Montpelier WWTF Discharge Permit

Opinion

STATE OF VERMONT ENVIRONMENTAL COURT } In re Montpelier WWTF Discharge Permit } Docket No. 22-2-08 Vtec }

Decision on Cross-Motions for Summary Judgment This appeal concerns a renewal permit (#3-1207) for the Montpelier Wastewater Treatment Facility (“Montpelier WWTF” or “Facility”) that was granted to the City of Montpelier (“City”) by the Vermont Agency of Natural Resources (“ANR”) on January 2, 2008. Appellant Conservation Law Foundation (“CLF”) challenges that permit as allowing discharges of phosphorus that exceed what is permissible under applicable federal and state laws and regulations. ANR and the Water Resources Panel of the Vermont Natural Resources Board (“Water Panel”), supported by the City, have asked for summary judgment upholding the permit as issued.1 CLF opposes these motions and requests summary judgment in its favor. In particular, CLF requests that this Court exercise its de novo review of the permit to impose a more stringent phosphorus limitation. We are therefore presented with cross-motions for summary judgment.2

Factual Background For the sole purpose of putting the pending motions in context, we recite the following facts, which we understand to be undisputed unless otherwise noted: 1. Lake Champlain (“the Lake”) is a large, freshwater lake bordered by Vermont, New York, and the Province of Quebec, Canada. 2. Lake Champlain has been identified as an impaired waterway due to the existence of high phosphorus levels that cause the Lake to exceed minimum water quality standards established by the State.3 As a result, section 303(d) of the Clean Water Act (“CWA”) requires states bordering

1 Although ANR’s initial motion in this matter was entitled a motion to dismiss, the Court gave notice to the parties at a pretrial conference on November 13, 2008, that the Court was converting this motion to a motion for summary judgment, pursuant to V.R.C.P 12(b). See Amended Scheduling Order (Nov. 14, 2008) (memorializing this conversion). 2 CLF is represented in these appeals by Anthony N.L. Iarrapino, Esq., and John L. Davenport, Esq.; ANR was initially represented by Aaron Adler, Esq., and is now represented by Warren T. Coleman, Esq., and Michael Steeves, Esq.; the Water Panel is represented by Mark L. Lucas, Esq.; and the City is represented by Steven F. Stitzel, Esq. 3 Phosphorus has been identified as contributing to the algae blooms that have been an unfortunate annual occurrence in Lake Champlain during the summer months.

1 Lake Champlain to develop a target maximum for additional pollutants delivered to Lake Champlain. See 33 U.S.C. § 1313(d). The terminology used in describing this target maximum of delivered pollutants is recognized in the regulatory and development industries as “total maximum daily load” (“TMDL”). See id. § 1313(d)(1)(C). 3. On September 25, 2002, ANR and the New York Department of Environmental Conservation issued the Lake Champlain Phosphorus TMDL (“Champlain TMDL”). This TMDL established a maximum level of phosphorus that can be delivered to Lake Champlain from all sources, for the stated purpose of eventually returning Lake Champlain to being a non- impaired waterway. 4. The Champlain TMDL divided the Lake into 13 segments and set annual loading limits of phosphorus for each segment. The only segment at issue in this appeal is the Main Lake Segment. 5. The Montpelier WWTF discharges into the Winooski River, which then travels about forty miles in a northwesterly direction before flowing into the Main Lake Segment of Lake Champlain. 6. The Champlain TMDL lists a total TMDL phosphorus loading capacity for the Main Lake Segment of 110.3 metric tons of phosphorus per year (“mt/yr”). See Champlain TMDL at 15 tbl.3 (allocating 76.6 mt/yr to Vermont and 33.7 mt/yr to New York). This total phosphorus loading capacity includes all point and nonpoint discharges from both Vermont and New York.4 7. According to the latest estimate, the Main Lake Segment currently receives phosphorus loads totaling 217.9 metric tons per year. See 2008 State of the Lake & Ecosystem Indicators, attached as CLF Ex. G, at 5 fig.4 (listing loads of 148.4 mt/yr from Vermont and 69.5 mt/yr from New York). 8. Under the previous ANR-issued discharge permit for the Montpelier WWTF, the Facility’s phosphorus discharges were limited to 4.388 metric tons per year. 9. The Champlain TMDL established a maximum total phosphorus discharge, referred to as a wasteload allocation (“WLA”), specifically from the Montpelier WWTF, of 3.290 metric tons per year, which converts to 7,253 pounds per year.

4 The Clean Water Act defines a “point source” as “any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged.” 33 U.S.C. § 1362(14). Any other source of pollution—such as general agricultural runoff—is a nonpoint source.

2 10. From 2002 through 2007, the Montpelier WWTF has never actually discharged more than 3,192 pounds of phosphorus in any given year. 11. On January 2, 2008, when ANR issued the Montpelier WWTF permit that is the subject of this appeal, ANR adopted—without further calculation—the allowance of 7,253 pounds per year of phosphorus discharge. ANR also listed a discharge limitation of a monthly average of 0.8 milligrams of phosphorus per liter of water discharged, since establishing such a limitation is now required under Vermont law. See 10 V.S.A. § 1266a(a).

Discussion The parties have presented the Court with cross-motions for summary judgment. CLF contends that the phosphorus limit in the permit is not stringent enough to meet the requirements of the federal Clean Water Act, the Vermont Water Pollution Control Act, and their respective implementing regulations. ANR, the Water Panel, and the City disagree; they support the permit as issued and ask this Court to summarily uphold the issued discharge permit. We may grant summary judgment only when “the pleadings, depositions, [and] answers to interrogatories, . . . together with the affidavits, if any, show that there is no genuine issue as to any material fact and that any party is entitled to judgment as a matter of law.” V.R.C.P. 56(c)(3). Generally, the burden of proof is on the party requesting summary judgment. Chapman v. Sparta, 167 Vt. 157, 159 (1997). When presented with cross-motions for summary judgment, we must consider each motion in turn and afford all reasonable doubts and inferences to the party opposing the particular motion under consideration. DeBartolo v. Underwriters at Lloyd’s of London, 2007 VT 31, ¶ 8, 181 Vt. 609 (citing Toys, Inc. v. F.M. Burlington Co., 155 Vt. 44, 48 (1990)). The parties in this appeal have worked together to narrow the issues on appeal, and each party has also produced detailed, thoroughly researched briefs and accompanying statements of undisputed material facts. These efforts have made it clear that all of the parties agree upon the facts that are material to resolving this dispute. The Court is therefore presented with a purely legal question that is appropriate for summary judgment. See V.R.C.P. 56(c)(3). The question before us is one of statutory interpretation. In particular, we must evaluate whether it is a violation of the Clean Water Act, the Vermont Water Pollution Control Act, or their respective implementing regulations, to base a discharge permit determination solely upon the applicable TMDL wasteload allocation, particularly when that TMDL is more than five years old. The process followed in the application proceeding now under appeal—a process that is

3 endorsed by ANR, the Water Panel, and the City (as permittee)—derives a water quality based effluent limitation (“WQBEL”) for the subject Facility, solely with reference to the Champlain TMDL and without further analysis of the historical discharges for the Facility, improvements that may be had through advances in technology, offset opportunities within the applicable watershed, the levels of phosphorus currently entering the watershed from other sources, or whether this particular discharge will cause or contribute to a violation of water quality standards.5 We conclude that federal laws and regulations require a more reasoned site-specific and time-specific analysis before a permit to discharge pollutants from the Facility is granted.6 We therefore conclude that the discharge permit issued by ANR must be vacated and the pending application remanded to ANR for that more specific analysis.7 The 2002 Champlain TMDL established a wasteload allocation for the Montpelier WWTF that set a maximum allocation of phosphorus discharges at 3.290 metric tons per year, also measured as 7,253 pounds per year. When ANR issued the Montpelier WWTF permit more than five years later (in 2008), ANR adopted—without further calculation—the limit of 7,253 pounds per year as the current permitted phosphorus discharge, thereby allowing the Montpelier WWTF to discharge up to 7,253 pounds of phosphorus per year8 from its outlets into the Winooski River. Thus, ANR concluded, by reference to the phosphorus discharge maximums established in 2002, that the Facility was entitled to a WQBEL equal to the 2002 TMDL maximum.9

5 We note at the outset that although the parties debate what level of phosphorus discharge “causes, has the reasonable potential to cause, or contributes to” a violation of water quality standards, 40 C.F.R. § 122.44(d)(1)(iii), it appears to be uncontested that some WQBEL is needed for phosphorus discharges from the Montpelier WWTF. 6 Our conclusion on the federal law and regulations at issue here makes it unnecessary for us to address CLF’s arguments regarding alleged violations of the Vermont Water Pollution Control Act and its implementing regulations. 7 Because we are remanding this appeal to ANR to begin and complete further analysis, we do not reach CLF’s arguments regarding the claim that a net zero discharge permit limitation is required as a matter of law. 8 This figure is more than twice the maximum phosphorus discharge actually experienced by the Facility in recent years (3,192 pounds per year). 9 ANR also listed a limitation of a monthly average of 0.8 milligrams per liter, as required under Vermont law. See 10 V.S.A. § 1266a(a). As is clear from the units used in this permit limitation—milligrams per liter—this restriction is calculated based upon the total volume of water discharged in any given month. Thus, as a practical matter, if the Montpelier WWTF experiences low discharge flows, the 0.8 milligrams per liter restriction will be the only restriction that matters because it will prevent the Facility from discharging anywhere near its 7,253 pound annual limit. If, on the other hand, the Facility experiences high enough discharge flows, it would run the risk of exceeding the 7,253 pound annual limit (and at maximum flows could actually discharge nearly 10,000 pounds of phosphorus), even while remaining consistently under 0.8 milligrams per every liter discharged. (We have come to this conclusion with the use of a mathematical calculation: The daily limitation of 3.97 million gallons converts to 15,028,085 liters per day. Applying the 0.8 milligrams per liter limitation to maximum daily flows results in a

4 According to ANR, the Water Panel, and the City, any permit limitation that is pulled directly from the TMDL is in all cases consistent with the assumptions and requirements of that wasteload allocation and therefore meets the requirements of 40 C.F.R. § 122.44(d)(1)(vii)(B). ANR, the Water Panel, and the City also assert that this type of process does not violate the five- year limitation on NPDES permits, see 33 U.S.C. § 1342(b)(1)(B), even in a circumstance where, such as here, the TMDL relied upon in issuing the discharge permit is more than five years old. We disagree on both points. Our Decision today has three components to it. First, we find that this is not an instance where there has been a specific determination from either ANR or EPA to which we can defer. Second, we conclude that the process of automatically adopting—without further analysis—a WQBEL directly from a maximum wasteload allocation in a more-than-five-year-old TMDL violates the five-year limitation on NPDES permits. See 33 U.S.C. § 1342(b)(1)(B). Third, we conclude that the Clean Water Act and its implementing regulations require specific analysis in each permit proceeding to determine whether a WQBEL that is derived from a TMDL is “consistent with the assumptions and requirements of any available wasteload allocation.” 40 C.F.R. § 122.44(d)(1)(vii)(B). Each of the latter two conclusions provides an independent basis for our ruling in this matter.

I. Deference to ANR or EPA As a threshold issue to our interpretation and analysis, we must first determine whether to grant what is known as Chevron deference to either ANR or EPA in the interpretation of the federal regulations at issue here. See generally Chevron, U.S.A. v. Natural Res. Def. Council, 467 U.S. 837, 842–43 (1984).

calculation that up to 12,022,468 milligrams—roughly 26.5 pounds—of phosphorus could be discharged daily, which in turn converts to over 9,672 pounds annually.) In other words, if flows into and out of the Facility are consistently high, the 0.8 milligrams per liter limitation on its own may allow for the annual limit to be violated. We therefore decline ANR’s invitation to evaluate this permit under the assumption that the 0.8 milligrams per liter limitation affords a more stringent limitation. We also recognize that some of these numbers may be inflated, as evidenced by affidavits that both ANR and the City have provided, which state that the historical—and predicted—discharges at the Facility are significantly lower than the amounts allowed under either the 3.97 million gallons per day flow limitation or the 0.8 milligrams per liter phosphorus limitation. However, our focus in this proceeding must be governed by the permit for which the City applied. If ANR and the City believed that more stringent limitations could be met, and wanted the permit to be evaluated it that light, it was incumbent upon those parties to write such limitations into the pending application. We decline to follow the suggestion, assuming such is being made, that we hold that an overly permissive limitation written into a permit is acceptable because the Facility would in fact meet more stringent limitations, due to its experience of lower flows than what is permitted. Until more stringent limitations are considered and adopted, we assume that the Facility wishes to receive a permit that authorizes the maximum amounts of pollutant discharges currently listed in the permit.

5 We conclude that there are two independent reasons for not deferring to ANR’s interpretation of the federal Clean Water Act or its implementing regulations. First, our Supreme Court has explicitly stated that courts should not defer to ANR—a state agency—in the interpretation of the federal Clean Water Act. In re Stormwater NPDES Petition, 2006 VT 91, ¶ 13 n.2, 180 Vt. 261; see also Jacobus v. Dep’t of PATH, 2004 VT 70, ¶ 23, 177 Vt. 496 (“The Court does not defer . . . to [a state agency’s] interpretation of federal law and regulations.”). Second, any deference here to the agency that made the decision on appeal would run astray of this Court’s statutory mandate to conduct a de novo hearing in this matter. See 10 V.S.A. § 8504(h) (requiring “a de novo hearing on those issues which have been appealed”); V.R.E.C.P. 5(g) (“All appeals under this rule shall be by trial de novo . . . .”); see also In re Stormwater NPDES Petition, No. 14-1-07 Vtec, slip op. at 9 n.2 (Vt. Envtl. Ct. Feb. 18, 2009) (Durkin, J.) (“ANR . . . asserts that this Court failed to show its determination on the pending CLF petition the proper deference, given ANR’s undisputed expertise in the field of stormwater assessment. ANR’s argument here confuses the proper deference shown to its determinations in separate permit proceedings with the necessary de novo review that must be conducted when ANR’s own determination has been challenged in an appeal.”). For these reasons, we cannot defer to ANR’s interpretation of the federal statutes and regulations at issue here. We also conclude that this is not a situation where deference to EPA’s interpretation of these statutes and regulations is appropriate, given that EPA has not yet spoken on the specific legal issues that have arisen in this appeal. EPA is not before the Court in this appeal, and we can find no instances where EPA has adopted an official regulation, ruling, policy, or any other position regarding the legal issues raised in this appeal. Although CLF claims that EPA interprets 40 C.F.R. § 122.44(d)(1)(vii)(B) as requiring an individual analysis in each permit proceeding to determine whether it is necessary to impose more stringent controls than those required under the applicable TMDL wasteload allocation, we cannot find such an interpretation by EPA. The closest that EPA has come to a general determination similar to that advocated by CLF is the following: While the governing regulations require consistency, they do not require that the permit limitations that will finally be adopted in a final NPDES permit be identical to any of the WLAs that may be provided in a TMDL. TMDLs are by definition maximum limits; permit-specific limits like those at hand, which are more conservative than the TMDL maxima, are not inconsistent with those maxima, or the WLA upon which they are based.

6 In re City of Moscow, Idaho, NPDES Appeal #00-10, 10 E.A.D. 135, 148 (EPA Envtl. Appeals Bd. July 27, 2001) (emphasis in original) (internal footnote omitted). While EPA’s statement in Moscow emphasizes that agencies like ANR can impose more stringent permit limitations where needed, it does not state that ANR is always required to do so. Indeed, given the procedural posture and the deferential standard of review in Moscow, EPA could not have reached that issue. Further, EPA’s statements in Moscow and in other proceedings that a permit limitation need not be “identical” to a wasteload allocation could simply be recognition that the units of a wasteload allocation often must be translated to different units when placed into an NPDES permit. See id. at 148 n.35 and sources cited therein. In short, nothing in Moscow or any of the other sources cited by CLF, and nothing that the Court has unveiled in its own independent research on this issue, reveals EPA’s position—one way or the other—on whether a state agency must analyze the underlying assumptions of an applicable TMDL wasteload allocation at each permit issuance to determine whether to impose more stringent permit limitations than those required under the applicable TMDL. Because EPA has yet to analyze or state a determination on this issue, we find that we cannot apply Chevron deference in this proceeding: “There is simply nothing here to which we may defer.” St. Agnes Hosp. v. Sullivan, 905 F.2d 1563, 1568 (D.C. Cir. 1990); accord id. (“[D]eference is not due . . . where the Secretary has never issued any regulation, ruling or policy statement . . . .”); Conn. Primary Care Ass’n v. Wilson-Coker, No. 3:02cv626, 2006 WL 2583083, at *8 (D. Conn. Sept. 5, 2006) (“Because [the agency] never engaged in any interpretation . . . , there is nothing for the Court to defer to.”). In short, Chevron deference is available only when, among other things, “the agency has issued regulations or taken other considered and official action, declaring that course.” Federal-Mogul Corp. v. United States, 63 F.3d 1572, 1579 (Fed. Cir. 1995). Further, as in the Federal-Mogul case, EPA’s non- participation in the current appeal before this Court “leaves us with some question about its current posture.” Id. at 1580; see also Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 212 (1988) (“[W]e have declined to give deference to an agency counsel’s interpretation of a statute where the agency itself has articulated no position on the question . . . .”). For these reasons, we find that we cannot defer to EPA on the interpretation of the statutes and regulations at issue here.

7 II. Whether the Phosphorus WQBEL Violates the Five-Year Limitation on NPDES Permits In our independent analysis of the legal issues in this appeal, we next must address whether the process of automatically adopting—without further analysis—a WQBEL directly from a maximum wasteload allocation in a more-than-five-year-old TMDL violates the statutory five-year limitation on NPDES permits. See 33 U.S.C. § 1342(b)(1)(B). For the reasons detailed below, we conclude that it does. It is helpful to start our analysis by looking at the holding of the former Vermont Water Resources Board in Re: Village of Enosburg Falls, No. WQ-03-03, Memorandum of Decision (Apr. 21, 2004), available at http://www.nrb.state.vt.us/wrp/decisions/wrbdecisions/2004/wq-03- 03mod.pdf. While we recognize that under 10 V.S.A. § 8504(m) this Court must give “the same weight and consideration” to decisions of the former Water Resource Board that we give to our own previous decisions, we are not bound by those previous decisions. At any rate, we find Enosburg Falls distinguishable from the appeal at hand. We first focus our analysis on the conclusion announced in Enosburg Falls (and repeated by some of the parties to this appeal) that if ANR is forced to analyze whether to impose more stringent permit limitations than what the TMDL requires, it “would render the TMDL process meaningless.” Id. at 6. Our legal analysis causes us to disagree. It is significant that the Water Resources Board reached this conclusion a mere year and a half after the Champlain TMDL was issued. At that time, so soon after the TMDL was developed, it probably would have been meaningless to engage in further analysis as to whether more stringent permit limitations were needed. But the appeal before us now arises more than six and a half years since the Champlain TMDL was issued. In that intervening period, the five-year statutory time limit for NPDES permits has run, and there has been ample time to study whether the underlying assumptions of the Champlain TMDL have been met so as to bring Lake Champlain into compliance with water quality standards. The undisputed present fact is that Lake Champlain currently experiences roughly twice the phosphorus load than the targeted loading capacity set by the 2002 Champlain TMDL. In light of the lengthy period of time that has passed since the 2002 Champlain TMDL was issued, we conclude that it would be meaningful—and is in fact required under the Clean Water Act and its implementing regulations—to analyze at each permit issuance whether more stringent permit limitations are required. This does not deprive the Champlain TMDL of its

8 import because the TMDL—a Lake-wide maximum daily load determination—remains a maximum, a ceiling that limits any WQBEL set for phosphorus discharges, while leaving the determination of whether to impose more stringent limitations to further analysis. “Maximum” means maximum, just like the D.C. Circuit has held that “[d]aily means daily.” Friends of the Earth v. EPA, 446 F.3d 140, 142 (D.C. Cir. 2006). As with budgetary decisions made every day, the imposition of one maximum (such as a credit limit) does not mean that no additional limits are needed.10 We find no support in the Act or its implementing regulations for the premise that perpetual use of total maximum daily limits in and of themselves satisfy the statutory goal. Rather, “the national goal” of the Clean Water Act was “that the discharge of pollutants into the navigable waters be eliminated.” 33 U.S.C. § 1251(a)(1) (emphasis added). In fact, the initial goal of the Clean Water Act was to eliminate all discharges “by 1985.” Id. In line with the goal of eliminating all discharges of pollutants into navigable waters, 33 U.S.C. § 1342(b)(1)(B) requires that each NPDES permit expire within five years of its issuance. As one scholar has noted, it is important to keep in mind that NPDES permits “were to be issued for just five-year terms, and businesses were to adopt new technology in the transition time to eliminate their discharges” in that five-year period. Mary Christina Wood, Nature’s Trust: Reclaiming an Environmental Discourse, 25 Va. Envtl. L.J. 243, 253 (2007) (emphasis added) (internal citation to 33 U.S.C. § 1342(b)(1)(B) omitted); see also Nw. Envtl. Advocates v. EPA, No. C 03-05760 SI, 2006 WL 2669042, at *12 (N.D. Cal. Sept. 18, 2006) (“[T]he requirement that NPDES permits last only five years serves to ensure that permits evolve to reflect advances in technology.” (emphasis added)), aff’d, 537 F.3d 1006 (9th Cir. 2008). But see Natural Res. Def. Council v. N.Y. State Dep’t of Envtl. Cons., 864 N.Y.S.2d 486, 488 (N.Y. App. Div. 2008) (holding without discussion that an expedited “administrative renewal” of a permit did not violate the requirement that permits must be renewed every five years).11 We find further

10 We believe an appropriate analogy may be gleaned from our current fiscal crisis, which appears to have been brought on in part by a practice of banks, creditors, and debtors all concluding that the appropriate level of personal debt should equate to the maximum allowed levels of debt. 11 At least one scholar has noted that New York’s expedited “administrative renewal” of permits “directly contradicts the CWA scheme mandating permit review based on the five year life of [a] permit.” Karl S. Coplan, Of Zombie Permits and Greenwash Renewal Strategies: Ten Years of New York’s So-Called “Environmental Benefit Permitting Strategy”, 22 Pace Envtl. L. Rev. 1, 23 (2005). Regardless of the propriety of New York’s expedited “administrative renewal” process, the process that ANR has adopted here goes far beyond that, since ANR never engaged—nor made any plans to engage—in even an expedited review of whether the wasteload allocation in the Champlain TMDL provided a stringent enough WQBEL.

9 support for our interpretation from the very title of the applicable portion of the Clean Water Act: the “permit system is called the National Pollution Discharge Elimination System, reflecting Congressional intent to phase out pollution to waterways.” Wood, supra, at 253 n.31 (emphasis in original). The five-year limitation on NPDES permits is significant here because unlike the permit at issue in Enosburg Falls, the Montpelier WWTF permit on appeal here was issued more than five years after the establishment of the Champlain TMDL. When ANR engages in the process of automatically adopting a wasteload allocation from a TMDL that was developed more than five years earlier and then uses the pollution discharge maximums as a WQBEL, it effectively authorizes a permittee to pollute for more than five years. (Notably, ANR does not contest that the Champlain TMDL is without any expiration date.) The Ninth Circuit has explicitly held that under the Clean Water Act, a state agency cannot do anything that effectively authorizes polluting beyond the five-year term of an NPDES permit. See Citizens for a Better Env’t—Cal. v. Union Oil Co. of Cal., 83 F.3d 1111, 1120 (9th Cir. 1996). In Union Oil, a state agency issued an amended permit order in 1991, which set a limitation on the selenium that permittee Union Oil could discharge, but the agency agreed that this limitation would not become effective until late 1993. Id. at 1114. As that deadline approached, Union Oil successfully negotiated with the state agency to extend the effective deadline until 1998—a full seven years after the selenium limitation was placed in Union Oil’s NPDES permit. Id. This agreement was memorialized in a “cease and desist order” (“CDO”)— a rather ironic term given that the CDO in effect actually gave permission for Union Oil to continue discharging above the final selenium limitation. Although the CDO appeased the concerns of the state agency (which received a $2 million payment in exchange for the extended deadline, see id.), it also led to a coalition of environmental groups filing a citizen suit against Union Oil for violating the terms of its NPDES permit. Id. at 1113 (citing 33 U.S.C. § 1365 (the citizen suit provision of the Clean Water Act)). Union Oil claimed that it had not violated its NPDES permit because the CDO effectively modified its permit to allow it to discharge at higher levels until 1998. Id. at 1118–19. The Ninth Circuit rejected this argument for several reasons, including that “there is a five year duration on the life of an NPDES permit that the ‘effective modification’ asserted here would violate.” Id. at 1120 (citing 33 U.S.C. § 1342(b)(1)(B)).

10 We find the situation in this appeal analogous to the situation that the Ninth Circuit faced in Union Oil, in that if ANR is allowed to automatically apply TMDL limits beyond the five-year duration of NPDES permits—without any further analysis of whether to impose stricter limitations—it would violate the requirement that NPDES permits be subject to review every five years. See id. (citing 33 U.S.C. § 1342(b)(1)(B)). Further, following in this de novo proceeding the type of process that ANR used below would flaunt clear congressional intent that requires agencies to use each five-year renewal period to impose more stringent requirements on a discharger. See 33 U.S.C. § 1342(o) (the anti-backsliding provision); see also, e.g., In re Keene Wastewater Treatment Plant, NPDES Appeal #07-18, 2008 WL 782613, slip op. at 7–8 (EPA Envtl. Appeals Bd. Mar. 19, 2008) (“In enacting the CWA, Congress expected that NPDES permits would be revisited every five years by permit issuers, new pollutant analyses conducted, and the permits reissued with new, sometimes more stringent terms and conditions incorporated, so that the express statutory goals of restoring and maintaining clean water would be achieved.”); United States v. Louisville & Jefferson County Metro. Sewer Dist., No. 91-6461, 983 F.2d 1070, 1993 WL 7516 (6th Cir. Jan. 12, 1993) (unpublished decision) (“NPDES permits have a duration of five years, and, as a general rule, subsequent permits are supposed to be no less stringent than the one before it.” (internal citation omitted)); Nicholas A. Robinson, Legal Systems, Decisionmaking, and the Science of Earth’s Systems: Procedural Missing Links, 27 Ecology L.Q. 1077, 1159 (2001) (“[E]ffluent permits expire no later than every five years and cannot degrade water quality standards, and since water quality standards must be reviewed for upgrade every three years, the effluent permit can be made stricter and stricter as water quality improves and as the water quality standards are increased.” (internal citations omitted)); Kurt Stephenson et al., Toward an Effective Watershed-Based Effluent Allowance Trading System: Identifying the Statutory and Regulatory Barriers to Implementation, 5 Envtl. Law. 775, 795 (1999) (“Permits are reissued every five years and are drafted with the intent to make effluent limitations progressively more stringent.”). Another analogous case on this issue is Natural Resources Defense Council v. EPA, 907 F.2d 1146 (D.C. Cir. 1990) (“NRDC”). In that case, the EPA had decided “that a permit issued under the underground injection well program of the Safe Drinking Water Act would satisfy . . . RCRA [the Resource Conservation and Recovery Act] permit requirement[s] . . . because such a permit qualifies as a RCRA permit under EPA’s permit-by-rule regulations.” Id. at 1165 (internal citations omitted). An environmental organization challenged this EPA combined

11 permitting process on several grounds, including alleged violations of the five-year review requirement on permits issued under RCRA. Id. (citing 42 U.S.C. § 6925(c)(3)). The D.C. Circuit reversed and remanded the case to EPA, holding that although EPA is free to integrate various permitting processes to avoid duplication, the agency cannot grant a RCRA permit until it “demonstrate[s] that [its] process meets the statutory requirements for such a permit.” Id. at 1165. Just as EPA was required in the NRDC case to show that any shortcuts it made in the permitting process did not evade the requirements needed in each permit application proceeding, we similarly conclude that ANR cannot use the Champlain TMDL as a substitute for undertaking its duties to establish whatever phosphorus WQBEL is statutorily required for the Montpelier WWTF, unless ANR can demonstrate that such a process meets the statutory requirements for an NPDES permit. Here, because the Champlain TMDL is more than five years old, and because the Clean Water Act requires that any permit term be limited to a five-year period, see 33 U.S.C. § 1342(b)(1)(B), the process ANR followed in analyzing the Montpelier WWTF application for an NPDES permit is deficient as a matter of law, since the very process allows for pollution discharge limits to be automatically renewed beyond the statutorily required five-year period. Our legal conclusion here is reinforced by the congressional intent expressed in the Clean Water Act, which aims to eliminate all discharges of pollutants into navigable waterways. See, e.g., 33 U.S.C. § 1251(a)(1); see also, e.g., Mark C. Van Putten & Bradley D. Jackson, The Dilution of the Clean Water Act, 19 U. Mich. J.L. Ref. 863, 893–94 (1986) (“[T]he Clean Water Act’s goal of continued progress towards eliminating all pollutant discharges conflicts with dischargers’ interest in avoiding water pollution control requirements . . . . Congress clearly sacrificed dischargers’ competitive interest in favor of the environmental interest to the extent that ambient water quality considerations require controls more stringent than national effluent limitation guidelines. . . . [T]he zero discharge goal [is] the operative principle of the Act . . . .”).12 We therefore conclude that the Clean Water Act does not allow the institutionalization

12 Our decision to look at congressional intent when interpreting the federal Clean Water Act and its implementing regulations is in accord with a number of Vermont Supreme Court cases interpreting federal law. See, e.g., Vt. Student Assistance Corp. v. Zeichner, 167 Vt. 616, 617 (1998) (looking at congressional intent to interpret a federal statute); White Current Corp. v. Vt. Elec. Co-op., Inc., 158 Vt. 216, 227 (1992) (looking at plain language and congressional intent to interpret a federal statute); see also Howard v. Dep’t of Soc. Welfare, 163 Vt. 109, 115 (1994) (looking at the expressed purpose of a federal statute to determine how to interpret it). Also, our analysis of congressional intent underlying the Clean Water Act leads us to note that we find a similar intent expressed by the Vermont legislature in state water quality laws.

12 of permit processing that effectively provides an automatic right to pollute for more than five years. In light of the clear congressional intent to eliminate pollution discharges altogether, 33 U.S.C. § 1251(a)(1), and in light of the requirement that a specific permit authorization to pollute at a certain level run for no more than five years, 33 U.S.C. § 1342(b)(1)(B), we reject ANR’s assertion that the Clean Water Act authorizes using any TMDL maximum wasteload allocation—including one that is more than five years old—as an automatic WQBEL for any NPDES permit. ANR’s interpretation, when taken to it its logical conclusion, would allow the terms of every subsequent effluent discharge permit to mirror any available maximum wasteload allocation in a TMDL for decades into the future. Such a process would render ANR’s responsibilities to be limited to a ministerial act, to be completed every five years when a permit came up for renewal, whereby ANR would issue each successive permit with the exact same effluent limitations as the previous permit. As a result, although ANR’s duties are undoubtedly much simpler under this type of process, it leads to the five-year limitation on NPDES permits becoming superfluous. We are reminded here of another court’s observation that “[w]hile perhaps the [agency] acts with good intentions, its policy arguments must fail in the face of clear statutory language.” Cape Hatteras Access Pres. Alliance v. U.S. Dep’t of Interior, 344 F. Supp. 2d 108, 123 (D.D.C. 2004). Here, the statute is quite clear that no permit—or permit limitation—can last more than five years without some further review and analysis. See 33 U.S.C. § 1342(b)(1)(B). In light of this statutory directive, we are compelled to vacate the permit at issue in these proceedings.

As mentioned earlier, the Montpelier WWTF permit includes a provision limiting the total phosphorus effluent concentration to a monthly average of 0.8 milligrams per liter. This provision is explicitly required by 10 V.S.A. § 1266a(a). The next subsection in that statute creates a savings clause: “Notwithstanding any provision of subsection (a) of this section to the contrary, the secretary shall establish effluent phosphorus wasteload allocations or concentration limits within any drainage basin in Vermont, as needed to achieve wasteload allocations in a total maximum daily load document approved by the United States Environmental Protection Agency, or as needed to attain compliance with water quality standards adopted by the Vermont water resources board pursuant to chapter 47 of this title.” 10 V.S.A. § 1266a(b). At first glance, this language appears to be an endorsement of ANR’s decision to adopt the wasteload allocation in the Champlain TMDL as a permit limitation without further analysis. However, even if the Vermont Legislature did endorse this type of procedure, it would not relieve ANR of its duties to impose stricter limitations when that is required under the federal CWA. See 33 U.S.C. § 1370 (noting that where there is a conflict between state and federal law, the more “stringent” provision prevails). We conclude that the savings clause in 10 V.S.A. § 1266a(b) serves a similar purpose to its companion language in 33 U.S.C. § 1370: both provisions aim to ensure that the most stringent water quality protections—wherever they arise in either state or federal law—be implemented to protect the State’s waters. See also In re NPDES Stormwater Petition, No. 14-1-07, slip op. at 8 (Vt. Envtl. Ct. Feb. 18, 2009) (Durkin, J.) (noting that the Clean Water Act calls for “multiple layers of protection” to prevent violations of water quality standards).

13 III. Whether the Phosphorus WQBEL Violates the Requirement that a WQBEL be Consistent with the Assumptions Underlying an Applicable Wasteload Allocation Turning to 40 C.F.R. § 122.44(d)(1)(vii)(B), we find additional legal foundation for why ANR must engage in some degree of site-specific and time-specific analysis for each discharge application to determine whether a suggested wasteload allocation provides a stringent enough permit limitation. Although ANR, the Water Panel, and the City are correct that any permit limitation pulled directly from the Champlain TMDL meets the “requirements” of the TMDL (since it establishes a limit for that facility that does not exceed the wasteload allocation in the TMDL), their interpretation renders the word “assumptions” in 40 C.F.R. § 122.44(d)(1)(vii)(B) superfluous, by never engaging in an analysis of what the underlying assumptions of the TMDL are and whether those assumptions have proved accurate and reliable in the years since the TMDL was issued. Our Supreme Court has directed us to “presume statutory language is inserted advisedly and not intended to create surplusage.” State v. Carroll, 2003 VT 57, ¶ 7, 175 Vt. 571; cf. also Friends of the Earth, 446 F.3d at 145 (“The existence of two conditions does not authorize EPA to disregard one of them.”). We therefore must make every effort to give meaning to the requirement that each permit limitation be “consistent with the assumptions” underlying a wasteload allocation. 40 C.F.R. § 122.44(d)(1)(vii)(B); accord, e.g., Jeffrey M. Gaba, Generally Illegal: NPDES General Permits Under the Clean Water Act, 31 Harv. Envtl. L. Rev. 409, 434–35 (2007) (“[I]n addition to consistency with the applicable WLA, EPA regulations also require that any WQBEL be consistent with the ‘assumptions’ of any applicable WLA.” (citing 40 C.F.R. § 122.44(d)(1)(vii)(B)). The Champlain TMDL is rife with assumptions that played a role in the setting of the wasteload allocation for the Montpelier WWTF. Most significantly, the Champlain TMDL explicitly states that “the balance between allowable point and nonpoint source loads is part of the allocation decision.” Champlain TMDL at 21. It is perfectly acceptable to engage in these types of tradeoffs when drafting a TMDL. See 40 C.F.R. § 130.2(i) (“If Best Management Practices (BMPs) or other nonpoint source pollution controls make more stringent load allocations practicable, then wasteload allocations can be made less stringent. Thus, the TMDL process provides for nonpoint source control tradeoffs.”); see also Douglas R. Williams, When Voluntary, Incentive-Based Controls Fail: Structuring a Regulatory Response to Agricultural Nonpoint Source Water Pollution, 9 Wash. U. J.L. & Pol’y 21, 82 (2002) (“EPA made clear that

14 in establishing TMDLs, states were free to make reduction trade-offs between nonpoint and point sources.”). In allocating phosphorus discharge reductions among various sources, the Champlain TMDL noted that “Vermont has already accomplished major reductions in point source phosphorus loading.” Champlain TMDL at 90. The Champlain TMDL therefore assumed that there would need to be significant decreases in nonpoint source discharges—an assumption that clearly played a role in the relatively lenient wasteload allocations for point sources. Even a cursory examination of the Champlain TMDL reveals that it places the bulk of responsibility for phosphorus reductions on nonpoint sources. See, e.g., id. at 89 (“The major responsibility for future phosphorus load reductions necessary under the Lake Champlain Phosphorus TMDL will fall on nonpoint sources.”).13 Given this allocation and the tradeoffs that were made, the success of the Champlain TMDL, as delineated when it was issued in 2002, depended primarily upon reductions in discharges from nonpoint sources. See, e.g., id. at 44 (“Vermont’s approach for controlling nonpoint sources . . . calls for a full implementation effort in each program area to address all controllable phosphorus sources.”). In fact, the Champlain TMDL rejected the idea of allowing trading between point and nonpoint sources because it “would run a high risk of trading a point source loading increase for a nonpoint source phosphorus reduction that is actually necessary to meet the nonpoint source portion of the TMDL.” Id. at 89. The drafters of the 2002 Champlain TMDL felt comfortable at the time placing the bulk of responsibility for phosphorus reductions on nonpoint sources because “there [wa]s ample reason to believe that the nonpoint source reductions will be met or exceeded.” Id. at 43 (emphasis added). The use of the phrase “to believe” here reveals an underlying assumption made in 2002: that efforts to reduce the phosphorus loading of Lake Champlain from nonpoint sources would be successful. Such assumptions are proper and in fact necessary when dealing with future actions. However, the TMDL assumptions that were made in 2002 become problematic when they are used as the sole basis for setting a WQBEL in 2008, particularly when these assumptions were never checked in the actual permit application process, despite evidence that Lake Champlain is currently receiving roughly twice the levels of phosphorus compared to what was allowed under its approved loading capacity in the 2002 Champlain

13 We find further support for this conclusion in the fact that the Montpelier WWTF was granted a permit limitation that was significantly higher than any level of phosphorus that the Facility had actually discharged in recent years.

15 TMDL. We therefore conclude that 40 C.F.R. § 122.44(d)(1)(vii)(B) directs that agencies not blindly accept such past assumptions, but rather analyze them at each permit issuance—or at least at each permit issuance that occurs more than five years after the issuance of the applicable TMDL—to determine whether those assumptions continue to have a basis of reliability. In addition to the general assumption that there would be massive decreases in nonpoint source discharges of phosphorus, the 2002 Champlain TMDL contains a number of specific assumptions as to how those decreases will come about. For instance, the 2002 Champlain TMDL assumed that agricultural land will continue to be protected “through the cooperative efforts of federal and state programs and willing land owners.” Id. at 55. The protection of agricultural land was important at the time of the drafting of the 2002 TMDL because “[c]onversion of agricultural land to some other non-agricultural use or purpose (e.g., residential or commercial) has been shown to result in the potential for significant increases in phosphorus.” Id. To accomplish the preservation of agricultural land, the 2002 Champlain TMDL also depended upon “willing land owners”—presumably referring to those noble farmers who choose to keep their land in agricultural use. Id. The existence of such “willing land owners” constitutes another underlying assumption of the 2002 TMDL. See id. Should those landowners (or their successors-in-interest) reverse course at any time (for instance, to generate needed income during difficult economic times), nonpoint sources of phosphorus will presumably increase, and the Champlain TMDL will be at risk of no longer meeting its goals for acceptable phosphorus levels. See id. The Champlain TMDL also explicitly stated in 2002 that the “progress to date in reducing phosphorus loads to Lake Champlain has been possible because of a sustained commitment of state and federal funding for point and nonpoint source programs.” Id. at 47. This track record of continued funding supposedly “provide[s] further reasonable assurances that progress will continue to be made in meeting the phosphorus load allocations established by the TMDL.” Id. But if reductions in phosphorus loads are “possible because of a sustained commitment of state and federal funding,” id., the implication is that reductions in phosphorus loads are dependent on such funding. See also id. at 56 (“Additional state and federal funds will be needed to continue this successful program.” (emphasis added)). Thus, far from providing “reasonable assurances,” the dependency on funding from multiple governmental bodies reveals that the TMDL is at risk of failing if the state or the federal government chooses to cut back on

16 funding—a possibility that is quite real given the current economic realities and budget shortfalls. Indeed, in addition to depending upon budgetary stability in programs that were already established, the TMDL also noted in 2002 that “[e]xpanded programs are needed” in certain areas to meet the nonpoint source load allocation goals. Id. at 57. To expand these programs, so that further reductions in phosphorus loading could be accomplished, “[i]ncreased governmental funding [will be] needed.” Id. at 58 (emphasis added). The TMDL authors made a number of other assumptions in their calculations, such as continued compliance with forestry Accepted Management Practices (“AMPs”). Id. at 77–78 (“For the purpose of the Lake Champlain Phosphorus TMDL, it will be assumed that compliance with the AMPs will prevent increases in phosphorus loading from logging activities in Vermont.” (emphasis added)). The TMDL is also premised upon the assumption that ANR or some other entity will be able to contain or offset the effects of urbanization: On average, developed land in the basin yields more phosphorus runoff per unit of area than either agricultural or forest land. The trend towards urbanization that is apparent in some Vermont portions of the Lake Champlain Basin is creating new phosphorus sources. In order to attain the wasteload allocation for developed land, phosphorus runoff generated by new development must be minimized through proper site design, construction techniques, and stormwater treatment, and phosphorus load reductions from existing developed areas must be achieved sufficient to offset the effects of new development. Id. at 58 (emphasis added) (internal citation omitted). The use of the word “must” here implies yet another assumption that must now be reviewed: that water quality standards will not be met if phosphorus runoff from new development is not contained. See id. Indeed, the Champlain TMDL explicitly states that “[e]xpanded efforts in the following areas will be needed to attain the phosphorus wasteload allocation for developed land,” and then lists “[s]tormwater discharge permitting,” “[e]rosion and sediment control at construction sites,” “[b]etter backroads,” and “[l]ocal municipal actions.” Id. (emphasis added). Taking just one of these factors, we note that it is a bit surprising to see that the Champlain TMDL’s success depends upon “better backroads”—a phrase that sounds somewhat oxymoronic in a state like Vermont, where potholes, frost heaves, washboards, free-flowing drainage ditches, and other mud-season obstacles can turn any outing into an adventure. Granted, the Champlain TMDL lays out an elaborate plan for upgrading Vermont’s backroads,

17 but at the end of the day this plan presumes that there will be a large amount of local and state funding, as well as changes to municipal regulations—all things that “need the support of town officials and voters” to be successful. Id. at 73. The TMDL notes that “[t]here are 136 Vermont cities and towns that are either wholly or partly within the Lake Champlain Basin.” Id. at 74. Further, “[i]n many cases, the delivery of phosphorus to Lake Champlain from developed land results from activities that are under the jurisdiction of municipalities.” Id. It is unclear what portion of those 136 cities and towns the TMDL presumed must vote to increase funding and to change local zoning regulations before the Champlain TMDL will meet its water quality goals. In summary, the Champlain TMDL itself states that its predicted success in protecting water quality standards depends upon an array of measures, including vast amounts of funding— at the federal, state, and local level—not only being sustained, but being increased by a total of nearly $10 million dollars per year. See id. at 95. This funding, along with a host of other measures described in the Champlain TMDL, in turn depends upon the state and possibly as many as 136 cities and towns all voting to implement funding as well as numerous other legislative changes in a relatively short period of time. See id. at 73. Other measures, such as the protection of agricultural land, depend not only upon governmental decisions, but also upon decisions by private actors, described in the TMDL as “willing land owners,” who will forevermore reject development on their lands. Id. at 55. All of these measures are necessary for the presumed success in phosphorus reductions anticipated by the 2002 Champlain TMDL, which placed the bulk of responsibility for phosphorus load reductions on nonpoint sources. Id. at 89. In fact, the TMDL noted that “the scope of nonpoint source control programs necessary to meet the target loads for the [L]ake is so extensive that nearly every feasible BMP is likely to be needed in some watersheds in order to attain the nonpoint source portion of the phosphorus load allocation.” Id. While the Champlain TMDL itself is not challenged in this lawsuit, the regulations implementing the Clean Water Act direct that its underlying “assumptions” must at least be reviewed in the course of a review of a new NPDES permit application. See 40 C.F.R. § 122.44(d)(1)(vii)(B) (requiring that WQBELs be “consistent with the assumptions and requirements of any available wasteload allocation” in an approved TMDL (emphasis added)). Here, there were a number of underlying assumptions—including doing nearly everything “feasible” to reduce phosphorus from nonpoint sources, see Champlain TMDL at 89—that led to

18 a tradeoff that allowed a facility such as the Montpelier WWTF to have the wasteload allocation that it was granted, rather than a more stringent limitation. Those assumptions might not be as worrisome if the Champlain TMDL incorporated a large enough “margin of safety,” as is required by 40 C.F.R. § 130.7(c)(1), to ensure that water quality standards will be met. But here, although the Champlain TMDL nominally included an “implicit margin of safety . . . by the fact that the model’s mean predicted phosphorus concentrations are below the applicable phosphorus criteria for most lake segments,” Champlain TMDL at 43 (emphasis added), the Main Lake Segment at issue here did not include such a margin of safety. See id. at 17 fig.4. Thus, it is all the more important that state agencies undergo an independent evaluation at each permit reissuance to determine whether the assumptions underlying the TMDL allocations (for point and nonpoint sources) still justify allowing the maximum wasteload allocation listed in the TMDL. Without the margin of safety that was built into the phosphorus allocations for most of the other Lake segments, and given how optimistic the projections of the Champlain TMDL are, stricter permit limitations may be needed to compensate for any exceedances in any other allocated discharges to the Main Lake Segment. In short, the Champlain TMDL was premised in 2002 on a number of assumptions about planned decreases in discharges from nonpoint sources. Given the tradeoff system used in the Champlain TMDL, those assumptions necessarily led to a more lenient wasteload allocation for the Montpelier WWTP than what might otherwise be expected to ensure that water quality standards are met. Although the Champlain TMDL requires “additional phosphorus removal” from wastewater treatment plants and other point sources, it notes that these measures are all “practical and cost-effective.” Id. at 90. But the reference to “practical and cost-effective” measures implies that the wasteload allocations in the TMDL are not really WQBELs at all— rather, they appear to be less stringent technology based effluent limitations (“TBELs”). Cf. 40 C.F.R. § 130.2(h) (noting that “WLAs constitute a type of water quality-based effluent limitation,” not a less stringent TBEL). While such an allocation might be acceptable if all of its underlying assumptions were met and nonpoint sources really decreased their discharges of phosphorus by the amounts called for in the Champlain TMDL, this allocation is not acceptable if those assumptions have not yet been realized. Hence, 40 C.F.R. § 122.44(d)(1)(vii)(B) requires that agencies analyze whether the underlying assumptions of a TMDL have come to pass when they consider the issuance of a new or renewed discharge permit.

19 To be clear, our Decision today does not determine whether the underlying assumptions of the Champlain TMDL have been met. Rather, we simply hold that the implementing regulations of the Clean Water Act require a permit review to include an analysis of whether past TMDL assumptions have some impact upon the discharge application being considered. One stark reality that appears to be undisputed is that the efforts to control or reduce phosphorus discharges into Lake Champlain have not yet been successful. According to the latest estimate, which none of the parties dispute, the Main Lake Segment currently receives phosphorus loads totaling 217.9 metric tons per year. See 2008 State of the Lake & Ecosystem Indicators, attached as CLF Ex. G, at 5 fig.4 (listing loads of 148.4 mt/yr from Vermont and 69.5 mt/yr from New York). This number is nearly double that segment’s phosphorus loading capacity of 110.3 metric tons per year. Thus, in ANR’s own words, “ANR does not contest that under current conditions, the Main Lake Segment has no remaining loading capacity for discharges of phosphorus.” (ANR’s Resp. to CLF’s Statement of Undisputed Material Facts, at 2 (emphasis in original)). Any review of the application for a renewed discharge permit for the Montpelier WWTF therefore must assess the reality that the Champlain TMDL is failing to achieve its mandate under the Clean Water Act to protect water quality standards, and that Lake Champlain remains an impaired water body.14 This current reality makes it all the more important for ANR to include in its analysis of any discharge permit application whether the relatively less stringent wasteload allocations listed in the Champlain TMDL are still protective enough to meet the requirements of the Clean Water Act. We pass no judgment on that specific issue today, but we do hold that ANR must engage in this analysis before it can establish the proper phosphorus WQBEL for the current Montpelier WWTF permit. We recognize that it is possible that it will be quite costly for the City to comply with more stringent phosphorus controls. On the other hand, the TMDL states that the Montpelier WWTF could use “anaerobic selector zones” to reduce phosphorus discharges in a way that simultaneously leads to “cost savings relative to current operating costs, even if more stringent phosphorus limits are required.” Champlain TMDL at 52; accord id. at 53 tbl.17 (listing the Montpelier facility as a site that could have significant savings in annual operating costs through use of this technology). As with any new technology, this would require an initial capital

14 We may be “beyond the stage when merely shifting the deck chairs on a sinking ship will suffice.” In re Alexandria Lake Area Sanitary Dist. NPDES/SDS Permit #MN0040738, 763 N.W.2d 303, 328 (2009) (Anderson, J., dissenting). While Alexandria addresses phosphorus discharges and we find the above quote relevant to our analysis, the Alexandria decision is otherwise distinguishable from the case before us in many ways.

20 investment, which will likely cause hardship for the City. Although we regret that the City might have to incur this—and possibly other—capital expenditures during difficult economic times, we have previously noted that we are without authority to take such monetary considerations into account in rendering our decision on this type of matter: We are obligated to determine the applicable law and apply it to the undisputed material facts. Similarly, while we are very mindful of the significant monetary obligations that may follow our decision here, we cannot allow those monetary consequences to impact our legal analysis at this stage of the proceedings. We remind the parties that we are a court of limited subject matter jurisdiction, and while acts or decisions of the ANR Secretary may be appealed de novo to this Court, we are not aware of any statutory provision allowing our Court to conduct an economic analysis in these types of proceedings. In re NPDES Stormwater Petition, No. 14-1-07, slip op. at 36 (Vt. Envtl. Ct. Aug. 28, 2008) (Durkin, J.). In addition to the jurisdictional limitations that prevent the Court from engaging in an economic analysis here, the structure of the Clean Water Act precludes any such analysis when it comes to protecting water quality standards. See, e.g., In re City of Moscow, Idaho, NPDES Appeal #00-10, 10 E.A.D. 135, 168 (EPA Envtl. Appeals Bd. July 27, 2001) (“[S]ection 301(b)(1)(C) of the CWA requires unequivocal compliance with applicable water quality standards, and does not recognize an exception for cost or technological infeasibility.”). In short, the decision that Congress made when it passed the Clean Water Act was that water quality takes priority over all other considerations, no matter how legitimate those considerations may be. Thus, according to the Clean Water Act, whenever ANR or this Court analyzes what must be done to bring a facility into compliance with applicable water quality standards, the costs—or even the feasibility—of such measures are not proper for our consideration, no matter how meritorious they may be. See id. Indeed, the EPA Environmental Appeals Board has explicitly stated that any agency implementing the Clean Water Act is without authority to grant variances from water quality standards based on cost or technological infeasibility. In re J&L Specialty Prods. Corp., NPDES Appeal #92-22, 5 E.A.D. 31, 48–49 (EPA Envtl. Appeals Bd. Feb. 2, 1994). Similarly, this Court—standing in place of ANR in this de novo appeal—is without authority to take such considerations into account. Cf. Friends of the Earth, 446 F.3d at 148 (“If adherence to this mandate leads to unintended consequences for water quality or for municipal pocketbooks, interested parties should direct their concerns to EPA or to

21 Congress, either of which can take steps to mitigate any fallout from the CWA’s unambiguous directive. We, however, have no such authority.”). In summary, we find that ANR must engage in an analysis, prior to the issuance of the discharge permit applied for here, to determine whether the wasteload allocation in the Champlain TMDL is protective enough of state water quality standards to be adopted as a WQBEL, or whether a more stringent limitation is required. We conclude that such an application review process is required as a matter of law, so that a WQBEL meets the five-year limitation on permits required by 33 U.S.C. § 1342(b)(1)(B), and so that a WQBEL is “consistent with the assumptions and requirements of any available wasteload allocation” in an approved TMDL, pursuant to 40 C.F.R. § 122.44(d)(1)(vii)(B). This application review process will not lead to a conflict with the Champlain TMDL, which maintains a ceiling beyond which the permit WQBEL cannot pass. Cf., e.g., Friends of the Earth, 446 F.3d at 145 (“[A]ll water bodies can achieve water quality standards if their TMDLs are set low enough—if all else fails, they can be set to zero—and the two requirements therefore never conflict with each other.”). As one scholar has noted, “[a]fter setting TMDLs, states must take additional measures to control both point and diffuse pollution through such measures as more stringent point source permits.” Cynthia J. Aukerman, Agricultural Diffuse Pollution Controls: Lessons for Scotland from the Chesapeake Bay Watershed, 20 J. Land Use & Envtl. L. 191, 241 (2004). We agree. A TMDL is meant to be a safety net. Whether the Champlain TMDL serves that purpose here is not at issue in this case. But we conclude that the Clean Water Act prohibits the sole reliance upon a TMDL that the record in this case suggests was all that occurred in the ANR review process below. Specifically, we conclude that the Clean Water Act requires that before this discharge permit can be issued, further analysis is needed to determine the proper phosphorus WQBEL for the Montpelier WWTF.

Conclusion For all the reasons more fully discussed above, we GRANT summary judgment to CLF and hold that the Montpelier WWTF permit is invalid as issued, and we DENY all other requests for summary judgment. Although we would normally maintain jurisdiction to engage in a de novo analysis of what phosphorus permit limitation is appropriate for the Facility, we instead choose here to follow the Supreme Court’s guidance that in a situation like this—where ANR is yet to engage at all in the analysis that the Clean Water Act requires, the proper recourse is to

22 remand the case to ANR to engage in that analysis in the first instance. See In re Stormwater NPDES Petition, 2006 VT 91, ¶ 30. We therefore REMAND the pending discharge permit application back to ANR to engage in the requisite analysis directed by this Decision.

Done at Berlin, Vermont, this 30th day of June 2009.

___________________________________ Thomas S. Durkin, Environmental Judge