Defining “Mitigation” for Hydropower Operations

Until recently, many measured society’s commitment to salmon by the amount of money devoted to salmon recovery. Most of the money comes from citizens of the Pacific Northwest, collected through their payments for electricity, and funneled through public utilities to the Bonneville Power Administration. A spectacularly successful coalition of state and tribal fishery agencies, allied with environmentalists and the media, have extracted higher and higher payments from BPA since 1980. Unfortunately, after the general collapse of legal constraints on agency behavior, nothing assures that all this money is spent in any sensible way. It isn’t.

The concept that the adverse effects of federal projects upon fish and wildlife should be mitigated is expressed in literally dozens of federal statutes and regulations. The concept is simple: a scientific effort is made to quantify the adverse effect, and then some sort of program is devised to offset that adverse effect. The U.S. Army Corps of Engineers, having the most experience in planning mitigation for the construction of federal water projects, has extensive regulations intended to produce a rational, cost-effective approach to fish and wildlife mitigation. They are all ignored.

The fish and wildlife statutes are so vague that Congress has essentially turned armies of fish bureaucrats loose to demand “mitigation” without any guidance whatsoever. Back before President Roosevelt’s threats to pack the Supreme Court produced the “switch in time that saved nine”, these sorts of statutes would have been held unconstitutional as “delegation run riot”.

It is not an easy problem to figure out what it means to “mitigate” for dam construction and operations. The first problem is that no one really knows what the adverse effects of dam construction were. Nobody really knows how many salmon ran in the Columbia River before it was dammed. As far back as 1940, observers recognized this, reporting that at Bonneville Dam "391,595 ascending chinooks were counted. Is this number large or small? Who knows? The biologists do not. They have no basis of comparison; never before have the fish in a vast stream been counted at all."8 In 1994, 400,000 chinook salmon entered the river.9 A higher percentage are hatchery fish now, but no one really knows how much higher.

The Northwest Power Planning Council is charged in the Northwest Power Act to devise a plan to mitigate for the adverse impacts of hydropower development in the Columbia River Basin. Obviously, the scope of the plan depends on some assessment of how many salmon have been lost, and why. By the early 1980s, salmon advocates began to realize that they needed a benchmark against which to measure the progress in salmon restoration, a benchmark against which present efforts would fall far short. They convinced the Council to settle on the number of salmon that had run in the Columbia before it was dammed, ultimately inducing the Council to declare that historical salmon runs averaged 16 million fish.

Dr. Don Chapman reviewed the historical evidence and concluded that the Council’s 16 million figure “cannot be supported by historical information and careful inference”.10 He reviewed the extensive data on historical catches of salmon and concluded that the total runs for all salmon and steelhead species probably ranged from 7.5 to 8.9 million fish.11 The Pacific Fishery Management Council, using available habitat areas to estimate runs, came up with a similar estimate of 6.2 million for salmon species.12

This huge disparity between the Council’s estimates and those of independent scientists continues to be noted in salmon recovery literature, with no resolution of which view is correct.13 Even quasi-scientific reviews, like the Northwest Power Planning Council’s Return to the River report, continue to cite the Council’s political estimates as scientific fact.14

Salmon advocates then induced the Northwest Power Planning Council to declare that dams were responsible for about 80% of the decline from historic peaks. This figure, widely publicized by the Council and others, has been been cited by dam opponents over and over and over again.15 Later, the figure was, by improper citation, recharacterized as an estimate of the National Marine Fisheries Service.16 It is now repeatedly echoed in solemn judicial opinions as a finding of fact: “NMFS estimates that approximately 80% of historical salmon losses are attributable to hydropower development and operation”.17 This statistic, unquestioned in the courts, media or elsewhere, is utterly false and a key supporting element of the Great Salmon Hoax.

Here is how the Northwest Power Planning Council got the number. In Appendix E to the 1987 Columbia River Basin Fish and Wildlife Program, the Council staff provided two alternative means of computing “numerical estimates of hydropower-related losses”. “Alternative 2”, the more simple alternative, simply posited the equation:

Total loss * Percent hydropower purpose of dams = Hydropower-related losses

This produced a “resulting hydropower contribution range” from “about 5 million to 11 million fish”.18 Under this approach, the Council was to pretend that dams caused all salmon losses since 1850, and then was to use pre-existing estimates (prepared for federal accounting purposes) of the proportion of hydropower-related investment at the dams to reduce hydropower’s share of the losses.

Alternative 1 was superficially more reasonable, but came to precisely the same result because of highly unreasonable assumptions. This approach attempted to estimate the ultimate survival of salmon with and without dams, going step by step through the stages of the salmon life cycle. The faulty assumptions were legion. To name just a few, the staff ignored pre-dam losses, assumed that each and every fish that died in the river died because of the dams, and ignored all beneficial effects of transportation and spill.19

The Council staff then proceeded on the “theory . . . that non-hydropower development effects (e.g., irrigation, fishing, logging, mining, grazing, agriculture, urbanization, pollution and other effects) are largely reversible so hydropower caused all the mortalities that would occur if predevelopment run sizes existed”.20 They did not explain how removing dams would, by hypothesis, de-populate the Pacific Northwest and restore pristine conditions prevailing prior to 1850. The idea that salmon runs, but for dams, would be much higher than the runs actually were before the dams has long been a trick used by fishery agencies seeking mitigation funding.21

In the 1987 Columbia River Fish and Wildlife Program itself, the Council summarized these estimates, baldly asserting that “all reasonable approaches would result in loss estimates in this range”.22 The Council concluded that “. . . the salmon and steelhead run size has declined by more than 10 million from all causes. Of that 10 million, about 8 million can be attributed to the hydropower system.”23 The U.S. Army Corps of Engineers commented that “there [wa]s no sound factual basis” for the Council’s conclusions, a view shared by all electric power interests at the time.24 In its 1994 Fish and Wildlife Program, the Council reaffirmed the estimate.

Thus it is an article of faith among most Northwestern policymakers that the dams singlehandedly destroyed 80% of the salmon runs. No one questions the logic of taking the historic peak in salmon runs, decades before the first dam was ever built, long before the sustained development of a salmon canning industry that sent Columbia River salmon to dinner tables around the world, and arbitrarily attributing 80% of the decline to dams.

Lacking a statutory definition of “mitigation”, everyone in the Columbia River Basin assumed for decades that replacing wild fish with hatchery fish was mitigation. The U.S. Army Corps of Engineers would figure out how many fish might be lost as a result of constructing a dam, and how many hatcheries would have to be built to replace them. This was a hugely successful, although expensive, effort.

Hatchery releases kept up total production for decades, even in the face of sustained or increasing fishing pressure. This has not made everyone happy. From the perspective of the tribes and upriver interests, what matters is the number of salmon they have been able to catch in the river. They look at the simple fact that the numbers of salmon they have been able to catch have dwindled as the number of dams has increased, and leap to the conclusion that the dams are responsible. They don’t consider the effect of others who have intercepted the fish first.

It also seems as if fishery managers made conscious choices to favor commercial harvest interests by siting hatcheries in downriver locations. In 1986, the Council staff concluded that a “dramatic effect of mitigation activities for hydropower and for multipurpose developments has been to strengthen fish propagation in the lower Columbia River Basin without attempting to rebuild upriver runs. A related effect has been to increase the proportion of hatchery fish to the overall outmigration.”25

For at least ten years, however, conservation biologists have been pushing to redefine “mitigation” to require the replacement of lost natural production of salmon.26 This is an enormous shift in public policy, and one that has so far occurred without any democratic input whatsoever, in the form of laws or even policy guidance from Congress or other elected representatives. To fully replace lost natural production of salmon is impossible without evacuating the Pacific Northwest. Efforts to replace lost natural production inevitably have the effect of limiting and even reversing development of the Pacific Northwest. This serves all the goals of conservation biologists, but may not serve the public interest.

8 R. Neuberger, "The Great Salmon Mystery", Saturday Evening Post, Sept. 13, 1941.

9 ISG, Return to the River 90.

10 D. Chapman, "Salmon and Steelhead Abundance in the Columbia River in the Nineteenth Century", Trans. Am. Fish. Soc. 116:662-70 (1986).

11 Id.

12 PFMC, “Freshwater habitat, salmon produced, and escapements for natural spawning along the Pacific Coast of the United States”, Report of the Anadromous Salmonid Environmental Task Force (1979).

13 See, e.g., NRC, Upstream at 81 (Prepub. ed.).

14 See, e.g., ISG, Return to the River 90 (citing predevelopment abundance estimate for chinook salmon of 4.7 to 9.2 million fish).

15 See, e.g., B. Harden, A River Lost 196.

16 See NMFS, “Factors for Decline: A Supplement to the Notice of Determination for Snake RIver Spring/Summer Chinook Salmon under the Endangered Species Act”, June 1991, at 7 (citing Northwest Power Planning Council estimate); NMFS, Biological Opinion, “Reinitiation of Consultation on 1994-98 Operation of the Federal Columbia River Power System and Juvenile Transportation Program in 1995 and Future Years”, Mar. 2, 1995, at 4 (misciting the “Factors for Decline” paper as reciting an estimate by NMFS).

17 See, e.g., American Rivers v. NMFS, No. 96-384-MA, slip op. at 12 (D. Or. April 3, 1997) (citing 1995 Biological Opinion); NRIC v. NWPPC, 35 F.3d 1371, 1376 (9th Cir. 1994).

18 “Numerical Estimates of Hydropower-Related Losses”, Appendix E of the Columbia River Basin Fish and Wildlife Program, at 7.

19 Id. at 3-4.

20 Id. at 6.

21 Reviewing the fishery agencies claims of “optimum sustainable runs” in the Snake River, prepared to support the Lower Snake River Compensation Program, Professor Salo found them “difficult to support”. E. Salo, “Special Report to the U.S. Army Corps of Engineers on two reports concerning proposed compensation for losses of fish caused by Ice Harbor, Lower Monumental, Little Goose, and Lower Granite Locks and Dam projects, Washington and Idaho”, at 26 (USACE Walla Walla Dist. June 26, 1974).

22 1987 Columbia River Basin Fish and Wildlife Program  203(b)(2), at 38.

23 Id.

24 Appendix D to 1987 Columbia River Basin Fish and Wildlife Program, at 6.

25 Id. at 212.

26 See, e.g., ISG, Return to the River 397.

Previous PageTable Of ContentsNext Page

This Web page was created using a Trial Version of HTML Transit 3.0.