The STATE OF OKLAHOMA, Oklahoma Scenic Rivers Commission and Pollution Control Coordinating Board; Save the Illinois River (STIR), a non-profit corporation of the State of Oklahoma; City of Fayetteville, Arkansas; the Beaver Water District; State of Arkansas; Arkansas Department of Pollution Control and Ecology, Petitioners, v. ENVIRONMENTAL PROTECTION AGENCY, Respondent, Oklahoma Wildlife Federation, Intervenor.
Nos. 89-9503, 89-9507 and 89-9516.
United States Court of Appeals, Tenth Circuit.
July 11, 1990.
908 F.2d 595 | 31 ERC 1741 | 59 USLW 2079 | 21 Envtl. L. Rep. 20,206
John Steven Clark, Atty. Gen. (R.B. Friedlander, Sol. Gen., Attys. for State of Ark., Steven Weaver, Anne Roberts Bobo, Attys. for Arkansas Dept. of Pollution Control & Ecology, Little Rock, Ark., James N. McCord, Atty. for City of Fayetteville, Ark., Niblock Law Firm, Atty. for Beaver Water Dist., Fayetteville, Ark., on the brief), for petitioners, City of Fayetteville, Ark., Beaver Water Dist., State of Ark., Arkansas Dept. of Pollution Control and Ecology.
Gary Guzy (Catherine A. Winer, Pat Rankin, U.S. E.P.A., and Richard B. Stewart, Asst. Atty. Gen., with him on the briefs), U.S. Dept. of Justice, Washington, D.C., for respondent.
Ed Edmondson, Muskogee, Okl., appearing for Save the Illinois River.
Before ANDERSON and BRORBY, Circuit Judges, and THEIS,* District Judge.
BRORBY, Circuit Judge.
In these consolidated appeals, appellants challenge certain actions of the U.S. Environmental Protection Agency (EPA) in issuing a discharge permit pursuant to the National Pollutant Discharge Elimination System (NPDES) of the Clean Water Act,
OVERVIEW
The city of Fayetteville, Arkansas, applied to EPA for an NPDES permit for a new municipal wastewater treatment plant. Fayetteville proposed to discharge treated wastewater via a split flow into the White River in Arkansas and into Mud Creek, a tributary of the Illinois River, an Arkansas-Oklahoma interstate stream. The State of Oklahoma and a nonprofit group, Save The Illinois River (STIR), requested denial of the permit. The State of Arkansas and the Oklahoma parties requested an evidentiary hearing on EPA‘s issuance of the permit. A hearing request was granted in part and denied in part by an Administrative Law Judge (ALJ), and the partial denial was upheld by the EPA Administrator acting through his Chief Judicial Officer (CJO). After the evidentiary hearing, the ALJ determined that the permit would not have an undue impact on water quality or violate Oklahoma‘s water quality standards (WQS). This initial decision was appealed by both Arkansas and Oklahoma. On appeal, the ALJ‘s decision was affirmed in part and reversed in part and remanded for a determination whether the record showed by a preponderance of the evidence that the permitted discharge would not cause an actual, detectable violation of WQS. On remand the ALJ reviewed the record and made detailed findings. He concluded that the permit could issue as written, finding that it would not result in any measurable violations of Oklahoma‘s WQS. The ALJ‘s decision on remand was appealed to the CJO who upheld it in a decision dated December 22, 1988. These petitions for review followed.
Appellants the State of Oklahoma, Oklahoma Scenic Rivers Commission, Oklahoma Pollution Control Coordinating Board, and STIR (the “Oklahoma parties,” or Oklahoma) set forth ten issues in their joint brief-in-chief. Essentially they contend that EPA erred in concluding that the permit would not violate Oklahoma‘s WQS; that EPA did not properly consider the Wild and Scenic Rivers Act,
BACKGROUND
The cornerstone of the Clean Water Act,
Discharges of pollutants must comply with limitations established in and pursuant to the Act. “Effluent limitations,” i.e., limits on “quantities, rates, and concentrations of chemical, physical, biological, and other constituents which are discharged from point sources,”
Federal and state effluent limitations and water quality standards are transformed into individual point source obligations through NPDES discharge permits.
EPA issued Fayetteville‘s NPDES permit because at the time this proceeding commenced Arkansas had not yet been delegated permitting authority pursuant to
The permit (NPDES Permit No. AR0020010) specifies that half of the city‘s treated wastewater will be discharged to the White River in Arkansas (this portion of the discharge is not in contention here), and half will be discharged to the Illinois River basin. Specifically, this latter effluent will be discharged to an unnamed stream in northwestern Arkansas, which flows approximately two miles before joining Mud Creek. Mud Creek flows three miles from that point to its confluence with Clear Creek, thirteen miles upstream from the Illinois River in Arkansas. Twenty-two miles downstream from Clear Creek--and thirty-nine miles from the Fayetteville plant--the Illinois River crosses the state line into northeastern Oklahoma and almost immediately flows into Lake Frances. A segment of the Illinois River (including Lake Frances) from the Oklahoma-Arkansas state line to Tenkiller Ferry Reservoir has been designated an Oklahoma state scenic river and was proposed for study as a potential addition to the National Wild and Scenic Rivers System when the WSRA was enacted in 1970.
The Fayetteville permit sets limits on the amounts of certain pollutants that may be discharged and establishes maximum or minimum effluent concentrations of those pollutants and other chemical parameters. Permit, EPA Supp. Addendum at 12-30. The permit prohibits the discharge of any incompletely treated effluent to Mud Creek.
ANALYSIS
I. Standard of Review
Review of the EPA rulings on appeal here is governed by the Administrative Procedure Act,
Nevertheless, the agency must examine the relevant data and articulate a satisfactory explanation for its action including a “rational connection between the facts found and the choice made.” In reviewing that explanation, we must “consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.” Normally, an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.
Id. at 43 (citations omitted).
Determining the extent of EPA‘s authority under the Clean Water Act is a question of law that we review de novo. “Our first inquiry is whether ‘Congress has directly spoken to the precise question at issue. If the intent of Congress is clear that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.’ ” Martin Exploration Management Co. v. FERC, 813 F.2d 1059, 1065 (10th Cir. 1987) (quoting Chevron, U.S.A., Inc. v. NRDC, 467 U.S. 837, 842-43 (1984), rev‘d on other grounds, 486 U.S. 204 (1988)). However, where the statute is ambiguous, EPA‘s construction, as that of the agency charged with administering the statute,1 is entitled to substantial deference. Chevron, 467 U.S. at 844. If EPA‘s interpretation of the Clean Water Act is reasonable, we should not disturb it unless it “is contrary to the policies Congress sought to implement in enacting the statute.” Martin Exploration Management Co., 813 F.2d at 1065; see also Chevron, 467 U.S. at 845.
II. Preliminary Procedural Matters
As an initial matter we address EPA‘s argument that Arkansas lacks standing to challenge EPA‘s interpretation of the Clean Water Act. EPA reasons that
Arkansas, by not challenging any of the terms of the Fayetteville permit, has failed to state a justiciable case or controversy.... Arkansas’ claim is purely hypothetical and would not be redressed by a favorable decision of this Court, just as its allegations as to future permit conditions are purely speculative.
EPA Brief at 13-14. EPA also rejects Arkansas‘s argument that it may be collaterally estopped in subsequent proceedings if it does not pursue this argument now by assuring Arkansas that EPA would not contest the justiciability of the claim if raised in the context of future permit decisions. EPA Brief at 14; see Arkansas Reply Brief at 11 n. 18.2
Section 509 of the Clean Water Act provides that “[r]eview of [EPA‘s] action ... in [, inter alia,] issuing or denying any permit under section 1342 of this title ... may be had by any interested person.”
Any person has standing in court to challenge administratively developed standards, rules and regulations under the Act. The courts are increasingly adapting this test to what administrative actions are reviewable.... The Courts have granted this review to those being regulated and to those who seek “to protect the public interest in the proper administration of a regulatory system enacted for their benefit.” Since precluding review does not appear to be warranted or desirable, the bill would specifically provide for such review within controlled time periods....
... For review of permits issued under section 402 [
33 U.S.C. Sec. 1342 ] ..., the section places jurisdiction in the U.S. Court of Appeals....
S. Rep. No. 414, 92d Cong. 2d Sess., reprinted in 1972 U.S. Code Cong. & Admin. News 3668, 3750-51 (emphasis added; citation omitted); see also Conf. Rep. No. 1236, reprinted in 1972 U.S. Code Cong. & Admin. News 3776, 3825. The Supreme Court reiterated the expansive language of the Senate Report in Middlesex County Sewerage Auth. v. National Sea Clammers Ass‘n, 453 U.S. 1, 14 n. 23 (1981) (“review provisions of Sec. 509 are open to ‘[a]ny person,’ S. Rep. No. 92-414, p. 85 (1971)“); cf. Montgomery Envtl. Coalition v. Costle, 646 F.2d 568, 576-78 (D.C. Cir. 1981) (
It would strain the meaning of “any interested person” to exclude from those eligible to obtain review of an EPA permit action the state in which the publicly owned treatment works seeking the permit is located, which partially financed the facility‘s construction, and which, among other entities, has review and approval authority over the facility‘s construction and operation. We conclude Arkansas does have standing to challenge EPA‘s determination that Oklahoma water quality standards apply to the plant.3
Before undertaking a discussion of the merits, we must consider a second procedural issue--whether Arkansas has exhausted its administrative remedies. EPA‘s regulations provide that a petition “for review of any initial decision ... is, under
Although the parties do not raise this issue, we have detected two arguable procedural deficiencies in Arkansas‘s appeal in light of
Secondly, EPA‘s action in this case was arguably not “complete” with respect to the Arkansas parties because the CJO failed to “issue an order either granting or denying [Arkansas‘s] petition for review.”
Nevertheless, under the circumstances of this case, we do not view these shortcomings as fatal to Arkansas‘s appeal. EPA‘s position on the basic issue raised by the Arkansas parties--whether the Fayetteville discharge must comply with Oklahoma WQS--has been clear since the ALJ‘s initial decision and is directly at odds with Arkansas‘s position. In his second and final opinion the CJO clearly affirmed his June 1988 ruling that Oklahoma WQS are applicable to the Fayetteville discharge. In so doing, he implicitly, if not expressly, denied Arkansas‘s petition for review of the ALJ‘s decision on remand. Thus, it would be fruitless to remand to the agency for mere technical compliance with subsection (c)(1)‘s requirement for “an order ... denying review.” Cf. Mathews v. Eldridge, 424 U.S. 319 (1976); Koerpel v. Heckler, 797 F.2d 858, 862 (10th Cir. 1986); Clonce v. Presley, 640 F.2d 271, 273 (10th Cir. 1981) (citing Lewis v. New Mexico, 423 F.2d 1048, 1049 (10th Cir. 1970)).
Moreover, no objection to Arkansas‘s failure to seek review of EPA‘s initial decision should now be allowed, given that Arkansas participated in the review of the initial decision (by responding to Oklahoma‘s petition) and the CJO expressly provided that the “parties will have the opportunity to petition for review of the ALJ‘s decision on remand.” Order on Petitions for Review, R., A-28, at 17 (emphasis added). Accordingly, the Arkansas parties’ appeal is ripe for our review, and we proceed with our discussion of the merits.
III. Statement of Issues
Arkansas poses the fundamental question in this case: Does the Clean Water Act require a point source of pollution to comply with the water quality standards of all affected downstream states? Oklahoma assumes such a requirement in that it challenges EPA‘s determination that the Fayetteville permit would not result in violations of Oklahoma‘s water quality standards and argues accordingly that no discharge to Oklahoma‘s Illinois River system should be allowed.
Oklahoma formulates the issues on appeal as “[w]hether the Chief Judicial Officer erred in denying review” of various ALJ rulings and whether the CJO and ALJ “erred in [refusing] to reopen the evidentiary hearing.” Despite this formulation, it seems clear that the Oklahoma parties’ chief concerns relate to the substantive issues underlying these procedural questions. The substantive issues are: (1) the adequacy of the treatment technology employed by the Fayetteville plant and the possible superiority of land application methods; (2) the propriety of considering evidence concerning future reductions in the discharges of other Arkansas cities; (3) the propriety of relying on “protective language” in the permit authorizing more stringent discharge limitations if shown to be necessary by an ongoing study of the Illinois River; (4) the correctness of EPA‘s interpretation and application of Oklahoma‘s beneficial use limitation, nutrient standard, and anti-degradation policy; (5) the relevance of new information concerning overflows at the old treatment plant; and (6) whether Fayetteville met its burden of proof in showing that a permit should be issued for its treatment plant. Our review of the record convinces us that we need not resolve many of the issues raised by the Oklahoma parties. In the following pages we address first the statutory interpretation question posited by Arkansas and then a significant issue not raised by any party--the significance of evidence of existing degradation of Illinois River water quality.
A. Construction of the Clean Water Act
1. The Opposing Views
The full ramifications of Arkansas‘s formulation of the Clean Water Act issue are exposed once it is realized that an upstream state has the ability (if not the legal right) largely to control the quality of certain of the waters of a downstream state. It can accomplish this simply by setting and enforcing its own water quality standards and releasing water of that quality to the downstream state. If the upstream state‘s water quality standards are lower than those considered desirable by the downstream state, so will be the actual quality of the interstate waters in the downstream state. In other words, the lowest common denominator will prevail. The ultimate question posed to this court is whose water quality standards take precedence under the Clean Water Act--the upstream state‘s, the downstream state‘s, the federal government‘s, or nobody‘s. We conclude that no state “imposes” its standards on another state, but rather that the Clean Water Act mandates compliance with federal law, including the federally approved water quality standards of affected states.
Specifically, Arkansas alleges an affected downstream state “may advise and make recommendations, but nowhere in the Clean Water Act did Congress authorize affected states such as Oklahoma to impose their water quality standards upon a discharger in another state.” Arkansas‘s Brief at 39. We treat this, the principal issue of this case, as whether the Clean Water Act requires that any discharge permitted under
EPA‘s Chief Judicial Officer, in his first order in this case dated June 28, 1988, stated the law and applied it as follows:
The CWA requires an NPDES permit to impose any effluent limitations necessary to comply with applicable state water quality standards.... The meaning of [
33 U.S.C. Sec. 1311(b)(1)(C) ] is plain and straightforward. It requires unequivocal compliance with applicable water quality standards, and does not make any exceptions for cost or technological feasibility....... In this case, the permit should be upheld if the record shows by a preponderance of the evidence that the authorized discharge would not cause an actual detectable violation of Oklahoma‘s water quality standards.
Order on Petitions for Review, R., A-28, at 11-13. The CJO explained that in an interstate dispute the “only applicable water quality standards are those that have been approved by EPA under the CWA.” Order on Petitions for Review at 11 n. 13 (citing Illinois v. City of Milwaukee, 731 F.2d 403, 413-14 (7th Cir. 1984), cert. denied, 469 U.S. 1196 (1985)). In noninterstate disputes, however, “the source state may impose more stringent non-EPA-approved water quality standards in NPDES permits under
On remand, the ALJ expressed similar views:
It is clear that an out-of-state source must meet the W.Q.S. of another downriver state. See
Sec. 401(a)(2) of the CWA [33 U.S.C. Sec. 1341(a)(2)] ;40 C.F.R. Secs. 122.4(D) and122.44(d)(4) ; International Paper Co. v. Ouellette, 479 U.S. 481 (1987). Therefore the Fayetteville discharge must meet Oklahoma‘s W.Q.S. as they exist at the border of the two states....... To accept [the Arkansas parties’ argument that the beneficial use limitations do not apply to Fayetteville] would violate the principals [sic] set out above since it is premised on the notion that such standards only apply to sources located in the State of Oklahoma. There is no factual issue among the parties that the Illinois River at the border of the two states is a Class (A) River and therefore the standards applicable to pollution crossing that border must comply with Oklahoma‘s W.Q.S. as they exist at that point. Any other interpretation would allow a source to locate its discharge just across the line in Arkansas and freely violate Oklahoma standards. Such a result is contrary to the [Clean Water Act], regulations and Court decisions.
Decision on Remand, R., A-33, at 4-5. The ALJ‘s interpretations of Oklahoma‘s WQS, including the Beneficial Use Limitations, were ultimately affirmed by the CJO. The CJO also reiterated the mandate of his first order--that ” ‘the permit should be upheld if ... the authorized discharges would not cause ... [a] violation of Oklahoma‘s water quality standards,’ “--and accepted the ALJ‘s conclusion that no violation would occur. Second Order on Petitions for Review, R., A-37, at 7-8.
The Arkansas parties contend we need look no farther than the Clean Water Act to decide this issue because “Congress has clearly manifested its intent [in the CWA] that affected states cannot impose their water quality standards upon dischargers in other states.” Arkansas Brief at 42; see
We do not find the Clean Water Act, on its face, quite as clear a manifestation of congressional intent on this issue as any of the parties suggests. Significantly, however, EPA‘s interpretation is not one the agency adopted only, or in the first instance, in the context of this permit proceeding. Rather, EPA‘s position herein is consistent with its CWA-implementing regulations. For example,
2. The Parties’ Statutory Arguments
In defending its construction of the CWA the EPA relies principally on
In order to carry out the objective of this chapter [i.e., to “restore and maintain the chemical, physical, and biological integrity of the Nation‘s waters,”
33 U.S.C. Sec. 1251 ] there shall be achieved ... not later than July 1, 1977, any more stringent limitation, including those necessary to meet water quality standards, ... established pursuant to any State law or regulations (under authority preserved bysection 1370 of this title) ... or required to implement any applicable water quality standard established pursuant to this chapter.
(Emphasis added.)
EPA rejects Arkansas‘s argument that these sections are “mere timing provisions.”
Arkansas counters that
First,
Second, the “waters ... of such States” language, which Arkansas deems significant, occurs in and applies only to the second of two principal provisions of
Third, thoughtful consideration of the language of
there shall be achieved ... any more stringent limitation, including those necessary to meet water quality standards ... established pursuant to any State law or regulations ... or required to implement any applicable water quality standard established pursuant to this chapter
(emphasis added)--exposes the irrationality of Arkansas‘s argument. In order to ensure that the EPA-approved water quality standards in all states are “met” or “implemented,” it is “necessary” to require dischargers to meet the applicable requirements of other affected states as well as those of the source state. There could be no assurance of achieving a state‘s more stringent WQS if an upstream, out-of-state discharger were not required to comply with those standards.
EPA concludes and we agree that Arkansas‘s construction of the Act would make achieving downstream water quality standards “impossible in many circumstances or ... possible ... only by imposing a disproportionate burden on dischargers located in the downstream state.” EPA Brief at 21.10 Moreover, rewarding sources for locating in states with less stringent water quality requirements (by relieving them from complying with more stringent downstream WQS) would also result in “pollution shopping,” contrary to Congress‘s intent in passing the 1972 CWA amendments.11
Arkansas counters that EPA‘s construction of the Act would have “chaotic” consequences because any downstream state could impose its requirements on proposed sources in any upstream state. Arkansas Brief at 46-47. Thus, Arkansas hypothesizes, a permit authorizing a discharge to the Mississippi River in Minnesota would be subject to challenge based on the water quality standards of each of the nine downstream states.
We find little practical merit in Arkansas‘s argument. The ability, as well as the authority, to require compliance with the WQS of downstream states is necessarily limited by the ability to measure a source‘s impact on the water quality of the receiving waters. At some point downstream, the impact on water quality of a particular pollution source becomes so attenuated as to be undetectable. Assuming the quality of the receiving waters currently meets or exceeds standards, there can be no violation of standards if the impact of the proposed source on the water quality could not be measured.12 Nor is it “impossible to predict the standard[s]” applicable to a new discharge, as Arkansas claims. First, EPA approval of state WQS determines the potentially applicable rules. Furthermore, the permitting system established in the 1972 and 1977 amendments to the CWA clearly provides for consultation with and input by states that may be affected. Finally, computer modeling (such as that performed for the Fayetteville plant) can predict the extent of a new source‘s potential impact, thus demonstrating which states’ WQS must be met.
3. Illinois v. Milwaukee and Ouellette
Arkansas cites International Paper Co. v. Ouellette and Illinois v. Milwaukee in support of its statutory construction argument, but that reliance is misplaced. In each of those cases an affected state was seeking to enjoin an ongoing discharge in another state by resort to its own state law nuisance remedies. 479 U.S. at 483; 731 F.2d at 404. In contrast, this case is a permitting, rather than an enforcement, action wherein Oklahoma seeks to ensure compliance with federal law, i.e., its EPA-approved WQS. The Seventh Circuit in Illinois v. Milwaukee recognized this distinction when it pointed out that
Illinois’ basic grievance is that the permits issued to Milwaukee pursuant to the [CWA] do not impose stringent enough controls on the discharges. Nevertheless, Illinois failed to participate in the permit issuing process when the Milwaukee permits were issued.... [T]hat process seems now to be the appropriate federal forum for adjusting the competing claims of states in the environmental quality of interstate waters.
731 F.2d at 412-13 n. 5. The court found that Illinois had “not sought to enforce an effluent limitation under Wisconsin [law] nor sought to enforce federal limitations as provided for under the 1972 [CWA]” and concluded that the CWA “preclude[s] the type of application of state law sought by Illinois.” 731 F.2d at 414.
The Supreme Court‘s decision in Ouellette is somewhat more problematic, even though, like Illinois v. Milwaukee, it is not factually or procedurally similar to this case. The specific issue in Ouellette was whether the CWA preempted a common law nuisance suit filed in a Vermont court under Vermont law against a New York discharger, which was the source of the alleged injury in Vermont. The Court concluded that “Vermont nuisance law is inapplicable to a New York point source,” 479 U.S. at 497; however, it chose to express its holding more broadly. The Court stated: “We hold that when a court considers a state-law claim concerning interstate water pollution that is subject to the CWA, the court must apply the law of the State in which the point source is located,”
While source States have a strong voice in regulating their own pollution, the CWA contemplates a much lesser role for States that share an interstate waterway with the source (the affected States). Even though it may be harmed by the discharges, an affected State only has an advisory role in regulating pollution that originates beyond its borders.... Significantly, however, an affected State does not have the authority to block the issuance of the permit if it is dissatisfied with the proposed standards. An affected State‘s only recourse is to apply to the EPA Administrator, who then has the discretion to disapprove the permit if he concludes that the discharges will have an undue impact on interstate waters.
Ouellette, 479 U.S. at 490-91, quoted in Arkansas Brief at 41.
While we agree these statements seem to undercut EPA‘s position, it is beyond dispute that they are dicta and not controlling here. Ouellette was an enforcement action in which the issue was the availability of a nuisance remedy under the common law of an affected state against an out-of-state discharger. Even the Court‘s broadly stated holdings were expressed in terms of “a state-law claim concerning interstate water pollution,” 479 U.S. at 487 (emphasis added), and “applying the law of an affected State,”
Moreover, Ouellette contains other dicta that mitigate the apparent impact of the language quoted above. For instance, the Court noted that nothing in its decision affected the plaintiffs’ right to “pursue remedies that may be provided by the Act.” 479 U.S. at 498 n. 18. The Court pointed out, as examples, that the plaintiffs had the opportunity to protect their interests before permit issuance by commenting on and objecting to the proposed permit conditions, and that they still were entitled to bring a citizen suit to compel compliance with the permit.
Ouellette also suggested that what the plaintiffs there sought to do was akin to establishing a second permit system, which the Court held is disallowed by
The Court‘s opinion in Ouellette corroborates this reasoning. As the Court observed:
Nothing in the Act gives each affected State this power to regulate discharges. The CWA carefully defines the role of both the source and affected States, and specifically provides for a process whereby their interests will be considered and balanced by the source State and the EPA. This delineation of authority represents Congress’ considered judgment as to the best method of serving the public interest and reconciling the often competing concerns of those affected by the pollution. It would be extraordinary for Congress, after devising an elaborate permit system that sets clear standards, to tolerate common-law suits that have the potential to undermine this regulatory structure.
479 U.S. at 497. Plainly, Ouellette was concerned not with the CWA‘s provisions for incorporating a downstream‘s water quality criteria in the permitting process, but with preventing a downstream state from circumventing or superseding that process by imposing on an already-permitted source additional requirements based on its own state law. So viewed, Ouellette is entirely consistent with EPA‘s interpretation of the applicability of Oklahoma‘s WQS. Cf. Champion, 648 F. Supp. at 1400 (concluding that nothing in Ouellette required a modification of the decision at 648 F. Supp. 1390 that a North Carolina discharge permit must require compliance with an applicable Tennessee WQS).
4. The Statutory and Regulatory Framework
The erroneous interpretation of Ouellette, which Arkansas advocates, runs aground when the Clean Water Act is considered as a whole. The Act contains several mechanisms for ensuring that minimum water quality and pollution criteria will apply to all navigable waters of the United States; for example, prohibiting the discharge of pollutants except pursuant to a permit,
a. 33 U.S.C. Sec. 1341
EPA finds support for its action here in certain of the foregoing sections. In addition, we consider
Whenever such a discharge may affect, as determined by the [EPA] Administrator, the quality of the waters of any other State, the Administrator ... shall so notify such other State.... If ... such other State determines that such discharge will affect the quality of its waters so as to violate any water quality requirement in such State, and ... notifies the Administrator ... and requests a public hearing ..., the licensing or permitting agency shall hold such a hearing.... [The licensing or permitting] agency, based upon the recommendations of such State, ... shall condition such license or permit in such manner as may be necessary to insure compliance with applicable water quality requirements. If the imposition of conditions cannot insure such compliance such agency shall not issue such license or permit.
” ‘[T]he purpose of the [
Arkansas disputes that “applicable water quality requirements” in
The legislative history of the certification statute sheds additional light on this matter. In 1977 Congress amended the statute to add section 303 [
S.Rep. No. 370, at 72, reprinted in 1977 U.S.Code Cong. & Admin.News 4326, 4397; H.Conf.Rep. No. 830, at 96, reprinted in 1977 U.S.Code Cong. & Admin.News 4424, 4471.16 According to the committees, the amendment was not meant to change the law but to follow and clarify the original congressional intent that “State water quality standards would be imposed through Section 301, and thus certification by the State would include consideration of water quality standards.” 1977 U.S.Code Cong. & Admin.News at 4397. The conference committee added that “[s]ection 303 is always included by reference where section 301 is listed.” H.Conf.Rep. No. 830, at 96, reprinted in 1977 U.S.Code Cong. & Admin.News 4424, 4471. The Senate committee offered this further explanation of the amendment:
[A]ll States have approved water quality standards. Thus, it is reasonable to require that Federal permits and licenses should take into account State water quality plans, standards and requirements adopted under section 303 to assure maintenance of water quality in the respective States.
Id. at 4398. Neither the statute as amended nor the committee reports concerning the bills distinguish between source and affected states. Thus, EPA‘s view that sources subject to NPDES permits must comply with all approved state water quality standards is a reasonable interpretation in light of this history.
b. 33 U.S.C. Sec. 1342
Also germane to EPA‘s construction of the CWA is the fact that, in those states authorized to issue NPDES permits, the EPA Administrator retains authority to veto any proposed permit if he objects to its issuance.
Although several of the CWA terms discussed in the foregoing paragraphs have no direct application to the permit in this case (because EPA, not Arkansas, is the permitting agency), they reflect the objectives and policies behind the Act and the statutory framework established for implementing them. Because nothing in the Act suggests that permits issued by states are subject to more stringent requirements than those issued by EPA--indeed,
The Arkansas parties construct a similar argument (based on the relation between EPA‘s and the states’ permitting responsibilities) to urge an opposite result, however. They contend:
Section 1342(b)(5) very clearly indicates that an affected state can only be an advisor to the source state in the permitting process when that process has been delegated to a state to administer.... Thus, it makes little sense to suggest ... that a source state discharger must comply with affected state water quality standards when the permitting authority is the EPA rather than a delegated state.
Arkansas Brief at 38-39.20 Arkansas correctly suggests it would make “little sense” if the applicability of a downstream state‘s water quality standards depended on what entity (EPA or the source state) issues the permit. However, Arkansas‘s argument that affected states are limited to an advisory role contains several fatal flaws.
First, in arguing, in effect, “an affected state may not require a source state to comply with the former‘s WQS; therefore, EPA may not require such compliance,” Arkansas‘s fundamental premise is faulty. The fact that an affected state may have only an “advisory role” under
Second, Arkansas‘s argument focuses on one paragraph in isolation, rather than in the context of the entire Act, or even in the context of
Arkansas‘s argument also overlooks the fact that
Finally, Arkansas‘s argument must fail in the face of other CWA provisions heretofore discussed--in particular, EPA‘s authority to veto permits and to suspend state programs if they do not meet the requirements of the Act,
c. EPA‘s “Upset” Regulation
We find still further support for EPA‘s construction of the CWA in the views the agency expressed in an earlier rulemaking proceeding. In the course of promulgating final regulations providing dischargers with a defense to violating effluent limits during unavoidable source “upset” conditions, EPA stated that “the CWA requires strict compliance with water quality standards“; thus, “water quality standards are ... legally required to be met at all times.” 49 Fed.Reg. 37,998, 38,038 (1984), quoted in Sierra Club v. Union Oil Co., 813 F.2d 1480, 1489 (9th Cir.1987), judgment vacated, 485 U.S. 931, 108 S.Ct. 1102, 99 L.Ed.2d 264 (1988); see
The final “upset” rule provides that in certain narrowly defined circumstances technology-based effluent limitations may be exceeded (i.e., failure of pollution controls may be allowed).
This view that all potentially affected stream segments would have to be monitored reflects EPA‘s conviction that an upstream source whose effluent might affect the water quality of downstream states must comply with the WQS of those states. It is also consistent with EPA‘s belief that “strict compliance” with water quality standards is required by the CWA, because such compliance would be impossible if sources could disregard the WQS of states other than the source state.
According to EPA, the requirement of strict compliance with WQS derives from
[T]he gains made as a result of the 1977 requirements could evaporate in the middle of the next decade if only the 1977 [effluent limitations] and new source performance standards are applied. Thus, for many riverways ..., pressure must be maintained to assure improved water quality and to avoid slipping back.
....
The Committee intends that current effluent limitations ... should represent a “floor” or minimum requirement of the modifications authorized by this section. Current levels of discharge must not be relaxed by this provision because that would imply additional treatment requirements on other point or nonpoint source dischargers.
Id. at 42, 44, reprinted in 1977 U.S.Code Cong. & Admin.News at 4367, 4369 (emphasis added).25 The Committee also stated: “There is nothing in these new provisions which in any way preempts the rights of States to have more stringent water quality standards or associated effluent limitations....” Id. at 43, U.S.Code Cong. & Admin.News at 4368.
d. 33 U.S.C. Sec. 1365
One final provision of the CWA deserves mention in our discussion of the statutory interpretation issue. Section 505(h),
Clearly, the injury sustained by a state for which
Section 1365 reminds us that, under the CWA, effluent limitations are not an end in themselves, but simply a means to an end--the desired water quality. The plainest evidence of this can be found in
Whenever, in the judgment of the [EPA] Administrator, discharges of pollutants from a point source or a group of point sources, with the application of effluent limitations required under section 1311(b)(2) of this title, would interfere with the attainment or maintenance of that water quality in a specific portion of the navigable waters which shall assure protection of public water supplies, agricultural and industrial uses, and the protection and propagation of a balanced population of shellfish, fish and wildlife, and allow recreational activities in and on the water, effluent limitations (including alternative effluent control strategies) for such point source or sources shall be established which can reasonably be expected to contribute to the attainment or maintenance of such water quality.
The limitations necessary to achieve a given level of water quality in one reach of a waterway may require more control of effluents than that attainable through application of the best available technology. Where that is desirable to implement the policies of the Act, and feasible, [this section] provides the authority to impose controls based on water quality.
....
The concept of “alternative effluent control strategies” is necessary to account for [certain] difficulties in simply setting more stringent effluent limitations.... [F]urther reduction of the level of effluent entering the affected waters may not be possible through control technology, yet essential to water quality. Alternative effluent control strategies, such as the transportation of effluents to other less affected waters or the control of in-plant processes would have to be developed.
S.Rep. No. 414, reprinted in 1972 U.S.Code Cong. & Admin.News 3668, 3712-13 (emphasis added).
This section and its legislative history reveal the preeminent importance of water quality--actual and desired--in the framework of the CWA. Significantly, they lack evidence of any intent to limit the scope of
Considered together, all of the provisions of the CWA discussed above (Secs.
B. Significance of Existing Violations of Illinois River Water Quality Standards
There is substantial evidence in the record of ongoing violations of Illinois River water quality standards, yet neither of the EPA judicial officers nor any of the parties addresses whether, or how, this is relevant to Fayetteville‘s application to discharge to the Illinois River. We believe this situation poses an issue of critical importance --whether a new discharge may be permitted when the applicable water quality standards are already being violated.28 Guided by the Supreme Court‘s pronouncement that an agency decision is arbitrary and capricious if the agency “entirely failed to consider an important aspect of the problem [or] offered an explanation for its decision that runs counter to the evidence before the agency,” Motor Vehicle Mfrs., 463 U.S. at 43, 103 S.Ct. at 2856, we conclude EPA‘s decision to issue the Fayetteville permit was arbitrary and capricious. The agency‘s decision is also flawed by misinterpretation and misapplication of two important Oklahoma water quality regulations and by arbitrary disregard for certain expert testimony. For these reasons, discussed more fully below, we hold that the Clean Water Act prohibits granting an NPDES permit under the circumstances of this case (i.e., where applicable water quality standards have already been violated) and reverse EPA‘s decision to permit Fayetteville to discharge any part of its effluent to the Illinois River Basin.
1. Law Applicable to Oklahoma Scenic Rivers
The Upper Illinois River, including Lake Frances, from the Arkansas state line down to the 650-foot elevation level of Tenkiller Ferry Reservoir, is designated an Oklahoma state scenic river.
As a preliminary matter, Oklahoma contends and we agree that EPA‘s judicial officers erred in concluding that Oklahoma‘s nutrients standard,
In addition to the nutrients standard, Oklahoma‘s “Anti-Degradation Policy,”
The Beneficial Use Limitations regulation provides that scenic rivers “are protected by prohibition of any new point source discharge of wastes ... except under conditions described in Section 3 [the Anti-Degradation Policy].”
Closer examination of the language and structure of the Anti-Degradation Policy, guided by the minimum requirements for such policies set forth in EPA‘s regulation, confirms our plain language construction.34 The Oklahoma regulation allows “no degradation” of water quality in designated scenic rivers. “Limited degradation” is permitted limited only in other “high quality waters” where the existing water quality “exceeds those levels necessary to support propagation of fish, shellfish, wildlife, and recreation.”
We conclude the requirements of the Beneficial Use Limitations/Anti-Degradation Policy are violated when the water quality of a scenic river undergoes any human-caused, detectable change. By “detectable change” we mean any detectable change in a water quality parameter such as turbidity or phosphorus (with the perhaps unnecessary qualification that an improvement in water quality is excepted). We do not mean a detectable change that violates a numeric criterion for that parameter (e.g., 25 NTUs for turbidity), which criterion would otherwise apply if the Beneficial Use Limitations were not applicable (i.e., if the receiving waters were not designated as a scenic river or otherwise as “(a)” in Appendix A).35 The Beneficial Use Limitations/Anti-Degradation Policy are designed to provide additional protection beyond that conferred by the numeric limits of other water quality standards. Interpreting these regulations as merely prohibiting violations of otherwise applicable WQS would render them a nullity because, as we have seen, WQS may not be contravened in any waters, regardless of whether these additional regulations apply.
The ALJ, on remand, did not explicitly address the Anti-Degradation Policy but did construe the Beneficial Use Limitations. The 1985 version of the Beneficial Use Limitations, which the ALJ deemed applicable, provides: ” ‘All streams and bodies of water designated as (a) ... are protected by prohibition of any new point source discharge which increases pollutant loading or increased load from an existing point source.’ ” Decision on Remand, R., A-33, at 4. Construing this regulation in light of the OWQS definition of “pollution,”36 he concluded: “[T]he Oklahoma parties must show by substantial evidence that the City‘s discharge will create a nuisance or render the Illinois River in Oklahoma harmful, detremental [sic] or injurious to any beneficial use of the river.” Decision on Remand, R., A-33, at 5. The CJO upheld this interpretation with minimal discussion. Second Order on Petitions for Review, R., A-37, at 8. He excused the ALJ‘s failure to discuss the Anti-Degradation Policy by explaining that the ALJ “implicitly addressed the policy in his detailed analysis of the discharge‘s potential impact on all relevant water quality parameters.” Id. at 9; see id. at 10 (if ALJ erred in this regard, it was “harmless error“). The CJO reasoned that “if the Fayetteville discharge will not cause a detectible change in any of the relevant water quality parameters [as the ALJ found], it logically follows that there will not be a ‘quality degradation.’ ” Id. at 9-10.
We have considerable difficulty with the agency‘s treatment of these crucial Oklahoma regulations. First, and most importantly, the ALJ‘s interpretation defies the plain language of the Beneficial Use Limitations and the Anti-Degradation Policy that it references.37 Secondly, the CJO ruled that the ALJ erred in applying the 1985, rather that the 1982, OWQS. Second Order on Petitions for Review, R., A-37, at 5-6. The CJO deemed this error harmless, but we disagree.38 The 1985 version of the Beneficial Use Limitations, which the ALJ improperly applied, states: “All streams ... designated as (a) in Appendix A are protected by prohibition of any new point source discharge which increases pollutant loading....”
Finally, the agency‘s construction of the Beneficial Use Limitation is further flawed by the ALJ‘s imposition of the burden on Oklahoma to prove that the discharge would “create a nuisance” or “render the Illinois River ... harmful ... or injurious to any beneficial use.” Decision on Remand at 5. Granted, the opponent of a permit has the “burden of going forward to present an affirmative case at the conclusion of the Agency case on the challenged requirement.”
As for the Anti-Degradation Policy, the CJO concluded there could be no violation of the policy if there would be no detectable change in water quality. However, it is not clear whether the CJO interpreted the policy as requiring that there be no detectable change in water quality, or whether he was merely reporting the legal significance of the facts found by the ALJ. Although the CJO determined in his first order that the applicable legal standard is “whether [Fayetteville‘s] discharges under the permit will result in a detectable violation of the applicable water quality standards,” Order on Petitions for Review, R. A-28, at 2, 12-13, his subsequent affirmation of the ALJ‘s erroneous construction of the Beneficial Use Limitations casts doubt on whether he intended the “no detectable change” test to apply to violations of the Beneficial Use Limitations/Anti-Degradation Policy as well. Because of this ambiguity and the errors in interpreting the Beneficial Use Limitations, we agree with the Oklahoma parties that the agency incorrectly construed and applied both Oklahoma regulations.
2. Existing Degradation of Illinois Scenic River
Under other circumstances, the errors described above might necessitate remanding to the agency with instructions to apply Oklahoma law as we have construed it. However, given the facts in this record, even proper interpretation and application of Oklahoma water quality standards cannot save this permit. The record contains substantial evidence from which the ALJ could have found that the water quality of the Illinois scenic river has been degraded and that water quality standards were being violated prior to the onset of Fayetteville‘s discharge to the river (see subpart a. below). We believe that, where a proposed source would discharge effluents that would contribute to conditions currently constituting a violation of applicable water quality standards, such proposed source may not be permitted.39 The ALJ and the CJO erred in failing to consider whether or how the river‘s existing degraded condition is relevant to the decision whether to permit a new source discharge.40
Three factual subissues are essential to our determination that the Fayetteville discharge to the Illinois River may not be permitted: (1) whether the Illinois scenic river is already degraded (i.e., whether its quality has deteriorated since the river‘s designation in 1970); (2) whether Fayetteville‘s effluent will reach the scenic river; and (3) whether and how the components of Fayetteville‘s discharge would contribute to conditions in the Illinois River. Although it is difficult to summarize a record that consists of five boxes and four years of briefs, orders, transcripts, prepared testimony, correspondence, technical reports and miscellaneous other documents, in the following few pages we attempt to capsulize the evidence relevant to these three issues.
a. Evidence of existing degradation. First, we address the subject of the degradation of the Illinois scenic river‘s historically pristine water quality. Our review of the record before the ALJ revealed ample evidence from which the ALJ could have concluded that the river‘s condition has deteriorated since its designation as a scenic river and that water quality standards are being violated. Examples of this evidence follow.
Myron Knudson, Director of the Water Management Division, EPA-Region 6, testified at the administrative hearing: “There has [sic] been many conversations as related to what could be done in order to clean up the Illinois River.” Tr. at 221. The Attorney General of Oklahoma Robert Henry, in a prepared statement delivered at the hearing, described the Illinois River as “degraded,” Tr. at 232, and stated that “the river cannot handle the existing load” of municipal wastewater treatment plant discharges, Tr. at 233. Dr. Stephen Threlkeld, witness for the Oklahoma Wildlife Federation and author of the EPA-funded “Clean Lakes” report on Lake Frances,41 summarized the results of the “Clean Lakes” study, stating: “Water quality violations of the Oklahoma Water Quality Standards in Lake Frances ... are in terms of bacteria and in terms of turbidity....” Tr. at 356. He explained that EPA funded the study “because they wanted to know what the problems were in Lake Frances.” Tr. at 359; see id. at 374.
Mike Schornick, Oklahoma witness and principal of Schornick/Roberts & Associates, consulting engineers, testified that significant degradation trends are and have been occurring in the Illinois scenic river, including Lake Frances. Tr. at 398-400 (citing prefiled testimony, R., OK-2, at 3-4). He stated that certain figures in his prefiled testimony, which reflect data obtained from regular water quality monitoring conducted by Oklahoma at several points along the Illinois River, illustrate the degradation trends. Tr. at 414, 439. He claimed dissolved oxygen concentrations are reaching levels that violate OWQS. OK-2 at 4. He also stated that Arkansas and Oklahoma monitor phosphorus in their regular 305(b) trend analysis reports42 to EPA, Tr. at 486, and that all of those reports (1976-81 and 1984) show increasing phosphorus concentrations, Tr. at 489-90. Accounting for the addition of Fayetteville‘s effluent, Schornick said phosphorus loading and concentrations in Lake Frances will have increased by 106 percent and 76 percent, respectively, over 1974-75 background levels. Tr. at 454-56 (citing prefiled testimony, OK-2, at 4).
Lawrence Edmison, Director, Oklahoma Department of Pollution Control, testified that his department has received “many complaints about odor problems and color problems on the river.” Tr. at 542. He also discussed the algae problem on the river in relation to the increasing phosphorus concentrations and decreasing nitrogen concentrations in the water. Tr. at 533-34 (citing prefiled testimony, OK-4, at 3). Based on his years of personal observation of the river and experience handling citizen complaints and looking at trends documented in 305(b) reports and other reports, Tr. at 546-48, he stated, “I know how bad the river is now; I anticipate that any increased load will only make it worse.” Tr. at 548. He testified that the 305(b) reports for both 1984 and 1986 related an “apparent increasing trend” in phosphorus concentrations at all four Illinois River sampling sites, an “apparent decreasing trend” in dissolved oxygen at the same sites (with the exception of the Baron Fork site in 1986), and an “apparent decreasing trend” in nitrogen levels at all four sites (with the exception of Tahlequah in 1986). OK-4, at 2-3. Decreasing nitrogen and increasing phosphorus, he claimed, are “indicative of the algae problem on the river.” OK-4, at 2-3.
Oklahoma witness and consultant Dr. William Walker reported that algae concentrations in Lake Frances already reach 90 parts per billion, which is three times the level typically considered indicative of severe nuisance conditions. Tr. at 609-10. Lake Frances is already “supersaturated with nutrients,” Tr. at 691; for example, existing concentrations of phosphorus in Lake Frances are more than ten times levels considered typical of eutrophic lakes, or where algae problems start to develop, Tr. at 701. According to Dr. Walker, a “plume [of] degraded water” exists in the river downstream from Lake Frances. Tr. at 701.
Jimmie Pigg, part-time ichthyologist with the Oklahoma Water Quality Division and science coordinator for an Oklahoma school district, Tr. at 65, narrated a slide presentation at the administrative hearing showing changes in the condition of the Upper Illinois River since 1972. In response to an objection concerning the relevance of part of the testimony, counsel for the State of Oklahoma stated that the evidence was offered for the purposes of “show[ing] that the Illinois River is already in a degraded state and cannot assimilate any more effluent” and that “the Illinois River has already exceeded [its] assimilative capacity.” Tr. at 72, 76.43 Mr. Pigg said Lake Frances “is really a sewage lagoon,” which “catch[es] and hold[s] the material from Arkansas.” Tr. at 73.
The Gakstatter study reported that “dense phytoplankton populations develop in Lake Frances and also adversely affect water clarity in the Illinois River for several miles downstream,” and that this growth is “stimulated by excessively high phosphorus levels originating from [the sewage treatment plants at] Springdale and Rogers [in Arkansas].” Report at 5.44 The Gakstatter Report also provides brief summaries of the results of several other studies. For example, the U.S. Geological Survey (USGS) in 1984 reported the Illinois River did not meet water quality standards for dissolved oxygen, phosphorus, and fecal coliform bacteria; Threlkeld (1983) described Lake Frances as “very eutrophic” due to phosphorus from Springdale and Rogers; the Oklahoma State Water Quality Laboratory (1977) reported Lake Frances was in the “late stages of eutrophication,” due partially to “elevated Illinois River nutrients“; and two EPA (1977) studies classified as eutrophic both Lake Frances and, to a lesser extent, Tenkiller Reservoir. Gakstatter Report at 7-9.
The evidence before the ALJ also included the record of a hearing conducted by the Arkansas-Oklahoma Arkansas River Compact Commission on June 3-4, 1985. See R., C-1, Tr. at 307. The subject of the Compact Commission hearing was the Illinois River situation and the (at that time) proposed Fayetteville permit. The Commission issued an order (also included in the administrative hearing record) containing several findings concerning the degraded condition of the river. Findings of Fact, Conclusions of Law, and Commission Order (“Compact Commission Order“), R., OK-5. The Commission began by observing that, historically, the Illinois River “has been recognized by Oklahomans as a watercourse of unique natural scenic beauty and high quality ... spring-fed waters [that] ran clear and plentiful.” para. 16. But, the Commission continued, the “Upper Illinois River System in Oklahoma has, in recent years, undergone a process of degradation in water quality, and the process appears to be on an escalating trend.” para. 32.
Other findings by the Commission include: “[The] Illinois River has degraded substantially over the past decade ... [including] radical changes in the river‘s water color and turbidity, and the existence of increased alga growth [and] offensive odors....” para. 33. Violations of the dissolved oxygen standard have been documented by the Oklahoma Department of Health immediately below Lake Frances. para. 35. Degradation of dissolved oxygen is also occurring farther downstream from Lake Frances. para. 36. A 1984 USGS study (presumably the one cited in the Gakstatter Report) showed violations of Arkansas WQS in the Arkansas portion of the river. para. 38. Phosphorus concentrations are continuing to increase in “significant and undesirable amounts.” para. 40. The “Arkansas guideline for maximum phosphorus concentration to prevent eutrophication has already been greatly exceeded in certain Arkansas and Oklahoma river segments.” para. 40. The “trend of phosphorus degradation of the Upper Illinois River appears to be occurring at all locations.” para. 41. Lake Frances is in “an obvious state of eutrophication, marked by putrid smells and dark brown turbid waters.” para. 45. Based on Oklahoma‘s 305(b) report for 1978-83, “[d]egradation trends also appear to be occurring with reference to levels of potassium, calcium, sodium, copper and hardness.” para. 46. The Commission acknowledged Arkansas‘s “exceptions” to certain of Oklahoma‘s sampling methods and conclusions, paras. 47-48, but decided Oklahoma‘s methods were generally accepted in the scientific community and sanctioned by EPA, para. 49. The Compact Commission concluded: “[M]an-made pollution (degradation) of the waters of the Upper Illinois River in both Arkansas and Oklahoma ... as defined by the compact, [has] occurred and said pollution continues to occur at increasing and alarming rates.... Further, said pollution is of grave interstate magnitude and significance.” Conclusions of Law para. 9.
The record before the agency also included the Fayetteville 201 Facilities Plan, prepared by Fayetteville in conjunction with its application for an EPA construction grant for its new treatment plant. See
At this juncture we note that the absence of any evidence in the record that enforcement efforts have been undertaken to remedy the pollution problems in the Illinois River does not undermine our conclusion that water quality violations have occurred and no doubt continue to occur. Enforcement actions are not necessary to document water quality degradation; it is only necessary that there be reliable evidence that water quality criteria have been exceeded.45 See
Similarly, a history of lax enforcement with respect to existing sources does not justify allowing a new source of pollution. Water quality standards prescribe the desired condition of surface waters to be met at all applicable times; they do not serve merely as a yardstick for enforcement efforts when enforcement personnel may be available to ascertain compliance.
Clearly then, the record before the ALJ contains substantial evidence from which it can be concluded that water quality in the Upper Illinois River is degraded and that Oklahoma water quality standards for nutrients, dissolved oxygen, and/or aesthetics have been and probably continue to be violated. The decisions of EPA‘s judicial officers ignore the bulk of this evidence. To our consternation, however, the ALJ believed some of the relevant testimony chronicled above, yet remained oblivious to its ramifications. In his Decision on Remand, for example, the ALJ stated that “dissolved oxygen violations in Oklahoma are occurring without [Fayetteville‘s] discharge.” R., A-33, at 19 (emphasis added). It also appears he accepted the testimony that nutrients, turbidity, and solids standards were being violated, although he disputed the conclusion that Fayetteville “would increase the spatial and temporal ... frequencies” of those violations. Id. at 14-15 (citing Dr. Walker‘s and Dr. Gakstatter‘s testimony). Significantly, no witness refuted the testimony concerning the river‘s currently degraded condition, nor did the ALJ discredit (or even comment on) any of that testimony. He simply failed to recognize the significance of this testimony with respect to the permitting decision at hand.
b. Downstream transport of pollution from Fayetteville. Next, we address the question of the downstream migration of Fayetteville‘s effluent. Our review of the transcript revealed that no person involved in the administrative hearing seriously disputed that pollution from Fayetteville would reach the state line; instead, the parties debated how much would reach Oklahoma and what effect, if any, it would have. Indeed, in his final opinion, the ALJ recites evidence that twenty to twenty-five percent of the nutrients (specifically, phosphorus) in Fayetteville‘s effluent would be “bio-available” at the Oklahoma state line. Decision on Remand, R., A-33, at 8. The evidence supporting downstream transport includes:
According to EPA witness and employee Garrett Bondy, the Waste Load Allocation Study performed by Arkansas predicts a six-percent increase in the phosphorous load to Lake Frances due to Fayetteville. Tr. at 147. Fayetteville witness Dr. Cliff Thompson testified to a 2.4-percent increase, based on a discharge of 35 lb. phosphorus per day. Tr. at 280-81. (Fayetteville‘s permit allows it to discharge 54 lb. phosphorus daily.) Thompson, whose firm CH2M Hill prepared the 201 Facilities Plan for the Fayetteville plant, said: “We recognized that we would be adding discharge to the Illinois River.” Tr. at 266-67. Dr. Robert Blanz, former deputy director of the Arkansas Department of Pollution Control and Ecology, testified that Fayetteville‘s waste could reach Lake Frances, the Illinois River below the lake, and Tenkiller Reservoir. Tr. at 308-11, 321-22. He “guessed” twenty to twenty-five percent of the phosphorus from the plant would be cycled through the Upper Illinois River system rather than taken out of it. Tr. at 311-12; cf. Decision on Remand at 8.
Mike Schornick, testifying for Oklahoma, suggested that sixty percent of Fayetteville‘s phosphorus discharge would reach Lake Frances (based on his review of existing data). Tr. at 454-56 (citing prefiled testimony, R., OK-2, at 4). He stated that treatment plant operation would result in measurable changes in Lake Frances during low flow conditions. Tr. at 461-62. Another Oklahoma witness, Dr. Walker, testified that Fayetteville would increase the phosphorus load to Lake Frances by 4.69 percent in an average flow year. Tr. at 610-11 (referring to amended Table 5 in his prefiled testimony, R., OK-9).
The Arkansas River Compact Commission found that Lake Frances no longer acts as a “nutrient trap“; thus, the “addition of any new waste effluent discharge into the Illinois River from above Lake Frances ... will be transmitted downstream below Lake Frances into the Illinois River in Oklahoma.” Compact Commission Order, R., OK-5, Findings para. 44. The Commission further stated that “the potential for or threat of an increased phosphorus loading ... from [Fayetteville‘s] discharge to the Illinois River in Oklahoma clearly exists.” Id. para. 73.
Finally, the 201 Facilities Plan reports “considerable downstream transport of enriched organic matter” in the Upper Illinois River. R., ARK-6, at 2-22. Citing the Oklahoma State Department of Health‘s conclusion that “nearly 60 percent of the nitrogen and 74.4 percent of the phosphorus load measured in the Illinois River at Tahlequah, Oklahoma was [sic] contributed by sources above Lake Frances,” the report concludes the “data clearly show that point and nonpoint sources in Arkansas are a major source of nutrients in the Illinois River of Oklahoma.” Id. at 2-24. The report states that additional nutrients introduced to Lake Frances “may be passed through [the lake] to downstream reaches of the Illinois,” id., and that “transport of dissolved and suspended nutrients from Arkansas sections of the Illinois basin may have some effects on Tenkiller Reservoir in Oklahoma,” id. at 2-28. “Fayetteville‘s treated effluent,” the report claims, “would increase downstream nutrient concentrations by ... perhaps 10-15 percent ... during low-flow conditions.” Id. at 4-13.
Based on the foregoing, which is just a sample of the record evidence pertaining to the downstream transport of Fayetteville‘s effluent, we conclude there was substantial evidence before the ALJ to support a finding that Fayetteville‘s effluent would reach the Illinois scenic river.
c. Significance of Fayetteville effluent to existing conditions. Lastly, we recite some of the evidence relevant to the third important subissue--whether and how the components of Fayetteville‘s discharge can be expected to contribute to water quality conditions in the Illinois River. Although this is more a scientific question than it is a legal one, the inquiry helps to tie together the conclusions drawn from the first and second subissues discussed above. The evidence includes:
Mike Schornick asserted that Fayetteville‘s effluent will result in increased algae in Lake Frances. Tr. at 434-35 (citing prefiled testimony, OK-2, at 15). He discussed the relation between phosphorus and nutrient loading and dissolved oxygen levels, Tr. at 436, offering a brief explanation of how algae can increase and decrease the concentration of oxygen in water, Tr. at 438.47 He noted that increased algae can result in many aesthetic problems, including taste, odor, and the appearance of a river. Tr. at 477-78. He reported that the decreasing dissolved oxygen trends in the Upper Illinois have paralleled increases in phosphorus and other nutrients, calling this “strong evidence that there is a definite relationship between the two phenomena,” as well as a logical consequence of increased biological activity. OK-2, at 4. In his opinion, the Fayetteville discharge will “precipitate lower dissolved oxygen concentrations and more frequent violations of the dissolved oxygen standards.” Id. at 5. He also contends Fayetteville will cause violations of the copper standard. Id.
Robert Blanz, testifying for Arkansas, stated that “scouring” (the action of high stream flows moving sediment on the stream bottom and along its banks) resuspends sedimented material, including algae, thus increasing turbidity. Tr. at 322. EPA official Garrett Bondy testified that the Fayetteville discharge “may raise” sediment oxygen demand, thus potentially contributing to reduced dissolved oxygen concentrations in the river. Tr. at 129; see also id. at 133, 135. Dr. Threlkeld stated that “algal growth and resuspension of sediments are a part of turbidity [in Lake Frances].” Tr. at 356.
Dr. Walker agreed that one cause of water clarity problems in Lake Frances is algae. (The other is inorganic turbidity.) Tr. at 680-81 (citing Gakstatter Report; see id. at 78). He believes the increase in nitrogen pollution of the Illinois River attributable to Fayetteville‘s discharge might increase the amounts of periphyton (surface algae) in backwater areas and under low flow conditions. Tr. at 693. He further believes these impacts would become more significant as the phosphorus discharges from other upstream sources decrease. Tr. at 694, 716-18.48 In Dr. Walker‘s opinion, an increased growth of periphyton (i.e., more floating algal “mats and scums” on the river) would violate Oklahoma‘s aesthetics standard. Tr. at 705. He further asserted that Fayetteville‘s discharge would increase turbidity in Tenkiller Reservoir. Tr. at 711.
Lawrence Edmison, director of the Oklahoma Department of Pollution Control, testified that algae degrades the river and violates the aesthetics and coloration standards. OK-4, at 3. He related the algae problem in the river to increasing phosphorus and decreasing nitrogen concentrations in the water. Tr. at 533-34 (citing prefiled testimony, OK-4, at 3). The record also includes a memorandum to Lawrence Edmison from Quang Pham, an Oklahoma State Department of Health employee, which references a recent EPA study “on eutrophication of the Illinois River [that] indicated that phosphorus plays a major role in the stimulation of algae growth in the river.” Addendum to OK-4, at 2. Gakstatter and Katko concluded tentatively that “controlling algal growth in Lake Frances will result in a marked improvement in water clarity in the reservoir and in the Illinois River reach below the dam.” Gakstatter Report at 76. Finally, the 201 Facilities Plan reports that the “high productivity of the Illinois [River] waters causes considerable downstream transport of enriched organic matter” and the “net impact of point and nonpoint sources [such as Fayetteville] is to increase nutrient loading, with consequent increases in algal growth and increased turbidity.” ARK-6, at 2-22.
We conclude from the foregoing three-part review of the record that there is substantial evidence that degraded water quality conditions currently exist in the Illinois River in Oklahoma and that these conditions have been caused at least in part by pollutants that are constituents of Fayetteville‘s effluent. There is also substantial evidence that Fayetteville‘s effluent will be transported downstream to Oklahoma; thus, the plant can be expected to contribute to the ongoing deterioration of the scenic river and possibly Tenkiller Reservoir as well. It is our inescapable conclusion, given this evidence and the requisites of federal-Oklahoma state water pollution control laws, that the Fayetteville discharge to the Illinois River may not be permitted.49
IV. Discussion and Conclusions
As explained in part I. of this opinion (Standard of Review), we normally give considerable deference to an agency‘s interpretation of its obligations and authority under a statute it administers. Here, EPA‘s view that no discharge to a navigable water may be permitted unless it will comply with the federally approved standards of all affected downstream states is consistent with the statutory language and EPA‘s implementing regulations, supported by the legislative history, and reasonable on its face; therefore, it is entitled to substantial deference. See Chevron, 467 U.S. at 844-45, 104 S.Ct. at 2782-83. As we discussed in part III.A. supra, we adopt the agency‘s view on this question of statutory interpretation as our first holding in this case.
The balance of the agency‘s actions, however, do not warrant similar respect. In part III.B. we have identified several errors or deficiencies in EPA‘s interpretation of the applicable Oklahoma regulations, in the agency‘s factual findings, and in its application of the law to the relevant facts. We believe the most serious of these errors is the failure to attribute any significance to the existing WQS violations. In this section we discuss the errors on which we found our conclusion that the Fayetteville permit decision must be set aside as “arbitrary, capricious, ... or otherwise not in accordance with law.”
As a preliminary matter, EPA undermined our usual deference to its special expertise by the failure of its presiding officer to consider an important scientific principle, the oxygen-reducing effects of algae respiration and decay, and by his incomplete understanding of phosphorus assimilation.50 “EPA‘s failure to base its position on scientific or policy considerations ... [is] cause for reduced deference.” National Wildlife Fed‘n v. Gorsuch, 693 F.2d 156, 169 (D.C.Cir.1982). Similarly, a lack of thoroughness on the part of the agency warrants reduced deference. Id. at 166 (” ‘thoroughness ... of an agency‘s reasoning’ bears on the proper degree of deference” (quoting Federal Election Comm‘n v. Democratic Senatorial Campaign Comm., 454 U.S. 27, 37, 102 S.Ct. 38, 44, 70 L.Ed.2d 23 (1981)). In light of other errors in the agency‘s reasoning, however, we need not decide whether these flaws alone constitute reversible error.
EPA also misinterpreted and misapplied the Oklahoma nutrients standard and the Beneficial Use Limitations/Anti-Degradation Policy. In these respects the permit decision is flawed as a matter of law and must be set aside.
Furthermore, the agency‘s judicial officers believed expert testimony that nutrients in Fayetteville‘s discharge would be transported downstream to Oklahoma, but they inexplicably rejected or discounted testimony concerning the probable eutrophying effects of these nutrients. This error may have resulted in part from the officers’ faulty understanding of eutrophication processes and/or their erroneous interpretation of the nutrients standard. In any event, the net result is that the agency‘s decision to permit the Fayetteville discharge to the Illinois River “runs counter to the evidence before the agency” and lacks a “satisfactory explanation ... including a ‘rational connection between the facts found and the choice made.’ ” Motor Vehicle Mfrs. Ass‘n, 463 U.S. at 43, 103 S.Ct. at 2856 (citation omitted). As such, it is arbitrary and capricious and must be set aside. Id.
Finally, we hold that EPA‘s decision is arbitrary and capricious on one significant, additional ground. We believe that EPA, in failing to consider the significance of the evidence of ongoing WQS violations, has not only rendered a decision that “runs counter to the evidence,” but has “entirely failed to consider an important aspect of the problem.” Id. We consider this the principal flaw in the agency‘s decision-making rationale.
It cannot be doubted that ongoing violations of federally approved water quality standards constitute “an important aspect” of the decision whether to permit an additional source of pollution on a waterway.
Adherence to EPA‘s treatment of the facts and law of this case would fatally undermine the federal water pollution control strategy engineered by the
[The 1972] Amendments introduced two major changes.... First, the Amendments are aimed at achieving maximum “effluent limitations” on “point sources,” as well as achieving acceptable water quality standards....
Second, the Amendments establish the National Pollutant Discharge Elimination System (NPDES) as a means of achieving and enforcing the effluent limitations....
Water quality standards are retained [in the amended Act] as a supplementary basis for effluent limitations ... so that numerous point sources, despite individual compliance with effluent limitations, may be further regulated to prevent water quality from falling below acceptable levels....
Id. at 204-05 & n. 12, 96 S.Ct. at 2024-25 & n. 12 (emphasis added).
APPENDIX
All streams and bodies of water designated as (a) are protected by prohibition of any new point source discharge of wastes or increased load from an existing point source except under conditions described in Section 3.
All streams designated by the State as “scenic river areas,” and such tributaries of those streams as may be appropriate will be so designated. Best management practices for control of nonpoint source discharges should be initiated when feasible.
The intent of the Anti-degradation Policy is to protect all waters of the State from quality degradation. Existing instream water uses shall be maintained and protected. No further water quality degradation which would interfere with or become injurious to existing instream water uses shall be allowed. Oklahoma‘s waters constitute a valuable State resource and shall be protected, maintained and improved for the benefit of all the citizens.
It is recognized that certain waters of the State possess an existing water quality which exceeds those levels necessary to support propagation of fish, shellfish, wildlife, and recreation in and on the water. These high quality waters shall be maintained and protected unless the State decides, after full satisfaction of the intergovernmental coordination, and public participation provisions of the State‘s continuing planning process, to allow lower water quality as a result of necessary and justifiable economic or social development. Furthermore, where limited degradation is justified, the State shall require that any new point source of pollution or increased load from an existing point source, protect all existing and attainable beneficial uses through the highest statutory and regulatory requirements, and feasible management or regulatory programs pursuant to Section 208 of Public Law 92-500 as amended by PL 95-217 for nonpoint sources.
No degradation shall be allowed in high quality waters which constitute an outstanding resource or in waters of exceptional recreational or ecological significance. These include water bodies located in National and State parks, Wildlife Refuges, and those designated “Scenic Rivers” in Appendix A.
As the quality of Oklahoma waters improves, no degradation of such improved waters shall be allowed. When the yearly mean standard for a specific parameter decreases to the point where the goals listed in Appendix E become attainable, degradation will be prohibited by incorporating the goal as a standard.
In those cases where potential water quality impairment associated with a thermal discharge is involved, the anti-degradation policy and implementation method shall be consistent with Section 316 of Public Law 92-500 as amended by PL 95-217.
Notes
Except as expressly provided in this chapter, nothing in this chapter shall (1) preclude or deny the right of any State or political subdivision thereof ... to adopt or enforce (A) any standard or limitation respecting discharges of pollutants, or (B) any requirement respecting control or abatement of pollution; except that if an effluent limitation, or other limitation, effluent standard, prohibition, pretreatment standard, or standard of performance is in effect under this chapter, such State or political subdivision ... may not adopt or enforce [any less stringent rule]; or (2) be construed as impairing or in any manner affecting any right or jurisdiction of the States with respect to the waters (including boundary waters) of such States.
The Committee has also been concerned with the possibility that any State or subdivision thereof might be deterred from adopting or enforcing more stringent standards because of the fear that their industry might then be at a competitive disadvantage with other States.... The Committee bill, however, is based on the policy that any State or political subdivision which adopts more stringent effluent [limitations] ... than those required [under the CWA] should be allowed to do so.S. Rep. No. 414, reprinted in 1972 U.S. Code Cong. & Admin. News 3668, 3751.
[T]he sponsors of the Act successfully insisted on a zero-discharge-of-pollutants goal despite strong objection from both within and without.... Senator Muskie, the Senate sponsor and principal force behind the bill, stated, in the post-conference debate on the bill: “These [goals] are not merely the pious declarations that Congress so often makes in passing its laws; on the contrary, this is literally a life or death proposition for the nation.”
693 F.2d at 179 (quoting 118 Cong.Rec. 33,693 (1972)).