NMFS' decision to list SONC coho must be set aside for three independent reasons. First, NMFS' has attempted to define "species" for purposes of the ESA in a fashion inconsistent with the definition Congress provided. Second, NMFS has abused its discretion in lumping Rogue River coho in with weaker California runs of coho salmon. Third, NMFS' implicit finding that Rogue River coho are "threatened" is not supported by any competent scientific data, and is hence arbitrary and capricious. Before turning to these legal arguments, it is important to understand Congressional intent in the question of "species" definition.
The United States cites at length legislative history regarding the importance of protecting endangered species from extinction. (U.S. Memo at 3-4) Losing the last pair of animals capable of interbreeding, so that the species vanishes from the face of the earth, is a very serious matter. Were that the case here, the District would not contest the entry of an injunction. However, "the purpose of the Endangered Species Act is to conserve endangered and threatened species and their critical habitats, not preserve every individual animal and plant". GAO, "Endangered SpeciesA Controversial Issue Needing Resolution", at 58 (July 2, 1979).28
Here the Secretary has listed a group of fish he calls Southern Oregon/Northern California ("SONC") coho. They are not a biological species, but a tiny fraction of the much larger coho salmon species Oncorhynchus kisutch. There is no chance that the biological species coho salmon, will vanish from the face of the earth. There are literally thousands of runs of coho salmon in rivers ranging from northern Japan through Kamchatka, across the Bering Sea to Alaska, and south through all coastal areas to California, as well as large transplanted populations in the Great Lakes. (See Response to RFA No. 1)
SONC coho are not a subspecies either. Thus they can only be protected under the Endangered Species Act only if they constitute a "distinct population segment that interbreeds when mature". 16 U.S.C. § 1531(16). Congress has warned that "distinct population segments" are not to be listed as a general matter. The Senate Report to the 1979 amendments stated that "the committee is aware of the great potential for abuse of this authority and expects the FWS to list populations sparingly and only when the biological evidence indicates that such action is warranted". S. Rep. No. 96-151, 96th Cong., 1st Sess. 7 (1979) (emphasis added). Senator Baker warned that listing authority must be "exercised with prudence and restraint", without which the ESA itself would "stand in jeopardy". Id. at 14.
The District asked NMFS in an Interrogatory to identify "each and every listing of a species present in the Pacific Northwest", and to state whether it had listed a "species", "subspecies", or "distinct population segment". The Response? NMFS has listed only distinct population segments. (Response to Interrogatory No. 1) Consistent with the overwhelming biases of its personnel, NMFS has never even exercised the discretion Congress expected it to exercise "sparingly."
Under the Chevron standard, "[f]irst, always, is the question whether Congress has directly spoken to the precise question at issue". 467 U.S. at 842. Congress has directly spoken to the question as to what sort of groups of animals may be listed for protection under the Endangered Species Act. There are three choices, for the Act declares that "[t]he term 'species' includes any subspecies of fish or wildlife or plants, and any distinct population segment which interbreeds when mature". 16 U.S.C. § 1531(16) (emphasis added). In this case, SONC coho are neither a "species" or "subspecies", but rather form a "distinct population segment".
Congress has considered the possibility that there may be population segments which interbreed when mature, and population segments that do not interbreed when mature. The SONC coho group does not interbreed when mature. It consists of numerous runs of salmon, none of which "interbreed when mature". They can't. When the breeding goes on, they are hundreds of miles apart. Indeed, there is little interbreeding even among the five stocks of wild Rogue River coho. (See Evenson Tr. 63)
Thus NMFS' species definition is flatly inconsistent with the text of the ESA, and must be set aside. That is not to say that NMFS' policy of protecting only "Evolutionarily Significant Units" or ESUs is erroneous per se. The ESU policy can be applied in perfect congruence with the statutory language. Here, for example, NMFS could have analyzed salmon stock-by-stock (the only rational approach to salmon management), and determined whether each stock met the "interbreeds when mature" test and the factors NMFS considers important in its ESU policy. But NMFS did not do so, at best for administrative convenience.29 Where the substantial rights of Grants Pass are at stake, better than administrative convenience ought to be required.
Careful review of the scientific evidence, or lack thereof, underlying NMFS' decision making ensures implementation of a policy which the Supreme Court has recently characterized as "primary" within the ESA:
"The obvious purpose of the requirement that each agency 'use the best scientific and commercial data available' is to ensure that the ESA not be implemented haphazardly, on the basis of speculation or surmise. While this no doubt serves to advance the ESA's overall goal of species preservation, we think it readily apparent that another objective (if not indeed the primary one) is to avoid needless economic dislocation produced by agency officials zealously but unintelligently pursuing their environmental objectives." Bennett v. Plenart, 117 S. Ct. 1154, 1168 (1997).
Here, NMFS officials have "zealously but unintelligently" pursued their "dambuster" objectives, with utter disregard to the actual likelihood that Rogue River coho will vanish from the face of the earth.
Where questions of statutory construction are secondary, and review of the agency's factual decision making primary, this Court is to apply the standard of review found in Motor Vehicle Mfg. Ass'n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 26, 43 (1983):
"Normally, an agency rule would be arbitrary and capricious [i] if the agency relied on factors which Congress has not intended it to consider, [ii] entirely failed to consider an important aspect of the problem, [iii] offered an explanation for its decision that runs counter to the evidence before the agency, or [iv] is so implausible that it could not be ascribed to a difference in view or the product of agency expertise."
NMFS' decision to include healthy Rogue River coho runs in a group of salmon given "threatened" status is "arbitrary and capricious" for all these reasons.
First, the only apparent (albeit unstated) rationale for NMFS' decision is that NMFS apparently prefer to pick two points on the coast and group all salmon within them for administrative convenience. But geographic proximity of distinct populations is not one of the statutory factors the Act tells the Secretary to consider in making listing decisions. See 16 U.S.C. § 1533(a)(1).
Second, NMFS really has "entirely failed to consider an important aspect of the problem" by failing to give any consideration whatsoever to giving individualized evaluation to Rogue River coho salmon.
Third, the vast majority of evidence before the agency concerning Rogue River coho was the evidence of long-term upward trends and some of the highest counts of returning adults ever measured. The decision to list the coho as "threatened" "runs counter to the evidence before the agency" and is frankly "implausible". As recently as March 24, 1998, one NMFS official opined that Rogue River coho represented "the only bright spot for coho salmon along the Oregon coast" (Becklin Decl. Ex. 1), yet they are the only Oregon coastal coho listed for protection under the ESA.
The Court may wonder how purported fishery scientists can look at a run of salmon that has been increasing for thirty years and determine that it is threatened with extinction. It is vital that the Court understand that the individuals involved in making this decision cannot be considered "scientists" under any conventional definition of "science".
In reviewing the admissibility of expert testimony under Rule 702 of the Federal Rules of Evidence, the United States Supreme Court recently issued a warning against the gradual dilution of scientific evidence in Federal Courts. Under Rule 702,
"The subject of an expert's testimony must be 'scientific . . . knowledge.' The adjective scientific implies a grounding in the methods and procedures of science. Similarly, the word knowledge connotes more than subjective belief or unsupported speculation. The term 'applies to any body of known facts or to any body of ideas inferred from such facts or accepted as truths on good grounds.' Webster's Third New International Dictionary 1252 (1986). Of course, it would be unreasonable to conclude that the subject of scientific testimony has to be "known" to a certainty; arguably, there are no certainties in science. [Citations omitted.] But, in order to qualify as 'scientific knowledge,' an inference or assertion must be derived by the scientific method." 509 U.S. at 589-90.
What then, is "the scientific method"? The Supreme Court explains:
"Ordinarily, a key question to be answered in determining whether a theory or technique is scientific knowledge that will assist the trier of fact will be whether it can be (and has been) tested. "Scientific methodology today is based on generating hypotheses and testing them to see if they can be falsified; indeed, this methodology is what distinguishes science from other fields of human behavior. Green 645. See also C. Hemple, Philosophy of Natural Science 49 (1966) ("[T]he statements constituting a scientific explanation must be capable of empirical test"); K. Popper, Conjectures and Refutations: The Growth of Scientific Knowledge 37 (5th ed. 1989) ("[T]he criterion of the scientific status of a theory is its falsifiability, or refutability, or testability") (emphasis deleted)." Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 593 (1993).
One can conceive of a scientific approach to assessing the risk of extinction that would qualify as "scientific knowledge". For example, one might use scientific data to make predictions of extinction risk for particular populations, and then test the predictions against reality: whether the species in question in fact goes extinct. And one need not wait for decades to accomplish this; one can use earlier historical data to predict abundance, and examine subsequent historical data to test the model.
As we will demonstrate below, NMFS has done no such thing. One of the problems is that "extinction science" is the province of a politically-driven group of individuals known as "conservation biologists". One law review article promoting the new "science" of "conservation biology" explained:
"A distinguishing feature of conservation biology is that it is mission oriented. Underlying any mission is a set of values. Philosophers of science now recognize that no science is value free [sic], despite all we were taught in school about the strict objectivity of the scientific method. Conservation biology is more value-laden than most sciences because it is not concerned with knowledge for its own sake but rather is directed toward particular goals. Maintaining biodiversity is an unquestioned goal of conservation biologists."30
Conservation biologists "view species extinction and loss as a crisis of major proportions that requires a drastic shift in our governing policies."31 In fact, the crisis is a hoax.32
Here, the focal point of NMFS' listing decision was a "Status Review of Coho Salmon from Washington, Oregon and California" issued by a group of conservation biologists known as the "Biological Review Team" or BRT.33 The Status Review very briefly discusses "assessment of extinction risk", citing papers by conservation biologists, (Soulé and Thompson, among others), which expressly eschew reliance on any approach that can remotely be characterized as scientific. For example, Thompson's paper cited in the Status Review lauds
". . . the legitimate role of subjective judgment in the scientific process. Marcot and Holthausen (1987) address this idea in the context of conservation biology, stating, 'There is much to be said for using subjective assessments and professional judgments for evaluating population viability, as the basic quantitative tools are still being hewn.' Soulé and Simberhoff (1986) argue similarly, writing, 'Intuition, common sense and judicious use of available data are still the state of the art.' Likewise, Zimmerman and Bierregaard (1986) advocate reliance on 'intuitive guessing . . . made by the field biologists involved.'
"This point is particularly important in an area such as conservation biology, where crisis is the norm rather than the exception. 'In crisis disciplines, one must act before knowing all the facts' (Soulé 1985). Stated alternatively, 'The luxuries of confidence limits and certainty are ones that conservation biologists cannot now afford' (Soulé 1980). Roughgarden (1983) describes the scientific process as one of building a convincing case, as distinguished from establishing proof by formal rules. In this context, the relevant question is not whether a state of endangerment can be proven; rather the question concerns the appropriate response to whatever amount of information exists (or can be gathered expeditiously) (Maguire 1991). If this view is accepted, it becomes clearly inappropriate to require that a finding of endangerment be corroborated by any particular model before taking action."34
In other words, a finding of "threatened" or "endangered" status is not a scientific finding at all. Rather, it consists of the subjective opinion of conservation biologists for whom "crisis is the norm rather than the exception". Congress never anticipated that the opinions of such individuals would rise to the level of the "best scientific and commercial evidence".
What is the legal significance of this philosophical detour? First, while we would not suggest that NMFS can only consider evidence that is admissible under the Federal Rules of Evidence, this Court ought to regard with some skepticism administrative decisions premised on "science" that, as explained below, miserably fails to meet the Daubert standard.
Second, the United States Supreme Court in Marsh v. Oregon Natural Resources Council, 490 U.S. 360, 378 (1989), has advised Federal district courts that "[w]hen specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own qualified experts even if, as an original matter, a court might find contrary reviews more persuasive". As set forth above, however, there are no such "qualified experts" in this case; only conservation biologists offering purely subjective opinions. Surely less deference is due purported experts with overwhelming biases based on their "crisis mentality". Moreover, the Court's review in Marsh was "'searching and careful'", id. (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971)), and considered in detail the relevant scientific evidence, id. at 378-85. In the final analysis, the Court affirmed the decision of the U.S. Army Corps of Engineers determining that asserted fisheries impacts were exaggerated and inaccurate. Id. at 385. There a federal dam operator was permitted to go forward notwithstanding the fishery concerns of state officials; here a state dam operator should be permitted to go forward notwithstanding the fishery concerns of federal officials.
Modeling runs by Mark Chilcote, filed herewith as DX64, describe the only quantitative assessment of extinction risk that NMFS bothered to obtain before listing Rogue River coho. (See generally Response to RFA Nos. 20-23.) It also provides a useful vehicle for summarizing the available scientific information on probabilities of extinction.
Mr. Chilcote reports that under 6% ocean survival conditions, the historical average for this stock during the modeling period (see also Mathews Decl. ¶ 7), the model produced a risk of "quasi-extinction" of 1.6%.35 The model results by themselves suggest that Rogue River coho are not likely to become endangered, i.e. are not "threatened", but the model grossly overstates the risk of extinction as well.
First, the model does not measure risk of extinction at all, but rather the risk that Rogue River coho populations will dip below 100. A NMFS technical paper explains "[g]iven that the ESA defines endangerment in terms of extinction (not just low numbers), perhaps the best alternative is to set [effective population size]=1, at least for indices that measure total population."36 In connection with assessing Snake River stocks, NMFS used a model of true extinction risk (a population of one), but abandoned it in this case with no explanation whatsoever.37
Second, Chilcote's model even includes a factor that tends to "trap" populations below 25 once they get there (see DX64, at 4 (referring to "forced extinction" at low densities), thereby increasing the number of model runs in which the end population is below 100. In fact, the classic "Ricker curve" that is the basis of the model predicts greater salmon productivity at lower number, which "tends to impart a certain resilience to the salmon population" (Satterthwaite Tr. 45) Rogue River coho salmon numbers are "well above any theoretical genetic 'bottleneck' level". (Mathews Decl. ¶ 8) 38
Third, the model does not purport to measure risk to the entire run of Rogue River coho; it was built only for fish spawning above Gold Ray Dam, so it is assessing risk to a population probably less than half the size of the total Rogue River population. To make matters worse, the model assumed that the "fully-seeded" population above Gold Ray Dam was only 2,907 (4,566 wild adult coho returned to Gold Ray Dam last year), and included arbitrary "caps" on the population (DX64, at 4).
Fourth, the model predicts greater extinction risk at higher populations.39 That alone ought to tell the Court that something is seriously wrong here. If the model did represent realitywhich it does notone could decrease the extinction risk to Rogue River coho by taking the screens off at Savage Rapids Dam.
Fifth, the model has a "double-whammy" effect that exaggerates the shock of random events on coho populations. Even though it includes a parameter to vary productivity in accordance with the observed variation in populations (including variability from varying ocean conditions), it also includes an "ocean" factor that is arbitrarily set to reduce productivity further if modeled ocean conditions are worse than average. It is only by setting the ocean survival factor at half the measured rate (see DX64, at 5, 9), and then adding the random factor that further reduces productivity, that any appreciable risks of "quasi-extinction" were generated.
In fact, future Rogue River coho salmon runs are likely to increase further because "the ocean can be expected to cycle back to a more productive phase, as it has tended to do through time". (Mathews Decl. ¶ 8) As Mr. Cramer points out, the fact that the Rogue River wild coho runs decreased only slightly in the face of four consecutive years of truly awful conditions is powerful evidence that they are not at risk of extinction. (See 2d Cramer Decl. ¶ 36)
The notion that random events other than human-caused impacts will cause extinction is purely theoretical. As one paper cited by the Status Review reported: ". . . no biologist has documented the extinction of a continental species of a plant or animal caused solely by non-human agencies such as competition, disease or environmental perturbation in situations unaffected by man."40 Harvest rates on adult coho salmon that "have at times exceeded 80%" (62 Fed. Reg. at 24,604) might cause extinction. The 1.3% take at Savage Rapids Dam never will; "observed large scale changes in smolt-to-adult survival and ocean harvest rate are orders of magnitude greater in their effect on Rogue coho returns than the small mortality on coho imposed at Savage Rapids Dam". (2d Cramer Decl. ¶ 37)
Sixth, the model contains no accounting of the effect of Cole Rivers hatchery. Common sense tells us that a hatchery producing fish of the identical ESU, some of which stray and go wild, will prevent Rogue River coho from ever vanishing so long as the hatchery is in operation, but NMFS nowhere addresses the stabilizing effect of the hatchery on extinction risk.
Seventh, the model was not tested or validated in any way by testing its predictions. When the authors of the textbook from which Mr. Chilcote drew his model tested it on shrew populations, they found "short expected persistence times" that were "surprising", given that the population "represents one of the larger and more stable populations in suburban areas near Lausanne".41 They speculated that recolonization, migration and dispersalall ignored in the modelmight account for the divergence from reality.42
The ability of salmon to stray and recolonize adjacent habitat is, of course, one of the features that make them nearly impossible to wipe out. (See Mathews Decl. ¶ 6) More sophisticated "meta-population" analyses that take account of this factor show no appreciable extinction risk for salmon,43 but NMFS refuses even to refer to such models. Mr. Michael Evenson, a declarant for the United States, acknowledged that given the presence of five subgroups of Rogue River coho salmon, Rogue River coho salmon were "unlikely" ever to become endangered. (Evenson Tr. 64-66 (emphasis added))
As to the likelihood of actual extinction:
"Q: You can agree, can you not, that there is no way that these fish will become extinct in the foreseeable future?
A: I would agree with that." (Evenson Tr. 56 (emphasis added))
Pressed to provide circumstances under which the Rogue River coho might go extinct, one ODFW witness suggested
". . . . Mount McLaughlin blowing up. National crisis. A great depression. It would be a guess on my part as to the risk associated with those particular events. Historically they happened in other societies. Historically, we've had catastrophic events such as Mount St. Helens blowing up and mud going down the Toutle River and killing everything that was living in the Toutle. So it's a possibility that's where the risk assessment comes in.
"Q: Ha[sn't] the Toutle River already been recolonized by the fish that were once in it when the mountain erupted?
"A: That is correct . . ." (Satterthwaite Tr. 47)
The District has obtained expert testimony from Steve Cramer and a professor at the University of Washington School of Fisheries, Stephen Mathews, to the same effect; these experts correctly "do not perceive Rogue River coho to be in immediate danger of extinction throughout all or a significant portion of their range, or likely to fall within this category in the foreseeable future". (Mathews Decl. ¶ 5; 2d Cramer Declaration ¶ 38) NMFS' witnesses do not appear to disagree with this view. (See Smith Tr. 82)
Worse still, it follows that the risk of extinction for the larger SONC coho group as a whole, with even more populations, is even more implausible. The more populations you have, the lower the risk of extinction for the group. (Evenson Tr. 64) Even NMFS acknowledges that the presence of multiple populations within the ESU "may provide some buffer against the ESU's extinction". 62 Fed. Reg. at 24,591.
A properly specified metapopulation model of extinction risk for the SONC coho population would demonstrate no extinction risk whatsoever, in accord with common sense. Such models (including those mentioned in the papers cited in the Status Review) are among the "best available scientific and commercial data" NMFS could have utilized, but did not.
28 Another principal purpose of the ESA is to avoid the irrevocable disappearance of genetic material (though advances in technology ŕ la Jurassic Park may have undermined Congress' premises). Yet "Pacific salmon populations are typically characterized by different frequencies of the same suite of alleles, rather than by qualitative differences in the types of alleles present". R. Waples, "Pacific Salmon, Oncorhynchus spp., and the Definition of 'Species' Under the Endangered Species Act", Marine Fisheries Review, 53(3), at 14 (1991). Thus the loss of any particular run of Pacific salmon does not mean that any unique genetic material is lost at all.
29 At worst, it is the product of improper political influence over the listing process; without additional discovery, the District and the Court will never know.
30 R. Noss, "Some Principles of Conservation Biology as They Apply to Environmental Law", 69 Chicago-Kent L. Rev. 893, 895 (1994).
31 R. Keiter, "Conservation Biology and the Law: Assessing the Challenges Ahead", 69 Chicago-Kent L. Rev. 911 (1994).
32 It is a popular fad that to imagine that species are dropping like flies, but for what journalist Gregg Easterbrook has called the Fly Corpse Factor. If species were dropping like flies, the corpses should be piling up by now. Instead species corpses turn out to be exceedingly difficult to locate. G. Easterbrook, A Moment on the Earth 558. He points out that because of the Spotted Owl listing, the Northwest forests are among the best-studied ecologies. Yet researchers have not uncovered a single actual extinction, which seems revealing given that every graduate student involved in a Northwest forest field study is acutely aware that documenting a species loss would make his or her academic career. Id. at 559. While protecting genetic diversity is important, it is worth remembering that
[s]cientists now believe Earths ecosphere has become progressively more diverse, playing host to a greater range of species and gene lines as the ages have passed. Edward O. Wilson of Harvard University, a leading contemporary biologist, thinks that at present global genetic diversity is the highest ever, with perhaps as many as 100 million species walking the earth. Id. at 36.
33 L. Weitkamp et al., "Status of Coho Salmon from Washington, Oregon and California", NOAA Technical Memorandum NMFS-NWFSC-24 (Sept. 1995).
34 G. Thompson, "Determining Minimum Viable Populations under the Endangered Species Act", NOAA Technical Memorandum NMFS F/NWC-198 (June 1991), at 30-31 (emphasis added).
35 These results are not presented in DX64, having been omitted from page 7 of that report because the report did not focus on Rogue River coho (see DX64, at 9, referring to higher ocean survivals for Rogue River fish). The District is in the process of obtaining testimony from Mr. Chilcote and assures the Court that the factual representations set forth herein concerning Mr. Chilcote's model are the product of work in progress.
36 G. Thompson, "Determining Minimum Viable Populations under the Endangered Species Act", NOAA Technical Memorandum NMFS F/NWC-198, at 35. This scientist also noted that "the available rules of thumb can provide a preliminary, order-of-magnitude diagnosis. If a population contains more than, say, 10,000 individuals, it would probably take some fairly extreme extenuating circumstances to consider it endangered." Id. at 33. Rogue River coho populations have been above 10,000 for yearseven before one considers a single year class of salmon as one-third of the total population. Though this paper is cited in NMFS' Status review, NMFS supplies no reason not to apply the "rule of thumb", and does not even discuss it.
37 S. Cramer & D. Neeley, "Evaluation of Delisting Criteria and Rebuilding Schedules for Snake River Spring/Summer Chinook, Fall Chinook, and Sockeye Salmon", Report for Contract No. DE-AM79-93BP99654, June 1993, at II ("We reviewed the extinction model that NMFS used as a basis for listing. The model is very sensitive to extinction number, and extinction was defined by NMFS in the model as the return of only one fish."); cf. Response to RFA No. 24.
38 But as one of the papers cited in the Status Review pointed out: "There is a danger that a student might . . . conclude that genetic malfunction was a significant cause of extinction in the wild. The literature of conservation genetics provides no hint that it might be otherwise. Yet no instance of extinction by genetic malfunction has been reported . . .". G. Caughley, "Directions in conservation biology", 63 J. Animal Eco. 215, 239 (1994) (emphasis added). The Status Review candidly concedes that "we are not aware of empirical evidence for inbreeding depression or loss of genetic variability in any natural or hatchery populations of Pacific salmon or steelhead". (Status Review at 100.) Moreover, very small populations that have "weathered a bout of inbreeding may come out of it with fitness enhanced because inbreeding exposes deleterious recessives and allows them to be purged from the gene pool. That is precisely the method used by animal breeders to remove deleterious alleles." G. Caughley, at 221. In other words, as a theoretical matter, swings down to small populations may improve the genetic stock of salmon.
39 This can be seen by examining the second line down (Rogue River) in the charts in Figure 2 of DX64, at 8. It is apparent that the "probability of quasi-extinction" is rising in the second and third charts as the "seeding level", a measure of abundance, rises. This effect appears to be more and more pronounced as more and more realistic measures of ocean survival are used.
40 C. Caughley, supra, at 240 (quoting and agreeing with another scientist cited by the Status Review, M.E. Soulé).
41 M.A. Bergman et al., Risk Assessment in Conservation Biology 104 (Chapman and Hall, London 1993).
43 See, e.g., J. Emlen, "Population Viability of the Snake River Chinook Salmon (Oncorhynchus tshawytscha)", at 14.
This Web page was created using a Trial Version of HTML Transit 3.0.