From Casetext: Smarter Legal Research

National Wildlife Federation v. Gorsuch

United States District Court, D. Columbia
Jan 29, 1982
530 F. Supp. 1291 (D.D.C. 1982)


indicating that penstocks in hydroelectric dams may constitute point sources

Summary of this case from Natl. Wildlife Fedn. v. Consumers Power


Civ. A. No. 79-0915.

January 29, 1982.

Patrick A. Parenteau, David G. Burwell, National Wildlife Federation, Washington, D.C., for plaintiffs.

Robert M. Lindholm, Jefferson City, Mo., for plaintiff intervenor State of Mo.

Fred R. Disheroon, Sr. Trial Atty., Dept. of Justice, Washington, D.C., for Anne Gorsuch in her official capacity as Administrator of the U.S. Environmental Protection Agency.

Arnold H. Quint, Washington, D.C., Turner T. Smith, Jr., Richmond, Va., for Electric Utilities-Alabama Power Co., et al.

Gerry Levenberg, Washington, D.C., for Northwest Utilities Group-Idaho Power et al.

Sam Kazman, Pacific Legal Foundation, Washington, D.C., for Ass'n of California Water Agencies (ACWA).

Robert L. McCarty, Washington, D.C., for Colorado Water Agencies-Colorado River Water Conservation Dist., et al.

Kenneth A. Rubin, James B. Vasile, Washington, D.C., for American Water Works Ass'n and Nat. Ass'n of Water Companies.

Frank E. Evans, Robert L. McCarty, Washington, D.C., for Nat. Water Resources Ass'n.


This case concerns the applicability to man-made dams of the Clean Water Act's permit system for pollutant discharges. Plaintiff National Wildlife Federation and plaintiff-intervenor State of Missouri seek a declaratory judgment that defendant Gorsuch, has violated her non-discretionary duty by failing to regulate the discharge of pollutants from dams under the National Pollution Discharge Elimination System (NPDES) as required by sections 301(a) and 402(a) of the Clean Water Act, 33 U.S.C. § 1311(a) and 1342(a), and a writ of mandamus or injunction compelling defendant to immediately promulgate regulations which designate dams as a point source category under CWA § 402, establish effluent limitations or other performance standards for dams on a categorical basis, and subject existing and proposed dams to all the NPDES requirements applicable to other categories of point sources. The case was brought under CWA § 505(a)(2), 33 U.S.C. § 1365(a)(2), vesting jurisdiction in the district courts over citizen suits against the Administrator of EPA "where there is alleged a failure of the Administrator to perform any act or duty under this Act which is not discretionary with the Administrator"; and under the Administrative Procedure Act, 5 U.S.C. § 706(2)(A) and (C), directing reviewing courts to set aside administrative action "not in accordance with law" and "in excess of statutory . . . authority . . .". There is no dispute as to EPA's non-discretionary duty to regulate dams under the NPDES program, if dams are in fact point sources which discharge pollutants into navigable waters within the meaning of the CWA. The case was tried to the Court in a three day trial.

The federal defendant will sometimes be referred to as EPA. Numerous electric utilities and water agencies, a water resources development association, and two trade associations of investor-owned water utilities have also intervened as defendants.

The Federal Water Pollution Control Act Amendments of 1972 constituted a major overhaul of the then existing Federal Water Pollution Control Act, and instituted the NPDES program at issue here. The Act was again amended by the Clean Water Act of 1977, in particulars not relevant to this action. The statute will be referred to in this opinion interchangeably as the FWPCA, the CWA, or the Act.

Although the parties were unable to agree to stipulated facts which would have allowed disposition of this case on cross motions for summary judgment, they are nevertheless in agreement that the essence of this case is a single legal question. As will be seen, remaining factual disputes are not material to the resolution of that issue. It is not disputed that the existence and operation of dams sometimes cause water quality problems. The question is whether, in terms of the statutory definitions in the CWA, the discharge of water of diminished quality from dams constitutes the discharge of one or more pollutants into the navigable waters of the United States from a point source. If so, the discharges would be unlawful, CWA § 301(a), 33 U.S.C. § 1311(a), unless permitted under the NPDES system. CWA § 402(a), 33 U.S.C. § 1342(a).

EPA has taken the position that dams are not point sources requiring NPDES permits, (See Exhibit 2 to Plaintiff's Amended Complaint), and has classified dams as nonpoint sources of pollution. In February 1978, plaintiff National Wildlife Federation petitioned the EPA to institute a rulemaking proceeding to set uniform effluent limitations for discharges from hydroelectric dams as a point source category under CWA § 402(a). EPA has not done so. Defendants argue that dams were not intended to be subject to NPDES regulation, and that the water quality problems associated with dams, can be, and are being addressed through other mechanisms in the CWA, such as the regulation of upstream point sources and the nonpoint source pollution program under § 208, and by other federal and state laws.

Pltfs.' Ex. 15.

The 1972 Federal Water Pollution Control Act Amendments marked a major transformation in the nation's approach to the control of water pollution. Prior to 1972, the program was based upon water quality standards promulgated and implemented by the states with some assistance and oversight from the federal government. The 1965 Act required each state to classify its streams (or stream segments) and waters according to their intended uses, such as agriculture, municipal water supply, fish and wildlife, or recreation; and set water quality standards, such as the allowable concentration of dissolved oxygen or suspended solids, appropriate for each category of use. The method of controlling water pollution was to work backwards from the desired water quality for the water body, and taking into account its capacity to assimilate pollutants, attempt to determine which sources were responsible for pollution causing violation of the standards. Pollution discharges did not violate the law unless they could be shown to cause the water body to fail to meet water quality standards. This process was inherently difficult and uncertain, and combined with the slow progress of the states in setting the standards and the cumbersome enforcement mechanisms, caused the Senate Committee on Public Works to conclude in 1972 that "the national effort to abate and control water pollution has been inadequate in every vital aspect", leaving many of the nation's navigable waters severely polluted, with major waterways near the industrial and urban areas unfit for most purposes.

2 A Legislative History of the Water Pollution Control Act Amendments of 1972 (Leg.Hist.) at 1254, 1271, 1280, and 1303.

2 Leg.Hist. at 1425.

In an endeavor to bolster the flagging antipollution effort, in 1970 federal officials instituted a permit system for pollutant discharges under Section 13 of the Refuse Act of 1899, 33 U.S.C. § 407, which prohibits the discharge of any matter into navigable waters without a federal permit. This system, for various reasons, also proved to be cumbersome and ineffective.

2 Leg.Hist. at 1423.

The new approach instituted by the 1972 legislation is based upon the principle that no one has the right to use the nation's waters to dispose of pollutants. Rather than allowing the disposal of pollutants in water up to the point where it causes water quality violations, the goal of the Act is to eliminate completely the discharge of pollutants into navigable waters. CWA § 101(a); 33 U.S.C. § 1251(a). However, since immediate total elimination of pollutant discharges is obviously impracticable, the Act provides a permitting system for discharges. To oversimplify somewhat, consonant with the philosophy of no right to pollute and the zero discharge goal, the permits are designed to allow the lowest level of discharge technologically feasible. For various categories of sources, EPA determines the pollution control processes which meet the appropriate level of technology, and sets effluent standards which can be achieved by application of those processes. The NPDES program converts these generalized standards into requirements for each individual source. Each permit contains effluent limitations, that is, restrictions of quantities, rates and concentrations of chemical, physical, biological, and other constitutents which may be discharged, CWA § 502(11), 33 U.S.C. § 1362(11), and a schedule of compliance. The permit program is administered either by EPA, or by states with EPA-approved administration programs. Water quality standards are not abolished, but supplement the technology-based program, and may be included in NPDES permits. CWA §§ 302, 303, 402(a)(1), 33 U.S.C. § 1312, 1313, 1342(a)(1).

2 Leg.Hist. at 1460-1461. Actually, there are various technological standards to be met by different dates for different pollutants. "Best practicable control technology currently available", "best available technology economically achievable," and "best conventional pollutant control technology" are terms of art defined by the EPA Administrator. CWA § 301(b), 33 U.S.C. § 1311(b).

In determining whether dams are subject to the NPDES program, the question is whether dams "discharge pollutants" within the meaning of the CWA. CWA §§ 301(a) and 402(a), 33 U.S.C. § 1311(a) and 1342(a). The definition of "pollutant" is "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." CWA § 502(6), 33 U.S.C. § 1362(6). The term "discharge of a pollutant" means "any addition of any pollutant to navigable waters from any point source", CWA § 502(12), 33 U.S.C. § 1362(12), and the term "point source" is defined as "any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged. This term does not include return flows from irrigated agriculture." CWA § 502(14), 33 U.S.C. § 1362(14).

Defendants have not argued that dams cannot be point sources. In the parties' Joint Statement of Material Facts No Longer in Dispute, they have characterized a dam as any structure which impounds water, and explained that all dams must have some structure to release excess water over, through, or around the dam, which is called a spillway. There are many different spillway designs. Some merely release water over the top of the dam. Large dams usually have multiple outlets, including one near the bottom of the reservoir so that most of the stored water is available for its intended use. Hydroelectric dams also have "penstocks", pipes which allow water to flow through the dam to the turbines before returning to the stream below. Dams constructed for navigation are equipped with locks, and those for irrigation have special outlets to the irrigation canals. Clearly at least some, if not all, of these discharge outlets from dams, are "discernible, confined and discrete conveyances" which meet the CWA's definition of point source, especially in light of the congressional intent to embrace the "broadest possible definition of any identifiable conveyance from which pollutants might enter the waters of the United States." United States v. Earth Sciences, Inc., 599 F.2d 368, 373 (10th Cir. 1979).

Tr. 11/5/80 (Defts'. Closing Argument) at 113.

Joint Statement of Material Facts No Longer in Dispute, (Paragraph # 2 of Pre-Trial Order), ¶¶ 4-6.

Rather than allege that dams cannot be point sources, defendants have argued that some of the water quality changes described by plaintiffs are not pollutants, and that none of them are added by dams to navigable waters. Each of the dam-created pollution problems presented by plaintiffs will be discussed to illuminate the determination of whether they constitute the discharge of pollutants by dams.

Low Dissolved Oxygen

The parties agree that certain dams which impound large, deep reservoirs and release water from deep in the reservoir discharge water low in dissolved oxygen (DO) content during the warm months of the year. The process responsible for depleting DO in the lower portion of standing water bodies is called thermal stratification. The surface of the water is heated by the sun, while the deeper water is not. Since warmer water is less dense than colder water, it has a tendency to stay at the top. For most of the year, wind-generated currents and waves mix the water so that its properties remain essentially uniform at all depths, but in the summer when the temperature differences are large, the currents cannot overcome the buoyancy of the warm water, the layers do not mix, and the water body becomes thermally stratified. The interface between the layers is known as the thermocline. The upper layer is called the epilimnion; and the lower layer is the hypolimnion. In the fall, the top layer of water cools, and becomes dense enough to sink and mix to some extent with the lower layers. This change reduces the temperature differential between layers and allows further wind-driven mixing until the waters gradually return to a fully mixed condition.

Id., ¶¶ 15, 16; Testimony of Gerald T. Orlob, expert in water quality, hydrology and hydraulics, Tr. 11/4/80 at 111-114; Pltfs.' Ex. 1 at 4-11.

The hypolimnion of a stratified water body generally coincides with the tropholytic zone. In that region, sunlight levels are insufficient to support photosynthesis, and the respiration and decomposition of animals and bacteria can result in a net consumption of oxygen. In a non-stratified water body, waters which have been rear-eated by the atmosphere at the surface mix with these deep waters, replenishing the oxygen. However, in a thermally stratified body, as we have seen, that mixing does not occur, and dissolved oxygen content in the hypolimnion can drop to very low levels.

Joint Statement of Material Facts No Longer in Dispute, ¶¶ 18 and 19; Orlob testimony, Tr. 11/4/80 at 116-118, 141-143.

The amount of dissolved oxygen in a lake or reservoir is strongly influenced by the amount of oxygen demanding organic material in the water body, which in turn is a product of both processes within it and of inputs of organic matter and nutrients from upstream sources. A lake or reservoir collects oxygen demanding material which would continue on downstream in a free-flowing stream. The decrease in water velocity when these materials reach the reservoir allows the materials to settle there. Some organic materials sink directly to the bottom. Inflowing nutrients like phosphorous and nitrogen promote the growth of plants such as algae in the epilimnion, which eventually sink to the hypolimnion below, and demand oxygen for their decomposition. Thus large reservoirs differ from free-flowing streams both in collecting and producing more oxygen-demanding material, and in forming a thermally stratified zone which is not rearated. The combination of those factors creates oxygen depleted water.

Joint Statement of Material Facts No Longer in Dispute, ¶¶ 19, 19a, and 20.

Id., ¶ 20; Pltfs.' Ex., 1 at 23-24.

Although these processes occur in large natural lakes as well as in man-made impoundments, natural lakes do not discharge oxygen deficient waters downstream. A natural lake will discharge well-oxygenated water from the surface, while the outlet from a dam may be in the hypolimnionic layer, where the dissolved oxygen has been depleted.

They do not occur in the free-flowing streams which are transformed into reservoirs by dams.

Testimony of Dr. James R. Whitley, expert in aquatic pollution and the effects of such pollution on aquatic ecology, Tr. 11/3/80 at 37-38, 117-118. It should be noted however, that low dissolved oxygen problems can occur in both standing and flowing water due to causes other than dams, such as the addition of large quantities of sewage or other oxygen-demanding organic matter, Tr. 11/3/80 at 111-117; Pltfs.' Ex. 50, or low stream flows. Orlob testimony, Tr. 11/4/80 at 161-162.

As almost all animals living in water require dissolved oxygen to carry out metabolic processes, DO is one of the more common measurements of water quality. Discharges from some dams are low enough in DO in the warm months to violate state water quality standards and to harm or destroy aquatic life, in some cases resulting in large fish kills. Low DO discharges can also reduce the assimilative capacity of the river downstream from the dam. That is, there is less oxygen available to break down organic matter, and additions from other sources might require more controls to avoid water pollution. The record reveals that the discharge of oxygen-depleted water from dams is a widespread and serious problem. Although defendants have questioned the severity and extensiveness of the problem claimed by plaintiffs, they have not disputed that it is a problem associated with the operation of dams which may require control techniques. The dispute is whether the release of oxygen-depleted water from dams constitutes a "discharge of pollutants" under the CWA.

Testimony of Dr. Atwell Ray Abernathy, expert in water pollution and aquatic biology, South Carolina Wildlife Federation v. Alexander, Pltfs.' Ex. 22 at 29. That case involves the same issue as the instant case, as it relates to certain dams in South Carolina. Defendant EPA was represented by the same counsel in both cases.

Abernathy testimony, Pltfs.' Ex. 22 at 95-96; Pltfs.' Ex. 23 at 19.

Whitley testimony, Tr. 11/3/80 at 36-54, 77-78, 130-131, 143-144; Pltfs.' Exs. 12, 23 at 19, 31, 52, 53 at 9-10, 58(c), and 59; United States Environmental Protection Agency, "Nonpoint Source Control Guidance: Hydrologic Modifications" (February 1977), Exhibit to Post-Trial Brief of Defendant-Intervenors Colorado Water Agencies, National Water Resources Association, American Waterworks Association, National Association of Water Companies, and the Association of California Water Agencies (hereinafter "Water Group Post-Trial Brief"), at I-10 (hereinafter "EPA Hydrologic Modifications Report").

Tr. 11/5/80 (Defts.' closing argument) at 114.

Dissolved metals

The dissolved metals problem is directly related to the oxygen depletion at the bottom of reservoirs which has just been described. Metals such as iron and manganese exist in particulate form in most bottom sediments of reservoirs. With sufficient dissolved oxygen in the water, a zone of oxidized iron, manganese and other constitutents develops between the water and the bottom sediments which effectively serves as a barrier preventing the bottom sediments from diffusing into the water. Yet, in a low DO condition, the iron and manganese in the interface zone will convert to more soluble forms and allow the metal in the bottom sediments to diffuse into the overlying water. These dissolved metals, unlike their particulate forms, can be harmful to aquatic organisms and can cause the water to require treatment to be usable for domestic, municipal, or industrial purposes. High concentrations of dissolved iron and manganese have been found in discharges from dams. Although defendants' expert, Dr. Orlob, testified that certain reservoirs have high "trapping efficiency", discharging less iron and manganese than the amount entering from upstream sources, defendants have not disputed that the above-described process does occur in some reservoirs, increasing the amount of metals in soluble form that are discharged downstream. Temperature

Abernathy testimony, Pltfs.' Ex. 22 at 35-37; EPA Hydrologic Modifications Report at I-10; United States Environmental Protection Agency, "The Control of Pollution from Hydrographic Modifications" (1973), Exhibit to Water Group Post-Trial Brief, hereinafter "EPA Hydrographic Modifications Report," at 75-76; Orlob testimony, Tr. 11/4/80 at 118-120.

Abernathy testimony, Pltfs.' Ex. 22 at 38-39; testimony of Dr. James Edward Schindler, expert in limnology and aquatic ecology, Pltfs.' Ex. 22 at 181.

Abernathy testimony, Pltfs.' Ex. 22 at 40-41, 45-46; Schindler testimony, Pltfs.' Ex. 22 at 178-182.

Orlob testimony, Tr. 11/4/80 at 128-138.

Defendants' Final Proposed Conclusions of Law ¶¶ 10-11; EPA Hydrographic Modifications Report at 75-76.

The discharge by dams of water which is warmer or colder than the natural stream temperature is also a function of thermal stratification and the level of the release structures in the dam. In a deep reservoir during a thermally stratified period, cold water would be released from the bottom of a reservoir, while warm water would be released from the top. The operation of a dam/reservoir facility can change the downstream temperature 20 to 30°(F. As various species of fish and other aquatic organisms are adapted to either warm or cold water conditions, dam-engendered temperature changes can, and have, resulted in the migration of fish away from certain areas, transforming a warm water fishery into a cold water fishery or vice-versa, and in fish kills. Again, defendants have not disputed that these temperature changes occur, and have adverse effects on aquatic life, but argue rather that they do not constitute discharges of pollutants as defined by the CWA.

Whitley testimony, Tr. 11/3/80 at 72-73.

Id., at 38, 74-76; Pltfs.' Exs. 2, 10, and 55.

Defendants' Final Proposed Conclusions of Law ¶ 14; Defendants' Joint Pretrial Brief, Statement of Facts ¶¶ 44-47; EPA Hydrographic Modifications Report at 79.


Streams carry sediment which consist of organic and inorganic material derived from runoff from the land and from the streambanks and channel bed. Sources of sediment in water bodies include farms, forest land, geologic outcroppings subject to erosion, vegetation, and construction sites. The amount of sediment which a stream will transport is generally dependent on the characteristics (size, shape, weight) of the sediment particles, and the hydraulics, or flow conditions, of the stream. Usually, the greater the velocity of the streamflow, the more sediment it will transport, as it is the flow velocity which keeps the particles in suspension rather than allowing them to settle to the bottom. When a portion of a stream is transformed into a reservoir by the construction of a dam, the normal flow of sediments downstream is intercepted by the reservoir, where the sediments tend to sink to the bottom. As the water from a river enters the reservoir, the flow velocity decreases and the jet of water expands. Sediment settles out, beginning with the coarsest particles. The ability of a reservoir to retain sediment is called trapping efficiency. Trapping efficiency is dependent on the size, shape and depth of a reservoir as well as the characteristics of the sediment particles. The greater the length of the reservoir, the more sediment will settle out before reaching the dam and the less sediment will be passed downstream. Larger, heavier particles will settle out more quickly than finer, lighter particles. The chief parameter related to the trapping efficiency of a reservoir is its retention time, defined as the volume of the reservoir divided by the rate of inflow. Since retention time derives from the size of the water body as well as its rate of flow, it encompasses the most important factors in trapping efficiency. The larger the retention time (i.e. a large reservoir with a small inflow), the greater the trapping efficiency.

Testimony of Dr. Daryl B. Simons, expert in hydrology, hydraulics, sedimentation, erosion, river mechanics and hydraulic structures, Tr. 11/4/80 at 40, 42.

Id., at 43.

Id. at 59, 62-64.

Another reason that many man-made impoundments are so effective in trapping sediments is that they are designed to cut off or moderate the transport of flood waters downstream. Since about 90% of sediments transported in river systems are transported during periods of major flood, these dams will significantly reduce the amount of sediment flowing past the dam.

Id., at 57-58.

Due to the various factors just discussed, the amount of sediment passed through dams can range from the full amount that was transported by the free-flowing stream to near zero. Most large reservoirs release far less sediment than the amount entering them, although occasionally when waters are released from a reservoir to augment low flows downstream, the amount of both water and sediment passing the dam will exceed the amount under natural conditions.

Id., at 63, 65.

Reservoirs which receive high loads of sediment and have high trapping efficiency can fill in at fairly rapid rates, impairing the functions for which the reservoir was designed. There are various possible techniques for removing the sediment from the reservoir bed, although they are generally not economically feasible for large reservoirs. Reservoirs may be dredged, that is, material is physically pumped out from the bottom; structures may be used to bypass heavy sediment loads around the reservoir; or they may be sluiced, a process consisting of drawing down the level of the reservoir to nearly or completely empty, and then allowing the stream to pass over the reservoir bottom and carry some of the sediment load with it downstream. The latter process can result in the release of water higher in sediment content than that which existed downstream before the dam was built, up to the capacity of the system to carry sediment. Dredging operations may also significantly increase downstream sediment loads if performed carelessly, so that not all of the dredged material is removed from the water, but rather is resuspended near the dam and then transported downstream.

Id., at 48.

Id., at 68.

Id., at 83-85, 90-91.

Id., at 81.

Illustrative is the sluicing of Guernsey Reservoir on the North Platte River in Wyoming which began in 1959, in order both to restore the volume of the reservoir, and to prevent seepage losses in downstream irrigation canals. Before the construction of the dam and reservoir, silt and clays entering the irrigation canals were to some extent implanted in their banks and beds and reduced seepage losses of water. The clear water being released from the reservoir failed to perform this function, and the affected irrigation districts insisted on the sluicing operation. The high sediment releases, however, caused violation of downstream water quality standards. In 1976, EPA issued an NPDES permit for the release of sediments from Guernsey, which in effect allowed only three more annual sluicing operations.

Id., at 86-87; Pltfs.' Ex. 30; Defendant's Response to Plaintiff's First Requests for Admissions No. 19.

Both the interruption of the flow of sediments and their release in large quantities by dams can degrade water quality. Increased sediment flow can reduce the quality of water for irrigation, hydro-power production, municipal use, and fisheries; while the release of water depleted of sediments can, among other things, induce erosion, increase seepage, and increase the growth of algae and aquatic plants.

Pltfs.' Ex. 21; Defendant's Response to Plaintiff's First Requests for Admissions No. 22.

Pltfs.' Ex. 21 at 97.

Id. at 103-104.

There is no dispute as to any of these facts, which were supplied almost entirely by defendants' expert, Dr. Simons. Again, the dispute is whether dam-induced effects on sediment loads constitute the discharge of a pollutant as defined by the Clean Water Act.


Supersaturated water contains more than the normal amount of dissolved atmospheric gases. It can occur under both natural and man-made conditions when atmospheric gases are driven deep into water, where the greater pressure causes more gas to go into solution than would at the surface. The condition is created at some natural waterfalls, and at some dams where water flows over spillways into stilling basins of at least a few meters depth. On the way down, the water becomes mixed with air. If the force and velocity of the water are sufficient, the water will plunge to a depth where the pressure causes air entrained on the way down to go into solution, and the water becomes supersaturated. Supersaturation has also been observed in water released from hydroelectric turbines where air has been injected into the penstocks. Although supersaturated water is not harmful to humans, and is suitable for most purposes, it can be lethal to fish and other aquatic organisms. Fish take in water through their gills and circulate it through their bodies. The level of dissolved gas in the water inside the fish will generally be the same as that of the surrounding water. When a fish takes in supersaturated water, some of the gases will come out of solution inside the fish, either due to normal physiological processes, or more rapidly if the fish swims into shallower or warmer water where the pressure or temperature causes gas to come out of solution. Bubbles form in the fish and may block the flow of blood in the blood vessels, in severe cases stopping the entire flow of blood and causing death from lack of oxygen. In less severe cases, the bubbles can cause breaks in the skin which may later lead to infection and death. The condition is called gas bubble disease.

Whitley testimony, Tr. 11/3/80 at 54-56; Testimony of Dr. Donald E. Weitkamp, expert in the field of supersaturation, Tr. 11/4/80 at 206-211; Pltfs.' Ex. 40 at 1; Pltfs.' Ex. 39 at 7.

Pltfs.' Ex.39 at 7-8; See Weitkamp testimony, Tr. 11/4/80 at 217-218.

Whitley Testimony, Tr. 11/3/80 at 57-58; Pltfs.' Ex. 58d at 2; Pltfs.' Ex. 40 at 2; See Pltfs.' Exs. 5 and 19.

Gas bubble disease was responsible for the largest fish kill in Missouri history, below Harry S. Truman Dam on the Osage River. In 1978, conservatively estimated, over 421,000 fish were killed. In 1979, the condition reoccurred, and 100,000 dead fish were counted in the first fifty miles of river below the dam. Supersaturation was caused by uncontrolled flows over the spillway of the dam during the construction stage. After the first fish kill in 1978, the Army Corps of Engineers, the agency building the dam, attempted to correct the problem by installing a spillway deflector designed to cause the flow to move horizontally, rather than plunge to a deep level, in the stilling basin. This would prevent the air and water mixture from reaching the deeper levels where increased pressure would force more gas into solution. Nevertheless, the following spring, the increased flow from heavy spring rains and snow melt overshot the spillway deflector and again plunged deep into the stilling basin, creating supersaturated conditions and another fish kill. In 1980, the spillway deflector was more successful, and no fish kill was observed. When the dam is fully operational, the water will pass through the hydroelectric turbines rather than over the spillway, and supersaturation is not likely to be a problem, except perhaps in flood conditions when all the water cannot be passed through the generators.

Whitley Testimony, Tr. 11/3/80 at 59-60; Pltfs'. Ex. 50, Abstract; Pltfs'. Exs. 3, and 14.

Pltfs'. Ex. 58d at 7-8.

Whitley testimony, Tr. 11/3/80 at 66-69, 92-93; Weitkamp testimony, Tr. 11/4/80 at 220-222.

Supersaturation was also at one time a severe problem in the Columbia River System in the Northwest. Although it probably existed since Bonneville Dam was constructed in 1938, it was first recognized as a problem in 1965. Because of the numerous dams in the watershed, high levels of supersaturation caused by water plunging over spillways existed in all reaches of the Columbia and Snake Rivers in the 1960s and '70s. Major losses of salmon and steelhead were attributed to gas bubble disease, and scientific studies in 1975 estimated that if the situation continued uncorrected, it would result in a total loss of two million salmon and steelhead between 1976 and 2000. A combination of the installation of spillway deflectors and greatly decreased flow over spillways due to the operation of additional turbines and completion of upstream water storage projects has effectively remedied the problem since that time.

Pltfs.' Ex. 39 at 3-4; Pltfs.' Ex. 5 at 11-13, 75-78; Pltfs.' Exs. 17, 18, 38 and 40.

Weitkamp testimony, Tr. 11/4/80 at 215-216; Pltfs.' Ex. 18 at 12, 22.

Fish kills due to supersaturation have occurred below dams in other parts of the country, and supersaturated conditions in violation of the EPA water quality criteria of 110% (water will normally be at 100% saturation) are not uncommon below dams. The Army Corps of Engineers has found that supersaturated conditions exist in streams below many of its dam/reservoir projects.

Pltfs.' Ex. 39 at 10.

Pltfs.' Ex. 58d at 2, 6-7. See also Pltfs.' Ex. 41.

Supersaturation can be predicted based upon the characteristics of the flow over the spillway and of the stilling basin, the water temperature, barometric pressure, and the dissolved gas levels in the reservoir. Changes in the design and/or operation of dams can prevent a supersaturation problem.

Pltfs.' Ex. 8.

Does the creation of supersaturated water by dams constitute the discharge of a pollutant under the Clean Water Act?

The essential difference between the parties in this case is that plaintiffs give a broad reading to the coverage of the NPDES program, arguing that any pollution ("the man-made or man-induced alteration of the chemical, physical, biological, and radiological integrity of water") emitted from a point source is a pollutant for which an NPDES permit must be obtained. Defendants on the other hand, contend that the NPDES program applies only to the pollutants listed in the Act, and then only if they are brought into the water from outside ("added") by a point source. Defendants do not dispute that heat, sediments, and dissolved metals come within the definitional list of pollutants, and would be subject to NPDES permitting if added to navigable waters from a point source. But low dissolved oxygen, cold, and supersaturation, in their view, can never be pollutants. In any case, defendants argue that all of the pollutants alleged by plaintiffs are changes in water quality occurring within navigable waters, namely reservoirs, or in the case of supersaturation, stilling basins, and cannot be considered pollutants added from a point source. The dam, as the purported point source, merely passes on water of already altered quality. A reservoir or stilling basin is part of the navigable water of a river, and a river cannot be said to be a point source polluting itself.

CWA §§ 502(19), 33 U.S.C. § 1362(19).

CWA § 502(6), 33 U.S.C. § 1362(6); see pp. 1296-1297, supra.

Nothing in the Clean Water Act or its legislative history addresses the question of whether dam-created pollution was intended to be regulated under the NPDES program. It does appear that the typical situation the legislators had in mind for the NPDES program was an industrial facility or a sewage treatment plant taking waste materials produced outside of the water and discharging them into water, essentially treating the waterway as a waste receptacle. Yet, in view of the broadly remedial purpose of the Act, "to restore and maintain the chemical, physical, and biological integrity of the Nation's waters", with the goal of eliminating the discharge of pollutants into water by 1985 and achieving fishable and swimmable water quality by 1983, it would disserve those mandates to give the Act a constricted reading or assume that because a particular problem was not specifically addressed in the legislative history, it was not meant to be subject to the federal regulatory program. In fact the courts as well as EPA have subjected any activity discharging pollutants from a discernible conveyance to the NPDES program, whether or not it fits the industrial or sewage treatment model, and whether or not there is evidence that the application of the NPDES program to the activity was specifically contemplated by the Congress. United States v. Earth Sciences, Inc., supra (sump overflow at mining operation); Sierra Club v. Abston Construction Co., Inc., 620 F.2d 41 (5th Cir. 1980) (sediment basin overflow and erosion of piles of discarded material at coal mine); Natural Resources Defense Council, Inc. v. Costle, 568 F.2d 1369 (D.C. Cir. 1977) (point sources from silvicultural and agricultural operations, confined animal feeding operations, irrigation return flows, and storm sewers); Applachian Power Co. v. Train, 545 F.2d 1351, 1373-1374 (4th Cir. 1976) (point source construction site run-off). In short, the NPDES program was intended to be comprehensive, and to cover any situation encompassed by the statutory language, whether or not the particular application was contemplated by Congress at the time of enactment.

CWA § 101(a), 33 U.S.C. § 1251(a).

The Supreme Court has said of Section 13 of the Rivers and Harbor Act of 1899, the predecessor to the NPDES program, "We read the 1899 Act charitably in light of the purpose to be served. The philosophy of the statement of Mr. Justice Holmes in State of New Jersey v. State of New York, 283 U.S. 336, 342 [ 51 S.Ct. 478, 479, 75 L.Ed. 1104] (1931), that `A river is more than an amenity, it is a treasure,' forbids a narrow, cramped reading of § 13. . . ." United States v. Republic Steel Corp., 362 U.S. 482,491, 80 S.Ct. 884, 890, 4 L.Ed.2d 903 (1960). The Sixth Circuit in United States v. Hamel, 551 F.2d 107 (6th Cir. 1977) applied this principle to the CWA, giving a broad interpretation to the definition of "pollutant", holding that it includes gasoline despite the fact that it is not specifically mentioned.

When a particular problem has not been addressed by Congress, the court must extrapolate congressional intent from areas where the expression of intent is clear, applying the more generally expressed intent to the specific issue before it. Montana Power Co. v. Federal Power Commission, 445 F.2d 739, 746 (D.C. Cir. 1970) ( en banc), cert. denied, 400 U.S. 1013, 91 S.Ct. 566, 27 L.Ed.2d 627 (1971); Portland Cement Association v. Ruckelshaus, 486 F.2d 375, 380, 383 cert. denied, 423 U.S. 1025, 96 S.Ct. 469, 46 L.Ed.2d 399 (1975). The Court will adopt this procedure in examining the application of NPDES program to dams. While the parties' dissection of the meaning of the words "pollutant", "discharge" and "point source" is relevant to the inquiry, ". . . the [CWA] must be given a reasonable interpretation, not parsed and dissected with the meticulous technicality applied in testing a common law indictment or a deed creating an estate in fee tail." Natural Resources Defense Council v. Costle, Inc., 564 F.2d 573, 579 (D.C. Cir. 1977). The statutory language will be given a common sense interpretation in harmony with the discernible intent of Congress.

There is no question but that Congress regarded the NPDES program as the heart of the Act and the most effective means of controlling water pollution. The 1972 FWPCA Amendments were enacted precisely because the existing water pollution control programs had failed, and a federally-controlled permit system based on effluent limitations was regarded as the remedy.

See pp. 1296-1297, supra; Environmental Protection Agency v. Caifornia ex rel. State Water Resources Control Board, 426 U.S. 200, 202-205, 96 S.Ct. 2022, 2023-25, 48 L.Ed.2d 578 (1976). In the Senate consideration of the 1972 Act, Senator Jackson stated that the failure of previous water pollution control laws because of lack of strict enforcement and meaningful deterrents would be remedied in this bill by the permit program. 1 Leg.Hist. at 216-217. Senator Muskie, the chairman of the subcommittee reporting out the bill, described effluent limits as "the best available mechanism to control water pollution." 2 Leg.Hist. at 1254. He also characterized the two major changes in the nation's water pollution control effort as the change from water quality standards to effluent limits as the control mechanism, and the use of the permit system against the discharge of pollutants, both features of the NPDES program. 2 Leg.Hist. at 1259. Senator Eagleton echoed this appraisal in the floor debate, remarking that the significant shift in focus reflected in the bill was the emphasis on the direct control of effluents. 2 Leg.Hist. at 1274. Senators Baker and Cooper also described the NPDES program as the most important feature of the bill. 2 Leg.Hist. at 1280, 1304. The Senate Committee Report declares that "[u]nder this Act the basis of pollution prevention and elimination will be the application of effluent limitations." 2 Leg.Hist. at 1426.

The NPDES system is the preferred method of control, and it appears that Congress would have put all pollution sources under that program had it been feasible. The Act generally divides pollution sources into two categories, point source and nonpoint source. EPA has characterized a nonpoint source in this litigation as "nothing more than a pollution problem not involving a discharge from a point source." Point sources fall under the NPDES program, while the regulation of nonpoint sources is left to states and localities under Section 208 areawide waste treatment management plans. EPA issues guidelines for identifying and controlling pollution from nonpoint sources, but has no direct authority over nonpoint source control. Point and nonpoint sources are not distinguished by the kind of pollution they create, or the kind of activity that creates the pollution, but rather on the basis of whether the pollution reaches the water through a confined, discrete conveyance. Although the statute lists runoff from agricultural, silvicultural, mining and construction activities as nonpoint sources, as we have seen, these sources are considered point sources subject to the NPDES program when the pollutants from those activities are emitted from a discernible conveyance. United States v. Earth Sciences, Inc., supra; Sierra Club v. Abston Construction Co., Inc., supra; Natural Resources Defense Council v. Costle, 568 F.2d 1369 (D.C. Cir. 1977). The only logical reason for exempting major sources of pollution from the regulatory program considered to be the most effective tool for pollution control in the CWA solely on the basis of the means of conveyance to the water is the infeasibility of subjecting such sources to a permit program based upon effluent limitations. When pollutants from, for example, a farm field, forest, or construction site are carried off by rainwater and joined by pollutants from other sources before they reach a waterbody, there is no way to measure the pollutants contributed by each source to enable enforcement of an effluent limitation. In contrast, it is a relatively simple matter to measure pollutants discharged from a pipe or other defined point. Also, the kinds of controls applicable to point source pollution, such as treatment of the wastes to remove pollutants or render them less harmful before they are returned to the water, changes in operational processes to reduce pollutant production, and collection of the pollutants for disposal on land, generally depend upon having the pollutants collected in one place which is under the source owner's control before they are discharged into water. These methods therefore generally cannot be applied to nonpoint sources. Nonpoint sources controls are of a different nature, for example, changes in farming and forestry practices to reduce erosion and land use controls. The NPDES program, which relies upon determining a technological process that will achieve maximum pollutant reduction from a particular type of source, and then monitoring discharges for compliance with effluent limitations, is simply inappropriate to nonpoint sources, at least based upon our current level of knowledge and control techniques. "It is clear from the legislative history that Congress would have regulated so-called nonpoint sources if a workable method could have been derived; it instructed the EPA to study the problem and come up with a solution." United States v. Earth Sciences, supra, 599 F.2d at 373.

Memorandum in Support of Defendant Costle's Motion to Strike Post-Trial Brief at 2.

CWA § 208(b)(2)(F)-(K), 33 U.S.C. § 1288 (b)(2)(F)-(K).

CWA § 304(f), 33 U.S.C. § 1314(f).

See Natural Resources Defense Council v. Costle, 564 F.2d 573, 580 (D.C. Cir. 1977).

CWA § 208(b)(2)(F)-(I), 33 U.S.C. § 1288 (b)(2)(F)-(I).

Congress recognized the significant contribution of nonpoint sources to water pollution, and the absolute necessity for their control if the goals of the CWA were to be met. 1 Leg.Hist. at 796; 2 Leg.Hist. at 1457.

Summarizing what can be discerned of congressional intent, the highly ambitious purpose of the CWA is to restore and maintain the integrity of the Nation's waters, with the goal of eliminating all pollutant discharges into water, and an interim goal of the achievement of fishable and swimmable water. The primary, and preferred method for achievement of this goal is the NPDES program. Nonpoint sources, the basic category of sources not subject to the NPDES program, have been exempted, it appears, solely because it would be impossible to apply an effluent limitation program to them. Within this framework, we must interpret the language of the Act, whenever possible and reasonable, to subject pollution sources to NPDES control. Defendants' argument that Congress intended only the definitional list of pollutants, strictly interpreted, to be controlled under the NPDES program, while all other problems subsumed under the broader definition of pollution ("the man-made or man-induced alteration of the chemical, physical, biological, and radiological integrity of water"), are relegated to other programs under the CWA and other federal and state laws, whether or not emitted from a point source and whether or not amenable to technological control and effluent monitoring, fails in the light of the obvious congressional preference for the NPDES program as the most effective means of pollution control. Exemptions of pollution sources from the NPDES program may not be based solely upon one possible interpretation of the statutory language when that interpretation frustrates the known purposes and policies of the Act.

There are some other minor exemptions which may be based on other considerations.

It is against this background then that we must construe the statutory language limiting the application of the NPDES program to the discharge of pollutants, defined as "any addition of any pollutant to navigable waters from any point source." It is not disputed that the pollution problems described by plaintiffs must come within this definition to be subject to NPDES control. The question is whether plaintiffs' or defendants' interpretation of this language is more consistent with the meaning of the words and the purpose and policies of the Act.

CWA § 502(12), 33 U.S.C. § 1362(12).

Putting aside for a moment the issue of whether some of the conditions described by plaintiffs qualify as "pollutants", we decide whether they are added to navigable waters from a point source. Defendants place great weight on the word "addition", claiming that dams do not add pollutants to navigable waters since the pollutants involved are already in the water, and in the case of low dissolved oxygen and cold, actually involve subtractions, rather than additions, of either oxygen or energy from the water. Also, a reservoir or stilling basin cannot be part of a point source (which the plaintiffs view as a dam/reservoir facility) which adds pollutants to navigable waters since it is itself a portion of the navigable water. Plaintiffs counter that the word "addition" is synonymous with "creation"; and that dam/reservoir facilities add pollutants to navigable waters because they create pollutants that would not exist but for the dam and reservoir. Further, plaintiffs contend, the word "addition" should not be interpreted so literally as to exclude the creation of pollutants which, technically speaking, subtract something from the water. The definition says that the point source must add pollutants, not that the pollutants must be "additions." The addition of water low in dissolved oxygen or reduced in temperature could clearly qualify. To the contention that a reservoir or stilling basin is simply a portion of navigable waters, plaintiffs respond that a dam project, which stores and utilizes navigable waters to achieve its purposes, such as power generation, flood control, or irrigation, and then returns the water to the stream below, should not be treated differently under the CWA from a power or industrial plant which diverts water from a river and uses it for cooling, cleansing or other purposes, and then returns the water to the stream with added pollutants. The latter situation is described in Appalachian Power Co. v. Train, supra, which held that a plant is responsible under the NPDES program for the pollutants it adds to the waters passing through it, but not for pollutants naturally occurring in the waterway or previously added by other sources. Likewise, a dam/reservoir project should be responsible for the pollutants it adds to the waters which pass through it.

The court in South Carolina Wildlife Federation v. Alexander, 457 F. Supp. 118 (D.S.C. 1978), in denying defendant's motion to dismiss, held that the release of water changed in quality because of the man-made impoundment of water could constitute the "addition" of pollutants to navigable waters.

The plaintiffs' interpretation of the statutory language is plausible, consistent with the common sense meaning of the words, and supportive of the policies and goals of the legislation. Defendants' interpretation, in addition to taking pollution sources out of the regulatory program preferred by Congress with no policy justification, is the more tortured. It gives words like "addition" an overly literal and technical meaning, and strains to characterize a man-made and operated dam/reservoir facility as merely part of a river discharging pollutants into itself through natural processes.

Defendants' interpretation also results in the anomaly of taking dams, which admittedly can be point sources and admittedly can emit pollutants, out of the NPDES program. Since the CWA cannot achieve its goals without attacking all pollution sources, dams cannot simply be ignored. EPA has resolved this dilemma by treating dams, which obviously fit the definition of "point source", as nonpoint sources. This clearly is not the most reasonable reading of the statutory language.

EPA has taken the awkward position that a "nonpoint source" is only a negative category for anything which does not meet the definition of a point source, (Memorandum in Support of Defendant Costle's Motion to Strike Post-Trial Brief), and that dams can be point sources, (Tr. 11/5/80 at 113), yet still must be treated as nonpoint sources, because under EPA's interpretation of the words, they do not "discharge pollutants." Even when a dam is a "confined, discrete conveyance" of substances classified as pollutants, it is classified as a nonpoint source unless the pollutants it conveys are brought in from outside the water for release through the dam.

We will therefore accept plaintiffs' reading of the meaning of "discharge of pollutants," and proceed to examine the application of the NPDES program to the undisputed pollutants discharged by dams: sediments, dissolved metals, and heat.


The sediments collected in a reservoir derive from upstream sources on land and from the channel banks and bed, as do sediments in free-flowing streams. They also derive from the land that was flooded by the reservoir. Nonetheless, it is the construction and operation of the dam/reservoir facility which causes the sediments to be trapped, and then possibly discharged downstream, as by sluicing, in far greater than natural quantities. This excess load of sediments, created by the dam and reservoir, and purposefully discharged through a point source, should be subject to the NPDES program. Construction or mining sites which by their activities merely augment the natural process of erosion are subject to NPDES permitting when they discharge the accumulated sediments from that erosion through a discrete conveyance. Similarly, when a reservoir modifies the natural process of the transport of sediments, and then discharges the accumulated sediments through the dam, it should be subject to NPDES regulation. EPA itself so recognized at least once, in requiring an NPDES permit for the sluicing of Guernsey Reservoir. The fact that sluicing operations may be unusual does not militate against the application of the permit program; it simply means that not many permits may be required. The problem is clearly amenable to point source controls, in that the dam operator can forego sluicing, control the amount of sediment released, or remove excess sediments by dredging, a procedure which can be accomplished without releasing excess sediments downstream. Dissolved metals

Pltfs.' Ex. 30; see p. 1301, supra.

In accordance with Appalachian Power Co. v. Train, supra, a dam owner would not be constrained from discharging sediment in quantities and at rates that were normal to the free-flowing stream.

Simons testimony, Tr. 11/4/80 at 80-81.

Although dissolved metals released by dams originally derive from sediments added to the stream from upstream sources and from native materials at the site, they are changed from the particulate to the dissolved form by the reservoir. Dissolved metals which did not exist before are created by the reservoir, and passed downstream through the dam. They can cause pollution problems which are not caused by the metals in their particulate form. The reservoir also adds these metals to navigable waters in the sense that they in part derive from areas which were dry land before innundation by the reservoir, and so entered navigable waters when the dam was built and impounded the river. The release of water containing these metals from a dam in greater quantities than existed in the free-flowing river constitutes the discharge of a pollutant from a point source, subject to NPDES permitting.

See discussion p. 1299, supra.


Some dams discharge water warmer than that of the previously free-flowing stream, because they release water from the upper layer of a thermally-stratified reservoir. The warm water releases are the product of the combination of thermal stratification, which does not exist in a free-flowing stream, and is caused by its impoundment by the dam, and release from the top layer, which is a result of the structure and operation of the dam. There is no doubt but that the dam created, and added, the pollutant. The process, and the impact, are comparable to a power plant which takes in water from a river and returns it in a warmer state, a situation which is unquestionably subject to NPDES permitting. Heat discharges from dams can be controlled technologically, by the use of multiple outlet works, by destratification of the reservoir, or by other means. The discharge of water of higher temperature than existed in the free-flowing state is subject to the NPDES program.

It remains to address the conditions which defendants claim do not fit the definition of "pollutant": low dissolved oxygen, cold, and supersaturation. The general question is whether the list of pollutants in CWA § 502(6), 33 U.S.C. § 1362(6) is intended to strictly limit the discharges subject to NPDES permitting, or is merely an hodgepodge attempt, not intended to be exclusive, at listing all of the substances and conditions causing pollution when discharged into water. The list is an odd assortment of terms, as specific as "cellar dirt" and as general as "industrial, municipal and agricultural waste." The legislative history is absent of guidance as to why the list was constructed in this manner, or why certain substances or conditions were or were not included. The definition goes on to specifically exclude sewage from vessels, and water, gas or other materials injected into wells in connection with oil or gas production, the latter of which appears not to be covered by the definition in the first place.

To repeat, the definition is: "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."

As discussed earlier, the definition will be construed broadly to further the purpose of the Act to eliminate all forms of water pollution. At least two courts have interpreted the Act in the manner urged by plaintiffs, including any form of pollution emitted by a point source in the definition of pollutant. In United States v. Earth Sciences, Inc., supra, 599 F.2d at 373, the court stated, "We believe it contravenes the intent of FWPCA and the structure of the statute to exempt from regulation any activity that emits pollution from an identifiable point." In determining, on a motion to dismiss, that low dissolved oxygen and high metallic concentrations in water discharged from dams could be considered "pollutants" under the CWA, the South Carolina district court reasoned that the conditions described by plaintiffs certainly comprised "pollution" as defined in the Act, and that "no reasonable purpose would be served by admitting pollution while denying the existence of a pollutant." South Carolina Wildlife Federation v. Alexander, supra, 457 F. Supp. at 125. While not going so far as to include all pollution emitted by point sources in the definition of "pollutant", (which was not necessary to its decision), the court in United States v. Hamel, supra, concluded that the failure to include petroleum products in the definition of pollutant did not evidence an intent to exclude them from the Act's prohibition of pollutant discharges without a permit. Since the definition of pollutant had been based on the 1899 Refuse Act's prohibition of the discharge of refuse matter of any kind (except liquid sewage) into navigable water, it was seen as encompassing at the minimum what was covered under the 1899 Act. The Supreme Court had found in United States v. Standard Oil Co., 384 U.S. 224, 86 S.Ct. 1427, 16 L.Ed.2d 492 (1966) that aviation gasoline was within the proscription of the 1899 Act, and that it was undoubtedly a pollutant.

"Bearing in mind that the FWPCA is not a model of clarity, when interpreting it we must consider the statutory scheme as a whole, including the object and policies contained within." United States v. City of Redwood City, 640 F.2d 963, 969 (9th Cir. 1981).

Only one case holding that a dam does not discharge "pollutants" under the CWA has come to our attention, and it is inapposite. Defendants have submitted to the Court the unreported opinion in State of Missouri v. Department of the Army, Corps of Engineers, C.A. No. 78-309 2-CV-S (W.D. Mo. 1980), holding that the operation of the Stockton dam and reservoir project "does not involve or result in the `discharge of a pollutant' as that term is defined in the FWPCA." Slip Op. at 36. The Missouri court, however does not explain how it reached this conclusion, except to say that the dislodging of silt and other pollutants from the downstream river bank caused by the rise and fall of the water level resulting from the operation of the dam's hydroelectric generator does not constitute "runoff of a pollutant" within the meaning of the FWPCA. The only conditions alleged by plaintiff which might be considered pollutants discussed in the opinion are the increased downstream erosion just referred to, and low dissolved oxygen, which the court found had been returned to acceptable levels by the construction of a skimming weir in the reservoir. Slip Op. at 22-23. That case, therefore, does not concern any of the pollutants alleged in this case. Its unexplicated conclusion that one particular dam does not discharge pollutants cannot lend support to any more general determinations concerning the discharge of pollutants from dams.

Further foundation for plaintiffs' interpretation comes from the statute itself, and EPA's implementation of it. In the 1977 Amendments to the Act, Congress directed EPA to publish information and guidelines concerning "conventional pollutants" regulated under the NPDES program, including pollutants classified as biological oxygen demanding ("BOD" is a measure of the amount of dissolved oxygen consumed through biochemical oxidation when the discharge is added to the waterbody), suspended solids, and pH. None of these are specifically listed in the definition of pollutant, although some of the constitutents causing biological oxygen demand and changes in acidity/alkalinity (pH) and some of the components of suspended solids are listed pollutants. However, pollutants not on the list can also be causes or constituents of these "pollutants." It therefore appears from the statute itself that the definitional list in CWA § 502(6) is not exclusive.

CWA § 304(a)(4), 33 U.S.C. § 1314(a)(4).

Defendants correctly argue that pollution parameters such as BOD, pH, and total suspended solids (tss), are not pollutants added to water, but are indicators of the presence of other pollutants. For example, biological materials and sewage create biological oxygen demand, and the discharge of acidic industrial wastes can cause changes in the pH of the water. Yet, by using pollution parameters as effluent limitations, EPA controls not only listed pollutants which may contribute to these measurements, but also any unlisted emissions so contributing. For illustration, nitrogen and phosphorous, although naturally occurring elements, and not included in the definitional list, can act as nutrients contributing to BOD. It is true that the pollution parameters are often applied to industrial discharges, which would come under the definitional term "industrial waste" no matter what they contain. However, this need not always be the case. Examine the situation of sediments and other materials derived from run-off and collected at a construction site. They are treated as pollutants when discharged from a discernible conveyance. This discharge cannot be considered industrial, agricultural, or municipal waste, and its constitutents, although perhaps partially consisting of listed materials such as biological materials and sand, do not consist entirely of listed pollutants. It is treated as a pollutant basically because it contributes to the total suspended solids parameter, which, when it reaches a certain level, has deleterious effects. The use of pollution parameters indicates that EPA is looking toward the pollution effect on the water, such as the total sediment load or the depletion of dissolved oxygen, rather than at whether the constituents of the discharge strictly conform to the definitional list. In FMC Corp. v. Train, 539 F.2d 973, 982-83 (4th Cir. 1976), EPA defended its use of chemical oxygen demand (COD) as a pollution parameter based on the definition of pollution as "the man-made or man-induced alteration of the chemical, physical, biological, and radiological integrity of water", and the court upheld EPA's action on that basis.

Further obvious evidence of EPA's broad construction of the definition of "pollutant" is the fact that defendants, because EPA has treated such as pollutants in other contexts, have not disputed here that sediments and metals, though not specifically listed, can be pollutants.

In short, the case law, the statute itself, and EPA's implementation of it all controvert defendants' position here that the definitional list of pollutants is exclusive.

Having established that pollution discharges need not be specifically listed in CWA § 502(6) in order to be regulated as "pollutants", it remains to examine the specific pollutants alleged here, both as to whether they are statutory pollutants, and as to whether they are added to navigable waters from a point source. There appears to be no logical reason to exclude cold from NPDES coverage when heat is a listed pollutant, thereby making the discharge of warmer, but not colder water subject to NPDES permitting. Both warm and cold discharges are caused by the operation of the dam, (i.e. added by the point source), and are controllable by the same techniques, such as the use of multiple outlet works or destratification.

Oxygen deficient water is the very condition that BOD effluent limitations are designed to prevent. It would serve no purpose to use a BOD parameter to regulate all substances causing water to be depleted of oxygen while leaving unregulated the discharge of oxygen deficient water from dams. The layer of oxygen deficient water is created by the reservoir in the first place, and allowed to travel downstream only because of the peculiar structure and operation of the dam which discharges from the lower layer of the reservoir. The dam/reservoir facility is a point source which creates water low in dissolved oxygen, just as point sources discharging oxygen-demanding materials create oxygen deficient water. Low DO discharges can be controlled by technological changes applied to the structure and operation of the dam/reservoir facility, such as the use of multiple outlet works and injection of oxygen into the discharge water.

Just as elements such as nitrogen and phosphorous, entirely natural and harmless under most circumstances, are considered pollutants when added to water because they contribute to biological oxygen demand, the entrainment and solution of naturally occurring atmospheric gases in water caused by the operation of a dam and resulting in supersaturation lethal to aquatic life should logically be considered a pollutant. There is no question but that the operation of the dam creates the pollutant and adds it to navigable water, and that effluent limits could be achieved by the application of technology, either in the design of dams before they are built, in their operation, as in controlling the amount of water released over the spillway, or by added control features such as spillway deflectors.

See pp. 1302-1303, supra.

It should be noted that plaintiffs have not argued, and the Court certainly does not hold, that all water quality changes occurring in man-made reservoirs constitute discharges of pollutants subject to the prohibition of CWA § 301(a). Congress has directed EPA to issue guidelines for identifying and controlling nonpoint source pollution caused by "changes in the movement, flow, or circulation of any navigable waters . . . including changes caused by the construction of dams . . .", CWA § 304(f)(2)(F), 33 U.S.C. § 1314(f)(2)(F), evidencing that dams cause nonpoint as well as point source pollution. A dam is a point source only when it discharges pollutants created by the dam/reservoir facility into navigable waters. Poorly oxygenated water and accumulated sediment deposits at the bottom of a reservoir are caused by changes in the movement, flow and circulation of the water due to the construction of the dam, but are not pollutants subject to NPDES permitting unless discharged from a point source, namely the dam. As long as the oxygen-depleted water or the sediments remain in the reservoir, they are subject only to nonpoint source controls, except in the sense that discharges of sediment or of oxygen demanding material from upstream point sources which contribute to these conditions are subject to NPDES permitting of those sources.

In concluding that EPA has violated the Clean Water Act in its refusal to regulate dams as point sources of the pollutants described by plaintiffs under the NPDES program, we are not unmindful of the deference normally afforded an agency in the interpretation of a statute which it administers. Udall v. Tallman, 380 U.S. 1, 16, 85 S.Ct. 792, 801, 13 L.Ed.2d 616 (1965). While a court need not find that the construction of the statute by the agency is the only reasonable one, or the one it would have reached had the question first arisen in judicial proceedings in order to affirm the agency's decision, Id., the court cannot simply echo an agency's construction and do no more. "Our duty is to see that the congressional mandate is carried out, and misinterpretation of that mandate will be corrected. . . . Where the Administrator's interpretation of the [CWA] coincides with the purpose and intent of the statute we defer to his construction. Where the issue presented involves questions of scientific expertise or a choice between reasonable interpretations of the statute we defer to the Administrator's interpretation; however no amount of deference can justify an interpretation of the statute that is contrary to law." Natural Resources Defense Council, Inc. v. U.S. Environmental Protection Agency, 656 F.2d 768, 774 (D.C. Cir. 1981).

The statutory interpretation involved here does not require scientific expertise, but rather is an interpretation of ordinary language used by Congress to identify the kinds of pollution it intended to control through the NPDES program. As we have seen, EPA's interpretation of the statute runs counter to expressed congressional intent, and is inconsistent with its own implementation of the Act in other contexts. EPA has given absolutely no reasonable basis, consistent with the purpose and policies of the Act, why dams should not be regulated as point sources. Its entire argument rests upon a dissection of the language of the Act, particularly the definition section, in an overly technical manner which has been rejected in the context of this broadly remedial legislation. Natural Resources Defense Council, Inc. v. Costle, 564 F.2d 573 (D.C. Cir. 1977). Under these circumstances, EPA's interpretation cannot be accepted merely because it is the agency administering the CWA.

It remains to consider the post-trial brief of defendant-intervenors Colorado Water Agencies, the National Water Resources Association, the American Water Works Association, the National Association of Water Companies and the Association of California Water Agencies, (Water Group), which does propound policy reasons for not regulating dams under the NPDES program, based upon the alleged division of state and federal authority in the CWA. Defendant EPA has stated its emphatic disagreement with the position of the Water Group, and asked that its post-trial brief be stricken. We allowed the post-trial brief to be filed, but deferred the question of its relevance until the time of determination of the merits. It cannot be said that legal arguments supporting the conclusion that dams should not be regulated by the NPDES program (which were addressed at trial in the testimony of Dr. Jeris A. Danielson,) are irrelevant to this action and should not be considered. Yet, while relevant, the defendant-intervenors' arguments are not persuasive. Essentially the Water Group argues that the mention of dams as a nonpoint source in CWA § 304(f)(2)(F), 33 U.S.C. § 1314(f)(2)(F), combined with expressions of congressional intent in the CWA and in other contexts to preserve the rights of states to allocate quantities of water within their borders, requires the conclusion that Congress intended dams to be regulated as nonpoint sources under the state-controlled § 208 program and under state law. Congress would so provide, they argue, because the operation of dams is essential to the system of water allocation rights in the arid West. There are two major fallacies in this argument. The first is that the reference to dams as a nonpoint source in § 304(f)(2)(F) means they can never be point sources. EPA's position, which has been affirmed by the courts, is that the nonpoint sources listed in § 304(f)(2) are to be considered point sources when they emit pollutants from discernible, discrete conveyances. United States v. Earth Sciences, Inc., supra; Sierra Club v. Abston Construction Co., Inc., supra; Natural Resources Defense Council, Inc. v. Costle, 568 F.2d 1369 (D.C. Cir. 1977). Therefore the inclusion of dams in CWA § 304(f)(2)(F) has no bearing whatsoever on the issue in this case, and is certainly not evidence that Congress intended to exclude dams from the NPDES program to protect state water rights. The second fallacy is that Congress expressed an intent to limit federal control of water quality in order to preserve state rights to allocate water quantity. The Water Group relies for this proposition primarily on CWA § 101(g), 33 U.S.C. § 1251(g), which sets forth the policy that "the authority of each State to allocate quantities of water within its jurisdiction shall not be superseded, abrogated or otherwise impaired by this Act. It is further the policy of Congress that nothing in this Act shall be construed to supersede or abrogate rights to quantities of water which have been established by any State. . . ." However, it is clear from the remarks of Senator Wallop, who offered the amendment which became § 101(g), that the purpose of the provision was to prevent unwarranted interference with state water rights under the guise of water quality regulation. The amendment was not intended to change existing law or interfere with incidental effects on water allocation stemming from legitimate federal water quality regulation. Assuming arguendo that the application of the NPDES permit system to dams would affect state water allocation systems, a proposition for which no supporting evidence has been introduced, this nonetheless would not negate the applicability of the federal permit program.

Memorandum in support of Defendant Costle's Motion to Strike Post-Trial Brief.

Tr. 11/5/80 at 25-86.

3 Leg.Hist. at 531-532.

Plaintiffs have proved that some dams discharge pollutants from a point source, subjecting them to the requirement of the CWA to obtain an NPDES permit for such discharges, and that therefore defendant Gorsuch has violated her non-discretionary duty to designate dams as a point source category subject to NPDES requirements. Judgment will be entered in favor of plaintiffs, and appropriate relief will be granted.

This decision is made in full cognizance of the enormous number of dams in this country (over two million total, including 50 to 60 thousand large dams) and their importance for many purposes, including flood control, water storage, irrigation, navigation and power production. Fortunately, as the parties recognize, the granting of this relief will not force EPA to require a permit for every dam in the United States. Many dams may cause no pollution problems whatsoever, and there are administrative options available to EPA, such as categorical exemptions, areawide permits and general permits, which can minimize the burden on both the Agency and dam owners and operators. Those dams that do or may discharge pollutants must be regulated according to law.

"Large" dams are those 25 or more feet in height with more than 15 acre-feet of storage capacity, or at least 6 feet in height with storage capacity of over 50 acre-feet. Joint Statement of Material Facts No Longer in Dispute ¶ 2a.

An appropriate judgment accompanies this memorandum opinion.


It is this 29th day of January, 1982, hereby

ORDERED, that declaratory judgment that defendant Anne Gorsuch, in her capacity as Administrator of the United States Environmental Protection Agency, has violated a non-discretionary duty under §§ 301(a) and 402(a) of the Clean Water Act, 33 U.S.C. § 1311(a) and 1342(a), and has acted arbitrarily and in excess of her statutory authority in failing to regulate dams as point sources of pollution under the National Pollutant Discharge Elimination System mandated by § 402(a) of the Clean Water Act, be and it hereby is entered in favor of plaintiff National Wildlife Federation and plaintiff-intervenor State of Missouri and against defendant Anne Gorsuch in her capacity as Administrator of the United States Environmental Protection Agency and against defendant-intervenors Alabama Power Company; Allegheny Power Company; American Electric Power Company, Inc.; Appalachian Power Company; Arkansas-Missouri Power Company; Arkansas Power Light Company; Baltimore Gas Electric Company; Boston Edison Company; Carolina Power Light Company; Central Illinois Light Company; Central Illinois Public Service Company; Cincinnati Gas Electric Company; Cleveland Electric Illuminating Company; Columbus Southern Ohio Electric Company, Commonwealth Edison Company, Connecticut Light Power Company; Consolidated Edison Company of New York, Inc.; Dallas Power Light Company; Dayton Power Light Company; Delmarva Power Light Company; Detroit Edison Company; Duke Power Company; Edison Electric Institute; Florida Power Light Company; Georgia Power Company; Gulf Power Company; Hartford Electric Light Company; Holyoke Water Power Company; Houston Lighting Power Company; Illinois Power Company; Indiana-Kentucky Electric Corporation; Indiana Michigan Electric Company; Indiana Michigan Power Company; Indianapolis Power Light Company; Iowa Public Service Company; Kansas City Power Light Company; Kentucky Power Company; Long Island Lighting Company; Louisiana Power Light Company; Madison Gas Electric Company; Middle South Utilities, Inc.; Mississippi Power Company; Mississippi Power Light Company; Monongahela Power Company; Montaup Electric Company; National Rural Electric Cooperative Association; New England Power Company; New Orleans Public Service, Inc.; New York State Electric Gas Corporation; Niagara Mohawk Power Corporation; Northeast Utilities; Northern States Power Company; Ohio Edison Company; Ohio Electric Company; Ohio Power Company; Ohio Valley Electric Corporation; Oklahoma Gas and Electric Company; Pacific Gas and Electric Company; Pennsylvania Power Light Company; Philadelphia Electric Company; Potomac Edison Company; Potomac Electric Power Company; Public Service Electric Gas Company; Public Service Company of Indiana, Inc.; Rochester Gas Electric Corporation; San Diego Gas Electric Company; South Carolina Electric Gas Company; Southern California Edison Company; Southern Company Services, Inc.; Tampa Electric Company; Texas Electric Service Company; Texas Power Light Company; Texas Utilities Generating Company; Toledo Edison Company; Union Electric Company; Virginia Electric and Power Company; West Penn Power Company; Western Massachusetts Electric Company; Wisconsin Electric Power Company; Wisconsin Power and Light Company; Wisconsin Public Service Corporation; Idaho Power Company; Montana Power Company; Puget Sound Power Light Company; Benton Rural Electric Association; Big Bend Electric Cooperative, Inc.; Blachly-Lane Electric Cooperative Association; Central Electric Cooperative, Inc.; Clearwater Power Company; Columbia Rural Electric Association, Inc.; Consumers Power, Inc.; Coos-Curry Electric Cooperative, Inc.; Eugene Water and Electric Board; Inland Power Light Company; Kootenai Electric Cooperative, Inc.; Lane Electric Cooperative, Inc.; Lincoln Electric Cooperative, Inc.; Lower Valley Power Light, Inc.; Midstate Electric Cooperative, Inc.; Orcas Power Light Company; Portland General Electric Company; Public Utility District No. 1 of Chelan County, Washington; Public Utility District No. 1 of Cowlitz County, Washington; Public Utility District No. 1 of Douglas County, Washington; Public Utility District No. 2 of Grant County, Washington; Raft River Rural Electric Cooperative, Inc.; City of Seattle, Department of Lighting; City of Tacoma, Department of Public Utilities; Umatilla Electric Cooperative Association; The Washington Water Power Company; The Association of California Water Agencies; Colorado River Water Conservation District; Southwestern Water Conservation District; Northern Colorado Water Conservancy District; City and County of Denver, City of Aurora, Colorado; City of Colorado Springs, Colorado; Board of Water Works; City of Pueblo, Colorado; American Water Works Association; National Association of Water Companies and National Water Resources Association, and it is

FURTHER ORDERED, that defendant Anne Gorsuch in her capacity as Administrator of the United States Environmental Protection Agency designate dams as a point source category under § 402 of the Clean Water Act, 33 U.S.C. § 1342; establish effluent limitations or other performance standards for dams on a categorical, as opposed to a case-by-case basis; and subject existing and proposed dams to all the National Pollutant Discharge Elimination System requirements applicable to other categories of point sources, and it is

FURTHER ORDERED, that final regulations implementing the above order be issued no later than 90 days from this date.


Plaintiff National Wildlife Federation and plaintiff-intervenor State of Missouri have requested an award of costs and attorneys fees, pursuant to 33 U.S.C. § 1365(d), section 505(d) of the Clean Water Act, which provides for awards of the costs of litigation, including reasonable attorney and expert witness fees, to any party whenever the court determines such award is appropriate. If plaintiffs wish to pursue their request, they must submit to the Court a detailed accounting and justification for the amounts requested within 15 days of this date. Responses by defendants, if any, shall be filed within 15 days of plaintiffs' filing.


Summaries of

National Wildlife Federation v. Gorsuch

United States District Court, D. Columbia
Jan 29, 1982
530 F. Supp. 1291 (D.D.C. 1982)

indicating that penstocks in hydroelectric dams may constitute point sources

Summary of this case from Natl. Wildlife Fedn. v. Consumers Power
Case details for

National Wildlife Federation v. Gorsuch

Case Details

Full title:NATIONAL WILDLIFE FEDERATION, et al., Plaintiffs, v. Anne GORSUCH…

Court:United States District Court, D. Columbia

Date published: Jan 29, 1982


530 F. Supp. 1291 (D.D.C. 1982)

Citing Cases

National Wildlife Federation v. Gorsuch

The National Wildlife Federation petitioned the district court for a declaration that the Administrator of…

Natl. Wildlife Fedn. v. Consumers Power

The district court in Gorsuch, moreover, found that although the parties were unable to agree to stipulated…