explaining that “failure to implement the TMDL [established by the EPA] could imperil federal funding”Summary of this case from Food & Water Watch & Friends of the Earth v. U.S. Envtl. Prot. Agency
No. C 99-01828 WHA.
March 30, 2000.
Fredrick S. Levin, of Mayer, Brown Platt, Los Angeles, CA, for plaintiffs.
S. Randall Humm, United States Department of Justice, Washington, DC, for EPA.
ORDER ON CROSS-MOTIONS FOR SUMMARY JUDGMENT REGARDING AUTHORITY OF ENVIRONMENTAL PROTECTION AGENCY UNDER THE CLEAN WATER ACT TO LIST SUBSTANDARD RIVERS AND WATERS AND TO ISSUE TMDLS FOR THEM
In this case of first impression, the issue is whether Section 303(d) of the Federal Water Pollution Control Act Amendments of 1972, later renamed the Clean Water Act, authorized the Environmental Protection Agency to determine "total maximum daily loads" for rivers and waters polluted only by logging and agricultural runoff and/or other nonpoint sources rather than by any municipal sewer and/or industrial point sources. 33 U.S.C. § 1313(d). The issue gathers importance from the fact that "nonpoint source pollution has become the dominant water quality problem in the United States, dwarfing all other sources of volume. . . ." According to EPA, 54% of California's substandard rivers and waters are impaired by nonpoint sources only and another 45% are impaired by a combination of both point and nonpoint sources (EPA Tab 23).
Houck, TMDLs, Are We There Yet?: The Long Road Toward Water-Quality Based Regulation Under the Clean Water Act, 27 ELR 10391, 10399 (Aug. 1997).
Plaintiffs Guido and Betty Pronsolino own forested land along the Garcia River in the North Coast of California. When they obtained a permit to harvest timber, the California Department of Forestry ("CDF") imposed restrictions designed to reduce soil erosion into the Garcia River. The restrictions include measures such as leaving certain large conifers standing. Plaintiffs contend that the conditions are onerous and costly. They argue that CDF imposed these restrictions in order to implement a criterion known as a "total maximum daily load" ("TMDL") set by EPA for the Garcia River. Seeking to strike at the root of their problem, the Pronsolinos brought this action under the Administrative Procedure Act, 5 U.S.C. § 701 et seq., to challenge EPA's authority to impose TMDLs on rivers polluted only by timber-harvesting and agricultural runoff and/or other nonpoint sources, as is concededly the case for the Garcia River. Joining them as plaintiffs are the Mendocino County Farm Bureau, the California Farm Bureau and the American Farm Bureau Federation, all of whom dispute EPA's authority to set TMDLs for such rivers.
The conditions included the following: that the Pronsolinos: (a) inventory controllable sediment sources from all roads, landings, skid trails and agricultural facilities by June 1, 2002; (b) mitigate 90% of controllable sediment volume at "road related" inventoried sites by June 1, 2012; (c) prevent sediment loading caused by road construction; (d) retain five conifer trees greater than 32 inches in diameter at breast height ("dbh") per 100 feet of all Class I and Class II watercourses (if the site lacks enough trees to comply, the five largest trees per 100 feet must be retained); (e) harvest only during dry, rainless periods between May 1 and October 15; (f) refrain from constructing or using skid trails on slopes greater than 40% within 200 feet of a watercourse; and (g) forbear removing trees from certain unstable areas which have a potential to deliver sediment to a watercourse. These recommendations were incorporated into the Pronsolinos' NTMP (Joint Stmt. ¶ 32).
The Garcia River runs through southwestern Mendocino County into the Pacific Ocean. The river was once flourished as a spawning ground for cold-water fish such as coho salmon and steelhead trout. Excess sediment from logging operations over many years in the region hurt, perhaps severely, the spawning and reproduction of these fish in the Garcia River (and other North Coast rivers). In 1966, one journal reported that one-half of "potential coho salmon's habitat in the Garcia River . . . was reported as moderately to severely damaged by ongoing logging practices" (quoted in Brown, et al., Historical Decline Current Status of Coho Salmon in California, 14 No. Am. J. of Fisheries Management 237, 251 (May 1994)). By 1998, a staff report on the Garcia River by the California Regional Water Control Board stated that "[t]he Garcia River and its tributaries have experienced a reduction in the quality and amount of instream habitat that is capable of fully supporting the beneficial use of cold-water fishery, due to increased sedimentation" (Exh. C to Pacific Coast Federation Memorandum at 4). Prior to 1992, California established water-quality standards for the river that include protection of these fish and their habitat (EPA Tabs 8-9). Recent years have seen improvement in the Garcia River, but the restrictions imposed by CDF are intended to further restore the fish habitat.
Coho salmon were recently listed by the National Marine Fisheries Service as a threatened species. 61 Fed. Reg. 56138 (Oct. 31, 1996). The Service also recently proposed to list steelhead trout as threatened for the same region. 63 Fed. Reg. 13347 (Mar. 19, 1998).
Although Section 303(d) of the Clean Water Act required the states and EPA to identify certain substandard waters and to set TMDLs for them a generation ago, the Garcia River and other North Coast rivers escaped their gaze until recently. In 1992, EPA required California to add the Garcia River and sixteen other North Coast waters to its list of substandard waters. Thereafter, California retained the same waters on its list in 1994, 1996 and 1998. Meanwhile, a group of fishermen and environmental groups sued EPA, alleging that the then-recent addition of the Garcia River and sixteen other water segments to California's list of substandard waters meant that California and/or the EPA had to prepare TMDLs for the rivers. That case ended in a consent decree in March 1997 requiring TMDLs for all the rivers. Consent Decree, Pacific Coast Federation of Fishermen's Association v. Marcus, et al., No. 95-4474 MHP (Mar. 6, 1997).
Pursuant to the consent decree, EPA set March 16, 1998, as the deadline for the establishment of a TMDL for the Garcia River. California's North Coast Regional Water Quality Control Board initiated public comment on a draft TMDL but missed the deadline. EPA immediately released its own TMDL for the Garcia River (which was only slightly different from the state draft (Pl. Tab 25)). The EPA TMDL was sensitive to the fishhabitat problem (EPA Tab 1 at 8, 9 and 12):
Brown et al. (1994) reports that coho salmon previously occurred in as many as 582 California streams from the Smith River near the Oregon border to the San Lorenzo River on the central coast. There are now probably less than 5,000 native coho salmon spawning in California each year, many in populations of less than 100 individuals. Coho populations today are probably less than 6% of what they were in the 1940s and there has been at least 70% decline since the 1960s. Brown et al. (1994) conclude that the reasons for the decline of coho salmon in California include: stream alterations brought about by poor land-use practices and by the effects of periodic floods and drought, the breakdown of genetic integrity of native stocks, introduced diseases, over harvest, and climatic change.
* * * * * *
The Garcia River watershed has experienced a reduction in the quality and quantity of instream habitat which is capable of supporting the cold water fishery, particularly that of coho salmon and steelhead. Controllable factors contributing to this habitat loss include the acceleration of sediment production and delivery due to land management activities and the loss of instream channel structure necessary to maintain the system's capacity to efficiently store, sort and transport delivered sediment.
Overall, the TMDL for the Garcia River called for a sixty percent reduction of sediment (Joint Stmt. ¶ 15). The TMDL set the total maximum amount of sediment loading at an average of 552 tons per square mile per year and allocated portions of this total load to various categories of nonpoint sources in the Garcia River watershed (Joint Stmt. ¶ 12). The various categories of nonpoint sources were: (a) mass wasting associated with roads; (b) mass wasting associated with timber-harvesting activities; (c) erosion related to road surfaces; and (d) erosion related to road and skid trail crossings and gullies from diversions on roads and skid trails (Joint Stmt. at ¶ 16). In order to achieve these load allocations, the TMDL called for percentage reductions in sediment loading from these nonpoint sources (Joint Stmt. at ¶¶ 14, 15). There were only "slight differences" between the regional board's pending TMDL and the EPA's TMDL as issued (Pl. Tab 25).
EPA's regulations on TMDLs were issued in 1985. 40 C.F.R. § 130.7. The record herein contains a summary of the TMDL process (EPA Tab 7 at 2):
The TMDL process, in essence, is the following: States identify specific waters where problems exist or are expected; States set priorities; States allocate pollutant loadings among point and nonpoint sources; and EPA approves State actions or acts in lieu of the State if necessary. Point and nonpoint sources then reduce pollutants to achieve the pollutant loadings established by the TMDL through a wide variety of Federal, State, Tribal, and local authorities, programs, and initiatives.
States have primary responsibility for developing lists and TMDLs under section 303(d). Section 303(d)(1)(A) and the implementing regulations (at 40 C.F.R. § 130.7(b)) provide States with latitude to determine their own priorities for developing and implementing TMDLs. In particular, the flexibility to States offered by the priority ranking process of section 303(d)(1)(A) is a good opportunity for incorporating rotating basin or other watershed approaches into the TMDL process.
The regional board concluded that if it did not implement EPA's TMDL, then EPA could withdraw federal funding to the state agency. CDF, the state agency charged with approving timber-harvesting plans, such as those required of plaintiffs under state law, also believed that failure to implement the TMDL would imperil federal funding. In this connection, the Clean Water Act calls upon the states to incorporate whatever TMDLs are authorized for listed rivers and waters — the question here being whether a TMDL was authorized at all.
Plaintiffs' forester estimated that TMDL compliance would cost the Pronsolinos upwards of $750,000. Larry Mailliard and Bill Barr, members of plaintiff Mendocino County Farm Bureau, are similarly situated. They estimated their compliance would cost $10,602,000 and $962,000 respectively. This suit was filed on April 12, 1999, seeking a determination whether a TMDL for the Garcia River was authorized by the Clean Water Act.
The general issue presented is the extent to which logging and agricultural runoff and other nonpoint sources of pollution are relevant in the listing-and-TMDL process of Section 303(d) of the Clean Water Act of 1972. 33 U.S.C. § 1313(d). Put more narrowly, the issue is whether listing and TMDLs are required for rivers and waters polluted only by logging and agricultural runoff and/or other nonpoint sources, such as the Garcia River.
The landscape is illuminated by the events leading to the enactment. Under the Federal Water Pollution Control Act of 1948, 62 Stat. 1155, the primary responsibility for control rested with the states. In 1965, the Water Quality Act required each state to develop comprehensive water-quality standards for interstate waters, taking into account, among other factors, the "propagation of fish and wildlife." 79 Stat. 903. Such standards did not identify and directly regulate pollutants. Rather, they stated a desired condition of the water. Reasonable discharges were inherently permitted under these standards. See William H. Rogers, Jr., Environmental Law 252, 259-62 (2d ed. 1994). In 1966, however, the Refuse Act of 1899, 30 Stat. 1152, was reinvigorated by the Supreme Court. United States v. Standard Oil Co., 384 U.S. 224, 230, 86 S.Ct. 1427, 16 L.Ed.2d 492 (1966), held that all discharges of all foreign substances and pollutants (except those flowing from streets and sewers as liquids) were illegal without a permit from the Army Corps of Engineers. This set up a clash between the absolute prohibition of the Refuse Act and the reasonable-discharge approach of the standards. The Court of Appeals for the Third Circuit resolved this clash in favor of the Refuse Act in United States v. Pennsylvania Industrial Chemical Corp., 461 F.2d 468 (3d Cir. 1972). These developments led to the legislation in 1972 now known, together with subsequent amendments, as the Clean Water Act. The 1972 Act was provoked by a groundswell of sustained popular support for genuine water restoration and environmental legislation. The Act became law over a presidential veto.
The Supreme Court, although it has never faced the issue now presented, has consistently referred to the 1972 Act as intended "to establish an all-compassing program of water pollution regulation" and "to establish a comprehensive long-range policy for the elimination of water pollution." Repeatedly, the Supreme Court has used the word "comprehensive" to describe the Act. E.g., Milwaukee v. Illinois, 451 U.S. 304, 318-19, 101 S.Ct. 1784, 68 L.Ed.2d 114 (1981); Arkansas v. Oklahoma, 503 U.S. 91, 99, 112 S.Ct. 1046, 117 L.Ed.2d 239 (1992). The Act anticipated "a partnership" between the states and the federal government, "animated by a shared objective: `to restore and maintain the chemical, physical, and biological integrity of the Nation's waters.'" Id. at 101, 112 S.Ct. 1046. Achievement of "water quality which provides for the protection and propagation of fish, shellfish and wildlife" was an express statutory objective. 86 Stat. 816.
The 1972 Act represented a major shift in enforcement policy — away from primary reliance on water-quality standards and toward primary reliance on specific effluent limits on all point sources, the latter being any discernible, confined and discrete conveyance such as a pipe or ditch. 33 U.S.C. § 1362(14). The Act established the National Pollution Discharge Elimination System ("NPDES") and required an NPDES permit for any discharge by any point source into any navigable water of the United States, interstate or intrastate. The new strategy sought to force the best technology practicable or achievable on dischargers. By 1977, industry was required to meet effluent limitations achievable through "best practicable control technology currently available." By 1983, it was to achieve control levels based on the "best available technology economically available." 33 U.S.C. § 1311. Instead of solely working backwards from the water-quality standards to develop acceptable levels of effluent from point sources, the new lead strategy was to require point sources to employ state-of-the-art treatment, even if it led, as a happy circumstance, to even cleaner water than called for by the standards. EPA was to issue NPDES permits except to the extent states adopted EPA-approved NPDES programs.
The Structure of the 1972 Act
Although the technology-based strategy of effluent limitations on all point sources (the NPDES permit program) was its capstone, the 1972 Act nonetheless carried forward the pre-existing regime of water-quality standards and even extended that regime to all navigable waters of the United States, interstate or intrastate. The Act explicitly recognized the separate problems of point versus nonpoint pollution and established different approaches to mitigate them. Point sources were subjected to NPDES regulation (under Sections 301-02 and 402). Nonpoint sources were left subject to state regulation. How TMDLs were supposed to fit into both branches of the solution is the problem presented by this case.
In analyzing this issue, it is important to bear in mind the comprehensive way in which all sources of pollution were addressed by the 1972 Act, albeit in different ways. It is important also to focus on the language actually adopted in 1972 (rather than in later amendments) because the issue here turns on the meaning of the 1972 language. The NPDES program has already been described. The following review of the structure of the 1972 Act focuses on those provisions arguably relevant to TMDLs and/or nonpoint-source pollution.
Under Section 102(a) of the 1972 Act, EPA was to cooperate with other state and federal agencies, municipalities and industry to "prepare or develop comprehensive programs for preventing, reducing or climinating the pollution of the navigable waters," giving "due regard" to "the protection and propagation of fish and aquatic life and wildlife," among other purposes. 86 Stat. 817. Under Section 104(n) and (p), EPA was to promote "continuing comprehensive studies of the effects of pollution, including sedimentation in the estuaries and estuarine zones of the United States on fish and wildlife, on sport and commercial fishing . . ." ( 86 Stat. 823) and to "carry out a comprehensive study and research program to determine new and improved methods and the better application of existing methods of preventing, reducing, and eliminating pollution from agriculture. . . ." 86 Stat. 824. In sum, while these provisions did not reference TMDLs, they called for "comprehensive" programs and studies, including protection of fish and wildlife.
Section 201 authorized grants for "waste treatment management" on an "areawide" basis to "provide control or treatment of all point and nonpoint sources of pollution. . . ." § 201(c). In turn, Section 208 called for "areawide waste treatment management" planning by the states, expressly including plans for "nonpoint source" pollution. To that end, EPA was required to publish regulations guiding the identification of areas with "substantial water quality control problems." § 208(a)(1). The states were then to identify such areas. § 208(a)(2). Within one year thereof, responsible state organizations were called upon to "have in operation a continuing areawide waste treatment management planning process" with initial plans certified to EPA within two years. § 208(b)(1). Such plans were to include a number of components, the most germane of which was explicitly directed at "nonpoint source" pollution. Under Section 208(b)(2)(F), for example, the plans had to include:
a process to (i) identify, if appropriate, agriculturally and silviculturally related nonpoint sources of pollution, including runoff from manure disposal areas, and from land used for livestock and crop production, and (ii) set forth procedures and methods (including land use requirements) to control to the extent feasible such sources [emphasis added];
The Section 208 process was summarized succinctly by a 1981 textbook on nonpoint-source pollution as follows:
Specifically, Section 208 of the act calls for area-wide water pollution planning in areas designed by the governor of each state that would include both point and nonpoint sources and pollution abatement programs. The plans should include: (a) identification of the treatment works necessary to meet the anticipated municipal and industrial waste treatment needs of the area with associated construction priorities, time schedules, and the establishment of regulatory programs for such treatment works, including urban runoff and storm water; (b) identification of the sources of nonpoint pollution — agriculture (including runoff from irrigated fields), silviculture, runoff from land used for livestock and crop production or land that has had manure applied to it, mining, saltwater intrusion, waste disposal on lands, disposition of all residual waste generated in the designated area, and land and subsurface excavations; (c) setting forth of a procedure and methods (including land-use requirements) that feasibly will control such sources.
Novotny Chesters, Handbook of Nonpoint Pollution 19 (1981). In short, Section 208 contemplated that nonpoint sources would be remedied through state regulation and required the states to develop programs to do so.
Turning to Section 303, at the core of this controversy, its title was "Water Quality Standards and Implementation Plans." Subsection (a) required the states to adopt water-quality standards promptly, to the extent not previously done, and to carry forward those already adopted (subject to further EPA approval). Standards were to be set, as stated, for both interstate and intrastate waters. Subsection (c) imposed periodic updating of the standards and submission to EPA for review and approval. Standards were to take into account the unique needs of each waterway, including "propagation of fish and wildlife" as well as "agricultural . . . and other purposes." 86 Stat. 848. Subsection (b) instructed EPA to impose its own standards on any state failing to set standards. 86 Stat. 847. These standards, the Supreme Court has said (once again), were meant by Congress to be "comprehensive." PUD No. 1 of Jefferson County v. Washington Dep't of Ecology, 511 U.S. 700, 704, 114 S.Ct. 1900, 128 L.Ed.2d 716 (1994).
Significantly, in the process of setting standards, Section 303 did not exempt any rivers or waters — all were covered to the full extent of federal authority over navigable waters. Nor was any distinction drawn between point sources and nonpoint sources. The goal was to set standards for all navigable waterways in America, balanced and tailored to accommodate the various needs of each, including, explicitly, the need for the protection of fish and wildlife. The standards-setting process of Section 303 plainly applied to waters polluted by point sources as well as nonpoint sources, either alone or in combination. All parties agree on this conclusion.
It was onto this comprehensive standards regime that Congress imposed the requirement at issue, a requirement subdivided into a listing and a TMDL. As to the first, Section 303(d)(1)(A) provided:
Each state shall identify those waters within its boundaries for which the effluent limitations required by Section 301(b)(1)(A) and 301(b)(1)(B) are not stringent enough to implement any water quality standard applicable to such waters. The State shall establish a priority ranking for such waters, taking into account the severity of the pollution and the uses to be made of such waters.
Section 303(d) thus became an intersection between the old and new strategies. It called for an assessment of the expected beneficial impact of the main innovation of the Act — imposition of the best effluent reduction technology could supply. If those reductions alone would bring a waterway into compliance with standards, well and good. If not, then Section 303(d)(1) required the waterway to join a list of unfinished business. That list, once completed, then had to be prioritized by the states. For each listed river and water, Section 303(d)(1)(D) of the Act next required the states to establish TMDLs:
Each State shall establish for the waters identified in paragraph (1)(A) of this subsection, and in accordance with the priority ranking, the total daily maximum load, for those pollutants which the Administrator identifies under Section 304(a)(2) as suitable for such calculation. Such load shall be established at a level necessary to implement the applicable water quality standards with seasonal variations and a margin of safety which takes into account any lack of knowledge concerning the relationship between effluent limitations and water quality.
Professor Oliver Houck states: "The rationale was that water quality standards would clean up waters which remained substandard after application of technology-based limits." Houck, 27 ELR at 10337.
TMDLs were thus required for all listed rivers and waters, at least as to pollutants identified by EPA as suitable for such calculation (and EPA long ago stated that "all" pollutants were suitable for such calculation). The controversy herein is whether the Garcia River should have been listed at all. Plaintiffs say no, it should not have been and, therefore, EPA should never have issued its TMDL. Before addressing this argument, it is instructive to complete the remainder of the structural review.
43 Fed. Reg. 60662 (Dec. 28, 1978).
EPA's regulations call it a list of "water-quality" limited segments. 40 C.F.R. § 130.7(b).
The next step — under Section 303(d)(2) — was for the State to submit the prioritized list and TMDLs for EPA review and for EPA to either approve or disapprove them. To repeat, EPA was statutorily required to approve or disapprove the lists and the TMDLs. If approved, then the state was "to incorporate" the lists and TMDLs into its "continuing planning process" under Section 303(e). If disapproved, EPA was to revise the list and/or TMDLs so as to implement the applicable water-quality standards. As revised, the state was then "to incorporate" them into its planning under Section 303(e). Either way, the lists and TMDLs were obliged to be incorporated into the states' continuing planning process, a process summarized momentarily.
The Act does not define total maximum daily load. EPA's regulations break it into a "wasteload allocation" for point sources and a "load allocation" for nonpoint sources. 40 C.F.R. § 130.2.
As to nonlisted rivers and waters, Section 303(d)(3) imposed an "informational" TMDL requirement regarding fish and wildlife:
For the specific purpose of developing information, each State shall identify all waters within its boundaries which it has not identified under paragraph (1)(A) and (1)(B) of this subsection and estimate for such waters the total maximum daily load with seasonal variations and margins of safety, for those pollutants which the Administrator identifies under Section 304(a)(2) as suitable for such calculation and for thermal discharges, at a level that would assure protection and propagation of a balanced indigenous population of fish, shellfish and wildlife.
As stated, this provision applied only to nonlisted waters. The informational TMDLs were not subject to EPA review. EPA was not authorized to review or to issue the "informational" TMDLs. This provision sought establishment of state-prepared TMDLs that would "assure protection and propagation of a balanced indigenous population of fish, shellfish and wildlife." They, too, were intended for the continuing planning process.
Although the water-quality standards themselves were supposed to take "into consideration" the "propagation of fish and wildlife," those standards also had to take into account the "use and value" of the waters for a number of other purposes. 33 U.S.C. § 1313(c)(2). As a result, a standard itself might be a compromise of competing considerations. The TMDL requirement of Section 303(d)(3) would specifically help the states identify the extent and types of pollution injurious to fish and wildlife.
Section 303(e) imposed on the states "a continuing planning process" reviewable by EPA for consistency with the Act. EPA was to approve "any continuing planning process" that would result in "plans for all navigable waters within such state . . . which were to include, among other things, total maximum daily loads for pollutants in accordance with subsection (d)" (emphasis added). Subsection (d), in turn, covered both the TMDLs for listed waters and "informational" TMDLs. The plans also were required to include the "areawide waste management plans" under Section 208. In turn, as seen, those plans had to include "a process" to identify "nonpoint sources of pollution" and "methods (including land-use requirements) to control" them. The Section 303(e) plans were also expected to include "adequate implementation, including schedules of compliance, for revised or new water quality standards." In short, the mandatory planning process of Section 303(e) covered "all navigable waters" and was to address "adequate implementation" of all water-quality standards, had to include plans incorporating TMDLs, and had to address "nonpoint sources of pollution." A wild river, therefore, polluted only by logging in its watershed, was clearly meant to benefit from the continuing planning process.
Section 304 was entitled "Information and Guidance." Section 304(a) generally called for EPA-set "criteria for water quality accurately reflecting the latest scientific knowledge" including impacts on fish and wildlife. EPA was to develop "information"
(A) on the factors necessary to restore and maintain the chemical, physical, and biological integrity of all navigable waters, ground waters, waters of the contiguous zone, and the oceans; (B) on the factors necessary for the protection and propagation of shellfish, fish, and wildlife for classes and categories of receiving waters and to allow recreational activities in and on the water; and (C) on the measurement and classification of water quality; and (D) for the purposes of section 303, on and the identification of pollutants suitable for maximum daily load measurement correlated with the achievement of water quality objectives.
§ 304(a)(2) (emphasis added). In requiring "identification of pollutants suitable" for the TMDL measurements of Section 303, no distinction was drawn between point and nonpoint sources in Section 304(a). That provision was comprehensive. In contrast, Section 304(b) focused only on point sources. It required EPA to issue guidelines for effluent limitations. Section 304(e) focused only on "nonpoint sources of pollution." It required EPA to issue "(1) guidelines for identifying an evaluating the nature and extent of nonpoint sources of pollutants and (2) processes, procedures, and methods to control pollution resulting from . . . agriculture and silvicultural activities, including runoff from fields and crop and forest lands." 86 Stat. 852. Section 304(a), in short, covered all sources of pollutants and called for a predicate step for the TMDL process at issue.
Section 305, labeled "Water Quality Inventory," called for an EPA-led report by January 1, 1974 that described the "specific quality," during 1973, "of all navigable waters" that included a roster of all point sources discharging into said waters, and that identified "specifically those navigable waters, the quality of which is adequate to provide for the protection and propagation of shellfish, fish and wildlife," among other things. Starting in January 1975, each state was to submit an annual report to EPA that described the water quality of all "navigable waters" within its borders, analyzing the extent to which "all navigable waters" within the state provided for the protection of shellfish, fish and wildlife, analyzing the extent to which eliminating the discharge of pollutants and a level of water quality provided for the protection and propagation of shellfish, fish and wildlife, and among other things (§ 305(b)(1)(E), 86 Stat. 854):
a description of the nature and extent of nonpoint sources of pollutants, and recommendations as to the programs which must be undertaken to control each category of such sources, including an estimate of the costs of implementing such programs.
Section 305 made clear that no river or water in the United States was immune from its process; all navigable waters were covered; nonpoint-source pollution was to be analyzed without exception.
In summary, the 1972 Act addressed all sources of pollution, although each type in different ways. It sought comprehensively to protect and to restore all navigable waters in America. Although Congress imposed direct NPDES regulation only on point sources, Congress plainly carried forward the pre-existing regime of water-quality standards and, indeed, expanded it to include intrastate waters. That regime was intended (in part) to mitigate nonpoint-source pollution through state land-use regulation. These general conclusions are not really in dispute in this case.
Construction of Section 303(d)
The issue on which the parties divide is the extent to which nonpoint sources of pollution were to count in assembling the substandard-waters list required by Section 303(d) and in preparing the corresponding TMDLs. In their opening brief, plaintiffs contended that the listing and TMDL requirements of Section 303(d) were "exclusively reserved for point sources" (Pl.Br.14) and that "Section 303(d) focuses solely on point sources" (Pl. Br.16). In their reply (at 8), plaintiffs stated: "A water body that is impaired by both point and nonpoint sources should be listed under both Section 303(d) and Section 319(a)(1)(A) and the point and nonpoint sources be addressed pursuant to those respective listings." A water polluted only by logging runoff or other nonpoint sources of pollution, like the Garcia River, plaintiffs argue, should not be listed and no TMDL should be prepared. Plaintiffs base their arguments on the fact that effluent limitations — which apply only to point sources — are referenced in the listing requirement of Section 303(d) whereas no reference is made to nonpoint sources:
Each state shall identify those waters within its boundaries for which the effluent limitations required by Section 301(b)(1)(A) and 301(b)(1)(B) are not stringent enough to implement any water quality standard applicable to such waters. The State shall establish a priority ranking for such waters, taking into account the severity of the pollution and the uses to be made of such waters.
All versions of plaintiffs' arguments must be rejected for four reasons.
First, the sole import of placing a river or water on a Section 303(d) list was that it would trigger the TMDL requirement. What use, then, did the statute contemplate for the TMDL? If the TMDL, for example, were to be used only to adjust NPDES effluent limitations for point sources, then plaintiffs' argument might have force. Such a narrower use, although a legitimate one, was not set forth in the statute as the sole use. Indeed, that use was not even expressly called out in the Act, although it was inferrable from Section 301(b)(1)(C), Section 302(a), and Section 303(d)(1)(C). The expressly contemplated use of TMDLs was their "incorporation" into the "continuing planning process" by the states under Section 303(e). That was the side of the equation, however, pertinent to nonpoint-source regulation (as well as to any state-administered NPDES program). Moreover, the TMDLs had to be set at levels that would "implement" the applicable water-quality standards. It would have been impossible to do so without taking any nonpoint sources into account as well as any point sources. It seems evident that TMDLs were intended, in part, to be used to help states evaluate and develop land-management practices to mitigate nonpoint-source pollution. Otherwise, as one court has stated, it would frustrate the "comprehensive approach" adopted in the 1972 Act. Natural Resources Defense Council, Inc. v. Fox, 909 F. Supp. 153, 156 (S.D.N.Y. 1995). In short, the statutorily-defined role of the TMDL is inconsistent with plaintiffs' argument.
The states are allowed but not required to adopt state-administered NPDES programs. In the absence of such a program approved by the EPA, the NPDES program within a state is administered by EPA itself. See Milwaukee v. Illinois, 451 U.S. 304, 311, 101 S.Ct. 1784, 68 L.Ed.2d 114 (1981). Even today, after nearly thirty years, two states do not have approved NPDES programs.
Second, plaintiff's argument is inconsistent with the logic expressed in Section 303(d). Section 303 was entitled "Water Quality Standards And Implementation Plans." Water-quality standards were required for all navigable waters, intrastate or interstate. The first sentence of Section 303(d) required each state to "identify those waters within its boundaries" for which the new effluent limits would not be stringent enough to meet the standards. The starting point was, therefore, each and every substandard navigable water within the boundaries of the state. Then, only those redeemable through the imposition of state-of-the-art technology on point sources, the lead strategy under the Act, were expressly excused from the list. Since all rivers and waters regardless of pollution source were included in the universe for which water-quality standards were required, all of them — again regardless of source of pollution — were included in the universe for which listing and TMDLs were required — save and excluding only those for which effluent limitations would be sufficient to achieve compliance with standards. Third, while it is true that nonpoint-source pollution was not mentioned in Section 303(d), the reason seems obvious. The 1972 Act superimposed the technology-driven mandate of point-source effluent limitations. To assess the impact of the new strategy on the monumental clean-up task facing the nation, Congress called for a list of the unfinished business expected to remain even after application of the new cleanup strategy. In calling for such a list, it was unnecessary to reference nonpoint pollution. Any polluted waterway — whether its sources were point, nonpoint or a combination — had to be listed if it would not be cleansed by the new approach. To have excluded the large number of rivers and waters polluted solely by agricultural and logging runoff would have left a chasm in the otherwise "comprehensive" statutory scheme. It would have crippled the continuing planning process by which the states were expressly required to confront nonpoint-source pollution and to incorporate TMDL data into their continuing planning process. To achieve the standards, an intermediate step was needed. That step required engineering data. That was the role of the TMDL. Similarly, to have limited TMDLs only to point-source loadings, as argued for by plaintiffs, would have left state agencies guessing at how to allocate the burden of cleanup between point and nonpoint contributions of the same pollutant. Fourth, the Ninth Circuit has already gone on record that the TMDL process covers nonpoint as well as point sources, as set forth below. None of these decisions is four-square on point but some come close. Plaintiffs' argument is hard to reconcile with the Ninth Circuit's caselaw.
This was subject to the further proviso that TMDLs only had to be set for those "pollutants" identified by EPA under Section 304(a)(2), 33 U.S.C. § 1314(a)(2), as suitable for TMDL calculation, a subissue discussed in detail below.
For these three reasons, the Court finds that Congress has directly spoken to the precise question at issue. There is, therefore, no need to resort to supplemental aids of construction. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-45, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984).
The Ninth Circuit Caselaw
Although the Ninth Circuit has not decided the precise issue raised, its Clean Water Act precedents support the conclusion reached above. The earliest-cited case to reach the Ninth Circuit involved a challenge by environmentalists to NPDES permits issued by EPA to Alaska gold-placer mines. The court held that EPA had erred in failing to require an effluent limitation for turbidity in the permit. The court distinguished between nonpoint-source pollution (no NPDES permit required) and point-source pollution (NPDES permit required), but held that water discharged through a sluice box was a point source. The EPA was required, therefore, to include in the permits whatever effluent limitations were necessary to achieve state water-quality standards (under Section 301(b)(1)(C)). Trustees for Alaska v. EPA, 749 F.2d 549, 557-58 (9th Cir. 1984).
In Oregon Natural Resources Council v. United States Forest Serv., 834 F.2d 842 (9th Cir. 1987), the plaintiff organization argued that Section 301, which established effluent limitations for point sources, should incorporate limitations designed to achieve state standards for achieving elimination of nonpoint runoff. The Ninth Circuit rejected the attempt to confound the two separate ways in which the Act distinguished between point and nonpoint pollution:
We recognize that nonpoint sources of pollution constitute a major source of pollution in the nation's waters [footnote omitted]. However, we do not believe that the Act allows for the enforcement of state water quality standards, as affected by nonpoint sources, under the citizen suit provision. When Congress established the National Pollutant Discharge Elimination System (NPDES) in 1972 and concomitantly created a new approach to regulating and abating water pollution, it drew a distinct line between point and nonpoint pollution sources. Point sources are subject to direct federal regulation and enforcement under the Act [footnote omitted]. See 33 U.S.C. § 1342. Nonpoint sources, because of their very nature, are not regulated under the NPDES. Instead, Congress addressed nonpoint sources of pollution in a separate portion of the Act which encourages states to develop areawide waste treatment management plans [footnote omitted]. See 33 U.S.C. § 1288 (emphasis added).Oregon Natural Resources Council, 834 F.2d at 849. The court remanded, however, for a determination on whether the timber sale would cause nonpoint-source runoff in violation of the water-quality standards set by Oregon. In no way does the decision address the Section 303(d) list or TMDL features. But the decision does recognize that the 1972 Act comprehended nonpoint-source regulations through state areawide waste treatment management plans.
Of most immediate significance are the following two decisions. In Alaska Center for the Environment v. Browner, 20 F.3d 981 (9th Cir. 1994), the district court ordered EPA to issue TMDLs for Alaska waters, after a long period of inaction by the agency. EPA appealed only on grounds of plaintiffs' alleged lack of standing. The Ninth Circuit affirmed. EPA's standing argument involved the fact that even if TMDLs were issued, Alaska would have, at least in part, discretion "with respect to non-point source pollution." The Ninth Circuit responded:
Here, by contrast, third party involvement does not render the relief sought completely speculative. Congress and the EPA have already determined that establishing TMDLs is an effective tool for achieving water quality standards in waters impacted by non-point source pollution.Alaska Center, 20 F.3d at 985. Significantly, therefore, the Ninth Circuit went on record that the TMDL process covered nonpoint-source pollution. At the very least, this statement covered rivers and waters affected by both point sources and nonpoint sources. Nothing in the opinion so limited it, however, and the court's rationale seems equally applicable to rivers spoiled only by logging runoff.
In Dioxin/Organochlorine Center v. Clarke, 57 F.3d 1517 (9th Cir. 1995), the court of appeals upheld EPA's TMDL for dioxin, which had been set at the lowest level measurable by the current technology. The court of appeals described a TMDL as follows ( id. at 1520):
A TMDL defines the specified maximum amount of a pollutant which can be discharged or "loaded" into the waters at issue from all combined sources. Thus a TMDL represents the cumulative total of all "load allocations" which are in turn best estimates of the discrete loading attributed to nonpoint sources, natural background sources, and individual wasteload allocations ("WLAs"), that is, specific portions of the total load allocated to individual point sources. When a TMDL and specific wasteload allocations for point sources have been established, any NPDES permits issued to a point source must be consistent with the terms of the TMDL and WLA. See 40 C.F.R. § 130.2.
This decision also treated TMDLs as applicable to nonpoint sources. The court then went on to hold that Section 303(d) did not require development and proven failure of best available technology before setting a TMDL for a toxic pollutant. Id. at 1528.
In summary, the Ninth Circuit has already stated that TMDLs are an "effective tool for achieving water quality standards in waters impacted by non-point source pollution" ( Alaska Center) and that "[a] TMDL defines the specified maximum amount of a pollutant which can be discharged or `loaded' into the waters at issue from all combined sources" ( Dioxin). In the face of these statements, it would be difficult for a district court within the Ninth Circuit to hold that TMDLs were not required for listed rivers and waters harmed only by nonpoint pollution.
In Natural Resources Defense Council v. EPA, 915 F.2d 1314 (9th Cir. 1990), the Ninth Circuit held under a 1977 Amendment, that EPA regulations had to include a requirement that the states identify all sources discharging any pollutant believed to be impairing water quality into a listed river or water. The decision explained the broad outline of the 1972 Act and stated that the fact that only point-source pollution was directly regulated but otherwise did not address the immediate issue. Oregon Natural Desert Ass'n v. Dombeck, 172 F.3d 1092 (9th Cir. 1998), held that a grazing permit issued by the United States Forest Service did not require prior state certification of compliance with state water-quality standards, holding that such certifications were required only for "discharges," which referred only to point sources. 172 F.3d at 1095.
The Legislative HistorySection 303(d) originated in the House of Representatives. The House committee report, on which plaintiffs base their legislative-history argument, stated:
Water quality standards will be utilized for the purpose of setting effluent limitations in those cases where effluent limitations for point sources would not be consistent with such standards. Even though all point sources must be January 1, 1976, as a minimum, meet the requirements of subsection (b)(1)(A) and subsection (b)(1)(B) of section 301 all point sources could be required to meet a more stringent effluent limitation consistent with water quality standards of the receiving waters if the effluent limitations set pursuant to subjection (b)(1)(A) and subsection (b)(1)(B) of section 301 are inadequate to meet those water quality standards. In this case a more stringent effluent limitation will be imposed.
Any required more stringent effluent limitations will be set on the basis of that reduction in the quantity and quality of the discharge of pollutants which would be required to make the total discharge load in the receiving waters from municipal and industrial sources consistent with water quality standards. This should not be interpreted to mean that such more stringent industrial and municipal effluent limitations will, in themselves, bring about a meeting of water quality standards for receiving waters. The Committee clearly recognize that non-point sources of pollution are a major contribution to water quality problems.
The Committee heard extensive testimony during the oversight and legislative hearings to the effect that it is extremely difficult to apportion the discharge load from all point sources along a waterway or section of a waterway. However, testimony was also heard from the more experienced States that they already have this capability. The Committee feels that with appropriate support from the Administrator, the required analysis can be completed by the State in a timely fashion.
H.R. Rep. No. 92-911, at 105-06 (1972) (emphasis added).
From this, plaintiffs conclude that while Congress understood the role of nonpoint sources to be a "major contribution" to water-quality problems, it elected to regulate only point sources through Section 303(d) and described load calculations only as a predicate step in adjusting effluent limits.
To be sure, the focus of the passage was on effluent limitations and their adjustment to meet water-quality standards. Nothing in this passage, however, expressly limited the role of water-quality standards or load calculations to this single purpose. The passage also recognized that "non-point sources of pollution are a major contribution to water quality problems." The passage also seemed, therefore, to recognize that mitigation of nonpoint-source pollution would also be required to meet standards. In conference, moreover, Section 303(d)(1)(C) was amended to amplify the TMDL text, although not directly as to the specific issue presented. The conference report then stated:
(1) Subsection (d)(1) requires each State to identify the waters within its boundaries for which effluent limitations required to section 301 are not stringent enough to implement a water quality standard applicable to the waters. The State is to establish a priority ranking for such waters, taking into account the severity of the pollution and uses to be made of the water.
* * * * * *
(3) Each State is to establish for waters identified under paragraph (1)(A) in accordance with the priority ranking the total maximum daily load for those pollutants which the Administrator identifies as suitable for such calculation. This is to be established at a level necessary to implement water quality standards with seasonal variations and a margin of safety.
* * * * * *
(5) The State is to submit to the Administrator from time to time the waters so identified and loads so established. The Administrator is to approve or disapprove the identification and load within 30 days after submission. If they are approved, the State must incorporate them into its plan under subsection (e). If he disapproves them, he is required to identify the waters and establish the loads, and the State is to incorporate that into its current plan.
* * * * * *
(7) Each State is required to have a continuing planning process consistent with this Act and to submit such plan within 100 days after the date of enactment of this Act to the Administrator for his approval. The Administrator must approve or disapprove such process within 30 days after submission, and he must, from time to time, review the State's approved planning process to insure that it is at all times consistent with the Act.
* * * * * *
(9) The planning process must include a process which will result in plans for all navigable waters within the State which include, among other things, total maximum daily loads for pollutants and thermal discharges.
The conference report followed the construction of the Act adopted by this Court, a construction that would, removing only those waters redeemed by the effluent limitations, result in a plan and TMDL for every substandard navigable water within a state.
The earlier Senate report stated, "[i]t has become clearly established that the waters of the Nation cannot be restored and their quality maintained unless the very complex and difficult problem of nonpoint sources is addressed." S.Rep. No. 92-414, 92nd Cong., 1st Sess. 39 (1971).
The Definition of PollutantFor this Court, the more troubling issue is one not raised by plaintiffs but one raised by EPA in a footnote (Def. Br. 25 n. 24) and on which the Court requested argument. It concerns the statutory definition of "pollutant." TMDLs were made obligatory, as stated, only for "those pollutants which the Administrator identifies under Section 304(a)(2) as suitable for such calculation." In turn, the latter provision called for EPA to consult with state and federal agencies and to develop and publish an identification of pollutants suitable for TMDL measurement correlated with the achievement of water-quality objectives. After enactment, EPA identified "all" pollutants as suitable. 43 Fed. Reg. 60662 (Dec. 28, 1978).
But what is a "pollutant"? "Pollutant" was defined in Section 502 as meaning "dredged spoil, solid waste, incinerator residue, sewage, garbage, sewage sludge, munitions, chemical wastes, biological materials, radioactive materials, heat, wrecked or discarded equipment, rock, sand, cellar dirt and industrial, municipal and agricultural waste discharged into water" (emphasis added). Two issues arise. First, the statutory definition did not include the word "sediment." To that, there is a dispositive answer. The Ninth Circuit has already held that "sediment" and natural material from the "bank alongside" a river is a pollutant (in the context of a placer-mining point source). Rybachek v. EPA, 904 F.2d 1276, 1285-86 (9th Cir. 1990); accord: Idaho Conservation League v. Thomas, 91 F.3d 1345, 1347 (9th Cir. 1996) (TMDL prepared for sediment); Driscoll v. Adams, 181 F.3d 1285, 1291 (11th Cir. 1999) (sediment is a pollutant); see also United States v. M.C.C. of Florida, Inc., 772 F.2d 1501, 1505-06 (11th Cir. 1985) (dredged spoil includes vegetation and sediment); Hudson River Fishermen's Ass'n v. Arcuri, 862 F. Supp. 73, 76 (S.D.N.Y. 1994) (pollutants include rock, sand and dirt). Moreover, the legislative history referred to "sediment" as a "pollutant," stating "sediment, often associated with agricultural activities, is by volume our major pollutant. . . ." S.Rep. No. 92-414, 92d Cong. 1st Sess. 52 (1971). This statement appears, incidentally, in the same report briefly explaining that the definition of pollutant was borrowed from the Refuse Act. Id. at 76.
Second, and more troubling, the italicized phrase above — "discharged into water" — was part of the statutory definition of pollutant. "Discharges" are uniquely associated with point sources under Section 502(12). Oregon Natural Desert Ass'n v. Dombeck, 172 F.3d 1092, 1095 (9th Cir. 1998). One might wonder, therefore, whether the entire list of statutory pollutants was confined to point sources. If so, then TMDLs were authorized only for point-source pollutants (as would be other features of the Act). Preliminarily, however, the statutory definition is ambiguous for it reads "dredged spoil . . . rock, sand, cellar dirt and industrial, municipal and agricultural waste discharged into water." The phrase "discharged into water" might have been intended to modify only the tag-end phrase "and industrial municipal and agricultural waste." Alternatively, it might have been intended to modify the entire list. To add confusion, the statutory definition of "discharge" in Section 502(12) itself incorporated the term "pollutant," thus injecting a circularity problem.
Significantly, the Act otherwise referred to "nonpoint sources" of "pollutants," including from "agricultural and silvicultural activities, including runoff from fields and crop and forest lands." E.g., §§ 105(d), 304(e), see also 305(b)(1)(E). The operative language of the Act, therefore, expressly treated pollutants as emanating from nonpoint sources. That usage was broader than and inconsistent with a narrow image of pollutants only flowing out of a pipe. Similarly, Section 201(d)(2) called for construction of facilities providing for, among other things, "the confined and contained disposal of pollutants not recycled." If pollutants had to be discharged into water, they could not also be confined and contained. These provisions make reasonably clear that pollutants could derive from any source, not merely from point sources. So too with the legislative history. While the background of the definitions is unilluminating, the legislative history otherwise referred to non-point-source "sediment" as a "pollutant," as quoted above. See H.R. Rep. No. 92-911, 92nd Cong., 2nd Sess. 102 (1972); S. Rep No. 92-414, 92nd Cong., 1st Sess. 10, 13, 52, 68 (1971); S.Rep. No. 92-1236, 92nd Cong., 2nd Sess. 126 (1972). To confine pollutants to point sources, finally, would impair the "comprehensive" fabric of the Act. Any residual doubt on this score is eliminated by deference to the reasonable construction adopted by the agency charged with enforcement of the Act. Chevron, 467 U.S. at 865, 104 S.Ct. 2778. Therefore, the Court holds that "pollutant," as used in the Act, includes sediment, regardless of whether it comes from a point source or a nonpoint source.
The 1987 Amendment and Section 319
In 1987, Congress amended Section 319 to the Clean Water Act. It was specifically directed to nonpoint-source management programs. 33 U.S.C. § 1329; 112 Stat. 3283. This enactment, plaintiffs urge, would have been unnecessary and superfluous if Section 303(d) already comprehended nonpoint sources. In brief, Section 319(a)(1) required each governor to submit to EPA a report that identifies:
Those navigable waters within the State which, without additional action to control nonpoint sources of pollution, cannot reasonably be expected to attain or maintain applicable water quality standards. . . .
The report was to identify categories and subcategories of nonpoint sources, the state's process for identifying best management practices, the state's measures to control each such category and subcategory, and the state and local programs for controlling pollution from nonpoint sources. Section 319(b) also required each state to submit a "management program" for controlling nonpoint-source pollution, including an identification of the best management practices which will be undertaken to reduce "pollutant loadings" resulting from each category and subcategory. Plaintiffs are correct that the 1987 amendment covered some of the same general ground that EPA contends was already enacted. Nonetheless, plaintiffs' argument is rejected for three reasons:
First, while Section 319 addressed non-point pollution, it did not conflict with or duplicate the Listing/TMDL provisions at issue. The Section 303(d) list called for all unfinished business after application of technology-driven effluent limitations. Section 319, however, sought instead to list those rivers and waters which could not achieve standards "without additional action to control nonpoint sources of pollution." The two lists would partially overlap, to be sure, but were not the same. A river ruined only or mainly by industrial waste might make the Section 303(d) list — but only that list — if the best available technology would be insufficient to meet state standards and any cleanup of non-point contaminants would make no material difference. In contrast, a remote river muddied by excessive logging might make both lists. Moreover, Section 319 was silent as to TMDLs whereas Section 303(d) required them. Section 303(d), therefore, supplied an important ingredient for the reports and plans under the 1972 Act as well as those later required under Section 319. Just as the TMDLs were input for the area-wide management plans under Section 208 and continuing planning process establish under Section 303(e), the TMDLs were needed for the planning required under Section 319. Finally, Section 310 authorizes federal grants to the states for nonpoint-source management programs, a new and additional feature to combat a continuing problem.
Second, while the 1987 enactment adopted newer and stronger measures to address the problem of nonpoint pollution, the 1972 enactment plainly spelled out — expressly so — medicine of its own. The phrase "nonpoint sources of pollution" was prominent in the 1972 Act (see Sections 201(c), 208(b)(2)(F), 304(e), 305(b)(1)(E)), as set forth above. It is inaccurate to argue, as do plaintiffs, that nonpoint-source pollution escaped attention under the 1972 Act.
Third, the Ninth Circuit has rejected a similar attempt to infer congressional intent for the 1972 Act from a later Clean Water Act amendment (in that case from the 1977 amendment):
This legislative history does not persuade us, because it is not part of the law, was written long after the law was passed, and seems inconsistent with the law passed when it was written. This is 1977 "history" about a 1972 law. Instead of giving us a window into the thinking of the legislators who wrote the bill, it gives us the advice of someone on a House Conference Committee staff five years after section 1369 was promulgated about how we should construe a law passed by an earlier Congress under a different president in a different political era. Subsequent legislative history in the form of committee reports of subsequent congresses are generally considered an "extremely hazardous basis for inferring the meaning of a congressional enactment." Consumer Product Safety Comm. v. GTE Sylvania, Inc., 447 U.S. 102, 118 fn. 13, 100 S.Ct. 2051, 2061 fn. 13, 64 L.Ed.2d 766 (1980).Longview Fibre Co. v. Rasmussen, 980 F.2d 1307, 1311-12 (9th Cir. 1992). So too here.
* * * * * *
It is true, as plaintiffs note, that several members of Congress stated that the 1987 amendment would be "a first step" or would "begin" the process of addressing this source of water degradation (see Pl. Br. 21). One must, however, remember the context. After 1972, EPA was exceedingly slow, even resistant to, the water-quality approach. Following the 1972 amendments, EPA was preoccupied in promulgating the technology standards required for point-source regulation. At least in the upper chamber, Senator Edmund Muskie, a champion of the technology approach, had urged EPA to give it top priority and relegate the standards approach to "secondary priority." Senate Report of Committee of Conference on S. 2770 in Lib. Of Cong., A Leg. Hist. Of the Water Pollution Control Act Amendment of 1972, 93d Cong. 1st Sess. 171 (1973). In that spirit, EPA did virtually nothing under Section 303(d) for six years.
The history is documented in Houck, TMDLs. Are We There Yet?: The Long Road Toward Water Quality-Based Regulation Under the Clean Water Act, 27 ELR 10391 (1997).
Under Section 303(e), EPA first published a water-quality standards regulations in 1975. 40 C.F.R. § 130.7; 40 Fed. Reg. 55334 (Nov. 28, 1975).
Then, in 1978, EPA was ordered to publish a final identification of pollutants, a preliminary step under Section 303(d) necessary before any state had to prepare a TMDL. Board of County Commrs. v. Costle, No. 78-0572, slip op. (D.D.C. June 20, 1978) (unpublished order); see 43 Fed. Reg. 42303 (Sept. 20, 1978). EPA stated that it had not considered such finalization "as a matter of high priority" because the "practical results" of TMDLs were already being accomplished, it felt, through basin planning. Id. at 42303. The final identification was published on December 28, 1978. It did not identify any pollutants by name but instead simply identified "all pollutants, under proper technical conditions, as being suitable for the calculation of total maximum daily loads." The phrase "proper technical conditions" was defined to mean "the availability of analytical methods, modeling techniques and database necessary to develop a technology defensible TMDL." Such availability, the EPA said, would have to be determined on a case-by-case basis. Id. at 60662. EPA then asked each state merely to identify "one or more" water-quality limited stream segments within 180 days, leaving it to the future as to when more than one would be due. Id. at 60666.
The states were also slow to respond. Most states submitted no lists under Section 303(d), although Nevada submitted a list in 1979 for the Walker River, a river polluted only by nonpoint sources. EPA then took the position that until a state submitted a TMDL, EPA had nothing to approve or disapprove. Houck, 27 ELR at 10393. Eventually, the Seventh Circuit held that the prolonged failure of a state to submit anything was a "constructive submission" of no TMDL at all, triggering EPA's duty to act. Doing nothing was, the court held, tantamount to approval of a "constructive submission." Scott v. City of Hammond, 530 F. Supp. 288 (N.D.Ill. 1981), aff'd in part, rev'd in part, 741 F.2d 992, 996 (7th Cir. 1984). Another suit in Oregon led to a consent decree in 1987 with a timetable for EPA to act if Oregon did not submit a list of substandard waters. Northwest Environmental Defense Center v. Thomas, No. 86-1578 BU (D. Or. June 3, 1987). In 1991, Judge Rothstein in Seattle held that EPA had flagrantly violated the Act by failing to implement the TMDL requirements for Alaska. Based on Alaska's "constructive submission" of no TMDLs at all, EPA was required to initiate its own process for promulgating TMDLs for that state. The Ninth Circuit affirmed. Alaska Center for the Environment v. Reilly, 762 F. Supp. 1422, 1429 (W.D.Wash. 1991), aff'd, 20 F.3d 981 (9th Cir. 1994). A series of other TMDL lawsuits against EPA were filed in Washington, Idaho, Georgia, New York. Houck, 27 ELR at 10395-96 (summarizing results). That EPA had virtually no TMDL program at all was the conclusion of a 1989 report on the TMDL process by the United States General Accounting. GAO, Water Pollution: More EPA Action Needed to Improve the Quality of Heavily Polluted Waters, GAO Report to the Chairman: Subcommittee on Regulation and Business Opportunities Committee on Small Business, House of Representatives (Jan. 1989).
In April 1991, finally, EPA began to warm to the TMDL process and published its first guidelines for state implementation of Section 303(d) (EPA Tab 16). EPA set October 1992 as a deadline for submission of the lists of substandard rivers and waters. Houck, 27 ELR at 10395. It was into this renaissance of Section 303(d) — twenty years after its passage — that the Pronsolinos were drawn.
Although EPA was exceedingly slow to implement the TMDL requirements, EPA has not taken a position in conflict with its construction urged in this case. To the contrary, EPA's first description of the TMDL process in its revised Water Quality Planning Management Regulations is fully consistent. 40 C.F.R. Parts 35 and 130 and 50 Fed. Reg. 1774-75 (Jan. 11, 1985). At oral argument, plaintiffs cited a recent Supreme Court decision concerning a government agency's regulatory authority. In Food and Drug Administration v. Brown Williamson Tobacco Corp., ___ U.S. ___, 120 S.Ct. 1291, ___ L.Ed.2d ___ (2000), the Court held the FDA had no authority to regulate tobacco products under the Food, Drug and Cosmetic Act. Congress had enacted six separate pieces of legislation addressing the problem of tobacco use and human health. Id. 120 S.Ct. at 1306. In adopting each statute, Congress had "acted against the backdrop of the FDA's consistent and repeated statements that it lacked authority under FDCA to regulate tobacco absent claims of therapeutic benefit by the manufacturer." Congress had considered, and rejected, legislation that would have given the FDA such authority. It is evident, under these circumstances, that the statutes enacted by Congress "effectively ratified the FDA's long-held position that it lacks jurisdiction to regulate tobacco products." Ibid. The Court concluded that Congress did not give the FDA the authority it now sought to regulate tobacco products. Here, on the other hand, EPA had never made any statements to Congress expressing a lack of authority to issue a TMDL for waters polluted by nonpoint sources. Furthermore, EPA has not been inconsistent in its position concerning TMDLs for nonpoint-source polluted waters.
In light of this history, it is no wonder that some elected representatives regarded the 1987 amendment as a "first step" toward controlling nonpoint-source pollution. The Court is convinced, however, that the first step was plainly authorized in 1972 — only to be little noticed due to other cleanup priorities until a generation later.
Grants Versus Regulation
The word "regulate" pervades plaintiffs' argument. Congress did not, they say, authorize EPA to regulate state land-use practices. The Court agrees. EPA agrees. Unlike EPA's authority to revise individual NPDES permits issued by states for individual point sources, EPA received no authority to review land-use restrictions placed (or not placed) on timber-harvesting permits by CDF or any other practice permitted for agriculture or silviculture. The 1972 Act was clear that states should finally decide whether, and to what extent, land-management practices should be adopted to mitigate runoff. To assist the states in gathering information, the statutory role of the TMDL was to identify the load necessary, as a matter of engineering, to implement the water-quality standards. Without such engineering data, states would be left to guess what needs to be done to meet those standards.
Under the Act, California must "incorporate" the TMDL in its planning. Nothing, however, requires that the TMDL be uncritically and mechanically passed through to every relevant parcel of land. California is free to select whatever, if any, land-management practices it feels will achieve the load reductions called for by the TMDL. California is also free to moderate or to modify the TMDL reductions, or even refuse to implement them, in light of countervailing state interests. Although such steps might provoke EPA to withhold federal environmental grant money, California is free to run the risk.
A practical reality, of course, is that once federal environmental grant money begins to flow, state regulatory agencies become dependent on it. They become sensitive to threats to terminate it — terminations that would entail job and programmatic cuts. This influences behavior. A state may knuckle under to coercive threats by EPA. A state may uncritically apply TMDL-loading reductions, like the ones at issue, without regard to other legitimate state interests or to the unique circumstances of an applicant. Even so, this is not direct federal regulation. The regulation is by California — though influenced by incentives established by Congress and the agency charged with protecting the environment. Cf., North Carolina Dept. of Transportation v. Crest Street Community Council, 479 U.S. 6, 8, 107 S.Ct. 336, 93 L.Ed.2d 188 (1986).
Landowners like the Pronsolinos have avenues of redress. They can appeal unreasonable or unauthorized restrictions within the state administrative system. Aggrieved landowners in a river basin might collectively or singly challenge a TMDL by EPA or a Section 303(d) listing under the Administrative Procedure Act as "arbitrary" or "capricious," or "unsupported by substantial evidence," or an "abuse of discretion." 5 U.S.C. § 702, 706. Landowners, for example, might try to show that EPA's engineering is manifestly wrong. No such claim, however, is made here. The only claim is that the Garcia River should never have been listed in the first place and that no TMDL at all was ever authorized, a claim that must be rejected for the reasons stated.
In summary, the Clean Water Act called for a comprehensive set of water-quality standards for every navigable river and water in America. For every substandard navigable river or water, Congress sought a determination whether the central innovation of the 1972 Act — technology-driven limits on effluent — would be sufficient to achieve compliance. If not, the river or water was required to go on a list of unfinished business and a TMDL calculation was required. The TMDL was to quantify the load improvements necessary to meet standards. If EPA disagreed with a state's list or any TMDL as inconsistent with the purposes of the Act, then EPA was required to revise the list or the TMDL. No substandard river or water was immune by reason of its sources of pollution. The process was made just as mandatory for wild but ruined rivers as it was for urban-blighted waters.
Once the TMDLs were prepared, they were intended to be applied to point and nonpoint sources differently. As to point sources, the TMDLs were to be taken into account in further restricting effluent, under NPDES permits, as authorized by Section 301(b)(1)(C). As to nonpoint sources of pollution, the TMDLs were to be incorporated into the continuing planning processes of the states. This conferred a large degree of discretion on the states in how and to what extent to implement the TMDLs for nonpoint sources. A state could even refuse to implement a TMDL, eschewing best management practices if it wished, although to do so might provoke EPA to curtail or to deny grant money to the state. But as to whether TMDLs were authorized in the first place for all substandard rivers and waters, there is no doubt. They plainly were and remain so today — without regard to the sources of pollution.
This resolves the issue raised over the power of EPA to list waters like the Garcia River or to issue TMDLs for such waters. The complaint raised no question whether the specific listing or specific TMDL was otherwise unlawful. The complaint did not, for example, challenge the specifics of the TMDL as arbitrary or capricious. The case now having been fully resolved on cross-motions for summary judgment, JUDGMENT shall be entered for defendants. The Clerk shall then close the file.