecticut.

As to direct particulate emissions, the court ruled that EPA reasonably concluded that there would be at most insignificant impacts on Connecticut. It rejected EPA's argument that the statute requires "significant" impacts before the § 110(a)(2)(E)(i)(I) protection comes into play. It ruled only that impacts "so insignificant as to be fairly described as minimal" do not violate the statute, citing Alabama Power Co. v. Costle.²⁰

As to the Agency's failure to estimate the secondary particulate impacts resulting from sulfate formation, the court found it consistent with the statute.It accepted EPA's assertion that it lacked models capable of accurately estimating sulfate formation and found that the evidence in the record which might be used to approximate such effects suggested that they were minimal. However, the court also went out of its way to tell EPA that it should not dally in developing the necessary sulfate model.²¹

* Interfere with PSD Measures. The court next addressed the third component of § 110(a)(2)(E), subsection (i)(II), which prohibits interference with measures required to be included in the SIP by the PSD program. Again it rejected EPA's statutory analysis, but accepted its review of the New York variances. EPA argued that there could be no PSD measures to interfere with, because such measures were not required until a PSD baseline was established. There are no PSD baselines for Connecticut, because there have been no applications for major new source permits. The court accepted this argument with regard to intrastate SO₂ pollution resulting from Connecticut's Energy Trade Program in Connecticut Fund III,²² but refused to extend it to the interstate pollution presented in this case. Such a holding, it reasoned, would erase § 110(a)(2)(E)(i)(II) by allowing New York to consume Connecticut's entire clean air margin²³ and to ignore any PSD measures the state had adopted voluntarily. However, since Connecticut had promulgated no such measures and there was no indication that the pollution in question would use most or all of Connecticut's clean air [13 ELR 10039] margin, the court held that EPA's approval was not an abuse of discretion.²⁴

The Connecticut Sulfur-in-Fuel Rule

In Connecticut Fund for the Environment, Inc. v. Environmental Protection Agency²⁵ the environmental group remained in the role of challenger, but Connecticut's position was reversed, because its own SIP relaxation was at stake. The Fund challenged EPA approval of a doubling from 0.5 percent to 1.0 percent in the allowable sulfur-in-fuel in Connecticut, arguing that EPA had acted unreasonably in failing to consider the impact of the SIP revision on attainment of the TSP standard in Connecticut. That the change would result in increased direct emissions of particulates and secondary formation of sulfate particulates was not disputed. The issue was whether the interaction between the two pollutants requires that the SO₂ SIP be treated as a TSP control strategy as well. The Second Circuit ruled that it does not.

The court's conclusion stems from the fact that particulate emissions are separately regulated under the Connecticut TSP implementation plan. Any direct TSP emissions resulting from the sulfur-in-fuel relaxation would have to comply with emission limits in the SIP for particulates, limits calculated to attain the NAAQS.²⁶ Although this mechanism would not address the problem of sulfate formation, it was nevertheless found adequate because EPA cannot predict the magnitude of this effect with current models.

Not only was the pollutant-specific approach functional, the court concluded, but it appeared to be the approach intended by Congress. The Connecticut SIP follows most states in prescribing different control strategies for different pollutants. Moreover, the statute refers to SIPs, PSD, and nonattainment programs in terms of individual pollutants.²⁷ The court found further support in two other court of appeals decisions²⁸ in which EPA approval of SIP provisions for one pollutant was upheld even though the Agency ignored the effect of the provision on other pollutants. It ruled that these considerations outweighed petitioner's meritorious argument that the pollutant-by-pollutant approach runs counter to the statement of purposes in § 101(b)(1)²⁹ — to protect and enhance the quality of the nation's air.

Having established the validity of pollutant-specific SIP revisions, the court easily rejected the argument that the nonattainment provisions of Part D prohibit relaxation of the SO₂ limit. Declining to accept EPA's argument that Part D does not ban relaxing part of a nonattainment area SIP so long as the whole demonstrates attainment, the court ruled instead that Part D is inapplicable to the SO₂ revision because Connecticut is in attainment for that pollutant. Only the particulate SIP is subject to Part D's requirements, and it is separate from the SO₂ plan even though Connecticut had described the 0.5 percent sulfur rule as a TSP control strategy in one part of that plan.

When the subject changed to interstate pollution, however, the court found the "apples and oranges" distinctions between SO₂ and TSP less persuasive. It noted that the language of § 110(a)(2)(E) is different from that of § 110(a)(1)(A) and (a)(2)(A), (B), and (H) governing intrastate impacts and seems to mandate consideration of the effect of one state's SIP for a pollutant on attainment and maintenance for all pollutants in other states. However, it avoided deciding the issue against EPA, concluding that under the facts of this case, the Agency did not abuse its discretion. Deferring to EPA's technical judgment that it lacks adequate models on sulfate formation, it found that failure to address the issue was not unreasonable. Nor was EPA's failure to address the interstate impacts of increases in direct emissions of TSP unreasonable. Because those emissions are regulated under the Connecticut particulate SIP, any increases would be minimal and within EPA's discretion to approve.

The Connecticut Energy Trade Program

The third Connecticut Fund case also named Connecticut Fund for the Environment, Inc. v. Environmental Protection Agency,³⁰ (Connecticut Fund III) concerns Connecticut's Energy Trade Program. The Fund and the City of Middletown challenged EPA's approval of a revision to Connecticut's SO₂ SIP establishing procedures through which the operative sulfur-in-fuel rules may be relaxed for plants which reduce their fuel use through energy conservation. Petitioners alleged that the revision (1) is a relaxation of the Connecticut SIP for particulates in violation of the Act, (2) allows use of the bubble policy³¹ in a nonattainment area in violation of the D.C. Circuit's recent decision in Natural Resources Defense Council, Inc. v. Gorsuch,³² and (3) violates PSD requirements for SO₂.

The Energy Trade Program³³ is a two part process allowing firms which conserve energy to burn higher sulfur fuel. In the first part of the analysis, a maximum allowable sulfur-in-fuel for the entire plant is calculated on the [13 ELR 10040] basis of a formula allowing 0.55 pounds of sulfur per million BTU of energy input.³⁴ Once the sulfur-in-fuel ratio is set, emission limits are calculated for the individual sources in the facility. The state then models the air quality impacts of the new limits and modifies them if they will cause violations of the NAAQS.³⁵

The court quickly dispatched petitioners' argument that EPA should have considered the impact of the program on TSP pollution, relying on its analysis in Connecticut Fund II concerning the separability of SIP provisions for different pollutants. It found further support for this ruling in the fact that the program itself is procedural. Each energy trade application must be approved as a SIP revision and Connecticut requires consideration of impacts on both SO₂ and TSP in this process. Nor does the fact that Connecticut allows minor increases in TSP change the result. First, under the different-pollutant, separate-SIP logic, EPA need not address the issue, since it is the SO₂ SIP that is involved. But even so, the court noted that minimal increases in nonattainment pollutants are allowable in EPA's discretion, citing Alabama Power and Connecticut Fund II.

Petitioners also argued that the Energy Trade Program is barred by the D.C. Circuit's decision in NRDC v. Gorsuch. The Second Circuit read NRDC to bar use of the bubble policy in nonattainment areas, but held that it did not apply in this case. It found that the bubble policy is only utilized in calculating the plant-wide SO₂ limitations, not in determining whether the new limitations are consistent with the SO₂ and TSP air quality standards. Since Connecticut has attained the SO₂ standard this application of the bubble policy does not involve nonattainment pollutants or nonattainment areas and the D.C. Circuit's decision is irrelevant.

In considering petitioners' argument that the Energy Trade Program is inconsistent with PSD requirements, the court first addressed the question of what PSD requirements are in effect in Connecticut. The Fund argued that 1974 regulations, which are inconsistent with the program, govern because Connecticut has not promulgated PSD measures for its SIP. The court disagreed. It found that EPA adopted a general PSD program for the state and that the1974 rules do not survive until the baseline is established where such a PSD program has been promulgated. Rather, Congress intended the PSD program to be suspended until baselines are established. Thus, by requiring PSD review of applications for energy trades affecting areas with PSD baselines, the Connecticut program did all it could to comply with PSD requirements, even though the requirement is meaningless at present.³⁶

Discussion

The Connecticut Fund cases provide a thorough review of the complex Clean Air Act framework governing relaxation of SO₂ emission limits. Connecticut Fund III raises interesting issues concerning use of the bubble policy, but the central questions in the cases concern the transformation of SO₂ into sulfates and interstate pollution provisions of § 110(a)(2)(E). The court had to traverse several uncharted areas of the law in addressing these issues. While relying heavily on EPA in making this journey, the court nevertheless interpreted the interstate pollution control mandate more liberally than did the Agency. The results in these cases will support EPA approval of other SO₂ relaxations that are being challenged by downwind states and environmental groups, but the court's insistence that § 110(a)(2)(E) is a vital provision of the Act which must be given effect will strengthen those challenges and add to the pressure on EPA to find a means of estimating the magnitude of sulfate pollution resulting from increased SO₂ emissions.

Interstate Air Pollution

* Finding Strength in § 110(a)(2)(E). The Second Circuit's major task in the Connecticut Fund trilogy was to bring a measure of clarity to the action-forcing terms of § 110(a)(2)(e) — "prevent attainment and maintenance" and "interfere with" PSD measures. It established that "prevent maintenance" means driving pollution concentrations to levels greater than the NAAQS; that "prevent attainment" means something more than contributing a minimal amount of a nonattainment pollutant; and that "interfering with" PSD measures requires some specific PSD measures to interfere with or air quality impacts approaching the magnitude of those preventing maintenance. With these rulings the court resolved much of the uncertainty surrounding § 110(a)(2)(E).

It also pointedly left open some questions about the provision's meaning. The court refused to adopt EPA's argument that to "prevent attainment" a state had to contribute "substantial" pollution to a nonattainment area in another state. But it did not suggest how much pollution is substantial or where between that level and minimal pollution the "prevent attainment" line might be drawn. Likewise, though rejecting EPA's argument that a state could not "interfere with" a neighbor's PSD measures prior to establishment of baselines for the state's PSD areas, the court left open the question of what types of voluntary PSD measures would qualify for protection from interstate pollution where baselines have not been established.

The court found more punch in § 110(a)(2)(E) than EPA had. Its rejection on EPA's arguments that substantial interstate pollution of nonattainment areas is needed to trigger the section and that § 110(a)(2)(E)(i)(II) does not apply until PSD baselines are established for the receiving area will allow somewhat greater control of air [13 ELR 10041] pollution exports. Similarly, the court's rejection of EPA's use of § 123 in modeling interstate impacts should tighten interstate pollution controls. Overall, the court seemed to be saying that interstate pollution is a special problem, and that it intends to scrutinize closely EPA's decisions on the subject.

That the court found the interstate pollution provisions more stringent than parallel provisions governing intrastate pollution makes sense given the institutional framework for air pollution control established by the Clean Air Act. The court said that under § 110(a) EPA could allow Connecticut to ignore the impacts of its SO₂ SIP revisions on Connecticut PSD and particulate attainment and maintenance. But in strongly worded dicta it suggested that under § 110(a)(2)(E) it could not allow New York to do the same. The result preserves Connecticut's ability to make the basic decisions affecting air quality and pollution control expenditures within its borders while restricting New York's ability to limit the choices available to Connecticut. If Connecticut ignores the TSP impacts of its SO₂ revision and those impacts are significant, it will have to pay the price in a stricter TSP SIP. If New York takes the same stance with pollution transported across Long Island Sound, it is Connecticut that picks up the tab.

* Limitations on § 110(a)(2)(E). While the Connecticut Fund decisions read some strength into § 110(a)(2)(E), they also reveal significant limitations in the provision. For attainment areas, § 110(a)(2)(E)(i)(I) offers at best a last line of defense, barring only pollution exports large enough to push air quality in the receiving area over the NAAQS. The New York variance allowed one power plant to raise SO₂ pollution in Connecticut by 87 micrograms per cubic meter (ug/m³).³⁷ This amounts to nearly one quarter of the total SO₂ pollution allowed by the primary NAAQS (365 ug/m³) and one third of the total SO₂ pollution recorded at the monitoring station with the highest readings mentioned in the Connecticut Fund I decision.³⁸ Thus, § 110(a)(2)(E) does not prevent New York from using up a substantial share of Connecticut's clean air margin.

The Second Circuit made a convincing argument that Congress intended the "prevent maintenance" provision to be secondary to the "interfere with" PSD measures provision in the protection of attainment areas. If the PSD program had been activated in Connecticut it probably would have restricted emissions from the New York plants.³⁹ It seems unlikely, however, that Congress foresaw how the nonattainment program could have the side effect of eliminating the primary protection of attainment areas.

The court suggested that Connecticut could provide added protection against New York pollution of its attainment areas by promulgating voluntary PSD measures, but apparently Connecticut now must accept the LILCO plants' SO₂ pollution as a given for the duration of the variance. This result seems dictated by the court's holding that EPA need consider the interstate impacts only of the sources affected by a given SIP revision.⁴⁰ In other words, the LILCO variance has consumed a portion of Connecticut's clean air margin and later revisions of the New York or Connecticut SIPs must be judged in terms of what is left. The further implication is that New York could use up all of Connecticut's clean air margin for SO₂ if it acted before Connecticut either promulgated voluntary PSD measures or used up the margin itself by relaxing its own SIP. Prompt enactment of voluntary PSD requirements would seem a prudent course for a state in Connecticut's situation. Under these circumstances there is also pressure to relax SO₂ limits for existing Connecticut sources. It seems unlikely that this race to relax SIPs was what Congress intended to accomplish with § 110(a)(2)(E).

As interpreted by the Second Circuit, § 110(a)(2)(E) gives nonattainment areas greater protection from interstate pollution than it gives attainment areas. The court refused to accept EPA's interpretation that attainment areas are protected only against substantial interstate pollution. On the other hand, by holding that EPA has the discretion to allow "minimal" increases in nonattainment area pollution, the court gave the Agency useful flexibility, but also made the determination of how much pollution is excessive a major subjective one.

One of the most important factors in determining the scope of the Clean Air Act's interstate air pollution restrictions is how broad should the focus be in defining the problem. Should it be on individual sources? On all sources covered by a given SIP revision? On all sources in the state which is charged with exporting excessive pollution? Or on all sources upwind of the complaining state? Section 110(a)(2)(E) appears to require that a single source have the prohibited effect (e.g., prevention of attainment) before the offending SIP must be changed.⁴¹ The Connecticut Fund cases did not get to the heart of this question, but they did shed some light on it.

Petitioners in Connecticut Fund I argued that EPA should have considered the cumulative effect of pollution from all sources in New York on Connecticut's air quality before approving the LILCO variances. The court rejected this argument summarily.⁴²

If EPA need not consider pollution from all sources in New York in evaluating the LILCO variances, must it limit its analysis to the impacts of individual sources? The Second Circuit did not address this question directly. It discussed the interstate impacts of the New York SIP revision in terms that indicated that it was considering them as a group.⁴³ In fact, however, in Connecticut Fund I, the [13 ELR 10042] impacts of the two LILCO plants were separate because the case involved the 24-hour standard. They were far enough apart that their emission plumes would not overlap significantly on any given day.⁴⁴

Though the Second Circuit did not answer the question, logic dictates that EPA consider all sources involved in a given SIP revision in applying § 110(a)(2)(E). If, unlike the situation in Connecticut Fund I, the interstate impacts of several sources affected by a given SIP revision overlapped significantly, considering them independently would open the door to approval of a revision which in fact would prevent attainment or maintenance of NAAQS in another state. To accept this result would undermine Congress' clearly stated intention that SIPs demonstrate attainment and maintenance.

Energy Bubbles

While the primary focus of the Connecticut Fund cases was interstate pollution, another important issue was the validity of the Connecticut Energy Trade Program. By upholding the program, the court may have given impetus to a mechanism for reducing reliance on costly low sulfur fuel without increasing SO₂ emissions or requiring installation of expensive SO₂ control technology. However, the court also gave a broad reading to limits on use of the bubble policy in nonattainment areas established in the D.C. Circuit's opinion in Natural Resources Defense Council, Inc. v. Gorsuch.⁴⁵ That decision barred use of the bubble policy to exempt major modifications of existing sources in nonattainment areas from the new source review otherwise required by the Act. Its rationale for this ruling was that earlier decisions of the D.C. Circuit stood for the proposition that for establishing the scope of new source review, the bubble policy had to be applied in attainment areas and could not be used in nonattainment areas. The opinion also included general statements to the effect that the bubble policy was completely inconsistent with nonattainment programs, implying that it could not be used with existing sources either. This issue was not before the court, but there has been speculation about whether other courts would adopt the broad rationale and even more sweeping language of the D.C. Circuit's opinion, or limit it to its facts.⁴⁶

The Second Circuit dodged the issue, but suggested that the D.C. Circuit's opinion should be given its broadest reading. It ruled that NRDC was simply not applicable, using the pollutant-specific SIP logic of Connecticut Fund II. It held that the bubble policy was only used in revising the SO₂ limits of the sources involved in energy trades and was thus not governed by the nonattainment programs of the Act. However, it characterized the NRDC decision as prohibiting any use of the bubble policy with sources covered by the nonattainment program.⁴⁷ While this is merely dictum, it illustrates the ease with which NRDC can be extended beyond its holding.

Deference

Finally, the Connecticut Fund cases illustrate the difficulty of overturning EPA's technical analysis in interstate pollution cases. The Agency prevailed on every technical question, from what models should be used to what the modeling showed. On some issues petitioners mounted sophisticated challenges to EPA modeling data and demonstrated serious weaknesses in EPA's analysis, but the court rejected them because EPA had presented reasonable counter arguments or petitioners had not presented alternative approaches without the flaws. However, the Second Circuit suggested that along with the deference goes responsibility; EPA has an obligation to develop the analytical tools needed to regulate the serious air quality problem of sulfate pollution.

Conclusion

What will be the effect of the Connecticut Fund decisions on efforts to use § 110(a)(2)(E) to force upwind states to take measures beyond those required to comply with the NAAQS locally in order to abate interstate pollution? The Connecticut Fund cases, especially Connecticut Fund I, are likely to dampen those efforts. The court's ruling that minimal increases in nonattainment pollutants do not "prevent attainment" makes the showing of an impermissible pollution export more difficult. Likewise, by upholding EPA's definition of "preventing maintenance" as pushing air quality all the way over the NAAQS, the court increased other petitioners' burdens, particularly since it ruled that the prohibition on interference with PSD measures is not applicable in all clean air areas. The court's repeated deference to EPA on the technical issues in the cases, particularly on the unavailability of a model for estimating SO₂ transformation into particulates over long distances, also should bolster EPA's case in the other § 110(a)(2)(E) matters now pending. The Second Circuit in effect said that EPA is doing the best it can with the tools available.

While supporting EPA's actions on all points in the Connecticut Fund trilogy, the Second Circuit also served notice on the Agency that it may not receive such a positive response in the future. The court refused to accept EPA's interpretation of the law on several key points, and, while not ruling against the Agency, presented detailed arguments against those interpretations. These dicta will strengthen petitioners' legal arguments in future cases, but they will not overcome the technical difficulty in making reasonable estimates of the long range impacts of SO₂ pollution. The absence of approved models showing greater interstate impacts and addressing the transformation of SO₂ into sulfate particulates will continue to hamper § 110(a)(2)(E) litigants. However, when EPA has the models (and it is working on them now) the dicta in the Connecticut Fund cases should take on greater significance.

1. 42 U.S.C. §§ 7401-7642, ELR STAT. 42201.

2. Connecticut v. EPA, Nos. 81-4217, -4228, 13 ELR 20135 (2d Cir. Dec. 1, 1982); Connecticut Fund for the Env't, Inc. v. EPA, No. 81-4227, 13 ELR 20146 (2d Cir. Dec. 1, 1982); Connecticut Fund for the Env't, Inc. v. EPA, No. 81-4202, 13 ELR 20151 (2d Cir. Dec. 1, 1982).

3. The court refers to Connecticut v. EPA as Connecticut Fund I. The Fund was a petitioner, although not the first-listed one. It denominates No. 81-4227 as Connecticut Fund II and No. 81-4202, as Connecticut Fund III. The short titles used by the court are used in this Comment.

4. See, e.g., Ohio v. EPA, Nos. 80-3575, -3576, -3579, -3581, -3582 (6th Cir., petition filed Aug. 1, 1980).

5. The high cost of low sulfur fuel is illustrated by the fact that in 1979 LILCO saved $52 million as a result of the variance at issue in Connecticut Fund I, 13 ELR at 20136 n.11, and one Connecticut utility estimated that it would save $24 million as a result of the SIP revision at issue in Connecticut Fund II, 13 ELR at 20148.

6. For a description of the Connecticut program, see 46 Fed. Reg. 24597 (May 1, 1981).

7. Clean Air Act §§ 171-178, 42 U.S.C. §§ 7501-7508, ELR STAT. 42238-39.

8. Clean Air Act §§ 160-169, 42 U.S.C. §§ 7470-7479, ELR STAT. 42233-36.

9. See Emissions Trading Policy Statement, 12 ELR 30006, 47 Fed. Reg. 15075 (Apr. 7, 1982), for a description of the bubble policy and related matters.

10. See, e.g., Wetstone, Air Pollution Control Laws in North America and the Problem of Acid Rain and Snow, 10 ELR 50001 (1980).

11. 42 U.S.C. § 7410(a)(2)(E), ELR STAT. 42210.

12. 42 U.S.C. § 7426, ELR STAT. 42229.

13. 13 ELR 20135 (2d Cir. Dec. 1, 1982).

14. Section 110(a)(2)(E), 42 U.S.C. § 7410(a)(2)(E), ELR STAT. 42212, provides that to approve a SIP or SIP revision the EPA Administrator must determine that

it contains adequate provisions (i) prohibiting any stationary source within the Statefrom emitting any air pollutant in amounts which will (I) prevent attainment or maintenance by any other State of any such national primary or secondary ambient air quality standard, or (II) interfere with measures required to be incluuluded in the applicable implementation plan for any other State under part C of this subchapter to prevent significant deterioration of air quality or to protect visibility, and (ii) insuring compliance with the requirements of section 7426 of this title, relating to interstate pollution abatement.

15. 42 U.S.C. § 7423, ELR STAT. 42228.

16. The court noted that § 123 was intended to prevent sources from using tall stacks to satisfy NAAQS at nearby monitoring stations by transporting much of the pollution beyond them. Use of the lower good engineering practice stack heights in estimating short range air quality impacts avoids this problem. However, it has the opposite effect with long range transport — it understates the pollution impact of a given level of emissions.Thus, the court found that this application would contravene the purpose of Congress in enacting § 123, 13 ELR at 20141, 20142.

17. The court suggested that "no provision in the Clean Air Act requires the EPA to consider the impact of a proposed revision of one state's implementation plan upon a restructuring of the state implementation plan of another state." It also added that it did "not mean to imply, however, that a situation could never arise where the general scheme of the Clean Air Act would require the Agency to consider the combined effect of two proposed revisions before approving either." 13 ELR at 20142.

18. Connecticut Fund for the Env't, Inc. v. EPA, 13 ELR 20146.

19. 42 U.S.C. § 7410(a)(1), (a)(2)(A), (B), & (H), ELR STAT. 42210. These sections are among those addressing the intrastate impacts of a SIP or SIP revision.

20. 636 F.2d 323, 10 ELR 20001 (D.C. Cir. 1979).

21. "Alternatively, the Agency may take so long to address, or at least measure, the sulfate problem that any further delay could be found to be an abuse of discretion" 13 ELR at 20144.

22. Connecticut Fund for the Env't, Inc. v. EPA, 13 ELR 20151 (2d Cir. Dec. 1, 1982). See note 3, supra.

23. The term "clean air margin" is used here to refer to the difference between existing air quality in an attainment area and the NAAQS.

24. The court concluded:

In the present case, Connecticut points to no specific measures it has taken to control sulfur dioxide levels with which LILCO's emissions will interfere. In addition, there is no evidence that the impact of LILCO's emissions will bring concentrations of SO₂ in Connecticut so close to the NAAQS as to preclude Connecticut from taking advantage of its PSD increment.

13 ELR at 20145. The New York power plants were estimated to contribute a maximum of 87 micrograms per cubic meter (ug/m³) to Connecticut's air when measured on a 24-hour basis. 13 ELR at 20141 n.29. Without this contribution from across Long Island Sound, Connecticut's clean air margin would have been 192.2 ug/m³. Id. The PSD increment for Class II areas, which include the affected areas of Connecticut, is 91 ug/m³. 13 ELR at 20144.

25. 13 ELR 20146 (2d Cir. Dec. 1, 1982) See note 3, supra.

26. The court noted that Connecticut had recently responded to its particulate nonattainment problem by promulgating a new, more stringent set of emission limits for that pollutant. 13 ELR at 20148 n.7.

27. 42 U.S.C. § 7410(a)(2)(A), (B), (H) refers to "a plan" as implementing "a standard." 13 ELR at 20149.

28. Connecticut Fund for the Env't, Inc. v. EPA, 672 F.2d 998, 12 ELR 20306 (2d Cir. 1982); Northern Ohio Lung Ass'n v. EPA, 572 F.2d 1143, 8 ELR 20213 (6th Cir. 1978).

29. 42 U.S.C. § 7401, ELR STAT. 42205.

30. 13 ELR 20151 (2d Cir. Dec. 1, 1982). See note 3, supra.

31. See note 12, supra.

32. 685 F.2d 18, 12 ELR 20942 (D.C. Cir. 1982). For a discussion of this case, see Comment, NRDC v. Gorsuch: D.C. Circuit Bursts EPA's Nonattainment Area Bubble, 12 ELR 10089 (1982).

33. See note 6, supra.

34. This is the same sulfur limit governing sources opting to burn fuel higher in sulfur than the 0.5 percent sulfur rule would allow, correcting for the overage with flue gas desulfurization under existing Connecticut SIP provisions. 46 Fed. Reg. 24597, 24598 (May 1, 1981). In calculating the baseline, energy inputs of all types are taken into account to encourage the firm to cut all energy use. The baseline is calculated for the year after 1971 in which the energy input per unit of production was highest for the facility in question. The facility may continue to use fuel containing that total amount of sulfur, regardless of changes in its energy input. Thus, as energy consumption drops through conservation, the percent of sulfur in the fuel burned can increase, up to a maximum of 2.2 percent.

35. While the total sulfur emitted from the facility apparently must remain the same, this approach could result in a substantial increases in sulfur emissions from individual smokestacks, and thus might produce localized NAAQS violations.

36. 13 ELR at 20155.

37. Connecticut Fund I, 13 ELR at 20141 n.29.

38. Id.

39. The variance would have used up virtually the entire Class II increment of 91 ug/m³. Connecticut Fund i, 13 ELR at 20144.

40. Connecticut Fund I, 13 ELR at 20146.

41. Section 110(a)(2)(E) prohibits "any stationary source" from preventing attainment or maintenance or interfering with PSD measures. 42 U.S.C. § 7410(a)(2)(E), ELR STAT. 42210.

42. Connecticut Fund I, 13 ELR at 20146, citing its earlier decision in Connecticut v. EPA, 656 F.2d 902, 11 ELR 20924 (2d Cir. 1981). It is somewhat puzzling that in the earlier decision the court had justified its refusal to consider cumulative effects in passing on SO₂ variances for two other power plants by noting that when later variances were considered, EPA could consider them in combination with the two it approved then. See 656 F.2d at 908, 11 ELR 20924 at 20927. In fact, however, EPA did include the effects of the two earlier variances in its modeling and found that they were negligible. (Phone conversation with Robert Predale, Environmental Scientist, U.S. EPA, Region II, Jan. 27, 1982.)

43. In Connecticut Fund I, the court referred to the pollution in collective terms such as as "LILCO's emissions." See, e.g., 13 ELR at 20141.

44. Phone conversation with Robert Predale, Environmental Scientist, U.S. EPA, Region II, Jan. 27, 1982.

45. 685 F.2d 718, 12 ELR 20942 (D.C. Cir. 1982).

46. See, e.g., Comment, NRDC v. Gorsuch: D.C. Circuit Bursts EPA's Nonattainment Area Bubble, 12 ELR 10094, 10095 (1982).

47. "Gorsuch forbids bubbling only when a NAAQS for the pollutant in question has not been attained." Connecticut Fund III, 13 ELR at 20154.

13 ELR 10036 | Environmental Law Reporter | copyright © 1983 | All rights reserved