holding the EPA has the discretion to consider an untimely certification by a state agencySummary of this case from Weaver's Cove v. R.I. Dept
Heard August 2, 1993.
Decided October 21, 1993.
Robert Brager with whom Richard S. Davis, Joseph C. Stanko, Jr., Patricia Ross McCubbin, Beveridge Diamond, P.C., Washington, DC, Leonardo Andrade-Lugo, Jose A. Cepeda-Rodriguez, Carlos A. Rodriguez-Vidal, Eli Matos-Alicea, Goldman Antonetti Cordova Axtmayer, Hato Rey, PR, and Edward J. Ciechon, Jr., Philadelphia, PA, were on brief, for petitioner.
Alan D. Greenberg, Environment Natural Resources Div., Environmental Defense Section, Dept., of Justice, with whom Myles E. Flint, Acting Asst. Atty. Gen., Randolph L. Hill and Meyer Scolnick, Asst. Regional Counsel, E.P.A., Washington, DC, were on brief, for respondent.
Petition for review from the Environmental Protection Agency.
Before SELYA, CYR and BOUDIN, Circuit Judges.
In August 1990, the Environmental Protection Agency issued a pollution discharge permit to Puerto Rico Sun Oil Company ("the Company"). In doing so EPA complied with the substantive requirements of the governing statute and the procedures set forth in the statute and EPA regulations. Only the result gives cause for concern, and that concern is not allayed by the agency's explanation for its decision. In our judgment, the result is so odd that either the EPA has abused its discretion or it has explained itself so poorly as to require further justification. On either view, we must vacate the agency's order adopting the permit and remand for further proceedings.
I. THE FACTS
The Clean Water Act, 33 U.S.C. § 1251, et seq., prohibits the discharge into protected waters of any pollutant by any person, id. § 1311(a), unless a discharge permit has been secured from EPA. Id. § 1342. The permitting regime is a hybrid one in which both EPA and the counterpart state agency play a role. The precise role depends on whether EPA has delegated permit issuing authority to the state; but no such delegation is present here. Puerto Rico is treated as a state for purposes of the Clean Water Act, id. § 1362(3), and its local agency is the Environmental Quality Board ("EQB").
To obtain a permit, the applicant must satisfy a variety of substantive requirements under the Clean Water Act but, in addition, no EPA permit can issue unless the state in which the discharge will occur gives its own approval (called "certification") or waives its right to do so. 33 U.S.C. § 1341(a)(1). Further, the state certification may impose discharge limitations or requirements more stringent than federal law requires, and those more stringent obligations are incorporated into the federal permit as a matter of course. See generally United States v. Marathon Development Corp., 867 F.2d 96, 99 (1st Cir. 1989) (describing state role). What lies at the heart of this case is EQB's effort to impose, and then back away from, such more stringent obligations.
For some years before this case began, the Company held a discharge permit for its oil refining facility at Yabucoa Bay, Puerto Rico, where it discharges pollutants from two different sources. On May 27, 1988, the Company submitted to EPA an application to renew the permit for its facility. On October 31, 1988, EPA forwarded the application to EQB, requesting that a draft certification be prepared promptly. EPA also warned EQB that under EPA regulations, Puerto Rico's right to impose obligations by certification would be waived if a final certification were not received within 60 days after EPA sent a copy of a (yet to be prepared) draft permit to EQB. 40 CFR § 124.53(c)(3) (60 day time limit).
The Clean Water Act provides that the state waives its certification rights if it fails to issue or to deny a certification "within a reasonable period of time (which shall not exceed one year) after receipt of such request. . . ." 33 U.S.C. § 1341(a)(1).
On January 25, 1989, EQB released a tentative certification — essentially a draft document that facilitates public comment on the proposed state certification and proposed federal permit. The draft certification in this case probably came as a surprise to the Company. The earlier permit had employed a "mixing zone" analysis in setting the pollution limitations for the Company's discharged effluent; the draft certificate did not include a mixing zone analysis. The difference, which is central to this case, needs a word of explanation.
A discharge permit under the Clean Water Act may include several types of requirements. One set concerns the technology used to limit pollution; another, pertinent here, requires that the amount of specified pollutants not exceed certain percentage levels. In theory, the percentage levels could be measured in the effluent itself — such as storm runoff or waste water — just as it drains into the stream, river or bay which is protected by the Clean Water Act; alternatively, it could be measured at the edge of a defined area of the receiving body of water after the pollutant has been diluted by that water.
Such a defined area is called a mixing zone, and it appears that measuring pollutants at the edge of the mixing zone is widespread in the application of the Clean Water Act. According to an EPA publication, "[w]hether to establish such a mixing zone policy is a matter of State discretion." EPA, Mixing Zones — Water Quality Standards Criteria Summaries: A Compilation of State/Federal Criteria 2 (September 1988) ( "Mixing Zones"). Practically every state and Puerto Rico have adopted mixing zone criteria, id., although the criteria appear to differ widely. Id. at 70-78 (Puerto Rico criteria as of 1988). The mixing zone concept is described in Marathon Oil Co. v. EPA, 830 F.2d 1346, 1349 (5th Cir. 1987), which concludes with the observation that "the `mixing zone' determination is basically a cost-benefit judgment on a given set of environmental facts, rather than any sort of `scientific' determination." Id. at 1351.
When in January 1989 EQB issued its draft certification for the Company's requested permit, the EQB was reformulating its mixing zone criteria. EQB's draft certification for the Company neither continued in force the old mixing zone criteria temporarily nor made the certificate subject to the new criteria still under development. Instead, the draft certification simply set further pollutant limitations which, absent the mixing zone analysis, apply directly to the effluent as it enters the receiving waters. Mixing Zones, supra, at 2 ("If no such mixing zone is recognized by a State, then the waters must meet the criteria at the point of discharge.").
The next event was EPA's release on August 11, 1989, of a draft permit and request for public comment. The draft permit incorporated the requirements of the draft certification issued by EQB and therefore used no mixing zone analysis. Although issuance of the draft permit meant that final EQB certification was now due in 60 days, 40 CFR § 124.53(c)(2), EQB apparently paid no attention to the deadline or to EPA's earlier warning that failure to meet the deadline would waive Puerto Rico's right to certify. Nevertheless, in October 1989 EPA told the Company's attorneys that it was extending the comment period on the draft permit "indefinitely" while awaiting the EQB's final certification. When the certification arrived, said EPA, it would set a "prompt" close to the comment period.
On July 24, 1990, almost a year after receiving the draft permit, EQB issued what it called its "final" water quality certification for the Company, again eschewing a mixing zone analysis. Both the timing and substance of this action are puzzling because, only four days before, on July 20, 1990, EQB had promulgated new regulations to be effective on August 20, 1990, adopting a new method of determining mixing zones. But if EQB's behavior was slothful and careless, EPA's reaction was even stranger.
At this point the EQB's final certification must have appeared a probable candidate for administrative or judicial revision in Puerto Rico. EQB had used a mixing zone analysis in the past and was proposing to do so in the future, and the use of such an analysis was likely to be significant; indeed, the Company later represented, and EPA has not disputed the claim, that its refinery cannot operate if forced to meet the pollution standards without the help of a mixing zone analysis. Yet just as the Company moved to correct the EQB certification, EPA moved even more swiftly to adopt a final permit based on the EQB certificate that omitted a mixing zone analysis.
The chronology can be compressed. On August 17, 1990, the Company asked EQB to reconsider its certification and include a mixing zone analysis. On August 21, 1990, EPA published a new draft permit incorporating EQB's final certification requirements, and it offered 30 days to submit comments. On September 7, 1990, EQB wrote to EPA saying that it was evaluating the Company's comments on reconsideration and that it might alter its certification. On September 10 and on September 21, 1990, the Company asked EPA to delay action on the permit to allow the EQB to complete its reconsideration. On September 28, 1990, EPA issued a final permit, based on the then July 1990 EQB certification and without provision for a mixing zone.
On November 7, 1990, the Company sought administrative review within EPA, an action that automatically stayed the new permit, 40 CFR § 124.15(b)(2), and left the old one in force on a temporary basis. On November 28, 1990, EQB adopted a resolution staying its certification pending reconsideration and announcing, for the benefit of EPA, that the certificate was "not to become final" until the reconsideration was completed. In February 1991, EQB wrote formally to EPA stating that the certificate should be treated as not final and urging EPA to leave the Company's previous permit in effect for the time being. In June 1992 EPA's regional administrator issued a decision reaffirming the new permit without a mixing zone provision but continuing the stay of the new permit pending a further administrative appeal.
In July 1992, the Company duly appealed the regional administrator's decision to EPA's Environmental Appeals Board, urging a number of the arguments discussed below, and making one further contention of note: the Company said that unless EPA modified the permit on direct review, the Company would likely be unable get the mixing zone analysis incorporated into the permit through subsequent proceedings. The reason, said the Company, was "the probable application of the anti-backsliding policy" of the Clean Water Act, 33 U.S.C. § 1342( o). On October 26, 1992, the EPA Environmental Appeals Board issued a lengthy decision refusing further review. The Company's appeal to this court followed. 33 U.S.C. § 1369(b)(1)(F).
Faced with what may be a disastrous outcome from its standpoint, the Company has offered this court a variety of procedural challenges to EPA. They range from a broad claim that EQB's final certification was ineffective (because Puerto Rico's time to certify had expired) to a trivial complaint that the EPA did not allow a 15-day extension to the comment period at one phase of the proceeding. We think virtually all of the procedural claims fail and, while addressing them at the close of the opinion, we prefer to begin by discussing EPA's central error.
EPA's action in adopting the permit in this case is not flawed by procedural mistake. On the contrary, EPA did a commendable job of dotting i's and crossing t's. Nor is there any violation of substantive provisions of the Clean Water Act; for example, nothing in that statute explicitly requires EPA to use mixing zone analyses in its permits. The problem with EPA's decision is simply that the outcome appears on its face to make no sense. We say "appears" because we cannot rule out the possibility that some further explanation could shore up the EPA's result. Either way, the EPA's present action cannot stand.
It may come as a surprise that agency decisions must make sense to reviewing courts. Agencies, after all, are normally entitled to substantial deference so long as their decisions do not collide directly with substantive statutory commands and so long as procedural corners are squarely turned. This deference is especially marked in technical areas. But in the end an agency decision must also be rational — technically speaking, it must not be "arbitrary or capricious," Administrative Procedure Act, 5 U.S.C. § 706(2)(A) — and that requirement exists even in technical areas of regulation. E.g., Public Citizens Health Research Group v. Tyson, 796 F.2d 1479, 1505 (D.C. Cir. 1986). The requirement is not very hard to meet, but it has not been met here.
The "arbitrary or capricious" concept, needless to say, is not easy to encapsulate in a single list of rubrics because it embraces a myriad of possible faults and depends heavily upon the circumstances of the case. Still, there are rules of thumb, e.g., Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983) (listing examples). In addressing individual aspects of EPA's decision, we cite to those requirements — discussion of relevant issues, consistency with past practice, avoidance of unexplained discrimination — that are pertinent to EPA's decision in this case.
On the surface of the administrative record, the following scene presents itself. EQB, having used a mixing zone analysis in past cases, neglected to include such a provision in its latest certification for this facility. EQB had previously used a mixing zone analysis for this very facility; and far from abandoning the concept, EQB was in the process of revising its regulations to prescribe such an analysis at the very time it was preparing the Company's certification. Four days before it issued the final certification in this case, omitting a mixing zone provision, it formally promulgated its new mixing zone regulations.
It is not clear whether in August 1990 EPA appreciated that EQB had probably misstepped. The Company's brief implies that the EPA, having obtained EQB's final certification, then proceeded with sinister speed — surely a rare accusation in administrative law — to mousetrap the Company by issuing a final permit before EQB's certification could be revised. An alternative explanation, to us entirely plausible, is that the EPA's patience with EQB had been exhausted and it wanted, as it had warned almost a year before, simply to get done with the permit as soon as it had EQB's final certification.
However this may be, both the Company and EQB made clear to the EPA at once, and before the final permit issued, that reconsideration was under way. EPA published its new draft permit for comment in August 1990; and in September 1990, before the EPA issued the final permit on September 28, 1990, EQB advised EPA (on September 7) that it was reconsidering its certification and might alter it, and the Company wrote letters (on September 10 and 21) begging the EPA to defer final action until the EQB acted. The EPA nevertheless proceeded to issue the final permit with no explanation for its refusal to wait.
Even at this stage, it appears that EPA was free to correct the problem on administrative review. There being no fixed timetable, the regional administrator presumably had discretion to defer action until EQB acted on the Company's reconsideration request and, if a mixing zone analysis were adopted by EQB in a revised certification, then to incorporate this revision into the new permit. One of EPA's regulations, 40 CFR § 124.55(b), which is discussed below, seems to contemplate just such a situation. During this same period EQB made crystal clear, by its resolution of November 28, 1990, and its formal letter of February 25, 1991, that it was planning to reexamine its certification and did not want the certification treated as final. Once again, EPA proceeded to reject the pleas and reaffirm the permit, sans mixing zone.
EPA has now explained its position at least three times administratively and for a fourth time in this court. Each time EPA deals deftly with the Company's procedural objections by showing why some regulation allowed EPA to await EQB's final certification, but to refuse to await EQB's attempt to repair the certification, and allowed EPA to adopt EQB's certification, but to reject EQB's retroactive attempt to brand it as non-final. The only thing that is missing, among this array of finely wrought explanations, is any reason why the EPA should want to frustrate the EQB's clumsy, long-delayed but increasingly evident desire to reconsider a mixing zone analysis for this permit.
Assuredly, some explanation is called for. The mixing zone analysis is not some freakish idea or whim of the Puerto Rico authorities. According to EPA's Mixing Zones publication, it is available for use in at least 49 states in varying situations; and the Company said that the refinery in question cannot operate if the permit limitations are applied, without a mixing zone analysis, at the point that the effluent enters the water. Patently, these considerations of history and practical effect would, in a rational decision, warrant at least some discussion. Motor Vehicle Mfrs. Ass'n, 463 U.S. at 43, 103 S.Ct. at 2867 (agency may not "entirely fail to consider an important aspect of the problem").
At oral argument, we inquired of counsel representing the EPA whether there were other situations in which EPA had refused to use a mixing zone analysis despite a state's desire that such an analysis be used. Yes, we were told, counsel for EPA knew of several such instances. On rebuttal, the Company's counsel responded that there were indeed other instances but they were limited to EPA's issuance of permits in Puerto Rico, in the same time frame as this case and to other applicants whose situations paralleled that of the Company. If this is the situation (counsel for EPA made no later effort to respond), then EPA's current posture is in some measure at odds with precedent. Cf. Atchison, T S.F. R.R. v. Wichita Bd. of Trade, 412 U.S. 800, 808, 93 S.Ct. 2367, 2375, 37 L.Ed.2d 350 (1973) ("departure from prior norms" must be explained).
The point is not that EPA has some overriding obligation under the Clean Water Act to do whatever it is that the state wants to do. On the contrary, EPA was entirely free, once Puerto Rico had ignored the clear deadlines for a final certification, to treat the Commonwealth as an interested bystander with no further veto authority. What is beyond explanation, or at least wholly unexplained, is why EPA should be intent on adopting half of what the Commonwealth wanted while systematically frustrating its attempt to secure the other half. The obligation, we repeat, is not one of deference to local authorities but of making sense.
There is also in this case an element of apparent irrational discrimination. See, e.g., Green Country Mobilephone, Inc. v. FCC, 765 F.2d 235 (D.C. Cir. 1985) (obligation to treat similar cases similarly). For all that appears, similarly situated facilities in Puerto Rico, if permitted for the first time next year, are likely to receive permits including a mixing zone analysis. Like facilities in other states, permitted in September 1990 at the same time as the Company, probably received the benefit of mixing zone analyses. Only Puerto Rican facilities that happened to be permitted or re-permitted in this strange "window," during which EQB was reformulating its mixing zone criteria, are left out in the cold — possibly forever if the anti-backsliding provisions apply.
Needless to say, we do not know whether the anti-backsliding provisions would produce this result. The provisions are complicated and contain certain exceptions. 33 U.S.C. § 1342( o). The Company's prediction is qualified, and EPA's brief is silent on this issue.
Perhaps there is some explanation for EPA's action other than a mechanical desire to reach a rapid conclusion without regard to whether the result is sound. Indeed, we suspect that there is an explanation. As noted, the Company insinuates that EPA deliberately took advantage of EQB's carelessness to mousetrap the Company into standards that could not later be relaxed because of the anti-backsliding provisions previously mentioned. Such a result would at least explain what happened, although it is doubtful that the explanation, if adopted by EPA, would commend itself to a reviewing court.
Or, there may be more benign reasons for EPA's action. Perhaps the Company's science is faulty and very slight adjustments in technology would permit it to meet the pollution limitations, and improve the environment to boot, without any mixing zone analysis. In all events, until EPA emerges from its fortress of procedural-rule citations and adopts a rationale for its action, any speculations are beside the point: the agency's decision cannot be supported on reasoning that the agency has not yet adopted. See SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 1577, 91 L.Ed. 1995 (1947).
We turn now to the Company's other arguments on appeal because some of them, if adopted, would alter the terms of the remand. The main thrust of the Company's various arguments is that, for various procedural reasons, EPA was not entitled to rely on the EQB certification. On this premise, the Company argues that EPA was required to formulate its own permit standards based upon the real requirements of Puerto Rico law, which the Company believes requires the use of a mixing zone analysis. Since we reject the Company's premise of procedural error, the further steps in the Company's argument need not concern us.
The Company's broadest procedural argument is that Puerto Rico's final certification came too late and therefore could not furnish the basis for EPA's own final permit. As already noted, the Clean Water Act required Puerto Rico to provide its certificate, or announce a decision not to certify, within a reasonable time not to exceed one year after the application, 33 U.S.C. § 1341(a)(1); and by regulation EPA required a certification decision within 60 days of the issuance of a draft permit. 40 CFR § 124.53(c)(3). Here, EQB apparently ignored both time limits, failing both to meet the statutory one-year deadline and the regulation-based 60-day deadline.
Under the statute and the regulation, the price of failing to meet the deadlines is that the state agency waives its right to dictate permit terms that go beyond what EPA would do on its own. Based on this waiver language, the Company argues that a state certification issued after the deadline is without legal effect. In reply EPA says it is free either to declare a waiver or, instead, to follow the course taken in this case and await the final, though belated, certification. The statute itself merely provides that the state must act within a reasonable period, not to exceed a year, or the certification requirement will be deemed "waived." 33 U.S.C. § 1341(a)(1).
Although we are provided no useful precedent or legislative history, our reading of the statute largely coincides with that of EPA. The statutory time limit and the word "waived" do not tell us the answer; Congress could have meant that a state certification issued after the deadline had to be ignored by EPA, or it could have meant only that EPA was free to do so. EPA interprets the statute to mean the latter and under the Chevron doctrine, Chevron U.S.A., Inc. v. NRDC, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), its view is entitled to weight. State of California v. FERC, 966 F.2d 1541 (9th Cir. 1992), cited by the Company as holding that the deadline cannot be waived, holds no such thing.
EPA's interpretation of its own 60-day regulation is even more compelling since it wrote the regulation. Gardebring v. Jenkins, 485 U.S. 415, 430, 108 S.Ct. 1306, 1314, 99 L.Ed.2d 515 (1988). In addition, agencies can usually (although not always) waive their own procedural regulations even where there is no express provision for waiver. American Farm Lines v. Black Ball Freight Service, 397 U.S. 532, 538, 90 S.Ct. 1288, 1292, 25 L.Ed.2d 547 (1970).
Further, EPA's reading both of the statute and its regulation seems to us a sensibly flexible one. EPA's reading gives it the practical benefit of the state process even if that benefit comes a little late. Indeed, where no one complains ( e.g., because the applicant is happy to operate under an earlier permit), it could be pointlessly rigid to insist that EPA begin its own calculations the moment the certification deadline expires for the state. The concern on the other side is that without a deadline, a new applicant could be left dangling forever. But we think the courts have adequate power to assure that flexibility does not become an excuse for permanent inaction.
See Administrative Procedure Act, 5 U.S.C. § 706(1) (power to compel agency action unduly delayed). The courts are normally deferential to the agency in such cases. See, e.g., Telecommunications Research Action Center v. FCC, 750 F.2d 70 (D.C. Cir. 1984). Here, however, Congress has expressed its intent that the state proceeding be completed in a year. If EPA wants to waive the state's failure to meet a deadline, and wait longer for its certification, we think that the propriety of its deferral might be open to judicial review that is somewhat more searching than customary.
The Company's remaining arguments require less discussion. The claim that EQB's certification was not final when EPA adopted it is unpersuasive. "Finality" is a concept with several shades of meaning in administrative law; but where, as here, the agency itself (rather than a subordinate body) has spoken and has explicitly labeled its action "final," we think that is enough, even though the agency may choose to reconsider or may be reversed on judicial review. The Company failed to get a stay of the EQB certification before EPA acted in reliance upon it. We agree with EPA that the subsequent decision of EQB to re-characterize its certification order as non-final cannot affect the procedural validity of EPA's decision to grant the permit.
In fact, EPA has regulations that govern the effect of a state stay or modification of a certification after a permit has issued. The pertinent regulation permits EPA's regional administrator under certain circumstances to incorporate the modifications into the permit so long as the state agency stays or modifies the old certificate and forwards a new one to EPA as a substitute. 40 CFR § 124.55(b). But this regulation does not apply in this case because EQB never sent a substitute certificate to the EPA.
The Company relies upon a different EPA regulation, 40 CFR § 122.44(d)(3). This provides in part that if a state court or board stays a certification, EPA shall notify the state that certification will be deemed waived unless a finally effective certificate is issued within 60 days; absent such a new certification, the regulation says that EPA shall impose its own requirements in the permit. In agreement with EPA, we read this regulation to apply only to stays that occur before EPA has issued its own permit. Once again, the agency's reading of its own regulation is entitled to deference. Gardebring, 485 U.S. at 430, 108 S.Ct. at 1314. Its reading also has the benefit of making this regulation, governing pre-permit stays, dovetail with section 124.55(b), governing post-permit stays.
In an attempt to bolster the importance of the EQB stay, the Company reminds us of the central role that the states were intended to play under the Clean Water Act. Yet that role is to be played within the framework of the procedures fixed by the statute and EPA regulations. Indeed, precisely because two different jurisdictions are expected to collaborate on a permit, there is a special need for compliance with the rules of the road. Here, the EQB stay came after the permit and — strictly from a procedural standpoint — EPA was entitled to disregard it, unless and until EPA's regulation governing a post-permit stay was satisfied.
In summing up, we stress again that the flaw in EPA's action is not a procedural defect. EPA's result is irrational, or at least inadequately explained, not because of EQB's hapless stay, but because of the substance of the EPA's permitting decision. To restate the gist of the matter, EPA has failed to explain why it makes sense, as a matter of substantive policy, to frustrate Puerto Rico's incipient desire to use the mixing zone analysis, and why those companies who fall in this "window" between Puerto Rico's old and new regulations should alone be denied the benefits of a mixing analysis. Those concerns would be virtually the same even if EQB had never used the word "stay."
We have not discussed the Company's separate claim that EPA abused its discretion by not extending the comment period for 15 days, as requested by the Company, to permit more time for comment on technical issues. This argument, summarily stated in a paragraph at the end of the Company's brief, is not seriously supported and is therefore not preserved for review. United States v. Zannino, 895 F.2d 1, 27 (1st Cir.), cert. denied, 494 U.S. 1082, 110 S.Ct. 1814, 108 L.Ed.2d 944 (1990).
In framing the remand, we begin by emphasizing what we have not decided. Whether the final certification issued by the EQB in August 1989 is vulnerable to attack under Puerto Rican law, if not altered by EQB on reconsideration, is an issue not before this court. Although state certification provisions are incorporated into federal permits, review of a state certification is a matter for local courts. Roosevelt Campobello Int'l Park Comm'n v. EPA, 684 F.2d 1041 (1st Cir. 1982). The apparent past and future inclination of EQB to employ mixing zone analyses is part of the background of this case, but nothing we have said should be taken to declare the law of Puerto Rico on this subject.
Similarly, we do not suggest that mixing zone analysis has a sacrosanct role under the Clean Water Act. Our impression from EPA's own publication is that the use of such analysis is widespread. But that impression is subject to correction. In any event, sound reasons may dictate that a mixing zone analysis not be used in certain cases or certain classes of cases, despite a possible hint to the contrary in Marathon Oil Co., 830 F.2d at 1349 ("By definition, the effluent itself does not meet water quality standards; otherwise, it would not be considered polluted."). There may even be reasons why, apart from EQB's procedural default, a mixing zone analysis is improper in this case.
All that we hold here is that EPA's decision to issue a permit in September 1990, adopting EQB's certification but refusing to await EQB's decision on reconsideration, produces a result that on the present record appears manifestly arbitrary and capricious. If legitimate reasons exist for such an outcome, then EPA is free to provide them and re-adopt the present permit (and the Company in turn is free to challenge those reasons and that action by petitioning again for judicial review). EPA, EQB, and the Company may find it possible to chart a more constructive course and make further litigation unnecessary.
The EPA order adopting the permit at issue in this case is vacated and the matter is remanded to EPA for further proceedings in accordance with this opinion. Costs are taxed in favor of the petitioner.