Addressing water pollution from livestock grazing after O.N.D.A. v. Dombeck: legal strategies under the Clean Water Act.I. INTRODUCTION In 1998 the Ninth Circuit ruled in Oregon Natural Desert Association v. Dombeck (ONDA v. Dombeck)(1) that issuing livestock grazing permits does not trigger the section 401 certification requirement of the Clean Water Act.(2) The Oregon Natural Desert Association (ONDA) alleged that the United States Forest Service had violated section 401 by issuing a grazing permit without first obtaining certification from the State of Oregon that the grazing would not violate the state's water quality standards.(3) Finding that the case turned on interpretation of the term "discharge" in the Clean Water Act,(4) the court determined that a "discharge" is limited to releases from point sources.(5) Thus, because the court rejected the suggestion that livestock constituted a point source under the Act, and because pollution from livestock grazing generally consists of surface runoff--a nonpoint source(6) of pollution--section 401 of the Clean Water Act does not regulate water pollution from livestock grazing. Almost two years later, environmental organizations continue in the struggle to develop alternative legal strategies to address grazing-related water pollution. As one of the most environmentally devastating uses of western public lands, livestock grazing causes a litany of environmental harms to fragile arid-land ecosystems. These harms include: decreased water quality (from increased sediment in streams due to soft and bank erosion, increased water temperature as a result of reduced vegetation shade cover, and fecal deposition in streams), the reduction of water resources (by watershed degradation and physical damage to waterways and riparian areas), soil erosion and compaction (including stream bank erosion and the trampling of fragile microbiotic crusts--valuable soil stabilizers and rare desert nitrogen-fixers), the introduction and spread of invasive weed species, the elimination of native plant species, a decrease in the frequency of natural fire regimes, and the overall destruction of critical riparian fish and wildlife habitat.(7) All of this environmental harm begins simply with livestock trampling vegetation and soil, grazing on native plants, and depositing waste onto the land and into the water. Although the western range contributes only a small portion of the nation's livestock production,(8) livestock graze vast acreages of the western states. The Bureau of Land Management (BLM) and the Forest Service administer grazing permits for over 240 million acres of public lands.(9) The modern incarnation of western ranching has its origins in the nation's early policy of western land disposition, when the federal government encouraged settlers to tame and utilize the western frontier.(10) Statutes such as the Homestead Act of 1862(11) and the Desert Lands Act of 1877(12) necessarily led to a western philosophy that a settler had an absolute right to use the public lands in any manner desired. As early as 1890, the Supreme Court ruled that Congress, by its silence, acquiesced to western settlers' use of the public lands for grazing.(13) The result of this early history was the classic "commons" situation,(14) in which ranchers grazed as many livestock as they could on the public lands. When the Great Depression and the Dust Bowl occurred, Congress finally recognized that the pervasive policy of unrestricted use of the public lands had left arid western rangelands in a perilous condition and decided to take action.(15) In 1934 Congress enacted the Taylor Grazing Act (TGA),(16) stating in the preamble that the purpose of the Act was "[t]o stop injury to the public grazing lands by preventing overgrazing and soil deterioration, [and] to provide for their orderly use, improvement, and development."(17) The TGA introduced a preference permit system for grazers and within a few years led to a withdrawal of all public lands into grazing districts.(18) Ranchers with lands adjacent to the publicly grazed lands dominated implementation of the new permit system.(19) In 1976 Congress enacted the Federal Land Policy and Management Act (FLPMA),(20) supplying the BLM with its first specific and centralized authority to manage the public lands under its administration.(21) The last thirty-five years have seen an explosion of significant environmental laws that address a wide range of issues,(22) not the least of which is the degradation in the quality of the nation's water resources. The most significant environmental law addressing water quality problems--with respect to grazing or any other source of pollution--is the Clean Water Act. It is the nation's comprehensive federal law for the control and abatement of water pollution. The section 401 certification theory espoused by the plaintiffs in ONDA v. Dombeck is the most direct strategy to attempt to legally harness the environmental degradation caused by grazing and, afar ONDA's success with the theory at the district court level,(23) it looked like environmental groups had found an answer. Now-at least in the Ninth Circuit--that route is foreclosed. Unless and until the Ninth Circuit or the Supreme Court reverses the holding in ONDA v. Dombeck,(24) environmental plaintiffs must focus their energies on several new or revitalized theories in order to address grazing-related water pollution.(25) By foreclosing regulation of grazing-related water pollution in limited contexts, ONDA v. Dombeck is a significant environmental defeat because the sections of the Clean Water Act that address nonpoint source pollution have been, by nearly all accounts, an utter failure.(26) Instead of directly regulating nonpoint source pollution, the Clean Water Act only provides grants to states that wish to opt into programs established by the Act.(27) For example, section 208 ostensibly requires states to develop "areawide waste treatment management" plans.(28) However, if a state chooses not to develop a plan, then it simply does not receive the grant money available for development and operation.(29) In 1987 Congress added section 319, rifled "Nonpoint source management programs."(30) Like section 208, this program requires states to develop state assessment reports and management programs, and it is also driven largely by federal grants.(31) Although more states are trying to develop section 319 programs, these programs would only be enforceable at the state level, if at all.(32) The Environmental Protection Agency (EPA) has no authority under section 319 to prepare or implement a management program or to require the use of specific best management practices (BMPs).(33) Though other statutes--for example, the National Wild and Scenic Rivers Act(34)--have enjoyed some significant successes in addressing grazing-related water pollution in limited contexts,(35) the Clean Water Act nevertheless still holds the most promising and widely applicable strategies to address the problem. This Chapter focuses on two major strategies under the Clean Water Act: total maximum daily loads (TMDLs) and section 313's federal facilities pollution control. Part II discusses the basic TMDL strategy with respect to the TMDL process. This process includes: 1) forcing states to list waters under section 303(d),(36) 2) challenging the adequacy of the lists, 3) imposing a schedule for preparation of TMDLs where they have not yet been prepared as required under the Act, and 4) challenging the substance of TMDLs as they are completed. Part I also analyzes the South Steens TMDL in Oregon--which EPA developed for an area that consists entirely of nonpoint source pollution--as a case study under this legal strategy at the conclusion of the TMDL discussion. Finally, Part II concludes with a short discussion of EPA's 1999 proposed revisions to the TMDL program. Because TMDLs are a long-term solution and potentially lack sufficient enforcement mechanisms for effective and immediate legal control of grazing-related water pollution, Part III of this Chapter focuses on what is perhaps a stronger strategy: section 313 of the Clean Water Act.(37) Part II develops a two-part argument regarding section 313: 1) the provision imposes a duty that federal agencies must adhere to state water quality regulation of both point and nonpoint source pollution, including livestock grazing permitted on federal lands; and 2) that duty is enforceable for nonpoint sources under the APA. Finally, Part IV briefly discusses three more limited strategies. These are the Wild and Scenic Rivers Act,(38) the Coastal Zone Management Act.(39) and a theory suggesting that livestock congregation areas on public lands--such as watering troughs or stream access points--might be treated as point sources under the "concentrated animal feeding operation"(40) component of the Clean Water Act's definition of "point source." Part V concludes that the long-term, broad-scale effectiveness of the TMDL program should combine well with the more immediate, site-specific potential of section 313's requirements on federal agencies that administer the public lands. II. TOTAL MAXIMUM DAILY LOADS A. Statutory and Regulatory Requirements The goal of the Clean Water Act is to "restore and maintain the chemical, physical, and biological integrity of the Nation's waters."(41) The Act implements several programs to achieve this goal. For example, section 301 initiates the National Pollutant Discharge Elimination System (NPDES) program, which prohibits the discharge of a pollutant unless authorized by a permit.(42) This program deals with point source pollution from direct dischargers.(43) NPDES permit issuance is governed by section 402,(44) which allows states to take over permit issuance from EPA if their programs are at least equivalent to federal standards. Other Clean Water Act programs address nonpoint source pollution,(45) as well as the implementation of water quality standards for the nation's waters.(46) 1. Water Quality Standards The primary environmental harms to clean water as a result of livestock grazing are additions of heat from decreased riparian vegetation and additions of sediment and biological pollutants from surface runoff.(47) Section 303 of the Clean Water Act addresses water quality through the mechanism of water quality standards,(48) which each state must develop and revise for every water body in the state. A water quality standard "defines the water quality goals of a water body" by specifying the appropriate uses of that water and then setting standards to protect those uses.(49) Section 303 and the regulations adopted by EPA(50) indicate that water quality standards consist of 1) a designated use for the water body,(51) 2) criteria that list the maximum allowable concentrations of pollutants that a water body can sustain while still preserving its designated use,(52) and 3) an antidegradation provision.(53) Designated uses include public water supplies, propagation of fish and wildlife, recreation, agriculture, and industrial purposes.(54) A water body may--and often does--have more than one designated use for which water quality is protected.(55) The criteria that protect the designated uses may be expressed in a number of ways, including numeric (e.g., no more than five micrograms of a certain pollutant in the water), biological monitoring and assessment (a test of the effect of a discharge on aquatic organisms), and narrative requirements (a general statement prohibiting "the discharge of toxic pollutants in toxic amounts").(56) Finally, the antidegradation policy mandates that a state will not permit actions that will impair existing uses, regardless of whether a given existing use is also a designated use.(57) If the water is cleaner than necessary to support the designated uses, then the state must maintain that level of water quality.(58) Also, the Clean Water Act contains a nondegradation standard for high quality "outstanding National resource" waters.(59) For example, Oregon's antidegradation policy is intended only to meet the minimum federal requirements.(60) It provides that the policy of the state is to "protect, maintain and improve the quality of the waters of the state for public water supplies, for the propagation of wildlife, fish and aquatic life and for domestic, agricultural, industrial, municipal, recreational and other legitimate beneficial uses."(61) 2. The Total Maximum Daily Load Process: Listing Waters and Establishing TMDLs Implementation of water quality standards requires states to place those waters not meeting water quality standards--water quality limited segments"(62)--on the "303(d) list."(63) States must then calculate total maximum daily loads (TMDLs) for those waters not meeting water quality standards.(64) Development of a TMDL assessment consists of two steps. First, the state must determine the total daily input of pollutants, on a pollutant-specific basis, that a water body can sustain and still meet the applicable water quality standards.(65) This total maximum daily load consists of the total of individual wasteload allocations for point sources and load allocations for nonpoint sources, along with the natural background level of pollutants.(66) Second, the state must apportion the total maximum daily load to particular polluting sources within the TMDL area. In doing so, the state must allocate loadings to various point and nonpoint sources.(67) In a watershed or TMDL area that consists of both point and nonpoint source pollution, the state must determine how much of the TMDL is coming from nonpoint sources in order to determine how much of the total loadings remain to be allocated to point source polluters.(68) In addition, a TMDL assessment must take into account seasonal variance and include a margin of safety for any lack of knowledge regarding effluent limitations and water quality.(69) 3. Implementation and Enforcement: A Brief History of TMDL Enforcement In general, the TMDL program is to be implemented by the states with EPA oversight.(70) States are required to set water quality standards, identify water quality limited segments, develop TMDL assessments, and submit their data and proposals to EPA for approval at each step along the way.(71) However, if a state does not act, or fails to adequately address any EPA concerns upon agency disapproval, EPA must step in and perform the task(s).(72) The program has not unfolded quite so smoothly, and as a result many water bodies throughout the country--especially in rural areas dominated by agricultural nonpoint source pollution and livestock grazing--still lack TMDL assessments.(73) For its first two decades, a lack of state action and citizen suits prevented the TMDL process from making appreciable headway.(74) For example, it was only in the early 1990s that a federal court finally ordered EPA and the State of Alaska to issue the first TMDL limits for that state.(75) The district court noted that although Alaska's 1988 section 305(b)(76) report identified several hundred water bodies as impaired or threatened by water pollution, the state had only identified a single water quality limited segment prior to the litigation, and the state had not completed "even the first stage of the TMDL process."(77) Similarly, a district court noted in 1999 that, at best, the State of Virginia had completed a single, disputed TMDL on a half-mile long, unnamed tributary to a creek in the nearly twenty years since the original statutory deadline.(78) From the inception of the Clean Water Act, the TMDL program has been characterized by inaction on the part of the states, and it was largely neglected by EPA as the agency dealt with seemingly more pressing matters in promulgating technology standards for point sources.(79) A 1978 lawsuit required EPA to publish the list of pollutants that would activate TMDL planning.(80) After most states failed to submit lists, EPA argued that section 303(d) required the agency to do nothing more than approve or disapprove state submissions.(81) If states submitted nothing, EPA argued, then the agency was not required to do anything.(82) In 1984, however, the Seventh Circuit rejected that line of reasoning.(83) Scott v. City of Hammond(84) introduced the concept of a "constructive submission," finding that the "prolonged failure" of a state (Illinois in that case) to submit any TMDL assessments to EPA might result in a constructive submission to EPA of no TMDLs, thus triggering EPA's duty to act.(85) Shortly thereafter, the Northwest Environmental Defense Center filed the first lawsuit in the country seeking to force EPA to develop a TMDL in Oregon--where the state had "constructively submitted" no TMDL to EPA--thus compelling the federal agency to perform the task.(86) Despite the slow start of the TMDL program, environmental plaintiffs have "launched a tidal wave of lawsuits"(87) in recent years, particularly seeking implementation of TMDLs.(88) The Alaska TMDL cases "took the next step" by requiring EPA to establish TMDLs where Alaska had submitted absolutely nothing to EPA.(89) In March 2000 EPA listed seventeen consent decrees or court orders in fifteen states that require EPA to establish TMDL limits if the states do not establish them--fifteen of those decrees occurred in 1997 or after.(90) Ultimately, this surge of lawsuits, which began in the early 1990s, has resulted in several developments over the past few years,(91) including new section 303(d) guidance from EPA,(92) the creation of a Federal Advisory Committee Act (FACA)(93) committee that convened to address the TMDL program,(94) new state lists of impaired waters,(95) the Clinton administration's Clean Water Action Plan.(96) and recently proposed rules revising the requirements for listing and TMDL development.(97) B. Legal Strategies While the TMDL program is often thought to have little teeth with respect to challenging nonpoint source pollution, environmental plaintiffs may nevertheless invoke a few important strategies to make TMDLs effective in this arena. Many commentators suggest that TMDLs offer the next logical point from which to address nonpoint source pollution under the Clean Water Act.(98) even though TMDLs historically have often lacked the power necessary for effective enforcement.(99) Until now, most challenges to TMDLs have been procedural in nature;(100) however, litigation will increasingly mm to substantive issues and challenges to implementation.(101) In general, citizens should address the listing of water quality-limited waters and the adequacy of the states' lists, state implementation and the schedules for preparation, and the substance of the individual TMDLs.(102) 1. Forcing States to List Waters and Challenging the Adequacy of the Lists Following the basic scheme established in the statute, the fundamental step in TMDL litigation is to verify that a state has listed its waters pursuant to section 303(d) and then to ensure that the list is complete. Each state had 180 days following EPA's 1979 publication of the first set of pollutants to provide EPA with its first 303(d) list.(103) Because EPA has the nondiscretionary duty to review the adequacy of the list and the priority ranking,(104) citizens may enforce this duty under the citizen suit provision of the Act.(105) Of course, if EPA does not act or acts arbitrarily and capriciously--for example, by accepting an allegedly incomplete list-citizens may bring suit under the APA. As of January 2000 EPA has listed forty-one states and Puerto Rico as having their most recent (1998) final list approved by their respective EPA Regions, eight states with final lists "Partially Approved/Partially Disapproved," and one state with its final list submitted and under review.(106) 2. Enforcing Implementation Aside from continually reviewing the adequacy and accuracy of the lists, step one is largely moot because all states have submitted their most recent 303(d) lists, and EPA has accepted nearly all of those lists.(107) At this point, the crucial push for implementation takes over. Section 303(d) not only requires states to identify waters that are not meeting water quality standards, but it also mandates that "[e]ach State shall establish for the waters identified ... and in accordance with the priority ranking, the total maximum daily load, for those pollutants which the Administrator identifies ... as suitable for such calculation."(108) a. Failure to Submit TMDLs to EPA as a "Constructive Submission" of No TMDLs Scott v. City of Hammond(109) introduced the "constructive submission" concept to early TMDL litigation. Since that case, courts have held that a failure by a state to submit a TMDL assessment to EPA by the statutory deadline amounts to a constructive submission of no TMDL.(110) EPA has the nondiscretionary duty to approve or disapprove the constructively submitted (non-) TMDL.(111) A recent example of such a claim occurred in a lawsuit against EPA in New York.(112) The Natural Resources Defense Council (NRDC) claimed that EPA's failure to establish TMDLs for the State of New York constituted a failure to perform a nondiscretionary duty under the Clean Water Act, because the state was over fifteen years late in submitting anything to EPA.(113) NRDC asked that EPA establish TMDLs and disapprove the state's antidegradation policy.(114) The district court found the constructive submission concept to be apt in these circumstances, but the court also found--grudgingly--that the question of whether New York had created and submitted TMDLs was a triable issue of fact.(115) In reaching that conclusion, the court addressed the appropriate standard for determining whether a constructive submission had occurred. While EPA argued that a court must find a subjective decision on the part of a state not to submit anything, the court was persuaded instead by the intention of the statutory scheme. Specifically, the statutory intention is to promote "prompt establishment of TMDLs, which ... would be thwarted by a subjective test of constructive submission."(116) The court concluded by stating that "[m]ere objective failure to submit TMDLs for water-quality-limited segments is enough to trigger the non-discretionary [sic] duties of the EPA."(117) Despite an apparent about-face by the New York court in a continuation of the same case a few years later (i.e., Fox II),(118) this remains the accepted interpretation of the constructive submission doctrine. Referring to Fox II as "[t]he one divergent decision," a district court in Virginia recently stated that "[w]ith one exception, every court that has considered the issue has followed [Scott v. Hammond]."(119) The Virginia court cited several recent decisions in support of its proposition that the constructive submission doctrine introduced in Scott v. Hammond (and interpreted in NRDC v. Fox) is still viable and the correct interpretation of that concept.(120) Rejecting the assertion in Fox H that "a statute creates a nondiscretionary, judicially-enforceable duty only when it provides a date-certain deadline for agency action," the Virginia court pointed out that the numerous original deadlines contained in the Clean Water Act serve as "readily ascertainable" bases for determining statutorily imposed deadlines.(121) b. What Suffices as Sufficient Implementation of the TMDL Program? Fox II(122) also raises the important issue of what exactly will suffice as sufficient implementation of a state TMDL program. Though the court found EPA's claim that New York had actually submitted some TMDL assessments dubious at best, it nevertheless conceded that there was a triable issue of fact.(123) EPA provided at least two declarations indicating that New York had submitted, and EPA had approved, at least a few TMDL assessments.(124) The court did note, however, that only a single document produced by the government actually included volume per day limits, and this document appeared highly inconsistent with the statute and was vague in its terms.(125) At least a few cases have attempted to answer the question of what constitutes sufficient implementation of an acceptable TMDL program.(126) In Idaho Sportsman's Coalition v. Browner, a district court found EPA's approval of only thirty-six TMDLs by 1992 to be arbitrary and contrary to law.(127) Under court order, EPA two years later approved a list identifying 962 water quality-limited segments, and subsequent stages of the litigation focused on development of an acceptable schedule for implementation.(128) EPA and Idaho submitted a schedule that contemplated at least another twenty-five years until the TMDLs would be completed--and even this time frame was couched in terms of "expected times and targets."(129) Finding this unacceptable and contrary to the statute's clear mandate for expediency in the TMDL program,(130) the court remanded to EPA with instructions to establish a reasonable schedule, "suggesting" that "a completion time of approximately five years would be reasonable."(131) A district court in Georgia reached a similar decision shortly thereafter.(132) In Sierra Club v. Hankinson, the Sierra Club challenged all aspects of Georgia's program, including the identification of water quality-limited segments; water quality monitoring; prioritization; and the number, adequacy, and pace of the development of TMDLs.(133) Georgia had submitted its first list of 123 waters in 1992 and by the time of trial had increased the list to 340 waters.(134) However, the state had only submitted two TMDLs (one in 1994 and one in 1995), claimed to be working on another two, and anticipated completion of an additional twenty-eight by the year 2005.(135) Though the court did not determine whether the list was adequate, it found the TMDL schedule insufficient, noting that it would take over 100 years at the state's current pace to prepare TMDLs for the currently listed waters.(136) c. Imposing a Schedule for Preparation Enforcing implementation of the program leads to the related goal of imposing a court-approved schedule for TMDL preparation in states that insist on dragging their feet with respect to implementation. The Idaho and Georgia cases provide good examples of this phase of TMDL litigation. While the district court in Idaho Sportsmen's Coalition remanded the scheduling decision to the agency with a suggestion that five years would be a "reasonable" amount of time,(137) the Sierra Club v. Hankinson court went further by ordering the completion of the TMDLs within five years on a basin-by-basin schedule.(138) Conditions of that order included the revision or termination of NPDES permits within one year following each new TMDL, permits for new dischargers into water quality-limited segments, and EPA revision of the NPDES program or revocation of state certification for state failure to comply.(139) The court also retained jurisdiction over the case and ordered EPA to submit detailed annual progress reports to it and the plaintiff.(140) Implementation schedules are often set in settlement, rather than by court order. In the Virginia case where the district court noted that the state and EPA had submitted either one or no TMDLs in the twenty years following the 1979 statutory deadline, the parties reached an agreement and the court approved the consent decree in 1999.(141) The consent decree included the following: 1) an eleven to twelve year schedule for Virginia to establish TMDLs for all waters on its 1998 section 303(d) list, 2) requirements that EPA establish the TMDLs if Virginia fails to do so, and 3) provisions addressing a number of other Clean Water Act and endangered species issues.(142) 3. Challenging the Substance of a TMDL Increasingly, litigation will focus on challenging the substance of individual TMDLs that states have submitted or EPA has approved. Such disputes will focus on whether the TMDLs are adequate under the requirements of section 303 and the EPA regulations. Substantive challenges to individual TMDLs will focus on: 1) whether the TMDL is scientifically defensible, 2) whether and when the established time frame will achieve compliance with water quality standards, and 3) whether the TMDL is sufficiently certain to be effective and actually achieve compliance. In perhaps the only true example so far of this stage of TMDL litigation, both environmental and industry plaintiffs challenged the adequacy of a TMDL established by EPA for dioxin in the Columbia River.(143) EPA estimated in that case that Washington, Oregon, and Idaho pulp and paper mills along the Columbia River would have to reduce their dioxin discharges by 95% in order to achieve the permissible 0.013 parts per quadrillion ("ppq") ambient concentration of dioxin.(144) Affirming the district court's decision to uphold the TMDL, the Ninth Circuit rejected several arguments, including a claim that the TMDL failed to consider the cumulative effect of dioxin with other chemicals present in the water.(145) The court pointed out that not only does the Clean Water Act not require TMDLs to be issued for all pollutants at once, but also that the regulations actually provide that TMDLs may be developed on a pollutant-by-pollutant basis.(146) The court also rejected an industry claim that EPA could not establish the TMDL until technology-based effluent limits for dioxin had actually been tested and proven ineffective.(147) Instead, the court agreed with EPA that "when a state has listed a water as impaired by toxic pollutants, the EPA has authority to implement TMDLs for that toxic pollutant under [section 303(d)] even before technological limitations have been developed and implemented pursuant to [section 301(b)(1)(A) or (B)]."(148) Although the court upheld the adequacy of this particular TMDL in the face of challenges from all sides, it is certainly among the first in a long line of cases that will begin to challenge the substance of the numerous TMDLs scheduled for implementation over the next decade. 4. Summary of the Legal Strategy Under the Current TMDL Framework In short, the legal strategy to prod the TMDL program along follows a chronological framework of first ensuring adequate state lists of section 303(d) waters, followed by implementation of state TMDL programs (including schedules for TMDL preparation) and finally turning to substantive challenges to individual TMDLs. Although EPA's power to enforce nonpoint source controls under the Clean Water Act is limited, it nevertheless has some options in this realm. Among these options is EPA's authority to promulgate regulations,(149) the full extent of which is still debatable.(150) Finally, as more and more TMDLs are produced and become available as examples of effective, acceptable models of TMDLs, substantive challenges to individual TMDLs will increase in number and enjoy increased leverage. C. Application to Oregon Nonpoint Source TMDLs Currently, two TMDLs in the State of Oregon consist entirely of nonpoint source pollution: the Sucker-Grayback Water Quality Management Plan (WQMP) and the South Steens WQMP/TMDL.(151) The Sucker-Grayback WQMP covers an area of primarily Forest Service and BLM lands in the Siskiyou Mountains of southern Oregon. The area has no point source discharges, and it is a key watershed in President Clinton's Northwest Forest Plan.(152) The South Steens WQMP/TMDL covers all BLM lands on and surrounding Steens Mountain in the high desert of southeastern Oregon. Because that area's water quality problems originate almost entirely from livestock grazing, it is analyzed here as a TMDL case study. 1. Water Quality Management Plans as TMDLs in Oregon Interestingly, the Oregon Department of Environmental Quality (DEQ) has issued guidance that describes how a WQMP may serve as a nonpoint source TMDL.(153) In Oregon, DEQ has the responsibility to identify and list section 303(d) waters.(154) DEQ states that "[i]n most cases, an approved WQMP can stand alone as a TMDL for any watershed where nonpoint sources are the only sources of water pollution."(155) According to the guidance, a WQMP must address several "basic elements," including identification of responsible participants.(156) This also involves providing reasonable assurance that the plan will be implemented.(157) The WQMP-TMDL guidance provides examples of the legal or contractual authority to enforce implementation, including state or federal enforcement authorities under the Clean Water Act; authorities under the Oregon Forest Practices Act(158) or the Agricultural Water Quality Act of 1993;(159) permit, lease, or contract authorities of public land management agencies; enforceable obligations arising out of funding sources that assist WQMP implementation; and land use and local ordinances.(160) Not surprisingly, DEQ suggests that WQMPs "should be voluntarily developed and voluntarily implemented, and the three requirements listed above should not be interpreted to mean that the plan must be narrowly prescriptive in nature or rely on regulatory mechanisms for success."(161) 2. South Steens WQMP/TMDL Steens Mountain is a sixty-mile (ninety-six kilometer) long, 9700-foot (2900-meter) elevation, fault-block mountain located in southeastern Oregon's high desert. There are no point sources of water pollution in the South Steens watersheds--according to BLM, the main sources of water pollution are livestock grazing, wild horse use, and fire suppression.(162) The Steens Mountain region is typical of the arid interior western United States, with a diverse array of unique and sensitive plant and animal species, an extreme landscape, and a long history of public lands livestock grazing.(163) The South Steens WQMP is based on two previous planning documents: the South Steens Allotment Management Plan and Environmental Assessment (AMP/EA),(164) and the Catlow Redband Trout and Tui Chub Conservation Agreement and Strategy (CCA).(165) The WQMP follows the outline set forth in DEQ's WQMP-TMDL Guidance. a. Pollutants, Water Quality Standards, and Surrogate Measures in Lieu of Daily Loads The South Steens TMDL addresses two pollutants: 1) heat, and 2) sedimentation from stream bank erosion.(166) Three water bodies in the South Steens area are included on Oregon's 1996 section 303(d) list, and five more were added to the 1998 list.(167) The applicable water quality standard for temperature for listed 303(d) South Steens area streams in the Malheur Lake Basin states that "[n]o measurable surface water temperature increase resulting from anthropogenic activities is allowed ... (i) [i]n a basin for which salmonid fish rearing is a designated beneficial use, and in which surface water temperatures exceed 64.0 [degrees] F (17.8 [degrees] C)."(168) For sediment, the water quality standard states that "[t]he formation of any organic or inorganic deposits deleterious to fish or other aquatic life ... shall not be allowed."(169) Because of the "inherent challenges" of TMDL development for nonpoint sources, the TMDL explains that it uses "surrogate measures" other than daily loads to meet the requirements of section 303(d).(170) The two surrogate measures used are "view to sky" and "active eroding stream bank."(171) An increased "view to sky" results from a lack of adequate riparian vegetation, which decreases shade cover over a stream and allows for greater incoming solar radiation to hit the water surface.(172) Actively eroding stream banks lead to wider channels, which in turn increase the amount of surface area of the water exposed to heat through incoming solar radiation.(173) Thus, rather than using heat loads in megawatt-hours per day, the TMDL employs these surrogate measures, which are presumably easier to understand, monitor, and implement.(174) Theoretically, "[al more efficient approach is to evaluate indicators which [sic] can be applied across a broader scale and have a demonstrated effect on reducing water temperatures."(175) b. Development of Loading Capacities and Allocations The South Steens TMDL expresses loading capacities as targets for "view to sky" and "active eroding stream bank."(176) A loading capacity is "the greatest amount of loading that a water can receive without violating water quality standards."(177) Allocations, in turn, are portions of the loading capacity that are allocated to point or nonpoint sources and the natural background;(178) a TMDL, then, is the sum of the allocations.(179) According to the analysis in the South Steens TMDL, targets are set at a maximum of twenty percent both for view to sky and for active eroding stream bank.(180) c. Measuring the South Steens WQMP/TMDL Against the TMDL Requirements The primary criticisms of the South Steens TMDL include its adjustments for seasonal variation and a margin of safety, its implementation mechanisms, and its less than candid treatment of current grazing practices as a major cause of water pollution in the area.(181) Section 303(d) requires that TMDLs be established at a level necessary to implement water quality standards with seasonal variations and a margin of safety.(182) This bare requirement leaves a good deal of room for agency discretion in defining the terms "seasonal variation" and "margin of safety." One criticism of the seasonal variation element of the South Steens TMDL is that the data is sparse and outdated. The TMDL states that the data used to represent "existing conditions for stream temperatures," which were collected between 1976 and 1981, is "the most comprehensive record of water temperature taken at one site over an extended period of time in the South Steens area."(183) Nevertheless, an agency likely cannot be compelled to produce current and comprehensive data unless the lack of data is so egregious as to leave the agency utterly uninformed of potential seasonal variations.(184) Even in such a situation, section 303(d) still requires the TMDL to "[take] into account any lack of knowledge concerning the relationship between effluent limitations and water quality."(185) With respect to the margin of safety, the TMDL states that "[t]he [margin of safety] may be implicit, as in conservation assumptions used in calculating the loading capacity, [wasteload allocations], and [load allocations]."(186) Again, the statute and regulations do not say whether the margin of safety must be explicitly stated in terms of a separate quantity in the TMDL calculation. Nevertheless, the so-called margin of safety in the South Steens TMDL is dubious because it consists of only a few sentences that describe certain assumptions with respect to the effects of air temperature and wind speed on water temperature.(187) Presumably, the TMDL calculations incorporate this margin of safety by way of conservative assumptions regarding the effects of these two factors on water temperature. However, it is unclear where these assumptions find their way into the actual calculations of loading capacities or load allocations. Finny, with respect to implementation, the TMDL's "Water Quality Standards Attainment Analysis"(188) proposes measurements of vegetation height along streams as a measure of the view to sky element, but it offers no measuring or monitoring system for the actively eroding stream bank element.(189) The document is also extremely vague in its description of implementation actions, providing only a bare list of such actions as "a management system to ensure proper grazing use," "livestock access control," "range and pasture rehabilitation," and "rangeland developments."(190) None of the suggested actions are described quantitatively or with adequate specificity. Even if the proposed rules discussed in the following section take effect, they would have little or no effect on present TMDLs such as the South Steens TMDL, as long as the TMDL meets the current requirements. EPA states in the proposed rules that it will approve any TMDL submitted to it within twelve months of the final rule's effective date if the TMDL meets either the current or new requirements.(191) One of the important lessons of the South Steens TMDL is that interested organizations need to take an active, early role in TMDL development for nonpoint sources, because effective quantitative requirements are essential to the TMDL's success. D. Proposed Rules Revising Listing and TMDL Requirements In August 1999 EPA issued proposed rules to "revise, clarify and strengthen" the current regulatory requirements for identifying and listing impaired waters and establishing TMDLs.(192) Several important revisions may significantly increase the effectiveness of the TMDL program. These revisions include updated definitions and distinctions between the terms "pollutant" and "pollution"; a requirement that states' listing methodologies be more specific, subject to public review, and periodically submitted to EPA;(193) a requirement for states to submit schedules for establishing TMDLs; clarification of the specific elements that a TMDL must contain, including an implementation plan; and increased public participation requirements.(194) Thus, the scope of the proposal covers the entire TMDL process--from the listing of waterbodies to development and implementation of the TMDL itself. The following discussion highlights a few of the most significant proposed revisions. 1. Revisions to the Listing Process EPA's proposed rules significantly revise the listing methodology, the scope of the lists, the components of the list, and the schedules for TMDL development.(195) As part of a broader effort to include the public at every step of the way, the states' methodologies for data collection and evaluation must not only be submitted annually to EPA, but also must be available for public notice and comment.(196) Importantly, EPA clarifies that the scope of the list includes waters impaired or threatened by point sources of pollution only, by both point and nonpoint sources, or by nonpoint sources only.(197) EPA correctly confirms that section 303(d) has ample authority to require listing of those waters degraded solely by nonpoint source pollution.(198) a. The Distinction Between "Pollutant" and "Pollution" In order to make the lists as comprehensive as possible, the proposed rules require that states list waters impaired and threatened both by pollutants and pollution.(199) "Pollution" is defined in the Act as "the man made or man-induced alteration of the chemical, physical, biological, or radiological integrity of water."(200) "Pollutant," on the other hand, is more narrowly defined by a long laundry list of very specific materials.(201) This requirement will certainly lead to very comprehensive lists of impaired and threatened waters. However, TMDLs are only required for those waters impaired by pollutants--because section 303(d) expressly requires TMDLs only "for those pollutants [identified by EPA]"--not pollution.(202) Nevertheless, the term "pollutant" is broad in at least some respects:(203) not only is one of the listed terms--"biological materials"--broad in and of itself, but the definition of "pollutants" in section 502(6) includes a wide variety of wastes that courts may construe as pollutants whether or not they are intended to be waste.(204) While requiring TMDLs only for pollutants does not appear to greatly hinder the establishment of TMDLs for nonpoint sources, a requirement that TMDLs be established for both "pollutants" and "pollution" would certainly be more comprehensive. For example, water quality degradation in the form of reduced stream flow or channelization--environmental harms as devastating to stream health as increased sedimentation or increased temperature(205)--is not covered under the definition of "pollutants," but it surely would be under the broad definition of "pollution." At least one commentator has suggested that the language of the Clean Water Act is inconsistent enough in its use of the two terms that EPA arguably has the discretion to require TMDLs for pollution as well as pollutants.(206) There are at least a few instances where the Act clearly suggests that the broader approach is preferable, including the Act's ecosystem-based goal of preserving the `biological integrity' of the Nation's waters and the requirement that states prioritize their 303(d) waters according to the severity of "pollution."(207) A strong argument exists for an ecosystem-level, "holistic watershed approach" to pollution control.(208) Nevertheless, because of the vagueness of the statutory language, EPA is almost certainly entitled to significant deference on the issue, and its interpretation still arguably encompasses the majority of nonpoint source pollution problems, including the most pervasive grazing-related problems of heat and sediment increases. b. Schedules for TMDL Development The other significant provision with respect to the listing process under the proposed rules is a requirement that states develop and submit to EPA schedules for establishing TMDLs for all impaired or threatened waters on the 303(d) list.(209) This is not required under the existing regulations, and EPA states in the proposed rule that this provision is the result of a strong recommendation by the FACA committee and numerous lawsuits in the 1990s.(210) Affirming a 1997 EPA policy memorandum,(211) the proposed rule requires that states complete TMDLs for all "Part One" waters(212) in not more than fifteen years, with a suggested target of eight to thirteen years and a maximum limit of five years for all high-priority waters.(213) By adding this time schedule requirement, EPA invokes its authority and discretion to interpret the statutory phrase "from time to time," which is not defined in the Act, to mean a reasonable pace.(214) 2. Revisions to TMDL Establishment and Review The proposed rule outlines ten minimum requirements for a satisfactory TMDL,(215) the most significant of which is the requirement that all TMDLs include an implementation plan. According to one commentator, "[i]mplementation is of course where all public laws live or die,"(216) and, for the Clean Water Act's first two decades, implementation under section 303(d) "languished."(217) The proposed obligation is derived from the requirement in section 303(d) that TMDLs "be established at a level necessary to implement the applicable water quality standards."(218) Perhaps anticipating strident challenges to EPA's authority to impose such requirements, the proposed rule makes clear EPA's stance that, under a Chevron(219) analysis, Congress did not speak clearly on the point and therefore EPA has permissibly construed the statute.(220) EPA supports the implementation plan requirement as "one additional appropriate way to enable EPA to determine properly whether or not a TMDL is established at the level necessary to implement the applicable water quality standards."(221) The mandatory implementation plan has several elements of its own, including two particularly critical elements: reasonable assurance and legal or regulatory controls. a. Reasonable Assurance of Implementation A reasonable assurance is "a high degree of confidence that wasteload allocations and/or load allocations in TMDLs will be implemented by Federal, State or local authorities and/or voluntary action."(222) For nonpoint sources, reasonable assurance means "nonpoint source controls are specific to the pollutant of concern, implemented according to an expeditious schedule and supported by reliable delivery mechanisms and adequate funding."(223) According to EPA, nonpoint source reasonable assurance might include state or tribal regulations or local ordinances, performance bonds, memoranda of understanding, contracts, or other voluntary actions or agreements.(224) This definition is somewhat troubling in that a "conservation agreement" between private parties and an administering agency might exclude other interested and affected parties--such as environmental groups or public land users--and may only be subject to future monitoring through enforcement action.(225) Thus, it may well be several years before any adjustments would be made to this kind of voluntary implementation mechanism because, for example, the parties might claim that they need more time to see whether it is working. A concern central to all TMDL enforcement issues is the extent of EPA's authority under the Clean Water Act to implement controls over nonpoint sources, should it disapprove of a TMDL. If EPA finds that a nonpoint source control offered as reasonable assurance of implementation--such as a conservation agreement between a private party and a state agency--is inadequate, does EPA have the authority to reject that TMDL? Though EPA characterizes its options as "strong and diverse authorities,"(226) the adequacy of the authorities is questionable and perhaps amounts to a bluff. EPA's authority rests on three different Clean Water Act provisions: 1) section 504 emergency powers,(227) 2) section 319 nonpoint source management programs, (228) and 3) section 501(a) authority to promulgate regulations.(229) First, the "emergency powers" under section 504 allow EPA to bring suit to restrain any person from causing "an imminent and substantial endangerment to the health of persons."(230) This seems to contemplate a rather grave situation and certainly does not mention endangerment of fish and wildlife or stream health. Second, EPA cites its authority to direct states' use of funding provided under section 319's "Nonpoint source management programs."(231) This power has proven inadequate far too often in this program's history, simply because states may choose whether or not they want to opt into the program.(232) A humor problem with both of these options from the environmental perspective is that they leave little or no room for citizen enforcement, should the public become dissatisfied with EPA's actions. The Clean Water Act's citizen suit provision allows citizens to sue EPA where the agency has failed to perform a nondiscretionary duty.(233) Here, however, EPA has complete discretion regarding both the emergency powers provision and the section 319 funding decisions, save for an egregious and difficult-to-imagine emergency situation where EPA refuses to act. As a result, if a state enters into a voluntary agreement as reasonable assurance of implementation of a nonpoint source TMDL, citizen groups may have little or no recourse should they feel that the agreement is not being faithfully executed. Finally, section 501(a) authorizes EPA to "prescribe such regulations as are necessary to carry out [its] functions ...."(234) The extent to which this generically stated provision allows EPA the discretion to find places in the Clean Water Act where R can implement nonpoint source controls is certainly subject to continual debate and challenge. b. Legal or Regulatory Controls Over Implementation The second significant requirement in the proposed rule's elements of an acceptable TMDL implementation plan is the inclusion of legal or regulatory controls. Any implementation plan must describe the legal authorities under which implementation will occur.(235) For nonpoint sources, EPA cites the Almanac of Enforceable State Laws to Control Nonpoint Source Pollution(236) as the source of each state's laws and regulations.(237) According to the Almanac, some state water pollution control law provisions in Oregon can be enforced against nonpoint sources, with the notable exception of forestry practices.(238) For example, where the State Department of Agriculture has developed a water quality management plan, grazing must be conducted "in full compliance with the plan and rules implementing the plan and with all rules and standards of the [state Environmental Quality Commission] relating to water pollution control."(239) While a requirement that a TMDL contain "legal or regulatory control" sounds valuable, reliance on state water pollution control law often amounts to a cadre of widely varying, exception-riddled requirements. Nevertheless, because state water pollution control requirements must at least meet the federal minimum, this obligation is at least valuable inasmuch as it forces a state to point to exactly where it feels that its implementation can be enforced. EPA's ability to enforce against nonpoint source pollution under the Clean Water Act is scant; to the extent that it has any power at all, the proposed regulations at least attempt to characterize that power in the strongest light possible. 3. EPA Establishment of TMDLs and Public Participation The proposed revisions expressly codify EPA's authority to establish a TMDL for a state if the state so requests or if EPA determines that the state is unlikely to meet its time schedule.(240) Because this authority is not explicit in section 303(d), EPA relies on its authorization in section 501(a) to promulgate regulations.(241) For this portion of the revisions, EPA cites Scott v. City of Hammond(242) and Dioxin/Organochlorine Center v. Clarke.(243) There is little argument that courts have found this authority to be within EPA's powers. As for public participation, the proposed revisions are significant because the current regulations contain no such requirements. The proposed rule provides for a comment period for all aspects of the list (including methodology), the schedule, and the TMDLs themselves. This comment period must occur prior to states' submissions to EPA.(244) In addition, not only may EPA comment during this period, but states must consider any comments received by wildlife agencies prior to submission to EPA.(245) In conjunction with the proposed substantive requirements of the TMDLs, these public participation requirements could be very important. For example, despite the potential weakness of permitting voluntary actions or agreements as reasonable assurance under an implementation plan, a citizens group may at least have the opportunity to voice its concern first to the state and then to EPA as it considers the state proposal. 4. Additional Potential Legal Strategies Under the Proposed Rules a. Listing A number of new strategies may evolve if EPA adopts some version of the proposed revisions to the water quality planning process. Most significant among the proposed revisions to the listing process is the provision for public notice and comment on all aspects of the listing process, including state submittals of listing methodologies. The public can attempt to influence decisions early in the process by ensuring that the methodology is valid and is capable of producing a comprehensive list of impaired and threatened waters. Citizens should also pay close attention to ensure that the lists include impaired and threatened waters. The comprehensive nature of the newly defined lists will serve as a valuable baseline of data to evaluate a state's progress toward achieving water quality standards in all of its waters. In addition, requiring a schedule with the list should help to ensure a more expedited TMDL listing process. b. TMDLs Even though TMDLs would only be required for "pollutants" and not "pollution," the statutory definition of "pollutants" adopted in the proposed regulations still contains terms inclusive enough to include pollution such as "heat, biological materials, rock, sand, and agricultural waste" from livestock grazing.(246) Also, by mandating an opportunity for public comment on TMDLs, the public should make sure not only to provide comments but also to ensure that the relevant wildlife agencies provide comments as well. Finally, citizens should get involved in the implementation plan portion of the TMDL in at least two ways. First, the public should attempt to become involved in any private or voluntary agreements intended to serve as reasonable assurance of implementation. Theoretically, agencies and private parties--for example, ranchers or timber companies--may be interested in having an environmental group involved in order to ascertain at the outset what are likely to be the most contentious aspects of environmental planning in a given area, or perhaps to share in monitoring efforts. Admittedly, this may be an overly optimistic scenario, but little doubt exists that inclusion of all parties with an interest in the public lands at issue would significantly improve the efficiency and effectiveness of the planning process. Second, citizens must ensure that implementation plans specifically describe adequate legal or regulatory authorities and controls on the plans to limit any potential ambiguity as to how the plan can (and will) be enforced down the road. Calling attention to vague language and dubious control sources early on may avoid future controversies. E. Conclusion: TMDLs As Effective and Evolving Mechanisms Addressing Nonpoint Source Pollution Trumpeted by some as being the most effective legal strategy to address nonpoint source pollution from livestock grazing and similar activities, the TMDL program holds great promise as it continues to evolve.(247) As the number of court orders, consent decrees, and settlement agreements across the country indicate, the next decade or so should push the program solidly into the phase of TMDL development and implementation. As more and more individual TMDLs are developed and the states, EPA, and the courts become increasingly familiar with what exactly constitutes an adequate and effective TMDL, plaintiffs' efforts to ensure that TMDLs meet substantive requirements should also become more and more successful. While there is cause for optimism, it will nonetheless be quite some time--potentially decades--before the nation has a comprehensive coverage of TMDLs for its impaired waters.(248) As a result, more immediate strategies are also needed that address nonpoint source polluting activities, such as grazing, on a site-specific basis. This is where another portion of the Clean Water Act, the federal facilities provision, enters the picture. III. SECTION 313 FEDERAL FACILITIES POLLUTION CONTROL A significant and virtually untested legal theory under the Clean Water Act involves section 313, a provision regarding federal facilities pollution control.(249) Section 313 provides that: Each department, agency, or instrumentality of the ... Federal Government (1) having jurisdiction over any property or facility, or (2) engaged in any activity resulting, or which may result, in the discharge or runoff of pollutants ... shall be subject to, and comply with, all Federal, State, interstate, and local requirements, administrative authority, and process and sanctions respecting the control and abatement of water pollution in the same manner, and to the same extent as any nongovernmental entity including the payment of reasonable service charges.(250) A seldom-used provision of the Act, section 313 requires federal agencies to adhere to state water quality regulation to the same extent as nongovernmental entities.(251) Only the President can grant an exception to this requirement.(252) The provision is particularly significant because it includes the phrase "discharge or runoff" in its language.(253) As the Ninth Circuit emphasized in ONDA v. Dombeck, "the word `discharge' [in the Clean Water Act] is used consistently to refer to the release of effluent from a point source. By contrast, the term `runoff' describes pollution flowing from nonpoint sources."(254) Indeed, the court flatly stated that section 313 "plainly applies to nonpoint sources of pollution on federal land."(255) The Ninth Circuit has repeatedly confirmed this requirement with respect to timber and road-building activities, and has held that the APA(256) provides a cause of action to enforce federal facilities nonpoint source pollution under section 313.(257) The power of the section 313 approach, then, is threefold:(258) 1) it expressly includes "runoff of a pollutant;" 2) it does not simply lead to (often meaningless) state certification; and, 3) coupled with a cause of action under the APA, it could prove to be very effective, regardless of what a state chooses to do under section 401 of the Clean Water Act. Thus, this overlooked statutory provision may in fact hold the elusive "key to unlocking the Clean Water Act."(259) A. Legislative History of Section 313: Purpose and Significance of Including the Term "Runoff" Though the absence of the term "runoff" proved to be the downfall of the section 401 theory in ONDA v. Dombeck, the term's presence in section 313 is very significant and more than coincidental. As has been noted elsewhere, Congress purposefully included both the words "discharge" and "runoff" in section 313.(260) First drafted in 1970 as section 21(a) of the Water Quality Improvement Act of 1970,(261) this federal facilities provision originally directed all federal activities to comply with state water quality standards for "any discharges allegedly contributing to pollution from any such Federal property."(262) With the emergence of the Clean Water Act in 1972, section 21(a) became section 313 of the Clean Water Act, and Congress amended "any discharge" to read "discharge or runoff of pollutants."(263) It is significant that even though Congress did not explicitly alter the permit scheme in section 401 to include both the terms discharge and runoff, it did so in section 313. Therefore, the presence of the term "runoff" in section 313 is more than congressional oversight or over-emphasis; it is a careful insertion that demands a plain-meaning interpretation. The Senate Report from the 1972 Act notes: Evidence received in hearings disclosed many incidents of flagrant violations of air and water pollution requirements by Federal facilities and activities. Lack of Federal leadership has been detrimental to the water pollution control effort. The Federal Government cannot expect private industry to abate pollution if the Federal Government continues to pollute.(264) The report accompanying the House bill also states that "[t]he lease of public lands by various Federal agencies to concessioners or other organizations for the purpose of providing regional facilities often results in serious sanitation problems."(265) Congress largely incorporated the Clean Water Act's federal facilities provision from a nearly identical provision set forth in the Clean Air Act(266) two years earlier.(267) Section 118 of the Clean Air Act stated at that time that the federal government "shall ... comply with, all Federal, State, interstate, and local requirements ... respecting control and abatement of air pollution in the same manner, and to the same extent as any nongovernmental entity."(268) According to one commentator, section 118 was significant in that it "made federal polluters subject not just to `standards,' but to a broader class of controls labeled `requirements.'"(269) Besides the Clean Water Act and the Clean Air Act, federal facilities provisions also appear in the Resource Conservation and Recovery Act(270) (RCRA) and the Comprehensive Environmental Response, Compensation, and Liability Act(271) (CERCLA).(272) B. Application of Section 313 to Nonpoint Source Pollution and Livestock Grazing Application of section 313 to address nonpoint source water pollution on federal lands requires two basic steps. First, one must establish that the federal government has a duty or obligation under the statutory provision. The language of section 313 clearly requires federal agencies to adhere to state water quality regulations to the same extent as non-governmental entities.(273) This requirement unquestionably includes nonpoint source pollution because of the inclusion of the term "runoff." Second, after establishing that the federal government has a duty under section 313, one must establish that the duty is enforceable. Under ONDA v. Dombeck, it is settled--at least in the Ninth Circuit--that the Clean Water Act's citizen suit provision does not apply to nonpoint source pollution. Therefore, if nonpoint source pollution control is enforceable against the federal government under section 313, it is the APA that must provide that enforceability.(274) This dual framework of first establishing a duty and then establishing enforceability guides the ensuing discussion. C. The Federal Government's Duty Under Section 313 to Adhere to Water Quality Standards 1. Case Law in the Ninth Circuit and Beyond a. The Ninth Circuit Cases: From Northwest Indian Cemetery Protective Association to Idaho Sporting Congress Over the past fifteen years, the Ninth Circuit has repeatedly confirmed that the federal government must comply with all federal, state, and local requirements respecting the control and abatement of water pollution, including state water quality standards and activities causing nonpoint source pollution.(275) The first in this line of Ninth Circuit cases, Northwest Indian Cemetery Protective Association v. Peterson,(276) involved a challenge to Forest Service timber harvesting and road building on the Six Rivers National Forest in the Siskiyou Mountains of northern California.(277) An association of Northwest Indian tribes alleged, among other claims, that the Forest Service's decisions violated the Clean Water Act because the federal agency was not adhering to California state water quality standards.(278) The government argued that the standards established by the state Water Quality Control Board were no longer applicable to the Forest Service because the Forest Service Best Management Practices (BMPs), which were accepted by the State of California and EPA, superceded those standards.(279) However, the court found that BMPs are merely means to achieve the state water quality standards, and neither the plan established by the Board, nor the agreements between the Forest Service and the Board, indicated that the BMPs themselves should be considered as standards.(280) Affirming the district court, the Ninth Circuit held that, because the Forest Service projects would have violated the state standards, the agency was in violation of section 3137.(281) A year later, the Ninth Circuit continued to define federal responsibilities under section 313 in another timber harvest and road construction case.(282) In Oregon Natural Resources Council v. United States Forest Service (ONRC v. USFS),(283) the plaintiffs alleged that the Forest Service's activities associated with a timber sale on the Willamette National Forest in Oregon violated state water quality standards and, therefore, were in violation of section 313.(284) Specifically, ONRC claimed that the defendants violated and planned to violate both Oregon's nondegradation standard that "existing high quality waters ... shall be maintained and protected" and a rule that activities in the Willamette Basin must not cause a ten percent or greater cumulative increase in natural stream turbidities.(285) Citing Northwet Indian Cemetery Protective Association, the plaintiffs argued that the Ninth Circuit had already "recognized the rights of citizens to enforce state water quality standards against the [Forest Service]."(286) The court accepted this duty under section 313 without further discussion and considered the source of plaintiffs' cause of action for the remainder of the opinion.(287) A few years later, in a case arising out of a fire-recovery timber sale on the Klamath National Forest in California, an environmental organization alleged that the proposed agency action would violate a state water quality control plan adopted by California's Water Quality Control Board.(288) While the state plan required that turbidity must not increase by more than twenty percent, the turbidity levels from the combined effects of the fire and the project would exceed that level.(289) Citing ONRC v. USFS, the Ninth Circuit reaffirmed that the Forest Service must comply with all state water quality standards, a duty that included violations from nonpoint sources.(290) Finally, in 1998 the Ninth Circuit stated that the requirement that all federal agencies comply with state water quality standards includes a state's antidegradation policy.(291) In Idaho Sporting Congress v. Thomas,(292) the plaintiff challenged two timber sales on the Targhee National Forest in southeastern Idaho, alleging that sedimentation from the logging operations would affect a species of trout that was a management indicator species for the forest.(293) The plaintiffs alleged that Idaho's antidegradation policy required the Forest Service to maintain existing water quality unless a formal determination was made that lowering water quality was necessary for social or economic reasons.(294) The court confirmed the agency duty but decided that sufficient facts did not exist to determine whether the antidegradation standard had been violated.(295) Even in ONDA v. Dombeck, the court acknowledged that section 313 differed from section 401 in the significant respect that the former included the term "runoff": "The word `discharge' [in the Clean Water Act] is used consistently to refer to the release of effluent from a point source. By contrast, the term `runoff describes pollution flowing from nonpoint sources."(296) The court flatly stated that "[s]ection [313] plainly applies to nonpoint sources of pollution on federal land."(297) b. Other Circuits and the Interior Board of Land Appeals The issue of whether section 313 imposes a duty on the federal government to adhere to all state requirements with respect to water pollution has been explicitly addressed only in the Ninth Circuit. No cases from other circuits cite ONDA v. Dombeck for its section 401-livestock grazing-nonpoint source holding. This is also true for the Oregon and Idaho district court cases,(298) which had split on the section 401 certification issue and consequently led to the ONDA v. Dombeck decision. There is very little direct case law on the issue of nonpoint source control under the Clean Water Act to indicate the likelihood of other circuits accepting a section 313 claim. In a recent Eighth Circuit case, an environmental organization challenging four timber sales on the Ozark National Forest in Arkansas claimed that, because the sales were contrary to the state's antidegradation policy, the sales violated section 313.(299) Although that court ultimately declined to accept plaintiff's Clean Water Act challenges, it did state that it would "assum[e] without deciding that compliance with a state antidegradation policy is a legitimate inquiry on APA review of this type of action."(300) Even assuming the legitimacy of the section 313 claim, the court rejected the challenge because it found Arkansas' state policy for nonpoint source pollution to be "so broadly stated that the Forest Service was not arbitrary or capricious in concluding this policy added nothing to its compliance obligations under federal environmental laws."(301) Under the right circumstances--a clear violation of water quality standards in a state with specifically stated, adequate standards--this court appears to leave the door open for a successful section 313 nonpoint source challenge. No other federal courts have addressed this type of challenge under section 313.(302) At the Interior Board of Land Appeals (IBLA) level of administrative review, a recent case, which included a section 313 claim over grazing impacts on a desert spring designated as an outstanding National resource water, led to an administrative stay of grazing.(303) In that case, the National Wildlife Federation (NWF) appealed a finding of no significant impact (FONSI) issued under the National Environmental Policy Act (NEPA)(304) by the Arizona BLM, approving the use of a spring in conjunction with grazing on a particular grazing allotment.(305) The spring is the source for the Peeples Canyon Creek, the entire length of which was designated by the State of Arizona as a "unique water."(306) State regulations provide that "[e]xisting water quality shall be maintained and protected in a surface water that is classified as a unique water ... pursuant to [state law]."(307) Besides a FLPMA claim that was not addressed by the administrative law judge (ALJ), NWF alleged violations of section 313 of the Clean Water Act as well as NEPA.(308) NWF claimed that BLM's approval of the use of the spring in conjunction with grazing on the allotment violated section 313 because the action would result in degradation of the water quality in Peeples Canyon Creek--that is, grazing would fail to "maintain and protect" as required by state law.(309) While BLM admitted in its EA that water quality for the spring might be changed under the preferred action alternative, it argued incorrectly that the spring was not connected to the "unique water" segment of Peeples Canyon Creek because no continuous stream of surface water runs from the spring to the creek.(310) The ALJ confirmed that section 313 requires the BLM to comply with all state water pollution requirements, including regulation of nonpoint sources of pollution.(311) Citing ONDA v. Dombeck,(312) the ALJ also stated that section 313 includes regulation of "nonpoint sources of pollution, including runoff [] associated with cattle grazing on Federal lands."(313) The ALJ found that BLM could not have made an informed judgment regarding its action because BLM did not consider the unique water designation to extend the entire length of the creek.(314) The success of NWF's section 313 claim thus seemed to turn on the fact that the FONSI was insufficient because the preferred alternative only considered the environmental impacts according to BLM's incorrect interpretation of the boundaries of the unique water. Perhaps most significant in this case, however, was the ALJ's unequivocal statement that section 313 applies to runoff associated with livestock grazing on federal lands. In short, no federal circuits other than the Ninth have directly addressed the issue of the federal government's duty under section 313 to adhere to water quality standards--and even the Ninth Circuit has not passed on the question in the context of a federal permit issuance for livestock grazing on public lands. To the benefit of potential plaintiffs, nothing in the limited precedent analyzing section 313 precludes a successful claim in the grazing context Moreover, the IBLA ruling in Arizona adds considerable support to the existence of a federal duty under section 313, even if the holding was primarily NEPA-dependent and lacks precedential value in a federal court. Despite the seemingly clear duty that the Ninth Circuit and other courts have articulated--often, but with very little discussion---under section 313, there are several potential counter-arguments that the provision does not impose such a duty with respect to nonpoint source pollution and livestock grazing. These include the meaning of the term "federal facility," the meaning of the phrase "to the same extent as any nongovernmental entity," the question of who is doing the actual polluting, and the subject matter of the Ninth Circuit line of cases. 2. The Meaning of "Federal Facility" One of the preliminary issues that will determine the success of the section 313 argument involves the question of what exactly the term "federal facility" means. Can the term be extended beyond its plain meaning of actual, physical facilities such as buildings? Does it include federal permits for use of public lands? The answer to these questions is undoubtedly "yes." This is critical, of course, because the Federal Land Policy and Management Act (FLPMA) provides for the Secretary of the Interior (or the Secretary of Agriculture in the case of National Forests) to issue grazing permits, which allow individuals to graze livestock on federal lands.(315) FLPMA defines the term "grazing permit and lease" as "any document authorizing use of public lands or lands in National Forests in the eleven contiguous western States for the purposes of grazing domestic livestock."(316) The plain language of section 313 itself appears clearly to include federally issued permits for the use of public lands. The federal facilities provision is phrased in terms of any "department, agency, or instrumentality ... having jurisdiction over any property or facility" or "engaged in any activity" which results or may result in the discharge or runoff of pollutants.(317) A search for an actual definition of the phrase yields nothing in the Clean Water Act, nor in the other environmental statutes that contain federal facilities provisions. The Office of Management and Budget (OMB) has defined the term administratively for pollution abatement purposes as the "buildings, installations, structures, land, public works, equipment, aircraft, vessels, and other vehicles and property, owned by, or constructed or manufactured for the purpose of leasing to, the Federal Government."(318) EPA uses a "property-based definition" for the term for hazardous waste issues, which "encompasses all contiguous land that is owned by a department, agency, or instrumentality of the United States."(319) The legislative history of the federal facilities provision in the Clean Water Act also supports a definition that includes federal permits to use public lands. The conference report states that section 313 applies to "every Federal agency with control over any activity or real property," and that "[e]vidence received in hearings disclosed many incidents of flagrant violations of air and water pollution requirements by Federal facilities and activities."(320) In addition, the report accompanying the House bill states that "[t]he lease of public lands by various Federal agencies to concessioners or other organizers for the purpose of providing regional facilities often results in serious sanitation problems."(321) In light of the OMB definition and the legislative history, it appears clear that the term "federal facility" should be read broadly to include federal control of real property, including federal permits issued for use of that property. 3. Federal Compliance to the Same Extent as Nongovernmental Entities Assuming that the federal facilities provision applies to federal permits for use of the public lands, perhaps the critical argument concerning whether to extend the duty under section 313 to livestock grazing--indeed, to nonpoint source pollution generally--involves the nature of the duty imposed. Section 313 requires the federal government to comply with all "requirements, administrative authority, and process and sanctions respecting the control and abatement of water pollution in the same manner, and to the same extent as any nongovernmental entity."(322) The obvious argument against extending section 313 to nonpoint source pollution is that private parties ("nongovernmental entit[ies]") have no obligations under the Clean Water Act with respect to nonpoint sources--and, therefore, neither can federal facilities. Section 301 of the Clean Water Act states that "[e]xcept as in compliance with this section and sections [302, 306, 307, 318, 402, and 404] of this title, the discharge of any pollutant by any person shall be unlawful."(323) Not only does the term "pollutant" refer exclusively to point sources of pollution, but each of the referenced sections in section 301 also only refer to pollutants--that is, point sources of pollution.(324) In the point source context, the government's obligation under section 313 is clear, because private parties must comply with the certification requirements of section 401 in order to obtain a section 402 permit to discharge pollutants into the navigable waters.(325) In the nonpoint source context, however, the argument is not so simple. The Ninth Circuit has stated with little discussion that the federal government must comply with all state water pollution control laws, including nonpoint source pollution.(326) One should also note that in some Ninth Circuit states--for example, Washington--state law regarding who must comply with nonpoint source requirements actually includes private parties.(327) While Ninth Circuit court precedent is highly beneficial to potential plaintiffs--even if court opinions have failed to fully explain their reasoning--there is a better reason than precedent for arguing that, in the nonpoint source context, the federal government is still required to abide state water pollution control laws even when private entities do not have such an obligation: statutory construction. The key is enforceability, considered in greater detail in Part II.D. Whereas citizen enforcement of point source violations under the Clean Water Act is available through the citizen suit provision,(328) citizens must enforce the nonpoint source requirements of section 313 via the APA because "there is no other adequate remedy" in the Clean Water Act.(329) In ONDA v. Dombeck, the Ninth Circuit again confirmed that citizens may not use the Clean Water Act's citizen suit provision to sue over nonpoint source water quality violations. The holding foreclosed any adequate remedy under the Clean Water Act against federal nonpoint sources in the Ninth Circuit. Therefore, the APA provides the only cause of action available to citizens to enforce section 313's nonpoint requirements. Since the APA applies only to federal administrative actions, the nongovernmental entities portion of the provision is simply not applicable in the nonpoint source context. Further, inclusion of the phrase "discharge or runoff" in section 313 would be meaningless if the section was not truly intended to require the government to follow water pollution control laws in both the point and nonpoint realms.(330) Courts have long recognized that statutory language should not be read in a way that would render certain provisions superfluous, unnecessary, or without meaning.(331) In addition, the phrase "to the same extent as any nongovernmental entity" means that federal facilities should be subject to both federal water pollution laws as well as state water pollution laws which are binding on private entities. To argue that the phrase means merely that federal facilities are subject only to any state water pollution laws binding on private entities is to deny any federal regulation of federal government facilities and actions, in the nonpoint source, even if such facilities and actions fall under federal law, and are reviewable in federal court. Such a result is counterintuitive to ordinary statutory construction, requires far too much inference and assumption on the part of a reviewing court, and cuts against the goals of the Clean Water Act generally and section 313's federal facilities provision specifically.(332) 4. Waiver of Sovereign Immunity With Respect to the Federal Government's Duty The issue of sovereign immunity arises twice in the section 313 theory: once with respect to the federal government's duty to adhere to water quality standards, and again regarding the enforceability of that duty, as discussed in Part II.D.3. Simply stated, the question central to both is whether Congress has waived the federal government's sovereign immunity to prosecution under the Clean Water Act. The Supreme Court has stated that the sovereign immunity doctrine simply means that "the United States cannot be sued at all without the consent of Congress."(333) A "necessary corollary" to this idea, according to the Court, is that a congressional waiver of sovereign immunity should be strictly construed in favor of the federal government if any doubt exists as to that waiver.(334) "[A] waiver of the traditional sovereign immunity `cannot be implied but must be unequivocally expressed'" by Congress.(335) Because the doctrine's history and application are often confused and contradictory, Congress has frequently expressly revoked the federal government's immunity in a variety of statutes.(336) Some commentators have suggested that this practice has "more firmly embedded the sovereign immunity doctrine in American jurisprudence by implying that Congress has the power to waive the judicially created immunity."(337) The issue of interpretation of waivers of sovereign immunity arose early on in the history of the Clean Water Act and other environmental statutes in the mid-1970s. Within a year of a Supreme Court decision holding that federal facilities need not comply with state NPDES permit requirements, Congress amended section 313 in order to reverse the decision.(338) In EPA v. California ex tel. State Water Resources Control Board, the Court stated that even though the federal government must conform to state water quality standards, section 313 did not contain a "clear and unambiguous" waiver of sovereign immunity sufficient to require the federal government to obtain a state permit.(339) At that time, section 313 required federal installations to "comply with Federal, State, interstate and local requirements respecting control and abatement of pollution to the same extent that any person is subject to such requirements."(340) As a result of the ensuing 1977 amendment, the Clean Water Act appeared to subject federal facilities to both procedural and substantive requirements of state water quality control law.(341) Congress accomplished this by adding several key phrases: "[federal facilities shall] comply with, all Federal, State, interstate, and local requirements, administrative authority, and process and sanctions respecting the control and abatement of water pollution in the same manner, and to the same extent as any nongovernmental entity."(342) A recent Clean Air Act case in the Sixth Circuit clarifies this waiver of sovereign immunity with respect to the government's duty under section 313.(343) In that case, the Tennessee Air Quality Control Board assessed a $2500 civil penalty against the United States Army for failure to comply with certain state air pollution regulations.(344) The Sixth Circuit considered the extent to which the federal government had waived its sovereign immunity via the Clean Air Act's federal facilities provision (section 118) and its citizen suit provision (section 304).(345) The court found that the citizen suit provision, which states that "[n]othing in this section or in any other law of the United States shall be construed to prohibit, exclude, or restrict any State, local, or interstate authority[,]"(346) clearly includes the law of sovereign immunity in the phrase "any other law."(347) Turning to the federal facilities provision in section 118(a), the court also found a waiver of sovereign immunity in that provision, which is nearly identical to the Clean Water Act's section 313.(348) The court stated that the reference to the federal facilities provision in the citizen suit provision "simply reminds the reader that [section 118] defines the United States' burden to comply with state laws; it is [section 304] that expansively and unambiguously removes any impediment to enforcement in the event of noncompliance."(349) Thus, in a statutory scheme essentially identical to that in the Clean Water Act (recall that Congress largely incorporated the Clean Water Act's federal facilities provision from the Clean Air Act's provision which was drafted two years earlier),(350) the Sixth Circuit correctly explained that the federal facilities provision places a clear duty of compliance on the federal government. There is a counter-argument that, as this duty with respect to nonpoint source pollution is reviewable under the APA, the Clean Water Act is an "other statute" under APA section 702, and therefore it limits APA review.(351) The limit that the Clean Water Act thus imposes on APA review is that federal facilities nonpoint sources are only subject to water quality requirements "to the same extent as any nongovernmental entit[ies]."(352) As discussed in Part II.C.3, this language does not impose a private entities limit in the nonpoint source context. Therefore, judicial review of the federal government's duty is available under the APA. 5. Who Is the Polluter? Next, there is the issue of who does the actual polluting. In reality, the actual polluter is the private individual who is grazing livestock on the public lands, rather than the federal government itself. Despite this technical argument, the federal agency (generally the BLM) issues the actual permit, allowing the individual to engage in the activity. Section 313 is phrased in broad language that covers not only actual polluting by federal employees and facilities (for example, actual building and vehicles) on federal lands, but also federal activities that directly affect federal lands--not the least of which are permit issuances for actual use of the public lands.(353) While citizens cannot use section 313 of the Clean Water Act to sue the rancher whose livestock are causing water pollution on federal lands, citizens can sue the government agency that issued the permit. 6. Past Cases' Subject Matter Finally, a thoroughly unconvincing argument against extending section 313 to the grazing context is a simple subject matter defense. Each of the Ninth Circuit's previous holdings that affirmed plaintiffs' right to bring a nonpoint source section 313 claim under the APA have dealt with either timber sales or road building.(354) This is generally tree outside of the Ninth Circuit as well.(355) Nevertheless, the underlying subject matter of each of those cases does affect their outcome. Rather, it is nonpoint source pollution on federal lands by federal agencies that is determinative of whether section 313 is invoked. The federal government must adhere to state water quality laws regardless of what federal activity might cause a violation. D. Enforceability of the Federal Duty Under Section 313: A Cause of Action Under the APA Having established a clear duty on the part of the federal government, the second step in a section 313 action is to ensure that the duty is enforceable by citizens. It is clear, as evidenced in ONDA v. Dombeck, that the Clean Water Act's citizen suit provision does not permit citizens to sue for nonpoint source violations.(356) However, the Ninth Circuit line of cases from Northwest Indian Cemetery Protective Association to Idaho Sporting Congress clearly states that plaintiffs may sue under the APA to enforce nonpoint source violations via section 313's federal facilities provision.(357) The APA provides that "[a] person suffering legal wrong because of agency action, or adversely affected or aggrieved by agency action within the meaning of a relevant statute, is entitled to judicial review thereof."(358) Agency action is reviewable when "there is no other adequate remedy,"(359) which is the case for nonpoint sources under the Clean Water Act. 1. Ninth Circuit Case Law Establishing a Cause of Action Under the APA In ONRC v. USFS, the plaintiffs claimed that the district court erred in concluding that they could not bring an action under the citizen suit provision of the Clean Water Act because they had failed to provide a sixty-day notice of intent to sue.(360) ONRC argued that, in fact, their suit was brought under the judicial review provisions of the APA.(361) The Ninth Circuit first examined the citizen suit provision, which permits a citizen suit "against any person ... who is alleged to be in violation of ... an effluent standard or limitation under this chapter...."(362) Plaintiffs conceded that the definition of an "effluent standard" refers specifically to point source discharges, which were not at issue in the case.(363) Dismissing the government's only argument against judicial review under the APA (i.e., that it is impliedly precluded by inclusion of the citizen suit provision),(364) the court held that "judicial review is available under the APA."(365) A few years later in Marble Mountain Audubon Society, the Ninth Circuit again examined the relationship between section 313 and the APA.(366) Citing ONRC v. USFS, the court reaffirmed that the Forest Service must comply with all state water quality standards, and that the APA's judicial review provision "permits private citizens to sue for alleged state water quality control violations from nonpoint sources."(367) The district court granted the Forest Service's summary judgment motion on the grounds that "1) the [state] Water Quality Control Board did not criticize the Draft EIS's application of [the state] plan's water quality objectives; and 2) although the plaintiffs claimed that the project did not meet state water quality objectives, they did not challenge the Forest Service's interpretation and application of the objectives."(368) However, because the APA allows citizens to sue for nonpoint source violations, the Ninth Circuit court found that the plaintiffs properly challenged the final EIS's interpretation of state water quality standards.(369) The Ninth Circuit therefore remanded the case to the district court.(370) Finally, by holding that the section 313 requirement that all federal agencies comply with state water quality standards includes a state's antidegradation policy, the Ninth Circuit in 1998 again confirmed without discussion that judicial review of that requirement is available under the APA.(371) Despite the seemingly clear state of the law in the Ninth Circuit, one recent anomalous case from the Western District of Washington appears to go against this clear and binding precedent.(372) In response to a claim that its BMPs regarding road maintenance on the Olympic National Forest were likely to violate the state antidegradation policy, the Forest Service asserted that regardless of what the agency's obligations under the Clean Water Act, the Tribe's action was barred because there was no private right of action--thus barring any action by the plaintiff based on the state water quality standards.(373) As alluded to in the context of the government's duty, the Forest Service argued that there is no private right of action against it because the Washington Water Pollution Control Act provides no private right of action and the state water quality standards are applicable to the Forest Service only as they would be against any other private entity.(374) The Tribe correctly argued that APA reviewability of agency action allows a cause of action for private actors even though they may lack a cause of action under state law.(375) Nevertheless, the district court, rejecting the Ninth Circuit's established (and certainly binding) holding on the issue, proceeded to rule in favor of the Forest Service's interpretation of the law.(376) Finding that the water quality violations were still actionable under the Forest Plan, the court offered its explanation as to why the previous Ninth Circuit decisions "have not questioned whether Clean Water Act violations can be brought under the APA; the challenged agency action was required to follow the Clean Water Act and so by challenging the action's failure to follow some other requirement which incorporated the Act, the Clean Water Act was applied"(377) Not only is this court's interpretation of the relationship between section 313 and the APA incorrect, but it also contradicts clearly settled Ninth Circuit law Therefore, it seems that, at least in the Ninth Circuit, the duty imposed under section 313 is enforceable by citizen plaintiffs under the APA. While this may be the current state of the law--even if binding only in the Ninth Circuit--the issue still is not foreclosed Two important questions remain unresolved: First, does the citizen suit provision implicitly cut off review of section 313? Second, is there a sovereign immunity problem in section 313 with respect to enforceability? 2. The Clean Water Act Citizen Suit Provision Does Not Impliedly Cut Off APA Review One potential counter-argument to the assertion that citizens may obtain judicial review of section 313 under the APA is that Congress impliedly cut off such review by including the citizen suit provision in the Clean Water Act. However, the Ninth Circuit discredited this argument in ONRC v. USFS.(378) In that case, the Forest Service relied upon a Supreme Court decision holding that an implied right of action did not exist where plaintiffs failed to give a sixty-day notice to sue, and that Congress could not have intended to preserve a Section 1983(379) right of action when it had created so many specific statutory remedies.(380) The Forest Service argued that this holding indicated that the citizen suit provision is the sole means by which an individual may enforce the Clean Water Act.(381) The Ninth Circuit rejected this argument for three reasons First, because citizens cannot use the citizen suit provision to enforce state water quality standards for nonpoint sources (the citizen suit provision refers only to provisions in the Act relating to point source pollution), they "have no exclusive and comprehensive remedy in the citizen suit provision of the Act, as did the plaintiffs in Sea Clammers."(382) Second, because the Supreme Court in Abbott Laboratories v. Gardener(383) established a presumption of judicial review absent clear and convincing congressional evidence to the contrary, Congress could not have intended to cut off review by merely adding a section to the Clean Water Act dealing specifically with nonpoint sources (section 319). Instead, Congress only "intended to protect the interests of persons aggrieved by nonpoint source violations of state water quality standards."(384) The Ninth Circuit also noted that the "savings clause" in the citizen suit provision preserves rights created by separate statutes such as the APA.(385) Finally, the court's decision was buttressed by precedent in Northwest Indian Cemetery Protection Association,(386) where the Ninth Circuit allowed citizens to sue the Forest Service for violations of water quality standards.(387) Therefore, the issue appears to be settled under the ONRC v. USFS court's clear and correct rejection of the theory of implied restriction of judicial review in section 313. In addition, the government did not raise this argument in the subsequent Ninth Circuit cases.(388) 3. Waiver of Sovereign Immunity by the Federal Government A second potential counter-argument on the enforceability issue is the argument that Congress has not clearly waived the federal government's sovereign immunity to prosecution under the Clean Water Act. In the point source context, "several courts have held that section 313 ... constitutes a waiver of sovereign immunity and authorizes the imposition of civil penalties against [violating] federal agencies.(389) In 1992, however, the Supreme Court took an extremely restrictive interpretation of congressional waiver with respect to civil fines in the Clean Water Act and RCRA context by reversing the Sixth Circuit's affirmation of waivers of sovereign immunity in those two statutes.(390) In United States Department of Energy v. Ohio, the state originally alleged violations under the Clean Water Act, CERCLA, and RCRA at a federal uranium-processing plant in Ohio.(391) Ohio claimed that the defendant federal facility improperly disposed of hazardous wastes; released radioactive materials into the air, water, and soil; and polluted surface and ground water with chemical and radioactive contaminants in violation of the effluent limitations contained in its permit.(392) The Supreme Court granted certiorari solely to decide whether Congress waived sovereign immunity from liability for civil fines imposed by a State for past violations of the Clean Water Act and RCRA.(393) In an opinion described by a dissenting Justice White as "analytic gymnastics" and "ingenuity to create ambiguity,"(394) the majority held that Congress did not do so in either statute.(395) Congress responded immediately by making RCRA's waiver "as clear and unambiguous as humanly possible"(396) in the Federal Facilities Compliance Act of 1992 (FFCA).(397) While the provision still looks very similar to the Clean Water Act's federal facilities provision, it is quite clear with respect to the waiver of sovereign immunity, and it also elaborates that the government must adhere to requirements that are "both substantive and procedural (including any requirement for permits or reporting or any provisions for injunctive relief and such sanctions as may imposed by a court to enforce such relief)."(398) In 1993, Congress proposed, but has not yet passed, similar amendments to section 313.(399) Notwithstanding this strict interpretation of Clean Water Act sovereign immunity with respect to civil fines, matters are significantly different in terms of enforcing the duty in the nonpoint source context.(400) Here, because plaintiffs must sue under the APA rather than use the Clean Water Act's citizen suit provision, the issue of a waiver of sovereign immunity does not pose a barrier to litigation. Again, the Sixth Circuit's analysis in United States v. Tennessee Air Pollution Control Board(401) clarifies this point. As noted above, that court stated that the Clean Air Act's reference to the federal facilities provision in the citizen suit provision "simply reminds the reader that [section 118, the federal facilities provision,] defines the United States' burden to comply with state laws; it is [section 304, the citizen suit provision,] that expansively and unambiguously removes any impediment to enforcement in the event of noncompliance."(402) Further, stating that the federal facilities provisions of the Clean Air Act and the Clean Water Act may not be read in isolation, the court noted that, "[s]ignificantly, as we have said, the Clean Water Act contains no counterpart to [section 304]," which allows states to bring any enforcement action or obtain any judicial or administrative remedy or sanction.(403) Thus, while the Clean Air Act has its federal facilities provision to impose a duty upon the federal government, and its citizen suit provision to provide enforceability of that duty, the Clean Water Act only contains the first part of that equation in the nonpoint source context Because plaintiffs must use the APA to enforce the federal facilities provision with respect to nonpoint sources, the issue of sovereign immunity waiver is inapplicable for nonpoint source enforceability. The APA's right of review provision, section 702, states that judicial review is available for any "person suffering legal wrong because of agency action, or adversely affected or aggrieved by agency action within the meaning of a relevant statute."(404) Amended in 1976, section 702 eliminates the federal government's sovereign immunity for most purposes: An action in a court of the United States seeking relief other than money damages and stating a claim that an agency ... acted or failed to act in an official capacity or under color of legal authority shall not be dismissed or relief therein be denied on the ground that it is against the United States or that the United States is an indispensable party. The United States may be named as a defendant in any such action, and a judgment or decree may be entered against the United States.(405) The only limits on this otherwise complete waiver of sovereign immunity under the APA are 1) the notion that the provision only applies to suits for "other than money damages," and 2) the restriction that specific statutes limiting the waiver of sovereign immunity prevail over the more general APA waiver.(406) Neither of these exceptions limits an APA suit alleging Clean Water Act section 313 nonpoint source pollution violations. If the APA cannot be used to expand a limited waiver of sovereign immunity in some "other statute,"(407) the question becomes what limitations there are on section 313's waiver of sovereign immunity. In section 313, the waiver of sovereign immunity for federal facilities is limited to "Federal, State, interstate, and local requirements, administrative authority, and processes and sanctions respecting the control and abatement of water pollution in the same manner, and to the same extent as any nongovernmental entity."(408) Because the "to the same extent as any nongovernmental entity" phrase only works to define federal facilities' point source requirements,(409) the sovereign immunity limitation in section 313 does not limit judicial review, for example, only to state nonpoint source management program requirements, such as section 319 BMPs. To argue that section 313 prohibits APA review of federal facilities nonpoint sources--and such an argument must assume that nongovernmental entities are not subject to any nonpoint source requirements under the Clean Water Act, federal facilities are therefore exempt from federal requirements--is particularly problematic because such a scheme would eliminate both a federal cause of action as well as a federal court forum for a suit against the federal government.(410) Indeed, one commentator writes "absent clearer language than is present in section 313, no court should hold that Congress waived the federal government's sovereign immunity for state judicial processes while keeping that immunity intact to bar identical suits in federal courts."(411) Therefore, the Ninth Circuit is correct in holding that section 313 federal facility nonpoint source pollution is indeed reviewable under the APA. Moreover, because a clear duty exists for the federal government to comply with all pollution control and abatement laws, compliance with that duty is thus reviewable under the APA. E. Conclusion: The Section 313 Federal Facilities Provision and Nonpoint Source Pollution In short, section 313 imposes a clear duty on the federal government that its facilities and activities--including the issuance of federal permits for livestock grazing on public lands--must adhere to state water pollution laws, including water quality standards. In addition to clear application to nonpoint sources, the provision is enforceable under the judicial review provisions of the APA. While the TMDL strategy discussed in the first half of this Chapter provides a steady, though long-term, strategy to address grazing-related nonpoint source pollution, the Clean Water Act's section 313 appears to provide a short-term, case-specific approach. Armed with a clear federal obligation and the means to enforce the obligation, environmental plaintiffs should be able to advance the goals of the Clean Water Act considerably with this strategy. IV. OTHER THEORIES The TMDL and section 313 strategies appear to hold the most promise at this time in addressing post-ONDA v. Dombeck grazing-related water pollution on public lands. However, there are at least a few other approaches that should be mentioned. In the non-Clean Water Act context, the Wild and Scenic Rivers Act is beginning to enjoy considerable success.(412) Of course, use of this statute is limited to the impacts of grazing within designated Wild and Scenic River corridors. At this time, 160 river segments have been designated in thirty-three states.(413) In addition, the Coastal Zone Management Act (CZMA), similarly limited in scope, was amended in 1990 to require states to adopt programs to control nonpoint sources of coastal water pollution(414) Perhaps the only other strategy under the Clean Water Act involves a scattering of court decisions interpreting agricultural operations as point sources.(415) Any application of these cases would likely be useful only in very discrete circumstances. For example, in 1994, the Second Circuit held that the liquid manure spreading operations of a large dairy farm in New York were a "point source within the meaning of Clean Water Act section [502(14)]."(416) Because the farm fell within the definition of a concentrated animal feeding operation (CAFO), it was specifically included in the Clean Water Act definition of a point source(417) and was not subject to the agricultural exemption of section 502(14).(418) In Southview Farm, a group of local landowners filed a Clean Water Act citizen suit, claiming that the farm's liquid manure spreading operations violated the Act when the liquid manure being applied to fields using a central pivot irrigation system, a hose piping system, and conventional spreaders pulled by tractors, overflowed its intended area on several occasions.(419) The first violation on appeal concerned an incident in which the liquid manure collected in a swale and flowed through a pipe in a stonewall to a ditch that eventually led to a river.(420) The court stated that "the swale coupled with the pipe under the stonewall leading into the ditch that leads into the stream was in and of itself a point source. As this court has previously noted, the definition of a point source is to be broadly interpreted."(421) In the alternative, the court stated that "the manure spreading vehicles themselves were point sources."(422) At least one commentator argues that the Southview Farm reasoning might be extended to areas, such as watering troughs ("sacrifice areas") or stream access points ("water gaps"), where livestock congregate and severely trample the vegetation surrounding water sources(423) While the CAFO exception is clearly not applicable for most, if not all, BLM grazing activities, there may be a small window of opportunity--given an appropriate situation--to work within the confining parameters set forth by the Ninth Circuit in ONDA v. Dombeck and still address some public lands grazing activities as point sources. V. CONCLUSION It is hoped that this Chapter will contribute significantly to legally addressing nonpoint source pollution. Although some of these strategies have been posited in a paragraph here or there in a small handful of previous articles, no article has combined these strategies in one place, or explored the section 313 theory in any depth. This is particularly true with respect to the application of these strategies to grazing-related water pollution. Finally, it is important to note that although this Chapter takes as its starting point water pollution in the livestock grazing context, the legal strategies discussed are generally equally applicable to other nonpoint sources of pollution on the public lands, such as activities related to timber harvesting or road-building. As is most often the case with any issue in environmental and natural resources law, the true key to developing effective legal controls over grazing-related water pollution on the public lands may well be a combination of several strategies. The half-political, half-legal, long-term effectiveness of the TMDL program combines well with the more immediate potential of section 313's requirements on federal agencies. Ultimately, the two strategies should provide a positive feedback mechanism through which the TMDL process can slowly but surely force states to develop adequate lists of waters, sufficient load allocations, and effective implementation. At the same time, section 313 suits force federal agencies to comply with those state water quality standards on an immediate, site-specific basis. In maintaining pressure through a wide variety of legal strategies, the movement to control nonpoint sources of water pollution on the public lands, especially from livestock grazing, will continue to see considerable success in the short- and long-term future. (1) 172 F.3d 1092 (9th Cir. 1998), cert. denied, 120 S. Ct. 397, 1999 WL 552731 (1999). (2) Federal Water Pollution Control Act, 33 U.S.C. [subsections] 1251-1387, 1341 (1994 & Supp. III 1997). (3) ONDA v. Dombeck, 172 F.3d at 1094. (4) Section 401 provides that: Any applicant for a Federal license or permit to conduct any activity ... which may result in any discharge into the navigable waters, shall provide the licensing or permitting agency a certification from the State in which the discharge originates ... that any such discharge will comply with the applicable provisions of sections 1311, 1312, 1313, 1316, and 1317 of this title.... No license or permit shall be granted until the certification required by this section has been obtained or has been waived. 33 U.S.C. [sections] 1341(a)(1) (1994). The court determined that each of the sections referenced in [sections] 401 relates to regulation of point sources. 172 F.3d at 1097. (5) 172 F.3d at 1097. The Clean Water Act defines the term "point source" as "any discernable, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged." 33 U.S.C. [sections] 1362(14) (1994 & Supp. III 1997). The definition also states that "[t]his term does not include agricultural stormwater discharges and return flows from irrigation agriculture." Id. The Ninth Circuit agreed with the Second Circuit that "the term `point source' does not include a human being or any other animal." 172 F.3d at 1098 (citing United States v. Plaza Health Labs, Inc., 3 F.3d 643, 649 (2nd Cir. 1903) (human beings are not point sources subject to criminal liability under the Clean Water Act)). (6) The Clean Water Act does not specifically define the term "nonpoint source," although it does appear in the Act. 33 U.S.C. [sections] 1288(b)(2)(F) (1994) ("Any [areawide wastewater treatment management plan] ... shall include, but not be limited to a process to (i) identify, if appropriate, agriculturally and silviculturally related nonpoint sources of pollution...."); Id. [sections] 1329 (1994 & Supp. III 1997) (titled "Nonpoint source management programs"). The Environmental Protection Agency (EPA) defines nonpoint source pollution as pollution "caused by diffuse sources that are not regulated as point sources and normally [are] associated with agricultural, silvicultural and urban runoff." ENVIRONMENTAL PROTECTION AGENCY, NONPOINT SOURCE GUIDANCE (1987), reprinted in ENVIRONMENTAL LAW INSTITUTE, CLEAN WATER DESKBOOK 177 (1991). See also Alia S. Miles, Searching for the Definition of "Discharge"; Section 401 of the Clean Water Act, 28 ENVTL. L. 191, 196-98 (1998) (discussing definitions of point and nonpoint sources of pollution). (7) For an exhaustive review of the science underlying grazing's myriad environmental harms, see A.J. Belsky et al., Survey of Livestock Influences on Stream and Riparian Ecosystems in the Western United States, 54 J. SOIL & WATER CONS. 419 (1999). See a/so A. Joy Belsky and Dana M. Blumenthal, Effects of Livestock Grazing on Stand Dynamics and Soils in Upland Forests of the Interior West, 11 CONS. BIOL. 315 (1997); LYNN B. JACOBS, WASTE OF THE WEST: PUBLIC LANDS RANCHING (1991); JOHN NORNING, GRAZING TO EXTINCTION: ENDANGERED, THREATENED, AND CANDIDATE SPECIES IMPERILED BY LIVESTOCK GRAZING ON WESTERN PUBLIC LANDS (1994). (8) See CHARLES F. WILKINSON, CROSSING THE NEXT MERIDIAN: LAND, WATER, AND THE FUTURE OF THE WEST 81 (1992). (9) Joseph M. Feller, `Til the Cows Come Home: The Fatal Flaw in the Clinton Administration's Public Lands Grazing Policy, 25 ENVTL. L. 703, 703-04 (1995) (citing Bureau of Land Management, U.S. Dep't of the Interior, Rangeland Reform 1994 Draft Environmental Impact Statement at 3-5 (1994)). Grazing is authorized on about 159 million acres (approximately 90%) of BLM lands in the western United States and about 85 million acres (approximately 60%) of the national forests in the lower forty-eight states. (10) For an engaging legal and environmental history of public lands ranching in the American West, see WILKINSON, supra note 8, at 80-92. A detailed history and analysis of early public rangeland management is also found in George Cameron Coggins & Margaret Lindberg-Johnson, The Law of Public Rangeland Management II: The Commons and the Taylor Act, 13 ENVTL. L. 1 (1982). Also, note a recently published and excellent environmental history of Oregon. WILLIAM G. ROBBINS, LANDSCAPES OF PROMISE: THE OREGON STORY 18(g)-1940 (1997). (11) 43 U.S.C. [subsections] 161-284 (repealed 1976) (as amended). (12) 43 U.S.C. [subsections] 321-339 (1994) (as amended). (13) Buford v. Houtz, 133 U.S. 320 (1890). (14) For leading explanations of the "commons" situation (also referred to as "the fisherman's problem"), in which individuals using common resources are motivated by a serf-serving desire to use as much of a resource as possible as quickly as possible in order to gain the maximum economic benefit, knowing that one's competitors are doing the same, see Garrett Hardin, The Tragedy of the Commons, 162 SCIENCE 1243 (1968); ARTHUR F. McEVOY, THE FISHERMAN'S PROBLEM (1986); Jeffrey C. Ellis, On the Search for a Root Cause: Essentialist Tendencies in Environmental Discourse, in UNCOMMON GROUND: RETHINKING THE HUMAN PLACE IN NATURE 256, 258-59 (William Cronon ed. 1996). Hardin and Paul Ehrlich (author of THE POPULATION BOMB (1968)) were leading proponents of the idea that unchecked population growth was the root of all environmental problems. McEvoy's book is a more recent environmental history that details the commons theory in the context of fisheries problems on the California coast (15) Coggins & Lindberg-Johnson, supra note 10, at 46-47. (16) Taylor Grazing Act of 1934, 43 U.S.C. [sections] 315-315r (1994). (17) Taylor Grazing Act of 1934, pmbl., 48 Stat. 1269 (1934). (18) Congress initially authorized withdrawal of eighty million acres of public lands for grazing districts under the TGA. Taylor Grazing Act of 1934, ch. 865, [sections] 1, 48 Stat 1269 (1934). In 1936 Congress increased that acreage to 142 million acres, Act of June 26, 1936, ch. 842, Title [sections] 1, 49 Stat 1976, and in 1954 the limit was eliminated altogether. Act of May 28, 1954, ch. 243, [sections] 2, 68 Stat 151 (codified at 43 U.S.C. [sections] 315 (1976)). Because the congressional action only authorized withdrawal of the lands, it took two Executive Orders to actually withdraw the land from settlement Exec. Order No. 6910, 43 C.F.R. 297.11 (1938) (withdrawing unreserved and unappropriated public lands in Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, North Dakota, Oregon, South Dakota, Utah, and Wyoming); Exec. Order No. 6964, 43 C.F.R. 297.12 (1938) (withdrawing public lands in Alabama, Arkansas, Florida, Kansas, Louisiana, Michigan, Minnesota, Mississippi, Nebraska, Oklahoma, Washington, and Wisconsin). (19) 43 U.S.C. [sections] 315(b) (1994). (20) Federal Land Policy and Management Act of 1976, 43 U.S.C. [subsections] 1701-1785 (1994 & Supp. III 1997). (21) See generally Coggins et al., The Law of Public Rangeland Management IV: FLPMA, PRIA, and the Multiple Use Mandate, 14 ENVTL. L. 1 (1983) (describing the public rangeland scheme prescribed by FLPMA and PRIA and analyzing the BLM's implementation of that scheme). (22) For example: Wilderness Act of 1964, 16 U.S.C. [subsections] 1131-1138 (1994); National Environmental Policy Act of 1969, 42 U.S.C. [subsections] 4321--4370e (1994 & Supp. III 1997); Clean Air Act, 42 U.S.C. [subsections] 7401-7671q (1994 & Supp. III 1997); Federal Water Pollution Control Act, 33 U.S.C. [subsections] 1251-1387 (1994 & Supp. III 1997); Endangered Species Act of 1973, 16 U.S.C. [subsections] 1531-544 (1994); Resource Conservation and Recovery Act of 1976, 42 U.S.C. [subsections] 6901-6992k (1994 & Supp. III 1997) (amending Solid Waste Disposal Act, Pub. L. No. 89-272, 79 Stat. 992); National Forest Management Act of 1976 (NFMA), 16 U.S.C. [subsections] 472a, 521b, 1600, 1611-1614 (1994) (amending Forest and Rangeland Renewable Resources Planning Act of 1974, Pub. L No. 93-378, 88 Stat 476); Federal Land Policy and Management Act of 1976, 43 U.S.C. [subsections] 1701-1785 (1994 & Supp. III 1997); Comprehensive Environmental Response, Compensation, and Liability Act of 1980, 42 U.S.C. [subsections] 9601-0675 (1994 & Supp. III 1997). (23) Oregon Natural Desert Ass'n v. |
