28 ELR 10415 | Environmental Law Reporter | copyright © 1998 | All rights reserved

TMDLs III: A New Framework for the Clean Water Act's Ambient Standards Program

Editors' Summary: For the past quarter century, the Clean Water Act has primarily relied on technological standards to abate point source pollution and achieve national clean water goals. Water quality standards lay largely dormant until the 1990s, when they were activated by citizen suits demanding implementation of § 303(d) of the Act — the abatement of pollution discharges based on total maximum daily loads. The first Article in this series, published in the July 1997 issue of this Reporter, described the enactment of § 303(d) and the active promotion of this approach to water pollution control by state governments and industry. A second Article, published in the August 1997 issue of ELR, described the subsequent neglect of § 303(d) by the states and EPA, and the eventual litigation that brought this issue to center stage. This Article describes the new § 303(d) program as it is emerging from the courts, EPA, a federal advisory committee, related federal programs, and the states.

The author is a Professor of Law at Tulane Law School. The research assistance of Eileen Budd, Tulane Law School '98, Louis Spencer '98, Ann-Marie Johnson '99, Scott Galante '99, and Christopher Moore '99 is acknowledged with gratitude.

$=S

[28 ELR 10415]

Almost 25 years after the passage of the [Clean Water Act], the national water program is at a defining moment. We — meaning each of you, each of our State, local, and Tribal partners, and all of us in the Office of Water — are making the transition from a clean water program based primarily on technology-based controls to water quality-based controls implemented on a watershed basis…. The [total maximum daily load (TMDL)] program is crucial to success because it brings rigor, accountability, and statutory authority to the process. These are the reasons the Administrator and I so strongly support the TMDL program in States, Tribes, and [U.S. Environmental Protection Agency] Regions. (Robert Perciasepe, Assistant Administrator for Water, U.S. Environmental Protection Agency)¹

Environmental law is a continuing experiment, and one ingredient of its success has been its tendency to throw several approaches at a problem and test their survival. The Clean Water Act (CWA)² presents nearly a dozen programs for reducing pollution and still other initiatives encouraging voluntary reductions.³ But the granddaddy of all approaches to pollution control was and is the regulation of discharges by ambient standards.⁴ The continuing vitality and attraction of this approach could be surprising, given the fact that it has never really worked for water pollution, air pollution, or anything else.⁵ Ambient standards regulation remains [28 ELR 10416] popular, however, because of its reliance on state and local management, and because of the illusion it offers that reduction measures can be calibrated by, and limited to, only those necessary to offset environmental effects. The assumptions that state and local authorities will actually act to control pollution, and that it will be possible for them to do so by calibrating causes and effects, are about to be put to a severe test in the emerging total maximum daily load (TMDL) program under § 303(d) of the Clean Water Act.⁶

Section 303(d) requires states to identify waters that remain polluted after the application of technological standards, to determine the total maximum loadings of pollutants that would bring these waters up to grade, and to incorporate allocations of these loadings into discharge permits and state water quality plans.⁷ The first Article in this series described the origins of § 303 and its retention in the Act at the insistence of both the states and the regulated community⁸; in 1972 and at every reconsideration of the Act since that time, the states and industry have insisted that this was the most sensible approach to pollution control and one that they were ready and able to carry out.⁹ The second Article in this series described the great disconnect between this enthusiasm for ambient-based pollution control and its practice: 25 years into the Act, few states had promulgated a single TMDL, and the U.S. Environmental Protection Agency (EPA) had taken few steps even to encourage them forward.¹⁰ In the early 1990s, this joint venture in nonperformance began to unravel through a series of lawsuits forcing EPA and the states to begin the necessary listings and load allocations (LAs).¹¹ Propelled by the litigation, EPA issued several interim guidance documents to its regional offices and states; convened a Federal Advisory Committee Act (FACA) committee; and came face-to-face with the problem it, Congress, states, and just about everyone else had been tiptoeing around for more than a quarter century — nonpoint source pollution.¹²

The smoke is now beginning to clear. The TMDL litigation has settled into a routine of predictable, if difficult, issues. EPA guidance is now published in operative, if not regulatory, form. The FACA committee has reported its findings. The White House has launched a Clean Water Initiative with primary focus on nonpoint sources of pollution. TMDL-related EPA programs and authorities of other federal agencies have been brought, if not to bear, at least closer to the scene. At the same time, state compliance with § 303 has faced its first test in a new round of listings that may double the number of identified, impaired lakes, estuaries, and stream segments. And in response to all of this pressure, the first rounds of TMDLs for these waters are emerging from state agencies.

This Article undertakes to assess these pieces of the action, and to put them into a single, contemporary frame. A final Article in this series will reflect on where this program is going, whether it makes sense, and what can be done to improve its chances of success.

Pieces of the Frame

The descriptions that follow are of complex events. Every lawsuit is its own private war; every initiative proposed by EPA, the FACA committee, the White House, and other agencies is underpinned by reams of paper, options, and thoughtful, at times argumentative, discussion; each state response is unique; and each TMDL relates to a hydrology all its own. The full story of any of them bears more detailed treatment than that offered here. The purpose of this discussion is to capture their commonality, and their places in this new armada on water pollution that is assembling around and through the use of TMDLs.

The Litigation

While I am pleased with the progress that many States are making, we still have a long way to go before we achieve our water quality goals everywhere. I remain very concerned about the extraordinarily high rate of litigation in this program and I think it is crucial that we Federal managers, together with our State partners, take every step we can to make sure that this program is carried out effectively and quickly. (Robert Perciasepe, Assistant Administrator for Water, EPA)¹³

Against a background of federal environmental programs in which litigation has played a central role, it is hard to think of any program more precipitously driven by citizen suits from absolute zero toward its statutory destiny than TMDLs. Short of some outside impetus, whatever Congress prescribed in § 303(d) was going to be ignored for no more complex reasons than (1) compliance was hard and (2) ignoring seemed possible.¹⁴

Starting in the early 1990s, a first wave of lawsuits established that ignoring § 303 was no longer possible and that [28 ELR 10417] continued state inaction constituted action, triggering EPA's duty to respond.¹⁵ As of the time of this writing, citizen suits have led to at least 11 consent decrees against EPA and the states in which those suits were filed; another 14 state cases are pending, and 4 states have received notices of intent to sue.¹⁶ Overall, litigation has challenged compliance in more than half the states of the country, and yet more is brewing. The issues in these cases have tracked the literal requirements of the statute, challenging (1) the failure to list state waters, then (2) the adequacy of these lists, and then (3) the failure to prepare TMDLs, leading to schedules for their preparation ranging from 12¹⁷ years to as few as 5.¹⁸ Up to this point, the alleged violations have been essentially procedural, and the courts have been able to avoid the content, or lack of content, of the TMDLs themselves.

This will change. A number of pending cases question the substance of what is currently being proffered as TMDLs.¹⁹ Under § 303 and EPA regulations, a TMDL is to include the sum of both point source waste load allocations (WLAs) and nonpoint source LAs, plus a margin of error for uncertainty and a margin for future growth.²⁰ The majority of current TMDLs in New York, Louisiana, and several other states — if not nearly all states — are alleged with credible specificity to be point source WLAs, plus nothing.²¹ This deficiency is probably the result of states, EPA, and, once a case is filed, the U.S. Department of Justice throwing any information available into the breach against charges that the environmental agencies had not been doing their job. On a clear day, and away from the shadow of a lawsuit, there is no reasonable way EPA could characterize much of the states' work here as TMDLs.²² More fundamentally, however, this limited focus on WLAs results from the tendency of state and federal regulators to finger identifiable point sources, even beleaguered municipal waste treatment systems, rather than tackle the timber industry, fertilizer manufacturers, and the rest of the nonpoint world.

Yet another round of citizen suits can be expected over the contents and, then, the implementation of TMDLs. Whether TMDLs should include provision for their own implementation is a gut issue unsettled by the statute and by EPA regulation, and was a live debate in the FACA committee.²³ Environmental interests see implementation as a necessary component of § 303(d), which is to say mandatory.²⁴ Agriculture and other interests see implementation carried out through § 319 planning for nonpoint sources or state water quality plans under § 303(e), which is to say voluntary.²⁵ EPA has proposed an interim compromise which leans both ways.²⁶ While these parties war over the issue of implementation at the national level, citizen suits and state practice may be establishing the precedent. A recent consent decree in the state of Washington requires a schedule for TMDLs to include plans for their own implementation.²⁷ A recent TMDL for the Newport Bay/San Diego Creek watershed in Orange County, California, includes a detailed implementation and nutrient management plan, with provision for monitoring and review.²⁸

The litigation will not be all from the citizen side. Industry has already mounted a serious challenge to one TMDL, albeit unsuccessful in outcome.²⁹ Agriculture and timber interests have given advance notice that they are ready to challenge the application of TMDLs to nonpoint sources³⁰ and, by implication, to the inclusion of implementation and enforcement measures.³¹ They have also given notice of their [28 ELR 10418] intent, in the name of "sound science," to challenge listings of waters and identification of sources based on anything short of dispositive evidence.³² Responding to this or other impetus, some states have taken more conservative positions in their latest submissions to EPA on the information needed to support the listing of impaired waters.³³ The adequacy of information needed to list these waters, as well as the adequacy of information quantifying loadings from particular sources, is certain to be an issue in cases to come. More of an issue, perhaps, than it should be. The statute anticipates uncertainty, and calls for a margin of error in its calculations³⁴; as one EPA official has put it, just add the margin "and keep moving."³⁵ The Agency's calculations, further, should be afforded considerable judicial deference.³⁶ Deference, however, is tested in courts every day.

The likely role of TMDL litigation in the foreseeable future is to keep the process moving at a pace that is plainly uncomfortable to EPA, the states, and the regulated community,³⁷ and to watchdog the resultant listing and TMDL decisions for those that are grossly insupportable in fact, either way. In short, the classic role of environmental litigation through the years.

Further EPA Guidance

The increased scrutiny that we all face as we assist States in implementing the TMDL program requires that we do our best to help States develop approvable and defensible section 303(d) lists in 1998. (EPA Memorandum)³⁸

Pressed by the litigation, in late 1997, EPA issued two guidance documents on the TMDL program to retake the initiative. The first document clarified the scope of listed waters.³⁹ The second set first-ever schedules for the resulting LAs and requirements for their content.⁴⁰ Although couched in the language of cooperation and assistance to state agencies, each was clearly intended to move the program forward — and ahead of the courts.

[] Listing. The listing guidance came first, perhaps to take advantage of state biennial water quality reviews that were then in motion. Section 305(b) of the Clean Water Act requires the states to report the condition of their water quality to EPA every two years.⁴¹ EPA regulations require state § 305(b) reports by April 1 of even years.⁴² Although § 303(d) leaves the timing for state listings of impaired waters more open-ended (e.g., to be submitted "from time to time"),⁴³ since 1991, EPA guidance has called for state § 303(d) lists to be submitted concurrently with § 305(b) reports, which scheduled the first real show for April 1998.⁴⁴

EPA's new listing document was a remarkable-to-read exercise in loophole closing. Generically, it opened by reiterating that states must consider "all existing and readily available" data in their listings, including information "actively solicited" from other agencies, the public, and academia⁴⁵; the water-watching public, for its part, alerted for the first time nationally on the nature and importance of this drill, would participate actively and, in the end, help boost the identification of impaired waters to uncomfortable new highs.⁴⁶ The document also reiterated that waters could be exempted from listings as "expected to meet" water quality standards only if those expectations were to be met in the next two years, a relatively small window.⁴⁷ The heart of the document, however, was in its appendix,⁴⁸ which proceeded to list and then summarily dismiss excuses, difficulties, and objections to the listing of impaired waters as if they were complaints from someone who really didn't want to take this course in the first place. A sample of the issues follows, each of which can be introduced by the question: Do we really have to include … :

… waterbodies where water quality standards are in the process of being revised (read: we are busy downgrading them)? Answer: yes. The current standards are the applicable standards.⁴⁹

… waters impaired by atmospheric deposition? Yes. Controlling airborne pollution "may be difficult," but waters impaired by "all sources" should be listed.⁵⁰

… waters impaired by temperature? Yes. Ditto, and heat is a pollutant.⁵¹

[28 ELR 10419]

… waters impaired by unknown sources, or by an unknown pollutant? Yes and yes. The question is pollution; take your best shot at its cause.⁵²

… waters impaired by nonpoint sources only? Yes, without argument and "consistent with long-standing EPA policy, regulations and practice."⁵³

In its listings guidance, then, EPA meant business and made sense. Polluted waters are not going to get addressed by the TMDL program — or even thought about — unless they are acknowledged in the first place. EPA's insistence on comprehensiveness was no doubt bolstered by the fact of its express statutory authority to disapprove the lists, and its capability, in the event of a state default, to do the listing job itself.⁵⁴ Although uncertain in its science and short on good monitoring data, listing would be a relatively easy task as compared to the steps to come.

[] Total Maximum Daily Loads. In its TMDL guidance, EPA faced a tougher chore in diplomacy, semantics, and, ultimately, the law. When we come to the root cause of water quality impairment in the United States — nonpoint source pollution — federal authority under § 303 only goes so far, and then ends.⁵⁵ EPA's guidance would extend this authority fully in some respects, but hold other aspects in reserve.

EPA TMDL guidance has been an iterative process since 1991, when the Agency first began to take its § 303 responsibilities in earnest. The Agency's April 1991 Guidance for Water Quality-Based Decisions: The TMDL Process remains in force and, in many ways, the most detailed prescription available for listings and LAs.⁵⁶ Subsequent directives clarified specific aspects of the program,⁵⁷ but insufficiently to stave off citizen suits and judgments that were imposing schedules and review requirements on their own.⁵⁸ The tail continued to wag the dog because the head had not taken control. In October 1997, EPA issued a memorandum to its Regional Administrators entitled New Policies for Establishing and Implementing TMDLs,⁵⁹ which announced two new steps for the program and, for the first time in print, its enforcement options to secure state compliance.

The first step was a process, a schedule for the development of all TMDLs by all states, and a deadline. Under pre-existing guidance, the states and EPA regions had the discretion to plan for TMDLs, literally, to infinity. It was indeed the open-ended nature of the process that impelled reviewing courts to reject state and EPA proposals that extended well into the next century, and to impose tight schedules on their own.⁶⁰ EPA now announced that the April 1998 state submissions were to include not only a complete list of impaired waters and proposed TMDLs for the next two years, but also a "specific written agreement" between the state and EPA for the completion of all TMDLs within from 8 to 13 years.⁶¹ The deadline chosen here reflected the generous end of the schedules that were emerging from ongoing litigation and consent decrees.⁶² It was an inevitable compromise between the eternal need for more information and the equally pressing need to, at last, get the show on the road.

The deadline set, EPA next turned to the controversy at the heart of the process — waters polluted primarily or exclusively by nonpoint sources. Acknowledging that its current guidance was "incomplete" in this regard, the Agency stated that "[i]mplementation of load allocations for nonpoint sources in these waters is essential if we are to maintain steady progress toward clean water goals."⁶³ As discussed in an earlier Article in this series, however, § 303 provides no direct authority for EPA to implement LAs for nonpoint sources,⁶⁴ and members of nonpoint industries were taking the position that § 303 did not reach their discharges at all.⁶⁵ On the latter issue, EPA could hang firm on its previous insistence that TMDLs include load allocations from nonpoint sources.⁶⁶ On the issue of implementation, new EPA guidance would begin by recognizing the voluntary measures of the § 319 program as the "primary implementation mechanism."⁶⁷ This said, however, the goal was "to achieve TMDL load allocations" for all nonpoint sources, and "all available Federal, State and local programs and authorities" were to be used.⁶⁸ Each state, therefore, was to "describe its plan for implementing" nonpoint LAs, water segment by water segment, or by watershed, or statewide; these "implementation plans" were to be submitted as § 303(e) plan revisions either "coupled with a proposed TMDL" or as part of an "equivalent" process.⁶⁹ Each plan would provide "reasonable assurances" that the LAs "will in fact be achieved."⁷⁰ What we have here is a new thing, a TMDL-oriented plan.

Because nonpoint LAs are so central to the success of the TMDL program and, indeed, to the restoration of water [28 ELR 10420] quality nationwide, it is worth a moment to examine the strengths and weaknesses of this new announcement. A first observation is that it is not really new: EPA has ostensibly required "reasonable assurances" of implementation of its WLAs and LAs since at least 1991.⁷¹ Stated for the first time, however, is the direct application of this requirement to exclusively nonpoint waters andthe additional requirement of an implementation plan. On the plan itself, however, the requirement begins to melt. For a state to have the option of offering an implementation plan for "all affected waters"⁷² runs a real risk of describing everything and requiring nothing. For the plan to be offered as part of a "geographic planning process"⁷³ or as part of state water quality management plans is also patently amorphous and runs a serious risk of continuing nonpoint source nonmanagement. As anyone who has dealt with state water quality plans knows, they are not "plans" in a dictionary sense of the word; rather, like state implementation plans under the Clean Air Act,⁷⁴ they are more a process composed of criteria, standards, and abbreviated assessments, some published and some in file drawers, an environment in which site-specific implementation measures can lose their focus, if not simply get lost.⁷⁵ On the one hand, it is hard not to sympathize with the flexibility EPA has offered here, given its shortage of authority over nonpoint sources and its obvious need for a significant state buy-in. On the other hand, it is hard to have confidence in "reasonable assurances" from plans that can appear in several different and diffuse ways.

The last and perhaps most significant aspect of the new TMDL guidance is its articulation of the tools in EPA's closet to enforce the implementation of nonpoint source TMDLs. The memorandum outlines a scenario of gradual escalation, beginning with "constructive and focused discussion" with state managers.⁷⁶ Given the position of at least some state agencies that they have no regulatory authority over nonpoint sources at all,⁷⁷ these discussions may be very constructive, no matter how focused. The ensuing steps include:

(1) requiring a state to "update" its water quality management plan,

(2) diverting "substantial grant dollars" to states that are providing genuine nonpoint LAs,

(3) initiating additional national pollutant discharge elimination system (NPDES) review of point sources "(including minors)" affecting the listed water, and

(4) denying a state "enhanced benefits status" under § 319, subjecting it to more oversight on its § 319 programs.⁷⁸

One finishes this list asking, is this all? The measures offered are both weak and, in some cases, counterintuitive. Updating a water quality management plan is, at best, a hassle and a diversion from resources the state could better spend on pollution control. Diverting grant assistance to performing states fails to deliver money to the states that need it the most (i.e., the ones with the most significant nonpoint source pollution and the most concomitant resistance to nonpoint controls). Ratcheting down further on point sources, even "minor sources," who are now carrying the entire brunt of the cleanup, presents some obvious equity problems — although it also presents the prospect of their alliance in efforts toward nonpoint source management. As a practical matter, furthermore, this leverage disappears entirely for polluted waters that have few or no point sources to blame. As for denying "enhanced benefits status" to non-complying states, this sanction may not be denying them anything to which they, as noncompliers, would be entitled anyway; at its strongest, it adds a hassle factor for nonpoint planning.

Which begs the question "could EPA do more?" The answer is clearly yes. The answer may also be that everybody already knows it, so EPA did not need to say it. The first unstated authority is EPA's ability to deny new point source permits in water quality limited (i.e., TMDL-required) waters. By long-standing EPA regulation, no new source may "cause or contribute to" the violation of water quality standards⁷⁹; if a water segment is already in violation, any "actually detectable or measurable" contribution would seem to be prohibited.⁸⁰ If a TMDL has been prepared, the new source must demonstrate that there is a remaining LA for its discharge and that all other dischargers are on compliance schedules to meet water quality standards.⁸¹ New sources, then, become leverage both for the preparation of TMDLs and their effective implementation.⁸² Corollary authority [28 ELR 10421] exists for modifying permits for existing sources based on new information, such as, arguably, the fact of water quality impairment or TMDL allocations, that would have justified different permit considerations and was not available at the time of issuance.⁸³ This regulation would seem to enlist the leverage of all point sources, new and existing, following the new information underlying the listings of April 1998.

A third unstated authority is EPA's ability to revoke a state-delegated program for failure to develop, implement, or enforce TMDLs.⁸⁴ This leverage is both more and less than meets the eye. The track record of EPA in revoking state delegated programs — or even threatening to revoke them — is minimal.⁸⁵ In the interests of federalism and political survival, EPA has steered a wide berth around underperforming state programs, some openly defiant of EPA requirements.⁸⁶ The threat in the case of unperformed nonpoint TMDLs is even more illusory, as it could involve EPA in permitting and land use decisions beyond its resources to administer or politically to survive. On the other hand, viewing the threat from a state perspective, EPA has competently administered many full CWA programs for years⁸⁷; sufficiently provoked, it could revoke a delegated program and do so again. The loss of such a program to a state is more than the loss of financial or other support; it is the loss of prestige and political power, and these are high stakes to risk in a game of chicken.

This last observation is one that EPA can best leave unsaid. To mention revocation would only alienate the states whose cooperation is badly needed to make nonpoint source management a reality. In a legal culture full of "gorillas in closets" that are rarely seen, but radiate their presence nonetheless, this gorilla, too, is there.

[] From Guidance to Regulation. EPA has been feeling its way with TMDL guidance to its regions and the states for the past seven years. Occupied by other priorities in the 1970s, the Agency began with regulations for a light and voluntary approach,⁸⁸ which met with a light and voluntary response.⁸⁹ As the heat intensified in the early 1990s, EPA responded with memoranda and guidance documents that could be issued quickly and be insulated from outside enforcement and judicial review. EPA's 1991 Guidance for Water Quality-Based Decisions: The TMDL Process begins:

This document provides guidance only. It does not establish or affect legal rights or obligations. This guidance may be reviewed and revised periodically to reflect changes in EPA's strategy for the implementation of water quality-based controls, to include new information, or to clarify and update the text. Decisions in any particular case will be made by applying the Clean Water Act and implementing regulations.⁹⁰

By spring 1998, § 303(d) had evolved into an elaborate federal program with increasingly specific requirements supported by very sketchy regulations and highly detailed informal guidance. With the close of the FACA process, basic principles of administrative law oblige EPA to pursue rulemaking that codifies the requirements of the program⁹¹ and, as importantly, provides the basis for citizen enforcement of these requirements, even against the Agency itself.⁹² The degree of specificity in its new regulations will be a litmus test for EPA. On the one hand, looking at a horizon of citizens keenly interested in TMDLs and battered by their enforcement suits, the Agency, and in particular its lawyers, would have an understandable temptation to write regulatory mush and then rest on "agency discretion" to defend its future action, or inaction. The result would be that EPA would start winning its lawsuits and losing on the timing, content, and enforcement of TMDLs. Citizen suits are part of the Agency's backbone, and the more difficult a program is politically, the more necessary their leverage becomes. And few programs are going to be more difficult to implement politically than nonpoint TMDLs.

The FACA Committee

The Committee, however, could not agree as to whether waters impacted only by nonpoint sources should be included on the § 303(d) list…. Some Committee members … observed that … the § 303(d) listing requirement may be duplicative and unnecessary. They were also concerned that TMDLs might result in (or, in and of themselves constitute) a mandatory water-quality based limit that could be enforced against nonpoint sources. (Facilitator's Report, FACA committee)⁹³

[28 ELR 10422]

In late 1996, seeing the storm on the near horizon and, indeed, breaking around its ears, EPA formed a FACA committee to attempt to arrive at a consensus on the direction and requirements of the TMDL program.⁹⁴ Members included state water program officials, local officials, a Native American representative, agribusiness, the timber industry, point source industry, municipal sewage plants, and environmentalists; officials of EPA, the U.S. Natural Resources Conservation Service (U.S. Department of Agriculture (USDA)), and the U.S. Forest Service served as ex officio members.⁹⁵ All the players were in the room. Their meetings were open to, attended by, and addressed by other members of the public. Their work was published, meeting by meeting and workgroup by workgroup, on the Internet. Their gradual consensus and hardening points of difference were described, as they evolved, in an iterative facilitator's report.⁹⁶ In all, this was a remarkably transparent process, and, given the diversity of interests represented, it led to a rather heartening convergence of viewpoints. Up to the real point: nonpoint source pollution controls. On this central issue, the FACA committee could only deliver part of the goods.

The final committee report contains more than 100 pages of recommendations, analysis, and appendices, and merits separate treatment — which it will surely receive as EPA moves toward rulemaking.In the main, the report and recommendations reinforce emerging EPA guidance, and the cross pollination from the FACA committee to EPA's latest memoranda is obvious even in the language used for EPA's new listing and TMDL prescriptions.⁹⁷ The committee reached considerable agreement on the need for better information,⁹⁸ for public and "stakeholder" participation,⁹⁹ for comprehensive listings,¹⁰⁰ for specificity in TMDLs,¹⁰¹ for an "implementation plan,"¹⁰² for "reasonable assurances" that the plan would be implemented,¹⁰³ and on mechanisms for monitoring and implementation.¹⁰⁴ The report offered consensus suggestions for adaptive management,¹⁰⁵ for "extremely difficult problems" such as atmospheric deposition and flow obstruction,¹⁰⁶ and even a detailed "hierarchy approach" to TMDL approval by EPA.¹⁰⁷ The honeymoon ended, however, at TMDLs for nonpoint sources, and the enforcement of TMDLs generally. Simply put, nonpoint sources wanted no part of a program that would be enforced.

The disagreement took two forms. First, to timber and agriculture representatives, waters impaired primarily or entirely by nonpoint sources should not be listed under § 303(d)(1), but rather under the voluntary and informational provisions of § 303(d)(3) or § 319¹⁰⁸; listing under § 303(d)(1) would be "duplicative" and "unnecessary."¹⁰⁹ As a second line of defense, they argued that the implementation of all TMDLs should not come through § 303(d) approvals but, rather, through the approval of state water quality plans under § 303(e).¹¹⁰ Both arguments threaten to emasculate the statute. To deflect nonpoint TMDL planning to the information-only framework of § 303(d)(3), or to the voluntary world of § 319, is to ensure that it will not bear fruit in the lifetimes of anyone now living; it also contradicts a sensible reading of § 303(d)(3), which seems directed to waters that currently meet water quality standards.¹¹¹ The argument for implementation through § 303(e) is at least facially better grounded in law, but is so clearly motivated by the fear that § 303(d) requirements could be enforceable ("there were some concerns that reliance on § 303(d) could lead to judicial enforcement of TMDL implementation plans in unexpected or unintended ways")¹¹² and by the confidence that implementation through § 303(e) will not ("[t]he extent and timing of EPA review of TMDL implementation plans submitted under § 303(e) is less clear [than review under § 303(d)]"),¹¹³ that this motive alone should be its undoing.

The issues left open by the FACA committee then — and through no lack of trying — are fundamental to the success of the TMDL program. If nonpoint waters are not included, the program is wasting everyone's time; EPA has long insisted [28 ELR 10423] on their inclusion,¹¹⁴ and the issue is in all likelihood headed for court. The mechanics for approving an implementation plan, however, are now before EPA to decide. Here, EPA should be guided more by the purpose of the statute and what can work to effectuate that purpose than by what will make nonpoint sources happy because it won't. To function at all, approvals under § 303(e) would require EPA to reinvent the same site-specific review for particular water bodies now called for under § 303(d)¹¹⁵; even then, § 303(e) approvals would not necessarily be based on regulatory criteria or subject to citizen suits for their timeliness or content.¹¹⁶ At bottom, the unresolved issue is enforcement. At the present time, in its latest, 1998 guidance, the Agency has left the doors to both § 303(d) and § 303(e) open,¹¹⁷ doubtlessly hoping that the FACA committee would ease its choice. The FACA committee worked hard but did not get there. EPA will yet have to make its most important call.

The Federal Family

TMDLs operate in an environment too hostile to succeed on their own. They face reluctance, to put it mildly, from point and nonpoint sources, state regulatory agencies, and state and federal land management agencies; they face serious shortages of funding; and they face a host of disincentives to conserve soil and water found in tax policies, property law, water law, and commodity support and subsidy programs. To expect desklevel, state water quality personnel to deal effectively with the timber industry, hog farmers, cattle ranchers, real estate developers, irrigation return flows, agricultural chemicals, and the rest of the nonpoint world without assistance from other state and federal agencies is to expect the impossible.

Over the past year, other federal programs began to rally around the TMDL approach in ways that give it at least a hope of accomplishing the job ahead. The rally was lead, or at least cheer-led, by the President, who announced a new Clean Water Action Plan focused primarily on nonpoint source pollution. It was supported by the acceleration of several other trends, including the increased treatment of diffuse sources (e.g., animal feedlots and municipal stormwater) as point sources under the Clean Water Act, the treatment of public lands uses (e.g., grazing and timber) as "dischargers" under the Act, and the emergence of quasi-technology standards for nonpoint source pollution under the Coastal Zone Management Act. While each, again, is properly the subject of its own article, if not its own library, they are summarized here for purposes of assessing their impact on the TMDL program.

[] The Clean Water Action Plan.

In most cases, the development of TMDL and wasteload allocations for specific impaired waters within the watershed will form the core of the Watershed Restoration Action Strategy.¹¹⁸

In November 1997, the Vice President of the United States announced a strategy for addressing the chronic and remaining pollution of the nation's waters, primarily from nonpoint sources.¹¹⁹ From a political standpoint, the Administration was recapturing the initiative in pollution control. Just two years earlier, Congress had taken a very direct run at the Clean Water Act with amendments that, inter alia, would have relaxed its already soft provisions for nonpoint source control.¹²⁰ After a highly visibly defeat on this issue, Congress had sidestepped the Clean Water Act and had gone off in search of other prey. Equally telling, from the standpoint of an Administration that reads its public opinion polls, was that when the Clean Water Action Plan emerged six months later, it was announced not by the Vice President but by the President of the United States.¹²¹ Whatever collateral damage the Presidency was suffering during this time, the announcement would focus several agencies of the United States on nonpoint source pollution.

The plan was refreshingly frank about the problem. Although sprinkled with success stories of waters that had been restored across the country, it presented the "bottom line" assessment that 50 percent of the nation's watersheds were polluted, including some 1,500 water bodies that did not meet water quality standards.¹²² Only 16 percent of U.S. watersheds had "good" water quality.¹²³ Turning to the causes, in a paragraph entitled Polluted Runoff is the Most Important Source of Water Pollution, the plan continued, "[n]ationally, agriculture is the most extensive source of water pollution, affecting 70 percent of impaired rivers and streams and 49 percent of impaired lake areas."¹²⁴ Based on state identification of water pollution sources, the plan quantified their relative contributions as follows:

Rank	Rivers	Lakes	Estuaries
1	Agriculture	Agriculture	Industrial Discharges
2	Municipal	Unspecified	Urban Runoff/
	Point Sources	Nonpoint	Storm Sewers
		Sources
3	Hydrologic	Atmospheric	Municipal Point
	Modification	Deposition	Sources
4	Habitat	Urban Runoff/	Upstream Sources
	Modification	Storm Sewers
5	Urban Runoff/	Municipal	Agriculture125
	Storm Sewers	Point
		Sources

[28 ELR 10424]

To environmentalists and, one supposes, no small number of regulated point sources, these admissions at this level of government were a breath of fresh air. To at least some nonpoint dischargers, on the other hand, the President had "declared war on the farmer."¹²⁶

The proposed solutions would be more numerous. A first was money: an increase of more than half a billion dollars in nonpoint pollution control spending for fiscal year 1999, and an increase of over $ 2.3 billion over the next five years.¹²⁷ A second was a bevy of existing programs of EPA, the USDA, the U.S. Department of the Interior, the U.S. Army Corps of Engineers, the U.S. Forest Service, and the National Oceanic and Atmospheric Administration (NOAA) that would be accelerated, reinforced, and given specific numerical goals.¹²⁸ The USDA would protect miles of riparian habitat and set aside acres of erodible lands; together with the Corps' permitting program, it would achieve a net gain in wetlands of 100,000 acres a year.¹²⁹ While few of the announced efforts were entirely new, and some not new at all, their emphasis and goals and the coordination offered were certainly new; like President Bush's "no net loss of wetlands" initiative several years before, they held the potential for at least legitimizing the effort within the federal family.

The plan filled EPA's plate, too, with new and old helpings of regulatory work, including quasi-point source initiatives described below and the long-overdue task of developing water quality criteria for nutrients.¹³⁰ In the middle of the plate, however, was EPA's old friend "watershed management," to be effectuated by new watershed restoration action strategies.¹³¹ This strategy would bottom — late in the plan and at a point probably reached only by the most attentive or suspicious reader — on … TMDLs.¹³² Indeed, TMDLs would "form the core" of the strategy. A watershed approach "can be a smarter, more effective and cost-efficient way to implement TMDLs," the report went on encouragingly, creating an opportunity to "bundle" TMDLs and "strike an appropriate balance" among dischargers.¹³³ In this construct — and to this author, appropriately — TMDLs are not an adjunct to watershed planning; rather,they are the basis of watershed planning, not because they are scientifically bulletproof, comprehensive, or efficient, but because they are objective, measurable, and the only approach so far that can be enforced by law.

[] From Nonpoint to Point Source Regulation. The great axiom of the Clean Water Act is that point sources are strictly regulated while nonpoint sources, in the delicate phrase of one commentator, are "immune from important features of the Act"¹³⁴ such as effluent standards, permits, and enforcement (i.e., those features that have made the NPDES program work). The history of nonpoint source pollution control since 1972 is of an attempt, to date largely unsuccessful, to find replacements for these features through voluntary, local programs. The rationales offered for treating nonpoint sources separately under the Act include:

(1) the alleged "number and variety of nonpoint sources";

(2) the "site-specific nature" of the pollution; and

(3) the "lack of known control technologies."¹³⁵

On reflection, none of these reasons are terribly convincing, because:

(1) we have a great number and variety of point sources as well (several hundred major industrial categories and subcategories, and more to come);

(2) each industrial discharge, too, has site-specific effects on its receiving water (effects that are irrelevant to the setting of technology-based guidelines); and

(3) the control technologies for nonpoint pollution (e.g., shelter-belts, nutrient caps, retention ponds) are anything but unknown, complex, technologically difficult, or even very costly.

In truth, we do not avoid regulating nonpoint source pollution because we are unable to figure out how to do it. Rather, we have deferred to the myth that its impacts are essentially local and of secondary importance, as we have deferred to legislatures dominated by rural constituencies unaccustomed to any regulation and ready to fight. Recently — albeit with glacial slowness — both the myth and the dominance have begun to melt, reopening the question of nonpoint source controls. One answer to the problem is simply to treat a greater number of dischargers as point sources, bringing them into the operational features of the Act.

EPA and Congress have been wrestling with the application of the NPDES program to agriculture, silviculture, and land-based pollution since the adoption of the Act in its modern form.¹³⁶ In 1973, the Agency adopted a definition of point sources that included runoff collected or channeled virtually in any way, but then proceeded to exempt discharges from all silviculture, all urban storm sewers, and all [28 ELR 10425] but the largest agricultural operations.¹³⁷ These exemptions were immediately challenged by a citizen suit and rejected by both the federal district and appellate courts of the District of Columbia.¹³⁸ Facing EPA arguments — similar to those noted above — that nonpoint sources were too numerous, diffuse, and difficult to regulate, these courts suggested the use of alternative permit conditions and general permits, and concluded with the inspiration that "[i]magination conjoined with determination will likely give EPA capability for practical administration. If not the remedy lies with Congress."¹³⁹

Congress did not wait to see. Firmly held by the myth that nonpoint source pollution was a local affair, Congress amended the Act in 1977 specifically to exclude irrigated agriculture from point source regulation,¹⁴⁰ and in 1987 went further to exempt all agricultural stormwater discharges.¹⁴¹ As a result, with the exception of concentrated animal feedlots, agricultural pollution was exempted from the Act even when it came from and through discrete collection systems. The scope of this exemption went largely unchallenged until 1994, when the Second Circuit ruled in Concerned Area Residents for the Environment v. Southview Farm¹⁴² that a large dairy farm, in its entirety, was a point source under the CWA.¹⁴³ The court reasoned in the alternative that the machines that spread manure on the farm, and the drainage system conveying these wastes to navigable waters, were point sources as well.¹⁴⁴ This ruling contained the seeds of reclassifying a great deal of mechanized agriculture as point sources, an invitation that EPA has not yet accepted.

Instead, under the impetus of other CWA amendments in 1987,¹⁴⁵ EPA has proposed to develop standards for a number of diffuse sources. The most important of these proposals, from the standpoint of abating serious pollution, involve animal feeding operations (AFOs) and municipal storm sewers. Each of these programs could serve to take considerable weight off of remaining nonpoint programs, and could give the TMDL approach for polluted waters the additional leverage that comes from the availability of point source-like permits and standards.

* Animal Feeding Operations and Concentrated Animal Feeding Operations.

On June 21, 1995, the eight-acre manure lagoon at Oceanview Farms in Onslow County burst through its dam. What followed was seen on news reports around the world, as 25 million gallons of excrement surged over a road, through a neighboring tobacco field and into the New River. The odoriferous tide was two feet deep and flowed for over two hours, ending up in the river, where it killed "virtually all aquatic life in the 17-mile stretch between Richlands and Jacksonville. When it was over, the New River had been the victim of a spill more than twice the size of the oil spill that followed the wreck of the Exxon Valdez. (Analysis of the North Carolina Hog Farming Industry)¹⁴⁶

EPA recognizes that its existing regulatory programs related to animal waste management are not being implemented consistently and have not kept pace with evolving technologies or industry practices that have altered the type and magnitude of related adverse environmental and public health impacts. (EPA Animal Feeding Operations Strategy)¹⁴⁷

As noted above, the CWA definition of point sources has included from the outset discharges from concentrated animal feeding operations (CAFOs). In 1976, EPA issued regulations for CAFOs that excluded all but industrial-sized operations: only CAFOs with more than, for example, 1,000 slaughter cattle, 2,500 hogs of more than 55 pounds apiece, or 30,000 chickens (or 100,000 chickens if a facility used "continuous overflow watering") were automatically included¹⁴⁸; smaller operations would be included if they discharged directly into U.S. waters or could be shown to impair water quality on a site-specific basis.¹⁴⁹ In theory, the requirement was "zero discharge" based on interim storage in lagoons and land application.¹⁵⁰ Overflow from waste lagoons during heavy storms was exempted, as was the practice of spreading animal wastes on land.¹⁵¹ This done, EPA turned to other business.

In practice, the EPA regulations were underinclusive and ineffective. Of the nearly half a million farms with confined (not pasture) feedlots in America, only 6,600 contained sufficiently high numbers of animals to require a permit; another 32,000 mid-sized operations were potentially subject to regulation depending on their method of discharge.¹⁵² As of 1995, 20 years after EPA's CAFO regulations were promulgated, only 25 percent of even the largest operations had NPDES permits.¹⁵³ Meanwhile, the industry was changing toward ever greater concentration of animals on smaller landscapes. Between 1991 and 1996, hog production increased 8 percent while the number of facilities dropped by 20 percent.¹⁵⁴ As the size of individual operations grew, traditional methods of waste management failed; lagoons overflowed, adjacent fields rejected overloads from increased land spreading.¹⁵⁵ The pollution from these wastes, always recognized as oxygen-depleting, was beginning to be seen as toxic as well.¹⁵⁶ In 1993, an EPA workgroup recommended [28 ELR 10426] additional "guidance" on CAFOs¹⁵⁷; a 1995 U.S. General Accounting Office report followed suit.¹⁵⁸ At about the same time the manure hit the fan on the New River delta of North Carolina, and then the Chesapeake Bay, in ways that could no longer be ignored.

The breakout of pollution from the mellifluously named Oceanview Farms was not an isolated event. Coastal North Carolina, home to a hog industry so new and so vigorous that it outsold tobacco as the state's number one crop,¹⁵⁹ began experiencing a "devastating series" of hog spills and fish kills.¹⁶⁰ By that time, pigs were outnumbering people by as much as 34 to 1 in some counties and, with pigs producing from two to four times the waste per day of an average human adult, the fuel for more contamination was on hand.¹⁶¹ The 1995 Oceanview disaster killed an estimated 10-125 million fish and closed 364,000 acres of commercial fishing grounds; it was accompanied by several other spills the same summer of more than a million gallons each.¹⁶²

One summer later along the Chesapeake Bay, Maryland officials closed a 17-mile stretch of the Pocomoke River where 30,000 fish had died and fishermen and scientists had become ill from exposure to the water.¹⁶³ Shortly thereafter, officials closed another waterway with fish infected by Pfiesteria piscicida, the so-called cell from hell, which flourishes in high-nutrient, low-oxygen conditions.¹⁶⁴ These conditions led investigators to the chicken industry on Maryland's Eastern Shore which, in the Pocomoke region alone, produces more than 100 million friers a year for Perdue and other large corporations.¹⁶⁵ Fearing the worst from an outbreak on the Chesapeake so close to Washington, D.C. — that EPA might actually come forward with regulations — the American Farm Bureau quickly announced that it advocated "nonregulatory solutions before you let the federal gorilla out of the closet and step on these folks."¹⁶⁶ But the events were too stark and too well publicized¹⁶⁷ and, for once, EPA was not inclined to listen.

In March 1998, on the heels of the President's Clean Water Action Plan, EPA announced a Draft Animal Feeding Operation Strategy directing a new series of initiatives at the problem,¹⁶⁸ led by politically correct offers of assistance to states on research, monitoring, enforcement, and watershed planning.¹⁶⁹ The strategy then moved to the heart of the matter: expanded permitting and more comprehensive effluent guidelines.¹⁷⁰ Beginning with CAFOs in the most heavily impacted watersheds, EPA and the states would place all large CAFOs under permit within seven years, including facilities previously exempted as "no discharge" and treated discharge operations.¹⁷¹ Adopting the flexibility suggested 20 years earlier by the D.C. Circuit,¹⁷² these permits would include individual permits for the largest CAFOS, "watershed-specific" permits for groups of facilities in a given area, and general permits for other operations; these permits would include "appropriate conditions" for the land application of animal wastes as well.¹⁷³

The permits would be based on new federal effluent guidelines, issued informally on "key permitting issues" by late 1998,¹⁷⁴ and as formal best available technology (BAT) and new source performance standards by December 2001.¹⁷⁵ The development of these standards would be a collaborative effort with the industry and the public — as effluent guideline development has become in the 1990s, but there is no mistaking the assertion of federal authority in this new strategy. And there is every reason to expect that, once these permits and guidelines are in place, pollution loadings will drop as dramatically as they have in every other NPDES-regulated category of the CWA.

When this comes to pass, discharges that in the aggregate contributed to an estimated 13 percent of the polluted rivers of the country will move, from what has largely been ad hoc water quality-based treatment, to federally prescribed management and technology practices. It bears remembering in all of this that EPA and the states have had, all along, the authority and duty to regulate AFOs and CAFOs individually and more strictly on the basis of water quality standards, whenever they found water quality impairment; North Carolina, furthermore, was and is considered to be a leader among the states in water pollution control.¹⁷⁶ Yet water quality-based controls did not happen, not in North Carolina until after a series of disasters, and not in several other states to this day.¹⁷⁷ All of which says something about both the ability of states to control local industry and the ability of anyone to control discharges by water quality standards. When things got serious, we turned once again to more extensive BAT.

* Stormwater.

Using Washington DC as an example, the group found that the amount of lead in the city's runoff in 1988 [28 ELR 10427] (26,000 pounds) was nine times the total amount discharged from all of Virginia's factories in 1987 (2,900 pounds). (BNA Environment Reporter)¹⁷⁸

Sediment loadings rates from construction sites are typically 10 to 20 times that of agricultural lands, with runoff rates as high as 100 times that of agricultural lands, and typically 1,000 to 2,000 times that of forest lands. (EPA)¹⁷⁹

EPA has approached stormwater regulation with the same awareness-cum-trepidation that it has agriculture, and for many of the same reasons. The Agency faces cash-strapped municipalities without either the threat of injunction which it holds over industrial sources¹⁸⁰ or the leverage of funding it provides for municipal sewage treatment.¹⁸¹ Stormwater remedies, further, while often obvious and available, smack of the kind of land use control of which no federal agency dares stand accused. On the other hand, urban runoff is major pollution, in coastal areas it is the dominant source of water pollution,¹⁸² and it is discharged for the most part through pipes, drains, and other discrete conveyances (i.e., point sources under the CWA).¹⁸³

EPA's first response was to duck, and in 1973 the Agency moved to exempt stormwater runoff sources from point source regulation as too diffuse, too impractical, too numerous, etc.¹⁸⁴ When this approach was invalidated, stormwater management fell into a cycle of inconclusive rulemakings, litigation, and paralysis for the next 15 years.¹⁸⁵ Finally, in 1987, prompted by increasing information on the volume, contamination, and outright toxicity of urban runoff, Congress stepped in.¹⁸⁶

The choice before Congress was between imposing industrial-type point source standards or continuing the limbo of nonpoint source management. Its decision was to advance the NPDES approach for major sources as far as possible, leaving some flexibility for smaller municipal stormwater systems. Industrial stormwater would henceforth be regulated through NPDES permits under effluent limitation guidelines¹⁸⁷ and a new, hybrid, technological standard, "maximum extent practicable," which was to include "management practices, control techniques and system, design and engineering methods" and "such other provisions" as EPA found appropriate for runoff pollution control.¹⁸⁸ On a slower track, EPA was to develop regulations for smaller communities that "may include performance standards, guidelines, guidance, and management practice and treatment requirements, as appropriate."¹⁸⁹ Stormwater runoff was thus moved from the de facto nonpoint to the permit world, with the apparent expectation that it would be made subject to explicit engineering and best management practices (BMPs).

For industrial sources this expectation is more or less coming to pass. EPA's program for industrial runoff is administered through individual (best professional judgment), group (multi-sector), and general permits, each based on BMPs.¹⁹⁰ General permits (available to smaller operations) are based on site-specific BMPs submitted by the applicant, while group permits (available, as an option, to larger activities by industrial category) are based on a list of BMPs developed, through rulemaking, with each affected industry.¹⁹¹ By whatever route, the end result is permits with a relatively specific set of "good housekeeping practices" ranging from required ground cover to the separate handling of construction debris.

For municipal sources, EPA has moved with more caution on a problem that by its own data is responsible, inter alia, for nearly 50 percent of the impairment of the nation's estuaries.¹⁹² By statute, EPA was to address large storm systems first and, in 1990, the Agency finally promulgated its Phase 1 regulations for communities of over 100,000 inhabitants.¹⁹³ The regulations consist largely of suggested practices (e.g., street sweeping) and public education campaigns (e.g., the encouragement of reduced lawn watering), coupled with discharge and water quality monitoring.¹⁹⁴ Municipal runoff control mechanisms well described in the literature — such as permeable surfaces, retention ponds, and greenbelts — are nowhere in view.¹⁹⁵ Actual Phase 1 [28 ELR 10428] permits, accordingly, consist of descriptions of such ongoing municipal operations as cleaning storm drains, and such public awareness initiatives as, in the case of the city of New Orleans, the Storm Drain Stenciling Project (e.g., "Drains to Lake"), on the apparent thesis that the large slugs of toxins that characterize the "first flush" from the city streets during each rainfall are the product of individual and uninformed toxic dumpers.¹⁹⁶ EPA has defended the insubstantial nature of its stormwater permit for the city of New Orleans by asserting that "the single most important pollution prevention practice for [municipal] permits is public education" which will "help create the changes that need to occur in how citizens view and approach potential urban runoff pollution sources and activities."¹⁹⁷ For the moment, then, and whatever Congress might have meant for the application of "maximum extent practicable" control standards to large municipal stormwater sources, we have business as usual, plus education, plus monitoring.

Meanwhile, EPA has also been under a statutory requirement and court orders to expand its coverage of runoff sources under Phase 2 regulations for smaller municipal systems and construction sites.¹⁹⁸ Stimulated, perhaps, by the President's Clean Water Action Plan, the Agency has now proposed new Phase 2 stormwater rules that will include permitting for an estimated 3,500 more municipalities and another 100,000 construction sites per year. The key word here is "permitting" because, although the statute does not specify permits as a necessary ingredient of EPA's Phase 2 program,¹⁹⁹ and although states have lobbied for the adoption of "non-NPDES" state programs instead,²⁰⁰ EPA has chosen to stay with a permits approach because permits are "enforceable under the CWA" and, more tellingly, "would ensure that citizens could participate through the permit issuance process as well as participate in enforcement proceedings"; a more frank acknowledgement of the need for citizen suits would be hard to find.²⁰¹ The net thus widens; it remains to be seen what will happen with the catch.

The first thing that will happen is, at least for Part 2 permits, potentially stronger BMPs. EPA's proposals for Phase 2 put out for public discussion contain significantly more substantive criteria, such as mandatory standards for "post-construction stormwater management in new development and redevelopment" which may require "limiting growth to identified areas," "minimizing impervious area," "maintaining open space," and "structural BMPs" (e.g., retention ponds, wetland falters, and porous pavement).²⁰² Here, at least we are beginning to talk about real pollution controls.

EPA's application of water quality standards is more ambiguous. The Agency has long insisted, indeed from the start, that municipal stormwater sources comply with water quality standards.²⁰³ The question has always been: how? Given the highly fluctuating rates and contamination levels of stormwater discharges, water quality standard-based effluent limitations would be a nightmare to develop, and the Agency turned to BMPs.²⁰⁴ EPA would impose more controls only where warranted by "cost considerations and water quality effects,"²⁰⁵ and its current Phase 1 regulations and permits require no more. The Phase 2 proposals continue this approach, but end the story as follows: "If additional specific measures to protect water quality were imposed, they would likely be the result of an assessment based on TMDLs, or the equivalent of TMDLs, where the proper allocations would be made to all contributing sources."²⁰⁶ Which returns us to TMDLs.

Eleven years after the 1987 CWA amendments calling for stormwater management under federal permit, EPA has effectively asserted sufficient jurisdiction to move stormwater pollution into the point source world through largely voluntary, largely preexisting, BMPs — plus education, plus monitoring to determine the need for more serious permit requirements in the future. TMDLs will provide additional, earlier pressure to get on with the next phase of the stormwater program. Conversely, although polluting cities may be able to fudge their discharge monitoring sufficiently to delay more serious requirements under the stormwater program, if they are discharging into impaired waters then § 303(d) assessments will bring the problem to their doors anyway, and the obligation to join industrial dischargers, municipal sewage treatment works, agricultural sources, and the rest of the TMDL bazaar.

[28 ELR 10429]

[] Grazing and Timber on Public Lands.

Nonpoint source water pollution on the public lands of the West is a problem of significant proportions, due to the immense land area involved, the ubiquity of the pollution sources, and the vulnerability of many surface waters …. The western states themselves have identified grazing (a subcategory of agriculture) as the [nonpoint source] category having the greatest impacts on the quality and beneficial uses of their waters. (Debra L. Donahue, The Untapped Power of Clean Water Act Section 401)²⁰⁷

In 1975 a severe storm hit the Mapleton District. A Forest Service survey of 70 percent of the district showed 245 landslides. Of these slides, 9 percent were natural events, 14 percent were road related, and 77 percent were in clearcut units apparently unrelated to roads or landings …. In 1982 a Forest Service aerial photo survey showed fifty road-related and seventy-nine clearcut related landslides. These slides scoured 14.29 miles of streams. (U.S. District Court for the District of Oregon)²⁰⁸

While we tend to think of the impacts of public lands grazing and timber harvest as alterations of the landscape and its wildlife populations, they have altered aquatic landscapes and aquatic life as dramatically and, in some cases, more irreversibly, through the degradation of water quality.

Cattle grazing has long been recognized as a primary source of pollution in western waters, which tend to be scarce to begin with and lack the volume to flush, mix, biodegrade, and otherwise accommodate the loads of oxygen demand and sediments input directly from livestock manure and indirectly from destabilized riparian zones.²⁰⁹ There are 2 million livestock across the West today, and there have been as many as 20 million at times in this century.²¹⁰ They congregate at, on, and in any water source at hand. An adult cow produces more than 80 pounds of manure per day, nearly 16 tons per year.²¹¹ In all bovine innocence, they trample streambanks into mudslides and pound their watering holes into so-called sacrifice areas and water gaps.²¹² With the loss of vegetation comes the loss of shade, the on-set of thermal pollution and serious algae bloom.²¹³ Overgrazing, even on lands miles distant from western water-courses, leads to accelerated runoff from periodic storms, further degrading water quality.

All of which has been documented for years. Wyoming, for example, reports sediments and nutrients, for both of which grazing is the primary source, as the number one and number two causes of water quality impairment in the state.²¹⁴ None of which has made any significant difference in western range management. To at least some western cattlemen, these concerns are a product of "ripariopsychorrhea," described as a "fictional mental disease with scatological overtones."²¹⁵ Western legislators and water quality regulators have responded in kind by allowing broad variances for grazing, logging, and similar activities or by exempting them from controls altogether.²¹⁶ Federal grazing managers, until recently, have been equally forgiving.

On the steeper slopes of national and other forests, timber harvests, almost uniformly by clearcut, have produced even more spectacular, if less ubiquitous, damage. Ironically, it was the effects of wildcat logging at the end of the last century that propelled the President and Congress toward national forest management, with the explicit purpose of protecting forest watersheds from further damage.²¹⁷ Despite the Organic Act of 1897²¹⁸ and the more recent National Forest Management Act,²¹⁹ however, the pressure for high timber yields drove the annual cut on public lands alone from 1 billion to more than 12 billion board feet per year,²²⁰ a harvest that could only be attained by practices that stripped the land of all trees and cover. Western mountainsides began collapsing into their streambeds.²²¹ Logging roads became gullies, conduits for silt and mud. With the increase in turbidity, salmon and other fisheries were eliminated from watersheds, then basins, then entire states.²²² Some runs have gone extinct, and others are perilously close to following suit.²²³ For a number of below-quality rivers of the West, logging roads are the only significant man-made source of pollution.²²⁴ On the steep slopes of the Pacific [28 ELR 10430] Range, clearcuts continue to cause blowouts and mudslides that take houses and small communities right into the riverbed.²²⁵ The more chronic and long-term damaging effects to water quality, of course, come with every rain.

Forest managers have been no more willing to come to grips with these effects on water resources than have their counterparts on the public range. As with the impacts of cattle on riparian areas, these impacts are well known and admit of rather obvious remedies: stepping cattle and clearcuts back from the watercourses and adopting a baseline for native groundcover. However well known, these remedies are firmly opposed by grazing and timber companies, even when the survival of species as cherished as the salmon are at stake.²²⁶ The strength of this opposition does not lie solely in economics, the lack of reasonable alternatives, or even the importance of these industries which contribute a surprisingly small and dwindling share to even their regional economies. Like agriculture and its image of the small farmer, timber and grazing are supported by images of Paul Bunyan and John Wayne that have dazzled legislators and intimidated regulators at all levels of government; outright violence against managers who attempt to impose environmental requirements is not unknown.²²⁷ The Forest Service itself seems intimidated. While the agency has regulations explicitly requiring mining operations, for example, to comply with state water quality requirements,²²⁸ one cannot find any similar regulation requiring timber operators to obey state water quality standards.²²⁹ The imposition of this requirement on the timber industry, then, would have to come from the outside. Which is exactly what is happening.

The pieces of this puzzle are not yet fully assembled. We have legal theories, pending lawsuits, cases on administrative appeal, cases on federal circuit appeal, and Administration initiatives all contributing to a picture that could have a dramatic effect on TMDLs for western waters. One theory would extend the reasoning of Southview Farm²³⁰ to consider grazing "sacrifice areas," for example, and forest logging roads as point source dischargers under the Clean Water Act.²³¹ While this theory percolates untested, a more modest theory emerges that even if these activities are not point sources under the Act, they are dischargers and therefore constitute activities requiring state certification under § 401 that their impacts will not violate state water quality standards.²³² The § 401 thesis has had a mixed reception, accepted in Oregon Natural Desert Ass'n v. Thomas²³³ (involving grazing permits) and rejected in Idaho Conservation League v. Casewell²³⁴ (involving the construction of two national forest roads). The split here centers on whether the term "discharge," in the context of § 401, is itself limited to point sources and "nonpoint sources with discrete discharge points," or whether it includes other means of conveyance as well.²³⁵ Reportedly, following the Oregon Natural Desertopinion, EPA began drafting regulations for grazing impacts²³⁶ which are now on hold, pending appeal of the case to the Ninth Circuit.²³⁷

Meanwhile, a third and more direct approach has arisen under § 313 of the Act, which mandates that all federal agencies "having jurisdiction over any property" or engaged in "any activity" that results or may result in "the discharge or runoff of pollutants" shall be "subject to, and comply with all … State … requirements … and process and sanctions respecting the control and abatement of water pollution" in the same manner as nongovernmental entities.²³⁸ This provision has led to declarations from the Ninth Circuit in Northwest Indian Cemetery Protective Ass'n v. Peterson²³⁹ and Marble Mountain Audubon Society v. Rice²⁴⁰ that the Forest Service must comply with state water quality standards. These holdings, both involving federal timber sales, were reinforced in Oregon Natural Resources Council v. U.S. Forest Service,²⁴¹ which found that a citizen suit over yet another timber sale based on § 313 would lie under the Administrative Procedure Act (APA),²⁴² even though it would not qualify under the citizen suit provision of the CWA itself.²⁴³ A similar § 313 case involving grazing impacts [28 ELR 10431] on a desert spring designated under the CWA as an "outstanding natural resource water," National Wildlife Federation v. Bureau of Land Management,²⁴⁴ has led to an administrative stay of the proposed grazing, in effect a preliminary injunction, foreshadowing the likelihood of success on the merits. The power of the § 313 approach is that its language is not restricted to "discharges" but includes the "runoff of pollutants" as well. Nor does it lead simply to state certification,²⁴⁵ which is in many states a routine and meaningless formality. Coupled with a cause of action by citizen groups under the APA, § 313 could go a long way to provide proactive and explicit scrutiny of timber and grazing impacts on water quality, independent of whatever a state chose to do under § 401.²⁴⁶

Against this background — and responding both to a heightened concern for water quality and to a lessening of the stranglehold that the extractive industries have had over the public and the politics of a rapidly urbanizing West²⁴⁷ — the Administration has been mounting an offensive of its own based largely on water quality impacts. In fact, this is an offensive that has been building for years in reaction to a series of "train wrecks" under the Endangered Species Act²⁴⁸ and to the evolving — if still elusive — principles of ecosystem management.²⁴⁹ Recognizing the importance of healthy riparian ecosystems, Secretary of the Interior Babbitt began a series of discussions on "rangeland reform," leading to the promulgation of regulations in 1995 on, inter alia, the restoration of "rangeland health."²⁵⁰ The national regulations contain specific goals for vegetation and water quality which, if not met, call for "appropriate action" including reductions in livestock use²⁵¹; among these goals are that "water quality [comply] with State water quality standards."²⁵² Bureau of Land Management (BLM) districts were to follow with their own more specific guidelines, which are now emerging. Guidelines for Utah are illustrative, stating that:

BLM will apply water quality standards established by the State of Utah and the Federal Clean Water and Safe Drinking Water Acts. BLM actions and other BLM land activities will notcontribute to violating those standards. As indicated by a measurement of nutrient loads, total dissolved solids, chemical constituents, fecal coliform, water temperature and other water quality parameters.²⁵³

On paper anyway, BLM has joined the party.

The Forest Service has joined too, if in a different fashion. The Service's first efforts were driven by the needs of endangered salmon and other aquatic and riparian species in the old growth forests of the Pacific Northwest; these efforts have led to greater "leave" areas along forest streams and the retention of native vegetation for the stated purpose of restoring water quality and aquatic populations.²⁵⁴ A second movement was generated over the long-simmering issue of "below-cost" timber sales on forest lands, sales whose revenue did not come close to matching the Service's costs in road construction, sale administration, and reforestation.²⁵⁵ In early 1998, the incoming Service Chief announced an 18-month moratorium on new road building in unroaded areas of greater than 5,000 acres, to be accompanied by a thorough review of road impacts and the generation of a new road policy.²⁵⁶

Extending this initiative, the President's Clean Water Action Plan announced that the Service would "substantially increase" maintenance of its existing road system "to protect water quality," and, for the same purpose, would relocate over 2,000 miles of existing roads and "decommission or obliterate" another 5,000 miles.²⁵⁷ Equally dramatically, the plan announced that EPA would be considering over the next year whether to require Clean Water Act permits for forest roads and to develop a pilot permit program.²⁵⁸ In light of the direct relationship between forest roads, erosion, and water quality, and the no less direct relationship between new roads, new clearcutting, and more erosion, these new requirements have the potential to work significant improvements in upstream, western water quality — taking considerable pressure off of TMDLs.

The key phrase for both the rangeland standards and the new forest policies is "potential." Both are under heavy fire from the timber and grazing industries and their champions in Congress²⁵⁹; neither may last long enough to be effectuated over the next several years. Even with the best of political [28 ELR 10432] support, moreover, it is not easy to calibrate the correlation between a given number of cattle and water quality several miles away; the most that can be said with confidence, is that more, or less, pollution will occur. For this reason, some states such as Arizona finesse these calculations altogether and substitute, for the purposes of certifying activities under § 401, compliance with a rather loose and casually supervised list of BMPs.²⁶⁰ These difficulties and this solution are, of course, precursors of the drill to come for all western waters through TMDLs. Precise calibration of activities to impacts will rarely be possible. Resort to BMPs-plus-monitoring and more-BMPs-plus-more-monitoring will be inevitable, until the water improves or the political will gives out.

What the new grazing standards and forest initiatives offer to the TMDL process is a more direct route to the steps everyone knows are necessary — reducing cattle and clearcuts and stepping them back from the water.²⁶¹ They also offer, to understaffed and outgunned state water quality employees who will be on the line for TMDLs, a sharing of the political heat for these steps with federal land managers. If and as these federal initiatives are allowed to bear fruit.

[] Coastal Nonpoint Programs.

The "dead zone" in the Gulf of Mexico, an area of water containing so little oxygen that fish and shrimp cannot survive in it, has grown to 7,032 square miles this summer, the biggest it's been in the 10 years the zone has been measured…. The nutrients in the rivers' water [come] from a variety of sources, including fertilizer runoff from farms and lawns throughout the Mississippi River watershed, which includes much of 33 states. ((New Orleans) Times Picayune)²⁶²

In 1990, faced with the reality that its CWA § 319 nonpoint source pollution control program was not getting the job done and that coastal waters were taking the biggest hit, Congress moved to bring yet another statute into play, the Coastal Zone Management Act (CZMA).²⁶³ As Congress noted, nonpoint sources were sending more than 100 million tons of sediments into the Great Lakes every year.²⁶⁴ They were also contributing more than half of the phosphorous, chromium, copper, lead, and zinc delivered to coastal waters.²⁶⁵ Nutrient loadings were in the process of creating "dead zones" from Long Island Sound to the Gulf of Mexico.²⁶⁶ On the receiving end of runoff from everywhere, and with enormous values in recreation, commerce, and biological productivity at stake, the coastal waters had good reason for special protection, and an available mechanism in state coastal management programs. The result was the Coastal Zone Amendments Reauthorization Act of 1990 (CZARA) § 6217.²⁶⁷

Originally enacted in 1972, the CZMA offered federal funding for voluntary state management programs, much along the lines of EPA's support for state nonpoint programs under the CWA.²⁶⁸ Funding for state plans and their implementation was administered by NOAA under statutory goals (both "to develop" and "to protect" the coastal zone)²⁶⁹ that were in obvious contradiction and under requirements that were the ultimate in ambiguity and discretion (e.g., to give "full consideration to ecological, cultural, historic and aesthetic values, as well as the needs for compatible economic development").²⁷⁰ For the next 17 years, NOAA struggled to bring forth programs of varying quality from 29 coastal states, reviewing, cajoling, and occasionally criticizing state performance but never once pulling the trigger on its only weapon, the diminution of federal funding for failing to "adhere to" an approved state program.²⁷¹ What is clear from the 1990 amendments is that Congress was impressed by impacts of nonpoint source pollution on coastal resources and, at least in this regard, wanted NOAA and the states to run a tighter ship.

CZARA called on states to identify both land use sources and coastal water quality,²⁷² to revisit the adequacy of the reach of their coastal zone jurisdictions with regard to nonpoint sources,²⁷³ and to develop "management measures" to "achieve and maintain applicable water quality standards under [CWA § 303]."²⁷⁴ Management measures were defined as "economically achievable measures" reflecting the "greatest degree of pollutant reduction achievable" through the "best available nonpoint pollution control practices, technologies, processes, siting criteria, operating methods or other alternatives."²⁷⁵ As EPA and NOAA, jointly given the federal authority to implement CZARA, explained, this approach was deliberately "technology-based rather than water quality-based" (i.e., based on "technical and economic achievability, rather than on establishing cause and effect linkages between particular land use activities and particular water quality problems").²⁷⁶ Rather than quibble over water quality impacts, "states would be able to concentrate their resources on measures that experts agree will reduce pollution significantly."²⁷⁷ Additional upgrade, where necessary, could be provided for "remaining coastal water quality problems."²⁷⁸ This approach, of course, is a mirror [28 ELR 10433] image of the BAT-then-TMDL approach of the CWA, and CZARA's initial management measures were to mimic in their objectivity and rigor those of point source BATs themselves.

Leaving less to chance than it had in 1972, Congress also mandated that state coastal plans contain "enforceable policies and mechanisms" to implement the nonpoint management measures, policies, and mechanisms that²⁷⁹ would have to be established under state law. State programs were to be submitted and approved within three years, or by May 1995.²⁸⁰ If a state "failed to submit an approvable program," NOAA "shall withhold" both CZMA and CWA § 319 funding by congressionally specified percentages, from 10 percent in the first year to 30 percent by 1999.²⁸¹ In the words of one enthusiastic commentator, CZARA contained "stringent enforcement requirements that mark the end of weak nonpoint source controls, at least in coastal states."²⁸²

Do not bet the farm on it.²⁸³ As of spring 1998, three years past the deadline, not one coastal state had submitted an approved nonpoint program.²⁸⁴ Twenty-two states had received "conditional approvals" for submissions that were inadequate in nearly every identified category.²⁸⁵ Seven states had submitted nothing at all.²⁸⁶ Relying on its "conditional approvals" of these submissions — a type of approval for which no provision is made in the statute and which circumvents the prescribed, statutory sanctions — NOAA has withheld no monies.²⁸⁷

Turning to the content of the state plans, not a great deal more that is encouraging meets the eye. Not for lack of technical guidance. In 1995, EPA and NOAA published a voluminous set of guidance with management measures for agriculture, forestry, transportation, and several other major coastal nonpoint sources.²⁸⁸ The guidance contained commendable specificity (e.g., new development construction in urban areas), it would achieve a reduction in total suspended solids of 80 percent or no greater than preconstruction loadings, and postconstruction peak runoff rate and volume would equal predevelopment levels.²⁸⁹ Easy enough to say, of course. But the hard part was yet to come — translating these measures into "enforceable" state plans. We may wait a long time.

A case in point is Louisiana, with perhaps the most important coastal zone in the country for commercial and recreational fisheries, and for the extent of coastal waters contaminated by fecal coliforms, nutrients, and other nonpoint pollution. Louisiana's plan is a telephone book-sized compendium of preexisting state programs and activities, proposing few new initiatives and nothing that could be characterized as "enforceable" authority.²⁹⁰ Indeed, as a 1995 state summary assures its readers:

Louisiana has taken the position that we will work to get passed a program that is as non-regulatory as possible, realizing that a minimum of enforceable policies must be identified as currently in existence, and deemed to be adequate, or new ones may need to be developed.²⁹¹

In August 1996, EPA and NOAA replied with "draft findings and conditions" that objected to the state's refusal to propose extending its coastal zone boundary to include more upstream sources, found that the state's agricultural management measures (consisting largely of the distribution of educational materials) did "not include enforceable policies and mechanisms to insure implementation," and found "insufficient justification" for Louisiana's proposal to exclude forestry sources altogether from its coastal nonpoint program.²⁹² The state responded with a brief in its defense, arguing, inter alia, that upstream sources had no adverse impact on coastal water quality, nor did forestry, and that existing law provided fully adequate enforcement activity for agricultural sources.²⁹³ A similar dialogue has begun with California.²⁹⁴ Indeed, similar conversations are now occurring with at least 29 coastal states, and no one can say how they will bottom out. One can say, however, in the words of Justice Holmes, that the "hydraulic pressures" of [28 ELR 10434] "immediate interests"²⁹⁵ are once again at work on statutory requirements, and something considerably less than technology-based, enforceable mechanisms will emerge from CZARA. What went wrong?

Whatever else went wrong, EPA and NOAA ran into more than they could handle from coastal states and their nonpoint source constituencies. As early as 1991, the Coastal States Organization was taking the position that CZARA was a "confusing piece of legislation" imposing "Draconian time frames" and, with some validity, that federal financial support would underwrite little more than "a nice slide show."²⁹⁶ In 1994, the organization formally petitioned EPA and NOAA to relax their view of the "enforceability" of state coastal programs, and to accept existing state authority as sufficient.²⁹⁷ As the state pressure mounted, the federal agencies issued program "clarification" in 1995. Entitled Flexibility for State Coastal Nonpoint Programs, it extended the time frames for state submissions, presented a "range of enforceable policies and mechanisms that could be used by states to implement their programs," and announced the policy of "conditional approvals" described above.²⁹⁸

The heat continued, however, and it came, predictably, from those same sources that have been opposing the application of TMDLs to nonpoint sources. Again by way of example, the Louisiana coastal nonpoint program record contains strong criticism from the Louisiana Forestry Association (forestry should be excluded from the state plan, reliance should be placed on voluntary practices),²⁹⁹ the Roy O. Martin Lumber Company (ditto),³⁰⁰ the Calcasieu Parish Cattlemen's Association (the plan should be delayed, agriculture should be exempted until specific impaired waterways are identified, a water quality approach rather than a technology-based approach should be used, and "all reference to the minimum backup enforcement recommended — the 'Bad Actor' law, or other enforcement measures — [should] be eliminated completely"),³⁰¹ the Louisiana Farm Bureau Federation, Inc. (ditto),³⁰² and Stream Property Management Inc. (ditto).³⁰³ The record also contained, for some indication of the lateral influence of these interests, two separate letters with three attachments from the Chancellor of the Louisiana State University Agricultural Center advocating no change in the coastal boundary, the exclusion of forestry, a water quality-based approach limited to impaired waters, an exclusively voluntary approach to implementation, and no additional enforcement mechanisms.³⁰⁴ Thoughtfully, the Chancellor copied every member of the Louisiana congressional delegation.

The same heat intensified on EPA and NOAA at the federal level.³⁰⁵ In April 1997, the agencies began "discussions" with states and "other interested parties" (not further specified, but one is allowed to guess) over "significant impediments" to implementing the coastal nonpoint program.³⁰⁶ These discussions pointed to the need for "additional flexibility" for the states, at which point the "parties proceeded to discuss in detail the specific aspects of the program that would require modification."³⁰⁷ All of which led to a public notice (for the first time) in April 1998 that the agencies were proposing more "administrative changes" to the program.³⁰⁸

The changes are more than administrative. After citing the President's Clean Water Action Plan for the proposition that NOAA and EPA will work with states to ensure the development of plans to control polluted runoff "with appropriate state-enforceable policies and mechanisms"³⁰⁹ — a reference that might lead one to think that stronger measures were to follow — the notice announced that the federal agencies were reconsidering the "level of detail" that should be required of states in linking the "implementing and enforcement agencies." I.e., the notice went on to explain, "should states be required to establish clear criteria to determine where voluntary efforts have been unsuccessful and that enforcement actions are necessary?"³¹⁰ With the statutory concept of "enforceability" thus put up for grabs, the notice went on to announce that since CZARA set no deadline for the implementation of state programs, the agencies were considering extending the time frame for state compliance from 5 to 15 years.³¹¹ As some indication of the distance the federal agencies had been moved here, the notice asked whether "a shorter [time frame], e.g., twelve years, is feasible."³¹² The notice concluded with the observation that, as administrative guidance to the states, none of these requirements or deadlines were to be treated as regulations³¹³ (i.e., as binding or enforceable by the agencies or, as importantly, environmental citizen groups).

[28 ELR 10435]

The messages here for TMDLs are both cautionary and direct. To begin with the obvious, coastal nonpoint programs are not likely to take any significant pressure off of TMDL development, even for highly impaired coastal waters. CZARA, at least in the more intransigent states — which tend also to be the more polluted states — is evolving toward a reenactment of CWA § 319, based on largely voluntary BMPs and minimal enforcement. The congressional concept for CZARA of technology standards first, additional water quality-based management measures later, is eroding, under state and agribusiness pressure, toward a TMDL-like focus on impaired waters. Absent, of course, the impetus of citizen enforcement, which is to say, and with good evidence from the above scenario, largely at the will of the coastal states and their nonpoint constituencies. Which is the final lesson of the implementation of CZARA: when push comes to shove, the federal environmental agencies will not be able to hold their own without the offsetting influence of direct citizen involvement and at least the possibility of litigation. A statute that tells federal agencies to get tough, even if it provides explicit standards and deadlines, will not succeed unless, among other things, people can ensure that it does. This is of course the reason the TMDL story may have a different ending.

The State Responses

State officials this week are drafting a new uniform position on Clean Water Act implementation which does not commit states to any specific [time frame] for making determinations on how much pollution a waterbody can receive without exceeding water quality standards, but instead calls for schedules to reflect the availability of sound science. (Inside EPA)³¹⁴

No one statement or article can fairly capture the range of state responses on any issue, and so it is with TMDLs. Some states, pressed early and hard by citizen suits and reviewing courts, have gotten with the program and produced significant local results.³¹⁵ Other states have moved in meaningful, if differing, and in a few cases truly innovative, ways on nonpoint source pollution, and the incorporation of these efforts into TMDLs is more administrative than revolutionary.³¹⁶ But for the greatest number of states, TMDLs are burdensome, new, and revolution-requiring in the demands they place on water pollution control agencies and the unhappy constituencies they bring to the table. Agriculture and timber interests in agricultural and forest product states are accustomed to a fair measure of local gratitude and a corresponding lack of regulation; many state laws exempt agriculture and silviculture from water quality regulation altogether.³¹⁷ Many state legislatures have, further, forbidden state environmental agencies from adopting requirements more stringent than federal law,³¹⁸ which in the context of a federal nonpoint source program limited to voluntary measures means any regulations or requirements at all. In these states (i.e., in most states), TMDLs arrive on the doorstep like a litter of stray cats — with many unpleasant requirements and little money to provide for them.

Many states, perhaps a majority, would like to rid themselves of § 303(d) requirements and return to the more comfortable world of § 319. This was a position heard throughout the just-completed FACA process. The August 1997 meeting of the Association of State and Interstate Water Pollution Control Administrators featured a proposal to remove all deadlines from the TMDL program, in favor of "unique, non-regulatory, cost effective approaches" to deal with nonpoint source pollution.³¹⁹

Individually, few states have taken up the challenge with enthusiasm.³²⁰ The most often heard responses from state officials to the looming tasks they face under § 303(d) are "more time" and "more money."³²¹ On the matter of money, the Administration has promised to commit major new resources [28 ELR 10436] to the process.³²² Whether states are willing to commit resources on their own remains to be seen³²³; it bears remembering that, under the CWA, water quality standards-based regulation was, supposedly, the states' responsibility and contribution to the shared goal of clean water.³²⁴ On the matter of time, there is an obvious tension between the need to do TMDLs "right" and the need to get going. As described earlier, EPA called for the first comprehensive § 303(d) lists of impaired waters by April 1998, and for the submission of TMDLs for all listed waters within no more than 13 years. Beyond the rhetoric, then, two useful measures of state responses are the 1998 lists and the emerging, first-round TMDLs.

[] Listings. State responses to the listing deadline may not have been enthusiastic, but they produced numbers that are likely to increase greatly the universe of identified waters, and raise the stakes for the game to come.

As described in an earlier Article, in late 1997 EPA put on a full-court press for its regions and the states to meet the April deadline with full and timely information.³²⁵ As of mid-May 1998, 28 states and territories had submitted final lists for EPA review, 26 were still in a stage of drafting and public review, and 2 (Idaho and Iowa) had yet to submit even a draft.³²⁶ None of which should be construed as indications of widespread recalcitrance or insurmountable difficulty; 50 percent on-time compliance with federal deadlines is about par for the course in environmental law. On the other hand, there is certainly evidence of both genuine difficulty and genuine recalcitrance over listing issues. Threshold problems arise over flow assumptions and wet-flow conditions.³²⁷ Additional difficulties arise over the use of "old" data, data from federal agencies, and the sufficiency of the data itself. The agriculture industry complains of "drive by" listings and threatens to sue over the adequacy of their supporting information.³²⁸ Environmentalists complain that while EPA guidance calls on states to use "all credible data," states have done little asking for outside data, and have invented such categories as "not credible enough" and "more data needed" to hold down the lists, and their subsequent workload.³²⁹ At least one state has indicated that it was going to reexamine (e.g., lower) the designation of stream uses that determine its water quality standards, explaining that if the uses are not "representative," the "system becomes flawed."³³⁰ At the far end of the process are disagreements over whether a water becomes delisted by the preparation of a TMDL or, as advocated by environmentalists and endorsed by the FACA committee, only after its water quality standards are attained.³³¹

These difficulties noted, there is no doubt that the universe of listed waters will expand. An EPA summary of the previous state lists of 1996 found 15,598 impaired waters, identifying (collectively) 33,000 sources.³³² An agriculture representative has predicted that these listings will double,³³³ which may be an indication of how much confidence this industry has had in the integrity of the exercise in prior years. EPA projections before the FACA committee rose throughout the process to an estimate of 20,000.³³⁴ We will soon see. The actual numbers will be arbitrary and in a sense misleading, because they are based on the arbitrary division of waters into stream segments for purposes of § 303 reports; small streams and large rivers, small stretches and long ones, apples and oranges, are lumped together toward one large number.³³⁵ On the other hand, the numbers do provide clear and convincing evidence that the nation's remaining water quality problems are far larger than admitted, which admission, as any recovering alcoholic will testify, is the first step toward recovery.

Equally as meaningful, the 1998 lists will contain first-ever state schedules for the development of TMDLs for all listed waters.³³⁶ Which brings us to the consideration, at last, of the ultimate product and purpose of the exercise, the TMDLs themselves.

[] TMDLs. At the end of § 303(d) and the long, contentious, and difficult road that leads to its implementation are the TMDLs themselves, the actual plans that identify pollution loadings, allocate them to sources, and present mechanisms for their abatement. Good plans and their execution in fact will vindicate the faith that at least some members of Congress, led by the House of Representatives, had in enacting these water quality-based requirements in 1972. Weak or unexecuted plans will confirm the skepticism of the Senate and others that this approach to water pollution control could ever work. We know enough from the process to date that this is an expensive and time-consuming courtship of state and regulated interests, particularly nonpoint discharge [28 ELR 10437] industries. What we don't know yet is whether what is coming out the end of the pipe is worth the effort.

But we do have some early data. Over the past few years, and in increasing numbers as the pressure of citizen suits is felt, states have begun submitting, and EPA approving, TMDLs for listed impaired waters. In the fall of 1997, this author sent Freedom of Information Act requests to each EPA region for five TMDLs illustrating the implementation of § 303 by their respective states. As of April 1998, 9 of the 10 regions had forwarded TMDLs; Region 7, as of that date, had no approved TMDLs. Additional TMDL summaries were obtained from EPA. In all, 55 TMDLs were reviewed in order to make very basic determinations of what pollutants, discharge sources, load calculations, and reductions were contained in these documents, as well as their inclusion of such regulatory criteria as margins for safety and for future development.

Summaries of this analysis are presented in Appendix A to this Article. An explanation of categories analyzed and the terminology used is presented in the Appendix as well. In addition to this explanation, a further caveat should be noted about the collection and use of this data. First, neither this author nor his research assistants are water quality experts; the analysis is limited to understanding what the documents contain and their compliance with the stated rules of the program. In the same vein, the review takes the documents at face value; if a calculation is said to have been made or if a BMP plan is said to exist, those statements are accepted as both true and sufficient. Doubtless, some of these statements will prove in fact to be truer, and more sufficient, than others.

The TMDLs reviewed reveal, and confirm, several features of the program that might have been predicted by its central conundrum: how to come to grips with nonpoint sources. These features include:

1. Nonpoint sources are targets of last resort. If at all possible, point sources will be identified as the sole source of even conventional pollution (e.g., solids and oxygen demand) and the sole instrument of cleanup. Despite EPA guidance defining TMDLs as the combination of point source WLAs and nonpoint source LAs and requiring both calculations for impaired waters, nearly 50 percent of the approved TMDLs (25 of a total of 55) did not even identify nonpoint source contributions. In effect, these documents described NPDES permit reductions, usually by municipal sewage treatment systems, that had taken place or would take place in the future.

2. Even where identified, nonpoint source reductions will frequently not be calculated, and, where calculated, even more frequently will not be implemented through identified abatement plans. Of the 30 TMDLs reviewed that acknowledged nonpoint contributions, 10 did not quantify these loadings, and only 13 projected actual, quantified reductions. Where projected, these reductions, further, were predicated in almost every case on unspecified BMPs contained in state nonpoint source or other documents and incorporated by reference. In only one case did it appear that nonpoint controls were going to be mandatory or enforceable.

3. Statutory margins for safety and to accommodate future growth are honored more in the breach than the observance. Only 20 documents contained margin of safety consideration, and only 6 made explicit provision for future growth. These conclusions, further, credit state assertions that margins of safety were provided not by allocating an increment for error but, rather, through the use of conservative estimates, assertions that might well be subject to question.

In sum, TMDLs have a long way to go before they contain the analysis or produce the results Congress intended in 1972. Taking the documents approved to date at face value, granting them the benefit of every doubt and making allowances for the fact that they are first efforts for many states and EPA regions, they provide little confidence that pollution sources are being addressed in any fashion beyond ratcheting down on NPDES permits where they are available and, where they are not, by passing the buck to unidentified and largely unenforceable nonpoint source BMPs. Congress, EPA, and citizen suits can lead this horse to water but it remains to be seen how much it is going to drink.

What We Have Framed

The machinery is now in place for an all-out assault on, essentially, nonpoint source pollution through the use of ambient water quality standards. Ahead are perhaps 20,000 impaired water bodies and a gallery of pollution sources that have largely escaped meaningful state or federal controls. At hand is a pollution control system that is beguilingly simple to describe and frustratingly difficult to implement: expensive, site-specific, heavily reliant on science, equally reliant on monitoring, and almost completely reliant on the ability and political will of states and local governments to carry it out. We have been here before with water pollution, indeed for nearly two decades prior to 1972. We have been here before with air pollution, in the years following the ambient air standards-based Clean Air Act, before it, too, gradually migrated toward technology standards. We did not make much progress this way on either occasion. The question is whether we will do better this time.

The answer is, of course, maybe. The responses of the state water pollution administrators, agriculture, and other nonpoint industries, and the early TMDLs themselves are not encouraging and could well drive this program back to another failed experiment in pollution control. On the other side is the statutory specificity of § 303(d), its favorable reception in the courts, and its empowerment of many Americans who consider clean water a birthright and are not going to let this opportunity slide. In the middle, of course, but with its hands now on the controls is EPA and, to its side, quiet so far but clearly hostile to anything regulatory, lies Congress. The smart money would hedge its bets, either way. Should the courts remain demanding, the states buckle down to the job, and Congress exercise the good judgment to let this process go forward, the only measure that counts will be how many TMDLs appear over the next few years, what they say, and whether what they say actually happens.

For 40 years, states have clamored for the authority to tackle water pollution themselves and by their own priorities. [28 ELR 10438] For an almost equal number of years, both they and industry have advocated that ambient water quality-based regulation was more practical, feasible, and fair. Section 303(d) responded directly to that advocacy, empowered the states, authorized a water quality approach, and provided the only logical sequence of steps imaginable, short of "do-whatever-you'd-like," for carrying it out. A major experiment in environmental federalism is about to begin.

[28 ELR 10439]

APPENDIX 1

Summary of Illustrative Total Maximum Daily Load Determinations, April 1998

The summary that follows tracks the steps of the Clean Water Act § 303(d) and of EPA regulations and guidance for their implementation. In brief, a TMDL is the sum of those waste load allocations (WLAs) of a given pollutant from point sources, load allocations (LAs) from nonpoint sources, margins of safety (MOS) to reflect uncertainty in the calculations, and margins to accommodate discharges from future development (MFDs) that are necessary to reduce their total in order for the receiving water to meet a state water quality standard. EPA's equation simply captures the obvious: TMDL = WLA + LA + MOS + MFD. Logically, a TMDL will first identify the total current loadings of an offending pollutant, then the total (lesser) loadings necessary to meet water quality standards, and the necessary allocations and margins and, at least in some fashion, a mechanism for their implementation. * These were the points of focus for the analysis below. A further description of this analysis and its conclusions is contained in the preceding text.

Location	Source	Pollutants	Pre-TMDL
(by EPA Region)		of Concern	Load
Region 1
Willimantic River,	PS	NH3-N, DO,	no
Connecticut		Phosphorus,
		TSS
Lake Champlain,	PS/NPS	Phosphorus	quantified
NY, VT, Quebec
Flints Pond	NPS	Phosphorus,	quantified
Hollis, NH		Chlorophyll-a
Lamprey River	PS/NPS	NH3-N, DO,	not
Epping, NH		CBOD,	quantified
		Phosphorus
Saco River	PS	DO, BOD	quantified
Estuary, Maine
Region 2
Bog Brook	PS, NPS	Phosphorus	quantified
Reservoir, NY
East Branch	PS, NPS	Phosphorus	quantified
Reservoir, NY
Croton Falls	PS, NPS	Phosphorus	quantified
Reservoir, NY
Diverting	PS, NPS	Phosphorus	quantified
Reservoir, NY
Middle Branch	PS, NPS	Phosphorus	quantified
Reservoir, NY
Region 3
Appoquininimink	PS, NPS	Phosphorus,	quantified
River, DE		CBOD, BOD
Upper Blackwater	PS, NPS	CBOD,	quantified
River, VA		NBOD
South Fork Branch	PS, NPS	Fecal coliform	quantified
Potomac River, WV
Region 4
Tar-Pamlico	PS, NPS	Nitrogen,	quantified
Basin, NC **		Phosphorus
Roanoke River,	PS	Dioxin	quantified
NC ***
Lake Wylie, NC	PS, NPS	Chlorophyll-a	quantified
Region 5
Roberts Creek	PS, NPS	CBOD, DO,	no
Brownsdale, MN		NH3-N
Center Creek	PS	TSS, DO,	quantified
Fairmont, MN		CBOD, NH3-N,
		Fecal coliform
Penobscot Creek	PS	DO, NH3-N,	quantified
Hibbing, MN		Temperature
Sycamore Creek	NPS	Sediment,	quantified
Ingham Co., MI		TSS
Lower Minn.	PS, NPS	CBOD,	quantified
River, MN		Ammonia
Region 6
North Canadian	PS	CBOD	quantified
River, OK
Coal Creek	PS	DO, CBOD,	no
Okamulgee, OK		NH3-N, TSS,
		BOD
Otter Creek	PS	CBOD,	no
E. Otter, OK		NH3-N
Arkansas River,	PS	DO	no
OK
Vermillion River,	PS	DO, Fecal	no
LA		coliform,
		Mercury,
		CBOD, TSS
Bayou Yarbor	PS	DO, CBOD,	no
Opelousas, LA		NH3-N
Foxskin Bayou	PS	DO, BOD,	quantified
Bossier Parish, LA		TSS, NH3-N
W. Pearl River	PS	DO, BOD,	quantified
Pearl River, LA		TSS, NH3-N
Bayou Teche,	PS	CBOD, TKN,	quantified
LA		DO
Hominy Creek	PS	DO, CBOD,	quantified
Osage Co., OK		NH3-N
Region 7
No TMDLS available
Region 8
W. Fork Clear	PS, NPS	Toxins,	not
Creek, CO		Metals	quantified
Boulder Creek,	PS, NPS	NH3-N,	quantified
CO		Nutrients
Deep Creek,	NPS	TSS	quantified
MT
Lake Poinsett,	NPS	Nutrients, TSS,	quantified
SD		Fecal coliform
Upper Big Sioux,	NPS	Nutrients, TSS	quantified
SD
Region 9
Truckee River,	PS, NPS	Nitrogen,	quantified
NV		Phosphorus,
		TSS
Laguna de Santa	PS, NPS	Ammonia,	quantified
Rosa, CA		DO
Region 10
Salmon River,	NPS	Sediments	quantified
ID
Columbia River	PS, NPS	Dioxin	quantified
Basin, WA, OR,
ID
Lake Chelan,	PS, NPS	Phosphorus,
WA		Bacteria	no
Tualatin River,	PS, NPS	Phosphorus	quantified
OR
Vanderbilt Creek,	NPS	TSS,	quantified
AL		Turbidity
Upper Birch Creek,	NPS	TSS,	quantified
AL		Turbidity
Lemon Creek,	NPS	TSS,	quantified
AL		Turbidity
Location	Gross Load	TMDL Allocation	Implementation
(by EPA Region)	Cal. (TMDL)	(PS/NPS)
Region 1
Willimantic River,	PS only	PS: quantified	WLA: NPDES
Connecticut
Lake Champlain,	quantified	PS: quantified	WLA: NPDES
NY, VT, Quebec		NPS: quantified	LA: BMPs
		only
Flints Pond	quantified	NPS: quantified	LA: zoning, BMPs
Hollis, NH
Lamprey River	quantified	PS: quantified	WLA: NPDES
Epping, NH	(seasonal)	NPS: quantified	LA: none
Saco River	quantified	PS: quantified	WLA: NPDES
Estuary, Maine	in part
Region 2
Bog Brook	quantified	PS: quantified	WLA: NPDES
Reservoir, NY		NPS: quantified	LA: BMP
			(reserved)
East Branch	quantified	PS: quantified	WLA: NPDES
Reservoir, NY		NPS: quantified	LA: BMP
			(reserved)
Croton Falls	quantified	PS: quantified	WLA: NPDES
Reservoir, NY		NPS: quantified	LA: BMP
			(reserved)
Diverting	quantified	PS: quantified	WLA: NPDES
Reservoir, NY		NPS: quantified	LA: BMP
			(reserved)
Middle Branch	quantified	PS: quantified	WLA: NPDES
Reservoir, NY		NPS: quantified	LA: BMP
			(reserved)
Region 3
Appoquininimink	quantified	PS: quantified	WLA: NPDES
River, DE		NPS: quantified	LA: not
			specified
Upper Blackwater	quantified	PS: quantified	WLA: NPDES
River, VA		NPS: quantified	LA: not
			specified
South Fork Branch	quantified	PS: quantified	WLA: NPDES
Potomac River, WV		NPS: quantified	LA: not
			specified
Region 4
Tar-Pamlico	quantified	PS: quantified	PS/NPS Trading
Basin, NC**		NPS: no	Program
Roanoke River,	quantified	PS: quantified	yes
NC ***
Lake Wylie, NC	quantified	PS: quantified	WLA: NPDES
		NPS: no	LA: BMP
Region 5
Roberts Creek	quantified	PS: quantified	WLA: NPDES
Brownsdale, MN		NPS: not	LA: no
		quantified
Center Creek	for NH3-N	PS: quantified	WLA: NPDES
Fairmont, MN	(seasonal)	NPS: not	LA: no
		quantified
Penobscot Creek	no	PS: quantified	WLA: NPDES
Hibbing, MN		(seasonal)	LA: no
		NPS: not
		quantified
Sycamore Creek	quantified	NPS: quantified	LA: BMP
Ingham Co., MI
Lower Minn.	quantified	PS: quantified	WLA: NPDES
River, MN	(seasonal)	NPS: quantified	LA: BMP
Region 6
North Canadian	quantified	PS: quantified	WLA: NPDES
River, OK	(seasonal)	NPS: quantified	LA: BMP
Coal Creek	quantified	PS: quantified	WLA: NPDES
Okamulgee, OK
Otter Creek	quantified	PS: quantified	WLA: NPDES
E. Otter, OK
Arkansas River,	quantified	PS: quantified	WLA: NPDES
OK
Vermillion River,	no	PS: quantified	WLA: NPDES
LA
Bayou Yarbor	quantified	no	no
Opelousas, LA
Foxskin Bayou	quantified	PS: quantified	WLA: NPDES
Bossier Parish, LA		NPS: quantified	LA: no
W. Pearl River	quantified	PS: quantified	WLA: NPDES
Pearl River, LA
Bayou Teche,	quantified	PS: quantified	WLA: NPDES
LA
Hominy Creek	quantified	PS: quantified	WLA: NPDES
Osage Co., OK
Region 7
No TMDLS available
Region 8
W. Fork Clear	quantified	PS: quantified	WLA: NPDES
Creek, CO	(seasonal)	NPS: quantified	LA: BMPs
Boulder Creek,	quantified	PS: quantified	WLA: NPDES
CO		NPS: quantified	LA: BMP
			(phased)
Deep Creek,	quantified	no	Monitoring
MT			only
Lake Poinsett,	quantified	no	WLA: no
SD			LA: BMP
Upper Big Sioux,	quantified	no	LA: BMP, control
SD		(NEWCOLUMN)mechanisms
Region 9
Truckee River,	quantified	PS: quantified	WLA: NPDES
NV		NPS: quantified	LA: stormwater
			permitting
Laguna de Santa	quantified	PS: quantified	WLA: NPDES
Rosa, CA		NPS: quantified	LA: stormwater
			permitting
Region 10
Salmon River,	not	NPS: not	LA: BMP
ID	quantified	quantified
		(25% goal)
Columbia River	quantified	PS: quantified	WLA: NPDES
Basin, WA, OR,		NPS: quantified	LA: BMP
ID
Lake Chelan,		PS: quantified	WLA: NPDES
WA	quantified	NPS: quantified	LA: quality
			committee
Tualatin River,	quantified	PS: quantified	WLA: NPDES
OR		NPS: quantified	LA: future
			BMPs
Vanderbilt Creek,	quantified	NPS: quantified	LA: BMP
AL			(phased)
Upper Birch Creek,	quantified	PS: quantified	WLA: NPDES
AL		NPS: quantified	LA: BMP
Lemon Creek,	quantified	NPS: quantified	LA: BMP
AL
			Projected
Location	Margin of	Future Dev.	NPS
(by EPA Region)	Safety	Margin	Reduction
Region 1
Willimantic River,	no	no	no
Connecticut
Lake Champlain,	quantified	quantified	yes
NY, VT, Quebec
Flints Pond	no	no	yes
Hollis, NH
Lamprey River	quantified	no	no
Epping, NH
Saco River	quantified	no	no
Estuary, Maine
Region 2
Bog Brook	quantified	quantified	no
Reservoir, NY			(deferred)
East Branch	quantified	not quantified,	no
Reservoir, NY		mentioned	(deferred)
Croton Falls	quantified	not quantified,	no
Reservoir, NY		mentioned	(deferred)
Diverting	quantified	not quantified,	no
Reservoir, NY		mentioned	(deferred)
Middle Branch	quantified	not quantified,	no
Reservoir, NY		mentioned	(deferred)
Region 3
Appoquininimink	not quantified,	no	yes (but
River, DE	mentioned		unspecified)
Upper Blackwater	not quantified,	no	yes (but
River, VA	mentioned		unspecified)
South Fork Branch	not quantified,	no	yes (but
Potomac River, WV	mentioned		unspecified)
Region 4
Tar-Pamlico	not quantified,	not quantified,	yes
Basin, NC**	mentioned	mentioned
Roanoke River,	no	no	no
NC ***
Lake Wylie, NC	not quantified,	quantified	no
	mentioned
Region 5
Roberts Creek	no	no	no
Brownsdale, MN
Center Creek	no	no	no
Fairmont, MN
Penobscot Creek	no	no	no
Hibbing, MN
Sycamore Creek	no	no	no
Ingham Co., MI
Lower Minn.	no	no	yes
River, MN
Region 6
North Canadian	quantified	no	no
River, OK
Coal Creek	no	no	no
Okamulgee, OK
Otter Creek	quantified	no	no
E. Otter, OK
Arkansas River,	quantified	no	no
OK
Vermillion River,	no	no	no
LA
Bayou Yarbor	no	no	no
Opelousas, LA
Foxskin Bayou	no	no	no
Bossier Parish, LA
W. Pearl River	quantified	no	no
Pearl River, LA
Bayou Teche,	no	no	no
LA
Hominy Creek	quantified	no	no
Osage Co., OK
Region 7
No TMDLS available
Region 8
W. Fork Clear	quantified	not quantified,	no
Creek, CO		mentioned
Boulder Creek,	no	quantified	no
CO
Deep Creek,	no	no	no
MT
Lake Poinsett,	no	no	no
SD
Upper Big Sioux,	no	no	no
SD
Region 9
Truckee River,	not quantified,	not quantified,	no
NV	mentioned	mentioned
Laguna de Santa	not quantified,	not quantified,	no
Rosa, CA	mentioned	mentioned
Region 10
Salmon River,	not quantified,	not quantified,	yes
ID	mentioned	mentioned
Columbia River	quantified	quantified	no
Basin, WA, OR,
ID
Lake Chelan,	not quantified,	quantified	yes
WA	mentioned
Tualatin River,	quantified	no	no
OR			(deferred)
Vanderbilt Creek,	quantified	no	yes
AL
Upper Birch Creek,	quantified	no	yes
AL
Lemon Creek,	quantified	quantified	yes
AL

Key:

PS = point source

NPS = nonpoint source

WLA = waste load allocation

LA = load allocation

NPDES = national pollutant discharge elimination system

BMP = best management practices plan

NH3-N = nitrogen

DO = dissolved oxygen

TSS = total suspended solids

BOD = biological oxygen demand

CBOD = carbonaceous biochemical oxygen demand

TKN = total Kjeldahl nitrogen

For a full description of these terms, categories, and analysis see preceding text.

1. New Policies for Establishing and Implementing Total Maximum Daily Loads (TMDLs), Memorandum from Robert Perciasepe, Assistant Administrator for Water, U.S. EPA, to Regional Administrators and Regional Water Division Administrators, U.S. EPA (Aug. 8, 1997) http://www.epa.gov/owowwtr1/watershed/ratepace.html [hereinafter Perciasepe Memorandum].

2. 33 U.S.C. §§ 1251-1387, ELR STAT. FWPCA §§ 101-607.

3. The Clean Water Act presents separate programs predicated on, inter alia, technology-based standards, id. § 1314(b), ELR STAT. FWPCA § 304(b); water quality standards, id. § 1313, ELR STAT. FWPCA § 303; water quality criteria for toxins, id. § 1314(c), ELR STAT. FWPCA § 304(c); accelerated review of toxic "hot spots," id. § 1313(c)(2), ELR STAT. FWPCA § 303(c)(2); whole effluent testing, id. §§ 1313(c)(2)(B), 1314(a)(8), ELR STAT. FWPCA §§ 303(c)(2)(B), 304(a)(8); in-stream biological monitoring, id.; municipal sewage treatment, id. § 1314(d), ELR STAT. FWPCA § 304(d); stormwater runoff, id. § 1342(p), ELR STAT. FWPCA § 402(p); nonpoint source pollution, id. § 1329, ELR STAT. FWPCA § 319; dredge and fill material, id. § 1344, ELR STAT. FWPCA § 404; ocean discharge, id. § 1343, ELR STAT. FWPCA § 403; and oil pollution id. § 1321, ELR STAT. FWPCA § 311. Voluntary reductions are sought, inter alia, through EPA's Common Sense and Project XL initiatives. See Rena I. Steinzor, Regulatory Reinvention and Project XL: Does the Emperor Have Any Clothes?, 26 ELR 10527 (Oct. 1996) (describing EPA's Project XL and related initiatives).

4. 33 U.S.C. § 1313(c)(2)(B), ELR STAT. FWPCA § 303(c)(2)(B).

5. For a discussion of the difficulties with ambient standards regulations, see, e.g., WILLIAM H. RODGERS JR., ENVIRONMENTAL LAW 196-230 (2d ed. 1994) (the Clean Air Act ambient standards program, introduced as "This is not a fairy tale"); id. at 342-66 (the Clean Water Act ambient standards program: "Of all the dissatisfactions associated with the water quality standards experience, perhaps most acute was the inability of the standards to serve as a planning and particularly an enforcement device," indeed, "they filled every expectation of a calculated misfit"). For further analysis of the reasons for the failure of ambient standards under state and federal water pollution programs, see Oliver A. Houck, The Regulation of Toxic Pollutants Under the Clean Water Act, 21 ELR 10528, 10529-33 (Sept. 1991) ("The Science and Not-So-Science of Water Quality Standards"). Similar difficulties with ambient standards regulation of toxic substances and hazardous wastes are legendary. See RODGERS, supra, at 393, 403-07 (describing "pandemic uncertainty" in ambient risk-based pesticide regulation); id. at 516-22 (similar uncertainty over "unreasonable effects" of toxic substances); id. at 741-45 (near-paralysis over ambient-based hazardous waste cleanup standards which "generate a variety of poor fits, misfits and distortions," and "have not proven to be the work-savers and decision-determinants they were thought to be").

6. 33 U.S.C. § 1313(d), ELR STAT. FWPCA § 303(d).

7. More specifically, this section requires states to (1) identify waters that will remain in violation of water quality standards after the imposition of technology controls, id. § 1313(d)(1)(A), ELR STAT. FWPCA § 303(d)(1)(A); (2) prioritize those waters for remedial action, id; (3) identify TMDL allocations that will meet water quality standards, id. § 1313(d)(1)(C), ELR STAT. FWPCA § 303(d)(1)(C); and (4) incorporate these allocations into discharge permits and state water quality plans, id. § 1313(e)(3)(C), (F), ELR STAT. FWPCA § 303(e)(3)(C), (F).

8. Oliver A. Houck, TMDLs: The Resurrection of Water Quality Standards-Based Regulation Under the Clean Water Act, 27 ELR 10329 (July 1997) [hereinafter TMDLs I].

9. Id. at 10332-35, 10338-44.

10. Oliver A. Houck, TMDLs, Are We There Yet?: The Long Road Toward Water Quality-Based Regulation Under the Clean Water Act, 27 ELR 10391 (Aug. 1997) [hereinafter TMDLs II].

11. Id. at 10395-96.

12. Id. at 10397-401.

13. New Policies for Developing and Implementing Total Maximum Daily Loads (TMDLs), Draft Memorandum from Robert Perciasepe, Assistant Administrator for Water, U.S. EPA, to Regional Administrators and Regional Water Division Administrators, U.S. EPA (Mar. 21, 1997) (on file with author).

14. See TMDLs II, supra note 10, at 10392-93.

15. See Alaska Ctr. for Env't v. Reilly, 762 F. Supp. 1422, 1426, 21 ELR 21305, 21306 (W.D. Wash. 1991); Scott v. City of Hammond, 741 F.2d 992, 14 ELR 20631 (7th Cir. 1984).

16. U.S. EPA, TMDL Litigation by State (Apr. 10, 1998) http://www.epa.gov/owow/tmdl/lawsuit1.html. Citizen suits that have led to consent decrees against EPA and their respective states have been filed in Alaska, Arizona, California, Delaware, Georgia, New Mexico, Oregon, Pennsylvania, and Washington. States (or other jurisdictions) with pending cases include Alabama, Colorado, District of Columbia, Kansas, Louisiana, Maryland, Mississippi, Montana, New Jersey, New York, North Carolina, Oklahoma, Oregon, and Wyoming. States that have received notices of intent to sue include Alabama, California, Florida, South Dakota, and Virginia. (While the majority of these suits challenged the absence of TMDLs statewide, several focused on watersheds within a state; hence, the presence of more than one lawsuit in several states.)

17. American Littoral Soc'y v. U.S. EPA, No. 96-330, http://www.epa.gov/owow/tmdl/lawsuit1.html#n (D. Del. filed June 1996).

18. Sierra Club v. Hankinson, 939 F. Supp. 865, 27 ELR 20280 (N.D. Ga. 1996).

19. See Natural Resources Defense Council v. Fox, 909 F. Supp. 153, 26 ELR 20732 (S.D.N.Y. 1995); Kansas Natural Resources Council v. Browner, No. 95-2490-JWL (D. Kan. filed Oct. 1995); Sierra Club v. Saginaw, No. 96-0527 (N.D. La. filed Feb. 1996).

20.

For pollutants other than heat, [WLAs/LAs and] TMDLs shall be established at levels necessary to attain and maintain the applicable narrative and numerical [water quality standards] with seasonal variations and a margin of safety which takes into account any lack of knowledge concerning the relationship between effluent limitations and water quality. Determinations of [WLAs/LAs and] TMDLs shall take into account critical conditions for stream flow, loading, and water quality parameters.

40 C.F.R. § 130.7(c)(1) (1997).

21. Plaintiff's Proposed Findings of Fact and Conclusions of Law, at 11, 12, Natural Resources Defense Council v. Fox, No. 94 Civ. 8424 (S.D.N.Y. filed Oct. 31, 1997) ("The purported TMDLs adopted by New York State for 51 waterbodies were at best WLAs for point sources and did not incorporate [LAs] for non-point sources."); see also Declaration of Jack D. Smith, at 24, 25, Sierra Club v. Saginaw, No. 96-9527 (N.D. La. case filed Feb. 1996):

The entire focus of Louisiana's procedure is on point source loads determined for conditions of flow and loads from nonpoint sources or background assumed but rarely encountered. Point source loads thus determined are then labelled WLAs or TMDLs. In no case are [LAs] for nonpoint sources is assumed, even where nonpoint sources are the inescapably dominating water quality determinant, no [LAs] are developed for individual nonpoint sources in the way that "WLAs" are developed for individual point sources.

22. See infra p. 10435, The State Responses, and p. 10436, TMDLs.

23. National Wildlife Fed'n, Saving Our Watersheds (June 19-20, 1998) (TMDL implementation summit).

24. Telephone Conversation with Rick Parrish, FACA committee member (May 28, 1998) [hereinafter Parrish Conversation]; see also TMDL FEDERAL ADVISORY COMMITTEE FACILITATORS, FACA ROUGH DRAFT TMDL REPORT (Jan. 15, 1998) [hereinafter ROUGH DRAFT TMDL REPORT].

25. See infra text accompanying notes 108-13.

26. See infra text accompanying notes 62-75.

27. Consent Decree, Northwest Envtl. Advocates v. Browner, No. 91427R (W.D. Wash. Jan. 20, 1998).

28. Carolyn Whetzel, EPA Intervenes in Load-Setting Process for California Streams Due to Local Delays, State Env't Daily (BNA), Mar. 16, 1998, available in LEXIS, Envirn Library, Bnased File.

29. Dioxin/Organochlorine Ctr. v. Clarke, 57 F.3d 1517, 25 ELR 21258 (9th Cir. 1995).

30. See TMDLs II, supra note 10, at 10399-401.

31. See infra text accompanying notes 108-13.

32. See infra text accompanying note 328.

33. See infra text accompanying note 330. This conservatism seems to contravene EPA regulations that require states to seek out and rely on all credible data. See 40 C.F.R. § 130.7(b)(5) (1997).

34. 33 U.S.C. § 1313(d)(1)(C), ELR STAT. FWPCA § 303(d)(1)(C). The TMDL "shall be established at a level necessary to implement the applicable water quality standards with seasonal variations and a margin of safety which takes into account any lack of knowledge concerning the relationship between effluent limitations and water quality." Id.

35. See TMDLs II, supra note 10, at 10395 (quoting EPA Region X official Tom Wilson).

36. See Udall v. Tallman, 380 U.S. 1 (1965).

37. See Whetzel, supra note 28 ("Joe Karkowski of EPA Region IX told regional water quality officials that the agency would prefer to allow the regional and state process to continue, but terms of an agreement with the environmental group Defend the Bay dictate that the total maximum daily loads be established by April 13."). For an indication that the environmental community pressure on TMDLs will continue, see NATIONAL WILDLIFE FED'N, SAVING OUR WATERSHEDS: A FEDERAL GUIDE TO WATERSHED RESTORATION USING TMDLs (Jan. 1998).

38. National Clarifying Guidance for 1998 State and Territory Section 303(d) Listing Decisions, Memorandum from Robert H. Wayland III, Director of Office of Wetlands, Oceans, and Watersheds, U.S. EPA 2 (Aug. 17, 1997) http://www.epa.gov/owow/tmdl/lisgid.html [hereinafter Wayland Memorandum].

39. Id.

40. See Perciasepe Memorandum, supra note 1.

41. 33 U.S.C. § 1315(b)(1), ELR STAT. FWPCA § 305(b)(1).

42. 40 C.F.R. § 130.8(a) (1997).

43. 33 U.S.C. § 1313(d), ELR STAT. FWPCA § 303(d).

44. U.S. EPA, GUIDANCE FOR WATER QUALITY-BASED DECISIONS: THE TMDL PROCESS 12 (Apr. 1991) (available from the ELR Document Service, ELR Order No. AD-3550) [hereinafter GUIDANCE FOR WATER QUALITY-BASED DECISIONS].

45. See Wayland Memorandum, supra note 38, at 2.

46. See infra text accompanying notes 325-36; see also NATIONAL WILDLIFE FED'N, supra note 37 (TMDL monitoring handbook of National Wildlife Federation); NATURAL RESOURCES DEFENSE COUNCIL, CLEAN WATER NETWORK (May 1998) (component of CWA implementation project of the National Resources Defense Council).

47. Wayland Memorandum, supra note 38, at 6. This directive reiterates guidance from as early as 1991. Id.

48. Wayland Memorandum, supra note 38, app.

49. Id. at 4.

50. Id. at 5.

51. Id.

52. Id.

53. Id. at 8.

54. 33 U.S.C. § 1314(l), ELR STAT. FWPCA § 304(l).

55. See TMDLs II, supra note 10, at 10399. Essentially, § 303(d) calls for the development of TMDLs, and for their implementation from state water quality management plans under § 303(e).

56. GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 44.

57. See Perciasepe Memorandum, supra note 1; Wayland Memorandum, supra note 38.

58. See Sierra Club v. Hankinson, 939 F. Supp. 865, 27 ELR 20280 (N.D. Ga. 1996); American Littoral Soc'y v. U.S. EPA, No. 96-489 (E.D. Pa. Apr. 9, 1997); Defenders of Wildlife v. Browner, 888 F. Supp. 1005, 25 ELR 21582 (D. Ariz. 1995).

59. See Perciasepe Memorandum, supra note 1.

60. See Idaho Sportsmen's Coalition v. Browner, 951 F. Supp. 962, 27 ELR 20771 (W.D. Wash. 1996) (The court found that EPA had violated a duty under the CWA and the TMDL regulations and had acted arbitrarily and capriciously in failing to develop, with Idaho, a reasonable schedule for the establishment of TMDLs for all impaired waters. The court then directed EPA to file such a schedule by May 19, 1996, which EPA did. On September 26, 1996, the court rejected the schedule submitted by EPA and remanded to EPA with instructions to submit a new schedule within six months. On April 8, 1997, EPA submitted an eight-year schedule developed by Idaho. All parties stipulated that the schedule was reasonable.).

61. Perciasepe Memorandum, supra note 1, at 2, 3.

62. See U.S. EPA, TMDL Litigation by State (Apr. 10, 1998) http://www.epa.gov/owow/tmdl/lawsuit1.html (summarizing TMDL court decisions and consent decrees with implementation timetables ranging from 5 to 15 years).

63. Perciasepe Memorandum, supra note 1, at 4.

64. See TMDLs II, supra note 10, at 10401.

65. Id. at 10399.

66. GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 44, at 9.

67. Perciasepe Memorandum, supra note 1, at 4.

68. Id. at 5.

69. Id.

70. Id.

71. GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 44, at 15.

72. Perciasepe Memorandum, supra note 1, at 5.

73. Id.

74. 42 U.S.C. § 7410(a)(1), ELR STAT. CAA § 110(a)(1) (state implementation laws (SIPs)). For a thoughtful discussion of the analogy of TMDLs to SIPs, see Robert Adler, Integrated Approaches to the Water Pollution Problem: Lessons From the Clean Air Act (draft article on file with author).

75. LOUISIANA DEP'T OF ENVTL. QUALITY, STATE OF LOUISIANA WATER QUALITY MANAGEMENT PLAN (1987) (the plan consists of 11 "volumes" entitled: Volume 1. Continuing Planning Process Document (CPP); Volume 2. Regulations; Volume 3. Water Quality Standards; Volume 4. Basin/Segment Boundaries and Inventories; Volume 5. Water Quality Assessment; Volume 6. Nonpoint Source Assessment; Volume 7. Municipal Treatment Needs; Volume 8. Industrial Treatment Needs; Volume 9. Residual Waste Treatment Needs; Volume 10. Wasteload Allocations; Volume 11. Groundwater).

76. Perciasepe Memorandum, supra note 1, at 6.

77. See U.S. EPA, Nonpoint Source Pollution Control Program, Executive Summary 1 (visited May 15, 1998) http://www.epa.gov/owow/NPS/elistudy/execsum.hmtl ("some states … have adopted explicit statutory or regulatory exemptions for agriculture or forestry activities").

78. Perciasepe Memorandum, supra note 1, at 6.

79. 40 C.F.R. § 122.4(i) (1997) (No permit may be issued "to a new source or a new discharger, if the discharge from its construction or operation will cause or contribute to the violation of water quality standards.").

80. See Arkansas v. Oklahoma, 503 U.S. 91, 22 ELR 20552 (1992) (striking down a construction of the statute that absolutely prohibited new discharges into water quality limited waters, but upholding an EPA interpretation prohibiting discharges that "effected an 'actually detectable or measurable' change in water quality." Although applied to interstate effects in this case, the same result would seem to obtain for intrastate discharges. The difficulty here, of course, is in the proof of a "measurable" change, an exercise that plagues all water quality-based regulation and is susceptible to considerable manipulation through, inter alia, base-flow assumptions and the use of mixing zones. See Houck, supra note 5, at 10544-46. On the other hand, "any detectable contribution" is a lower burden of proof than, say, "harm"; at the least, any new source contributing pollution to a water already in violation for that pollutant would seem to carry the burden to show no such detectable contribution, which could prove to be a high standard to meet).

81. 40 C.F.R. § 122.4(i) (1997).

82. Potentially effective leverage comes from the ability of a new source to offset its loadings by trading, abating, or buying out pollution from existing sources. See GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 44, at 51.

83. 40 C.F.R. § 122.62(a)(2) (1997):

The Director has received new information. Permits may be modified during their terms for this cause only if the information was not available at the time of permit issuance (other than revised regulations, guidance or test methods) and would have justified the application of different permit conditions at the time of issuance.

84. Id. § 123.63(a)(5); see also Sierra Club v. Hankinson, 939 F. Supp. 865, 27 ELR 20280 (N.D. Ga. 1996) (citing with approval EPA's authorization under 40 C.F.R. § 123.63(a)(5) to revoke a state program for failure to perform TMDLs).

85. See RODGERS, supra note 5, at 367-68.

86. Id. at 368; see also Amendments to Requirements for Authorized State Permit Programs Under Section 402 of the Clean Water Act, 60 Fed. Reg. 14588 (Mar. 17, 1995) (EPA's response to a petition filed by the Environmental Defense Fund and the Chesapeake Bay Foundation seeking revocation of Virginia's delegated program CWA authority for failure to provide for citizen enforcement).

87. As of 1994, EPA still administered CWA NPDES programs in 10 states. U.S. EPA, NATIONAL WATER QUALITY INVENTORY: 1994 REPORT TO CONGRESS 390 (1994) [hereinafter NATIONAL WATER QUALITY INVENTORY].

88. See TMDLs II, supra note 10, at 10392-93.

89. Id. at 10393-96.

90. GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 44.

91. Under the Administrative Procedure Act, 5 U.S.C. § 553, available in ELR STAT. ADMIN. PROC., the imposition of such generally applicable and detailed requirements call for notice and comment rulemaking. United States v. Florida E. Coast Ry., 410 U.S. 224 (1973). At last report, EPA intends to move forward with TMDL rulemaking and additional guidance. Parrish Conversation, supra note 24. How much of the program emerges as enforceable rules, and how much continues as discretionary guidance, remains to be seen.

92. The importance of citizen suits in requiring EPA and the states to live up to their responsibilities under § 303(d) is apparent not only from the decided cases but from EPA guidance which explain tight compliance schedules and criteria as needed to avoid future litigation. See supra text accompanying note 13. Further, the number of times EPA personnel say to a regulated entity, "We would love to cut this corner for you but the environmentalists would take us to court," is unrecorded but, from all anecdotal evidence, and in this author's personal experience, significant. For one such piece of evidence, see Whetzel, supra note 28 (comments of EPA Region IX official); supra note 37.

93. ROUGH DRAFT TMDL REPORT, supra note 24, at 51-52.

94. National Advisory Council for Environmental Policy and Technology — Total Maximum Daily Load Committee; Public Meeting, 61 Fed. Reg. 54438 (Oct. 18, 1996) (notice of first FACA committee meeting).

95. TMDL FEDERAL ADVISORY COMMITTEE, FINAL REPORT app. B (May 20, 1998) (List of Committee Members) [hereinafter FACA REPORT].

96. See ROUGH DRAFT TMDL REPORT, supra note 24.

97. Compare for example the listing and TMDL guidance for "expected to meet waters" and "atmospheric deposition" and the development of "implementation plans," supra text accompanying notes 38-92, with the same in the FACA reports, ROUGH DRAFT TMDL REPORT, supra note 24.

98. FACA REPORT, supra note 95, at 10-11.

99. Id. at 10-14.

100. Id. at 53-57.

101. Id. at 30-34.

102. Id. at 36-40.

103. Id. at 39-40.

104. Id. at 43-46.

105. Id. at 46-52.

106. Id. app. G (Outline of the Hierarchy Approach to TMDL Approval (with examples)).

107. FACA REPORT, supra note 95, at 43.

108. ROUGH DRAFT TMDL REPORT, supra note 24, at 29-30. Section 303(d)(3) provides:

For the specific purpose of developing information, each State shall identify all waters within its boundaries which it has not identified under paragraph (1)(A) and (1)(B) of this subsection and estimate for such waters the [TMDL] with seasonal variations and margins of safety, for those pollutants which the Administrator identifies under [§ 304(a)(2)] as suitable for such calculation and for thermal discharges, at a level that would assure protection and propagation of a balanced indigenous population of fish, shellfish, and wildlife.

33 U.S.C. § 1313(d)(3), ELR STAT. FWPCA § 303(d)(3). The nonpoint industry's viewpoint is consistent with its position that, since § 303(d)(1)(A) speaks only in terms of waters impaired after the implementation of point source controls, nonpoint sources were meant to be excluded from the subsequent drill. For reasons stated in an earlier article, this author finds that argument unpersuasive. See TMDLs II, supra note 10, at 10399-400 and sources cited therein.

109. ROUGH DRAFT TMDL REPORT, supra note 24, at 52.

110. Id. at 52-53.

111. See supra note 108. A logical reading of the statute, and one that effectuates its clean water goals, is that § 303(d)(1)(A) applies to all waters still polluted after the implementation of technological standards, and that § 303(d)(3) sets up a separate process for ensuring the maintenance of waters that currently meet standards, similar to the prevention of significant deterioration program under the Clean Air Act.

112. ROUGH DRAFT TMDL REPORT, supra note 24, app. I, at 1. The intent to avoid enforcement could not be more plainly stated; in the stated point of view, any judicial enforcement is "unexpected" and "unintended."

113. Id. at 2.

114. See TMDLs II, supra note 10, at 10400.

115. See ROUGH DRAFT TMDL REPORT, supra note 24, app. I, at 3.

116. As the FACA final report hopefully words the enforceability of § 303(e) planning, "[a]ccountability mechanisms under § 303(e) could include a variety of oversight and leadership tools through which EPA generally influences state action." FACA REPORT, supra note 95, app. H, at H-1. These are, of course, the same oversight and leadership tools that have produced such unremarkable results under § 319. See David Zaring, Agriculture, Nonpoint Source Pollution, and Regulatory Control: The Clean Water Act's Bleak Present and Future, 20 HARV. ENVTL. L. REV. 515 (1996).

117. See supra text accompanying note 69.

118. Id.

119. Clean Water Act; Vice President's Initiatives, 62 Fed. Reg. 60448 (Nov. 7, 1997).

120. H.R. 961, 104th Cong. (1995). Inter alia, the bill would have extended a voluntary approach toward nonpoint pollution over the next 20 years. See TMDLs I, supra note 8, at 10343.

121. See President William J. Clinton, Remarks Made at the Living Classroom Foundation, Baltimore, Md. (Feb. 19, 1998), reprinted at U.S. EPA, Clean Water Action Plans: Restoring and Protecting America's Waters (visited Feb. 26, 1998) http://www.epa.gov/cleanwater/presrem.html.

122. See U.S. EPA, Clean Water Action Plan: Setting the Stage: Successes, Challenges, and New Directions 8-9 (visited Feb. 26, 1998) http://www.epa.gov/cleanwater/action/c1a.html.

123. Id.

124. Id.at 10.

125. Id. at 11. With respect to estuarine impairment, other EPA data rank "urban runoff/storm sewers" as the primary culprit. See infra note 181.

126. Linda Korn Levy, Assistant Secretary, Office of Water Resources, Louisiana Department of Environmental Quality, Oral Briefing on State of Louisiana Water Pollution Control Program, at Tulane Law School, New Orleans, La. (Mar. 10, 1998).

127. U.S. EPA, Clean Water Action Plan: Clean Water — The Road Ahead 9 (visited Feb. 26, 1998) http://www.epa.gov/cleanwater/action/overview.html [hereinafter CWAP — The Road Ahead].

128. For example, the Office of Surface Mining Reclamation and Enforcement in the U.S. Department of the Interior would increase by 50 percent its number of projects to clean up water polluted by acid mine drainage. EPA, Clean Water Action Plan: Actions to Strengthen Core Clean Water Programs 8 (visited Feb. 26, 1998) http/www.epa.gov/cleanwater/action/c2b.html.

129. Id. at 13.

130. U.S. EPA, Clean Water Action Plan: America's Watersheds: The Key to Clean Water 2 (visited Feb. 26, 1998) http://www.epa.gov/cleanwater/action/c3b.html.

131. Restoring Aquatic System on a Watershed Basis, CWAP — The Road Ahead, supra note 127, at 2.

132. Id. at 3.

133. Id.

134. 2 WILLIAM H. RODGERS JR., ENVIRONMENTAL LAW 127 (1986).

135. These oft-cited reasons are taken, inter alia, from Rodgers, id. at 124-27; George Gould, Agriculture, Nonpoint Source Pollution and Federal Law, 23 U.C. DAVIS L. REV. 461, 470-71 (1990); Daniel Mandelker, Controlling Nonpoint Source Water Pollution: Can It Be Done?, 65 CHI.-KENT L. REV. 479, 481-82 (1989); and EPA's position in Natural Resources Defense Council v. Train, 396 F. Supp. 1393, 5 ELR 20401 (D.D.C. 1975), aff'd, 568 F.2d 1369 (D.C. Cir. 1977).

136. See Gould, supra note 135; Zaring, supra note 116.

137. 40 C.F.R. § 125.4(j) (1975).

138. Natural Resources Defense Council v. Train, 396 F. Supp. 1393, 5 ELR 20401 (D.D.C. 1975), aff'd, 568 F.2d 1369 (D.C. Cir. 1977).

139. 568 F.2d at 1383.

140. Pub. L. No. 95-217, § 33(b), 91 Stat. 1577 (1977).

141. Pub. L. No. 100-4, § 503, 101 Stat. 75 (1987).

142. 34 F.3d 114, 24 ELR 21480 (2d Cir. 1994), cert. denied, 115 S. Ct. 1793 (1995).

143. Id.

144. Id. at 118-19, 24 ELR at 21482-83.

145. 33 U.S.C. § 1314(m), ELR STAT. FWPCA § 304(m).

146. John Burns, The Eight Million Little Pigs — A Cautionary Tale: Statutory and Regulatory Responses to Concentrated Hog Farming, 31 WAKE FOREST L. REV. 851, 851 (1996).

147. U.S. EPA, DRAFT STRATEGY FOR ADDRESSING ENVIRONMENTAL AND PUBLIC HEALTH IMPACTS FROM ANIMAL FEEDING OPERATIONS 8 (Mar. 1998) [hereinafter EPA DRAFT STRATEGY].

148. 40 C.F.R. § 122.23 & app. B (1997).

149. Id.

150. 40 C.F.R. pt. 412 (1997).

151. Id.

152. See U.S. EPA, Compliance Assurance Implementation Plan for Concentrated Animal Feeding Operations 2 (visited May 15, 1998) http://es.epa.gov/oeca/strategy.html.

153. Id.

154. OFFICE OF THE INSPECTOR GENERAL, U.S. EPA, ANIMAL WASTE DISPOSAL ISSUES, ch. 3, at 2 (Apr. 21, 1997) [hereinafter ANIMAL WASTE DISPOSAL ISSUES].

155. See Burns, supra note 146, at 859-62.

156. Id. at 862 (describing toxic effects of Pfiesteria on humans).

157. See ANIMAL WASTE DISPOSAL ISSUES, supra note 154, at 2 (citing the Report of the EPA/State Feedlot Working Group).

158. U.S. GAO, ANIMAL AGRICULTURE: INFORMATION ON WASTE MANAGEMENT AND WATER QUALITY ISSUES (1995).

159. Burns, supra note 146, at 854.

160. Id. at 851.

161. Anthony E. Ludd & Bob Edwards, It's a Dirty Business and the People Want Stricter Science Regulations, Results From Recent Surveys of Eastern North Carolina 2 (materials presented at Louisiana Environment '98: Law, Science and the Public Interest, Conference, Tulane University Law School, Mar. 6, 1998) (on file with author).

162. Id.; ANIMAL WASTE DISPOSAL ISSUES, supra note 154, at 1.

163. Margaret Kriz, Pfiesteria Hysteria, 29 NAT'L J. 1783 (1997); Margaret Kriz, Fish and Fowl, 30 NAT'L J. 450 (1998).

164. Supra note 163.

165. Id.; see also Elaine Bueschen, Pfiesteria Piscicida: A Regional Symptom of a National Problem, 28 ELR 10317 (June 1998) (describing recent outbreaks and state responses).

166. Pfiesteria Hysteria, supra note 163, at 1783.

167. As one state agency lobbyist observed, "I don't expect to see a poultry voluntary program to continue." Fish and Fowl, supra note 163, at 451 (quoting Robin Savage of the Association of State and Interstate Water Pollution Control Administrators).

168. EPA DRAFT STRATEGY, supra note 147.

169. Id. at 10-11.

170. Id. at 14, 16.

171. Id. at 15. It should be noted, however, that this EPA proposal only consolidates and strengthens the regulation of its previously described universe of large CAFOs. See supra text accompanying note 148. EPA has not (yet) proposed to expand this universe to smaller operations, although, under its regulations, it has the authority to do so on a case-by-case basis. Supra text accompanying note 149.

172. See supra note 138.

173. EPA DRAFT STRATEGY, supra note 147, at 15-16.

174. Id. at 15.

175. Id. at 16.

176. Hog Waste Spills Result in Regulatory Changes in North Carolina, in ANIMAL WASTE DISPOSAL ISSUES, supra note 154, ch. 2, at 1.

177. What Can EPA Do?, in ANIMAL WASTE DISPOSAL ISSUES, supra note 154, ch. 3, at 1.

178. See NRDC Offers Solutions to "Poison Runoff," Calls It Leading Source of Water Pollution, [20 Current Developments] Env't Rep. (BNA) 569 (July 21, 1989); see also Nonpoint Sources Found by INFORM to Be Major Contributors of Toxics to Hudson River, [18 Current Developments] Env't Rep. (BNA) 1263 (Aug. 28, 1987).

179. 55 Fed. Reg. 47990, 47992 (Nov. 16, 1990).

180. EPA's enforcement leverage begins, of course, with monetary fines, which, in the context of municipalities, is uncertain leverage even in the context of underperforming municipal sewage treatment systems. See Air and Water, in 2 RODGERS, supra note 134, at 605-09. Admittedly, on rare occasions, EPA may use the prospect of serious fines to leverage improved municipal performance. See Bill Voelker, S&WB, U.S. Make a Deal: Agency to Clean Drainage to Lake, (NEW ORLEANS) TIMES PICAYUNE, Apr. 9, 1998, at 1 (describing consent decree requiring $ 200 million cleanup of New Orleans municipal runoff).

181. The federal government has spent more than $ 54 billion on municipal public sewage treatment plants since 1972. NATIONAL WATER QUALITY INVENTORY, supra note 87, at 179.

182. EPA and state water quality officials ascribe 46 percent of estuarine impaired waters to "urban runoff/storm sewers." Id. at ES-25.

183. See 55 Fed. Reg. 47991 (Nov. 16, 1990) ("From a legal standpoint, however, most urban runoff is discharged through conveyances such as separate storm sewers or other conveyances which are point sources under the CWA."). Whether, before the 1987 CWA amendments, these point sources discharged a "pollutant" within the meaning of the CWA was an interesting question. See National Wildlife Fed'n v. Gorsuch, 693 F.2d 156, 13 ELR 20015 (D.C. Cir. 1982). Since 1987, Congress has explicitly called for their regulation under the NPDES program. 33 U.S.C. § 1342(p), ELR STAT. FWPCA § 402(p).

184. See 38 Fed. Reg. 1350 (May 22, 1973).

185. For a summary of these developments, see ROBERT ADLER ET AL., THE CLEAN WATER ACT 20 YEARS LATER 152-53 (1993), and sources cited therein.

186. 33 U.S.C. § 1342(p), ELR STAT. FWPCA § 402(p).

187. Id. § 1342(p)(3)(A), ELR STAT. FWPCA § 402(p)(3)(A).

188. Id. § 1342(p)(1) & (2), ELR STAT. FWPCA § 402(p)(1) & (2) (requiring regulation of municipal sources serving a population of 100,000 or more); id. § 1342(p)(3)(B), ELR STAT. FWPCA § 402(p)(3)(B) ("maximum extent practicable" standard).

189. Id. § 1342(p)(6), ELR STAT. FWPCA § 402(p)(6).

190. 40 C.F.R. § 122.26(c)(1) (1998). See generally Brian Weeks, Trends in Regulation of Stormwater and Nonpoint Source Pollution, 25 ELR 10300 (June 1995).

191. See 40 C.F.R. § 122.26(b)(14) (1998) (categorical permits are available to 11 classes of industry, including manufacturing, mining, utilities, landfills, transportation, and construction disturbing more than five acres of land). EPA's exception for light industry and small construction was invalidated in National Resources Defense Council v. U.S. EPA, 966 F.2d 1292, 22 ELR 20950 (9th Cir. 1992), leading to new proposals. See infra text accompanying note 196.

192. See supra note 182.

193. 55 Fed. Reg. 47990 (Nov. 16, 1990).

194. 40 C.F.R. § 122.26 (1998).

195. See OLIVER A. HOUCK ET AL., TO RESTORE LAKE PONTCHARTRAIN 82-93 (Greater New Orleans Expressway Comm'n 1989) and sources cited therein (describing urban stormwater control mechanisms).

196. See U.S. EPA, Authorization to Discharge Under the National Pollution Discharge Elimination System, Sewerage and Water Board of New Orleans et al., NPDES LA 5000301, at 5-27 (Mar. 1, 1997) (on file with author). Within a year, the stencils have largely washed away. Author's personal observation.

197. Letter of Jack V. Ferguson, Chief, NPDES Permits Branch, Region VI, U.S. EPA, to Robert R. Kuehn, Director, Tulane Environmental Law Clinic, encl. at 1 (Feb. 11, 1997) (on file with author).

198. See National Resources Defense Council v. U.S. EPA, 966 F.2d 1292, 22 ELR 20950 (9th Cir. 1992). Regulations in 1995 proposed to include virtually all remaining sources over the next six years. At the same time, EPA began consulting with a Stormwater Phase II subcommittee of a separate FACA committee addressing urban, wet-water flows. 60 Fed. Reg. 21189 (May 1, 1995); see also U.S. EPA, NPDES STORM WATER PHASE II FACT SHEET 6 (Jan. 1998) [hereinafter NPDES FACT SHEET]. As of April 1998, EPA had issued 25 notices of meetings of this FACA committee and its sub-committees. Westlaw search (May 1998). No final FACA report had issued.

199. NPDES FACT SHEET, supra note 198, at 3.

200. 33 U.S.C. § 1342(p)(6), ELR STAT. FWPCA § 402(p)(6).

201. NPDES FACT SHEET, supra note 198, at 9.

202. Id. at 20. It seems anomalous that EPA is proposing far more prescriptive regulations for smaller municipal sources than for larger ones; the Agency's rationale is that its regulation of smaller municipal sources is a one-shot affair with little subsequent monitoring, while its regulation of larger sources will be iterative and progressive. Id. at 2-4. Perhaps so. And perhaps the Agency is simply reluctant to face down large municipalities without hard water quality impact data.

203. Id. at 2-4.

204. See 61 Fed. Reg. 43761 (Aug. 26, 1996); see also U.S. EPA, Questions and Answers Regarding Implementation of an Interim Permitting Approach for Water Quality-Based Effluent Limitations in Storm Water Permits, 61 Fed. Reg. 57425 (Nov. 6, 1996). Municipalities have also questioned the fairness of blaming them for water quality violations if the receiving waters are already contaminated beyond state standards. Telephone Conversation with Peter Lehrer, Natural Resources Defense Council (June 12, 1998) (Mr. Lehrer is a member of the stormwater FACA committee, see supra note 198). Of course, this problem confronts the application of water quality standards to all point sources, and the CWA's solution is to require them to abate their pollution emissions collectively, through TMDLs.

205. See 60 Fed. Reg. 17950, 17952 (Apr. 7, 1995) (describing the President's 1994 Clean Water Initiative as "clarifying that the Maximum Extent Practicable Standard should be applied in a site specific, flexible manner taking into account use considerations as well as water quality effects").

206. 63 Fed. Reg. 1595 (Jan. 9, 1988).

207. See Debra L. Donahue, The Untapped Power of Clean Water Act Section 401, ECOLOGY L.Q. 201, 278-79 (1996).

208. National Wildlife Fed'n v. U.S. Forest Serv., 592 F. Supp. 931, 14 ELR 20755 (D. Or. 1984).

209. See Donohue, supra note 207; see also Richard H. Braun, Emerging Limits on Federal Land Management Discretion: Livestock, Riparian Ecosystems and Clean Water Law, 17 ENVTL. L. 43 (1986). Braun describes, inter alia, EPA prescriptions for livestock grazing BMPs as a component of state nonpoint source programs. Id. at 72. These prescriptions, of course, fall, as all EPA nonpoint source regulations, on the voluntary side of state implementation.

210. See Ed Marston, The Old West Is Going Under, HIGH COUNTRY NEWS, Apr. 27, 1998, at 1.

211. See Larry C. Frarey & Staci J. Pratt, Environmental Regulation of Livestock Production Operations, NAT. RESOURCES & ENV'T, Winter 1995, at 8 ("An average 1,000 pound milk cow produces approximately eighty-two pounds of wet manure per day — twenty times that of an adult human. Thus, the 225,000 cows on 275 dairies in Tulare County, California, produce over 3.4 million tons of wet manure each year.").

212. See Susan E. Schell, The Uncertain Future of Clean Water Act Agricultural Pollution Exemptions After Concerned Area Residents for the Env't v. South View Farm, 31 LAND & WATER L. REV. 113, 127 n.111 (1995).

213. See Braun, supra note 209, at 44-49.

214. See Donahue, supra note 207, at 279-80.

215. See Braun, supra note 209, at 50 n.16 (comments ascribed to the Idaho Cattle Association).

216. See Jeffrey W. Styron, Regulation of Nonpoint Sources of Water Pollution on Public Lands, 41 NAVAL L. REV. 97, 112 (1993) ("Some states apparently except logging, grazing and other nonpoint sources of pollution from complying with water quality criteria, while other states provide liberal guidelines for allowing temporary variances or modifications for nonpoint source activities." (citations omitted)).

217. The Organic Administrative Act of 1897 reads, "No national forest shall be established, except to improve and protect the forest within the boundaries, or for the purpose of securing favorable conditions of water flows, and to furnish a continuous supply of timber to the United States …." 16 U.S.C. § 475 (1994).

218. Id.

219. National Forest Management Act of 1976, Pub. L. No. 94-588, 90 Stat. 2949.

220. CHARLES WILKINSON, CROSSING THE NEXT MERIDIAN: LAND, WATER AND THE FUTURE OF THE WEST 136-37 (1992).

221. National Wildlife Fed'n v. U.S. Forest Serv., 592 F. Supp. 931, 14 ELR 20755 (D. Or. 1984).

222. Of the 400 known stocks of Pacific salmon, 214 were identified as endangered by the early 1990s; 106 were extinct. Murray D. Feldman, National Forest Management Under the Endangered Species Act, NAT. RESOURCES & ENV'T, Winter 1995, at 32, 34.

223. Id.

224. See IDAHO DIVISION OF ENVIRONMENTAL QUALITY TOTAL MAXIMUM DAILY LOAD (TMDL) PROVISIONS, PROBLEM ASSESSMENT: SOUTH FORK SALMON RIVER AND WATERSHED 5 (1991).

225. See Jeffrey St. Clair, U.S. Environment: Logging Practices Unleash Lethal Landslides, INTER PRESS SERV., Jan. 6, 1997, at 115, available in 1997 WL 7073076 ("In the [third week of November 1996] Oregon rains, landslides killed eight people. And in all but one of the cases, the slides occurred in sites that had been clear cut in the last 10 years by timber giants.").

226. See WILKINSON, supra note 220, at 136-37.

227. See Bomb Rips Building in Nevada, (NEW ORLEANS) TIMES PICAYUNE, Nov. 1, 1993, at A5 (describing a bomb explosion at Bureau of Land Management (BLM) offices); Pipe Bomber Targets Forest Service Region, LAND LETTER, Sept. 1, 1995, at 6 (bomb attack at Forest Service employee's home). Violence against environmentalists is also not unknown. See Tony Davis, A Struggle for the Last Grass, HIGH COUNTRY NEWS, May 2, 1994, at 1.

228. 36 C.F.R. § 228(b) (1997).

229. The closest such requirements in the Service's timber regulations appear in 33 C.F.R. § 219.13 (1997), but provide no quantitative measures nor do they reference state water quality standards.

230. See Schell, supra note 212, at 126-30.

231. CWA § 401 requires in pertinent part:

Any applicant for a Federal license or permit to conduct any activity including, but not limited to, the construction or operation of facilities, which may result in any discharge into the navigable waters, shall provide the licensing or permitting agency a certification from the State in which the discharge originates or will originate … that any such discharge will comply with the applicable provisions of [CWA §§ 301, 302, 303, 306, and 307].

33 U.S.C. § 1341, ELR STAT. FWPCA § 401.

232. See Oregon Natural Desert Ass'n v. Thomas, 940 F. Supp. 1534, 27 ELR 20221 (D. Or. 1996) (appeal pending); see also Paul Larmer, Judge Sends Message to Cows, HIGH COUNTRY NEWS, Oct. 29, 1996, at 4 (quoting chief counsel for the Oregon Farm Bureau who acknowledged that "it doesn't take much of a leap to realize that this could be applied to BLM lands, and to activities such as timber harvesting….").

233. 940 F. Supp. 1534, 27 ELR 20221 (D. Or. 1996).

234. No. CV 95-394-S-MHW, 1996 WL 938215 (D. Idaho Aug. 12, 1996).

235. See Oregon Natural Desert Ass'n, 940 F. Supp. at 1539, 27 ELR at 20222.

236. See EPA Drafts Regulations for Federal Nonpoint Source Activities, INSIDE EPA, Feb. 28, 1997, at 20.

237. See Captain DeRoma, Has EPA Deserted Oregon Natural Desert?, ARMY LAW., July 1997, at 43.

238. 33 U.S.C. § 1323, ELR STAT. FWPCA § 313.

239. 565 F. Supp. 586, 13 ELR 20793 (N.D. Cal. 1983), aff'd in part, vacated in part, 764 F.2d 581, 15 ELR 20682 (9th Cir. 1985), on reh'g, 795 F.2d 688, 17 ELR 20021 (9th Cir. 1986), rev'd & remanded sub nom. Lyng v. Northwest Indian Cemetery Protective Ass'n, 485 U.S. 439, 18 ELR 21043 (U.S. 1988).

240. 914 F.2d 179, 21 ELR 20023 (9th Cir. 1990).

241. 834 F.2d 842, 18 ELR 20450 (9th Cir. 1987).

242. 5 U.S.C. §§ 701-706, available in ELR STAT. ADMIN. PROC.

243. See 834 F.2d at 849, 18 ELR at 20454 (holding that a citizen suit under CWA § 505 would not lie because the violation was not of an "effluent limitation.").

244. No. EA A2-026-92-24 (Dep't of the Interior, Interior Board of Land Appeals, Nov. 18, 1996) (on file with author).

245. Telephone Communication with Thomas Lustig (Apr. 3, 1998) (Mr. Lustig is lead counsel in National Wildlife Fed'n, No. EA A2-026-92-24).

246. On the other hand, the language of § 313 could be read to require only that federal agencies comply with state standards to the extent that state agencies and private parties do — raising the question whether, if a state imposes no water quality requirements on these activities, this section requires more.

247. For a glimpse of these changing dynamics, see Dustin Solberg, Timber Town Opts for Water Over Logs, HIGH COUNTRY NEWS, Apr. 27, 1998 at 10.

248. 16 U.S.C. §§ 1531-1544, ELR STAT. ESA §§ 2-18. For the initiative of the Secretary of the Interior to get ahead of Endangered Species Act "train wrecks," see Bruce Babbitt, The Endangered Species Act and Takings: A Call for Innovation With the Terms of the Act, 24 ENVTL. L. 355, 364 (1994).

249. For a description of several administrative "ecosystem management" initiatives, see Oliver A. Houck, On the Law of Biodiversity and Ecosystem Management, 81 MINN. L. REV. 869 (1997).

250. 43 C.F.R. § 4180.1 (1995).

251. Id.

252. Id. § 4180.1(c).

253. See BUREAU OF LAND MANAGEMENT, COMPARISON OF UTAH STANDARDS AND GUIDELINES WITH THE MINIMUM ROLLBACKS CONTAINED IN THE PREFERRED ALTERNATIVE FOR THE RANGELAND REFORM '94 EIS 1-1, para. 2 (undated) (on file with author).

254. See U.S. DEP'T OF AGRICULTURE & U.S. DEP'T OF THE INTERIOR, ENVIRONMENT ASSESSMENT FOR THE IMPLEMENTATION OF INTERIM STRATEGIES FOR MANAGING ANACHRONOUS FISH-PRODUCING WATERSHEDS IN EASTERN OREGON AND WASHINGTON, IDAHO AND PORTIONS OF NORTHERN CALIFORNIA (Mar. 1994); Pacific Rivers Council v. Robertson, 854 F. Supp. 713 (D. Or. 1993), aff'd in part, rev'd in part sub nom. Pacific Rivers Council v. Thomas, 30 F.3d 1050, 24 ELR 21367 (9th Cir. 1994) (describing salmon-based forest planning).

255. This issue has been brewing since at least 1985. See New Forest Service Management Scheme Proposed by Conservation Advocates, LAND LETTER, Aug. 14, 1985, at 1.

256. See Todd Wilkinson, Forest Service Seeks a New (Roadless) Road to the Future, HIGH COUNTRY NEWS, Apr. 27, 1998, at 8; see also Enhance Natural Resources Stewardship, CWAP — The Road Ahead, supra note 127, at 5.

257. Enhance Natural Resources Stewardship, CWAP — The Road Ahead, supra note 127, at 5.

258. Id.

259. See Wilkinson, supra note 256 (identifying "powerful opponents" in Congress as including congressional committee chairs Rep. Helen Chenoweth (R-Idaho), Sen. Frank Murkowski (R-Alaska), Sen. Larry Craig (R-Idaho), and Rep. Don Young (R-Alaska)).

260. ARIZ. REV. STAT. ANN. § 49-202.01 (1988).

261. As one commentator has observed:

Federal agencies can accomplish a great deal toward the goal of reduction of nonpoint source pollution if they are willing to diligently develop the best BMPs for a given area during the operations phase. The biggest piece of the regulations puzzle that is missing at present is the lack of federal enforcement in monitoring the effectiveness of BMPs on public lands.

Styron, supra note 216, at 113.

262. Mark Schlefstein, "Dead Zone" in Gulf Biggest in Decade, (NEW ORLEANS) TIMES PICAYUNE, July 28, 1995 at B-12.

263. 16 U.S.C. §§ 1451-1465, ELR STAT. CZMA §§ 302-319.

264. H. REP. No. 101-535, at 9 (1990).

265. Id.

266. See Carol Kaesuk Yoon, A "Dead Zone" Grows in the Gulf of Mexico, N.Y. TIMES, Jan. 20, 1998 at B-1.

267. 16 U.S.C. § 1455b, available in ELR STAT. CZMA.

268. For a description of the CZMA and its implementation, see Oliver A. Houck & Michael Rolland, Federalism in Wetlands Regulation: A Consideration of Delegation of Clean Water Act Section 404 and Related Programs to the States, 54 MD. L. REV. 1242, 1294-99 (1995).

269. 16 U.S.C. § 1452(1), ELR STAT. CZMA § 303(1).

270. Id. § 1452(2), ELR STAT. CZMA § 303(2).

271. See Houck & Rolland, supra note 268, at 1297-98.

272. 16 U.S.C. § 1455b(b)(1) & (2), available in ELR STAT. CZMA.

273. Id. § 1455b(b)(7), available in ELR STAT. CZMA.

274. Id. § 1455b(b)(3), available in ELR STAT. CZMA.

275. Id. § 1455b(g)(5), available in ELR STAT. CZMA.

276. U.S. Department of Commerce & U.S. EPA, Coastal Nonpoint Pollution State Program Guidance Documents, 58 Fed. Reg. 5182, 5184 (Jan. 19, 1993).

277. Id.

278. Id.

279. 16 U.S.C. § 1455(d)(16), ELR STAT. CZMA § 306(16).

280. Id. § 1455b(a)(1), available in ELR STAT. CZMA (30 months for state submission); id. § 1455b(c)(1), available in ELR STAT. CZMA (six months for review and approval).

281. Id. § 1455b(c)(3), available in ELR STAT. CZMA.

282. Clare F. Saperstein, State Solutions to Nonpoint Source Pollution: Implementation and Enforcement of the 1990 Coastal Zone Amendments Reauthorization Act Section 6217, 73 B.U.L. REV. 889, 890 (1995).

283. See U.S. Dep't of Commerce & U.S. EPA, Availability of Proposed Administrative Changes to Coastal Nonpoint Pollution Programs Guidance, 63 Fed. Reg. 12078 (Mar. 12, 1998) [hereinafter Availability of Proposed Administrative Changes].

284. Id. at 12078.

285. Id.

286. Id.

287. Telephone Conversation with Stuart S. Tuller, Office of Water, U.S. EPA (Feb. 25, 1998) [hereinafter Tuller Telephone Conversation].

288. U.S. EPA, GUIDANCE SPECIFYING MANAGEMENT MEASURES FOR SOURCES OF NONPOINT POLLUTION IN COASTAL WATERS (Jan. 1993) (a bible of data on and prescriptions for nonpoint sources running over 1,000 pages, separate chapters address agricultural, forestry, urban, marina and recreation, and hydro modification sources; wetlands systems; and monitoring).

289. Id. at 4-12.

290. See COASTAL MANAGEMENT DIV., LOUISIANA DEP'T OF NATURAL RESOURCES & OFFICE OF WATER RESOURCES, LOUISIANA DEP'T OF ENVTL. QUALITY, LOUISIANA'S COASTAL NONPOINT POLLUTION CONTROL PROGRAM (Oct. 1995) [hereinafter LOUISIANA'S COASTAL NONPOINT POLLUTION CONTROL PROGRAM]. The document does reveal rather significant gaps in statutory enforcement authority over, for example, agricultural "erosion and sediment control," "grazing," and "nutrient management" and forestry "preharvest planning," "road management," "timber harvesting," and "revegetation in distributed areas." Id. at I-5. Which raises the question: what's left?

291. COASTAL MANAGEMENT DIV., LOUISIANA DEP'T OF NATURAL RESOURCES & OFFICE OF WATER RESOURCES, LOUISIANA DEP'T OF ENVTL. QUALITY, SUMMARY: LOUISIANA COASTAL NONPOINT POLLUTION CONTROL PROGRAM 10/10/95 (on file with author).

292. Letter of Joseph A. Utravitan, Acting Chief, Coastal Programs Division, Office of Ocean and Coastal Resource Management, NOAA, to Terry Howie, Director, Coastal Management, Louisiana Department of Natural Resources, and Glenda Levy, Administrator, Office of Water Resources, Louisiana Department of Environmental Quality (Aug. 22, 1997), and attachment COASTAL MANAGEMENT DIV., LOUISIANA DEP'T OF NATURAL RESOURCES & OFFICE OF WATER RESOURCES, LOUISIANA DEP'T OF ENVTL. QUALITY, LOUISIANA COASTAL NONPOINT PROGRAM FINDINGS AND CONDITIONS, DRAFT FINAL 8/5/96.

293. COASTAL MANAGEMENT DIV., LOUISIANA DEP'T OF NATURAL RESOURCES & OFFICE OF WATER RESOURCES, LOUISIANA DEP'T OF ENVTL. QUALITY, LOUISIANA COASTAL NONPOINT POLLUTION CONTROL PROGRAM, COMMENTS IN REPLY TO DRAFT FINDINGS AND CONDITIONS, DRAFT FINAL 8/5/96 (on file with author).

294. See U.S. EPA, California Nonpoint Source Program, Final Draft California Coastal Nonpoint Program Findings and Conditions (visited May 15, 1998) http//www.epa.gov/region09/water/nonpoint/cal/finaldraft.html.

295. Northern Sec. Co. v. United States, 193 U.S. 197, 400-01 (1904) (Holmes, J., dissenting) ("These immediate interests exercise a kind of hydraulic pressure which makes what previously was clear seem doubtful, and before which even well settled principles of law will bend.").

296. See Comment Period for Nonpoint Source Guidance Extended as Industry, States Question Feasibility, [22 Current Developments] Env't Rep. (BNA) 1795 (Nov. 22, 1991).

297. See Letter of H. Wayne Beam, Chairman, Coastal States Organization, to Carol M. Browner, Administrator, U.S. EPA, and Dr. James Baker, Undersecretary of Commerce, U.S. Department of Commerce (Dec. 5, 1994) (cited in Saperstein, supra note 282, at 921 n.117).

298. See Availability of Proposed Administrative Changes, supra note 283 (describing a January 6, 1995, letter and the March 16, 1995, document).

299. Written Testimony Received at Public Meetings, 2 LOUISIANA'S COASTAL NONPOINT POLLUTION CONTROL PROGRAM, supra note 290, § G.

300. Id.

301. Written Comments Received After Public Meetings, 2 LOUISIANA'S COASTAL NONPOINT POLLUTION CONTROL PROGRAM, supra note 290, § H.

302. Id.

303. Id.

304. Id.

305. See Tuller Telephone Conversation, supra note 287.

306. Availability of Proposed Administrative Changes, supra note 283.

307. Id.

308. Id.

309. Id. at 12079.

310. Id.

311. Id.

312. Id.

313. The closed-door nature of this guidance continues; in April 1998, EPA and NOAA scheduled a "workshop" with coastal states, and undisclosed "other partners," to discuss the implementation of the program and of this notice; at the request of the states who wanted to be able to "shoot from the hip without broadcasting to the world," members of the public were not allowed to attend. Tuller Telephone Conversation, supra note 287. The contrast between this process and that of the open, FACA meetings on the TMDL program is striking.

314. States Drafting New Position on Key Clean Water Act Program, INSIDE EPA, Aug. 22, 1997, at 3 (describing meeting of Association of State and Interstate Water Pollution Administration, August 1997).

315. Oregon, an early subject of TMDL litigation, has turned out to be a strong proponent of the TMDL program. See States Lack Resources Needed to Implement TMDL Strategy, EPA Told, Daily Env't Rep. (BNA), May 12, 1997, at A-8 (describing Oregon support for EPA requirements). For a description of TMDLs catching up with nutrient loadings into Long Island Sound, see Ann Powers, Reducing Nitrogen Pollution on Long Island Sound: Is There a Place for Pollutant Trading?, 23 COLUM. J. ENVTL. L. 137 (1998). For a website listing of other state TMDL programs, see http://www.epa.gov/owow/tmdl/links.html (Maryland, Montana, Oregon, South Dakota, and Washington).

316. For a comprehensive and current summary of state nonpoint programs, see U.S. EPA, Nonpoint Source Pollution Control Program (visited May 15, 1998) http://www.epa.gov/owow/NPS/elistudy/ [hereinafter Nonpoint Source Pollution Control Program]; see also U.S. EPA, Section 319 Nonpoint Source Success Stories, No. 841-S-94-004 (visited May 15, 1998) http://www.epa.gov/owow/NPS/sec319.html/. For a comprehensive state approach, see North Carolina Div. of Water Quality, Draft Plan, Neuse River Nutrient Sensitive Waters (NSW) Management Strategy (July 12, 1996) (unpublished, on file with author). For an indication that state programs are not so substantive, however, see the following description: "Virginia continues to develop watershed projects that combine educational programs and demonstrations, technical assistance, financial assistance and water quality monitoring … Over 600 students participated in an educational program for all local schools, which included t-shirts and logo and poster contests." http://www.epa.gov/owow/cgi-bin/imagemap/owow-usa?433,203. The narrative continues by describing a "pilot total maximum daily load program," which turns out to be a study leading to the following remarkable conclusion: "although more research is required, the completed project shows a correlation between land use changes and change in water quality." Id. It sounds like money well spent.

317. See supra note 77.

318. See State "No More Stringent" Laws, in Nonpoint Source Pollution Control Program, supra note 316, app.

319. States Drafting New Position on Key Clean Water Act Program, supra note 314.

320. For a recent assessment of state TMDL compliance, see NATIONAL WILDLIFE FED'N, POLLUTION PARALYSIS: STATE INACTION PUTS WATERS AT RISK (Oct. 9, 1997). Comparing the performance of 53 states and territories in TMDL lists and plans, the report ranks no programs as "good," 17 as "weak," 16 as "poor," and 20 as "failing." Id. at 17.

321. See Water Pollution: States Lack Resources Needed to Implement TMDL Strategy, EPA Told, supra note 315, at A-8 (describing state concerns with funding and timing; for example,

Bruce Anderson, an official with the Hawaii Department of Health, said in an April 14 letter, "in a period of declining state budgets, public demand for no new taxes, and resistance to regulation, we do not expect to obtain sufficient funding to establish scientifically-defensible numeric targets for polluted runoff control"

see also Western Governors' Forum Outlines States' TMDL Strategies; More Money Needed, Daily Env't Rep. (BNA), Oct. 2, 1997, at A-7 ("Let's get more money from EPA now," said Dennis Hemmer, director of the Wyoming Department of Environmental Quality. Hemmer said he appreciated the opportunity to exchange information with other states on implementing TMDLs, but he said the main issue common to all states is "wanting more money.").

322. See supra text accompanying note 127.

323. See States Lack Resources Needed to Implement TMDL Strategy, EPA Told, supra note 315, at A-8 (comment of Hawaii official); see also supra note 321.

324. See 33 U.S.C. § 1251(b), ELR STAT. FWPCA § 101(b) (recognizing the "primary responsibilities and rights of states" to abate water pollution); see also TMDLs I, supra note 8.

325. TMDLs II, supra note 10, at 10397.

326. U.S. EPA, Status of 1998 Section 303(d) Lists 5/15/98 (visited May 15, 1998) http://www.epa.gov/owow/tmdl/tmdlmap.htm.

327. See ROUGH DRAFT TMDL REPORT, supra note 24, at 10-11.

328. See Susan Bruninga, Time Allowed for States to Do TMDLs Too Little for Adequate Job, Group Told, Daily Env't Rep. (BNA), Mar. 10, 1998, at A-8.

329. E-mail Communication from Nina Bell, Northwest Environmental Advocates, to Ray Gorning (Mar. 10, 1998) (on file with author).

330. See Bruninga, supra note 328 (the referenced state is Kansas). Not to be outdone in any race-to-the-bottom, Louisiana has recently relaxed its water quality standards for fecal coliforms, limiting the application of strict criteria to summer months on the stated (and amazing) rationale that primary contact recreation does not occur in Louisiana at other times of the year. Jennifer Anderson, Water Quality, LA. ENVTL. COMPLIANCE UPDATE, May 1998, at 5.

331. See TMDLs II, supra note 10, at 10397-98.

332. See Bruninga, supra note 328. Of these 33,000 cases of impairment, 1,000 were identified as impaired by pesticides. Id.

333. Id.

334. Parrish Conversation, supra note 24. As of May 1998, EPA was projecting a new total of 22,851 waters which, if borne out, would represent a 50 percent increase over 1996. U.S. EPA, SUMMARY STATUS OF 1998 § 303(d) LISTS MAY 29, 1998 [hereinafter SUMMARY STATUS] (on file with author).

335. The lack of comparability may be illustrated by projected listings in the state of Washington at 4,400 and in the state of Louisiana — an equally wet and agricultural state — at a mere 176. SUMMARY STATUS, supra note 334.

336. See supra text accompanying note 61. As of late May 1998, EPA had received the required TMDL schedules from 40 states; 33 states projected TMDL completion within the prescribed 13-year term, while 7 projected taking from 14 to 22 years. These disparities may well reflect relative difficulties in the task among the various states; they most certainly as well reflect differences in commitment to the TMDL process.

* For a discussion of the current dispute over mechanisms for implementation, see the preceding text at supra text accompanying notes 93-117.

** This TMDL presents a system of trading of pollution rights among point and nonpoint dischargers in order to attain nutrient reduction goals. Widely heralded as an example of regulatory flexibility, the approach is further described, inter alia, in note 316 (Neuse River Nutrient Sensitive Waters (NSW) Management Strategy Draft Plan).

*** This TMDL addresses a pollutant from a single point source that, through process changes, has ceased discharging the pollutant.

28 ELR 10415 | Environmental Law Reporter | copyright © 1998 | All rights reserved