32 ELR 10385 | Environmental Law Reporter | copyright © 2002 | All rights reserved

The Clean Water Act TMDL Program V: Aftershock and Prelude

[Editors' Note: This Article will be included in a forthcoming Second Edition of the Environmental Law Institute's monograph THE CLEAN WATER ACT TMDL PROGRAM: LAW, POLICY, AND IMPLEMENTATION by Oliver A. Houck. For further information and to order a copy of the Second Edition, call (800) 433-5120 ((202) 939-3844 in the Washington. D.C. area) or e-mail orders@eli.org.]

Oliver Houck is a Professor of Law at the Tulane University. The research assistance of Andrew L. Adams, LL.M. 2002, is acknowledged with gratitude.

[32 ELR 10385]

The U.S. Environmental Protection Agency (EPA) is in the process of redesigning the Clean Water Act's (CWA's) total maximum daily load (TMDL) program.¹ Section 303 of the Act requires states and, if necessary, EPA to: (1) identify waters that do not meet water quality standards; (2) establish the TMDLs for pollutants discharged into these waters that will achieve these standards; and (3) incorporate these loads into state planning.² These are of course the classic steps of ambient-based water quality management.

[32 ELR 10386]

Ambient-based management has not worked well in any media—air, water, or waste.³ It requires enormous amounts of data. It requires analysis that is rarely definitive and nearly always litigable. It launches a process that never ends. These same factors plagued the CWA's predecessors and the similarly constructed Clean Air Act (CAA). They have prolonged and frustrated decisions under federal pesticide, toxic substances, safe drinking water, hazardous waste, and Superfund laws as well. EPA's task with TMDLs is not an easy one, then, under the best of circumstances.

The circumstances of TMDLs are not ideal. The Agency has a skimpy statute to work with, and the Clinton Administration's efforts to flesh it out met with a firestorm of opposition. EPA now faces states on red alert for anything that sounds like a federal environmental requirement, and is well below baseline in their ability to staff, finance, maintain, or enforce a program of this complexity. It also faces powerful resistance from major industries that have, to date, enjoyed virtual immunity from the CWA, and from their allies in the U.S. Congress and the current Administration.⁴ In January 2002, the Office of Management and Budget, after private meetings with industry, placed TMDLs on a short list of "outdated or outmoded" rules.⁵ It did not need to. The old rules were already dead, and EPA had given sufficient notice that the program would be changed.

This Article attempts to summarize the events leading to the adoption of final TMDL rules in the year 2000, and recent studies, agency initiatives, and litigation that will affect the new program to come. It concludes with an assessment of the program now forming, and with recommendations that would assist such a program in reaching the CWA's clean water goal.

I. The Showdown

The year 2000 saw a dramatic showdown between Congress and the executive branch over TMDLs. In retrospect, it was unavoidable. The Agency, which was trying to build a carrot-and-stick program, ran into constituencies implacable in their opposition to sticks, too powerful to need to compromise, and with ready access to Capitol Hill. What ensued was a high-stakes clash of power in which EPA won the battle but its opposition, strongly backed by Congress, won the war.

The showdown concerned EPA's first ever, comprehensive regulations for the TMDL program.⁶ The Agency and the states had been operating under skeletal regulations originating in 1978 that repeated requirements of the statute but provided little additional guidance.⁷ As written earlier, the stage was set for underperformance, and underperformance happened.⁸

Shocked into motion by citizen suits in the early 1990s, EPA struggled to get ahead of district court mandates with a flurry of TMDL guidance to the states that fleshed out preliminary requirements for impaired water listings and for the contents and schedules for TMDLs themselves.⁹ At the same time EPA convened a Federal Advisory Committee Act (FACA) committee with representation from state water agencies, agribusiness, timber, industrial point sources, municipal sources, environmentalists, Native Americans, academics, and, ex officio, EPA and the U.S. Department of Agriculture (USDA). Its mission was to forge a consensus on program regulations.¹⁰

Consensus was not in the cards. At bottom, the timber and agriculture industries were not going to accept that CWA § 303(d) covered nonpoint sources or that, if it did, it required implementation plans. To these industries, the CWA's only lawful vehicle to abate their discharges was the voluntary, grant-in-aid program under § 319. TMDLs smacked of first ever regulation, a causa belli.¹¹ A Farm Bureau Federation representative wrote that the program had been "hijacked by a vast national bureaucracy of parasites."¹² An article in the fall 1999 issue of Range Magazine called TMDLs and the Administration's supporting Clean Water Action Plan "perhaps the most anti-agricultural document [32 ELR 10387] ever produced in Washington, DC."¹³ Before EPA had even proposed regulations the hype was reaching best available technology (BAT) levels for the English language.

EPA, for its part, could not let the issue slide. Environmental lawsuits were setting its agenda, case by case, state by state, with differing and at times highly demanding schedules for impaired waters listings and TMDLs.¹⁴ The Agency and the states were making decisions on the basis of guidance memoranda¹⁵ that had never been subjected to notice-and-comment rulemaking under the Administrative Procedure Act (APA)¹⁶ and that carried no force of law.

In July 1999, EPA proposed draft, comprehensive regulations for the TMDL program¹⁷ and launched a series of meetings, briefing sessions, and telethons in an attempt to calm the waters. To no avail. Most states and affected industries remained adamantly opposed. It is likely that in the tenor of the times—anti-government sentiment at a high fever and anti-Clinton sentiment in Congress even higher—no regulatory effort would have met a different reception. At which point EPA's choices were to drop nonpoint sources and implementation plans from the program or face the consequences. Drop them it would not do. Instead, the Agency softened other aspects of its proposal, clarified that TMDL implementation plans could be satisfied through "voluntary means and education," lengthened the deadlines for TMDL submissions, and hoped for the best.¹⁸

The rulemaking took a year. More than 34,000 comments were received, which, even discounting mass-mailings from industry trade associations, indicated no small controversy.¹⁹ At the same time, with little confidence in the administrative process, states and nonpoint industries turned up the heat and took their case to Congress.²⁰ They were warmly received. Congressional committees held field hearings stocked with farmers and small woodlot owners afraid, in some cases told, that EPA was going to require permits for the use of their land.²¹ Committees held hearings on Capitol Hill in which their mistrust of EPA was so marked that one took the unprecedented step of insisting that Agency witnesses provide their testimony under oath.²² Committee members accused EPA of "secret meetings" with environmentalists.²³ Learning that an Undersecretary of Agriculture had cooperated with EPA in developing a timetable for nonpoint source TMDLs, a representative from Arkansas inserted a budget rider in the agriculture appropriations bill removing the Undersecretary from authority in the matter.²⁴ The legislation passed.²⁵

By April 2000, battered by the House and Senate, EPA began adding concessions to its most vocal opponents. It would drop its proposal to include "threatened waters" in TMDLs, despite its belief that attending to these waters early would save time and money later on.²⁶ It would drop a requirement that gave waters used for human consumption and for endangered species priority attention, despite its belief that these priorities were required by law.²⁷ It would drop a process allowing the public to petition for review of TMDL decisions, despite its belief that administrative review was preferable to litigation.²⁸ It would promise flat out that forestry practices would not be subject to national pollutant discharge elimination system (NPDES) permitting "under any circumstances."²⁹ And at seven congressional hearings, as if the white flag on these issues were not fully visible, the Agency stressed that it was trying to enhance flexibility and leave TMDL implementation to the states.³⁰

Too much flexibility, for some parties. Following publication of the proposed rules in 1999 and continuing into the early months of 2000, several environmental groups, including lead TMDL litigating groups, became disenchanted with EPA's concessions. They saw in theproposals a recipe for endless delay, well beyond the deadlines they were establishing through citizen suits.³¹ They also saw little assurance in implementation programs based on voluntary measures and education. On another side, however, were environmental groups that, having litigated over lists and [32 ELR 10388] TMDL schedules, were now engaged in their states on implementation issues. In their view, without regulations setting implementation standards, the rest of the TMDL process—listings, schedules of submissions etc.—even on accelerated schedules, led to no productive end. The environmental community split, some seeing the doughnut, others the hole.³²

In May 2000, the split broke open when six national environmental organizations wrote the EPA Administrator calling on her to scrap the regulations and go back to the drawing board.³³ The letter could not have caught the Agency at a more vulnerable moment. Already under attack from the states, industry, and Congress, EPA was left without a single, unified constituency, while its opponents gained the argument that here was a rule that nobody liked.³⁴ Such an argument is persuasive in Congress; if nobody likes it, there are no political consequences to ditching the rule. Months later, seeing the entire program under vigorous attack on Capitol Hill, several of the dissenting groups switched sides to support the rule, but the political damage had been done. With the exception of a few environmental groups, a few of the more progressive states, e.g., California, Maryland, and Vermont,³⁵ and the Association of Metropolitan Sewerage Agencies (who saw TMDLs for nonpoint sources as a way to begin to share the burden of cleaning up impaired waters that otherwise fell largely on their shoulders),³⁶ EPA was on its own.

In full cry now, Congress hosted a competition of bills seeking to derail EPA's regulations. The lead vehicle, Senate Bill 2417, labeled with no apparent irony the "Water Pollution Program Enhancements Act of 2000," characterized the regulations as "hasty," "unscientific," "one-size-fits-all," and "unlikely to improve water quality."³⁷ It would put the EPA program on hold pending studies on its supporting science and on the effectiveness of alternative, voluntary programs and their costs.³⁸ Other bills would simply repeal the TMDL rules in full or exempt particular industries, e.g., forestry, from their scope.³⁹

Meanwhile, the Administration, convinced that it could not persuade Congress that its process over the previous three years had been anything but hasty, that its regulations were reasonable, and that an approach preserving state authority to determine where, what, and how pollution would be abated was hardly "one-size-fits-all,"⁴⁰ began a race to beat the clock. Passing whatever bill Congress chose would take time. EPA accelerated its regulations review.

Outraged, now, by what appeared to be a stiff-arm of its concerns and an end-run on its authority, Congress looked for another vehicle to block the Administration. They found it in the Fiscal Year (FY) 2001 Military Constructions/FY 2000 Urgent Supplemental Appropriations Bill, an obscure funding measure that in neither House contained any mention of TMDLs.⁴¹ Overnight and without notice, Congress tacked on a provision that no funds would be used in FY 2000 or 2001 to "make a final determination on or implement any new rule relative" to EPA's regulations.⁴² It further mandated a study of the program by the National Academy of Sciences (NAS) to determine the adequacy of the scientific basis for the TMDL program.⁴³ Among other things, the study would delay the program; even better, the study might derail it.

As appropriations for expenditures that were unavoidable and in some cases already expended, the supplemental appropriations bill was must-sign legislation.⁴⁴ It was also, in a decade marked by environmental-legislation-by-appropriations-rider, a high watermark for this political art.⁴⁵ President William J. Clinton appeared check-mated; in order to cover, inter alia, monies spent for hurricane relief in Nicaragua, he would have to sign away the TMDL rule. A small window of opportunity remained, however, a gap of several days before which, with or without the president's signature, the bill would become law.⁴⁶ The EPA Administrator signed [32 ELR 10389] off on the regulations on July 11.⁴⁷ The final rule was published on July 13.⁴⁸ The same day, the last available, President Clinton signed the appropriations bill⁴⁹ prohibiting half of what his Administration had just done.

In the final regulations, EPA acknowledged the appropriations rider by delaying their effect until October 31, 2001, or until the expiration of the rider, whichever came first. As a practical matter, with a ban on funding program implementation, it would be delayed until whichever came second. In the meantime, EPA's skeletal regulations from the 1980s and subsequent guidance would remain in effect. The program would continue under the old rules. Environmentalists left the field bloodied but relieved that the TMDL program had not been repealed. The agriculture and timber industries left triumphant but insecure in their victory. The threat of TMDL-imposed requirements on nonpoint sources had been parried but it had not been eliminated.

II. The Aftermath

On the heels of the TMDL showdown came the national presidential elections, and, when the dust finally settled, a new philosophy toward environmental protection. EPA Administrator Christine Todd Whitman was a governor from a state with a proven, if not unmixed, capability and record in pollution control,⁵⁰ one of the more successful examples of federalism in a country where federalism can also simply mean the abandonment of national goals.⁵¹ While TMDL issues continued to play out in the courts and percolated below the surface of a still-suspicious Congress, the lead action, although not all of the action, rebounded to EPA. We are a year away from a new TMDL program. From EPA's recent initiatives and statements, however, the contours of the new program are evident. The sections that follow attempt to summarize what is happening and likely to happen over the next year—the critical period in which the program jells.

A. Sobering News

If the flap over EPA's regulations produced no final resolution, it stimulated four studies on the underpinnings of the TMDL program. One was a state-of-readiness survey by the U.S. General Accounting Office (GAO) assessing the capabilities of water regulatory agencies to assess and manage water quality. In a second, previously mentioned study, the NAS would ascertain the state of the science supporting TMDL decisionmaking. A third study by EPA would project the total short- and long-term costs of the TMDL program. A fourth, by EPA's Inspector General, audited state water program enforcement. Taken individually, they provide answers to the questions asked. Taken together, they raise a larger question that is far more difficult to answer: will this dog really hunt?

1. State Capacity

As described in an earlier discussion of TMDLs, ambient-based water quality management has always been severely challenged by its demands for current, continuous, and definitive data.⁵² The information needed to support water quality criteria, to identify polluted waters, to determine the causal pollutants, to identify the sources of these pollutants, to sift out background, natural, atmospheric, and other exogenous sources, to quantify these loadings, and to apportion load reductions among these sources, all in ways that are defensible in a court of law (since any significant restrictions will be challenged), was a primary reason Congress abandoned ambient-based management in 1972 and turned to BAT.⁵³

In the years that followed, the picture did not improve. A 1984 EPA survey of state water quality management needs identified more than 30 "difficulties" in state capacity, including: "lack of knowledge of how to do realistic modeling of a pollutant's concentration and fate in a water body"; "lack [of] data on the water quality of a given site"; and "lack [of] data necessary to determine the design discharge of the receiving water."⁵⁴ A 1987 Office of Technology Assessment report found that "only limited data are available on ambient pollutant concentrations in receiving waters, variability in these concentrations, and the fate of these pollutants and their impacts on indigenous organisms," and that "our ability to monitor water quality in relation to potential environmental or human impacts is relatively primitive."⁵⁵ A 1993 GAO report found that even the U.S. Geological [32 ELR 10390] Survey faced "formidable data management challenges" in assessing natural water quality because, inter alia, efforts to collect, analyze, and store data were so costly and labor intensive.⁵⁶ A 1999 report issued by former EPA and state agency employees called state monitoring and assessment "variable," "manipulated," and "erroneous"; a game of "politics, burcaucratic incrtia and bad science."⁵⁷ Reports on the similar CAA program have consistently found the same.⁵⁸

Which brings us to today. In March 2000, the GAO released its study of the adequacy of the information underlying the evaluation of national water quality and pollution control strategies.⁵⁹ Because the federal CWA program relies on state assessments, the report focused on state capabilities, surveying all 50 states and probing the programs of 4 representative states in more detail.⁶⁰ In sum, the GAO found wide inconsistency and incomplete information on existing water quality, the foundation information for everything that is supposed to follow.⁶¹

According to the GAO, as of the year 2000, a majority of the nation's waters remain, in practice, unmonitored and unassessed, even by the most primitive of indicators. Assessment levels range from 72% for estuaries, to 40% for lakes, ponds, and reservoirs, to 19% for rivers and streams, and to 6% for oceans shoreline waters.⁶² The numbers are worse than they look. Only one-half of the 19% of assessments for rivers and streams was performed by actual monitoring; the other one-half by estimates, extrapolations, and unspecified "other means."⁶³ Further, the monitoring may be infrequent, e.g., monthly or, for pesticides, annually, for only a few contaminants, e.g., oxygen levels, solids, and bacteria, and not representative of more serious pollution.⁶⁴

The states themselves were equally pessimistic about their data-readiness. Only 6 states reported sufficient information to "fully assess" state waters, and only 18 reported sufficient information even to identify waters as polluted for TMDL listings.⁶⁵ Moving on down the TMDL program requirements, while 40 states reported confidence in their ability to identify point sources of pollution (not that hard a trick, given that point source loadings are characterized in each NPDES permit), only 3 reported confidence in their ability to identify nonpoint sources, the sources primarily at issue in the TMDL program.⁶⁶ Similarly, while 29 states felt that they had sufficient data to develop point source TMDLs, only 3 had data sufficient to develop nonpoint source TMDLs.⁶⁷ In other words, for the type of pollution for which TMDLs are most important—that from crops, cattle, clearcutting, roadbuilding, and suburban and urban construction—only 3 states of 50 have enough data, or will admit to having enough data, to act.

This lack of basic information—historic, present, and incontrovertible—has always placed state water quality regulators in a tactical bind. On the one hand, the facts speak for themselves. On the other hand, state water quality agencies have fought tenaciously since the 1960s for the primary responsibility to manage the nation's waters.⁶⁸ Ambient-based programs give them the power to determine water use categories, a threshold and essentially political decision that allows low water quality in states with little inclination to face pollution dischargers.⁶⁹ It is easier to declare the use of a somewhat polluted river to be "secondary contact recreation," e.g., boating, than to be "primary contact recreation," e.g., swimming, and then have to clean it up.⁷⁰ States have additional leeway under ambient management programs to allocate pollution cleanup-responsibilities among dischargers, a leeway that carries its own, obvious political clout. In effect, the CWA's NPDES program robbed states of both authorities.⁷¹ It further starved technocracies of state water engineers, private consultant firms, and water research institutions of federal and state funding for data collection, modeling, and analysis, chores that—in a program based on ambient impacts and constant adjustments to those impacts—never end. Authority and funding are strong motivators, and it has been no surprise to see states consistently assert before Congress complete confidence in their ability to manage ambient-based programs and at the same time assert poverty when it comes to the tools—raw data, trained personnel, assessment techniques, and modeling—needed to make them work.⁷²

Over time, wearied of technology-standard battles, EPA has joined the state chorus. In 1987, with amendments to the CWA launching new ambient-based programs for toxic dischargers [32 ELR 10391] and inaugurating the § 319 nonpoint source program, EPA's Assistant Administrator for Water would testify to Congress:

Before the 1972 law, you'd get into these long, long debates with dischargers who would say[:] "No let me prove to you this isn't a problem." We've learned something in the past 20 years. Our monitoring technology is much better than it was 20 years ago. So the combination of the much better information and the much better permitting and enforcement base means that I think we have a chance we didn't have then.⁷³

Whether this testimony was actually believed, wishful thinking, or simply expedient in an era when EPA was looking to shed environmental responsibility to the states, time would show otherwise.

In October 2001, notwithstanding the GAO's findings, EPA would pronounce:

In recent years, EPA and the States have made great strides in implementing the existing [§]303(d) program to list impaired waters and develop and implement TMDLs. States have substantially improved their TMDL programs while the Agency has provided the States with significant increases in technical and financial support to expand and strengthen all elements of their programs.⁷⁴

There is something about the phrase "making great strides" that, like the phrase "full and frank discussion" in the field of diplomacy, seems to confirm that the goal, if attainable at all, is still a long way away.

2. Supporting Science

As noted earlier, in July 2000, Congress not only deferred the regulatory program until further notice but mandated a study by the NAS as well. The requested points of study were the sufficiency of knowledge about point and nonpoint sources and the state of monitoring and modeling needed to predict pollution loadings and develop TMDLs.⁷⁵ The study began in January 2001, and was due in June—tall orders for a short amount of time.⁷⁶ In effect, the report would be written by a few individuals, and, as it happened, ones with strong, preexisting points of view on the TMDL program.

On one level, the NAS report is an affirmation of the TMDL program. The message on the adequacy of the supporting science was blunt: the science is there, its uncertainty can be reduced, and uncertainties that remain should not be used as an excuse for delay or inaction.⁷⁷ In words that echo those of water quality administrators in the 1960s and EPA's Assistant Administrator for Water in the 1980s, the committee reported that "the data and science have progressed sufficiently over the past 35 years to support the nation's return to ambient-based water quality management."⁷⁸ These messages may not have been welcome news to those in Congress who saw "bad science" as a reason to oppose EPA's TMDL regulations.⁷⁹ They seem to have been as unwelcome to—or unread by—EPA, which quickly used the National Research Council (NRC) report as a reason to suspend its regulations into the year 2003.⁸⁰

Buried not very deeply in the report, however, were three additional themes that cast a longer shadow over the program. The first relates to the technology and personnel demands of adequate ambient-based management. The science might be there in theory, but in practice it certainly was not, either in models, data collection, training, or state capacity.⁸¹ Further, the committee added two technology demands that threaten to push the current state-of-unreadiness over the edge. One was an insistence on greatly increased monitoring in order to produce a sufficient number of current and statistically accurate samples that would, in turn, reduce uncertainty and inconsistency in decisionmaking and provide a basis for "adaptive management," i.e., management responding to changed and changing conditions.⁸² Criticizing EPA and state reliance on static and insufficiently comprehensive monitoring,⁸³ good science would require increased and constantly maintained data collection. At which point, given an already taxed system producing minimal data on only 19% of the nation's waterways, we may be at the limits of funding and political will.

The other high-tech demand came with the committee's support for biological monitoring and biological criteria.⁸⁴ As a theoretical matter, few could contest that the living environment of an aquatic system is a more complete measure of its health than are numerical, chemical criteria; the real [32 ELR 10392] thing is always better than a partial surrogate.⁸⁵ As a practical matter, however, biological monitoring is infinitely more complex in what is getting monitored and how impacts are measured, e.g., the number of minnow per one-quarter mile and their relative states of health, than the use of chemical or physical parameters, e.g., 0.04 micrograms/liter of oxygen, 0.10 parts per million of dissolved solids. Recognizing this fact, the report recommends the use of both chemical and biological monitors in order to assure scientific accuracy.⁸⁶ At this point we may have crossed the limits of funding and political will.

A second theme of the report, indeed the central theme to those who chaired and wrote it, was the recommendation that the TMDL process approach the identification of impaired waters in two stages, a preliminary stage of eligible waters and a second stage of formal listing.⁸⁷ The data requirements for eligible waters would be broad and permissive, including water quality data from several years back and additional data such as fish kills or fish advisories (for public health purposes).⁸⁸ No legal consequences would flow from this preliminary list, however, other than additional monitoring. For the formal listing a rigorous set of continuous data would be required proving water quality impairment to a 90% confidence level.⁸⁹ The advantage of a two-tiered listing process would be final listings that are more resistant to litigation, more persuasive to pollution dischargers, and more focused on badly polluted waterways. The minus would be the potential for a limbo of eligible-but-unproven waters—a phenomenon already witnessed in the CAA and Endangered Species Act—that escape attention or treatment.⁹⁰

The report cast a third shadow over the TMDL program by recommending, repeatedly, that states redetermine their water standards and uses for each impaired water body before TMDLs are performed.⁹¹ It explained:

It does not follow that a water body lacking integrity [, e.g., polluted,] is impaired or that restoring biological integrity is either possible or desirable. A water body that is described as lacking "biological integrity" should not be assumed to be in a less-than-desirable state.⁹²

What we have here is an open invitation to lower environmental protection. Were the invitation overlooked, a sidebar in the report entitled, "Six Reasons for Changing Water Quality Standards"⁹³ provides a recipe for dropping a designation from, say, primary contact recreation to secondary contact recreation when utilities, the sugar industry, or pulp and paper manufacturers dig in their heels.

The report concludes that, "the scientific foundation for adaptive implementation must rely on state initiative and leadership."⁹⁴ "Today," it notes with no approval, "EPA retains an extensive oversight" of the TMDL program⁹⁵; instead, the report's approach will "require increased state assumption of responsibility for individual TMDLs, with EPA oversight focused at the program level instead of on each individual water segment."⁹⁶ One does not have to be a proponent nor an opponent of devolution to recognize that at this point the report is not treating the adequacy of science to support the TMDL program. It is taking sides in a political debate over federalism.

In the final analysis, reports are written by people. Under normal circumstances, NAS reports are written over several years, by scientists with little financial or political connection to the issues before them, and with a balance of points of view.⁹⁷ As the NRC staff director of this report has complained, however, this one was otherwise, a rush job with little time for panel building or, for that matter, report writing.⁹⁸ The result was a small group with an attitude. The panel chair, a witness for the Florida Pulp and Paper Association in hearings on the above-mentioned Florida rules who is also reported as having accused EPA of "heavy handed 'intrusion'" in North Carolina TMDLs, has stated that he "jumped at it [the chairmanship], because it was a great opportunity to make some of the arguments I've been making"; he is also quoted as referring to two subsequent TMDL-related consultant contracts as "an opportunity to push my agenda."⁹⁹ Florida had its own representative on the committee as well, who happened to write the section endorsing Florida's two-tiered program, a "great job on the committee" according to his state agency supervisors.¹⁰⁰ [32 ELR 10393] The report's principal author¹⁰¹ had, just two months before the committee geared up, published an article through the Cato Institute entitled The Trouble With Implementing TMDLs¹⁰² —a title remarkably similar to that of the final chapter of the NRC report¹⁰³ —expressing the fear that EPA would turn the TMDL program into a NPDES-like permit system.¹⁰⁴ Such an extension would be "costly, contentious, environmentally suspect, and often inequitable."¹⁰⁵ Warming to a close, he concluded: "Congress should explicitly affirm that the ends of the Act are to secure ambient water quality goals, not to eliminate all discharge."¹⁰⁶ Thus concluded the NRC report as well.¹⁰⁷ Congress, of course, decided otherwise in the CWA of 1972,¹⁰⁸ but the argument lingers on.

At bottom, a small group of industry consultants, state agency representatives, and water quality engineers have affirmed their faith in an ambient-based water quality program. In response to Congress' question, they found that TMDLs are supportable by science—lots more of it. Which leads to the next question: costs.

3. Costs of the TMDL Program

As it put EPA's final TMDL regulations on hold in July 2000, Congress also directed EPA to study the total costs of implementing such a program.¹⁰⁹ In December 2000, EPA announced the study and published a call for information in the Federal Register.¹¹⁰ Fewer than one-half of the states submitted data, the majority on listing and load allocation costs, which was about as far as most states had progressed.¹¹¹ Based on these data and the contributions from industry and other participants in a few (very costly) TMDLs, EPA took its best shot and published a draft report for public comment in August 2001.¹¹² The comment period closed at the end of 2001,¹¹³ and a final report is expected in the spring of 2002. While the final version is not available, its basic content, like that of the TMDL program itself, is visible.

The scope of the work was defined by the 1998 state lists of impaired waters under CWA § 303(d), the last of such submissions. About one-third of the nation's waters were polluted, including 300,000 miles of rivers and shoreline and 5 million acres of lakes.¹¹⁴ After 28 years of pollution control efforts under the CWA of 1972 and another 20 years of previous federal assistance programs, nearly 22,000 water bodies were identified as not meeting state water quality standards.¹¹⁵ These waters would require an anticipated 36,000 TMDLs.¹¹⁶

Turning to the sources of pollution, state data identified less than 5% of impairment from point sources only and about 25% from a combination of point and nonpoint sources.¹¹⁷ Nonpoint sources were the exclusive sources of pollution for 50% of the listed waters, with the remaining waters polluted by combinations of nonpoint and "other" sources.¹¹⁸ The leading source of impairment was agriculture at 24.6%, with another 11.4% attributed to unspecified "nonpoint" causes.¹¹⁹ The three leading pollutants were agriculture-driven: sediments, pathogens, and nutrients.¹²⁰ In overview and subject to regional exceptions, e.g., forestry in the northern Rockies and acid mine drainage in the Appalachians, the task at hand would be more than 20,000 TMDLs for nonpoint sources, primarily for agricultural runoff.

EPA estimated the costs for impaired water listings, for TMDL load calculations, and for implementation in two stages: state costs in TMDL development, and source costs in their implementation. The state costs, more easily calculated, were found to be between $ 63 and $ 69 million per year nationwide, which amounts to approximately $ 1 billion over the next 10 to 15 years to get the job done.¹²¹ Another $ 17 million a year would be needed for additional water quality monitoring required by the program.¹²² While many knowledgeable commenters will have their crack at these estimates, the assumption that the anticipated load of TMDLs can be handled for a little over $ 1 million a year per state seems optimistic, and the assumption that only an additional $ 300,000 or so per state would meet the levels of monitoring and data called for in the NRC study seems optimistic as well.

[32 ELR 10394]

EPA approached source implementation costs under three "scenarios."¹²³ The first, obviously disfavored and something of a stalking-horse for the two that followed, was described as the "least flexible TMDL program" model and based on requiring—either through NPDES permits or by aggressive "inducement" under CWA § 319 and other funding programs—all point and nonpoint sources to abate their discharges by the "next treatment step."¹²⁴ The second scenario, the "moderately cost-effective" model, would more finely calculate the reductions to be achieved by the "next treatment steps" and allocate them in a way that "let off" some or many dischargers and did not "overshoot" the water quality goal.¹²⁵ The third and favored approach, labeled "more cost-effective," played with the second scenario to assign reductions to sources that have low reduction costs either through targeted allocations or by emissions trading.¹²⁶

Under EPA's more cost-effective scenario, TMDL implementation was estimated to cost $ 900 million to $ 3.2 billion per year.¹²⁷ Under the disfavored least flexible scenario, annual costs would range between $ 1.9 and $ 4.3 billion per year.¹²⁸ Two noteworthy facts appear from the data. The first is that under all scenarios, both point and nonpoint sources took approximately equal cost hits, although nonpoint sources were the overwhelming sources of pollution; i.e., point sources, already regulated, present marginal cost savings compared to agriculture and other sources that have been heretofore requirement-free. The second fact is that even with the most optimistic, cost-minimizing, market-oriented trading imaginable, the high-end projections for the best and worst scenarios were at $ 3.2 and $ 4.3 billion, respectively, which is not all that far apart. In other words, you can save money by trading, up to a point, but it is no panacea.

As this Article is being written, EPA is closing its comment period on the cost analysis, but the comment it has received from all sides indicates that the estimates are low. A joint coalition of industrial dischargers and municipal wastewater treatment facilities calling itself the "Federal Water Quality Coalition" criticized several assumptions relative to costs to its members,¹²⁹ as has the American Forest and Paper Association, which feared that "likely administrative and political pressures" would require "a much broader application of the TMDL program" to forest industries.¹³⁰ The American Farm Bureau took issue with the best management practices that EPA defined as "cost-effective reductions."¹³¹ A conservative think tank offered annual cost ranges of $ 2.5 to $ 5.26 billion.¹³² Even an environmental representative stated: "I think they have undershot their numbers somewhat."¹³³

The commentators probably have undershot the numbers, but so what? No one familiar with the soaring costs of a public works project or a home improvement contract can be surprised by undershooting. On the other hand, with a little more experience and a little less resistance from the states and industry, the numbers could come way down. The first can openers are always the most expensive.

In the final analysis, what is striking about cost estimates for the TMDL program is not the extent to which they differ but rather their common ground. Cost estimates for implementation of the CAA Amendments of 1990 were at odds by a factor of 10. If, all told, TMDL cost estimates come in at somewhere between $ 25 to $ 50 billion over 15 years, that is a pretty strong concurrence for a large federal program. Whether Congress is willing to fund such an exercise over the long haul is something very different. In May 2001, the chairman of the NRC' study was testifying on Capitol Hill against a proposed $ 30 million cut in the Geological Survey's toxic substances hydrology and water quality assessment programs,¹³⁴ which provide a starting point for ambient water quality management. Congress may or may not listen, and it can always listen differently from year to year, Ambient-based programs from start to finish are a very expensive way to go.

4. State Enforcement

Effective, state-based ambient water quality programs depend at the front end on raw data and, at the far end (following use designation, standard-setting, monitoring, modeling, impact analysis, load reduction negotiations, trading, permitting, reporting, and inspections) on enforcement. Under the CWA, 44 states have assumed enforcement authority along with all other aspects of delegated programs,¹³⁵ with EPA financial assistance and oversight. In practice, with the exception of significant violations or citizen suits, most water program enforcement begins and ends with the states. Here, too, comes the question of whether state institutions are up to the job.

The question is not new. For the past 20 years, reports by the GAO, the EPA Inspector General, academics, and the press have assessed state environmental law enforcement programs and concluded that they lacked adequate monitoring of regulated entities, failed to act in a timely fashion when violations were noted, imposed minor and ineffective sanctions, failed to recover the economic benefit gained from noncompliance, and varied widely from state to state.¹³⁶ The CWA audits have been particularly damning. [32 ELR 10395] During 1992-1994, for example, 18 to 27% of all major CWA-regulated facilities were in significant noncompliance.¹³⁷ From October 1998 to January 2000, 26% of major facilities were in significant violation, 159 of them in significant violation throughout the entire 15-month period.¹³⁸

Neither is the question one of personnel. There is no reason to believe that were federal and state enforcement personnel to replace each other, enforcement results would differ. The problems reflect a chronic shortage of resources, made more significant as states assume increasing responsibility for environmental programs.¹³⁹ They also reflect inconsistency in federal oversight from administration to administration, from EPA region to region, and from the ambiguity in knowing when to intervene.¹⁴⁰ They also reflect differences in the philosophy of compliance—carrots or sticks—that pervade all environmental law.¹⁴¹ And at bottom they reflect the gut-level difference in opinion over the importance of environmental protection so notable between, say, states of the northern tier and the deep South. Unfortunately, the facts underlying the U.S. Supreme Court's recent CWA standing decision in Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc.¹⁴² —a state enforcement order drawn up by the violating discharger itself and imposing sub-minimal sanctions for the violations—was no anomaly. State enforcement actions are characterized by concessions to dischargers,¹⁴³ and a tougher line risks political consequences. The Louisiana Secretary of Environmental Quality was dismissed by the governor for bringing an enforcement action against a hazardous waste incinerator.¹⁴⁴ Enforcement is hard, state agencies are human, and they are very close to the heat.¹⁴⁵

In 2001, EPA's Office of the Inspector General conducted another audit of state CWA enforcement programs to determine whether they adequately "protect the environment and human health."¹⁴⁶ The published results concluded in polite language, e.g., state enforcers "could be much more effective in deterring noncompliance,"¹⁴⁷ that the states were wanting in every aspect of enforcement from identifying violators, to taking enforcement actions, to the assessment of penalties and fines.¹⁴⁸ More particular shortfalls included: a basic lack of data on dischargers; serious, unreported toxic discharges; incomplete identification of stormwater dischargers; enforcement actions delayed by a year or more following violations; and penalties that failed to recover the economic gains from noncompliance.¹⁴⁹ Nothing new here; the old patterns continue. Of more direct concern to the TMDL program, however, the report noted a lack of focus on all but major industrial point source dischargers, basic gaps in water quality data, incomplete compliance data, a lack of compliance data for watershed-based programs, and a "reluctance" on the part of states to address small business on the one hand and "economically vital business and industries" on the other.¹⁵⁰ The reader is left to wonder, with these categories excluded, what entities were getting addressed.

Two aspects of the Inspector General report are particularly relevant to TMDLs. The first is that much of the audit focused on point source dischargers, all of whom are covered by permit requirements designed to facilitate oversight and enforcement. The limits are fixed, derived from national standards, committed to writing, usually numerical, and compliance is self-reported and routine. From an enforcement standpoint, this is the easy stuff. Yet even in this universe, only 10 states reported a compliance rate of 90% or better with permit requirements during FY 2000; 20 states did not even reach 75% compliance, where a finding of noncompliance required 2 significant permit violations repeated within 2 consecutive quarters, i.e., egregious conduct.¹⁵¹ In more than one-third of the states, over one-half of the facilities in significant violation in 1999 were significant [32 ELR 10396] violators in 2000.¹⁵² Even for point source permitting, state enforcement is not securing compliance.¹⁵³

Of equal relevance is the finding that, beyond major point source dischargers, there was less enforcement for several (more diffuse) categories of dischargers, including concentrated animal feeding operations (CAFOs) and urban runoff, that had received increased attention under the CWA because of their impacts on water quality.¹⁵⁴ Permit systems were in place but even basic monitoring and reporting was lacking.¹⁵⁵ The problem here was not a reluctance to sanction violators but, rather, an unwillingness even to find out who the violators were. CAFOs and urban stormwater are, of course, the two CWA permit categories most similar to the nonpoint runoff at issue with TMDLs.

Viewed as a whole, the Inspector General's report presages more trouble for an effective TMDL program. TMDLs are an ambient-based strategy whose enforcement depends not on self-reported compliance with numerical limits based on national standards but, rather, on management practices and impacts on receiving waters that are subject to interpretation and contest. If states as a whole have difficulty enforcing the NPDES program and the monitoring requirements contained in municipal stormwater and CAFO permits, it is not easy to see how they will do better with the more diffuse ambient game. Thirty years ago, the demands of proof inherent in ambient violations and the demands of political will led to little state enforcement at all.¹⁵⁶ Congress' answer in 1972 was to provide for EPA oversight and, with dramatic success, citizen enforcement as well.¹⁵⁷ Citizen suits are what brought the TMDL program out of remission.¹⁵⁸ It remains to be seen whether they will continue to exercise this same influence on its implementation.

B. EPA and the Regulatory Progrant

In August 2001, citing the need to build a "foundation of trust" among stakeholders and a "more workable program," EPA proposed to delay for 18 months the effective date of the final TMDL rule.¹⁵⁹ A new rule would be proposed in 2002 and adopted in 2003. Environmentalists booed. Agriculture cheered. Nobody was in the least surprised. It was a given to all concerned that with the arrival of a new Administration, [32 ELR 10397] the TMDL regulations as written were history and it was open season again on the issues involved.

1. The July 2000 Rule

The Clinton Administration regulations might well be history, but they are also a useful benchmark from which to measure what will come. The final rule dropped required listings of "threatened" waters not yet fully impaired, listings of waters experiencing "pollution," e.g., from whatever cause, as opposed to those affected by "pollutants," e.g., only from discharges, pollution offsets for new dischargers into impaired waterways, separate forestry controls, and an administrative process for citizen challenges.¹⁶⁰ The periods for reporting were extended from 2 years to 4, and the deadlines for TMDL development were extended to up to 15 years.¹⁶¹ EPA, further, would not act to develop a TMDL in the place of a defaulting state unless the state failed to show "reasonable progress" toward its development.¹⁶² Each of these issues, some of which were vital to individual environmental organizations (and in particular the deadlines),¹⁶³ was conceded to states and discharger industries.

What the final rule retained was the core of the program the Agency had been shaping the previous decade: (1) TMDLs would cover nonpoint sources; (2) would include a plan for implementing the necessary load reductions; and (3) would include reasonable assurances that the proposed reductions would be implemented.¹⁶⁴ Such assurances could include voluntary market-and incentive-based initiatives, so long as they applied specifically to the water body at issue and would be implemented expeditiously, through reliable mechanisms, and with adequate funding.¹⁶⁵ Thus, the bones of contention would remain, as they had always been, nonpoint sources and implementation plans.

2. The Listening Sessions

EPA's first action was to take the kettle off of boil. It set up a round of "listening sessions" in five major cities, ostensibly to elicit more "stakeholder perspectives" on "key issues" of the program.¹⁶⁶ While each session was associated with a set of themes, e.g., "are TMDLs appropriate for all impaired waters and pollutants?",¹⁶⁷ their panels and attendance were heavily weighted with state water program administrators and nonpoint industries who, whatever the assigned topic, produced familiar sounding recommendations, e.g., that EPA had no business regulating nonpoint sources; that TMDLs should not be implemented until other nonpoint programs had been given "time to work"; that TMDLs needed to avoid being "specific" and "inflexible"; that "functional equivalents" to TMDLs were sufficient; and that the process needed "off ramps" (a more expedited process) to delist impaired waters.¹⁶⁸ There were few new perspectives. It is hard to imagine, after all that TMDLs have been through, that there could be perspectives that were new. Viewed optimistically, the sessions served to blow off steam and provide an extra measure of participatory democracy for EPA's new program design. Viewed realistically, they provided a record, were any needed, for softening the program.

3. Monitoring and Assessment Guidance

Concurrently EPA moved to propose new guidance to facilitate state acceptance of TMDLs, each containing a nugget of what is likely to appear in EPA's new program. In November 2001, the Agency issued the 2002 Integrated Water Quality Monitoring and Assessment Report Guidance, intended, inter alia, to provide more flexibility for state listing decisions.¹⁶⁹ It allowed states to list several categories of impairment, including one for waters with "insufficient data" to determine impairment and another for those whose impairment is "not known."¹⁷⁰ Waters categorized with insufficient data or other unknowns would not require TMDLs. Applauded by state agencies dealing with large numbers of waters and small amounts of data,¹⁷¹ the guidance was viewed skeptically by environmentalists who envisioned a [32 ELR 10398] black hole for data-short, impaired waters into which many, if not most, polluted water bodies would fall.¹⁷²

Removed from the final version of the guidance, but perking along on its own burner within the Agency, is the consolidated assessment and listing methodology (CALM), a recipe intended to provide more consistent and reliable data collection and reporting.¹⁷³ Originally presented as part of the guidance along with the carrot of an additional $ 25 million in state grants toward these ends,¹⁷⁴ CALM ran into rough sledding with state officials fearing that it could impose "baseline" conditions and, worse, include "requirements" for water quality monitoring programs.¹⁷⁵ The Association of State and Interstate Water Pollution Control Administrators issued a position statement that it did not support the use of "critical elements" as a "strict evaluation checklist" against which the "adequacy of monitoring programs is determined."¹⁷⁶ States to EPA: no requirements allowed.

4. The CWA § 319 Program

The most significant EPA action and the most indicative of its future intentions was its September 2001 Supplemental Guidelines for the Award of Section 319 Nonpoint Source Grants to States and Territories in FY 2002 and Subsequent Years, which was intended to provide a "concentrated focus" for its voluntary, grant-based, nonpoint source pollution control program.¹⁷⁷ The concentrated focus would be TMDLs. The memorandum began by reminding the states of their oft-stated position that § 319 "provides an appropriate and effective programmatic framework" for nonpoint pollution and for the implementation of TMDLs.¹⁷⁸ TMDLs would provide the "analytical link" between "actions on the ground and the water quality results to be achieved."¹⁷⁹ The memorandum then dropped, or rather, phased in, a soft hammer: starting in FY 2002, and more fully effective in FY 2003, EPA would require the dedication of a significant portion of § 319 funding to the development of nonpoint source TMDLs and to watershed plans implementing them.¹⁸⁰ Approved § 319 plans would require "reasonable assurance" that the load allocations identified in the TMDLs would be achieved.¹⁸¹ The TMDLs themselves would require a source-based quantification of load reductions, specified implementation measures, an identification of technical and financial assistance needed, a "reasonably expeditious" schedule for implementation, and a description of interim milestones, monitoring, and criteria to determine if "reasonable progress" is being made.¹⁸² The ingredients should look familiar. These are, nearly jot-for-jot, the requirements of the suspended TMDL rule¹⁸³ transferred to the voluntary § 319 program.

[32 ELR 10399]

The American Farm Bureau was not pleased. In a stinging letter to EPA's Administrator in October 2001, the bureau protested that EPA had overstepped its bounds in using "the power of the purse" to force states to develop nonpoint source TMDLs.¹⁸⁴ The bureau found EPA's guidance "wholly inappropriate" given the litigation surrounding TMDLs and their application to nonpoint sources and indicated, joined by the National Cattlemen's Association and others, that a legal challenge to the guidance was a possibility.¹⁸⁵ It is clear that the agriculture industry, having stalled if not won the war on TMDLs and implementation plans for nonpoint sources under CWA § 303(d), was now turning to fight TMDL-based grant-in-aid programs to states under § 319. When you are a major polluter, the threats just keep on coming.

5. Trades and Markets

On another front, by late 2001 and early 2002, EPA was preparing a new policy statement outlining market-based approaches, including a trading system, for CWA programs including TMDLs.¹⁸⁶

Emissions trading is of course not new and was a central theme of the Clinton Administration. In 1996, EPA issued an Effluent Trading in Watersheds Policy¹⁸⁷ and a Draft Framework for Watershed-Based Trading¹⁸⁸ identifying, inter alia, the social benefits of trading as "encouraging dialogue among stakeholders and fostering concerted and holistic solutions."¹⁸⁹ The 1996 policy required trading to result in "an equivalent or better water pollutant reduction."¹⁹⁰ The Devil, of course, resides in the details.

EPA's renewed program is said to use state water quality standards as an emissions "cap" similar to the sulfur dioxide caps under the CAA acid rain program.¹⁹¹ While any analysis of EPA's proposal needs a proposal to analyze, pollution control programs have sufficient experience with trading to know where the upsides and downsides lie. The upsides lie with cost savings, economies of scale, and political acceptance.¹⁹² The downsides lie with what gets traded for what, in practice, and whether things really get cleaned up in the bargain.

A threshold question is who will want to trade, and the answer is those parties that would otherwise be compelled to expend greater costs in abating their pollution directly. In the current CWA scheme, those parties are industrial and municipal point sources operating under national technology standards. What they will be willing, indeed eager, to trade for is less-expensive abatement from nonpoint sources. Unless nonpoint sources are under a legal obligation to abate their discharges, however, these sources will have no incentive to buy less costly reductions from other nonpoint or point source players. Under a TMDL program that imposes no enforceable abatement requirements on nonpoint sources, therefore, trading only runs in one direction, from point to nonpoint sources. The danger is that the trades become escape hatches for point sources from having to meet the technology standards that have or that are being met by their competitors.¹⁹³ The only way to close such a hemorrhage would be to require full BAT compliance for point sources as a precondition for trading whatever additional compliance measures are imposed on them by TMDLs,¹⁹⁴ and to impose specific abatement obligations on nonpoint sources as well, something that EPA at the moment does not seem disposed to do.

A second difficulty with trading schemes is that, even for single-source pollutants, they often upon closer scrutiny turn out to be bogus.¹⁹⁵ In a practice so widespread that it has earned its own label—"gaming"—claims are made for wholly past reductions, for the same reductions several times, and for reductions that never happened, happened only ephemerally, or never will happen.¹⁹⁶ The monitoring, supervisory, and accounting costs of trading schemes are formidable and beyond the capacity of at least some states.¹⁹⁷ The state of Louisiana's accounting program for nonattainment ozone trading has become so hopelessly confused and so completely reliant on the accounting practices of industrial applicants for reduction credits that EPA was forced to disavow it¹⁹⁸; the U.S. Department of Justice is currently auditing New Jersey's emissions trading program [32 ELR 10400] for possible CAA violations.¹⁹⁹ It should be kept in mind that these problems arose over emissions from discrete point sources, identifiable by concentration and volume, of single precursor pollutants. Tracking the reductions, proposed and real, from nonpoint sources across a watershed will present another order of difficulty.

One preview of this additional difficulty is the mitigation program under CWA § 404²⁰⁰ in which wetland values destroyed by development programs are to be offset by the acquisition and improvement of other wetlands.²⁰¹ Both nonpoint and § 404 trading exchange "harm X" for "benefit Y"; the "X" is known but the "Y" is speculative and depends upon the on-the-ground application, success, and long-term maintenance of land use measures that will, in the one case, abate pollution and, in the other case, enhance wetlands values. In practice, the mitigation/trade measures—the improvement of streamside vegetation, for example—may be identical. As studies consistently show, the problem is that the Y mitigation is not implemented, does not work well, or does not work at all.²⁰² The success rate for wetland mitigation projects hovers near 50%.²⁰³

A final aspect of trading that glimmers within EPA's new trading initiative is its announcement of an "enhancement of economics" in its water programs.²⁰⁴ What this term apparently includes is the economic valuation of aquatic resources—"what's a loon worth?" was offered as an example²⁰⁵ —an exercise that has confounded resource economists since the 1970s and presents a significant problem in the natural resources damages programs of the U.S. Department of the Interior under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)²⁰⁶ and the U.S. Coast Guard under the Oil Pollution Act.²⁰⁷ Whatever a loon or a river is said to be worth in dollars, the inquiry in a permitting context opens a Pandora's box for decisions based not only on a comparison of costs to impacts but on a comparison of costs to benefits²⁰⁸ that can never be adequately assessed.²⁰⁹

In short, while trading among sources has always been inherent in ambient-based management systems and indeed is touted as one of their major virtues, the formalization of a market in pollution rights raises serious issues of trading obligations, baselines, performance, and accountability that will need to be addressed. The war over permits and enforcement will not end; it will simply obtain new venues.

6. Water Quality Standards

EPA is also pursuing several initiatives to shore up the foundation for the ambient water quality program—state water quality standards. It is a tricky problem. On the one hand, the existing standards are in most cases out of date and vary widely in their content and application.²¹⁰ Identical industries in adjoining states may face permit limits for the same pollutant that differ by more than 10,000 times.²¹¹ Until recently these differences did not matter because water quality standards did not matter; few states bothered to apply them. TMDLs, however, now put pressure on the whole foundation. On the other hand, reopening water quality standards reopens questions of baselines, costs, and competing economic and political interests that could well weaken the criteria, standards, antidegradation policies, and other safeguards of the program.²¹²

In 1998, the Agency issued a notice of proposed rulemaking on water quality standards across the board, asking a number of questions about the most critical issues with standards and criteria.²¹³ Perhaps realizing that the rulemaking had opened too large a can of worms, or perhaps [32 ELR 10401] overtaken by events more specifically related to TMDLs, this generic rulemaking has been on a slower track.²¹⁴ Spurred forward by the NRC report,²¹⁵ however, the new rules will likely facilitate, if not require, a wholesale, nationwide "use attainability" analysis for all impaired waters,²¹⁶ a recipe for accepting degradation that many states will not miss.²¹⁷ Unaddressed, meanwhile, are issues that have chronically plagued the CWA and its predecessors, including the high degree of play in mixing zones, application factors, variances, antidegradation policies, and enforcement.²¹⁸ There is no more uneven playing field in federal environmental law than water quality management. To the extent that EPA follows its current polestar of flexibility in state programs, the more that water quality standards matter the more this unevenness will increase.

These problems noted, a special challenge presented to water quality standards by TMDLs has been the absence of meaningful water quality criteria for the primary impacts of the primary pollution science, nutrient loadings from agriculture.²¹⁹ The question is difficult enough technically, as the relationship of loadings to impacts vary with local environments.²²⁰ It is also difficult politically, as shown by the agriculture industry's effective resistance to measures to reduce nutrient loadings to the Mississippi River.²²¹ EPA's current nutrient guidance, issued informally over the last two years, provides for maximum state flexibility.²²² As with all other national water quality criteria, EPA developed suggested limits for nutrient-related pollutants, e.g., phosphorous, nitrogen, chlorophyll-a, and turbidity.²²³ It then departed from its normal practice by dividing the nation into 14 "eco-regions," each to adopt its own criteria, from which the states would then derive their standards. States could adopt regional numerical criteria, narrative criteria, or any other "scientifically defensible" method. State standards were to be inplace by 2004.

Obviously shy of creating another tempest, EPA will not require states to adopt nutrient management plans, although such plans were "strongly encouraged."²²⁴ Whatever comes beyond "encouragement," therefore, will come from the TMDL program.

7. Focus Groups

A reader diligent enough to have pursued the topic to this point is entitled to some lighter reading, and in this vein we consider the EPA Nonpoint Source Pollution Focus Group Final Report.²²⁵ The report is an outgrowth of the Nonpoint Source Management Partnership, a collaboration between EPA and the states to "identify, prioritize and solve nonpoint source problems."²²⁶ The problem here was image, and EPA turned to a consultant organization to get to the bottom of it.

The consultant organized focus groups in four major cities across the United States.²²⁷ Two groups, divided by age, were assembled at each site and were asked to identify the visibility of nonpoint source pollution and to suggest measures to increase public awareness of the problem.²²⁸ By overwhelming consensus, the groups concluded that nonpoint source pollution had near zero visibility, and that even the term "nonpoint source pollution" was a non-starter: "It doesn't tell you anything"; "it sounds like there's nothing you can do."²²⁹

By way of remedy, the focus groups recommended a "bold, hard-hitting and provocative" advertising campaign featuring videos, music, and the techniques of successful television advertisements.²³⁰ Among the more specific suggestions:

. Disgusting works; the grosser the better."²³¹

[32 ELR 10402]

. "Have Titanic hit a great big pile of trash, or oil cans."²³²

. "Britney Spears with a gas mask on."²³³

. "For me, everything today has to be hot girls: bathing suits, Maxim, and beer commercials."²³⁴

Exactly how EPA works these suggestions into its collaboration with stakeholders and a more flexible, state-based TMDL program remains to be seen.

8. The State of the Game

As the year 2001 drew to a close, the states were still using their approved 1998 lists of impaired waters identifying 36,000 water bodies that failed to meet water quality standards for at least one pollutant.²³⁵ Some states, spurred forward by environmental litigation, had begun to make significant investments in TMDL implementation: $ 5 million in Iowa, $ 10 million in California.²³⁶ Other states, as described in an earlier article, have resorted to a variety of mechanisms to avoid their responsibilities, including revision of antidegradation policies, downgrading, use unattainability, insufficient data, and this author's favorite, "swamp waters," to trim their workload.²³⁷ For the waters remaining, EPA had approved almost 4,000 TMDLs submitted by the states, a pace of action that will only accelerate.²³⁸

Of course, what matters more than the volume is the content. A review of 55 TMDLs by this author in 1998 found that the early approved submissions met few of the requirements of the statute and Agency guidelines and regulations.²³⁹ Variously, they did not quantify overall loadings, did not quantify loadings from individual sources, neither identified nor quantified nonpoint sources, focused almost exclusively on reductions from point sources, and made little explicit provision either for margins of growth or margins of error; only a few contained implementation plans, and none provided objective assurance that the necessary load reductions would be attained.²⁴⁰ While a similar analysis of the second-wave TMDLs is beyond the reach of this Article—and perhaps beyond the reach even of EPA—a December 2001 study of TMDLs in West Virginia produced similar findings.²⁴¹ The study reviewed 25 TMDLs prepared by EPA and the state for a range of point and nonpoint sources, among them current and abandoned mining operations and agriculture.²⁴² Not surprisingly, the point sources received specific abatement proposals while the nonpoint sources did not. While new nonpoint source programs had been initiated in four impaired watersheds, "no concrete steps toward implementation had been undertaken in the remaining seven watersheds."²⁴³ No nonpoint source water TMDLs contained implementation plans.²⁴⁴ Virtually all anticipated nonpoint source abatement was to be paid for through federal funding and supplemented by state funding.²⁴⁵

Large-scale water quality improvement projects show the same characteristics: long on goals, short on requirements and deadlines. Three of the most famous and ambitious restoration efforts in the country are those for the Florida Everglades, the Gulf of Mexico, and the Chesapeake Bay. In each case, years of planning has led to agreement on pollution reduction goals; a 40% reduction in nutrient loading in the case of the Chesapeake,²⁴⁶ and in the case of the Gulf of Mexico,²⁴⁷ a desirable outcome of 30% reductions achieved by upstream states—the word "goal" was apparently too strong for the participants. Neither program contains enforceable requirements or time lines.²⁴⁸ In the case of the Everglades, the year 2002 saw the first regulations for a $ 7.8 billion, 30-year, federal-state restoration plan criticized by environmentalists for its lack of specifics and praised by U.S. Sugar as "fluid."²⁴⁹ By contrast, restoration planning [32 ELR 10403] for water quality in the San Joaquin delta of California has moved forward with significant abatement requirements from large irrigators and other dominant water users in the region.²⁵⁰ Then again, the San Joaquin delta contains the Delta Smelt, an endangered species. When we have to, we get serious and abate.

What we have so far from § 303(d) in practice, then, is a process in which TMDLs provide goals with varying levels of specificity. What we also have is a marked dichotomy between the treatment of point and nonpoint sources in the exigence of load reductions and in the question of who pays for them. For the life of water pollution control in this country, while public sewage treatment has been publicly funded, private industry point sources have paid their own abatement costs. Nonpoint source abatement, however, although private and including some of the largest corporations in America, is paid for largely by the government. The state of Wisconsin is in the final stages of adopting rules imposing mandatory controls on nonpoint source runoff but require the state to provide 70% of the costs.²⁵¹ The state of New York is providing over $ 14 million to farmers for runoff abatement,²⁵² and the city of New York is kicking in another $ 10 million to protect its drinking water resources.²⁵³ The CWA, by creating a dichotomy between point and nonpoint sources, also created an attitude within the nonpoint industry of an entitlement to pollute akin to a property right. The attitude is that if you, the public, want to abate nonpoint pollution, then you, the public, will foot the bill.

C. Recent TMDL Litigation

TMDL litigation is undergoing a seismic shift from the offense to the defense, as federal and state programs come into rudimentary, facial compliance with § 303(d) and as point and nonpoint sources see abatement requirements on the horizon. The litigation through the 1990s was brought almost exclusively by the environmental community trying to get the program in gear and, more specifically, requiring schedules for impaired water lists and TMDLs. As the lists and initial TMDLs appeared, nonpoint industries and municipal sewage treatment systems have reacted and have taken to court both the core concepts of the TMDL program and its application to particular sources. This litigation promises to be the wave of the future unless EPA relaxes the program's requirements, which it is in the process of doing. Relaxation at the federal level, however, will not wish away these issues at the state level, particularly in those states that continue to pursue abatement under the impetus of TMDLs.

Environmental litigation has continued to push the TMDL agenda forward. At the start of 2002, EPA was under court order by decision or consent decree in more than 20 states to establish TMDLs on timetables ranging from 6 to 12 years unless the states stepped forward.²⁵⁴ Recent decisions and settlements in Hawaii, Iowa, and Tennessee kept this ball rolling despite minimal compliance by EPA and these states to list their waters and develop TMDLs.²⁵⁵ On the other hand, an increasing number of courts are beginning to show a weariness in overseeing the process and are accepting an any-progress-is-sufficient-progress attitude toward this same low level of performance. Viewed on the "constructive submission" theory, courts in California, Maryland, and Oklahoma have recently held that the submission of something, anything, sufficed, and that it sufficed under the APA's "arbitrary and capricious" standard as well.²⁵⁶ In the words of the California court, "California and the EPA have both been doing something about TMDLs, albeit not as rapidly as contemplated by the passage of the [CWA]."²⁵⁷ The logic of these cases in giving EPA and the states, if they are intent on implementing the program, some leeway in going forward is difficult to gainsay. On the other hand, against a history of noncompliance and a recent, explicit federal policy change from hands-on to hands-off, more rigorous court review may be necessary to keep the program viable.

Environmental lawsuits have also challenged other compliance issues, particularly the exercise of EPA's discretion in approving the content of TMDLs and the question of permitting new activities in impaired waters for which TMDLs have not yet been prepared. In Natural Resources Defense Council, Inc. v. Muszynski,²⁵⁸ the latest in a series of cases gigging TMDLs forward in the state of New York, the U.S. Court of Appeals for the Second Circuit reversed and remanded EPA's approval of a TMDL based on "annual loadings," reasoning that limits based on seasonal variations of flow and temperature would make more sense.²⁵⁹ The concept of "averaging" pollution loadings over long periods of time, high and low flows, and growing and nongrowing seasons is another way of gaming the TMDL program and, indeed, all ambient programs; the more averaging allowed, the more acceptable that periodic exceedances, even severe exceedances, become. Lawsuits in Maryland and Vermont have also raised the issue of permitting in the absence of TMDLs and appear to be headed toward allowing no new discharge, net, until TMDLs are prepared.²⁶⁰ The content of [32 ELR 10404] TMDLs is also on appeal on the important question of whether TMDLs require an implementation plan. While EPA has done an abrupt about-face on this issue and is arguing, now, in the negative, the Agency had earlier represented to a court in Georgia that such plans were both required and going to be included in Georgia TMDLs, giving rise to an estoppel theory on which this case may be decided without reaching the general issue of whether such plans are required by law.²⁶¹ A similar case is pending in New Mexico, absent the estoppel argument but based on EPA guidance that required, for an approved TMDL, "reasonable assurance" that its load reductions would be accomplished.²⁶²

One more technical environmental challenge is pending against the state of Florida's adopted rules for determining water impairment, rules that formed the basis for much of the NRC report described above.²⁶³ At bottom, Florida has provided a two-tiered approach to decisions on impairment and imposed a higher standard of proof for those waters subject to § 303(d).²⁶⁴ EPA has long required a minimum of three samples, 10% of which show impairment, for a decision to list and has accepted other data than sampling, such as fish kills and drinking water advisories, to serve as data sources as well. The Florida rule relegates all data other than monitoring to its initial candidate list.²⁶⁵ It further requires greater certainty for waters that have fewer samples in order to correct for the possibility that these fewer samples might reflect unusual conditions.²⁶⁶

The state defends its rule as more objective and scientific than that formerly taken by EPA, and even predicts that its approach, although heavily favored by the sugar and pulp and paper industries, may lead to more listings than those currently named.²⁶⁷ The reasons supporting this prediction are not obvious. If 100 samples are now required to match EPA's former threshold of 10% impairment, and less than 100 samples require a more than 10% showing, then unless the states do a lot more sampling waters are going to drop out of the program and into a limbo of the unproven. In terms of data requirements, taking the state of Maryland's list of 1,000-plus impaired waters as an easy number, and multiplying it by the 100 samples needed to prove it, yields 100,000 samples needed to maintain that state's list alone. Multiplying these samples by 50 states yields 5 million samples. Cutting these 5 million in half, to account for states with fewer waters, Oklahoma's 400 for example, yields a workload of 2.5 million samples. Those numbers are in the range of McDonald's hamburgers sold. Scientific confidence in an ambient-based system does not come cheap.

While the sum of the environmental litigation has been to continue to pressure the program forward, important litigation is brewing on the other side. The mega-action was filed in late 2000 against EPA's July 2000 TMDL rule by the American Farm Bureau Federation, the American Forest and Paper Association, the American Crop Protection Association, the National Pork Producers Council, the National Corn Growers Association, the National Chicken Council, the National Cattlemen's Beef Association, the National Cotton Council of America, the Fertilizer Institute, the Utility Water Act Group, and the TMDL Coalition, representing industrial dischargers and municipal publicly owned treatment works (POTWs).²⁶⁸ Is there anyone whose name we have not yet called? On the environmental side are the Friends of the Earth and the Water Keeper Alliance. While the petitions for review in these cases raise a bevy of issues, at the heart of them are the two that have always drawn the fire of the nonpoint industries, the coverage of nonpoint sources and the requirement for implementation plans. The proceedings in the case have since been suspended with the consent of all parties while EPA reworks its way through these same issues.²⁶⁹ As discussed below, it is likely that the Agency will, in the end, concede most if not all of the industry complaints.

Meanwhile, the defense bar is encouraging its clients to pool their resources to challenge individual TMDLs from start to finish, and predicting an "increase in challenges to lists of impaired water bodies and state methodology of establishing TMDLs."²⁷⁰ POTWs, feeling the pinch of tightened permit requirements imposed by states unwilling to impose abatement requirements on nonpoint sources, are leading the way and have raised procedural issues that will add new layers of complexity to the program.²⁷¹

A threshold issue is whether listings of impaired waters and TMDLs, at the state level, are decisions that require full notice-and-comment rulemaking.²⁷² Courts are arriving at opposite conclusions here, but the majority are answering in the affirmative, and the argument that these decisions are "statements of general applicability which impact, interpret, [32 ELR 10405] or prescribe law or policy" seems hard to gainsay.²⁷³ The effect of these decisions will be, in theory, to provide review by both the environmental community and by affected sources. As a practical matter, however, setting aside the question of resources to litigate, in a program with so few affirmative requirements at the federal or state level the challenges will largely come in response to a state decision to list or impose TMDL allocations, i.e., from the affected discharge community. If full administrative process is required for all listings, TMDLs, state decisions "translating" state narrative criteria to TMDL allocations, and each plan of whatever nature to set allocations on whatever basis, including markets and trades for the estimated 40% of the nation's waters that are impaired and the 36,000 TMDL decisions awaiting them, we have a recipe for tying implementation in knots. Ambient-based management meets the APA.

Process issues aside, affected dischargers have also given more than notice of their intent to challenge the data and scientific underpinnings of TMDL decisions. Two cases on behalf of a Maryland POTW challenge the validity of the state's nutrient-related chlorophyll-a criterion and the calculations of a TMDL, e.g., "calibration runs of the model used to develop the TMDL predict impossible levels of dissolved oxygen" and "are irreconcilable with the available real world data."²⁷⁴ Ambient-based management meets aggressive lawyering, an encounter it has often lost over the years.

Perhaps the safest conclusion from the current trend in TMDL litigation is that it is likely to grow and to contest every aspect of the program at the implementing state level. Ambient management is very much prove-it-or-lose-it and will lead, in the words of a defense lawyer, "to a diversion of a large amount of time and money into litigation of both regulated sources and the regulators."²⁷⁵ The message from the defense is clear: we'll sue.

III. Prelude

When Congress adopted the TMDL program in 1972, it had a smaller job in mind.²⁷⁶ Basing its approach to pollution control on national technology standards, it retained ambient-based management as a safety net, where technology-based permits would violate water quality standards, § 303, and where residual pockets of pollution remained, § 303(d): TMDLs. No one at the time believed that TMDLs would do more than come in late in the game and tidy up.²⁷⁷

Congress was half right. The national technology standards were more effective than it had reason to hope. Within the first years of application of BAT to industrial and municipal sources pollution loadings dropped precipitously,²⁷⁸ in some industrial categories such as the pulp and paper industry they dropped by more than 90%.²⁷⁹ Receiving water quality improved by 29%.²⁸⁰ With later rounds of permitting, some industries and industrial categories have moved to closed system recycling of used waters, meeting the CWA's zero discharge goal.²⁸¹ Municipal treatment systems, meanwhile, dropped in pollution loadings by 50% while their populations served more than doubled.²⁸² Reviled by economists and the more hard-nosed industries as [32 ELR 10406] "inefficient," "dictatorial," and "one-size-fits-all," the NPDES system has had but one defense: it worked.²⁸³

Congress was also half wrong. Focused on evidence of poisoned lakes and rivers so contaminated by industrial chemicals that they were catching fire,²⁸⁴ it failed to see the problem that nonpoint sources were to become. By the 1990s, these sources, primarily agriculture untouched by the NPDES program, had grown to elephantine proportions threatening waters the size of the Chesapeake Bay and the Gulf of Mexico and nearly one-half of all waters in between. Enter TMDLs, Congress' intended offense to handle the bits and pieces remaining after implementation of the NPDES program, suddenly facing a problem as large, varied, and intractable as industrial sources ever were. The question is whether the TMDL program will be up to the job.

A. Forward Into the Past

It is oft said that those who ignore history are doomed to repeat it. The inverse of the statement is only partially true. We may at times repeat history because we ignore the past. But we may also repeat history because other goals are more important to us, no matter what the lessons of the past. So it will be with TMDLs.

1. The Past as Prologue

Ambient-based pollution control strategies have a strong grip on the American psyche.²⁸⁵ They rely on science, an objective and steady guide. They conform to our desire for local government and local controls. And they are apparently rational, requiring abatement and cleanup only where contamination interferes with human uses. It is no surprise then that ambient strategies were at the heart of our first environmental legislation, our almost instinctual response to a pollution problem. They formed the basis for the Water Quality Act of 1965 and its predecessors.²⁸⁶ They formed the basis of the CAA of 1970.²⁸⁷ They were the guideposts for the Toxic Substances Control Act (TSCA), Resource Conservation and Recovery Act, Federal Insecticide, Fungicide, and Rodenticide Act, CERCLA, and the Safe Drinking Water Act.²⁸⁸ Unfortunately, all of these statutes tripped over the same thresholds.

One difficulty is that ambient systems require more of science than it can deliver. The issues turn out to be more complex than imagined. They require extrapolations of causes and effects—be they over toxicity, carcinogenicity, persistence, bioaccumulation, exposure pathways, synergy, dilution, or distribution—that are rarely dispositive and highly susceptible to challenge.²⁸⁹ The federal pesticide and toxic substances control programs bogged down under the number of subject chemicals and the burden of proving their effects.²⁹⁰ The CAA hosts the same arguments over air pollutants, standards, transport, and attainment.²⁹¹ The Superfund has become a legend of litigation, a "full employment act for lawyers" over the issue of how clean is clean?²⁹² Setting safe levels even for known toxins has confounded the toxic emission programs of the Air and Water Acts,²⁹³ and scientists are still arguing over dichlorodiphenyl-trichloroethane (DDT), dioxin (a pollutant once characterized by the EPA as "by far the most potent carcinogen evaluated [32 ELR 10407] to date by this agency"),²⁹⁴ and even the health effects of tobacco. Risk assessment may be science but it is anything but a steady guide.

No more steady a guide is risk management, which is frankly political and, in the case of air and water regulation, a decision conceived of as best lodged with state and local authorities who are said to be familiar with the problem and in tune with local needs.²⁹⁵ In most states these needs are aligned with economic and development interests whose local influence—be it chickens in Arkansas, sugar in Florida, the timber industry in Idaho, wheat in Kansas, oil and gas in Louisiana, cattle in Nevada, coal in Wyoming, and real estate nearly everywhere—is magnified by being the dominant game in town. Trying to achieve a national interest in clean air or water through state and local governments, however appealing in theory, is like trying to encourage spaghetti through a keyhole.²⁹⁶

A third difficulty with ambient-based management is its insatiable demand for data, manpower, expertise, and detailed, site-specific analysis.²⁹⁷ Every water segment has its own flow regime, natural and unnatural background levels of contaminants, contaminant-neutralizers, soils, subsoils, slope, streamside vegetation, rainfall, permeability, meanders, tides, pools and drops, flora, and fauna. One size does not fit all; each water body calls for separate analysis. The discovery of adverse impacts triggers even larger resource demands in identifying, and proving, the causes and their relative contributions to the problem. More data, more resources, more time.

Each of these factors plagued the ambient-based programs of the CWA and its predecessors.²⁹⁸ In the 1950s and 1960s, many states did not even bother to set water quality standards, and those that did so kept an open eye toward encouraging economic development.²⁹⁹ Enforcement was paralyzed by politics and lack of proof.³⁰⁰ Federal-state abatement conferences for the Potomac River and Puget Sound went staggering into their second decade without a solution in sight.³⁰¹ By the late 1960s, meanwhile, municipalities were discharging untreated waste from 75 million people, and industrial sources more than doubled that figure in organic wastes alone, without beginning to account for heavy metals and hydrocarbons.³⁰² By 1971, after two decades of federal assistance to state water quality-based programs, more than 300,000 industrial sources were discharging 22 billion gallons of wastewater a year, less than one-third with any form of treatment.³⁰³

The years following the CWA of 1972 saw a repeat performance in the § 208 program.³⁰⁴ Section 208 provided assistance to states and local bodies for areawide water quality planning. In the most logical way, over many years, and through cooperating agencies, detailed investigations, and countless stakeholder meetings, the program undertook to identify all sources of pollution and develop plans to abate them.³⁰⁵ It ate of millions of dollars, produced several hundred studies, and cleaned up very little water.³⁰⁶ By the 1980s, the § 208 program was quietly shelved. In its place, in 1987, came the § 319 program³⁰⁷ which attempted, through grants to participating states, to encourage nonpoint sources toward reducing their pollution loads.³⁰⁸ As with § 208, there were few standards and fewer sticks.³⁰⁹ More studies ensued; nonpoint source pollution increased.³¹⁰

We are now in the year 2002. The reports noted earlier in this Article tell us that states lack the basic tools for water quality management: monitoring, modeling, and site-specific analysis.³¹¹ State water quality standards and application factors remain highly variable, with abundant opportunity to game the system.³¹² State enforcement remains variable as well and influenced by interests of local importance.³¹³ As one indicator of political will, many states have adopted policies forbidding the imposition of environmental [32 ELR 10408] requirements more demanding than those required by federal law,³¹⁴ and more have enacted laws prohibiting the regulation of agriculture, and in some cases all nonpoint sources.³¹⁵ Nor do states have the money for increased water quality management responsibilities³¹⁶; as of January 2002 nearly two-thirds were facing deficit budgets,³¹⁷ and that number is apparently growing. For the first time in years, the federal government is in deficit spending as well.³¹⁸ And the costs of this ambient program, by any measure, will be high.

These problems noted, EPA and the states continue to pronounce their faith in water quality standards-based management, seconded by the Congress, the scientific-consulting community, and, of course, regulated and nonregulated industry.³¹⁹ With all of these handicaps, and since none of them are secrets, one might well ask why ambient-based management would hold such sway? The answer is that it feeds into the primary agendas of many different constituencies for reasons other than the attainment of clean water:

. To the states, it offers the opportunity to regain authority over public policy;

. To water quality scientists, it offers a lead role in that policy and a continuous source of public funding;

. To water quality engineers, it returns water management, like the management of oil and gas, to logical outputs and human ends;

. To economists (somehow overlooking transaction costs)³²⁰ it is efficiency, avoiding the sin of waters made too unpolluted;

. To federalists it is good government, as in the Articles of Confederation;

. To discharge industries, it puts water quality management back on the site-specific, prove-it-or-lose-it basis that worked so well for them before;

. To EPA, it is a way to share the burden of pollution control, a burden with few friends and many enemies;

. And to Congress it is a way of appearing to address an environmental problem while directing large amounts of money to agencies and institutions back home.

Ambient-based management may not work very well, but it is compatible with the pursuit of many different stars.

A current analogy may be useful. In the wake of the collapse of the Enron corporation in January 2002—produced, inter alia, by a lapse in federal regulation and a self-interested gaming of the system—it turned out that while most investment firms continued to promote Enron stock despite clear signals that the company was in trouble, at least one firm saw Enron failing and did otherwise. In order to proof-check that it had not misjudged Enron, the firm invited analysts from other firms to come in and tell it why its conclusions were wrong. "It's not that they didn't see what we saw," the firm's president is quoted as saying.³²¹ "They either chose to come to the wrong conclusion because it suited them for a variety of other reasons, or to put their faith in management."³²²

Everyone who looks knows the problems with ambient-based water quality management as well. We are, nonetheless, moved by other reasons.

2. The New TMDL Program

I envision TMDLs to be a kind of information-based strategy which, if done properly, can inform, empower, and energize citizens, local communities and States to improve water quality at the local, watershed level. The basic information derived from a sound TMDL could liberate the creative energies of those most likely to benefit from reduced pollutant loadings to their own waters.

—EPA Assistant Administrator for Water, November 2001³²³

[32 ELR 10409]

EPA is redesigning a TMDL program that very much resembles the Water Quality Act of 1965 and § 208 of the subsequent CWA—neither a marked success. Guided by the stated goals of "flexibility, cost-effectiveness, and efficiency,"³²⁴ the new program will provide states the happy combination of increased federal funding and reduced federal oversight.³²⁵ EPA's challenge, in a context in which the states and discharge industries hold all the important cards, is to come up with a program that is, on the one hand, acceptable to these same states and industries and, on the other hand, does not simply recreate a failed past.

The Agency has identified the key issues of its new rules and rulemaking to include: the listing of impaired waters; implementation plans and reasonable assurance of their implementation; time frames and deadlines; stakeholder involvement; and EPA's role in the process.³²⁶ Not so identified, but very clearly at the threshold of these issues, is the coverage of nonpoint sources and nonpoint source contaminated waters.

a. Nonpoint Sources

EPA has been committed to the view that CWA § 303(d) includes nonpoint source pollution and pollution sources since its first program regulations in 1985.³²⁷ As a practical matter, nonpoint sources are the dominant source of pollution in every state in the country, and the near exclusive sources in several western states.³²⁸ An interpretation of § 303(d) without nonpoint sources would be like an interpretation of Shakespeare without the plays, interesting poetry but not very important.

The nonpoint industries have taken the position that, since § 303(d) speaks in terms of impairment remaining after the application of technology standards, it must contemplate only waters with point sources that are subject to technology standards, or, as a fallback, that it contemplates only waters that include point source discharges.³²⁹ Such a reading of § 303(d) is, of course, by no means compelled by its language; it is as easy to read the section as referring to all water impairment after the application of technology standards, not simply impairment in particular, point source-contaminated water bodies.³³⁰ At best (or worst), the statute is ambiguous on the question, and the combination of EPA's inclusion of nonpoint source impairment in its regulations for nearly 20 years and the fact that this inclusion best furthers the CWA's clean water restoration goal should allow the Agency's interpretation to survive judicial review under Chevron³³¹ principles before even the most hostile court.

The inclusion of nonpoint source pollution remains a causa belli for the agriculture and timber industries, however, and statements from EPA officials indicate that the Agency is at least rethinking its position.³³² A compromise position that may appear at least as an option in the new regulations will be to include only those nonpoint sources to waters where point source pollution also exists, exempting thereby the so-called nonpoint-only waters.³³³ Given the large number of nonpoint-only waters in the American West and the historic intransigence of nonpoint sources to abatement, this interpretation would reduce the impact of the TMDL program by as much as one-third, and by as much as 90% in some states—to state and industry opponents, a victory. A result so at odds with the Act's goals, the Agency's regulatory history, and the Agency's current view of TMDLs—as mere numerical targets, nonbinding calculations of load reductions to be achieved through other programs—will certainly be challenged in court. One would think such a challenge a clear winner until one recalled that the facts in Chevron were of a very similar pirouette by EPA, reversing its previous interpretation of a pollution control statute, which a majority of the Supreme Court found to be within its discretion.³³⁴

b. Listings

The new rules will likely approve the five-part listing process described earlier, with only the fifth and last category identifying impaired waters that require TMDLs.³³⁵ The fourth category, just below, will include waters that are impaired but that: (1) have TMDLs; (2) are impaired by pollution but not by a pollutant; or (3) are "expected to meet" water quality standards.³³⁶ Although current Agency guidelines limit the expected-to-meet category to waters achieving standards within two years,³³⁷ the Agency may well follow [32 ELR 10410] its rule of flexibility and allow expectations over longer periods of time and perhaps more based on hope than on performance.³³⁸ One issue sure to arise in practice is the extent to which unproven "best management practices" by nonpoint sources will provide a reasonable basis to "expect" attainment over a given period of time.³³⁹ Another issue is whether there will be a given period of time at all.³⁴⁰

Other approved listing categories will include "insufficient data" and "additional monitoring needed" waters, an approach that has already significantly reduced the lists of several states, and hence their workload in implementing the program.³⁴¹ How large a limbo of unproven impaired waters this approach creates remains to be seen, but EPA seems unlikely to impose deadlines for determinations under these lesser categories, or "hammer" provisions requiring full listing after fixed periods of time. According to an EPA spokesman, such time frames would "limit flexibility"; instead, per the Agency, compliance times should vary with the nature of the problem, local water conditions, existing abatement measures, etc.,³⁴² an approach that will invite case-by-case arguments in practice.

EPA also appears ready to approve, if not require, Florida's statistical threshold, described earlier, for determining impaired waters.³⁴³ The data demands of this approach will almost certainly increase the list of unproven waters for which sufficient samples are lacking to establish the requisite certainty of impairment. Further, EPA has approved a Virginia program allowing the state, following listings, to delimit these same waters if voluntary efforts by sources in the watershed bring the water into compliance for two consecutive years.³⁴⁴ Virginia officials see the program as an incentive for voluntary abatement measures, which it surely is.³⁴⁵ Environmentalists see the program as an opportunity to delay working on TMDLs for two years or more, pending the development and then the assessment of the voluntary abatement measures³⁴⁶ — which it also surely is.

In sum, the new listing process, while likely to focus initial cleanups on the most provable contaminated waters, is likely to lead as well to significant "gaming" toward the lesser and "expected to meet" categories, to postponement of the development of TMDLs, and to a large body of unproven waters which remain outside the system.

c. Source Loadings

Section 303(d) requires states to derive the TMDLs of pollutants necessary to achieve water quality standards.³⁴⁷ In its 1985 regulations, EPA defined "loading capacity" as the "greatest amount of loading that a water can receive" without violating these standards,³⁴⁸ and TMDLs as the sum of individual wasteload allocations from nonpoint sources and load allocations, plus natural background levels, an allocation for future growth, and a margin of safety.³⁴⁹ In its subsequent guidance and in its approvals of TMDLs since this time, EPA has, in theory at least, required not only the identification of an overall load but also the ascription of this load to contributing sources.³⁵⁰ The question is whether the Agency will continue this approach, or, rather, will accept TMDLs that do no more than put all loadings in a single bucket, end of calculation.³⁵¹

The question is both legal and practical. As a legal matter, the 1985 regulations require more than the bare language of the statute. If challenged, the Agency would have to rely on its historic interpretation and practice and its general rulemaking authority, which should suffice³⁵²; with the wide differences of opinion over the scope of agency discretion now found on the federal bench, however, any result is possible. As a practical matter, the identification of the loads of contributing sources is a vital tool to the next steps in the process, abatement planning and its implementation. The who-caused-it question is a starting point for all remedial actions, including the well-known "whose mess is this in the kitchen?" To be sure, the relation between contribution and the ultimate cleanup responsibility may not be one-for-one in federal programs under CERCLA³⁵³ or even the CAA,³⁵⁴ but some relationship between causation and liability suits our sense of justice. Identifying this relationship also provides a good basis for subsequent trading and for the application of other factors such as economic efficiency, i.e., tweaking cleanup responsibilities to favor the least expensive.³⁵⁵ [32 ELR 10411] It, finally, provides a persuasive psychological handle over individual sources, particularly nonpoint sources over whom there is little other leverage, both in internal cleanup negotiations and in external publicity. If EPA wishes to rely on public pressure, the process needs to begin by fingering the blame.

d. Implementation Plans

Assuming that EPA holds the line on nonpoint source coverage and the identification of individual source loadings in approved TMDLs, the next question is the explicit inclusion of implementation plans. Indeed, this is the mega-question of the program. The absence of effective and enforceable implementation mechanisms—and the consequent absence of implementation—was a key to the underperformance of all previous water quality programs. Even the most antagonistic states and industries to the TMDL program concede the importance of implementation to water quality improvement. They simply oppose being required and being reviewed. They contest the inclusion of implementation plans as part of TMDLs under § 303(d) because the adequacy of these plans and their fulfillment would then be reviewable by EPA and, worse, at the behest of citizen groups, by courts of law.³⁵⁶ For these reasons, implementation plans are both the most important and the most controversial part of the entire program.

The outcome on this issue is not in doubt. EPA will throw § 303(d) implementation plans overboard, first thing out of port.³⁵⁷ As described earlier, the Clinton Administration's TMDL rules, jettisoning nearly all else, hung on to § 303(d) plans and required objective assurance of their fulfillment.³⁵⁸ With these plans now eliminated, the TMDL process is reduced to a calculation of hypothetical load reductions that would be necessary to attain water quality standards, useful numbers for "an information-based strategy" in the words of the current EPA.³⁵⁹

A reviewing court's treatment of this administrative pirouette is more problematic. On its face, the statute refers implementation planning to § 303(e),³⁶⁰ and the Agency's explicit inclusion of this planning in the § 303(d) context is relatively recent.³⁶¹ The Clinton Administration's decision to require § 303(d) plans was supported both by the logic of the requirement in determining, as required by the statute, what load reductions were necessary, i.e., an ephemeral reduction from source "A" would make a greater reduction from "B" necessary, and by the Agency's generic authority to issue regulations necessary to carry out its functions.³⁶² As noted earlier, at least one federal court has ruled in favor of their inclusion,³⁶³ but others lean in the opposite direction.³⁶⁴ On balance, an administrative decision to jettison § 303(d) implementation plans would be difficult to reverse.

In lieu of § 303(d) implementation plans, EPA is now moving to vest implementation in § 303(e), a never-never land of the CWA.³⁶⁵ Section 303(e) requires a continuing planning process (CPP) approved by EPA that "will result in plans" that "include," among a variety of elements, TMDLs.³⁶⁶ So far, then, we have a required "process" subject to federal review.³⁶⁷ EPA regulations—understandably, given the hodge-podge nature of the statute—combine the planning processes of § 303(e) (CPP plans), § 208 (areawide waste treatment management plans), and § 319 (nonpoint source plans) into water quality management plans (WQM plans), which are intended to "identify priority point and nonpoint water quality problems" and "control measures," including time frames and financing, necessary to "carry out the plan."³⁶⁸ Among the WQM elements are said to be, without further elaboration, TMDLs.³⁶⁹ Whatever its review authority in the planning process, EPA has no explicit review authority in its regulations over individual WQM plans, nor is the Agency authorized by the statute to promulgate plans should the states fail to do the job. In practice, state WQM plans are something of a misnomer and tend to consist of a series of publications on such elements as state water quality standards and the identification of impaired waters. The state of Louisiana's TMDL element, for example, simply lists impaired waters and summarizes state authority, e.g., permitting, to address the problem.³⁷⁰ Small wonder, then, that TMDL implementation enters this land of no definition and gets lost.

The effect of reverting TMDL plans to § 303(e) depends on what EPA now makes of the process. As presently structured and implemented, the section provides few standards for plans in general, none for TMDL implementation, and no process for either challenge or Agency or public review of individual plans. To those guided by the concept of states-rights—or those whose clients see advantages in a [32 ELR 10412] flexible program—this is a correct result.³⁷¹ It is also a facially legal result, given the shortage of statutory guidance. It is not, however, a necessary result. EPA has full authority to redefine implementation under § 303(e), to include the implementation standards found in its July 2000 TMDL rule, and to include, as well, both as a condition of delegated programs and as a condition of CPP approval, a process for administrative decisionmaking at the state level based on these standards and subject to judicial review.³⁷² Whether the Agency has the heart and the political capital to do this, however, is an open question.

As we have earlier seen, EPA has moved to incorporate TMDL planning standards into its § 319 program.³⁷³ As also noted, this move has drawn vigorous protest from the agriculture industry³⁷⁴ and its survival, and its rigor if it survives, is an open question. Certainly, if the inclusion of TMDL plans and assurances are incorporated into § 319 they will improve the effectiveness of that program. Further, it could be argued that since EPA's primary leverage over nonpoint sources is essentially a power of the purse, vesting the plans under § 319 preserves the same purse power through § 319 or ancillary federal funding.

The argument overlooks two realities. The first is that the threat of withdrawing federal funds weakens leverage over nonperforming states. Federal funding for delegated programs under the CWA or under the similarlyconstructed Coastal Zone Management Act (CZMA) has not been withdrawn even in the face of dramatic state noncompliance, indeed, provocation.³⁷⁵ Further, removing monies from states not interested in using them to abate pollution is a counterintuitive way to reach clean water.

The second reality allies to the first: the primary moving force in the § 303(d) program for the past decade has been citizen enforcement of the section's requirements.³⁷⁶ Complain as the Agency does about these suits as hindrances to rational administration of the statute—which they may be—they are also the primary reason the statute is being implemented at all. Were these suits removed, the TMDL program would fade as quietly into the night as its unsuccessful predecessors. Which may be no small part of the agenda. The worst fear of the nonpoint industry throughout the FACA and since has been that the program might lead to plans and abatement measures reviewable, first by the Agency, and then by a court of law.³⁷⁷ It is also EPA's worst fear. Because listings and TMDL development were explicit requirements of § 303(d), reviewable in a court of law, EPA, in effect, had to share enforcement authority over listings and TMDLs with environmentalists. By shifting implementation planning to §§ 303(e) and 319, without more, EPA regains control. In the shift, however, it loses the important, indeed essential, pressure that citizen enforcement provides.

e. Time Frames and Deadlines

EPA believes in long time frames and few deadlines. It will combine listings under §§ 305(b) and 303(d) and extend the period of reporting to several more years, probably five.³⁷⁸ The Agency will probably review the state § 303(e) planning process on the same schedule, although exactly what the Agency will review it for, given the minimal requirements of the statute, is not clear. Deadlines for TMDL development may well be accelerated, given the fact that TMDLs will have no planning involved, although this acceleration is not certain.³⁷⁹ On the other hand, courts for their part appear increasingly willing to accept any progress toward TMDL development as sufficient to comply with the statue.³⁸⁰ The outlook is for ample flexibility in meeting whatever minimum requirements the new program entails.

f. Stakeholder Involvement

There are essentially three types of stakeholders in TMDL development: the states, thepollution sources, and citizens. The new rules will certainly provide for close collaboration among states and the point and nonpoint source communities over allocation issues and for the development of trading systems to achieve more cost-effective strategies. The more open question is how they will involve local citizens and environmentalists. On this issue, there appears to be a disconnect between the talk and the walk.

In testimony before Congress, EPA professes a desire to "empower and energize" citizens and local communities in the TMDL process, liberating "the creative energies of those most likely to benefit from reduced pollution loadings to their own waters."³⁸¹ By which beneficiaries the Agency must mean fishers, boaters, swimmers, paddlers, collectors, biologists, school kids, dreamers, and all the great number of Americans who care about clean water more than they care about clean anything else in the world. There are of course millions of them, many of them organized and almost all of them willing to get involved. They are for the most part volunteers, with a relatively few, low-paid, overextended professionals. The question is how they will be liberated and empowered to participate: as kibitzers, or as players.

EPA appears to imagine a world where TMDL information, in the hands of the public, will influence significant pollution reductions from nonpoint source industries, somewhat akin to the publicity effects of the toxic release inventory (TRI) list.³⁸² While the TRI is undoubtedly an effective tool, the Agency may be overlooking a significant difference [32 ELR 10413] in public perception between the release of toxins and the release of sediments. As EPA's own polling shows, the two are worlds apart.³⁸³ EPA also overlooks the obvious fact that the practice of annual TRI disclosures has in no way obviated the continuing need to limit, monitor, and enforce compliance by the chemical industry.³⁸⁴ The TRI is a complementary program; it is in no way the game-in-chief.

The basics to effective citizen participation in environmental programs are access to decisionmaking at all stages of the process, decisions based on objective standards, and the opportunity to have decisions reviewed when they depart from these standards or are based on demonstrably false information. The National Environmental Policy Act (NEPA)³⁸⁵ works not because its standards are enforced by a supervising federal agency but rather by citizens and reviewing courts. The CWA NPDES program works the same way, a way explicitly promoted in the citizen suit provisions of the Act,³⁸⁶ to the point that it is impossible to imagine the accomplishments of the Act without the constant pressure of citizen enforcement.³⁸⁷ The genius of the Act was to invest enforcement at all three levels, federal, state, and citizen, in order to maximize the chances that at least one would get the job done.

EPA's current rules leave very little to which citizen involvement can attach. Access to TMDL calculations might be interesting information and the stuff of a good press release, but it is no substitute for an equal seat at the table. The Agency may aspire to liberated energy and community pressure rising up to secure clean water, but by the removal of implementation plans from § 303(d) and the failure to recreate a similar process elsewhere, the Agency has, and one would have to say quite consciously, put that aspiration on short rations.

g. Summing Up

EPA is moving toward a TMDL program that trades standards, oversight, and citizen enforcement for volunteerism, flexibility, and local decisionmaking. No one can gainsay the desirability of state leadership and voluntary participation in a program of this size and diversity. Nor can anyone gainsay the desirability of flexibility in designing solutions. The question is reasonable assurance that the solutions will work and the enforcement of their terms. At bottom, law is about rules, and any system that swaps rules for hope in dealing with actors who have historically refused to move forward is asking for failure.

One can imagine a scenario in which the TMDL program, made user-friendly and relieved of deadlines and enforcement, is accepted by the states with enthusiasm. Congress in turn funds it to the hilt, the post-September 11 economy and deficit budget notwithstanding. States, for their part, despite the fact that one-half are in deficit spending, more are heading in that direction, and nearly all have passed laws expressly prohibiting their regulation of agricultural and nonpoint sources, contribute a match in both funding and political will. Local governments hold the line on uses requiring clean water and maintain high water quality standards; a vast network of chemical monitors and biological surveys provides all necessary data; engineering models and scientific estimates provide persuasive, definitive answers to causes and impacts, fate and persistence, and the effectiveness of particular control measures; markets and trades take the sting out of abatement controls that, in turn, are scrupulously observed and self-enforced by dischargers, at which point the nonpoint industries, U.S. Sugar, Boise Cascade, Archer Daniels Midland, join in; those who do not are embarrassed into joining by adverse, TRI-like publicity; and we leave the lawyers and the environmental community for that matter looking for other things to do. We can imagine all this. EPA's new rules seem to. It would be a historic first. But anything is possible.

A more likely scenario is that the TMDL program bottomed on carrots, but far fewer than those needed, and will sputter forward unevenly even in the most well-intentioned and best-resourced states, which, in turn, are not legion. A program that simply provides a TMDL cap, or even load calculations, and leaves the rest to the states provides little backup for the hard decisions—hard in science, hard in proof at law, and hardest in politics—that will have to be made. Federal requirements may be a state governor's worst enemy, but they are a state environmental agency line player's best friend.³⁸⁸ A program that depends, further, on the threat of EPA stepping in to do TMDLs should the states underperform, as many will surely do, depends largely on an illusion; through "conditional approvals" and other evasions already seen in EPA's responses to state failures in the CAA program,³⁸⁹ this is one gorilla that simply does not come out of the closet.³⁹⁰ The missing link of the new program is something in between: objective performance standards and the enfranchisement of citizens to secure compliance, the same ingredients that have brought the NPDES program home.

[32 ELR 10414]

B. Forward Into the Future

The TMDL program needs help. As it is currently headed it will need a great deal of help, and this help will have to come from other, complementary approaches to pollution control.

There are at least three strategies for abating pollution.³⁹¹ One is to approach it from the backend, by its effects on the environment, ambient-based controls. Another is to approach it from the source, through technology and practice standards. A third is to approach it from the pollutant, either circumscribing or banning its use. With intractable problems—automobile emissions, for example—all three strategies are necessary to address the problem. We have ambient standards for the primary contaminants from automobile emissions³⁹² and highly complex requirements for abatement planning in nonattainment areas.³⁹³ We have technology requirements as well,³⁹⁴ technology-based emissions limitations,³⁹⁵ fuel standards,³⁹⁶ and low emission vehicles.³⁹⁷ We have taken the third step, over time, of eliminating lead in gasoline altogether.³⁹⁸

The other lesson from the automobile, of course, is that where all three elements are not pursued vigorously the problems are forever managed and never solved. Each of these strategies is potentially available to the TMDL program. All three will be necessary for it to succeed.

1. Enhanced Ambient Management

The TMDL program has often been described as an "air SIP," referencing the state implementation plan process under the CAA.³⁹⁹ The analogy works in theory, but less well in practice and law. Unlike the TMDL program, added as a last-minute compromise to the CWA in 1972, the SIP program was intended to be the lead strategy for achieving national clean air. As such, Congress afforded it important federal, structural features not found in the CWA: uniform national ambient air quality standards (NAAQS)⁴⁰⁰; implementation plans with prescriptive standards for federal approval⁴⁰¹; deadlines for their submission and review⁴⁰²; sanctions for underperformance and noncompliance⁴⁰³; and citizen suits with the potential of hounding the process forward.⁴⁰⁴ If one were to imagine a program combining state management with close and rigorous federal supervision, this would be the one.

This said, one would not wish the CAA SIP program on one's worst enemy. The complexity of the system is mind-boggling and, in the view of many commentators, self-defeating.⁴⁰⁵ NAAQS and their supporting science have been in controversy since day one, with few conclusive results.⁴⁰⁶ Models of short- and long-range transport vary widely, at times in close coincidence with the interests that fund them.⁴⁰⁷ Basic data and monitoring are spotty, selective, and widely unavailable.⁴⁰⁸ Large airsheds remain in a limbo of insufficient information.⁴⁰⁹ Delays in plan preparation, amendments, and approvals are multi-year.⁴¹⁰ Load allocations are unequally distributed, with demanding requirements on the more easily regulated point sources and almost hortatory planning provisions (e.g., "high occupancy lanes" for commuters, which in Louisiana means a second person in the car) for the diffuse sources that are often at the root of the problem.⁴¹¹ The "adaptive management" required for changes in air quality and in source loadings is so [32 ELR 10415] constant that if a "plan" exists few even in the Agency know what and where it is.⁴¹² Point sources are able to "game" the system at every point—monitoring,⁴¹³ false upgrades,⁴¹⁴ claimed reductions,⁴¹⁵ inspection and maintenance programs⁴¹⁶ — and trading systems add difficulties that confound both EPA and state administrators.⁴¹⁷ None of which should be news to the reader; the parallels with the CWA TMDL program are striking.

Prof. Robert Adler has written a thoughtful article on Lessons From the Clean Air Act for TMDLs in which, while recognizing essential differences between air and water quality management, he makes several recommendations for a strengthened TMDL program.⁴¹⁸ These include: consistency in water quality standards, monitoring, and listing; watershed, as airshed, planning areas; consolidated planing; implementation criteria and substantive implementation plans; attainment deadlines; and administrative review and enforcement authority.⁴¹⁹ In the end, however, he does not make them with great confidence in their effectiveness. While conceding that SIPs have succeeded "to some degree" in reducing air pollution, Professor Adler concludes that "no one would propose to repeat the problems of delay, uncertainty, complexity and political divisiveness that have plagued the SIP process over the years."⁴²⁰ He is not alone in this conclusion; nearly all commentators on the program agree.⁴²¹ So did Congress, which, by the late 1970s, began adding technology requirements to the CAA⁴²² and adopted them full bore in the Amendments of 1990.⁴²³ Ambient standards remained in the mix, but they were no longer the main game.

Several years ago this author presented an examination question describing the plight of Mexico City, overrun at the time with smog and motor vehicles and without an air quality program. The proposal was, starting from scratch, to recommend an air pollution abatement strategy. Many students undertook to rewrite the SIP provisions of the CAA. They missed the obvious. The problem was not the absence of air standards and monitors, although one could doubtless spend considerable time and money on them. The problem was factories and cars.

2. Technology and Practice Standards

To paraphrase a campaign slogan: "It's about agriculture, stupid!" The nonpoint source pollution that has swamped the nation's waters has many and diverse sources, but the lion's share are agricultural: crops and animals.⁴²⁴ More than 50% of water impairment nationally comes from agricultural runoff.⁴²⁵ In some western states dominated by cattle, the number reaches 90%.⁴²⁶ More than 80% of the eutrophication of the Gulf of Mexico dead zone is attributable to farm loadings over 500 river miles away.⁴²⁷ There is a time in the life of a household when one has looked at a messy room long enough and, without further dialogue about impacts and the rights of occupants, says the only reasonable thing: "Clean it up!" In the life of the CWA, that time has arrived for agriculture.

Agriculture is one of the largest industries in the United States, with a total production value of near $ 200 billion.⁴²⁸ Its principle field crops are corn, soybeans, hay, and wheat⁴²⁹; it also produces on an annual basis 98 million cattle, 366 million egg-laying chickens, 6.75 billion meat chickens, and 61 million hogs.⁴³⁰ It spends $ 18 billion on agricultural chemicals a year,⁴³¹ more than twice what it spends on gasoline⁴³² and electricity.⁴³³ Its payroll reaches $ 16.9 billion in farm and contract employees.⁴³⁴ Its larger operations are industrial, look industrial, and are managed as any national or multinational corporation.

The industry is both highly diverse and highly stratified, concentrating more by the year into large factory operations in which former individual farmers become sublet contractors or employees. The total number of American farmers has dropped from 7 million to 1.9 million in recent years.⁴³⁵ [32 ELR 10416] Roughly one-half of this population, small farms, provide only 1.5% of total farm production value while, at the other end, 3.6% account for more than 50% of the total value.⁴³⁶ The industry and its many subcategories are also highly subsidized by the federal government, at the rate of $ 20 billion per year, with the top 10% of farm operations receiving two-thirds of the benefits.⁴³⁷ The industry is supported by federal and state agricultural research institutions and by entire universities.⁴³⁸ It is also well defended through the American Farm Bureau Federation, a $ 200 million a year organization that produces another $ 18.5 billion through its insurance companies and provides aggressive campaign, political action, lobbying, and litigation services.⁴³⁹

With even this rudimentary understanding of the industry, the exemption of American agriculture from the CWA is either gross negligence or gross politics.⁴⁴⁰ Whichever, few of the reasons that have been given for exempting farm sources from the NPDES program obtain today. Far from being trivial, they are mammoth. Far from being site-specific and local, their impacts are multi-state, regional, and even international. Far from being without available control technologies, the control strategies for nonpoint source pollution, e.g., shelterbelts, cover crops, and cattle fences, are orders of magnitude less complex and less costly than other industry controls. To be sure, farms are diverse, but they are no more diverse by crop and production method than the many categories and subcategories of pulp and paper manufacturers, petrochemicals, metals, rubber, and other industries subject to the NPDES program.⁴⁴¹ Nor is it a persuasive distinction that farm runoff does not emerge from a pipe. Construction, municipal, feedlot, and acid mine drainage do not emerge from pipes either. They all arise from identifiable technologies and practices. They all pollute. They should all share the burden of cleanup and this burden in America begins with adopting BAT. There is something wrong with a picture that regulates the discharges of small Pacific Coast canning factories,⁴⁴² while Boise Cascade, U.S. Sugar, and Archer Daniels Midland walk free.

Prof. J.B. Ruhl has published a well-documented study of farm pollution, Farms, Their Environmental Harms and Environmental Law, suggesting a multifaceted strategy for imposing meaningful pollution controls.⁴⁴³ His proposal relies on a mix of emissions trading, disclosure, taxes, and economic incentives familiar to any reader of the literature on reinventing environmental law.⁴⁴⁴ His first and most persuasive strategy, however, and certainly the most concrete, is the application of technology standards to factory farm and large crop operations, what he calls the "agro-industry low-hanging fruit."⁴⁴⁵ The proposal neatly severs the relatively few major farm players from the general population of farmers, subjecting only the former to federal regulation; such distinctions are already made in law between large and small municipalities⁴⁴⁶ and large and small CAFOs.⁴⁴⁷ There can be no doubt that sugar farming, rice farming, and other high-impact activities would benefit from the same multimedia, air, water, and waste analysis that EPA has provided for the pulp and paper industry.⁴⁴⁸ Indeed, such analysis and abatement measures are already on hand in best practices manuals of the USDA, EPA, and many states.⁴⁴⁹ They are undeniably effective.⁴⁵⁰ The only question is whether they will, at last, be mandatory and, therefore, be implemented widely and evenhandedly across the country.

Of course, the proposal is heresy to the agriculture industry, which has tied up EPA's proposed BAT for CAFO's for several years⁴⁵¹ and is presently offering an alternative to technology standards based on "environmental management systems."⁴⁵² On the other hand, Congress and EPA have been nibbling around the edges of diffuse sources since the 1980s, bringing municipal runoff, construction runoff, and CAFOs under permit.⁴⁵³ Environmental lawsuits are [32 ELR 10417] pushing the envelope farther with cases seeking permit controls for dairy farms,⁴⁵⁴ stormwater runoff,⁴⁵⁵ spraying of aquatic herbicides,⁴⁵⁶ ballast waters,⁴⁵⁷ and logging.⁴⁵⁸ In the heat of battle, of course, EPA pulled back from further mention of the application of practice standards to logging roads,⁴⁵⁹ but no reason in the world other than politics can distinguish between runoff from a construction site bulldozer, a forest road bulldozer, and a plow.

In short, technology controls were the genius of the CWA and have been the primary engine of its success. To continue to exempt agriculture from these controls requires an act of willful blindness. Whatever the reasons offered exempting agriculture in the past, those reasons no longer obtain. If we are truly serious about cleaning up the nation's waters, the agriculture industry should apply what everyone else is coming to apply: BAT.

3. Limiting the Pollutant

Faced with an overload of mule pollution, one has several options. The first is an intensive examination of the Mule-Manure-Nitrogen-Cycle, followed by detailed studies of persistence, fate, and distribution, their relation to downstream algae blooms, and the post-baseline composition and distribution of aquatic communities. The second is to go to Best Available Mule Practices: we step 'em back from the creek. A third is to reduce the number of mules.

As a national environmental strategy, we have made such reductions before. We have eliminated chlorofluorocarbons (CFCs) and DDT. We have severely limited uses of lead, mercury, and polychlorinated biphenyls. We have learned two lessons from these experiences. The first is that when we limit the pollutants, the environment improves, dramatically.⁴⁶⁰ Lead levels in urban air and in the bloodstreams of urban children dropped sharply⁴⁶¹; the Brown Pelican and the American Bald Eagle, virtually exterminated by DDT and related pesticides, are coming off of the endangered species list.⁴⁶² The second lesson is that, no matter how loudly their manufacturers proclaimed that DDT, CFCs, and lead were indispensable, more benign alternatives were found that filled the niche.⁴⁶³ Granted, these are drastic measures and ones limited to major, harmful chemical actors. They are also highly effective measures, and in terms of transaction costs, highly efficient as well.

The pollutant at issue in TMDLs is fertilizer.⁴⁶⁴ To be sure, fertilizers are only part of the stew but they are the dominant man-made ingredient.⁴⁶⁵ The contamination of the Louisiana dead zone is directly correlated to fertilizer use; the hypoxia has grown in direct proportion to fertilizer application upstream and rises seasonally with increased runoff.⁴⁶⁶ This correlation is no secret to those working on cleanups for the Chesapeake Bay, the Apalachicola Bay, the Everglades, and many other anoxic bodies of water across the country.⁴⁶⁷ If the primary proposition is: "It's about agriculture, stupid," then the sub-prime is: "It's about fertilizer."

More than 54 million tons a year—110 billion pounds—of commercial fertilizers and related liming materials are consumed in the United States.⁴⁶⁸ Fertilizers with the primary nutrients—nitrogen, phosphorous, and potassium—account for 91% of the total.⁴⁶⁹ The states of highest fertilizer consumption are, not surprisingly, agricultural states in the cornbelt and California.⁴⁷⁰ While the potato crop is the most fertilizer-intensive for all primary nutrient fertilizers at 195 pounds per acre of nitrogen, 173 pounds per acre of phosphorous, and 139 pounds per acre of potassium . . . i.e., one-half ton of fertilizers per acre of crop⁴⁷¹ . . . [32 ELR 10418] corn, covering 70 million acres, leads the nation in total loadings.⁴⁷² Nearly 6 million tons of phosphates and nitrogen run back into the nation's waters.⁴⁷³ Considering that almost one-half of the land mass of the United States is in field-crop agriculture,⁴⁷⁴ what we have here is a very profitable business⁴⁷⁵ and massive pollution.⁴⁷⁶

We also have, of course, a necessary ingredient to American farming and a principal factor in its success. On the other hand, fertilizers are over-marketed and overused in virtually every crop-growing sector,⁴⁷⁷ not out of ignorance or low costs (indeed, fertilizers may be the largest cost in a growing season)⁴⁷⁸ but because individual farmers find themselves in a prisoner's dilemma: if their neighbors do it, they cannot afford not to.⁴⁷⁹ Meanwhile, of course, the environmental costs are "externalized," i.e., floated downstream to, say, Louisiana and the Chesapeake Bay. The question becomes, then, how to limit the application of fertilizers in a way that will not penalize those who are willing to restrict both their use and thus, to a degree, their production. At least three tools available for doing so.

The first is federal regulation to limit fertilizer application and use. In late 1985, EPA exercised its authority under TSCA to begin an investigation of fertilizers and the contamination of groundwater.⁴⁸⁰ TSCA § 8 was invoked to identify groundwater contamination problems,⁴⁸¹ § 4 to test groundwater quality and health effect,⁴⁸² and § 6 to determine appropriate controls.⁴⁸³ The decisionmaking framework foresaw first-round controls—described as emphasizing "integrated problem solving," i.e., the one of best control, whether TSCA or another method, in 1988-1989, with further study leading to additional means as needed in the future.⁴⁸⁴ Whether from the cumbersome nature of TSCA or the political force of the agriculture and fertilizer manufacturing industries, the initiative quickly died and has not reappeared.⁴⁸⁵ The Agency is now looking in a more limited way at the impact of heavy metals from fertilizers, although it has proposed no strategy to deal with them.⁴⁸⁶ This history notwithstanding, if ever there were a product with the type of multimedia impacts that TSCA was enacted to address, fertilizers certainly come to mind. The TSCA approach may not have died because it was a bad idea but, rather, precisely because it might have worked.

A second approach, developed by the USDA, several states, and the European Union (EU) and its Member countries are fertilizer/crop/soil ratios that determine optimal application regimes.⁴⁸⁷ Yet a third approach, adopted by a few states and by several agriculture-intensive countries of the EU, is to impose fertilizer taxes, either across-the-board or in relation to optimal application standards.⁴⁸⁸ The general question of pollution taxes is beyond the reach of this Article, but commentators note the difficulties they pose in calibrating the relationship of tax increases to a desired level of abatement, in fairness to smaller competitors, and more recently in environmental justice for communities impacted by larger competitors, e.g., the Louisiana chemical corridor.⁴⁸⁹ The proposal here, however, is not to use taxes as a substitute for regulation—an approach considered and rejected by Congress in enacting the CWA of 1972—but, rather, as a supplement to incentivize abatement from its dominant and most intransigent source.

[32 ELR 10419]

In 1992, the EU amended its organizing treaty to include a statement of environmental principles that would guide EU and Member State decisionmaking.⁴⁹⁰ The first of these principles were: pollution should be rectified at the source, and the polluter pays.⁴⁹¹ Why not?

IV. Concluding Thoughts

The CWA was enacted with the mission of clean water. It was that simple, and it is still on the books. The approach was to clean up the country's major pollution sources through technology standards and to treat those remaining through federal-state, ambient-based programs: TMDLs. The technology standards—despite their subsequent compromises, litigation delays, loopholes, and enforcement lapses—worked remarkably well. What no one anticipated was that nonpoint sources would come along to eat up the gains. TMDLs are now facing a problem of a scale they were not designed to solve, and by themselves will never solve.

The emerging TMDL program, with its emphasis on incentives and voluntary measures, is based more on faith than on fact. In a previous series of articles on the TMDL program, this author concluded that, despite its indirectness, uncertainties, and steep maintenance costs, the program was worth the candle because it gave Americans who care about clean water the chance to make it happen.⁴⁹² The unstated corollary was that, absent an enforcement role for citizens, the program was more waste than product. While it is impossible to gainsay the need to involve state agencies in water pollution control, it is equally impossible to predict that—given the challenges in science, funding, and political will that this program faces—they will do the job any better than they have done for the past 50 years. The TMDL program needs carrots but it also needs consequences, and it could use help from other pollution control strategies that have proven themselves effective. With these additional points of leverage, TMDLs have the chance to make a significant contribution to clean water. Without them, all bets are off.

1. TMDLs are referred to under § 303(d) of the CWA, 33 U.S.C. § 1313(d), ELR STAT. FWPCA § 303(d). The evolution of the program implementing § 303(d) is described in OLIVER A. HOUCK, THE CLEAN WATER ACT TMDL PROGRAM: LAW, POLICY, AND IMPLEMENTATION (Envtl. L. Inst. 1999) [hereinafter TMDL PROGRAM].

2. Section 303(d) provides in relevant part:

(d) Identification of areas with insufficient controls; maximum daily load; certain effluent limitations revision

(1)(A) Each State shall identify those waters within its boundaries for which the effluent limitations required by section 1311(b)(1)(A) and section 1311(b)(1)(B) of this title are not stringent enough to implement any water quality standard applicable to such waters. The State shall establish a priority ranking for such waters, taking into account the severity of the pollution and the uses to be made of such waters.

(B) Each State shall identify those waters or parts thereof within its boundaries for which controls on thermal discharges under section 1311 of this title are not stringent enough to assure protection and propagation of a balanced indigenous population of shellfish, fish. and wildlife.

(C) Each State shall establish for the waters identified in paragraph (1)(A) of this subsection, and in accordance with the priority ranking, the total maximum daily load, for those pollutants which the Administrator identifies under section 1314(a)(2) of this title as suitable for such calculation. Such load 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. . . .

(2) Each State shall submit to the Administrator from time to time, with the first such submission not later than one hundred and eighty days after the date of publication of the first identification of pollutants under section 1314(a)(2)(D) of this title, for his approval the waters identified and the loads established under paragraphs (1)(A), (1)(B), (1)(C), and (1)(D) of this subsection. The Administrator shall either approve or disapprove such identification and load not later than thirty days after the date of submission. If the Administrator approves such identification and load, such State shall incorporate them into its current plan under subsection (e) of this section. If the Administrator disapproves such identification and load, he shall not later than thirty days after the date of such disapproval identify such waters in such State and establish such loads for such waters as he determines necessary to implement the water quality standards applicable to such waters and upon such identification and establishment the State shall incorporate them into its current plan under subsection (e) of this section.

3. See infra section III.A.1. See also WILLIAM H. ROGERS JR., ENVIRONMENTAL LAW (2d ed. 1994) 55-57 (risk-based management):

The strategy of prospective regulation based on risk assessment has not been successful. Legislative deadlines are often missed. Regulatory endeavors often barely scratch the surface of the list of pollutant candidates. Known problem chemicals linger in commercial chemicals for years. The methodology demands more time, information, and bureaucratic effort than is available to devote to it. It requires a number of hard policy choices. It affords opportunities for refuge and obstructionism by parties not interested in cooperating.

161-64 (ambient air management), 342-52 (ambient water management), 741-45 (hazardous waste cleanup); Oliver Houck. Of Bats, Birds, and B-A-T: The Convergent Evolution of Environmental Law, 63 MISS. L.J. 403 (1994) (arguing that failure of ambient standards management has led to alternative standards in both pollution control and natural resources law).

4. For insight into the relative strength of EPA in the current Administration, see Whitman Takes Back Seat to White House, OMB Policy Priorities, INSIDE EPA, Jan. 4, 2002, at 15 ("The Bush Administration has centralized its environmental policymaking decisions in the White House more than virtually every previous administration, investing much more of the authority for broad environmental policy priorities with Office of Management and Budget (OMB) regulatory chief John Graham, observers say.").

5. See New OMB List Targets Eight Major EPA Rules for Possible Rollback, INSIDE EPA, Jan. 4, 2002, at 1.

6. U.S. EPA, Revisions to the Water Quality Planning and Management Regulation and Revision to the National Pollutant Discharge Elimination System Program in Support of Revision to the Water Quality Planning and Management Regulation, 65 Fed. Reg. 43586 (July 13, 2000).

7. EPA's first TMDL regulations were promulgated in 1978, U.S. EPA. Total Maximum Daily Loads Under the Clean Water Act, 43 Fed. Reg. 60664 (Dec. 28, 1978). They were amended in 1985 as part of the Agency's consolidated water quality management program, U.S. EPA. Water Quality Standards Regulation. 50 Fed. Reg. 1774 (Jan. 1, 1985), and again in 1992, 57 Fed. Reg. 33040 (July 24, 1992), following court decisions that were beginning to enforce TMDL requirements to impose deadlines for the submission of listed waters; no other deadlines nor substantive standards were provided for the TMDLs themselves. See generally TMDL PROGRAM supra note 1, at 49-53.

8. See TMDL PROGRAM, supra note 1, at 49-55.

9. Id. at 56-58, 77-82.

10. Id. at 57-58, 82-84.

11. Id. at 83-84.

12. Letter from John Barrett to Oliver A. Houck (Sept. 17, 1999) (on file with author). Mr. Barrett represented the American Farm Bureau Federation on the FACA and in subsequent lobbying against EPA's TMDL rules.

13. J. Zane Walley, The Water Wars, RANGE MAG., Fall 1999, at 30.

14. See TMDL PROGRAM, supra note 1, at 76 (court-ordered schedules for TMDL preparation ranged from 12 years to as few as 5).

15. See id. at 7-82 (describing EPA guidance on listing and TMDL content).

16. 5 U.S.C. § 553, available in ELR ADMIN. STAT. PROC. (notice-and-comment rulemaking requirements).

17. U.S. EPA, Proposed Regulations to the Water Quality Planning and Management Regulation, 64 Fed. Reg. 46012 (Aug. 23, 1999).

18. See Water Quality Standards: Offset Requirements, Other Provisions Eliminated From TMDL Rule, EPA Says, 31 Env't Rep. (BNA) 685 (Apr. 14, 2000).

19. U.S. EPA. Revisions to the Water Quality Planning and Management Regulations, 65 Fed. Reg. 43586, 43589 (July 13, 2000).

20. See State Programs: Officials Want More Guidance, Flexibility, to Implement Programs, Subcommittee Told, 33 Env't Rep. (BNA) n.p. (Mar. 2, 2001) (describing testimony from two state governors and two state agencies approving EPA TMDL rules); National Association of Convention Districts, TMDL Update (2000), at http://nacdnet.org/district/leader/ (describing agriculture interest lobbying).

21. T. Robert Braile, Clean Water Bill by Smith Meets EPA Opposition, BOSTON GLOBE, May 7, 2000 at 1.

22. Water Quality Standards: Agency Accused of Secret Meetings, Shoddy Science in TMDL Proposed Rule, 31 Env't Rep. (BNA) 1429 (July 7, 2000).

23. Id.

24. See Notice and Comment, ENVTL. F., Nov/Dec. 2000, at 17 ("A senior environmental official was stripped of power by a budget rider, reports the Washington Post. The measure, added to the Department of Agriculture spending bill, removes all authority for the under secretary responsible for the Forest Service and the Natural Resources Conservation Service, James Lyons, to run the two agencies.").

25. Id.

26. See Water Quality Standards: Agency's Draft Final Regulation to Revise TMDL Program Sent for White House Review, 31 Env't Rep. (BNA) 1325 (June 23, 2000).

27. Id.

28. Id.

29. Id.

30. See Claudia Copeland, EPA's Total Maximum Daily Load (TMDL) Program: Highlights of the Final Revised Rule, Cong. Research Serv., July 18, 2000, at 4, available at http://www.cnie.org/NLE/h2o-36.html (last visited Feb. 21, 2002); see also Senate Bill Would Insure TMDL Funding, Delay EPA Implementation of Final Rule, WATER ENV'T & TECH. NEWS WATCH (undated) (on file with author) (referencing EPA assurances to House Transportation and Infrastructure Committee Chairman Rep. Bud Shuster (R-Pa.)).

31. See comments of Earthjustice Legal Defense Fund representative on the proposed TMDL deadlines, Water Quality Standards, supra note 26 ("'Ten years plus five years is still 15 years,' she said. After that the 10 years states have to develop an implementation plan, and it pushes back the attainment of water quality standards even further, Mulhern said. Moreover, she added, the goal or attaining water quality standards is not enforceable. 'Fifty years from now, if the water is still not clean, will EPA still be on the hook to go and clean it up?' she asked.").

32. Personal conversation and e-mail correspondence with representatives of Earthjustice Legal Defense Fund, the Natural Resources Defense Council, Widener Law School Environmental Law Clinic, Southeastern Legal Foundation, Northwest Environmental Advocates, the National Wildlife Federation, the Sierra Club, and other national, state, and local citizen organizations throughout 1999-2000.

33. See Water Quality Standards: EPA Plans to Move Forward With Issuance of TMDL Rule Despite Obstacles, Fox Says, 31 Env't Rep. (BNA) 1123 (May 26, 2000).

34. Sen. Max Baucus (D-Mont.), for example, vowing to "reverse the [TMDL] decision at the earliest opportunity," characterized the rules as "roundly criticized by states, environmental groups, business organizations, and agricultural interests." See Supplemental Spending Bill, infra note 44.

35. See States Split on Bush Review of Clinton-EPA Air, Water Initiatives, INSIDE EPA, Aug. 31, 2001, at 1.

36. See Nonpoint Sources Need to Be Included in TMDL Program. Treatment Officials Say, 32 Env't Rep. (BNA) n.p. (Aug. 20, 2001) (questioning the General Counsel to her Association of Metropolitan Sewerage Agencies as stating: "What [the rule] did was to provide teeth to the program. . . . It made the program more meaningful.").

37. See Water Quality Standards: Agency TMDL Proposal Costly, Technical, Unenforceable, Senate Chairman Tells Group, 31 Env't Rep. (BNA) 389 (Mar. 3, 2000); Press Release. Office of Sen. Mike Crapo (R-Idaho). Senator Crapo's Floor Statement on S. 2417, The Water Pollution Program Enhancements Act of 2000 (undated), available at http://www.senate.gov/-Crapo/tmdl_floor_statement.htm.

38. See Clean Water Network, Summary of Water Pollution Program Enhancement Act of 2000 (2000), at http://www.cwn.org/docs/programs/tmdl/s2417factsheet.htm.

39. See H.R. 3609, 106th Cong. (2000) and S. 2041, 106th Cong. (2000) (exempting silvicultural activities from the NPDES program): H.R.J. Res. 105, 106th Cong. (2000) (nullifying TMDL rules).

40. See Water Quality Standards: Final TMDL Rule Signed by Browner, Styming Bid by Congress to Block Measure, 32 Env't Rep. (BNA) 1468 (July 14, 2000) (quoting EPA Administrator Carol Browner: "It is up to the states to make the decisions. It is not a federal permitting program.").

41. H.R. 4425, 106th Cong. (2000). See also Copeland, supra note 30.

42. H.R. 4425; see also Copeland, supra note 30; Water Quality Standards: Final TMDL Rule Signed by Browner, supra note 40 (quoting EPA Administrator Browner: "It was put on the supplemental [appropriation to the military construction spending bill] late at night. It was a surprise, done very late at night, without public knowledge or public review.").

43. H.R. 4425; see also infra section II.A.2.

44. The bill funded, inter alia, funding for military activities in Kosovo, aid to Columbia, and domestic disaster relief. Id. See also Ground Water Protection Council, Supplemental Spending Bill Approved by Congress Stops TMDL Rulemaking Effort (2000), at http://gwpc.site.net/News/nws-tmdlrulemaking.htm.

45. For a discussion of the use and abuse of appropriations riders, see Sandra A. Zellmer, Sacrificing Legislative Integrity at the Altar of Appropriations Riders: A Constitutional Crisis, 21 HARV. L. REV. 457 (1998). In this case, President William J. Clinton complemented Congress for dropping "several anti-environmental riders," without mention of its rider blocking the TMDL rule—which he knew he could finalize before he signed the rider. See H.R. 4425, Such is the language of politics.

46. Seth Borenstein & Steven Chomma, Clinton Hurries EPA to Beat Ban by Congress: Clean Water Bill Needs Signature by July 13, TIMES PICAYUNE, July 6, 2000, at A-4.

47. See Water Quality Standards: Final TMDL Rule Signed by Browner, supra note 40.

48. 65 Fed. Reg. at 43585.

49. See Water Quality Standards: Final TMDL Rule Signed by Browner, supra note 40; see also Copeland, supra note 30, at 4.

50. For one criticism of Administrator Whitman's environmental record while governor of New Jersey, see RICHARD CAPLAN, POLLUTERS' PLAYGROUND: HOW THE GOVERNMENT PERMITS POLLUTION 10, 11 (U.S. PIRG Educ. Fund., May 2001).

In her first budget, she cut more than 200 positions from the department. Other anti-enforcement actions included: eliminating the Office of Environmental Prosecutor and the Office of the Public Advocate; downsizing the DEP from 3.729 staff in 1992 to a low of 3.022 in 1998; eliminating the DEP lab; eliminating more than 100 stations that monitor water quality; removing more than 1,000 chemicals from the right-to-know list and thus placing them out of state inspection requirements; and conducting far fewer inspections by environmental regulators.

51. See Rena L. Steinzor, EPA and Its Sisters at 30: Devolution, Revolution, or Reform?, 31 ELR 11086 (Sept. 2001) ("As long as EPA . . . mollifies the states by turning the other way when they shove unpopular federal programs off the table, the faster the crises threatening the rule of law will grow.").

52. TMDL PROGRAM, supra note 1, at 131-41. See also WILLIAM H. ROGERS JR., ENVIRONMENTAL LAW 288-89 (1986).

53. See Oliver A. Houck, The Regulation of Toxic Pollutants Under the Clean Water Act, 21 ELR 10528 (Sept. 1991) (quoting floor statements of seven Senators and Representatives identifying weaknesses in the water quality standards program); see also the statement of the Senate conferencees on the 1972 Amendments to the CWA:

The Committee adopted this substantial change because of the great difficulty associated with establishing reliable and enforceable precise effluent limitations on the basis of a given stream quality. Water quality standards, in addition to their deficiencies in relying on the assimilative capacity of receiving waters, often cannot be translated into effluent limitations—defendable in court tests, because of the imprecision of models for water quality and the effects of effluents in most waters.

S. REP. NO. 92-414 (1972), reprinted in 1972 U.S.C.C.A.N. 3675.

54. OFFICE OF WATER QUALITY, U.S. EPA, ASSESSMENT OF STATE NEEDS FOR TECHNICAL ASSISTANCE IN NPDES PERMITTING (1984).

55. OFFICE OF TECHNOLOGY ASSESSMENT, U.S. EPA, WATERS IN MARINE ENVIRONMENTS 206 (1987).

56. Congress Lacks Conclusive Data Needed to Assess Water Act, Hydrologist Says, 25 Env't Rep. (BNA) 3235 (Apr. 30, 1993); see also U.S. GAO, NATIONAL WATER QUALITY ASSESSMENT, GEOLOGICAL SURVEY, FACES FORMIDABLE DATA MANAGEMENT CHALLENGES (1993).

57. PUBLIC EMPLOYEES FOR ENVIRONMENTAL RESPONSIBILITY, MARKING WATERS, OFFICIAL WATER QUALITY REPORTS ARE ALL WET: AN INSIDE LOOK AT EPA'S IMPLEMENTATION OF THE CLEAN WATER ACT 2, 3, (1999).

58. See U.S. GAO, AIR POLLUTION: NATIONAL AIR MONITORING NETWORK IS INADEQUATE (1989).

59. U.S. GAO, KEY EPA AND STATE DECISIONS LIMITED BY INCONSISTENT AND INCOMPLETE DATA (2000) [hereinafter KEY EPA AND STATE DECISIONS].

60. Id. at 5.

61. Id. at 5-7. The Office of the Inspector General of EPA reaches the same conclusion in OFFICE OF INSPECTOR GENERAL, U.S. EPA, INCONSISTENCIES IN STATES' LISTINGS OF IMPAIRED WATER BODIES MAY DELAY TMDL DEVELOPMENT AND IMPLEMENTATION (2001).

62. U.S. GAO, KEY EPA AND STATE DECISIONS, supra note 59, at 9, fig. 2.

63. Id. at 8.

64. Id. at 28, 31, 45.

65. Id. at 7, fig. 7; id. at 43, 44.

66. Id.

67. Id.

68. See TMDL PROGRAM, supra note 1, at 10-34 (state efforts in the 1960s and 1970s); id. at 133-34 (state efforts today).

69. Water quality-based management begins with the designation of a water "use," which determines water quality "standards," which determine, in turn, the amounts of pollution permissible. See 33 U.S.C. § 1313(c)(2)(A), ELR STAT. FWPCA § 303(c)(2)(A) (describing state designation of water uses); Mississippi Comm'n on Natural Resources v. Costle, 625 F.2d 1269, 10 ELR 20931 (5th Cir. 1980) (affirming broad state discretion in setting use-based standards).

70. See Donald W. Stever, Waste Load Allocation, in 2 LAW OF ENVIRONMENTAL PROTECTION 12-13 to 12-14 (Sheldon M. Novick et al. eds., 1998) ("the political difficulties interest in such a scheme are obvious. Given a choice between a sometimes intolerable burden on existing dischargers or saying 'no' to a new industry and its local economic benefits, the state agencies would be pressured to go along with a third alternative: reclassifying the stream segment to downgrade it.").

71. See TMDL PROGRAM, supra note 1, at 10-34. By imposing national technology standards, the CWA limited state authority to the application of these standards in state-delegated programs and to the imposition of additional water quality-based limits where necessary.

72. Id.

73. Special Report, The Clean Water Act Amendments of 1987. Daily Env't Rep. (BNA), Sept. 4, 1987, at 4 (quoting EPA Assessment Administrator for Water, Rebecca Hanmer).

74. Susan Bruninga, TMDL Rule to Be Delayed Until April 2003; More Time Allowed for Impaired Waters Lists, 32 Env't Rep. (BNA) 2065, Oct. 26, 2001.

75. See WATER SCIENCE AND TECHNOLOGY BOARD, NATIONAL RESEARCH COUNCIL, NATIONAL ACADEMY OF SCIENCE, SURVEY DATA, ASSESSING THE SCIENTIFIC BASIS OF THE TOTAL MAXIMUM DAILY LOAD APPROACH TO WATER POLLUTION REDUCTION (undated) (on file with author) (identifying task as focused on "(1) what information is needed to determine TMDLs for impaired waters, (2) the sufficiency of knowledge about point and nonpoint sources of pollution, (3) the state of monitoring and modeling to assess and predict pollutant loads, and (4) the effectiveness of management approaches in controlling nonpoint source pollution").

76. Indeed, the NAS Staff Director of the study called the deadline "ungodly." Craig Pittman, Was Panel on Water Cleanup Biased?, ST. PETERSBURG TIMES, Sept. 3, 2001, at 1A.

77. NRC, ASSESSING THE TMDL APPROACH TO WATER QUALITY MANAGEMENT 3. 4, (June, 2001) [hereinafter NRC REPORT]. In subsequent testimony before Congress, the report panel chair, Dr. Kenneth Reckhow, summarized the report as finding that "we have the scientific capacity to identify the nation's polluted water and develop plans for their cleanup." National Association of Flood and Stormwater Management Agencies. TMDL Hearing on Science Report: NRC Witnesses Stress That Science Exists to Identify Polluted Waters and Plan for Cleanup, But Improvements to TMDL Program Needed, at www.nafsma.org/tmdlsciencereport.htm (last visited Feb. 13, 2002).

78. NRC REPORT, supra note 77, at 3.

79. Susan Bruninga, Water Quality Standards: Agency TMDL Proposal Costly, Technical, Unenforceable, Senate Chairman Tells Group, 31 Env't Rep. (BNA) n.p. (Mar. 3, 2000) (comments of Sen. Robert Smith of New Hampshire).

80. See Pittman, supra note 76 ("Whitman had the perfect excuse for postponing them [the TMDL rules]: The program had been criticized in a recent report by a panel from the prestigious National Academy of Sciences.").

81. NRC REPORT, supra note 77, at 81 (identifying "extremely large data requirements"), 82 (finding nonpoint source data "much less available and reliable" and "data generally unavailable" for the evaluation of the efficacy of nonpoint management controls).

82. Id. at 5.

83. Id. at 32-38.

84. For a discussion of the use of biological criteria, see Houck, The Regulation of Toxic Pollutants Under the Clean Water Act, supra note 53, at 10558-59. See also NRC REPORT, supra note 77, at 44-49.

85. NRC REPORT, supra note 77, at 33-38.

86. Id. at 41. How these two measuring sticks are applied together is an important question. If, for example, biological surveys are used to override chemical monitoring showing pollution, they will simply create one more loophole in the program. For this reason, the state of Ohio, a leader in biological assessment of water quality, only allows these assessments to upgrade control requirements, not both to lower them. See E-mail correspondence with Elaine Marsh, Project Director. Ohio Greenways (Jan. 17, 2002). Unfortunately, the Ohio program is "being systematically dismembered by budget cuts." E-mail correspondence with Michael Utl, President, Ohio Smallmouth Alliance (Jan. 15, 2000).

87. NRC REPORT, supra note 77, at 53-55. This approach is modeled after that proposed in Florida. Id.

88. NRC REPORT, supra note 77, at 53.

89. Id. at 58. See also infra section III.A.2.b.

90. The report recommends, in this regard, that if a water body is not removed from the preliminary list at the end of a listing cycle, perhaps five years, it would automatically be placed on the TMDL list, a "hammer" provision seen in RCRA and other programs to action-force a decision. NRC REPORT, supra note 77, at 56. As a practical matter, whether such a redesignation could resist legal challenges on the basis of insufficient data, once the state had raised the data bar, is at least problematical. Also problematical, of course, is whether the states and EPA would ever accept such a regulatory hammer.

91. The report starts beating the drum for use-based management at the very beginning and continues the call throughout the report. Indeed, its first recommendation is that "the TMDL Program should focus first and foremost on improving the condition of waterbodies as measured by attainment of designated uses." Id. at 3. On page 6, it provides a flow chart that inserts a "Review Use/Standard" to the § 303(d) process between listing and the development of TMDLs: such review is not contained in the statute or in EPA regulations or guidelines. On page 90, the report is still arguing that before a waterbody is listed, the state conduct a review of "the appropriateness of the water quality standard." Id. at 90. In short, in the mind-set of the report's authors, TMDLs are a drastic measure of last resort.

92. Id. at 49 (emphasis in the original).

93. Id. at 93.

94. Id. at 101.

95. Id.

96. Id.

97. See Pittman, supra note 76. The author of this instant Article has served on two NAS panels and has found this observation to be correct.

98. See supra note 76 and accompanying text.

99. Id. The committee's activism even brought it into conflict with congressional staff. See Susan Bruninga, Scientists, Hill Staff Differ Over Charge of Science Panel Examining TMDL Program, Envt Rep. (BNA) n.p. (Jan. 26, 2001).

100. Id.

101. Dr. Leonard Shabman, a professor of research and environmental emissions who was working with the NRC at the time, is credited as "having played a key role in drafting the text and developing the regulations." Pittman, supra note 76.

102. Kurt Stephenson & Leonard Shabman, The Trouble Water Implementing TMDLs: A New Federal Rule Would Hurt Free Market Efforts to Protect Waterways, REG., Spring 2001, at 28.

103. The report's concluding chapter is entitled "TMDL Implementation Challenges." NRC REPORT, supra note 77, at 97.

104. Stephenson & Shabman, supra note 102, at 29.

105. Id.

106. Id. at 32.

107. NRC REPORT, supra note 77, at 101.

108. The goal of the CWA, adopted in 1972 and maintained against heavy fire from states and industry, is "to restore and maintain the chemical, physical and biological integrity of the nation's waters." 33 U.S.C. § 1251(a), ELR STAT. FWPCA § 101(a). In order to achieve this end, "it is the national goal that the discharge of pollutants in the navigable waters be eliminated by 1985." Id. § 1251(a)(1), ELR STAT. FWPCA § 101(a)(1).

109. See U.S. EPA, THE NATIONAL COSTS OF THE TOTAL MAXIMUM DAILY LOAD PROGRAM (DRAFT REPORT) 1 (2001) (available from the ELR Document Service, ELR Order No. AD-4655) [hereinafter EPA COSTS REPORT]. The directive came as a "request" in the Conference Report on the VA/HUD and Independent Agencies Appropriations Act for FY 2001. Id.

110. Id. at 2. EPA's previous estimates of TMDL costs were widely criticized as incomplete and low-end. See U.S. GAO, CLEAN WATER ACT: PROPOSED REVISIONS TO EPA REGULATIONS TO CLEAN UP POLLUTED WATERS (2000). See also Copeland, supra note 30 (noting state criticisms of cost estimates).

111. EPA COSTS REPORT, supra note 109, at 2.

112. U.S. EPA, Notice of Availability of Draft Report on Costs Associated With the Total Maximum Daily Load Program and Request for Costs, 66 Fed. Reg. 41876 (Aug. 9, 2001).

113. EPA COSTS REPORT, supra note 109, at i, 12.

114. Id.

115. Id.

116. Id. at 16.

117. Id. at 12.

118. Id.

119. Id. at 13, tbl. IV-1.

120. Id.

121. Id. at ii, 15-21. The average costs per TMDL per pollutant was estimated at 28,000, with a typical range of $ 6,000 to $ 154,000. On a water body as opposed to an individual pollutant basis, the costs were estimated at $ 52,000, with a range of $ 26,000 to $ 500,000. Complex TMDLs, characterized as "outliers," were estimated at up to $ 1 million.

122. Id. at iii.

123. Id. at 28-31.

124. Id. at 28, 29.

125. Id. at 29.

126. Id. at 30, 31.

127. Id. at 35, tbl. VI-1.

128. Id.

129. Susan Bruninga, Water Pollution Costs of Implementing TMDL Regulation Underestimated in EPA Report, Groups Say, Daily Env't Rep. (BNA), Dec. 14, 2001, at A-1.

130. Id.

131. Id.

132. Id. The comments were submitted by the Mercatus Center for Regulatory Studies at George Mason University. Id.

133. Susan Bruninga, Agency Seeks Public Comment on Costs of Implementing Rule on Impaired Waters, 32 Env't Rep. (BNA) n.p. (Aug. 10, 2001) (quoting Rick Parish of the Southern Environmental Law Center).

134. See Kenneth H. Reckhow, Cuts in USGS Budget Undermine Sound Water Science, cited in e-mail from river info-administrative river network.org (May 2, 2001).

135. OFFICE OF THE INSPECTOR GENERAL (OIG), EPA STATE ENFORCEMENT OF CLEAN WATER ACT DISCHARGERS CAN BE MORE EFFECTIVE 8 (2001) [hereinafter OIG REPORT].

136. A representative list of the more recent reports on enforcement include: U.S. GAO, ENVIRONMENTAL ENFORCEMENT: EPA CANNOT ENSURE THE RECOVERY OF SELF-REPORTED COMPLIANCE MONETARY DATA (1993); OFFICE OF ENFORCEMENT AND COMPLIANCE ASSURANCE, U.S. EPA, THE STATE OF FEDERAL FACILITIES (2000) (EPA 300 R-00-00); U.S. GAO, WATER POLLUTION: DIFFERENCES AMONG THE STATES IN ISSUING PERMITS LIMITING THE DISCHARGE OF POLLUTANTS (1996) (GAO/RCED-96-42); U.S. GAO, WATER POLLUTION: MANY VIOLATIONS HAVE NOT RECEIVED APPROPRIATE ENFORCEMENT ATTENTION 23 (1996) (GAO/RCED-96-23); Clifford Rechtschaffen, Deterrence vs. Cooperation and the Evolving Theory of Environmental Enforcement, 71 S. CAL. L. REV. 1181 (1998); John H. Cushman, EPA and States Found to Be Lax on Pollution Law: Enforcement Is Faulted: Agencies Are Failing to Inspect, Issue Permits and Report Violations, Audit Says, N.Y. TIMES, June 7, 1998, at 1.

137. David L. Markell, The Role of Deterrence-Based Enforcement in a "Reinvented" State/Reduced Relationship: The Divide Between Theory and Reality. 24 HARV. ENVTL. L. REV. 1. 55 (2000).

138. CAPLAN, supra note 50, at 5.

139. State agencies issue more than 90% of all CWA permits and conduct over 75% of CWA enforcement. OIG REPORT, supra note 135, at 6. See also R. Steven Brown, The States Protect the Environment, ECOSTATES MAG, (Summer 1999), available at http://www.sso.org/ecos/publications/statesarticle.htm ("A remarkable, and largely unnoticed, change in environmental protection has occurred over the past five to 10 years. The states have become the primary environmental protection agencies across the country."). Mr. Brown is Director of Research for the Environmental Council of the States.

140. See Markell, supra note 137; ENVIRONMENTAL LAW INST., REPORT OF THE COLLOQUIUM ON FEDERAL-STATE RELATIONS IN ENVIRONMENTAL ENFORCEMENT (1992); David R. Hodas, Enforcement of Environmental Law in a Triangular Federal System: Can Three Not Be a Crowd When Enforcement Authority Is Shared by the United States, the States, and Their Citizens?, 54 MD. L. REV. 1552 (1995).

141. See Clifford Rechtschaffen, Competing Visions: EPA and the States Battle for the Future of Environmental Enforcement, 30 ELR 10803 (Oct. 2002). Professor Rechtschaffen concludes: "Our experience to date . . . suggests that it would be ill-advised to make a wholesale shift away from deterrence-based practices." Id. at 10828.

142. 120 S. Ct. 693, 30 ELR 20246 (2000).

143. See the discussion of nonenforcement against Bethlehem Steel and Smithfield Farms in Steinzor, supra note 51, and Dereck Yeo & Roy A. Hogland, United States v. Smithfield: A Paradigmatic Example of Lax Enforcement of the Clean Water Act by the Commonwealth of Virginia, 23 WM. & MARY ENVTL. L. & POL'Y REV. 513 (1999). The same record of nonenforcement is found against the Exxon Mobil refineries in OMB Watch, States Stack Off on Environmental Enforcement, at http://www.ombwatch.org/execreport/enforcement.html (last visited Feb. 22, 2002). For a report of similar underperformance in Michigan, see Michigan Environmental Council, Protect Our Children's Health, at http://www.mecprotects.org/HEALTH.html (last visited Feb. 22, 2002).

144. See Robert Anderson, Rollins Told to Close Site, MINING ADVOC. (Baton Rouge), Aug. 7, 1985, at A1 (loss of controls at plant led to shutdown order); Bill Grady, Former DEQ Chief Is Still in the Fight, TIMES PICAYNE, June 1, 1992, at B-I.

145. A survey of Massachusetts environment enforcement officers reported that managers had "inappropriately intervened in a criminal investigation" and that nearly one-third "fear retaliation from [chain of command] for advocating strong environmental enforcement." Press Release, Public Employees for Environmental Responsibilities, Environmental Cops Cannot Enforce the Law (Apr. 11, 2001), cited in CAPLAN, supra note 50, at 8.

146. OIG REPORT, supra note 135.

147. Id. at 8.

148. Id. at i-iii, 5-16, 36-42.

149. Id. at iii, 20-34, 43-50.

150. Id. at i, 5-13.

151. Id. at 17.

152. Id. at 43.

153. According to a study released in January 2002 by a Wisconsin environmental organization, the state lost between $ 14 and $ 248 million in 1998 by failing to prosecute water pollution law violations. Among the findings:

. From 1990 to 1998, on average, the Department of Natural Resources (DNR) sent notices of violation to only 10% of all municipal and industrial facilities that were in significant noncompliance with their Wisconsin pollutant discharge elimination system (WPDES) permits.

. During that same time period, the DNR only referred to the U.S. Department of Justice for prosecution 2.5% of all industrial and municipal facilities that were in significant noncompliance with their WPDES permits.

. In 1999, the DNR failed to inspect 53% of all major industrial facilities with WPDES permits.

. In 2000, only four industrial and municipal facilities were prosecuted for water pollution violations. Of these four, they paid only $ 212.217 dollars to the state in penalties.

Press Release, Midwest Environmental Advocates, Millions in State Revenues Lost Due to Lack of Enforcement of Water Laws (Jan. 28, 2002).

154. OIG REPORT, supra note 135, at 6-8. See also TMDL PROGRAM, supra note 1, at 88-94 (describing CAFO and municipal runoff programs).

155. OIG REPORT, supra note 135, at 20 ("The question, variety and complexity of the regulated community had greatly outstripped the system's capabilities. Dischargers not monitored by the [EPA compliance system] include: stormwater, [CAFOs], and sewer overflows."). Focused on the reporting and paperwork requirements for major dischargers, states tend to let the smaller discharges—which are precisely those at issue in nonpoint source positions—go unattended, even where they are purportedly regulated under the NPDES program. See the following notices of intent to file citizen suits of the Tulane Environmental Law Clinic:

Notice of Environmental Action Network re: General Animal Hospital, Inc. (failure to operate and maintain pollution control equipment, and to maintain effluent limitations) (Jan. 29, 2002).

Notice of Louisiana Environmental Action Network re: Bank One—Covington Banking Center (failure to operate and maintain pollution control equipment, and to maintain effluent limitations) (Jan. 29, 2002).

Notice of Louisiana Environmental Action Network re: Beau Amis Lounge, Inc. (failure to operate and maintain pollution control equipment and to maintain effluent limitations) (Jan. 29, 2002).

Notice of Louisiana Environmental Action Network re: Zelden Physical Therapy (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Jan. 25, 2002).

Notice of Louisiana Environmental Action Network re: Trinity Baptist Church (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Jan. 25, 2002).

Notice of Louisiana Environmental Action Network re: Gloria Coker, M.S. (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Jan. 23, 2002).

Notice of Louisiana Environmental Action Network re: Garrity, Sanders, Reed & Caire (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Jan. 23, 2002).

Notice of Louisiana Environmental Action Network re: Holton Enterprises of Covington, Inc. (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Dec. 6, 2001).

Notice of Louisiana Action Network re: Northlake Moving and Storage, Inc. (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Dec. 6, 2001).

Notice of Louisiana Environmental Action Network re: Plaza Professional Center, Inc. (failure to operate and maintain pollution control equipment and to file discharge monitoring reports to ensure compliance) (Dec. 6, 2001).

156. See Weverhauser Co. v. Costle, 590 F.2d 1011, 9 ELR 20284 (D.C. Cir. 1978) (describing the ineffectiveness of water quality enforcement prior to the 1972 CWA Amendments).

157. 33 U.S.C. § 1365, ELR STAT. FWPCA § 505 (citizen suits). For the effectiveness of CWA citizen suits, see Hodas, supra note 140, at 1617-55; JEFFREY G. MILLER. CITIZEN SUITS: PRIVATE ENFORCEMENT OF FEDERAL POLLUTION CONTROL LAWS (1987).

158. See TMDL PROGRAM, supra note 1, at 55-56.

159. U.S. EPA, Delay of Effective Date of Revisions to the Water Quality Planning and Management Regulation, Notice of Proposed Rulemaking, 66 Fed. Reg. at 41877. The decision was made final in October. U.S. EPA, Effective Date of Revisions to the Water Quality Planning and Management Regulation, 66 Fed. Reg. 53044 (Oct. 18, 2001).

160. See CAPLAN, supra note 50, at 10, 11.

161. Id.

162. Id. at 12. The final rule defines "reasonable assurance" as "management means or other control actions (regulatory or voluntary)" that (1) apply specifically to the impaired water body, (2) will be accomplished through "reliable and effective" mechanisms, (3) will be implemented "as expeditiously as practicable," and (4) will be supported by "adequate funding." Id.

163. See comments of the Earthjustice Legal Defense Fund representative on the proposed TMDL deadlines, Water Quality Standards: Agency's Draft Final Regulation, supra note 26.

164. See Copeland, supra note 30, at 9-10.

165. Id.

166. Press Release, U.S. EPA. Public Meetings on the TMDL Program and Related Areas of the NPDES Program (undated), available at http://www.epa.gov/owow/tmdl/meetings/ (last visited Oct. 25, 2001).

167. The listening sessions took place in five cities, each with an ostensibly different theme: Chicago (nonpoint sources); Sacramento (TMDL content): Atlanta (EPA role, permitting); Oklahoma (listing); Washington, D.C. (all issues). Press Release, U.S. EPA, Meeting Dates, Times. Locations, and Meeting Themes (undated), available at http://www.epa.gov.owow/tmdl/meetings.html (last visited Oct. 25, 2001). In fact, the meetings were open to all comments and produced the same range of complaints heard since the days of the FACA. See Susan Bruninga, Water Pollution: EPA Listening Sessions on TMDLs Covers Debate on Implementation, More Information, 32 Env't Rep. (BNA) n.p. (Dec. 12, 2001).

168. Bruninga, Water Pollution: EPA Listening Sessions, supra note 167 (e.g., comments of Plum Creek Timber Company representative, "there are many ways to solve water quality problems, and the federal government should steer away from an approach that is too prescriptive and inflexible"; discussion of deferring TMDLs until other nonpoint programs "have had the time to work").

169. Susan Bruninga, Water Quality Standards: Monitoring Guidance Draws Support From Interested Parties; Categories Backed, 32 Env't Rep. (BNA) 2309-10 (Nov. 30, 2001).

170. Id. See also Susan Bruninga, Water Quality Standards: Draft Guidance on Integrating State Reports Defines Five Categories of Attainment States, 32 Env't Rep. (BNA) 2062-63 (Oct. 26, 2001). The five categories are for waters:

1. attaining all water quality standards and no standards are threatened;

2. attaining some water quality standards, no standards are threatened, and insufficient data are available to determine if the remaining standards are met or threatened;

3. for which insufficient data are available to determine of water quality standards are met;

4. impaired or threatened for one or more water quality standards but not needing a [TMDL]; and

5. impaired or threatened for one or more water quality standards and a TMDL is needed.

Id. Only the last category requires a TMDL.

171. Id. ("This allows us to focus on waters that we really know are impaired, the industry official said.").

172. Id. ("This is the next step in a pattern developed by EPA of allowing, or encouraging, states to remove from the list waters requiring a TMDL, she said," quoting an attorney for Earthjustice.).

173. See Susan Bruninga, EPA Draft Plan to Assist Streamlining in State Monitoring Program Expected Soon, 32 Env't Rep. (BNA) 693 (Apr. 13, 2001). CALM elements include:

. guidance or attainment and nonattainment of water quality standards;

. comprehensive monitoring coverage;

. presentation of data;

. elements of an increasingly comprehensive state monitoring system;

. causes and sources of impairment; and

. sections of "discrete types" of pollutants such as pathogens, nutrients, sediments, and fish advisories.

Id. The director of Science and Technology in EPA's Office of Water is quoted as stating that these elements are "hugely underfunded." Id.

174. See id. See also Susan Bruninga, Guidance to Integrate Water Act Reports Should Be Delaved, State Officials Tell EPA, 32 Env't Rep. (BNA) 2116-17 (Nov. 2, 2001).

175. See Bruninga, EPA Draft Plan, supra note 173.

176. Id. Over the years, dealing with the Association of State and Interstate Water Pollution Control Administrators must appear to EPA like dealing with the teachers union in a problem-ridden public school system; at some point they become part of the problem. See TMDL PROGRAM, supra note 1, at 31-33.

177. U.S. EPA, Supplemental Guidelines for the Award of Section 319 Nonpoint Source Grants to States and Territories in FY 2002 and Subsequent Years, 66 Fed. Reg. 476.53 (Sept. 13, 2001), available at http://www.epa.gov/owow/nps/cwact.html.

178. Id. at 47653.

179. Id. The "analytical link" language and role for TMDLs is identical to that of the Clinton Administration's Clean Water Action Plan. See TMDL PROGRAM, supra note 1, at 85-86.

180. 66 Fed. Reg. at 47654.

181. Id.

182. More specifically the new § 319 TMDL guidance requires:

a. An identification of the sources or groups of similar sources that will need to be controlled to achieve the load reductions established in the NPS [nonpoint source] TMDL (and to achieve any other watershed goals identified in the watershed-based plan);

b. A description of the NPS management measures that will need to be implemented to achieve the load reductions established in the NPS TMDL (as well as to achieve other watershed goals identified in the watershed-based plan); an estimate of the load reductions expected to these management measures (recognizing the natural variability and the difficulty in precisely predicting the performance of management measures over time); and an identification of the critical areas in which those measures will need to be implemented to achieve the NPS TMDL;

c. An estimate of the sources of technical and financial assistance needed, and/or authorities that will be relied upon, to implement the plan. As sources of funding, States should consider the use of their [§]319 programs. State Revolving Funds, USDA's Environmental Quality Incentives Program and Conservation Reserve Program, and other relevant Federal, State, local, and private funds that may be available to assist in implementing the plan;

d. An information/education component that will be used to enhance public understanding of the project and encourage their participation in selecting, designing, and implementing the NPS management measures;

e. A schedule for implementing the NPS management measures identified in the plan that is reasonably expeditious;

f. A description of interim, measurable milestones (e.g., amount of load reductions, or improvement in biological or habitat parameters), for determining whether NPS management measures or other control actions are being implemented;

g. A set of criteria that can be used to determine whether substantial progress is being made towards attaining water quality standards and, if not, the criteria for determining whether the NPS TMDL needs to be revised; and

h. A monitoring component to evaluate the effectiveness of the implementation efforts, measured against the criteria established under item (g) immediately above.

Id.

183. See supra section II.B.1. They are also, jot-for-jot, the requirements of 1997 EPA guidance entitled New Policies for Establishing and Implementing Total Maximum Daily Loads (TMDLs) (1997) (available from the ELR Document Service. ELR Order No. AD-3467). See TMDL PROGRAM, supra note 1, at 77-82.

184. See Farm Groups Seek Major Revisions to Recent Water Grants Guidance, INSIDE EPA, Oct. 26, 2001, at 20.

185. Id.

186. See Susan Bruninga, Water Quality Standards, Draft Policy Statement Being Crafted by EPA on Market-Based Approaches, 32 Env't Rep. (BNA) 2351 (Dec. 7, 2001).

187. U.S. EPA, Effluent Trading in Watersheds Policy Statement, 61 Fed. Reg. 4994 (Feb. 9, 1996).

188. U.S. EPA, Draft Framework for Watershed-Based Trading, 61 Fed. Reg. 29563 (June 11, 1996).

189. Id.

190. Id.

191. See Bruninga, Water Quality Standards, Draft Policy Statement, supra note 186, at 2352.

192. See 61 Fed. Reg. at 4994 (presenting advantages of trading programs).

193. "For instance, requiring point sources to obtain [NPDES] permit modification and requiring point source requirements to be incorporated into NPDES permits is not generally conducive to implementing market-based strategies." See Bruninga, Water Quality Standards, Draft Policy Statement, supra note 186 (quoting EPA Associate Administrator for Water). On the contrary, the case could be made that such permit requirements are indispensable to market-based strategies; without them, there is neither obligation nor accountability.

194. The Clinton Administration policy statement was clear on this point but, as with any policy, it is subject to change. See EPA Eyes Replacing Facility Water Permits With Watershed Approach, INSIDE EPA, Feb. 8, 2002.

195. See ROGERS, supra note 3, at 221-23; see also Citizens Against the Refinery's Effects v. EPA, 643 F.2d 183, 11 ELR 20176 (4th Cir. 1981) (manipulation of air emission offsets); Bethlehem Steel Corp. v. EPA, 782 F.2d 645, 16 ELR 20268 (7th Cir. 1986) (same).

196. For a discussion of these difficulties in the CAA state implementation plan (SIP) program, see Robert W. Adler, Integrated Approaches to Water Pollution: Lessons From the Clean Air Act, 23 HARV. ENVTL. L. REV. 203, 243-45, 282. ("Critics charge that states produce 'cheater SIPs' that the state never intends to implement, and that emission inventories base attainment or maintenance predictions on unrealistically optimistic assumptions, questionable baselines, paper offsets or emission reductions that never materialize.") Id. at 245, citations omitted. The process is known as "gaming," described by a former EPA Administrator as "reaching the same high art of gamesmanship as lawyering." Id. (citing David Schoenbrod, Goals or Rules Statutes, the Case of the Clean Air Act, 32 UCLA L. REV. 740, 773 (1983).

197. See Adler, supra note 196, and Schoenbrod, supra note 196. See also William F. Pedersen Jr., The Limits of Market-Based Approaches to Environmental Protection, 24 ELR 10173 (Apr. 1994).

198. See Joint Motion for Voluntary Remand. Louisiana Envtl. Action Network v. EPA, No. 99-60570 (5th Cir. Oct. 9, 2002). (Louisiana Department of Environmental Quality failed to assess validity of emission reductions credits).

199. See DOJ Audit of New Jersey Trading Plan May Be Sign of CAA Violation, INSIDE EPA, Jan. 18, 2002, at 14.

200. 33 U.S.C. § 1344, ELR STAT. FWPCA § 404.

201. See WILLIAM L. WANT, THE LAW OF WETLAND REGULATION § 6.10 (1990).

202. See NRC, COMPENSATORY FOR WETLAND LOSSES UNDER THE CLEAN WATER ACT (2001) (severely criticizing administration and performance of mitigation under § 404); see also U.S. GAO, WETLANDS PROTECTION: ASSESSMENTS NEEDED TO DETERMINE EFFECTIVENESS OF IN LIEU FEE MITIGATION (2001); J.B. Ruhl & R. Juge Gregg, Integrating Ecosystem Services Into Environmental Law: A Case Study of Wetlands Mitigation Banking, 20 STAN. ENVTL. L.J. 365 (2001); see U.S. Army Corps of Engineers, Regulatory Guidance Letter, Guidance for the Establishment and Maintenance of Compensatory Mitigation Projects Under the Corps Regulatory Program Pursuant to Section 404(a) of the Clean Water Act and Section 10 of the Rivers and Harbors Act of 1999 (2001).

203. See Oliver A. Houck, More Net Loss of Wetlands: The Army-EPA Agreement for Mitigation, 20 ELR 10212 (June 1990). See also Endangered Species Coalition, GREENLines (Feb. 15, 2002) (citing a study by the Washington Department of Ecology concluding that "only about 13% of the man-made wetlands in Washington are fully successful").

204. See Bruninga, Water Quality Standards, Draft Policy Statement, supra note 186, at 2352.

205. Id.

206. 42 U.S.C. § 9607(a)(4)(C), ELR STAT. CERCLA § 107(a)(4)(C); see also 43 C.F.R. 11 (National Resource Damage Assessments). Ohio v. Department of the Interior, 800 F.2d 432, 19 ELR 21099 (D.C. Cir. 1989) (invalidating the U.S. Department of the Interior's original rule).

207. 33 U.S.C. § 2706, ELR STAT. OPA § 1006; see also 15 C.F.R. § 990 (National Resource Damage Assessments); General Elec. Co. v. U.S. Department of Commerce, 128 F.3d 767, 28 ELR 20263 (D.C. Cir. 1997) (invalidating, in part, National Occanic and Atmospheric Administration (NOAA) original rule).

208. See Bruninga, Water Quality Standards, Draft Policy Statement, supra note 186 (quoting EPA Assistant Administrator for Water's interest in the question of "natural capital and its benefits").

209. See Douglas R. Williams. Valuing Natural Environments: Compensation, Market Norms, and the Ideal of Public Goods, 27 CONN. L. REV. 365 (1997).

210. See ROBERT W. ADLER ET AL., THE CLEAN WATER ACT 20 YEARS LATER 118-28 (NRC 1993); see also Houck, The Regulation of Toxic Pollutants, supra note 53, at 10544-54.

211. See Houck, The Regulation of Toxic Pollutants, supra note 53, at 10549-54 (describing wide discrepancies in state water quality standards for dioxin). See also ZYGMUNT J.B. PLATER ET AL., ENVIRONMENTAL LAW AND POLICY: NATURE, LAW, AND SOCIETY 538-540 (2d ed. 1998) (describing differences in 775 times between New York and Pennsylvania emission limitations, also describing "play" in state management systems and "race of laxity" to lower standards); U.S. GAO, DIFFERENCES AMONG THE STATES IN ISSUING PERMITS LIMITING THE DISCHARGE OF POLLUTANTS (1996).

212. Two key safeguards of the water quality program are restrictions on antidegradation (lowering existing water quality within a given use). 40 C.F.R. § 131.12, and downgrading (lowering the use), 40 C.F.R. § 131.10. Both restrictions contain escape hatches, however, where maintaining existing water quality and use would have significant economic and social impact. Id. §§ 131.12(a)(2), 131.10(g)(6). Needless to say, a state unwilling to maintain high water quality could drive a truck through these loopholes without scraping the fenders. The fact that unwilling states have not done so in a widespread fashion to date reflects more their failure to apply water quality standards at all than their resolve to hold the line. With the advent of TMDLs, these restrictions and loopholes will be put to the test.

213. See U.S. EPA, Advanced Notice of Proposed Rulemaking Water Quality Standards Regulation, 63 Fed. Reg. 36742 (July 7, 1998).

214. According to EPA's Office of Science and Technology within the water program, a new draft strategy would be developed in "early 2001 (sic)." Press Release, U.S. EPA, Strategy for Water Quality Standards and Guidance (Dec. 2001).

215. Telephone conversation of Andrew L. Adams with Fred Leutner, Chief, Office of Water, U.S. EPA (Jan. 19, 2002).

216. See Upcoming EPA Water Use Guidance Sets Stage for Activist Fight, INSIDE EPA, Feb. 8, 2002, at 3 ("EPA is preparing a new guidance for streamlining the process states use to downgrade water bodies' 'designated uses,' raising fears among environmentalists that the new process will significantly increase pollution discharges.").

217. See supra notes 91-93 and accompanying text. See also Nonpoint Sources Need to Be Included in TMDL Program, Treatment Officials Say, supra note 36 (citing Association of Municipal Sewer Authorities white paper "calling on states to review and revise as needed their designated uses and water quality criteria"); one can be sure that the Association of Municipal Sewer Authorities does not have upwards revisions in mind.

218. See Comments of Jacqueline Savitz, Coast Alliance et al., to Fred Leutner, Chief, Water Quality Standards Branch, Answers to Questions on Strategies for the Water Quality Standards Program (Sept. 18, 2001). See also Ohio Valley Envtl. Coalition v. EPA, No. 7:020059 (S.D. W. Va. Jan. 23, 2002) (citizen suit challenging alleged state weakening of antidegradation standards); TMDL PROGRAM, supra note 1, at 140-42.

219. See TMDL PROGRAM, supra note 1, at 151, See also U.S. Geological Serv., National Water Quality Assessment Program, Nutrients in the Nation's Water Too Much of a Good Thing?, at http://water.usgs.gov/nawqa/cire-1136/h8.html (last visited Feb. 21, 2002); Larry J. Puckett, U.S. Geological Serv., National Water Quality Assessment Program, Nonpoint and Point Sources of Nitrogen in Major Watersheds of the United States, at http://water.usgs.gov/nawqa/wri94-4001/wri94-4001main.html (last visited Feb. 21, 2002).

220. See Susan Bruninga, Some Flexibility Afforded to States in Agency Guidance on Nutrient Criteria, 32 Env't Rep. (BNA) 2252 (Nov. 23, 2001); U.S. EPA, Background Paper No. 4; Science and Technology, FACA Committee Meeting (Nov. 19-21, 1996).

221. See TMDL PROGRAM, supra note 1, at 144 (quoting Iowa Corn Growers Association President as saying: "It doesn't jive. Two and two is not adding up to four. Agriculture is being hung with the blame and we don't think it can be substantiated."). See Mark Schleifstein, Panel Splits on River Dead Zone Goal, TIMES PICAYUNE, June 16, 2000, at A-3 (quoting the Iowa Secretary of Agriculture as stating "before she could even consider [cleanup] goals, she will have to be shown direct benefits from the reduction of fertilizers to Iowa residents"); Agriculture Department May Withhold Support of Gulf Hypoxia Study, INSIDE EPA, Dec. 24, 1999, at 7 ("the [USDA] is threatening to withhold its approval of an interagency assessment of the cause of low water oxygen levels, or hypoxia, in the Gulf of Mexico unless other federal agencies delete language from the study"): the offending language was that suggesting a 20% cut in fertilizer use.

222. See Bruninga, Some Flexibility, supra note 220.

223. Id. The description of the nutrient standards program that follows is taken from this source. For a critique of the program, see Evan Hansen & Martin Christ, EPA's Nutrient Criteria Recommendations and Their Application in Nutrient Eco Region XI (May 20, 2001), at West Virginia Rivers Coalition, http://www.wvrivers.org (on file with author).

224. See Bruninga, Some Flexibility, supra note 220, at 2253.

225. U.S. EPA, EPA NONPOINT SOURCE POLLUTION FOCUS GROUPS FINAL REPORT (2001) (on file with author).

226. Id. at 2.

227. Id. at 3.

228. Id.

229. Id. at 4.

230. Id. at 1.

231. Id. at 7.

232. Id.

233. Id.

234. Id.

235. See supra section II.A.3.

236. See E-mail From Allene Levine, Marine Conservation Network, State Water Resources Declares TMDLs (Water Quality Control Pollution Programs) Highest Priority (Oct. 15, 2001); see generally TMDL PROGRAM, supra note 1, at 144 (describing positive initiatives in, inter alia, California, Missouri, and New York).

237. The Kansas Senate has recently enacted legislation requiring a cost-benefit analysis for designating state waters "primary contact recreation," and allowing downgrades to "secondary contract recreation" on the basis of denial of stream access by adjacent landowners. Communication with Charles M. Benjamin. Legislative Coordinator for Kansas Chapter of the Sierra Club, Lawrence, Kansas (Jan. 31, 2001); see also American Wildlands v. EPA, 260 F.2d 1192, 32 ELR 20860 (10th Cir. 2001) (approving Montana's exclusion of nonpoint sources from antidegradation review). Ohio Valley Envtl. Coalition v. Whitman, No. 3:02-0059 (S.D. W. Va. Jan. 23, 2002) (challenging West Virginia antidegradation program). See generally TMDL PROGRAM, supra note 1, at 144 (describing negative initiatives in other states). The state of Wyoming, for example, reduced its lists of over 400 impaired waters to 61, with 315 needing "further monitoring"; 275 of the 315 waters so delisted were impaired by nonpoint sources, primarily cattle. See id. at 138, 139. Of course, the relationship between the cattle industry and the deferred listings could be coincidental.

238. See EVAN HANSEN, TOTAL MAXIMUM DAILY LOADS IMPLEMENTATION IN WEST VIRGINIA: A STATUS REPORT 1 (Dec. 2001).

239. See TMDL PROGRAM, supra note 1, at 105-09, 185-89.

240. Analysis of each TMDL for each factor is provided in TMDL PROGRAM, supra note 1, at app. C, 185-89.

241. See HANSEN, supra note 238.

242. Id. at 2, 3.

243. Id. at 12.

244. Id. at 13.

245. Id. at 4. See also EPA Backs Innovative Virginia Program, infra note 344 ("The [Virginia] program will also target dischargers of nonpoint sources of pollution . . . in order to utilize state and federal grants for controlling contamination through best management practices."). The assumption is, for nonpoint source abatement, the government, i.e., the general public, pays. See also infra notes 251-53 and accompanying text.

246. Telephone conversation with Allison Wiedman, Chesapeake Bay Foundation (Feb. 18, 2002). Apparently, the 40% target is now being revised downwards. Id.

247. See MISSISSIPPI RIVER/GULF OF MEXICO WATERSHED NUTRIENT TASK FORCE, ACTION PLAN FOR REDUCING, MITIGATING, AND MANAGING HYPOXIA IN THE NORTHERN GULF OF MEXICO 21 (2001).

248. Telephone conversation with Wiedman, supra note 246 (Chesapeake Bay plan). The Gulf of Mexico "Action Plan" avoids even mentioning a "goal" of 30% reduction, stating instead that implementing strategies "should be aimed at" achieving this reduction, the goals for which begin with the following principle: "Encourage actions that are voluntary, practical and cost-effective." MISSISSIPPI RIVER/GULF OF MEXICO ACTION PLAN, supra note 247, at 21. In short, a plan it may be, but a plan of action it is not.

249. See Michael Grunwald, Plan to Revive Everglades Brings Renewed Dispute: Environmentalists Say Draft Rules Offer No Gain, WASH. POST. Dec. 29, 2001, at A-3. An attorney for the National Resources Defense Council is quoted as stating: "This is a joke . . . . They are basically saying let's keep doing everything the way it's been done in the past. That's what destroyed the Everglades in the first place." Id.

250. See Elizabeth Ann Rieke, The Bay-Delta Accord: A Stride Toward Sustainability, 67 U. COLO. L. REV. 341 (1996).

251. See Landmark Nonpoint Rule Requires Wisconsin to Shoulder Most Costs, INSIDE EPA, Jan. 18, 2002, at 1.

252. See TMDL PROGRAM, supra note 1, at 144.

253. New York Farmers Paid to Shove Cows Off of City's Watershed, N.Y. TIMES, Aug. 26, 1998, at B7.

254. Office of Water U.S. EPA, TMDL Litigation by State, at http://www.epa.gov/owow/tmdl/lawsuitl.html (last visited Nov. 5, 2001).

255. See Hihiwai Stream Restoration Coalition v. Whitman, No. 00-00477 (D. Haw. Sept. 5, 2001) (ordering EPA to comply with § 303(d) and identify all of Hawaii's polluted streams and other water bodies); Save All Iowa Lakes, Oxbows, Rivers & Streams v. EPA, No. C98-134 (D. Iowa Oct. 26, 2001) (consent decree requiring Iowa to complete TMDLs for 1998 lists within 10 years); Tennessee Envtl. Council v. EPA, No. 3-01-9932 (M.D. Tenn. May 10, 2001) (consent decree requiring Tennessee to set 7,890 TMDLs under a 10-year enforceable schedule).

257. San Francisco Baykeeper, 147 F. Supp. at 1002.

258. 268 F.3d 91, 32 ELR 20203 (2d Cir. 2001).

259. Id. at 99, 32 ELR at 20207.

260. See In re Hannaford Bros. Co., No. WQ-01-01 (Vt. Water Resources Bd. June 29, 2001). See also Water Permitting Dispute May Alter Rules Setting Discharge Limits, INSIDE EPA, Mar. 2, 2001, at 18-19.

261. Sierra Club v. EPA, No. 01-145876gg (11th Cir. 2002) (oral argument set for Mar. 8, 2002).

262. Amigos Bravos v. Cook, No. 00-1615 (D.D.C. 2001) (before Judge Walton, no date set for oral argument).

263. Young v. Department of Envtl. Protection, No. 01-1462RP (Fla. Admin. Hearings 2002) (full history of the Administrative proceedings and the full text of documents can be accessed at http://www.cwn-se.org).

264. See Petition for Admin. Hearing, Young v. Department of Envtl. Protection, No. 01-1462RP (Fla. Admin. Hearings Apr. 13, 2001), at 150.

265. Identification of Impaired Surface Waters, 27 Fla. Admin. Weekly 1395, at 62-303.300 et seq. (proposed Mar. 23, 2001).

266. The example of a flock of geese landing and dropping their waters is given. Thus, a water body with only 10 to 15 samples would require 3 positives, raising EPA's 10% from 20% to 30%, and requiring 100 samples before the 10% was statistically confirmed.

267. Telephone Interview with Jan Mandrup-Poulsen, Florida Department of Environmental Protection (Oct. 2001).

268. American Farm Bureau Fed'n v. Whitman, No. 00-1320 (D.D.C. July 18, 2000).

269. In October 2001, the lawsuit was suspended for 18 months at EPA's request so that it could consider potential revisions before issuing the final rule. See Court Grants EPA Suspension of Lawsuit Over Impaired Waters Rule, INSIDE EPA, Oct. 19, 2001, at 5-6.

270. John H. Stam, Court-Ordered Deadlines for TMDLs Have Negative Consequences, ABA Told, 32 Env't Rep. (BNA) 1589 (Aug. 10, 2001); see also Susan Bruninga, Challenging TMDLs May Require Lawsuits, Alliances With Affected Parties, Lawyer Says, 31 Env't Rep. (BNA) 1575-76 (July 28, 2000).

271. See Susan Bruninga, Nonpoint Sources Should Not Be Excluded From TMDL Program, Government Argues, 31 Env't Rep. (BNA) 547-48 (Mar. 24, 2000); Suit Challenging EPA Authority to Set TMDLs for Nonpoint Sources Concerns Cities, 30 Env't Rep. (BNA) 216 (June 4, 1999).

274. Somerset County Sanitary Dist., Inc. v. Maryland Dep't of the Env't, No. 19-C-01-007932 (Md. Cir. Ct. May 11, 2001); City of Salisbury v. Maryland Dep't of the Env't, No. C-01-622 (Md. Cir. Ct. May 17, 2001).

275. Stam, supra note 270 (quoting Steven Koorse, attorney with Hunton & Williams, Richmond, Va.).

276. See TMDL PROGRAM, supra note 1, at 20-24.

277. Id. at 24 (House committee staff viewed § 303 as a "game plan for the next generation"; a Senate staff member stated that, "we didn't take it seriously and thought it would be foolish for EPA to waste time and money to implement it"). The massive irony of § 303(d) is that it was only written into the statute, as a last-minute compromise, at the insistence of the same states, industries, and trade associations who resist it today. Id. at 14-24.

278. See U.S. EPA, WATER QUALITY IMPROVEMENT STUDY (1989) (showing total loadings from industrial categories and quality of receiving waters before and after the application of the first BAT requirements).

279. Id. tbl. 1-2 (showing RAW (pre-BAT) pulp and paper industry emissions of total suspended solids at 10.8 million pounds (lbs.)/day and post-BAT at 0.89 million; the industry's organic priority pollutant discharger dropped from 32.8 thousand lbs./day to 3.3 thousand). Other industrial categories emissions fell dramatically as well, including:

280. Id. tbl. 3-1, which shows the following water quality improvement:

281. See Chemical Mfrs. Ass'n v. EPA, 870 F.2d 177, 19 ELR 20989 (5th Cir. 1989), cert. denied sub nom. PPG Indus. v. EPA, 110 S. Ct. 1936 (1990) (requiring closed cycle recycling for some operations); Rybachek v. EPA, 904 F.2d 1276, 20 ELR 20973 (9th Cir. 1990) (zero discharge for placer mining).

282. EPA Finds Significant Progress in Controlling Pollution of Water, N. Y. TIMES, Feb. 12, 1984, at A-31 (claiming an improvement of 65% in POTW discharges in the first 10 years of the CWA POTW program, and that while the population served increased by 18 million people, total pollutants dischargers had stayed "roughly constant").

283. It has not worked completely, of course, and the later history of BAT and its application to large, rather-fight-than-switch industries led to compromised standards that abandoned the CWA's "action-foreing" and best achievable goals. See Special Report, Effluent Guidelines Rulemaking Nears End: Litigation, Compliance Extensions Expected, 15 Env't Rep. (BNA) 1629-30 (Jan. 21, 1983) (describing compromises in final rules for petroleum and leather tanning industrial categories). See also Philip D. Reed, New BAT Standards: Lowering the Ceiling or Raising the Floor?, 13 ELR 10002 (Jan. 1983) ("It can be argued that EPA has coupled relatively tough [best practical technology] with relatively lenient BAT.").

284. For the collapse of Lake Erie, see BARRY COMMONER, THE CLOSING CIRCLE 94-111 (1971); for rivers catching fire, see Patricia Howard, A Happier Cleveland, HOUS. POST, Oct. 24, 1990, at A2 (describing 1969 fire on the Cuyohoga River in Cleveland, Ohio).

285. For a general discussion of the theory and attractions of water quality-based regulation, see N. William Hines, Nor Any Drop to Drink: Public Regulation of Water Quality Part I: State Pollution Control Programs, 52 IOWA L. REV. 186 (1966); 2 WILLIAM H. ROGERS JR., ENVIRONMENTAL LAW: AIR AND WATER 242-53 (1986).

286. Water Quality Act, ch. 758, tit. III, § 303, 62 Stat. 1155 (1948). For a history of federal water quality legislation prior to the 1972 Amendments, see TMDL PROGRAM, supra note 1, at 12-14.

287. 42 U.S.C. §§ 7408, 7409, ELR STAT. CAA §§ 108, 109; see also infra section III.B.1.

288. Toxic Substances Control Act of 1976, Pub. L. No. 94-469, 90 Stat. 2003 (current version at 15 U.S.C. §§ 2601-2692, ELR STAT. TSCA §§ 2-412); Resource Conservation and Recovery Act, 42 U.S.C. §§ 6901-6992k, ELR STAT. RCRA §§ 1001-11011; Federal Insecticide, Fungicide, and Rodenticide Act of 1972, Pub. L. No. 92-516, 86 Stat. 987 (current version at 7 U.S.C. §§ 136-136y. ELR STAT. FIFRA §§ 2-34); Comprehensive Environmental Response, Compensation, and Liability Act of 1980, Pub. L. No. 96-510, 94 Stat. 2767 (current version at 42 U.S.C. §§ 9601-9675, ELR STAT. CERCLA §§ 101-405); Safe Drinking Water Act of 1974, Pub. L. No. 93-523, 88 Stat. 1660 (current version at 42 U.S.C. §§ 300f to 300j-26, ELR STAT. SDWA §§ 1401-1465).

289. For a view of these difficulties under the CWA, see Environmental Defense Fund v. EPA, 598 F.2d 62, 8 ELR 20765 (D.C. Cir. 1978) (upholding EPA extrapolations of human harm from tests on laboratory mice, stating, "as a practical matter, scientific knowledge about the effect of industrial chemicals cannot keep pace with the ability of industrial laboratories to create new ones," and "what scientists know about the causes of cancer is how limited is their knowledge"). Id. at 82, 8 ELR at 20780. See also Hercules, Inc. v. EPA, 598 F.2d 91, 8 ELR 20811 (D.C. Cir. 1978) (upholding EPA standards for the pesticides, derived through a complex mix of extrapolations and arbitrary safety factors).

290. See ENVIRONMENTAL DEFENSE FUND, TOXIC IGNORANCE: THE CONTINUING ABSENCE OF BASIC HEALTH TESTING OF POP-SETTING CHEMICALS IN THE UNITED STATES (1997); ROBERT V. PERCIVAL ET AL., ENVIRONMENTAL REGULATION 522 (2d ed. 1996) ("Despite amendments to FIFRA, the process of canceling a pesticide remains fraught with considerable procedural difficulties.").

291. For difficulties in CAA standard setting, see Natural Resources Defense Council v. EPA, 824 F.2d 1146, 17 ELR 21032 (D.C. Cir. 1987) (vinyl chloride regulations); Natural Resource Defense Council v. EPA, 695 F.2d 48, 19 ELR 20344 (D.D.C. 1988) (benzene emissions). See PERCIVAL ET AL., ENVIRONMENTAL REGULATION, supra note 290, at 780 ("The regulatory burden in establishing a NAAQS is so demanding that EPA has strong incentives to avoid making frequent charges on such standards, much less to promulgate new ones."); id. at 822 ("Uncertainty over the causes and effects of acid deposition contributed to political gridlock.").

292. While an entire subject to itself, for a critical commentary on CERCLA risk-based decisionmaking, see STEPHEN BREYER, BREAKING THE VICIOUS CYCLE (1993) (arguing for the creation of an independent science board to make correct and apolitical federal risk decisions); for a more technical walk through CERCLA's risk-assessment woods, see Lawrence E. Starfield, The 1990 National Contingency Plan—More Detail and More Structure, But Still a Balancing Act, 20 ELR 10222 (June 1990); Linda Malone, Bioavailability: On the Frontiers of Science and Law in Cleanup Methodologies for Contamination, 31 ELR 10800 (Aug. 2001).

293. See supra note 291 (air toxics): Houck, The Regulation of Toxic Pollutants, supra note 53 (water toxics).

294. U.S. EPA. INTEGRATED RISK ASSESSMENTS FOR DIOXINS AND FURANS FOR CHLORINE BLEACH IN PULP AND PAPER MILLS 1 (1990). For an update on the dioxin wars, see Dioxin Litigants May File a Second Case Under New Data Guidelines, INSIDE EPA, Jan. 11, 2002, at 14.

Industry plaintiffs are considering dropping a portion of their current lawsuit challenging EPA's classification of dioxin as a known human carcinogen in its upcoming dioxin risk assessment of the chemical byproduct, to instead file an additional lawsuit challenging the study under a newly released federal data quality guideline, sources say.

Id.

295. See NATIONAL WATER COMM'N, WATER POLICIES FOR THE FUTURE 105-06 (1973) (water pollution standards should be determined by local authorities).

297. See infra section II.A.2.

298. See ROGERS, supra note 3; see also TMDL PROGRAM, supra note 1, at 131-34.

299. ROGERS, supra note 3; see also TMDL PROGRAM, supra note 1, at 131-34.

300. See S. REP. NO. 92-14 (1972), reprinted in 1972 U.S.C.C.A.N. 3671. See also Weyerhauser Co. v. Costle, 590 F.2d 1011, 9 ELR 20284 (D.C. Cir. 1978).

301. William H. Rogers Jr., Industrial Water Pollution and the Refuse Act. A Second Chance for Water Quality, 119 U. PA. L. REV. 761, 803 (1971).

302. Id. at 764-65.

303. Id. at 803.

304. 33 U.S.C. § 1288, ELR STAT. FWPCA § 208.

305. See JACKSON BATTLE & MAXINE LIPELES, WATER POLLUTION 538-40 (3d ed. 1998).

306. For a discussion of CWA § 208 and its underperformance, see id.; see also ROGERS, ENVIRONMENTAL LAW: AIR AND WATER, supra note 285, at 319-30; ANDERSON ET AL., ENVIRONMENTAL PROTECTION LAW AND POLICY 384-86 (1990).

307. 33 U.S.C. § 1329, ELR STAT. FWPCA § 319.

308. See TMDL PROGRAM, supra note 1, at 30, 31.

309. Under § 319, EPA does not review nonpoint plans but, rather, a state program to develop and implement them; EPA program disapproval, further, leads only to the possible loss of future grant funding. 33 U.S.C. § 1329(h), ELR STAT. FWPCA § 319(h). Senate leaders, in passing the measure, expressed their concern for lack of regulatory standards.

310. See David Zaring, Agriculture, Nonpoint Source Pollution and Regulation, Catch the Clean Water Act's Bleak Present and Future, 20 HARV. ENVTL. L. REV. 515 (1996) (concluding: "unfortunately, Section 319 has failed to reduce nonpoint source pollution. Its failings can be characterized as not enough carrot, not enough stick, and too much of the planning syndrome that had characterized Section 208"): George Gould, Agriculture, Nonpoint Source Pollution, and Federal Law, 23 U.C. DAVIS L. REV. 461 (1990) (ditto).

311. See supra section II.A.1.

312. See supra section II.B.6.

313. See supra section II.A.4.

314. See ENVIRONMENTAL LAW INST., ENFORCEABLE STATE MECHANISMS FOR THE CONTROL OF NONPOINT SOURCE WATER, Appendix: State "No More Stringent" Laws (1997), available at http://www.eli.org.

315. Id. at 1 (Executive Summary) ("Some states . . . have adopted explicit statutory or regulatory exemptions for agriculture or forestry activities.").

316. See States Back Revenues Needed to Implement TMDL Strategy, EPA Told, Daily Env't Rep. (BNA), May 12, 1997, at A-8 (quoting Hawaii official: "In a period of declining state budgets . . . we do not expect to be able to obtain sufficient funding to establish scientifically-defensible numeric targets for polluted runoff control").

317. See Bruninga, Water Quality Standards: Agency TMDL Proposal Costly, supra note 79 (comments of Sen. Robert Smith of New Hampshire); Notes on Hearings Before the Subcommittee on Water Resources and the Environment of the Committee on Transportation and Infrastructure, U.S. House of Representatives, Nov. 15, 2001 (notes on file with author) (comments of Rep. Gene Taylor (R-Miss.)) (with three-fifths of the states in deficit spending. EPA should not expect additional state effort without funding support). State environmental agencies are, in turn, taking the largest budget cuts in a decade. See Greenwire (Feb. 15, 2002), at http://www.greenwire.com ("The economic downturn that blew in over the past year has chilled the budgets of many state environmental agencies, forcing them to enact cuts not seen in a decade, and the outlook for the future is no brighter, according to a new report by the Environmental Council of the States."). Id. at 1. See also Press Release, Michigan Environmental Council, Engler Budget Cuts Fall Hard on MDEQ and MDNR (Nov. 7, 2001), available at http://www.mecprotects.org/pr11_06_01.html (citing a 24% cut in budget of the Michigan Department of Environmental Quality).

318. See David Broder, Pretty Pictures Can't Hide the Red Ink, TIMES PICAYUNE, Feb. 10, 2002, at A-7. The administration has proposed significant budget cuts for EPA in FY 2003, including CWAprograms. See also Press Release, U.S. EPA, Administration Proposes $ 280 Million Cut in EPA's FY 03 Budget (Feb. 5, 2002), at 1.

319. See TMDL PROGRAM, supra note 1, at 31-33, 145-46: see also NRC REPORT, supra note 77.

320. See Howard Latin, Ideal Versus Real Regulatory Efficiency: Implementation of Uniform Standards and "Fine Tuning" Regulatory Reforms, 37 STAN. L. REV. 1267, 1270-71 (1985) ("The academic literature on regulatory reform reflects an excessive preoccupation with theoretical efficiency, while it places inadequate emphasis on actual decisionmaking costs and implementation constraints."); Daniel H. Cole & Peter Z. Grossman, When Is Command and Control Efficient? Institutes, Technology, and the Comparative Efficiency of Alternative Regulatory Regimes for Environmental Protection, 5 WIS. L. REV. 887, 935 (1999) ("specifically, standard economic accounts of the comparative efficiency of alternative regulatory regimes are insensitive to historical, institutional, and technological contexts. Most importantly, they tend to assume 'perfect (and, incidentally, costless) monitoring,' or they assume that monitoring costs are the same regardless of the control regime that is chosen").

321. Richard W. Steverman & Jeff Garth, Web of Safeguards Failed as Energy Giant Fell, TIMES PICAYUNE, Jan. 20, 2002, at A-23 (quoting Jim Chanos, President of Kynikos Associates).

322. Id.

323. Testimony of G. Tracy Mehan III, Assistant Administrator for Water, U.S. EPA. Before the Subcommittee on Water Resources and the Environment of the Committee on Transportation and Infrastructure, U.S. House of Representatives (Nov. 15, 2001), available at http://www.house.gov/transportation/water/11-15-01/mehan.html.

324. Presentation of Charles Sutfin. U.S. EPA Office of Water, to ASIWPCA Meeting, Seattle, Washington, Jan. 14, 2002; see also testimony of G. Tracey Mehan. III, supra note 323 (describing EPA's efforts to provide greater flexibility).

325. See EPA Preparing to Cede More Impaired Water Responsibility to States, INSIDE EPA, June 29, 2001, at 1.

326. U.S. EPA Briefing Notes (on file with author), overhead 12, "Key Issues for Rulemaking" Including: 303(d) Listing; Implementation Plans: Reasonable Answer; Timeframes; Stakeholder Involvement; EPA TMDL Rule; and EPA Permitting Rule.

327. See TMDL PROGRAM, supra note 1, at 61.

328. See infra notes 424-27 and accompanying text; see also TMDL PROGRAM, supra note 1, at 95.

330. See TMDL PROGRAM, supra note 1, at 61-62. The statute reads in pertinent part: "Each state shall identify these sections within its boundaries for which the effluent limitations required by Section 1311(b)(1)(A) and Section 1311(b)(1)(B) of this title are not stringent enough to implement any water quality standard applicable to such waters." 33 U.S.C. § 1313(d)(1)(A), ELR STAT. FWPCA § 303(d)(1)(A).

331. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 14 ELR 20507 (1984) (announcing principle of deference to Agency interpretations of statutes).

332. Presentation of Sutfin, supra note 324 (indicating his personal agreement with the position not to list or require TMDLs for waters polluted by nonpoint sources only).

333. Id. This compromise was not stated, but is inferable from the previous comment cited.

334. See Chevron, 467 U.S. at 837, 14 ELR at 20507. The Court approved a turn-about by EPA on pollution abatement requirements in nonattainment areas under the CAA.

335. See supra section II.B.3.

336. Id.; see also Presentation of Sutfin, supra note 324.

337. See U.S. EPA, GUIDANCE FOR WATER QUALITY-BASED DECISIONS: THE TMDL PROCESS (1991) (available from the ELR Document Service, ELR Order No. AD-3550); see also U.S. EPA, NATIONAL CLARIFYING GUIDANCE FOR 1998 STATE AND TERRITORY SECTION 303(d) LISTING DECISIONS (1997) (available from the ELR Document Service, ELR Order No. AD-3504).

338. Presentation of Sutfin, supra note 324.

339. This issue is particularly relevant given EPA's recognition that the science is weak on the effectiveness of particular best management practice strategies. Id.

340. Id.

341. See the reduction in the impaired waters list for Wyoming, from over 400 waters to 61, with 315 now identified as requiring "further monitoring." TMDL PROGRAM, supra note 1, at 138.

342. See Presentation of Sutfin, supra note 324.

343. See supra section II.A.2.

344. See EPA Backs Innovative Virginia Program to Avoid Setting TMDLs, INSIDE EPA, Aug. 10, 2001.

345. Id.

346. Id.

347. See supra note 2.

348. 40 C.F.R. § 130.2(c).

349. Id. § 130.2(h). Hence the equation: TMDL = WLA (waste load allocation) + LA (load allocation) + BL (baseline condition) + MOS (margin of safety) + FG (future growth). See Adler, supra note 196, at n. 113.

350. See U.S. EPA, GUIDANCE FOR WATER QUALITY-BASED DECISIONS, supra note 337.

351. See EPA Preparing to Cede, supra note 325 ("Other changes include lifting a requirement that TMDL discharge limits be set for certain point and nonpoint source dischargers, and instead would essentially alleviate the individual accountability of particular dischargers for controlling dischargers.").

352. See 33 U.S.C. § 1361(a), ELR STAT. FWPCA § 501(a) (EPA authority to adopt rules "as broad as necessary" to carry out its CWA responsibilities). This authority has been interpreted broadly in the past. See American Paper Inst. v. EPA, 890 F.2d 869, 875-878, 20 ELR 20482, 20485-86 (7th Cir. 1989); see also American Petroleum Inst. v. Knecht, 456 F. Supp. 889, 8 ELR 20853, aff'd, 609 F.2d 1306, 10 ELR 20083 (9th Cir. 1979) (upholding federal requirements under the Coastal Zone Management Act. "Under our so called federal system . . . the federal bureaucracy is legally permitted to execute the congressional management with a high degree of befuddlement as long as it acts no more befuddled than the Congress must reasonably have anticipated." 456 F. Supp. at 931, 8 ELR at 20873).

353. CERCLA, a remedial statute seeking to spread the costs of cleanup beyond the general public, imposes joint and several liability for all (but minor) contributors. See Starfield, supra note 292, at 10226.

354. The CAA SIP process allows the state the freedom to allocate cleanup responsibilities among sources without regard to their contributions. See Union Elec. Co. v. EPA, 427 U.S. 246, 6 ELR 20570 (1976).

355. EPA has released model allocation options that are "efficient" and "equitable." E-mail correspondence with Merritt Frey, Clean Water Action Network, Boise, Idaho (Jan. 16, 2002).

356. See TMDL PROGRAM, supra note 1, at 84 (citing "some concerns that relevance on § 303(d) could lead to judicial enforcement of TMDL plans in unexpected and unintended ways").

357. See EPA Advances Set of Proposals for Upcoming Impaired Waters Rule, INSIDE EPA, Aug. 17, 2001, at 11 (EPA is "looking at throwing out a requirement in the July 2000 rule that forced states to submit implementation plans with their impaired water lists").

358. See supra section II.B.1.

359. See Testimony of G. Tracy Mehan III, supra note 323.

360. "The state shall incorporate [TMDLs] into its current plan under subsection (e) of this section." 33 U.S.C. § 1313(d)(2), ELR STAT. FWPCA § 303(d)(2).

361. While EPA had required "reasonable assurance" as early as 1991, specific implementation plans were not required by the Agency until 1997. See TMDL PROGRAM, supra note 1, at 80-81.

362. See supra note 352.

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

365. See EPA Preparing to Cede, supra note 325 ("The shift would move the agencies' oversight of state implementation plans from Section 303(d) of the Clean Water Act to Section 303(e) essentially throwing to state officials the responsibility for reviewing implementation plans, rather than EPA.").

366. 33 U.S.C. § 1313(e), ELR STAT. FWPCA § 303(e).

367. No one familiar with federal "process" review under, for example, the federal aid highway program or the coastal zone management program, can have much faith in its ability to provide environmentally protective results. See Oliver A. Houck & Michael Rolland, Federalism in Wetlands Regulations, 54 MD. L. REV. 1242, 1249-99 (1995) (describing spotty-at-best effectiveness of NOAA process review of state coastal management programs).

368. 40 C.F.R. § 130.6.

369. Id. § 130.6(1).

370. See Louisiana TMDL Program, at http://www.deq.state.la.us/technology/tmdl/index.htm (last visited Feb. 18, 2002). The state of Virginia has recently proposed to repeal all of its § 303(e) plans as obsolete. Press Release, Virginia Department of Environmental Quality, Total Maximum Daily Loads—Proposed TMDL Planning Regulation (Jan. 18, 2001).

371. See NRC REPORT, supra note 77, at 101 (EPA should set "broad guidelines" and no more); Stevenson & Shabman, supra note 102 (ditto; then again, he drafted the report). See supra notes 101-07 and accompanying text.

372. See supra note 352.

373. See supra section II.B.4.

374. Id.

375. See Houck & Rolland, supra note 367, at 1289-99 (describing EPA reluctance to sanction delegated NPDES programs and similar NOAA reluctance under the CZMA); Steinzor, supra note 51 (describing ineffectiveness of federal sanctions generally).

376. See TMDL PROGRAM, supra note 1, at 55-56, 75-77, 168.

377. Id. at 84 (quoting concerns of the nonpoint source industries in the FACA "that reliance on § 303(d) could lead to judicial enforcement of TMDL implementation plans in unexpected or unintended way" (citing the draft FACA report)), id. at 84 n.93.

378. See EPA Advances Set of Proposals, supra note 357.

379. Id.

380. See supra notes 257-58 and accompanying text.

381. See Testimony of G. Tracy Mehan III, supra note 323.

382. See Mehan Says Toxics Inventory Holds Lessons for TMDL Program [20 Current Developments] Env't Rep. (BNA) 2400 (Dec. 14, 2001). The TRI is a requirement of the Emergency Planning and Community Right-To-Know Act, 42 U.S.C. § 11001, ELR STAT. EPCRA § 301, producing an annual record of industrial toxic dischargers nationwide. Id. § 11023, ELR STAT. EPCRA § 313.

383. See U.S. EPA, EPA NONPOINT SOURCE POLLUTION FOCUS GROUPS FINAL. REPORT, supra note 225 and accompanying text.

384. Louisiana industries released, under permit, over 150 million pounds of toxins in 1999; the national total exceeded 7.5 billion pounds. U.S. EPA, TRI Onsite and Offsite Releases by State, available at http://www.epa.gov/triinter/tridata/tri99/press/state_all.pdf (last visited Feb. 22, 2202).

385. 42 U.S.C. § 4321, ELR STAT. NEPA § 2.

386. 33 U.S.C. § 1365, ELR STAT. FWPCA § 505 (citizen suit provisions).

387. See Houck, The Regulation of Toxic Pollutants, supra note 53, at 10537 n.43 (recording the role of one citizen organization, the Natural Resources Defense Council, in the implementation of the Act. For a recent example of the action-forcing nature of a citizen lawsuit, see Paper Mill Agrees to Pay $ 30 Million to Upgrade Plant, $ 2 Million for Restoration, 32 [Current Developments] Env't Rep. (BNA) 1685 (Aug. 24, 2001) (describing proposed settlement agreement in Pennsylvania in Public Interest Research Group of N.J. v. P.H. Glatfelter Co., No. 1 CV 99-09040 (M.D. Pa. Aug. 20, 2001); the pulp and paper plant at issue will, inter alia, eliminate dioxin-producing chlorine from its bleaching process, id., a result that EPA has yet to achieve through regulation).

388. See Steve Novik & Bill Westerfield, Whose SIP Is It Anyway? State-Federal Conflict in Clean Air Act Enforcement, 18 WM. & MARY J. ENVTL. L. 215, 220 (1994) (quoting a Wisconsin air management official: "From direct personal experience I can tell you it is extremely difficult for a state to adopt and implement control measures which have not been specifically required by U.S. EPA.").

389. See THEODORE L. GARRETT & SONYA WINNER, CLEAN AIR DESKBOOK 16 (ELI 1992) (describing use of and court-imposed limits on EPA "conditional approvals" of state SIPs); in addition, partial and conditional approaches have subsequently been authorized by Congress, 42 U.S.C. § 7410(k)(3), (4), ELR STAT. CAA § 110(k)(3), (4).

390. See Steinzor, supra note 51. See also FREDERICK R. ANDERSON ET AL., ENVIRONMENTAL PROTECTION LAW AND POLICY 451-52 (3d ed. 1999) (federal implementation plans have been few, far between, and controversial).

391. Market-based approaches are not included in this analysis because they presuppose a target and an obligation to abate. See William F. Pedersen Jr., Why the Clean Air Act Works Badly, 129 PA. L. REV. 1059 (1981). Pollution trading in the context of the TMDL program is neither standard-setting nor obligation-imposing; rather, it is a potentially useful means of allocating abatement requirements.

392. 42 U.S.C. § 7408, ELR STAT. CAA § 108 (designation of criteria pollutants); § 7409, ELR STAT. CAA § 109 (establishment of national ambient air quality standards (NAAQS)). NAAQS have been established for, inter alia, nitrogen oxide and ozone.

393. Id. § 7511, ELR STAT. CAA § 181 (imposing more onerous SIP requirements in nonattainment areas according to severity of pollution); id. § 7511(a), ELR STAT. CAA § 181(a) (mandatory enhanced vehicle inspection and maintenance programs in more severely polluted nonattainment areas).

394. Id. § 7521(m), ELR STAT. CAA § 202(m) (emission control diagnostics).

395. Id. § 7521(b)-(k), ELR STAT. CAA § 202(b)-(k) (emissions standards).

396. Id. § 7545, ELR STAT. CAA § 211 (regulation of fuels).

397. See PERCIVAL ET AL., supra note 290, at 849-55 (discussing low-emission and zero-emission vehicle regulation).

398. See Amoco Oil Co. v. EPA, 501 F.2d 722, 4 ELR 20397 (D.C. Cir. 1972) (upholding regulations barring leaded gasoline in automobiles with catalytic converters, and requiring widespread marketing of unleaded gasoline); Ethyl Corp. v. EPA, 541 F.2d 1, 6 ELR 20267 (D.C. Cir. 1976) (en banc), cert. denied, 426 U.S. 941 (1976) (upholding phase-down in lead content in gasoline); 42 U.S.C. § 7545(a), ELR STAT. CAA § 211(a) (prohibiting the sale of leaded gasoline).

399. See Adler, supra note 196.

400. 42 U.S.C. §§ 7408, 7409, ELR STAT. CAA §§ 108, 109.

401. Id. § 7410, ELR STAT. CAA § 110; see also 40 C.F.R. § 51.112 (spelling out SIP requirements more fully).

402. 42 U.S.C. § 7410(a)(1), ELR STAT. CAA § 110(a)(1) (the submissions are due within three years of the promulgation of a NAAQS).

403. EPA enforcement options include a federal SIP, 42 U.S.C. § 7410(c), ELR STAT. CAA § 110(c), and the withholding of federal highway monies and other federal assistance. Id. § 7509, ELR STAT. CAA § 109.

404. Id. § 7604, ELR STAT. CAA § 304. Approved SIPs, further, become enforceable by governments and citizens as a matter of federal law. See Adler, supra note 196, at 235 n.1902 and cases cited therein.

405. See U.S. Steel Corp. v. EPA, 444 U.S. 1035, 1078, 10 ELR 20081, 20082 (1980) (Rehnquist, J. dissenting) (the provisions of the CAA "virtually swim before one's eyes"). See also Adler, supra note 196, at 207-09; Pedersen, supra note 391; Schoenbrod, supra note 196; Arnold Rietze Jr., A Century of Air Pollution Control Law: What's Worked; What's Failed; What Might Work, 21 ENVTL. L. 1549 (1991); JOHN-MARK STENSVAAG & CRAIG N. OREN, CLEAN AIR ACT: LAW AND PRACTICE (1994).

406. See ROGERS, supra note 3, at 161-68 ("consistently with their dignity as national health standards, the NAAQS have continued to be a tumultuous subject of public policy since their adoption in the early 1970s." Id. at 162). For a continuation of the controversy, see Whitman v. American Trucking Ass'n 121 S. Ct. 903, 31 ELR 20512 (2001) (ozone standards).

407. See ANDERSON ET AL., supra note 390, at 4545; ROGERS, supra note 3, at 266-71. See also Air Pollution Control Dist. of Jefferson County v. EPA, 739 F.2d 1071, 14 ELR 20573 (6th Cir. 1984) (conflicts of modeling long-range transport).

408. See ROGERS, supra note 3, at 269 ("the empirical bottom line on source monitoring is that it is a surprising no show; continuous monitoring is largely a fiction. It has been estimated that fewer than five percent of major air pollution sources have had their allowable emissions verified by stack testing").

409. The CAA provides a loophole of "unclassifiable" attainment for criteria pollutants based on insufficient information. 42 U.S.C. § 7407(d)(1)(A)(iii), ELR STAT. CAA § 107(d)(1)(A)(iii). This section has "apparently been abused to avoid more stringent source controls and has provided an incentive to avoid collecting the data needed to determine attainment." See Adler, supra note 196, at 233.

410. See Alder, supra note 196, at 241-42 and sources cited therein.

411. Id. at 294.

412. See ROGERS, supra note 3, at 202-10 ("a SIP with pages turning, some on the shelf and some not yet arrived, hardly makes for ascertainable legal obligations"). Id. at 207.

413. See LESTER B. LAVE & GIBERTS OMENN, CLEANING THE AIR: REFINING THE CLEAN AIR ACT 42-43 (1981) (states discontinued monitoring in order to avoid nonattainment designation): id. at 46 (attainment status depends on a monitoring system that is characterized by siting to avoid pollution sources, lax controls, and other manipulation to avoid nonattainment).

414. See Adler, supra note 196, at 282 and sources cited therein.

415. Id. at 233-34.

416. See EPA Audits of State I/M Programs Reveals Series Enforcement Problems, 16 Env't Rep. (BNA) 325 (1985) (noncompliance reaches 60% of inspection and maintenance facilities).

417. See supra notes 195-99 and accompanying text. See also Craig N. Oren, Prevention of Significant Deterioration: Control Compelling Versus Site Shifting, 74 IOWA L. REV. 1, 5 (1988) (calling CAA prevention of significant deterioration trading an "elaborate regulatory hocus-pocus.") An enormous amount of resources have been spent on litigation simply attempting to keep the SIP process honest. See ROGERS, supra note 3, at 204-10 and case cited therein.

418. Adler, supra note 196, at 250-75.

419. Id.

420. Id. at 294.

421. See id. and sources cited therein. See also supra note 415.

422. See 42 U.S.C. § 7501(3), ELR STAT. CAA § 171(3) (lowest achievable emission rate); id. § 7502(c)(1), ELR STAT. CAA § 172(c)(1) (reasonably available control technology); id. § 7475(a)(4), ELR STAT. CAA § 165(a)(4) (best available control technology).

423. 42 U.S.C. §§ 7661 et seq., ELR STAT. §§ 501 et seq.

424. TMDL PROGRAM, supra note 1, at 85-86 (identifying agriculture as the number one cause of impairment of the nation's rivers and lakes (citing the President's Clean Water Action Plan)), id. at 85 n.98.

425. Id.

426. Id.

427. See DONALD A. GOOLSBY ET AL., FLUX AND SOURCES OF NUTRIENTS IN THE MISSISSIPPI-ATCHAFALAYA RIVER BASIN 14 (1999) (attributing 90% of nitrogen and phosphorous loadings to nonpoint sources upstream); National Oceanic & Atmospheric Administration, Hypoxia in the Gulf of Mexico, at http://www.noaa.gov/products/pubs_hypox.html (principal Gulf hypoxia source areas Illinois, Indiana, Iowa, Minnesota, and Ohio).

428. NATIONAL AGRICULTURAL STATISTICS SERV., U.S. DEPARTMENT OF AGRICULTURE, 1997 CENSUS OF AGRICULTURE, available at http://www.nass.usda.gov/census/ [hereinafter CENSUS], cited in J.B. Ruhl, Farms, Their Environmental Harms, and Environmental Law, 27 ECOLOGY L.Q. 263 (2000). The agricultural data that follow are taken largely from this Article.

429. See CENSUS, supra note 428, at 8, fig. 5.

430. Id. at 10, tbl. 1.

431. Id. at 23, tbl. 15.

432. Id. at 22, tbl. 14 (gasoline expenditures over $ 6 billion).

433. Id. at 100, tbl. 49 (electricity at our $ 2.75 billion).

434. Id.

435. See Changing Rural America, in VICKI MARKS. AMBER WAVES OF GRAIN 30 (Defenders of Wildlife 2000) (quoting a former Farm Bureau representative as having written that "in the past 50 years, farm numbers have dropped from 7 million to less than 2 million").

436. See CENSUS, supra note 428, at 6, fig. 2.

437. Elizabeth Becker, Airing of Farm-Subsidy Details Cultivates Envy: Published Secrets Show Reliance on Government. TIMES PICAYUNE, Dec. 28, 2001, at 11-13. A Republican state Senate candidate is quoted as saying: "You can see from the payment lists how the rich farmers can afford to buy up all the small ones like cannibals, all subsidized by the government." Id. (quoting candidate Rod Thorson). See also George Will, Demos Head Into the Wild Blue Yonder, TIMES PICAYUNE, Jan. 14, 2002, at B7 (two-thirds of farm subsidies go to 10 percent of the farming community, "most of whom carn more than $ 250,000 annually"—a phenomenon he then attributes to the Democratic Party).

438. There are 65 agricultural colleges and universities in the United States. Agriculture Colleges and Universities, Agriculture Colleges and Universities Index, at http://www.oneglobe.com/agriculture/agcolleg.html (last visited Feb. 19, 2001).

439. See MARKS, supra note 435 (detailing dominant insurance business of the Farm Bureau, and its political action, lobbying, and litigation initiatives).

440. Professor Ruhl characterizes the treatment of the agriculture industry as a series of "safe harbors" from environmental law and the CWA's treatment of nonpoint source pollution as a "classic example of passive nonregulation." Ruhl, Farms, supra note 428, at 293, 298. He is of course not the only one to note the anomaly. See John Davidson, Conservation Agriculture: An Old New Idea, 9 NAT. RESOURCES & ENV'T 20 (1995); C. Ford Runga, Environmental Protection: From Farm to Market, in THINKING ECOLOGICALLY: THE NEXT GENERATION OF ENVIRONMENTAL POLICY (Marion R. Chertow & David C. Esty eds., 1997); Tim Chen, Get Green or Get Out: Decoupling Environmental From Economic Objectives in Agricultural Regulation, 48 OKLA. L. REV. 333 (1995). Indeed, it is difficult to find any literature on agriculture and the environment that does not (1) note the anomaly and (2) conclude that the time has long come to end it.

441. For the diversity and complexity of industrial sources regulation under the NPDES program, see 40 C.F.R. §§ 401.10 et seq. (industrial category and subcategory technology-based emissions limitations).

442. See Association of Pac. Fisheries v. EPA, 615 F.2d 794, 10 ELR 20336 (9th Cir. 1990) (seafood processing standards will lead to closure of more than 10% of industry).

443. Ruhl, Farms, supra note 428, at 333.

444. Id.

445. Id. at 335-37.

446. See TMDL PROGRAM, supra note 1, at 92.

447. Id. at 89.

448. U.S. EPA, National Emission Standards for Hazardous Air Pollutants for Source Category: Pulp and Paper Production; Effluent Limitations Guidelines, Pretreatment Standards, and New Source Performance Standards, 63 Fed. Reg. 18504-751 (Apr. 15, 1998). The multimedia rule produced a 60% reduction in toxic air pollutants and a 96% reduction of dioxin discharges to waterways, leading to an "expedited cleanup" of 73 rivers and streams nationally. Press Release, U.S. EPA, EPA Eliminates Dioxin, Reduces Air and Water Pollutants From Nation's Pulp and Paper Mills (Nov. 14, 1997).

449. See TMDL PROGRAM, supra note 1, at 143.

450. See id. (describing a 40% drop within one year in nutrient loadings from Florida sugar farms through the requirement of best practices in the application of fertilizers).

451. See EPA Data Appears to Undermine CAFO Rule, INSIDE EPA. Nov. 26, 2001, at 1; see also TMDL PROGRAM, supra note 1, at 88.

452. See Industry Pushes Regulatory Alternatives in EPA Feedlot Rule, INSIDE EPA, Dec. 21, 2001, at 11.

453. See TMDL PROGRAM, supra note 1, at 87-99. See also Construction Coalition Forms to Oppose Upcoming Effluent Rule, INSIDE EPA, Jan. 4, 2002, at 5.

454. See Community Ass'n for Restoration of the Env't v. Henry Bosma Dairy, 52 ERC 1167 (E.D. Wash. 2001).

455. See Miccosukee Tribe v. Southern Fla. Water Mgmt. Dist., 49 ERC 2067 (11th Cir. 2002).

456. See Headwaters, Inc. v. Talent Irrigation Dist., 243 F.3d 526, 31 ELR 20535 (9th Cir. 2001).

457. See Northwest Envtl. Advocates v. EPA, No. Co11297-BZ (N.D. Cal. Apr. 2, 2001).

458. See Press Release, Cynthia Elkins & Alan Levine, Coast Action Group, Environmentalists Sue to Halt Federal Clean Water Violations by Pacific Lumber Company in Northern California (July 29, 2001).

459. See supra note 160 and accompanying text.

460. See B. Commoner, Failure of the Environmental Effort, 18 ELR 10195, 10197, tbl. III (June 1988) (showing reductions of 70 to 92% for lead, DDT, PCBs mercury, strantium 90, and phosphates).

461. See MARK SQUILLACE, ENVIRONMENTAL LAW: AIR POLLUTION 423 (1992) (ambient air levels of lead dropped sharply); Richard B. Alexander & Richard A. Smith, Trends in Lead Concentrations in Major Rivers and Their Relation to Historic Changes in Gasoline-Lead Consumption, 25 WATER RESOURCES BULL. 1275 (1989) (ambient water quality levels of lead dropped sharply as well).

462. See U.S. Department of the Interior, Proposed Rule to Remove the Bald Eagle in the Lower 48 States From the List of Endangered and Threatened Wildlife, 64 Fed. Reg. 36454 (July 6, 1999); 50 C.F.R. pt. 70 (removal of Brown Pelican from list); U.S. Department of the Interior, Final Rule to Remove the American Peregrine Falcon From the Federal List of Endangered and Threatened Wildlife, 64 Fed. Reg. 46542 (Aug. 25, 1999).

463. See Moving Fast to Protect Ozone Layer, N.Y. TIMES, May 15, 1991 (rapid progress in reducing CFCs in the electronic industry, despite industry claims that alternatives not available, and quoting an Apple Computer. Inc., chemist as stating: "If the federal government set a goal that there would be no fossil fuels sold by 1999, I bet you would see electric cars by then"). See also Houck, The Regulation of Toxic Pollutants, supra note 53, at 10554 (alternative bleaching agents replacing dioxin-producing chlorine in the pulp and paper industry). As for the economic impact of environmental requirements including these bans and reductions, between 1970 and 1990 the U.S. economy experienced a growth in real terms of 72% despite industry prophesies of lost jobs and economic doom. COUNCIL ON ECONOMIC QUALITY, 21ST ANNUAL REPORT 6-9 (1990).

464. Technically, the pollutants are phosphorous, nitrogen, and other components of fertilizers, but for the sake of convenience they are aggregated here as "fertilizers."

465. Commercial fertilizers are the leading cause of nitrate loadings in surface and groundwaters, see NATURAL RESOURCE CONSERVATION SERV., USDA, GEOGRAPHY OF HOPE 48 (1996), cited in Ruhl, supra note 428, at note 34. Fertilizer loadings increased by up to 10 times during the 20th century. Id. at 41-45.

466. NATIONAL SCIENCE AND TECHNOLOGY COMMITTEE ON ENVIRONMENT AND NATURAL RESOURCES, INTEGRATED ASSESSMENT OF HYPOXIA IN THE NORTHERN GULF OF MEXICO (2000), at http://www.nosnoaa.gov/products/pubs_hypox.html (last visited Feb. 22, 2002).

467. See Thomas E. Jordan et al., Effects of Agriculture on Discharges of Nutrients From Coastal Plain Watersheds of Chesapeake Bay, 26 J. ENVTL QUALITY 836 (1997). The correlation is found in Europe as well. See EPA [European Protection Agency] Says Farm, Wastewater Practices at Fault in Eutrophication of North Seas, 24 Int'l Env't Rep. (BNA) 969 (Nov. 7, 2001).

468. OFFICE OF POLLUTION PREVENTION AND TOXICS, U.S. EPA, BACKGROUND REPORT ON FERTILIZER USE, CONTAMINANTS, AND REGULATIONS i (1999) [hereinafter U.S. EPA, BACKGROUND REPORT].

469. Id.

470. Id.

471. Id.

472. Id.

473. See Ruhl, Farms, supra note 428, at 285, citing Charles M. Cooper & William M. Lipe, Water Quality and Agriculture: Mississippi Experiences, 47 J. SOIL & WATER CONSERVATION 221 (1992).

474. See Ruhl, Farms, supra note 428, at 272 n.20 (45% of the land mass of the United States is in field crop agriculture; another 30% is in silviculture and livestock).

475. In the late 1980s farmers were spending $ 6.7 billion on fertilizer alone; by the late 1990s they were spending $ 9.6 billion, an increase of nearly 50%. See id. at 113.

476. See supra note 474 and accompanying text. See also Ruhl, Farms, supra note 428, at 282-85, 287-91 (extent of farm runoff).

477. See Less Fertilizer, TIMES PICAYUNE, Apr. 3, 1998, at C1 ("the use of less fertilizer at precisely the right time can cut costs up to 17% for farmers in developing countries and reduce damage to the environment, according to a study of Mexican wheat").

478. See supra notes 431-33 and accompanying text (costs of agricultural chemicals as compared to costs for gasoline and electricity).

479. See Garrett Hardin. The Tragedy of the Commons, 83 SCIENCE 1234 (1968).

480. See Groundwater: Draft Strategy Under TSCA Aims for Data on VOCs, Fertilizers, Septic System Additives, 16 [Current Developments]Env't Rep. (BNA) 1799 (Jan. 24, 1986).

481. 15 U.S.C. § 2607, ELR STAT. TSCA § 8. For a discussion of the (intended but little-used) TSCA mechanisms, see ROGERS, supra note 3, at 488-92.

482. 15 U.S.C. § 2603, ELR STAT. TSCA § 4.

483. Id. § 2605, ELR STAT. TSCA § 6.

484. See Groundwater: Draft Strategy Under TSCA Aims for Data on VOCs, supra note 480.

485. Telephone conversation of Andrew L. Adams with David Fagan, Office of Solid Waste, U.S. EPA (Jan. 30, 2002). Mr. Fagan was involved in the preparation of U.S. EPA BACKGROUND REPORT, supra note 468.

486. See U.S. EPA, BACKGROUND REPORT, supra note 468.

487. For an illustrative state program, see Pamela S. Clarke & Stacey M. Crank, The Pennsylvania Nutrient Management Act: Pennsylvania Helps to "Save the Bay" Through Nonpoint Source Pollution Management, 6 VILL. ENVTL. L.J. 319 (1995). For developments in the EU, see New French Water Law to Extend Polluter-Pays Principle to Agriculture, 24 Int'l Env't Rep. (BNA) 544 (July 4, 2001) (first-ever taxes on agriculture water use and runoff); Government Tells Commission It Will Hasten Efforts to Curb Nutrients to Meet EU Directive, 21 [Current Developments] Env't Rep. (BNA) 1234 (Dec. 9, 1998) (EU directive imposes "nitrate accounts" and "nitrate crops"); Andrew P. Manale, European Community Programs to Control Nitrate Emissions From Agriculture, 14 Int'l Env't Rep. (BNA) 345 (June 19, 1991).

488. See MARC RIBARDO & DANETTE WOO, U.S. DEPARTMENT OF AGRICULTURE, SUMMARY OF STATE WATER QUALITY LAWS AFFECTING AGRICULTURE 54 (undated) (identifying fertilizer taxes in California, Iowa, South Dakota, and Wisconsin).

489. See JOHN-MARK STENSVAAG, MATERIALS ON ENVIRONMENTAL LAW 547-95 (1999); see also Pedersen, supra note 197, at 10173:

The most potent signal alternative—imposing a tax on pollution—has thus far failed to be adopted widely. In addition, because the impact of a tax on pollution is hard to predict in advance, and because the government must also decide how to use the revenues collected, a tax raises policy and program design issues even more complicated than those that attend emissions trading.

Id. at 10173 n.4. These difficulties noted. states do impose permit fees based on the volume and toxicity of discharge, see LA. ADMIN. CODE tit. 33, §§ 1301-1313 (1988) (water program fee regulations), as do member countries of the EU. See Brennan & Johnson, Pollution Control by Effluent Charges: It Works in the Federal Republic of Germany, Why Not in the U.S., 24 NAT. RESOURCES J. 929 (1984). For two symposia pushing the envelope on environment and taxation generally, see Pollution Tax Forum: Colloquium, 12 PACE ENVTL. L. REV. 1 (1994); Vermont Law School, The Third Annual Global Conference on Environmental Taxation (2002) (brochure) (on file with author).

490. Treaty on European Union: (Maastricht Treaty), Feb. 7, 1992, 31 I.L.M. 247, available at http://www.hri.org/docs/Maastrecht92.

491. Id. art. 130r(2) ("Action by the Commission, relating to the environment shall be based on the principles that preventative action should as a priority be rectified at the source, and that the polluter should pay.").

492. See TMDL PROGRAM, supra note 1, at 168-69.

32 ELR 10385 | Environmental Law Reporter | copyright © 2002 | All rights reserved