Power-Plant Cooling: How Many Fish Per kWh?


EPA flounders on the Clean Water rule, while producers tackle the real enemy—shortage.

Fortnightly Magazine - July 2005

The U.S. Environmental Protection Agency (EPA) says that a typical sport fisherman working the Great Lakes would pay $4.58 for the privilege of catching a single walleye/pike, but would gladly fork over $7.99 to land a trout, or as much as $11.19 for a salmon.

Sound fishy? Yet the EPA would rely heavily on these data, and other figures quite similar, to justify its proposed “Phase III” rule to regulate cooling water intake structures (CWIS) at small power plants and other similarly sized facilities, to preserve aquatic and marine life in the nation’s lakes, rivers, streams, bays, and estuaries.

The idea is to estimate the dollar benefits of reducing fish mortality through impingement (being pinned against screens or other parts of a cooling water intake structure) or entrainment (being drawn into cooling water systems and subjected to thermal, physical or chemical stresses) at cooling water intake structures. So the EPA has done more than just measure losses to the commercial fish harvest. It seeks to measure the value of unharvested fish. It seeks to calibrate the dollar value of the pleasure from fishing for fun, or at least for knowing that we could if we wanted:

“Society may value both landed and unlanded fish for reasons unrelated to their use value. Such non-use values include the value that people may hold simply for knowing these fish exist. While non-use values are difficult to quantify, EPA believes it is important to consider such value.”

(See, National Pollutant Discharge Elimination System — Proposed Regulations to Establish Requirements for Cooling Water Intake Structures at Phase III Facilities, U.S. Environmental Protection Agency, Docket OW-2004-002 [Phase III Rule], 69 Fed.Reg. 68444, at page 68514, Nov. 24, 2004. See p. 68517 for discussion of valuation of individual fish.)

In fact, if you don’t count this “non-use” value for all those walleye, trout, and salmon that got away, then the costs of complying with the Phase III rule just plain overrule the meager commercial benefits.

And so in early spring, after the deadline had passed for the general public to submit comments on the proposed rule, the EPA sought authority from the Office of Management and Budget to go on a fishing expedition to collect more evidence to bolster its theory. As the agency said, it would begin conducting as many as a dozen focus groups to “better understand the public’s perceptions of fishery resources.” (See, 70 Fed. Reg. 15079, March 24, 2005.)

When it comes back with new results, the EPA may well attempt to build a new cost-benefit case for its proposed rule. So stay tuned.

All the same, this piquant debate over the economic justifications for environmental protection should not obscure the fact that water usage by power plants deserves close examination on its own merits, for many other reasons firmly rooted in dollars and sense.

California’s drought of 1987-1992 illustrates the severe limitations of that state’s water supply, whether for power-plant cooling or otherwise. According to the California Energy Commission, the economic, environmental, and social consequences from the drought echoed throughout the state, affecting every aspect of life—from agricultural production to recreational fishing. For example, the San Joaquin Valley experienced water shortages of 390,000 acre-feet, leaving 166,000 acres of agricultural land idle. The overall economic repercussions totaled $179 million in lost revenue.

Wildlife Protection: The Cost-Benefit Analysis

Acting under section 316(b) of the Clean Water Act, the EPA had successfully issued two rules prior to the current Phase III proposal to regulate the cooling water intake structures (CWIS) at facilities, such as power plants, that take in large quantities of water. These two rules replaced a prior regime whereby CWIS facilities would fall subject to permit regulation only by state officials, if at all, and then only on a case-by-case basis. State regulation had prevailed because the EPA’s first attempt (proposed in 1976) to regulate CWIS facilities under the Clean Water Act Sec. 316(b) had been struck down by a federal appeals court in 1977, on procedural grounds, and was not again taken up by the agency. (See, Mercatus Center, George Mason University, “Public Interest Comment on the EPA’s Proposed Rule: Phase III Cooling Water Intake Structures,” EPA Docket OW-2004-0002-0919, filed March 24, 2005.)

In Phase I, in a final rule issued in 2003, after various iterations, the EPA issued rules governing cooling water intake structures at “new” facilities, including power plants, constructed after Jan. 17, 2002. It governed facilities with a design capability to withdraw more than 2 million gallons per day (MGD) from waters of the United States, provided they used 25 percent or more of their intake water for cooling. The EPA projected in 2001 that its Phase I rule would apply to some 120 facilities built over the next 20 years, at a compliance cost of less than $48 million per year.

In July 2004, the EPA issued new Phase II rules governing cooling water intakes at large “existing” power plants (construction begun on or before Jan. 17, 2002) with a total design intake flow of 50 MGD or more. It estimated an average annual pre-tax compliance cost of about 0.01 cents/kWh, or about one-tenth of one percent, at most, tacked on to the average consumer’s retail electric rate.

These rules covered the low-hanging fruit. Phases I and II together covered existing and future power plants using the very largest quantities of water for cooling. In each case, the agency conducted extensive studies of facilities of the type to be regulated in order to confirm that the benefits of protecting aquatic and marine life from impingement and entrainment caused by cooling intake structures would outweigh the costs of compliance.

Then came the Phase III rule, proposed in November of last year. The EPA offered three different options—three different definitions of the types of facilities that the rule would cover, depending upon the design capacity of the intakes, and the type of resource from which the facility would draw its cooling water (see “EPA’s Phase-III CWIS Proposal”). However, in contrast to Phases I and II, the EPA chose not to conduct extensive additional studies at facilities of the size and type that might be covered in order to finalize its cost-benefit analysis for Phase III. Instead, it simply extrapolated the results from its studies of Phase I and Phase II facilities in order to estimate potential costs and benefits from applying similar rules to Phase III units.

This leap of faith, however, has drawn considerable criticism from rule opponents in the public comments filed to date. The opponents argue that the EPA has failed to justify its cost and benefit estimates because the experience from large-capacity Phase I and Phase II units does’t give a clear idea of how the smaller-sized intakes at Phase III units would affect the environment.

Consider arguments offered by the scholars of regulatory law at the Mercatus Center, George Mason University, in Fairfax, Va.

They note that the Phase II rule covered 257 plants with intakes greater than 500 MGD, and 112 plants greater than 1,000 MGD. By contrast, the Phase III universe would include only 10 facilities larger than 500 MGD, and only 2 above 1,000 MGD. Thus, it would be natural to assume that fish mortality (and the benefits of protection) would run much higher at Phase II facilities than for Phase III units. And, as one might expect, the EPA’s smaller estimates for Phase II units reflected that general assumption. However, the Mercatus scholars still find fault with EPA’s numbers.

For example, EPA admits that water intake volumes likely would run nearly 10 times higher at Phase III facilities than for Phase II—214 billion gallons per day versus 23 billion. It adds that “larger withdrawals of water may result in commensurately greater levels of entrainment.” Yet, at the same time, EPA claims that “even on a flow-weighted basis, the number of organisms impinged and entrained by Phase III facilities is approximately one-third of the number of organisms impinged and entrained by Phase II facilities.”

Mercatus replies simply: “This claim does not make sense.”

These words reflect the general tenor of comments from two other key industry groups. One is the Cooling Water Intake Structure Coalition, led by the American Petroleum Institute and the American Forest & Paper Association. The other is the Utility Water Act Group, an association of 198 individual electric utilities and four national trade groups: Edison Electric Institute (EEI), the Nuclear Energy Institute, the American Public Power Association (APPA), and the National Rural Electric Cooperative Association (NRECA).

For example, the APPA highlights just how much the EPA is relying upon its novel theory that protection of purely recreational fishing can justify the rule:

“There are substantial, unquantified non-use benefits associated with the Rule,” APPA notes. “In fact, EPA states that 96.7 percent of the organisms expected to be protected by the Rule will not be harvested and, thus, have no direct use value.”

In particular, the APPA notes that some 60 of its members appear subject to a potential Phase III rule, out of an estimated universe of 121 utilities threatened by Phase III. And for those 60 members, the APPA agrees with its colleagues at EEI and NRECA:

“The costs of complying with the Rule’s performance standards far outweigh the potential benefits.”

For the small-sized power plants and other facilities that the EPA proposes to regulate through the Phase III rule, the industry groups would rather see the agency simply decline to set a national standard. They would prefer the status quo, whereby such small facilities may obtain permits from the appropriate state agencies on a case-by-case basis.

In fact, in the preamble (the regulatory explanation) of its proposed Phase III rule, the EPA anticipates just such a result:

“EPA is also considering an alternative under which EPA would not promulgate, at this time, categorical requirements under section 316(b) for cooling water intake structures unregulated by Phase I and Phase II. Rather, EPA would continue to rely on the best professional judgment of the permitting authority to determine the best technology available to minimize adverse environmental impact, in order to allow these requirements to be better tailored to local conditions.”

For its part, the Mercatus Center strongly urges EPA to settle on the so-called “no-rule” option and maintain the status quo. “The rule does not explain,” the scholars say, “why the practice for the past 25 years is flawed.

“The fact that EPA has been unable to justify that this regulation is cost-effective … reinforces the superiority of a case-by-case approach.”

Supply Shortage: Water as a Commodity

Industry experts, including Kent Zammit, manager, Cooling Water Technologies at the Electric Power Research Institute (EPRI), point to the highly visible Riverkeeper case in the Northeast as the latest word on water conservation and protection of aquatic and marine life. (See, Riverkeeper, Inc., v. EPA, 358 F.3d 174, 2d Cir., 2004.)

In that case, environmental groups had challenged the EPA’s Phase I rule and had prevailed on the claim that the Clean Water Act does not allow “restoration” measures to mitigate damage after the fact to comply with Section 316(b), but instead requires affirmative action beforehand to avoid damage. The electric generating and manufacturing industries also had challenged EPA’s Phase I rule, but their claims were denied.

In wake of Riverkeeper and other cases, New York’s Department of Environmental Quality now is perceived as taking the position that it doesn’t even want to discuss what is an “adverse” im-pact; instead, it wants to look at every plant and what it can do to reduce the impact it might have. Some industry reps see New York’s conservation agencies unyielding to the point where actions affecting “one fish, one egg, one larva”
is considered an adverse environmental impact. That has created more interest—not only to make water in-takes safer,
but to switch perhaps to dry cooling technologies, which use less water.

Already power plant designers as far West as New Mexico, Nevada, and California are redesigning cooling strategies in an effort to avoid permitting obstacles related to water use and availability.

In Nevada, for instance, the need to conserve water has driven utilities to search for alternatives to wet evaporative cooling systems. But the water issues go much further, says Zammit.

“In some areas around the country—not just the arid West—stakeholders are opposing new plants on the basis of water use. These plants will operate for 40 to 60 years, and the issue is the water they use may be very, very valuable 50 years from now. That can create timing issues as far as getting timely permits to be able to build the plants.

“Even in areas where water has historically not been an issue, we’re starting to see it become an issue. For instance in Georgia, they are starting to have trouble meeting their interstate river compacts due to drought and growth issues. There’s a concern about where they're going to get water for future development for business.

“You think of Georgia as being a relatively wet state and not really needing a water plan and yet with the booming growth of Atlanta, it's really putting pressure on their water usage. There are water impacts in other water-rich states, like saltwater intrusion of fresh water aquifers along the Gulf coast, in Louisiana and Florida. These kinds of developments will continue to increase the pressure to reduce water use in power generation.”

The State Regulatory Role

Ted Long, formerly supervisor, Environmental, for Reliant Energy, and now with Texas GenCo, gives some background on implementing EPA rules already issued under Sec. 316b, in Phase I and Phase II.

“The states have to implement the federal rule. … Basically we have to submit to the state something called a PIC, ‘Proposal for Information Collection,’ and that’s the first regulatory document that we submit.”

The applicant’s PIC outlines what it needs to do so that data collection and information collection comply with rule 316b.

“That’s the first thing, and that’s across the country,” he adds. In Texas one would go before the TCEQ, the Texas Commission on Environmental Quality, he explains.

“So much of this is state dependent. I would imagine that as far as dealing with 316b what you’re going to face in Texas—even though it’s the same regulation—could be a lot different than how it’s going to be looked at in California and New York.”

Long continues: “The state is supposed to get back and say, ‘Okay we agree.’ Then the next thing gets real complicated. It is part of a comprehensive demonstration study. Basically, you have several years to pull together your study where you’ve looked at all your different compliance options, your outline, how you want to go about trying to comply with the rule. And you have to have all that stuff submitted with your next permit renewal.”

Time is critical, however. And time is money.

“There are places,” Long explains, “where they say you need to get two years’ worth of data. Well, if you get two years’ worth of data, you’re running into some time constraints in getting everything ready.”

Kerry Whelan, a principal with Reliant, echoes Long’s comments.

“Most of the recent activity in California,” he notes, “has been a result of re-licensing projects by other entities where they are expanding or modifying existing intakes or creating new intakes. And that work has been subject to review by the California Energy

Referring to Reliant’s power facilities at Mandalay Beach, and Ormond Beach in Oxnard, Calif., Whelan says, “For us and for our process we’re working closely with the L.A. Regional Water Quality Control Board. … They’re trying to understand what their role is. There are areas of ambiguity in the regulation that we’re talking about together and that we’re talking about with EPA headquarters.

“How do you proceed when there’s this sort of cloud over the regulation? The answer is, if you’re a regulated entity you have to proceed, but it does raise some issues schedule-wise. These are general things that we see. … None of that is really unique to California.”

The Future of Cooling Technology

Referring to Reliant Energy’s newer Nevada gen plants (El Dorado and Big Horn, located southeast and southwest, respectively, from Las Vegas), Long says, “I think the reason they went to dry cooling … was strictly water supply and use of water.

“If your facility has cooling towers or does not have once-through cooling, then you’re basically off the hook for Phase II of 316b,” Long says.

“So if I was going to build a new plant, I would probably not even think about it or and wouldn’t even want to go there with once-through cooling because of all this. It would be easier to just go ahead and throw in cooling towers.”

“With El Dorado in the design stage, water supply was coming from Lake Mead. To minimize water use (they) went with the air cool condenser. As far as the wastewater permitting side of it, it was set up where everything would go into evaporation ponds. Big Horn was similar except their water supply was coming from wastewater from a casino just up the road.”

According to EPRI’s Zammit, the Big Horn and El Dorado sites mark just two of a handful of new power plants that are representative of new building trends in the West. These newer facilities tend to be combined-cycle, natural-gas-fired plants. With approximately two-thirds of the energy from these plants generated by gas turbines, the potential load for dry cooling already is cut by two-thirds over a traditional simple-cycle plant. The remaining one-third of the plant capacity is generated by Heat Recovery Steam Generators (HRSG) and traditional steam turbines that can be cooled by existing once-through wet-cooling towers (typically for re-powering only), or air-cooled condensers.

Other designs also are in the works. An air permit application was filed with the EPA (Region 9) for plans for a design not seen before in the United States. The Dine Power Authority (DPA) has contracted with Steag Power (a German company recently purchased by Sithe Global) to develop an electric power generation facility on the Navajo Nation called “Desert Rock Energy Facility.” The facility, to be located 25 miles southwest of Farmington, N.M., would produce 1,500 MW (gross). It would be composed of two units of 750 MW (gross), 683 MW (net) each. The applicant proposes to be water efficient because it will use a “once through, supercritical steam cycle”—a system with dry cooling. If everything goes according to plan, Dine authorities expect the plant to break ground in the latter part of 2005.

John Howard, an energy environmental and public policy attorney and former deputy director for the White House Council on Environmental Quality, predicts that EPA, Department of Energy, and the power generation sector will continue to push for technological innovation that will accelerate efficiencies and environmental progress.

“Hopefully,” he adds, “without breaking the bank.

“We should know whether dry cooling clears, or fails to clear, these hurdles for application across the United States in the next couple of years.”


EPA’s Phase-III CWIS Proposal

Empowering Act. Section 316(b) of the Clean Water Act (CWA) requires EPA to ensure that the location, design, construction, and capacity of cooling water intake structures (CWIS) reflect the best technology available to protect aquatic organisms from being killed or injured by impingement (being pinned against screens or other parts of a cooling water intake structure) or entrainment (being drawn into cooling water systems and subjected to thermal, physical, or chemical stresses).

General Provisions. Proposed in November 2004, with a projected effective date of June 1, 2006, Phase III of the EPA’s rule for CWIS will apply to electric generating plants using relatively small amounts of cooling water—less than 50 million gallons per day (MGD), on the basis of design capability.

(The rule also would apply to similarly situated industrial and manufacturing facilities and new offshore and coastal oil and gas extraction facilities that use at least 25 percent of the withdrawn water exclusively for cooling.)

The facilities subject to the Phase III rule would have to meet the same requirements as those set in Phase II for large-flow power plants, for reducing mortality of aquatic and marine life from impingement or entrainment. However, the EPA’s Phase III rule, as proposed in November, leaves the matter open as to the exact size of power plant and type of water body that would be covered under the regulation.

Alternative Regulatory Regimes. The EPA in fact proposes several alternative optional regimes, with each one providing for a different definition of the size of facility and type of water resource subject to the rule:

  • Option 1 The rule would apply only to facilities that withdraw at least 50 million gallons per day (MGD) of cooling water from any type of water body. Choosing this option would mean that the rule would apply only to about 23 percent of the universe of power plants potentially subject to Phase III (some 155 facilities), and would cover about 75 percent of the total water intake in that universe.
  • Option 2 The rule would apply only to facilities that withdraw at least 200 MGD from any type of water body. This option would cover only 5 percent of potential Phase-III facilities, but 45 percent of the total potential water intake.
  • Option 3 The rule would apply only to facilities that withdraw at least 100 MGD from an ocean, estuary, tidal river, or one of the Great Lakes. This option would cover only the most sensitive water bodies. It would cover only 4 percent of the facilities potentially covered under Phase III, and 18 percent of the total potentially covered water intake. (Note, for example, that this option would not apply to a small gen plant taking cooling water from the Colorado or Snake Rivers in the West, as these water bodies are not “tidal.”)

Cost-Benefit Analyses. EPA estimates that the proposed options would affect anywhere between 19 and 136 existing facilities, and would protect between 30 million and 50 million aquatic organisms annually from death or injury by cooling water intake structures. However, the associated “use benefits” (identifiable dollar benefits from commercial and recreational fishing) would range only from $1.3 million to $1.9 million per year, depending on the design intake flow threshold of the proposed options. That would fall way short of the estimated compliance cost, which EPA has estimated as between $17.4 million and $46.8 million per year, depending upon which of the 3 regulatory options is chosen:

  • Option 1  Estimated cost outweighs estimated benefit by about 26 to 1. (EPA estimates compliance cost at $47.3 million to $50.1million per year, or between $348,000 and $368,000 on average annually per covered facility. Total annual “use” benefits run from $1.5 million to $1.9 million.
  • Option 2  Cost outweighs benefits by about 19 to 1. (Cost runs $22.8 million to $24.1 million, or between $912,000 and $964,000 annually per covered facility. “Use-value” benefits run $0.98 million to $1.26 million per year.)
  • Option 3  Cost outweighs benefits by about 13 to 1. (Cost runs between $17.6 million to $18.2 million, or $926,000 to $958,000 per year per covered facility. Benefits range from $1.1 million to $1.4 million in annual use-value.)

The “No Rule” Option. If it so chooses, EPA may decide to refrain from issuing any sort of national standard for Phase III facilities, and instead would throw the matter back to the states for permitting on a case-by-case basis.

(See generally, http://www.epa.gov/waterscience/316b/ph3.htm, and public comments filed in EPA Docket OW-2004-0002.)—BWR


A Primer on Dry Cooling

In General. Dry-cooling systems transfer heat to the atmosphere without the evaporative loss of water. As a result, water consumption rates run very low for dry cooling as compared to wet cooling systems. However, since the dry-cooling unit does not rely on evaporative cooling as does a wet cooling tower, larger volumes of air must be passed through the system. As a result, dry-cooling towers need larger heat transfer surfaces and, therefore, tend to be larger in size than comparable wet cooling towers.

Technology Types. There are two types of dry-cooling systems for power-plant applications: direct and indirect.

  • Direct Dry Cooling. Uses air to condense steam directly. The most common version of direct dry cooling adapted for new power plants is a recirculated cooling system with mechanical draft towers. Natural draft towers are used infrequently for power-plant installations in the United States.
  • Indirect Dry Cooling. Uses a closed-cycle water system to condense steam; the heated water is then air cooled. Indirect dry-cooling systems are used generally in retrofit situations at existing power plants, because a water-cooled condenser would be in place already for a once-through or recirculated cooling system.

Water Consumption. As explained by Joe O’Hagan, specialist with the California Energy Commission, the cooling water demand for a 500-MW, combined-cycle power plant using once-through cooling is about 15,000 gallons per megawatt-hour, whereas closed loop (re-circulating) cooling requires about 200 to 250 gallons per MWh. The same type of plant using dry cooling will use no water for cooling, while a hybrid system’s water use will be somewhere between a wet and dry system.

Electric Output and Cost. Dry cooling offers certain environmental advantages, but also has serious economic drawbacks. According to statistics from the Electric Power Research Institute of Palo Alto, electric output from gen plants using dry cooling output runs about 2 percent less than with wet energy cooling systems. EPA estimates that the penalty in electric output for dry cooling on average over the year can run as high as 8.6 percent versus once-through wet cooling, and as much as 10 percent during peak periods. Yet the costs of dry cooling run anywhere from 3.5 to 4.5 times more for dry cooling than for wet cooling.

Aesthetics. “Dry Cooling involves a bigger facility and may be noisier,” says John Howard, an energy environmental and public policy attorney and former deputy director for the White House Council on Environmental Quality. “You’re relying on ambient air, which is harder to cool off,” and it requires special fans, he says. Estimates vary greatly, but a standard fan and its gear box can run about $30,000 or more.
A special low-noise fan can cost one to two times that amount.


(See Technical Development Document for the Final Regulations (EPA-821-R-01-036) November 2001, Chapter 4—Dry Cooling, at http://www.epa.gov/waterscience/316b/technical/ch4.pdf)—CB