Power Plant Cooling Water, Hudson River Fish - Again

New York Law Journal

February 23, 2007
by Stephen L. Kass, Jean McCarroll and Ethan I. Strell

The recent decision of the Second Circuit Court of Appeals in Riverkeeper v. U.S. Environmental Protection Agency (Riverkeeper II),[1] remanding a substantial portion of EPA’s final regulations concerning cooling water intake structures at large existing power plants, reminds us of where modern environmental law began.

The harm to fish, fish eggs, and larvae that would be drawn into the power plant structure the Consolidated Edison Co. proposed for Storm King Mountain on the Hudson River in the 1960s was one of the principal environmental issues raised by the petitioners in Scenic Hudson Preservation Conference v. Federal Power Commission,[2] the environmental litigation that spawned many of the environmental statutes (including the National Environmental Policy Act) (NEPA) and court decisions that underlie U.S. environmental law.

Following the Second Circuit’s Scenic Hudson decisions in 1965 and 1971, Congress enacted the Clean Water Act.[3] Section 316(b) of the act provides for the regulation of cooling water intake structures by requiring “that the location, design, construction, and capacity of cooling water intake structures reflect the best technology available for minimizing adverse environmental impact.”[4] Section 316(b) has been the subject of considerable litigation and delay since its enactment in 1972. The EPA first attempted to regulate cooling water intake structures in 1976, but in a challenge by the power industry, the regulations were set aside on procedural grounds. Environmental groups eventually sued the EPA and won a consent decree, under which the EPA was required to issue rules in three phases: Phase I to concern new facilities, Phase II (at issue in Riverkeeper II) to concern large, existing power plants, and Phase III to concern existing power plants not covered under Phase II.

The Phase I regulations for new facilities were issued in 2001. Although most of those regulations withstood legal challenge, some were set aside by the Second Circuit in Riverkeeper v. Environmental Protection Agency (Riverkeeper I).[5]  The Phase II regulations at issue in Riverkeeper II were published in the Federal Register on July 9, 2004.[6]

Impacts of Cooling Water Withdrawals

By the EPA’s own account, the withdrawal of cooling water, nearly half of which is by the electric power industry, has profound environmental impacts on aquatic environments. These include reductions of phytoplankton, zooplankton, fish, and shellfish, including the loss of threatened and endangered species; damage to aquatic organisms, including important elements of the food chain; reduction of populations and their compensatory reserves; losses of commercial and recreational fisheries; and stresses to overall communities and ecosystems.

These impacts result principally from “entrainment” of small organisms into a plant’s cooling system and “impingement” of larger organisms against a plant’s intake screen by the force of the water entering the cooling system.[7]

Three Types of Cooling Systems

Most existing plants use “once-through” cooling, which takes water in to absorb heat and returns it to the waterbody at a higher temperature. This is relatively inexpensive and efficient for power generation, but results in high entrainment and impingement losses. “Closed-cycle” systems recirculate the cooling water, cooling it off in a reservoir or tower, and use far less water than once-through systems, but can be expensive and unsightly. Dry cooling systems use air to transfer heat and use little or no water, but, according to the EPA, cost more than 10 times per year as much as closed-cycle wet cooling and reduce water intake by only slightly more than closed-cycle cooling. Dry cooling also requires more energy, results in greater air emissions, and is less effective in warmer climates.

The Phase II Regulations

As noted above, § 316(b) directs that the “cooling water intake structures use “the best technology available for minimizing adverse environmental impact.” The Clean Water Act, however, provides little guidance for the EPA in determining what constitutes that technology, and the term appears nowhere else in the act.

Rather than requiring existing plants to employ a particular type of technology, such as closed cycle cooling, the Phase II Rule attempts a more flexible regulatory approach by setting forth six alternatives[8] that a facility may use for complying:

(1)(a)      Demonstrating that the facility’s flow is or will be reduced commensurate with a closed-cycle system;

(1)(b)   Demonstrating that the facility has or will reduce the water velocity at the intake screen to a maximum of 0.5 feet/second;

(2)               Demonstrating that the facility’s existing design and construction technologies, operational measures, and/or restoration measures meet certain performance standards or restoration measures described in the rule;

(3)               Demonstrating that the facility has selected and will install and properly operate and maintain design and construction technologies, operational measures, and/or restoration measures that will meet certain performance standards described in the rule;

(4)               Demonstrating that the facility has installed or will install an “approved design and construction technology;” and

(5)               Demonstrating that the plant qualifies for a site-specific variance. The grounds for a variance is either that the cost of complying with the other alternatives (a) “would be significantly greater than the costs considered by the Administrator [for a similar facility] in establishing the applicable performance standards…. (the “cost-cost” variance), or (b) “would be significantly greater than the benefits of complying with the applicable performance standards….” (the “cost-benefit” variance).[9]

The principal “performance standards” are: (1) reducing impingement mortality of fish and shellfish by 80 to 90 percent, and (2) reducing entrainment by 60 to 90 percent.[10]

Facilities choosing certain alternatives can comply based on either the implementation of a “Technology Installation and Operation Plan” (TIOP), indicating how the facility will install and ensure the efficacy of its design and construction technologies, its operational measures, or a “restoration plan.” A restoration plan would allow off-site restoration measures that “will produce ecological benefits (fish and shellfish), including maintenance or protection of community structure and function in [a] facility’s waterbody or watershed, at a level that is substantially similar to the level [the facility] would achieve by meeting the applicable performance standards….”[11] The regulations state that if a facility remains in compliance with the TIOP, it may request that compliance for the next permit term be determined solely by remaining in compliance with the TIOP.[12]

Riverkeeper II

In Riverkeeper II, EPA’s Phase II regulations were challenged on two fronts. Environmental groups and eastern states[13] asserted that the regulations were under-protective, while three groups of industry petitioners[14] alleged that they went too far. The Second Circuit agreed with many of the states’ and environmental groups’ claims and with virtually none of the industry groups’ claims.

Best Technology Available

The states’ and environmental groups’ principal argument was that EPA improperly considered costs in rejecting closed-cycle cooling as the “best technology available” in favor of a suite of alternative technologies. The court generally agreed, holding that, although § 316(b) does not necessarily require the selection of one technology, the Clean Water Act does require a “technology-forcing” result with a high level of environmental benefit. The court concluded “that the language of 316(b) itself plainly indicates that facilities must adopt the best technology available and that cost-benefit analysis cannot be justified in light of Congress’s directive.”[15]  However, while § 316(b) does not permit cost-benefit analysis, the EPA may consider the cost-effectiveness of its regulation. Thus, the EPA must first determine what the benchmark “best” technology is. “Once this determination has been made, the EPA may then consider other factors, including cost-effectiveness, to choose a less expensive technology that achieves essentially the same results as the benchmark.”[16] The court held that the basis for EPA’s selection of technologies in the record was unclear, and remanded the Phase II Rule to EPA for clarification of the basis for EPA’s decision and, potentially, for a reassessment of “best technology available.”

Standards as Ranges

The environmental petitioners and states argued that the Phase II rule’s performance standards - - reduction of impingement mortality by 80 to 90 percent and reduction of entrainment by 60 to 90 percent - - were impermissible because the statute requires a single standard. The court held that, although expressing performance standards as ranges was allowable because measuring impingement and entrainment reduction is imprecise, the regulation nevertheless was invalid because it did “not require facilities to choose technologies that produce the greatest reduction possible.”[17] In other words, the rule must establish a high benchmark and then could allow for a margin of error from that high standard, rather than allowing facilities to be in compliance with the low end of a range when a higher standard or performance is achievable.

Restoration Measures

As noted above, the Phase II rule allows facilities to meet performance standards through “restoration” measures, such as restocking fish and habitat improvement projects. EPA had included a similar restoration provision in the Phase I rule for new facilities, which the Second Circuit had already invalidated in Riverkeeper I, holding that restoration measures were “plainly inconsistent with the statute’s text and Congress’s intent in passing the 1972 amendments.”[18]

Quoting its decision in Riverkeeper I, the court stated that, “however beneficial to the environment, [restoration measures] have nothing to do with the location, the design, the construction, or the capacity of cooling water intake structures, because they are unrelated to the structures themselves.’ ‘Restoration measures correct for the adverse environmental impacts of impingement and entrainment,’ we noted, but ‘they do not minimize those impacts in the first place.’”[19]


The Phase II rule includes two variance provisions. The first, a “cost-cost” variance, allows the EPA to make a “site-specific determination of the best technology available for minimizing adverse environmental impact,” if data specific to a facility “demonstrate that the costs of compliance… would be significantly greater than the costs considered by the Administrator” for a similar facility.[20] The second variance, a “cost-benefit” variance, also allows EPA to make a site-specific determination of best technology available if “the costs of compliance… would be significantly greater than the benefits of complying with the applicable performance standards….”[21]  The court remanded the cost-cost variance on procedural grounds and the cost-benefit variance on substantive grounds.

In remanding the cost-cost variance, the court found that the EPA failed to comply with the notice and comment provisions of the Administrative Procedure Act because the proposed rule had estimated compliance costs for model plants, while the final rule assigned cost estimates to specific, named facilities. The court held that the final rule was not a “logical outgrowth” of the proposed rule, and that using actual costs was an impermissible “surprise switcheroo.”[22]

The court invalidated the cost-benefit variance on two similar substantive grounds. First, it held that in enacting § 316(b), Congress did not permit cost-benefit considerations. Second, the court held that the cost-benefit variance constituted a “water-quality based standard” that is inconsistent with the overall thrust of the Act, which enacted a technological approach because of the difficulties of proving that a particular polluter had caused the water quality to dip below the applicable standard.

The TIOP Provision

The Phase II rule permits compliance with the performance standards by implementing a TIOP. The court invalidated this provision on two procedural grounds: first, the EPA gave inadequate notice of the provision, and second, the TIOP depends on the selection of the best technology available, for which EPA’s determination was also remanded. The court found inadequate notice because the proposed rule indicated that a facility could be in compliance for a limited duration (one permit duration of up to five years) by implementing a TIOP, whereas the “TIOP provision in the final rule. . . appears to permit a facility to satisfy the rule’s requirements in subsequent permit terms, for an indefinite period, without ever demonstrating compliance with the performance standards, so long as the facility has adhered to its TIOP….”[23]

Industry Arguments

Three industry groups also challenged the Phase II Rule. The only successful industry challenge was to a provision of the rule governing a facility that purchases cooling water from an “independent supplier.” One of the industry groups, the Utility Water Act Group, alleged that the EPA provided inadequate notice that the Phase II Rule would impose Phase II requirements on Phase III facilities that supply cooling water to Phase II facilities. The court agreed and remanded. The industry petitioners also challenged other aspects of the Phase II rule, but the Court either rejected those arguments or dismissed them for lack of jurisdiction.[24]

Future Challenges

Given the glacial pace at which the § 316(b) regulations have been implemented by the EPA to date and the range of issues that the court remanded in Riverkeeper II, both renewed administrative hearings and judicial challenges almost certainly lie ahead for the Phase II Rule. At some point, we suspect, Congress might also address these issues, not to introduce cost-benefit analysis into § 316(b), but to permit the EPA to consider non-aquatic environmental impacts, such as air emissions and energy efficiency, in selecting the best technology for cooling water systems in U.S. power plants.

Stephen L. Kass and Jean M. McCarroll, together with Clifford P. Case, direct the Environmental Practice Group at Carter Ledyard & Milburn. Ethan I. Strell, an associate at the firm, assisted in the preparation of this article.

Reprinted with permission from the February 23, 2007 edition of The New York Law Journal
© 2007 ALM Properties, Inc. All rights reserved. Further duplication without permission is prohibited.

[1]   F.3d, No. 04-6692-ag (L), 2007 WL 184658 (2d Cir. Jan. 25, 2007). 
[2] 354 F.2d 608, 623-24 (2d Cir. 1965), cert.denied sub nom. Consolidated Edison Co. of New York v. Scenic Hudson Preservation Conference, 384 U.S. 941 (1966); 453 F.2d 463, 469, 476-77 (2d Cir. 1971), cert. denied, 407 U.S. 926 (1972). 
[3] Federal Water Pollution Control Act, 33 U.S.C. §§ 1251-1387 (“Clean Water Act” or the “act”). 
[4] 33 U.S.C. § 1326(b). 
[5] 358 F.3d 174 (2d Cir. 2004). 
[6] 69 Fed. Reg. 41576 (July 9, 2004). 
[7] Although much of the data on impingement and entrainment is somewhat dated and imprecise, the EPA estimates that entrainment and impingement result in the annual loss of over 3.4 billion fish and shellfish, with the greatest losses in the mid-Atlantic region. Studies done in the 1980s considering the water withdrawals of five Hudson River power plants predicted year class reductions ranging from 6 to 79 percent, depending on the species - - for example, between 20 and 43 percent for striped bass, bay anchovy, and Atlantic tomcod. Entrainment by these facilities also removed food from, and altered the composition of food in, the food chain by killing and crushing fish and by leaving less prey available for larger species and comparatively more food available for lower trophic level organisms, disrupting the entire aquatic community. 
[8] The rule says there are five alternatives, but in reality the first of the five includes two different approaches. 
[9] 40 C.F.R. § 125.94(a).
[10] 40 C.F.R. § 125.94(b).
[11] 69 Fed. Reg. 41591-92 (July 9, 2004). 
[12] 40 C.F.R. §125.94(d)(2). 
[13] Riverkeeper, Inc., Natural Resources Defense Council, Waterkeeper Alliance, Soundkeeper, Inc., Scenic Hudson, Inc., and several other “keeper” groups; Rhode Island, Connecticut, Delaware, Massachusetts, New Jersey, and New York. 
[14] Entergy Corporation, the Utility Water Act Group, PSEG Fossil LLC, and PSEG Nuclear LLC. 
[15] Riverkeeper II, slip op. at 22.
[16] Id. at 24.
[17]  Id. at 35.
[18] Riverkeeper I, 358 F.3d at 189. 
[19] Riverkeeper II, slip op. at 39-40 (citations omitted; brackets and emphasis in original). 
[20] 40 C.F.R. § 125.94(a)(5)(i). 
[21] Id. § 125.94(a)(5)(ii). 
[22] Riverkeeper II, slip op. at 46. 
[23] 54.
[24] The industry petitioners alleged that (1) § 316(b) does not apply to existing facilities; (2) the definition of “adverse environmental impact” is not supported by the record; (3) EPA’s assumption of zero entrainment survival is not supported by the record; (4) the rule did not account for a disproportionate impact on nuclear facilities; and (5) EPA did not provide adequate notice of its definition of “Great Lakes.” All were rejected by the Second Circuit. 

© Copyright 2020 Carter Ledyard & Milburn LLP