COMMENTS TO NJPDES PERMIT NO. 0005622

CONCERNING § 316(B) OF THE CLEAN WATER ACT

SUBMITTED BY THE MID-ATLANTIC ENVIRONMENTAL LAW CENTER

ON BEHALF OF:

AMERICAN LITTORAL SOCIETY; CLEAN OCEAN ACTION; COALITION AGAINST TOXICS; COALITION FOR PEACE & JUSTICE; DELAWARE AUDUBON SOCIETY; DELAWARE RIVERKEEPER; DELAWARE SIERRA CLUB; GREEN OCEAN SOCIETY; NEW JERSEY ENVIRONMENTAL FEDERATION; NEW JERSEY PUBLIC INTEREST RESEARCH GROUP; NEW JERSEY SIERRA CLUB; STOCKTON PEACE ACTION; SURFERS' ENVIRONMENTAL ALLIANCE; UNPLUG SALEM ALLIANCE

Section 316(b) of the Clean Water Act, 33 U.S.C. §1326(b), requires that the "location, design, construction, and capacity of a cooling water intake structure reflect best technology available for minimizing adverse environmental impact." There is no question that Public Service Electric and Gas Co.'s ("PSE&G") Salem Nuclear Generating Station ("Salem") SNGS is a point source with a cooling water intake structure ("intake structure") subject to § 316(b). What is at issue is whether permit No. NJ0005622 ("Permit") requires the design, capacity and location to "reflect best technology available" ("BTA") for "minimizing adverse environmental impact."

Commentators submit that the Permit is not consistent with CWA § 316(b) for two fundamental reasons. First, the Permit impermissibly considers costs in rejecting true BTA at Salem. Particularly in light of last month's unanimous Supreme Court Opinion in Whitman v. American Trucking, it is improper for the Permit to consider the costs of BTA at all in making a BTA determination because § 316(b) does not allow consideration of costs. Even if costs were relevant, the Permit should have assessed the marginal increase in costs to ratepayers for requiring BTA, instead of rejecting BTA based in part on PSE&G's flawed comparison of total costs of conversion solely to the commercial and recreational worth of lost fish. In any event, by disregarding societal costs and benefits, NJDEP misapplies the cost/benefit test.

Second, the Permit does not "reflect" BTA. NJDEP agrees that Salem is having adverse environmental effects. Yet the Permit fails to "minimize" such impacts. To minimize such impacts, the Permit should require a technological conversion at Salem that adopts or reflects BTA, which is "dry cooling," a type of technology that reduces adverse impacts by as much as 99 percent. The Permit's purported BTA conditions are not BTA. They do nothing to reduce entrainment losses, which account for 99 percent of Salem's adverse impact, and reduce impingement losses under the best circumstances, by only 0.2 percent.

Following a background intended to supplement that found about the permit in NJDEP's Fact Sheet, this comment has four parts. Part One discusses why NJDEP erred in considering costs. Part Two supports NJDEP's findings that Salem is having adverse environmental impacts, but disagrees that the Permit minimizes these impacts; and, Part Three submits that Salem must adopt or reflect dry cooling -- a technology NJDEP did not consider -- because it is BTA. Part Four explains how the design, capacity and location of Salem's intake do not reflect BTA.

I.        BACKGROUND TO NJPDES Permit No. 0005622

Along the estuarine banks of one of the most ecologically important flyways and ecosystems in the world, Public Service Electric and Gas Company (PSE&G) operates the Salem Nuclear Generating Station (Salem), one of the largest powerplants in the world. The SNGS is located along the Delaware River Estuary at Artificial Island on the eastern shore of the Delaware River in Salem County, New Jersey, approximately fifty miles northwest of the mouth of Delaware Bay and thirty miles southwest of Philadelphia.[1] Salem has two nuclear reactors, Units 1 and 2, which came on-line in 1977 and 1981, respectively.[2] Both Units 1 and 2 employ once-through cooling. Once-through cooling results in the continuous withdrawal of water from, and the discharge of wastewater into, the Delaware Estuary.[3]

Salem withdraws cooling water from the estuary by means of a cooling water intake structure virtually unparalleled in size anywhere else in the world. The intake structure has twelve intake bays with associated screens that rotate at an average of 0.9 inches per second, and twelve water pumps.[4] Each pump has a capacity of 266 million gallons per day, giving Salem a cooling water intake capacity of 3.2 BGD.[5] After use, Salem's cooling water is discharged into the Delaware Estuary near the New Jersey-Delaware state line.[6]

The SNGS has been operating (and withdrawing water) under the aegis of the CWA's National Pollutant Discharge Elimination System (NPDES) permit program since the early 1970's. In the early 1980's, EPA delegated NPDES permitting authority to New Jersey, and with it, authority over Salem's CWA permit. PSE&G has been operating Salem pursuant to an NJPDES permit since September 1, 1994, which expired August 31, 1999, but has since been administratively extended.[7]

In 1978, EPA (which at the time had permitting authority over Salem) issued a consent decree requiring PSE&G to submit a Section 316(b) demonstration.[8]  Thereafter, EPA provided PSE&G with technical guidance to assist the company with submission of the Section 316(b) demonstration.[9]  PSE&G subsequently submitted its first Section 316(b) demonstration in 1984.[10]  Following delegation of the NPDES program to New Jersey in 1982, the State undertook the task of evaluating the demonstration.[11]

On October 3, 1990, DEP issued a Draft Permit (1990 Draft Permit) that adopted Versar's findings, and proposed requiring Salem to convert to a closed-cycle cooling system.[12]  On June 24, 1993, however, DEP issued a second draft permit for Salem (1993 Draft Permit) that reconsidered the 1990 Draft Permit. The 1993 Draft Permit proposed a mitigation experiment and associated measures in lieu of the more stringent technology-based controls (i.e., a closed-cycle cooling system) required by the 1990 Draft Permit.[13]

Following extensive public comment, DEP issued a final permit (1994 Permit) on July 20, 1994, which allowed PSE&G to continue to operate Salem with little change. The 1994 Permit's conditions ostensibly combine both technological and mitigative means for attempting to minimize adverse environmental impacts. The 1994 Permit contains seven special conditions.[14] NJDEP contended that three of these conditions represent compliance with Section 316(b): (1) reduction of the permitted intake flow of Salem from its maximum design capacity to its maximum actual operating capacity;[15] (2) modification of the design of the screens on Salem's intake structure;[16] and (3) development and implementation of a plan to study the feasibility of deterring fish from entering the area surrounding Salem's intake structure through the use of underwater speakers and/or sound projectors.[17]

Non-BTA conditions were included purportedly as further means of minimizing adverse environmental impact.[18] These included: (1) a wetlands restoration and enhancement program in and around the Delaware Estuary;[19] (2) establishment of a Management Plan Advisory Committee to provide technical advice to PSE&G concerning the development and implementation of the mitigation project;[20] (3) spending at least $425,000 to build and maintain five fish ladders at approved locations for tributaries to the estuary;[21] and (4) conducting a baywide biological monitoring study.[22] To help execute the biological monitoring study, DEP required PSE&G to establish a Monitoring Advisory Committee.[23] The Committee was empanelled to advise PSE&G regarding biological monitoring design, implementation, modifications and results interpretation.[24] Instead of explicitly infusing Salem's permit with a condition requiring that Salem comply with Section 316(b) (as the New York DEC had in Hudson Riverkeeper Fund), NJDEP required: "With respect to Section 316(b), the Department's determination [at the time of permit reissuance, scheduled for September 1, 1999] will include, but not be limited to, an evaluation of whether technologies, their costs and benefits, and potential for application at the Station have changed."[25] The 1994 permit produced two challenges.

On or about August 19, 1994, both the State of Delaware and a group of fourteen environmental, conservation, and fishing interests (Coalition) appealed issuance of the 1994 Permit.[26]  The State and the Coalition's legal arguments were, in principle, aligned.  They each argued that the 1994 Permit did not ensure that the design, capacity, and location of Salem's intake structure reflect BTA for minimizing adverse environmental impact in accordance with Section 316(b);[27] that the costs of certain technology-based compliance measures not required by DEP were not wholly disproportionate to the environmental benefits conferred by such measures;[28] and that Salem's intake structure was having an undue impact on interstate waters.[29] 

PSE&G settled both challenges. PSE&G's settlement with the Coalition included requirements to reevaluate its fish screen and return systems, and evaluate newly-developed technologies to reduce impingement and entrainment. PSE&G's settlement with the State of Delaware's required it to spend roughly $10 million on mitigation and restoration projects in Delaware waters over the term of the permit. It also agreed to review Salem's intake structure problems with fish entanglement with marsh grasses or other debris found in the intake screens and to test new technology designed to reduce associated fish mortality if necessary, and to fund any NJDEP review of PSE&G's evaluation of alternative intake technologies for the Salem plant.

In March of 1999, PSE&G submitted its permit renewal application.[30]  The permit application consists of 36 volumes and 167 volumes of reference material.[31] Due to the large amount of materials, “[NJDEP] contracted the services of ESSA Technologies, Ltd. of Richmond Hill, Ontario, Canada, for those issues associated with Section 316(b) of the Clean Water Act . . .”[32] Under the terms of its settlement, PSE&G paid for these services.

NJDEP announced the issuance of a draft permit for public comment on or about December 28, 2000 (hereinafter, "2000 draft permit"). In the draft permit, NJDEP "determined that [SNGS's] existing once-through cooling system in conjunction with an intake flow limitation, an enhanced fish return system and the study and potential implementation of a multi-sensory hybrid system constitutes best technology available.”[33]

II.        NJDEP'S BTA DETERMINATION MAY NOT CONSIDER COSTS AND IN ANY EVENT THE COSTS OF BTA ARE NOT WHOLLY DISPROPORTIONATE TO THE ENVIRONMENTAL BENEFITS CONFERRED

NJDEP states "BTA is intended to mean the best technology available commercially at an economically practicable cost and, further, that the costs of a technology must not be wholly disproportionate to the environmental benefit to be gained.[34] NJDEP then posits that it -- and not PSE&G -- shoulders the burden of demonstrating that any given BTA is cost effective.[35]

Applying this standard, NJDEP "determined that the Station's existing once-through cooling system in conjunction with an intake flow limitation [Special Condition 1], an enhanced fish return system [Special Condition 2, as modified] and the study and potential implementation of a multi-sensory hybrid system [Special Condition 3, as modified] constitutes best technology available."[36] NJDEP says that "the Department is committed to requiring implementation of any cost-effective alternate technologies that will minimize impingement an/or entrainment effects,"[37] but evidently accepts PSE&G's position "that the costs of any [other technological] alternatives would be wholly disproportionate to any benefits provided," and thus that "none . . . is available for Salem."[38]

NJDEP's use of cost considerations to influence its BTA determination is incorrect in both invocation and application. Section 316(b) neither requires nor allows costs considerations in making BTA determinations. Section 316(b) of the CWA requires that an intake structure "reflect [BTA] for minimizing adverse environmental impacts," and NJDEP engrafts onto the statute what is plainly not there. Assuming arguendo that costs are germane, NJDEP misapplies cost considerations by employing a cost/benefit ratio approach instead of evaluating the marginal rate increases ratepayers would have to bear if BTA were installed at Salem. This misstep proves to be outcome predictive and one that avoids application of BTA at Salem. Applying the correct cost methodology, assuming the most expensive technology (closed-cycle cooling), complete pass-through of costs to ratepayers, and no other sources of revenue, the maximum rate increase to PSE&G's ratepayers would be only one and one-tenth of one percent (1.1), or about $1.00 per month per ratepayer with an average bill of $100.00 per month. Contrary to NJDEP's contention, legal precedent holds that such costs are certainly not wholly disproportionate to environmental impacts. Assuming arguendo a cost/benefit ratio approach is allowed, NJDEP misapplies it by failing adequately to consider societal benefits of BTA. 

            A.    NJDEP's BTA Determination May Not Consider Costs

In contrast with other aspects of the CWA that specifically contemplate economic considerations when addressing adverse environmental impacts, §316(b) allows no such thing.[39] By omitting any reference to costs, Congress simply did not wish for costs to be considered.[40]

Section 316(b) of the Clean Water Act requires that “the location, design, construction, and capacity of cooling water intake structures reflect the best technology available for minimizing adverse environmental impact.”[41] The statutory language makes no direct or inferential reference to economic considerations. ." [42] If Congress intended economics to be a factor in determining BTA, it would have specifically worded the statute to do so; NJDEP may not engraft such considerations at Salem.

The most recent statement of controlling law reiterates this basic tenet of administrative law that an agency may not consider costs in implementing a law unless Congress allows it to do so. See Whitman v. American Trucking Co. et al. (Feb. 27, 2001), 2001 U.S. LEXIS 1952, *17 (Scalia, J.)[43] In Whitman, the Court considered whether §109(b)(1) of the Clean Air Act (CAA) allows EPA to "consider the costs of implementation in setting national ambient air quality standards (NAAQS) under Section 109(b)(1)."[44] Section 109 of the CAA requires the Administrator inter alia to set primary ambient air quality standards “the attainment and maintenance of which . . . are requisite to protect the public health” with “an adequate margin of safety.”[45] Respondents American Trucking argued, among other things, that the phrases "public health" and "adequate margin of safety" allowed EPA to consider the costs of setting the standards.[46]

By unanimous decision, the Court disagreed and ruled that not only may not agencies consider costs, but they must not do so, unless Congress explicitly allows it: "the textual commitment [to consider costs] must be a clear one.[47] Noting that the provisions at issue "say not a word about costs," the Court said: "we find it implausible that Congress would give to [an agency] … the power to determine whether implementation costs should moderate national air quality standards."[48] (“Congress . . . does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provisions – it does not, one might say, hide elephants in mouseholes.”[49]). Absent language allowing consideration of costs, the Court does not even believe it is a close call, observing:

Were it not for the hundreds of pages of briefing respondents have submitted on the issue, one would have thought it fairly clear that this text does not permit the EPA to consider costs … [t]he language is 'absolute.' … Nowhere are the costs of achieving such a standard made part of that initial calculation.[50]

Thus, Congress does not intend for an agency to consider significant issues like costs unless it explicitly says so. In the words of the Court, this "unambiguously bars cost considerations … and thus ends the matter for us as well as [an agency]."[51]

The same holds for §316(b). It says "not a word about costs." CWA § 316(b), like CAA § 109(b)(1), § 316(b), is a technology-forcing[52] provision that does not allow EPA, or a state with delegated authority to issue NPDES permits with conditions under § 316(b), to consider costs. [53] Nowhere is NJDEP given the power to determine whether costs should moderate implementation of §316(b). It was improper for the agency to do so here.

That this is so is made all the more clear by noticing, as the Court did in American Trucking, that Congress is well able to say when and how costs may be considered. Indeed, In American Trucking, the Court noted the string of provisions in the CAA in which Congress allows for consideration of implementation costs.[54] Citing a string of cases fundamental in Administrative Law, the Court thus "refused to find implicit in ambiguous sections of the CAA an authorization to consider costs that has elsewhere, and so often, been expressly granted."[55]

Again, the same holds true for §316(b) and the CWA. As with § 109(b)(1) of the CAA, CWA § 316(b) can easily be distinguished from other provisions that "specifically contemplate economic considerations when addressing adverse environmental impacts."[56] For example, Congress required 'best technology economically achievable' in defining how effluent standards for specified categories of pollution sources are to be created pursuant to CWA § 301(b)(1)(A),  33 U.S.C. § 1314(b)(1)(B), and 'best available technology economically achievable for implementing standards for discharges of toxic pollutants. 33 U.S.C. § 1317(a). Section 316(b), however, makes no mention of costs.

Last, assuming the "legislative history" of a single statement linking § 316(b) and economic considerations even relevant [57]  -- which American Trucking says it is not -- the statement does not establish whether or how costs are to be considered otherwise, and in any event cannot be inflated to re-write the provision as enacted.[58]

B.            Considering Marginal Cost Increases Attributable to BTA Would Have Supported Closed-Cycle Cooling

Assuming, arguendo that an economic analysis is appropriate for CWA §316(b), NJDEP should have assessed the marginal costs of BTA to ratepayers, instead of considering total costs versus total economic benefit. EPA has applied the "wholly disproportionate" test as a marginal cost increase -- and not a total cost ratio -- analysis. Instead, NJ used a total cost versus benefit ratio as the guiding principle of the "wholly disproportionate" test. This is incorrect. Had NJDEP applied the correct test, it would have to require PSE&G to convert to closed-cycle cooling.

A cost-benefit analysis is clearly out of bounds.[59] Based on the Fourth Circuit's holdings in Appalachian Power Co. v. Train[60] and E.I. du Pont de Nemours & Co. v. Train,[61]  PSCO argued in Seabrook II that a BTA determination should be accompanied by a formal cost/benefit analysis.[62] Consistent with its own guidance, [63] EPA found in Seabrook II, that "[t]here is nothing in Section 316(b) indicating that a cost/benefit analysis should be done."[64]

Instead, the standard that EPA has applied when it has considered costs is whether the cost of the technology to be required is wholly disproportionate to the environmental benefit to be gained.[65] In Seabrook II, EPA remarked that "it is [not] reasonable to interpret Section 316(b) as requiring use of technology whose cost is wholly disproportionate to the environmental benefit to be gained," reasoning that "some consideration ought to be given to costs . . . otherwise the effect would be to require cooling towers at every plant . . . regardless of whether any significant degree of entrainment or entrapment [impingement] was anticipated."[66]

Nonetheless, rather than turning to a cost/benefit ratio, the "wholly disproportionate test" requires an evaluation of the marginal rate increase to ratepayers of BTA.[67] EPA has ruled that § 316(b) cost considerations require consideration of the marginal cost to each ratepayer of implementing a proposed technology, and not the total cost of the technology to a utility.[68] In Brunswick, for example, EPA found that while a reduction in capacity of 96% would cost a total of $106,300,000,[69] the monthly marginal cost to each residential ratepayer would be a mere $.77 to $.85 per month, or an increase of approximately 2.5%.[70]  When EPA compared this 2.5% rate increase with the 96% reduction in the adverse environmental impact associated with conversion to a closed-cycle system, EPA concluded that "an increase of approximately 2.5% from the . . . average monthly residential [electric] bill is not wholly disproportionate to a 96% reduction in the severe adverse environmental impacts of the plant."[71]

Applying a marginal cost increase test here would have established that the cost of closed cycle cooling is not wholly disproportionate to the environmental benefits. Indeed, applying this test, NJDEP has found that the marginal cost of closed cycle cooling at Salem is not wholly disproportionate to environmental benefits.[72] NJDEP has acknowledged such costs to be a mere $0.20 per month ($1989).[73] The Permit should have considered such marginal cost increases.

Natural draft towers save about 86% of the fish for only a 1.1% increase in revenue, [74] thus equating, at the most, a 1.1 percent increase in rates.[75] Assuming the typical PECO or PSE&G ratepayer uses 1000 kWh a month and has a monthly bill of $100, the 1.1% revenue increase needed to install natural draft towers at Salem would translate to an per ratepayer increase of $1.10 per month, or $13.20 per year. The Permit should have considered such marginal cost increases. No such comparison can be attempted for dry cooling because NJDEP did not even consider it.

                        C.            Even Though a Cost/Benefit Test is Impermissible, NJDEP Misapplies It

In addition to being incorrect in the first place to consider a cost/benefit analysis, NJDEP nonetheless misapplies it. Although NJDEP makes a strong statement of its commitment, it incorrectly accepts, without question, the economic analysis conducted by PSE&G in support of its permit application.[76] PSE&G contends "[a] permitting authority cannot require an alternative as BTA when its economic and environmental costs are wholly disproportionate to its environmental benefits."[77] PSE&G studies six potential alternatives to the existing once-through intake structure at Salem and finds "that, in the case of each alternative, costs are wholly disproportionate to benefits."[78] PSE&G submits that "NJDEP's 1994 determination that the existing fish protections system is BTA for Salem"[79] is correct, and that "[t]here are no additional intake technologies or measures that are functionally suitable for adoption at Salem at an economic and environmental cost that is not wholly disproportionate to the environmental benefits provided."[80] 

To support its contention, PSE&G utilizes a cost-benefit ratio analysis (CBA).[81]  PSE&G's CBA calculates the cost of the alternatives by calculating: (1) the costs associated with construction and installation; (2) incremental operating and maintenance costs; and (3) the value of lost power at Salem as a result of construction and changes in continuing plant operations.[82]

PSE&G calculates the benefits of the alternatives by assigning a monetary value to the commercial and recreational fishing benefits expected if alternatives were implemented.[83]  The assigned values for commercial fishing range from $0.07 per pound for Atlantic menhaden to $3.05 per pound for striped bass.[84]  The assigned value for recreational fishing is $3.52 per pound.  The estimates of pounds of fish caught commercially and recreationally are based on 1990 to 1996 harvests.[85]

PSE&G then estimates the total cost for natural draft towers and mechanical draft towers to be $712.0 million and $849.2 million respectively.[86] PSE&G estimates the benefit of implementing closed cycle cooling to be a mere $58 million.[87] PSE&G further calculates the cost-benefit ratio for natural draft towers to be 12.3:1 and the ratio for mechanical towers to be 14.7:1. Applying these figures, PSE&G concludes that closed cycle cooling is not BTA for Salem because the costs associated with its installation are wholly disproportionate to the environmental benefits that would be gained.[88]

NJDEP thus concludes that the only alternative to the existing intake structure that is not wholly disproportionate to the environmental benefits gained is the potential installation of strobe light/air bubble curtain technology: "Therefore, the Department, has determined that study of a multi-sensory hybrid system, of which strobe light/air bubble is a component of, is an available technology at a cost which is not wholly disproportionate to the environmental benefits to be realized."[89] Nonetheless, NJDEP, however, does not require the installation of the strobe light/air bubble system. Instead, NJDEP curiously considers the study of a multi-sensory system part of the BTA determination.

PSE&G's CBA approach, however, is flawed because it fails adequately to take into account societal benefits. Economic theory states clearly that the appropriate level of analysis for all costs and benefits estimates is the societal level.[90] Thus, when considering the BTA, NJDEP should have considered "external" environmental costs of failing to implement a particular technology.[91]

Yet, as ESSA concludes, PSE&G's CBA is fundamentally flawed because it does not adequately reflect societal (so-called "external") costs and benefits.[92] A study of PSE&G's economic analysis supports ESSA's finding that the estimated costs for installing closed cycle cooling should be from the societal perspective.[93] "Social welfare increases when private parties who use the public's environmental goods and services face the full cost of their actions."[94] Anything less than full cost leaves unpaid the full cost of producing electricity, resulting in falsely low consumer costs and a depreciated appreciation of the environmental resources used to produce the electricity.[95] Cost estimates from the utility perspective are much higher than from the societal perspective.[96]

Unfortunately,  cost/benefit data used by PSE&G in their CBA "are not useful for conducting any test of proportionality between benefits and costs."[97] To correctly determine the price of environmental resources, both the active and passive component of the environment must be considered.[98] On average, for every dollar of lost active use, $1.90 of passive loss occurs.[99] The active component consists of actual use of the Delaware Estuary recreation such as fishing, swimming and boating. The passive component consists of two parts, option and existence values. Option value refers to an increase in use as conditions change, i.e., even if someone does not currently use the estuary, they may choose to use it (and may be willing to pay more to use it) as the environmental quality of the estuary enhances. Existence values refer to the amount that individuals are willing to pay to preserve the estuary out of a sense of responsibility.

PSE&G's cost-benefit analysis is inadequate because it limits the active component and does not address the passive component.[100] As discussed, to calculate the benefits of closed-cycle cooling, PSE&G simply assigns a monetary value to commercial and recreational fish caught. PSE&G underestimates the active benefit by ignoring the value realized by fishermen who enjoy fishing even if they do not catch anything, or practice catch and release.[101] This is inconsistent with the CWA. Section 101 of the CWA states:  "It is the national goal that wherever attainable, an interim goal of water quality which provides for the protection and propagation of fish, shellfish, and wildlife and . . . for recreation in and on the water should be achieved"[102]; § 316(b) of the CWA states that intake structures shall "reflect the best technology available for minimizing adverse environmental impact."[103]   

Additionally, the true active benefit of saving the aquatic life in the Delaware Estuary is the enhancement to such ecosystem services as detoxification and decomposition of waste; maintenance of biodiversity; food; recreation; and aesthetic beauty and intellectual stimulation that lift the human spirit.[104]  Yet PSE&G does not take these (and other) passive values into account and therefore incorrectly devalues the benefit of alternate technologies, such as closed cycle cooling.[105] EPA has established that as the environmental benefits associated with a technology increase, so too do the acceptable cost levels associated with a proposed technology.[106] Clearly, a principal priority in the implementation of the provisions of these laws is the protection of aquatic life and associated passive values.

Next, incorrectly inflated costs and underestimated benefits cause PSE&G's CBA ratios used by PSE&G to be inaccurate and artificially high, which in turn allows PSE&G to erroneously conclude that the cost of installing closed-cycle cooling is wholly disproportionate to the environmental benefits.[107]

Last, NJDEP inappropriately fails to consider to what extent PSE&G had reason to believe that such costs would be required when the plant was constructed.  For instance, during the course of the Administrator's review of the RA's ID in Brunswick II (requiring CP&L to reduce its intake capacity by 96%), CP&L claimed that a requirement to retrofit the Brunswick facility would be financially prohibitive.[108]  CP&L suggested that existing cooling systems should be subject to a different standard than new cooling systems.[109]  In response, without passing on the merits of the issue, EPA stated:

I suggest that it would be useful for the RA to include in his revised decision, . . . findings concerning the timing of construction of the once-through cooling system. Specifically, did CP&L have reason to believe cooling towers might be required when it constructed its once-through system, and if so, what alternatives were available to it?[110]

On notice for thirty years (since enactment of §316(b) in 1972) that a design change, or reduced capacity, or some other technology, may in the future be required, PSE&G should not be rewarded for ignoring that information and should instead be held responsible for technologies that were achievable, in the least, at the time of construction, or were foreseeable.

II.                The permit does not Minimize Adverse Environmental Impacts

A.             Salem's Intake Structure is Having an Adverse Environmental Impact

Section 316(b) is applicable to intake structures that are having an "adverse" environmental impact.[111] NJDEP considers "[t]he cornerstone of PSE&G's [permit] application [to be] the [adverse] impact assessment,"[112] and that "[m]any of the analyses including in [PSE&G's] application feed into [the adverse impact] assessment."[113]

EPA defines "adverse" to mean "unfavorable, harmful, difficult, or detrimental," but not "irreversible" or "irretrievable."[114] EPA maintains that any detrimental environmental impact is "adverse,"[115] including those inflicted by impingement or entrainment,[116] and that "fish losses of any given quantity" must be considered and "minimized" as required by the CWA.[117] EPA says that such adverse impacts occur "whenever there is entrainment or impingement damage as a result of the operation of a specific [intake structure]."[118] EPA uses both short-term and long-term criterion measurements to determine the extent of an intake structure's impact on the environment, viewing both absolute and percentage damage.[119] Consistently, NJDEP defines "adverse environmental impact" to be the death of any fish: "NJDEP and other states, such as New York, have considered the death of any fish at or through a cooling water intake structure to be an adverse [environmental] impact . . ."[120]

There is no question that Salem has an adverse impact; fish losses from Salem are astounding. The U.S. Fish and Wildlife service estimates combined loss of finfish and blue crab as a result of the operation of Salem to be 2.9 million kilograms per year.[121] This translates to approximately 5.5 million weakfish, striped bass, white perch, blueback herring, spot, over 800 million bay anchovy, and other fish due to impingement. In addition, 3,327.9 million fish are lost due to entrainment.[122] Worse still, USF&WS believe these figures undercount annual aquatic losses from Salem. Indeed, Salem is responsible for adult losses for herring, spot, and white perch [that] exceed the average commercial or recreational fishery for the Delaware Estuary for the period of 1975-1980 stretching from Trenton to the Atlantic Ocean, 120 miles away.[123]

NJDEP has agreed that "the [ongoing] adverse impacts of the SNGS are large and indicate the potential for substantial long-term population and ecosystem level impacts is great," [124] and that the "once-through cooling system employed at Salem adversely affects the Delaware Estuary and threatens the protection and propagation of balanced indigenous populations."[125]

Reading "adverse" to "consist[ ] of the harm to populations and communities, rather than individuals . . . .," however, PSE&G disputes that Salem is having any adverse effect. [126] PSE&G maintains that "adverse" requires evidence of the intake structure causing (1) imbalance in the indigenous populations of fish, and (2) jeopardy to the sustainability of fish stocks of Representative Important Species (RIS).[127] Employing this standard, PSE&G concludes that "there is no evidence that Salem's intake structure is causing or will cause an adverse environmental impact," and that there is "no justification or legal basis for requiring changes at the intake."[128]

PSE&G's position is clearly absurd.[129] As NJDEP remarks, "PSE&G and [NJDEP] do not agree regarding the definition of 'adverse environmental impact,'"[130] and has determined that "it is justified in requiring the pursuit of alternate intake protection technologies, to further minimize the impacts at the Salem Station."[131] NJDEP's interpretation of the term "adverse" to constitute the loss of "any fish" is consistent with EPA's interpretation, while PSE&G's is flatly not. As the State of Delaware's consultant commented, environmental harm is adverse long before it causes either an "imbalance" (whatever that means) to indigenous populations (whatever that is), or "jeopardizes" (read, extirpate or threaten) fish stocks of species PSE&G believes are "important."[132] ESSA rejects PSE&G's "population approach" and the notion that an impact is not adverse unless is causes an "imbalance" to or "jeopardizes" fish stocks on similar grounds: [133] 

Each of the three assessment endpoints chosen in the PSE&G Application (i.e., historical trends, long term sustainability, fish community structure) are confounded by changes in other stressors (i.e., water quality, changes in harvest).  Inferences made on these assessment endpoints are therefore dependent on historical and future assumptions regarding other stressors.  By contrast, assessment endpoints such as fish killed by entrainment and impingement and foregone production, are related directly to the impacts of the power station intakes, are less confounded by other factors, and require fewer assumptions about unknown parameters.

ESSA says that although "assessing impacts on the populations is necessary, [ ] it is not sufficient."[134] In so doing, ESSA rejects PSE&G's application of EPA's Guidelines on Ecological Risk Assessment,[135] concluding that PSE&G misapplies the Guidance.[136] Indeed, EPA finds that the impact of Salem's intake structure is both adverse and underestimated.[137]

Therefore, there is no doubt that the intake structure at Salem is having an "adverse environmental impact" that PSE&G must "minimize."

            B.    The Permit Does Not Minimize Salem's Adverse Effects

Section 316(b) requires that the "design, location, … and capacity of a intake structure reflect best technology for minimizing adverse environmental impact."[138] The Permit, however, does not minimize adverse effects.

EPA interprets the term "minimize" to mean a reduction to the "smallest possible amount or degree." [139] The American Heritage Dictionary defines "minimize" as: "[t]o reduce to the smallest possible amount, extent, size, or degree."[140] Although this standard does not require elimination of all losses of organisms due to plant operations,[141] minimization of adverse environmental impact is required regardless of whether the adverse environmental impact is significant; in EPA's words, "[a]ll environmental harm should be avoided."[142] 

A decade ago, NJDEP agreed that "[t]he only means of reducing the risk of long-term population and ecosystem level impacts from occurring, would be to institute major reductions in entrainment and impingement losses … at Salem." Yet to minimize impacts, the Permit requires PSE&G (1) to limit is withdrawal amounts to no more than the most Salem could withdraw, (2) continue to operate a fish return system, and (3) study various techniques for reducing impingement and entrainment.[143] 

These measures clearly do not reduce impacts "to the smallest possible amount, extent, size, or degree." Nor does it meet the requirement that "[a]ll environmental harm should be avoided." First, most important, the Permit requires no reduction in entrainment losses. Entrainment losses constitute 95 percent of the intake structure's impact. The Ristroph Screens, the Fish Return System and Sound Study are designed only to reduce impingement..[144] In any event, PSE&G's studies on entrainment contain biases in sampling procedures that necessitate data adjustments.[145]

Second, the Permit does not require different technologies to minimize impacts. The design features contemplated by Section 316(b) revolve around technologies that will reduce fish losses due to both entrainment and impingement effects.[146] Presently, however, the technologies used by the PSE&G SNGS largely involve antiquated fish screening and fish return devices designed only to prevent debris from entering the cooling water system.[147]  These systems "are limited in their abilities to minimize adverse aquatic impact."[148] For example, the ("enchanced") Ristoph screens at Salem do little to protect the mortality rate of weaker species such as bay anchovy and alewife.[149] PSE&G's application contains “inconsistencies and points of confusion in the documentation” as it relates to impingement.[150] Yet fish return systems remain unchanged. The intake flow limitation is the same flow limit that was specified in the July 20, 1994 permit (3,024 million gallons per day) and is based on "the impingement/entrainment loss estimates presenting in the March 4, 1993 PSE&G Renewal Application Supplement . . ." [151]  Other techniques are left for future study, but not implementation.

Third, the Permit ignores the cumulative effects of the intake structures of other facilities in the Delaware Estuary in determining how much "minimizing" need occur. In Brunswick I, the Office of Regional Counsel for EPA Region IV suggested that it is impossible to determine the impact of a single intake structure on an estuary and its aquatic life and, therefore, that location decisions should include consideration of the existence of other intake structures in the area.[152] The RA in Brunswick I stated:  "The environmental decision to be made by the regulatory agency is: at what point do we draw the line at this powerplant, or the next, or the next?  It appears that Congress has answered this question by requiring best technology to minimize impact at all plants."[153]  Brunswick I suggests that, although an intake structure may appear to have little impact on populations, the cumulative effect of many intakes "must eventually spell doom for important marine resources."[154]    Thus, it is evident that Section 316(b) requires consideration of the impacts of all intake structures on a source waterbody. The Permit does not do so.

III.             SALEM MUST ADOPT OR REFLECT DRY COOLING BECAUSE IT IS

                                    BTA TO MINIMIZE ADVERSE ENVIRONMENTAL IMPACT

            What reflects BTA minimizing adverse environmental impact "is an issue of fact."[155] BTA is "that technology which produces the greatest reductions in damage inflicted upon aquatic resources by entrainment and impingement."[156] EPA maintains that "[t]he reference point for best technology is closed cycle cooling."[157] The United States Fish and Wildlife Services maintains that "closed-cycle cooling [is] the best technology available" for Salem.[158]  ESSA (like Versar before it), finds that "cooling towers . . . would minimize entrainment/impingement events . . .," as required by the CWA.[159] The 1989 expert report prepared by Versar found that "a closed cycle cooling system would reduce Salem's cooling water intake requirements by more than 95% and result in a concomitant 95% reduction of entrainment and impingement losses, "[160] and would reflect BTA at Salem.

Fact is, the proven best technology available for "minimizing" adverse environmental impact is closed-cycle cooling. "[T]he only real means to reduce fish mortality is to significantly reduce intake water flow thru the implementation or a recirculating cooling water system . . ."[161]

Closed-cycle cooling technologies consist of dry cooling, hybrid (wet/dry) cooling, and wet evaporative cooling towers. Hybrid and wet evaporative cooling towers require roughly the same amount of water intake because "the dry sections of the hybrid cooling towers are only operated to abate plumes, and the time frame for visible plumes is relatively small given the total number of hours of operation."[162] These technologies reduce water withdrawal, and therefore, impingement and entrainment losses by as much as 95 percent.[163]

The Commentators submit that dry cooling technology in particular "reflects" BTA and thus Salem must minimize its impact commensurate with dry cooling. Dry cooling requires the least amount of water intake and reduces the adverse effects of impingement and entrainment, thereby minimizing the adverse environmental impacts.[164] Dry cooling technology is the most effective technology for minimizing adverse environmental impacts because it requires the least amount of water intake.[165] The water in dry cooling systems does not come in contact with the air, rather it travels through closed pipes in the tower.[166]  Keeping the water out of contact with the air reduces the amount of evaporation and consequently reduces the amount of water needed. Dry cooling systems have been installed in power plants of varying type and sizes for over 60 years.[167] "[I]nstallations of dry cooling technology in the U.S. have been growing rapidly since the early 1980's."[168] "There are at least 50 power plants with dry cooling systems in operation in the U.S. today . . ."[169] The amount of generating capacity using dry cooling systems in the U.S. has increased by over 15.4 percent per year between 1985 and 1998.[170] Conversely, the amount of generating capacity using evaporative cooling systems has increased by only 0.2 percent per year over the same time period.[171] Dry cooling technologies reduce water withdrawal, and therefore, impingement and entrainment by as much as 99 percent.  "[Salem] currently uses 2.1 million gallons per minute to produce 2200 MW which is equivalent to 954 gpm/MW. Assuming a 99.98% reduction in water usage with the use of dry cooling, water usage could be decreased to just 420 gpm or 0.19 gpm/MW."[172]  A 99.98% reduction in water usage would result in a concomitant 99.98% reduction in impingement and entrainment events at Salem.

Importantly, the Commissioner of the New York Department of Environmental Conservation (NYDEC) held that dry cooling is the most recent technology, uses markedly less water than any other intake structure, and minimizes the environmental impacts.[173] Unfortunately, NJDEP did not consider dry cooling technologies as BTA. This is inconsistent with §316(b). As a consultant not paid by PSE&G determined: "[U]se of dry cooling reduces environmental impact to the greatest degree. NJDEP must consider a dry cooling water system BTA for Salem."[174]

PSE&G does not address dry cooling either, and discounts the possibility of installing any closed-cycle cooling at Salem because it "would be an extremely complicated and expensive engineering construction project."[175] PSE&G also contends that cooling towers, as opposed to the existing intake structure, would cause a loss in power generating capacity.[176] "Even minor changes to the cooling water supply (for example a temperature increase a few degrees above design or a reduction in flow) can result in a large decrease in the Station's ability to achieve its rated capacity."[177]

But whether the installation of the cooling towers would be complicated to install or would result in diminished power generation capacity is not determinative under CWA § 316(b). The statute does not render installation difficulties or potential reductions in power generation determinative of whether compliance is required. There is no waiver for difficult jobs; indeed, there is no waiver at all.

In contrast, none of the technologies, considered as BTA for Salem, discussed in the next part to this comment, "can be considered BTA for minimizing environmental impacts at the [SNGS] because they fail to address entrainment losses."[178] As discussed supra, “once-through cooling, even with sub-littoral intakes and fish protection devices, does not offer equivalent levels of environmental protection to that afforded by dry cooling.”[179] Comparison of impingement losses versus entrainment losses using PSE&G's 1978-1982 and 1998 data show that "[f]rom 1978-1982 entrainment accounted for 99.9% of the fish lost . . . and in 1998 entrainment accounted for 99.8% of the fish lost."[180] The comparison results reflect the minimal impact of reducing impingement without reducing entrainment.  "Even if impingement can [be] reduced 100%, this would save only 0.2% of the fish lost at the [SNGS]."[181] The technologies considered BTA in the draft permit reduce impingement but do nothing to reduce entrainment. This is inconsistent with § 316(b).

IV.       THE PERMIT DOES NOT REQUIRE THAT THE DESIGN, CAPACITY AND LOCATION OF SALEM’S INTAKE STRUCTURE REFLECT BTA FOR MINIMIZING ADVERSE ENVIRONMENTAL IMPACT

Rather than require BTA that reflects closed-cycle cooling, NJDEP determined BTA for Salem to be the "existing once-through cooling system in conjunction with an intake flow limitation, an enhanced fish return system and the study and potential implementation of a multi-sensory hybrid system. . ."[182] Regardless, the design, capacity and location of Salem's intake structure do not reflect BTA.

A.            The Design of Salem’s Intake Structure Does Not Reflect BTA

The intake flow limitation in the 2000 draft permit is the same volume limit (3,024 million gallons per day (MGD)) as specified in the July 20, 1994 BTA determination. The figure represents no actual flow reduction whatsoever.  PSE&G discharge monitoring reports demonstrate this figure merely matches (or exceeds) the maximum current operating capacity of Salem.[183] Indeed, the intake flow limitation is based on “the impingement/entrainment loss estimates presented in the March 4, 1993 PSE&G Renewal Application Supplement . . .”[184] NJDEP has retained the same intake flow limitation as it did in the 1994 permit based on the fact that Salem has, on average, limited its intake flow amount to 2,968 MGD.[185] 

The ultimate success or failure of a specific cooling water intake design will depend upon the individual characteristics of a powerplant, the waterbody from which it is drawing its cooling water, and the aquatic organisms affected by the cooling water intake structure.[186]  Most  "designs" involve techniques to reduce impingement. These include screening devices,[187] use of fish handling and bypass systems,[188] pumps,[189] trash racks,[190] and behavioral barriers.[191]  Additionally, as EPA noted in Seabrook and Brunswick I and their progeny, the "design factor contemplates such items as fish ladders, fish buckets, improved fish screens, protective coatings, intake depth adjustment, modified velocity caps, and sound deterrent devices."[192]

The most direct influence on intake structure design involves fish screening devices.  Screening and fish return designs employed to minimize adverse environmental impact of intake structures can be divided into three basic types:  (1) removal and return technologies, namely, traveling, disk, drum, and fixed screens; (2) exclusion technologies, namely, wedgewire screens, perforated pipes, radial wells, porous dikes, and artificial filter beds; and (3) diversion and/or avoidance technologies, such as louvers, velocity caps, and physical and hydraulic barriers, discussed below. NJDEP, however, rejects all alternate technologies as not available at Salem, except for limited screening methods, and a behavioral study.  

The most common intake structure configurations, like those used at Salem, have a fish and debris removal system followed by a system that returns impinged fish and debris back to a waterbody.[193] The removal entry systems are usually designed with front-end, fixed-bar trash racks to keep large debris out of the intake structure, followed by a phalanx of traveling screens designed to prevent smaller debris from bypassing the intake structure and clogging condenser tubes.[194] The return systems usually involve high- and low-pressure sprays followed by combined or separate fish and debris return troughs, which re-introduce impinged fish and debris to their aquatic environment. The design of the fish removal and return systems typically distinguish one intake structure from the next. Designs of such systems include an assortment of screens, including traveling,[195] modified traveling,[196] inclined traveling,[197] Passavant,[198] dual flow,[199] horizontal traveling,[200] fine mesh,[201] horizontal drum,[202] vertical drum,[203] rotating disk,[204] and fixed.[205] One commonality shared by all screens is that they do not reduce entrainment and have only a limited ability to minimize impingement losses, and are generally ineffective in their ability to treat fish less severely than debris,[206] a feature of Salem.

Exclusion technologies attempt to dissuade aquatic organisms from entering the intake structure in the first place, thereby reducing impingement and entrainment losses. In order of general effectiveness, exclusion technologies include wedgewire screens,[207] perforated pipes,[208] radial wells,[209] porous dikes,[210] and artificial filter beds.[211] The Permit rejects some of these, incorrectly based on application of costs and benefits, and ignores others.

Behavioral barriers are designed to take advantage of fish behavioral patterns which would cause them to avoid intake structures altogether.[212]  Behavioral barriers include louvers,[213] velocity caps,[214] fish barrier nets,[215] and air bubble, electrical, light, sound, cable, chain, and water jet barriers.[216] The permit requires PSE&G to study sound deterrence and other behavioral barriers. The principal drawback of behavioral barriers, like those in the Permit, is that they do not reduce entrainment, and have only limited capacity to reduce impingement losses.[217]

B.            The Capacity of Salem’s Intake Structure Does Not Reflect BTA

The intake flow limitation in the 2000 draft permit is the same volume (3,024 million gallons per day (MGD)) as specified in the July 20, 1994 BTA determination. The figure represents no actual flow reduction whatsoever. PSE&G discharge monitoring reports demonstrate this figure merely matches (or exceeds) the maximum current operating capacity of Salem.[218] Indeed, the intake flow limitation is based on “the impingement/entrainment loss estimates presented in the March 4, 1993 PSE&G Renewal Application Supplement . . .”[219] NJDEP has retained the same intake flow limitation it did in the 1994 permit based on the fact that Salem has, on average, limited its intake flow amount to 2,968 MGD.[220] 

EPA defines "capacity" to mean "the volume of water withdrawn through a cooling water intake structure."[221] There is good reason to equate capacity with volume. Doing so provides Section 316(b)'s best means of minimizing entrainment losses, and associated impingement losses. Indeed, the principal factor in determining the amount of entrainment damage an intake structure will have is the volume of cooling water a plant is withdrawing from its source waterbody and the density of organisms located within the water withdrawn.[222] Thus, the level of entrainment caused by a given intake structure is proportionally decreased by reducing its intake volume.[223]

Decision of the General Counsel No. 41, Brunswick, Seabrook, Big Bend, and the 1980 Hudson River Agreement have each followed this reasoning. In Decision of the General Counsel No. 41, the EPA General Counsel employed the Random House Dictionary definition of "capacity."  EPA quoted "capacity" to mean "cubic contents; volume; that which can be contained."[224] The EPA Office of General Counsel also posited that "the volume [of water] withdrawn is the principal determinant of entrainment damage which is the major adverse environmental effect associated with most cooling water intake structures."[225] The EPA Office of General Counsel noted that the major adverse environmental impacts of intake structures are those affecting aquatic organisms in the water drawn through the cooling system, and that such adverse impacts may be reduced by restricting capacity.[226] The EPA Office of General Counsel found that the legislative history of the CWA supported its decision.[227] In In re Central Hudson Gas & Electric Corp., the EPA Office of General Counsel reinforced its findings.[228]

Next, in Brunswick I, EPA stated that "one of the most effective methods that can be used to reduce [entrainment] impact is the reduction of volume of water withdrawn."[229] In Brunswick I, EPA considered what would constitute BTA for the Brunswick powerplant located on the Cape Fear River near Wilmington, North Carolina. The Brunswick plant uses once-through cooling and has an intake capacity of eighty-one cubic meters per second (approximately 1.85 BGD).[230]  In Brunswick I, EPA initially decided that BTA mandated a permanent capacity restriction.[231] After the utility challenged this decision, EPA ultimately required that the plant reduce its flows during part of the year.[232] The Regional Administrator held that "[t]he best technology available to minimize the adverse environmental impacts of the plant is to restrict the capacity of the plant's intake structure."[233] Accordingly, EPA required CP&L to both institute seasonal flow reductions during periods when entrainment of larval forms would be highest,[234] as well as upgrade screening.

The powerplant in Big Bend was able to meet Section 316(b) requirements by implementing a 36% reduction in intake water volume and installing fine mesh screens.[235] Big Bend powerplant Units 1-3, were designed for once-through cooling. The intake flow water withdrawal capacity for the station's Units 1-3 was 70.8 m³/sec (approximately 1.6 BGD). EPA found that the intake capacity did not reflect BTA because of the facility's entrainment effects on the bay.[236] Therefore, EPA required the utility to reduce the intake flow by more than one-third to 45 m³/sec (approximately 1.03 BGD).[237]  When the utility proposed to add another unit to the site, which would have increased total intake volume to 61 m³/sec (approximately 1.39 BGD),[238] EPA required the utility to place a fine mesh screening apparatus on two of the four units, and to monitor their performance.[239]

The 1980 Hudson River Agreement required periods of reduced flow, partial power output reduction and seasonal shut-downs over substantial portions of the year during those times when aquatic organisms were most affected by the entrainment impact of the intake structure.[240]  The restrictions were to be attained by complete unit outages.  These outages were potentially coincident with the peak summer electric demands, and were not at times that would be economically more convenient to the utilities. The settlement also included modifications at several plants to further reduce adverse environmental impacts and the operation of a hatchery.[241]

In Seabrook II, the powerplant employed a design intake capacity of 1.2 billion gallons per day and a design intake velocity of one foot per second.[242] Because the powerplant operated at substantially less than full design capacity, EPA determined that actual capacity was not large enough to require further reduction in flow to meet BTA.[243] Nonetheless, some utilities seeking NPDES permits have repeatedly attempted to redefine "capacity" as the intake velocity. Each time this canard has been served, as in Decision of the General Counsel No. 41, In re Central Hudson Gas & Electric Corp., and Seabrook II, EPA has rejected it.[244] For example, in Decision of the General Counsel No. 41, EPA found that "capacity does not refer to the velocity of water withdrawn through a cooling water intake structure,"[245] nor is it limited to the physical size of the inlet of a cooling water intake structure.[246] An interpretation that narrows the definition of "capacity" to velocity alone (to the exclusion of volume) would effectively remove capacity as a potential factor meaningfully reducing adverse environmental impacts.

As in Brunswick and Seabrook II, an agency may restrict the capacity of a cooling water intake structure to such an extent that use of closed-cycle cooling would be a "predictable consequence of the capacity limitation."[247] For instance, Brunswick I involved an intake structure with an intake capacity of 1.85 billion gallons per day, located in an estuarine environment.[248] After considering the impingement and entrainment effects the powerplant was having on the aquatic organisms in its source waterbody, the RA determined that the capacity of the plant did not fulfill the BTA requirements of Section 316(b).[249] Information provided by CP&L indicated that use of closed-cycle cooling at the plant would provide a 96% reduction in its intake capacity.[250] As a result, based on Decision of the General Counsel No. 41, the RA determined that BTA for the Brunswick intake structure could be achieved by reducing the plant's intake capacity by an amount approximating 96%.[251]

Most recently, in Athens, the NYDEC Commissioner, applying Brunwsick I, reasoned that "capacity" in CWA § 316(b) refers to "the volume of water withdrawn through a cooling water intake structure."[252] In Athens, the Commissioner considered whether to require the installation of a hybrid cooling system or dry cooling. The Commissioner found that a hybrid system required an average 4.2 mgd to operate while a dry cooling system at the same location would require only 0.18 mgd.[253]  "I cannot discount the uncontroverted facts that application of dry cooling would use markedly less water and reduce the adverse effects of entrainment, thereby minimizing adverse environmental impacts to a greater degree that a hybrid cooling system."[254] The Commissioner subsequently directed the draft SPDES permit to be revised to limit the capacity of the intake structure to 0.18 mgd, i.e. to "reflect" dry cooling.[255]

Closed-cycle cooling technologies use considerably less water than once-through cooling systems,[256] which in turn reduces the incidents of aquatic life impingement and entrainment.  “[I]t is impossible to remove any significant volume of water from . . . a river . . . without also removing some of the organisms that are living within it.”[257] “[O]nce-through cooling, even with sub-littoral intakes and fish protection devices, does not offer equivalent levels of environmental protection to that afforded by dry cooling.”[258] Because once-through cooling stations use extremely large volumes of water, there is no once-through technology available that can ensure the level of fish protection that is achieved with less consumptive forms of closed-cycle cooling.[259]

C.            The Location of Salem’s Intake Structure Does Not Reflect BTA

Intake structures must be located so as to minimize adverse environmental impact. EPA maintains that the "location" of an intake structure "can be the most important consideration relevant to applying the best technology available for cooling water intake structures."[260] When cooling water intake structures draw water from a source waterbody, they also pull into the system all of the organisms living within that water. "The quantities of life destroyed by impingement on the screens or entrainment through the condenser cooling system are a direct function of the density of organisms residing in the source water[body]."[261] Therefore, by locating an intake structure where the quantity of aquatic organisms is high, a powerplant inevitably increases the adverse environmental impacts it will have on the source water.

EPA says that decisions concerning the location of a intake structure require an "extensive ecological survey in the vicinity of the proposed site."[262] This survey should be used to determine the potential impacts an intake structure may have "to important wildlife and aquatic breeding, nursery, feeding, and/or migration areas," and it may "enable determinations to be made with regard to concentrations of aquatic life within specific and proposed siting areas."[263] EPA makes clear that determinations regarding the location of an intake structure are dependent on the physical characteristics of the source waterbody to be used as well as hydraulic and economic factors.[264]

Under Section 316(b), "location factors primarily concern the siting or alternate placement of the intake structure with respect to the affected environment."[265] Siting and alternate placement issues involve proximity to estuaries and the cumulative impact of other intake structures, as the following paragraphs explore in greater detail.

While few generally applicable rules exist regarding Section 316(b) and its BTA requirements, one general rule has been stressed by EPA (intake structures should be located so as to avoid highly productive biological areas and estuarine environments.)  EPA says that when considering the location of a intake structure for minimization of adverse environmental impacts, permit writers should avoid spawning areas, fish migration paths, shellfish beds, and locations where aquatic life is concentrated.[266] Additionally, EPA has made clear that a proposed intake structure should be placed at a depth so as to avoid aquatic life.[267] EPA followed this reasoning in Seabrook, Brunswick, the 1980 Hudson River Agreement, and Big Bend.

In Seabrook II, the powerplant was located near Hampton Harbor Estuary, about two miles from the Atlantic Ocean.[268]  In upholding once-through cooling as constituting BTA, EPA noted that "the general design and location of Seabrook Station's intake and discharge offshore, and not in an estuary, were . . . special considerations that have been factored into the environmental characteristics of the area."[269] In the extensive Section 316(b) proceedings in Seabrook, the EPA Administrator cautioned that the "conclusions [are] based on the . . . ecology of this area . . . and [are] not transferable to another area with different, and its own special, environmental characteristics. . . . [I]ntake and discharge offshore, and not in an estuary, are also special considerations."[270]  The SNGS is, of course, in an estuary.

In Brunswick I, the "intake canal (a part of its intake structure) beg[an] in a dredged ship channel (a migratory pathway for aquatic organisms), interrupt[ed] a tidal creek (another migratory pathway), and cut[] across 6,000 feet of highly productive marshland (a nursery area)."[271]  In addition to the fact that the intake structure was located in such a highly productive area, the intake canal was attracting populations of fish which exacerbated the entrainment and impingement impacts created by the intake structure.[272]  As a result of its location in these highly productive biological areas, the RA determined that the Brunswick intake structure was having significant adverse impingement and entrainment impacts on the source waterbody and, therefore, that the location of the Brunswick intake structure did not comply with the requirements of Section 316(b).[273] EPA noted that

[t]he proper location of intake structures is the most important consideration relevant to applying best technology available for cooling water intake structures. Careful location of intake structures can greatly minimize adverse environmental impacts.  For example, intake structures should be located to withdraw water from areas that are the least productive biologically and contain the lowest population densities of critical aquatic organisms.  Additionally, they should be located to avoid spawning areas, nursery areas, fish migration paths, shellfish beds, or any location containing a high concentration of aquatic life.[274]

EPA later reiterated its support for this position in the Brunswick Memorandum.[275] The 1980 Hudson River Agreement involved powerplants along the Hudson River estuary, and was concerned with long-term regional population, forage organism and food web impacts.[276] In making its initial BTA decision, EPA focused on the impacts of the facility on aquatic life in the affected area and required these powerplants to institute closed-cycle cooling.[277] In Big Bend, EPA took note of the sandy, tidal-swept nature of the estuary subject to impact by the intake structure in making its Section 316(b) determination.[278]

EPA's focus on fish population, forage, and food web impact stresses the importance of avoiding the siting of intake structures near spawning areas, fish migration paths, or any location with a particular concentration of aquatic life. Poor location may cause adverse environmental impacts that may be offset only by a reduction in intake capacity, and by improved design, discussed supra.  

CONCLUSION

Commentators respectfully submit that NJPDES draft permit NJ0005622 is not consistent with § 316(b) of the CWA. First, the Permit impermissibly considers costs in rejecting true BTA at Salem. Particularly in light of last month's unanimous Supreme Court Opinion in Whitman v. American Trucking, it is improper for the Permit to consider the costs of BTA at all in making a BTA determination because § 316(b) does not allow consideration of costs. Even if costs were relevant, the Permit should have assessed the marginal increase in costs to ratepayers for requiring BTA, instead of rejecting BTA based in part on PSE&G's flawed comparison of total costs of conversion solely to the commercial and recreational worth of lost fish. In any event, by disregarding societal costs and benefits, NJDEP misapplies the cost/benefit test.

Second, the Permit's does not "reflect" BTA. NJDEP agrees that Salem is having adverse environmental effects. Yet the Permit fails to "minimize" such impacts. To minimize such impacts, the Permit should require a technological conversion at Salem that adopts or reflects BTA, which is "dry cooling," a type of technology that reduces adverse impacts by as much as 99 percent. The Permit's purported BTA conditions are not BTA. They do nothing to reduce entrainment losses, which account for 99 percent of Salem's adverse impact, and reduce impingement losses under the best circumstances, by only 0.2 percent.

                                           Respectfully submitted,

                                           James R. May

                                           Executive Director, Center

                                           Director, Clinic

                                           Scott Fazio (student intern, 2000-01)

                                           Glenn Elters (student intern, 1999-00)