Defenders of Wildlife v. Hall

565 F. Supp. 2d 1160, 68 ERC 1146, 68 ERC (BNA) 1146, 2008 U.S. Dist. LEXIS 55071

• Court: District Court, D. Montana
• Decided: July 18, 2008
• Docket: CV 08-56-M-DWM
• Status: Published

Opinion

ORDER

DONALD W. MOLLOY, District Judge.

I. Introduction

This case, like a cloud larger than a man’s hand, will hang over the northwest states of Montana, Idaho, and Wyoming until there has been a final determination of the complex issues presented. Those issues must be answered in accordance with the intent of Congress as stated in the Endangered Species Act and its implementing regulations. Here, Plaintiffs challenge the U.S. Fish & Wildlife Service’s decision to designate and delist a northern Rocky Mountain gray wolf distinct population segment under the Endangered Species Act (“ESA”), 16 U.S.C. § 1536. In seeking to alter the course of that decision, Plaintiffs move for a preliminary injunc tion. They ask the Court to reinstate ESA protections for the wolf, at least while this lawsuit is pending. In support of their motion, Plaintiffs argue (1) even though the environmental impact statement on wolf reintroduction specifically conditions the delisting decision on a Finding of Sub-population Genetic Exchange, the Fish & Wildlife Service delisted the wolf when there is no plausible showing of that genetic exchange between the Greater Yellowstone core recovery area and the northwestern Montana and central Idaho core recovery areas; (2) the Service approved Wyoming’s 2007 wolf management plan even though the Wyoming plan still contains provisions that the Service earlier deemed inadequate; and (3) the Fish & Wildlife Service did not consider the several states’ liberal defense of property laws in concluding the states’ wolf management plans were adequate. The argument concludes with the claim that a preliminary injunction is necessary because wolves are not likely to survive the increased incidents of human-caused mortality that will occur under state management.

In my view, Plaintiffs are likely to succeed on the majority of the claims relied upon in their request for a preliminary injunction. In particular, (1) the Fish & Wildlife Service acted arbitrarily in de-listing the wolf despite a lack of evidence of genetic exchange between subpopula-tions; and (2) it acted arbitrarily and capriciously when it approved Wyoming’s 2007 plan despite the State’s failure to commit to managing for 15 breeding pairs and the plan’s malleable trophy game area. In both instances, the Fish & Wildlife Service altered its earlier position without providing a reasoned decision for the change based on identified new information.

As recently as 2002, the Service determined genetic exchange between wolves in the Greater Yellowstone, northwestern Montana, and central Idaho core recovery areas was necessary to maintain a viable northern Rocky Mountain wolf population in the face of environmental variability and stochastic events. The Fish & Wildlife Service nevertheless delisted the wolf without any evidence of genetic exchange between wolves in the Greater Yellowstone core recovery area and the other two core recovery areas. To justify its decision, the Service relied on the same information that was available to it when it determined genetic exchange was necessary in 2002.

In 2004, the Fish & Wildlife Service rejected Wyoming’s 2003 wolf management plan. The Service determined the 2003 plan was inadequate to protect wolves because it permitted Wyoming state officials to classify the wolf as a predatory animal throughout the state and then failed to clearly commit the state to managing for 15 breeding pairs within its borders. Before delisting the wolf, the Fish & Wildlife Service approved Wyoming’s revised 2007 plan. This revised plan suffers from the same deficiencies as the 2003 plan: it classifies the wolf as a predatory animal in almost 90 percent of the state and only commits the state to managing for 7 breeding pairs outside the national parks. In supporting its decision to approve Wyoming’s 2007 plan, the Service does not offer any information not available to it when it rejected the 2003 plan. Armed with the same information, the agency flip-flopped without explanation. While the Fish & Wildlife Service can change its recovery criteria, it must nevertheless provide a reasoned analysis for the change of position and if it does so, its decision is entitled to deference. The Service has failed to do so here. Thus, in my view, Plaintiffs are likely to succeed on several of their claims.

Plaintiffs have also shown a significant possibility of irreparable injury. More wolves will be killed under state management than were killed when ESA protections were in place. Idaho, Montana, and Wyoming each have public wolf hunts scheduled for this fall. Additionally, the states’ defense of property laws permit the killing of wolves in more circumstances than defense of property regulations under the ESA. The killing of wolves during the pendency of this lawsuit will further reduce opportunities for genetic exchange among subpopulations. Genetic exchange that did not take place between larger subpopulations under ESA protections is not likely to occur with fewer wolves under state management. Absent genetic exchange, the viability of the wolf will be threatened by future environmental variability and stochastic events.

Because Plaintiffs have demonstrated a likelihood of success on the merits of several of their claims and the possibility of irreparable injury, their motion for a preliminary injunction is granted. The limited preliminary relief will reinstate ESA protections for the northern Rocky Mountain gray wolf to ensure the species is not imperiled during the pendency of this lawsuit.

II. Factual Background

The gray wolf is the largest wild member of the dog family. 72 Fed.Reg. 6106 (Feb. 8, 2007). Wolves generally live in packs of 2 to 12 animals and have strong social bonds. Id. at 6107. Wolf packs consist of a breeding pair (the alpha male and alpha female), their offspring from previous years, and an occasional unrelated wolf. Id. Generally, only the alpha male and alpha female of a pack breed. Id. Litters are born in April and average around 5 pups. Id. All pack members help feed and protect the pups as they grow. Pups are weaned at 5 to 6 weeks and then are mature enough to travel with the pack by around October. Packs typically occupy territories from 200 to 500 square miles. Each pack will defend its territory against other wolves and wolf packs. Id.

Wolves were once abundant throughout most of North America. Id. at 6106. Wolf hunting and an active, government-sponsored eradication program resulted in the extirpation of wolves from more than 95 percent of their range in the lower 48 states. Id. at 6106, 6125. They were exterminated in Idaho, Montana, Wyoming, and adjacent southwestern Canada by the 1930s. Id. at 6107.

The Northern Rocky Mountain gray wolf was listed under the ESA in 1974. 39 Fed.Reg. 1171 (Jan. 4, 1974). In 1987, the Fish & Wildlife Service developed a wolf recovery plan. 72 Fed.Reg. at 6107. This plan established a recovery goal of at least 10 breeding pairs and at least 100 wolves for three consecutive years in each of three core recovery areas: northwestern Montana, central Idaho, and the Greater Yellowstone area. Id.

In 1994, the Fish & Wildlife Service proposed designating portions of Idaho, Montana, and Wyoming as two nonessential experimental population areas for the gray wolf under § 10(j) of the ESA. 59 Fed.Reg. 60,252 (November 22, 1994); 59 Fed.Reg. 60,266 (November 22,1994). Before introducing the experimental wolf populations, the Service prepared an Environmental Impact Statement on the Reintroduction of Gray Wolves to Yellowstone National Park and Central Idaho (the “1994 EIS”). 72 Fed.Reg. at 6107. Northwestern Montana was not included because the wolves had moved naturally into that part of the state. The 1994 EIS evaluated whether the population goals for delisting wolves contained in the 1987 recovery plan would result in a viable wolf population. 1994 EIS, App. 9. The same EIS concluded that the plan goal of 10 breeding pairs and 100 wolves in three separate recovery areas for a period of three consecutive years was “somewhat conservative ... and should be considered minimal.” 1994 EIS, App. 9, at 42. It noted “[tjhirty or more breeding pairs comprising some 300+ wolves in a meta-population (a population that exists as partially isolated sets of subpopulations) with genetic exchange between subpopulations should have a high probability of long-term persistence.” 1994 EIS, App. 9, at 42. Over the next two years, in 1995 and 1996, the Service reintroduced wolves captured in southwestern Canada into central Idaho and into the Greater Yellowstone' Area. 72 Fed.Reg. at 6108.

The northern Rocky Mountain wolf population met the Fish & Wildlife Services’ numeric recovery goal of 30 breeding pairs and 300 wolves for the first time in 2000. Id. In late 2001 and early 2002, the Service conducted another evaluation of what constitutes a recovered wolf population and reaffirmed the recovery criteria set forth in the 1994 EIS. Id. at 6107. The criteria included genetic exchange between the three subpopulations. By the end of 2007, the northern Rocky Mountain wolf population had achieved the Service’s numerical recovery goal for eight consecutive years. Id. at 6108. It had not achieved genetic exchange between the three subpopula-tions.

The Service then asked the states of Idaho, Montana, and Wyoming to prepare wolf management plans, specifying how each state would manage wolves after de-listing. The Service approved the Idaho and Montana wolf management plans in January 2004, but it rejected Wyoming’s plan. Id. at 6127; First Harbine Deck Ex. 11. In early 2007, the Service proposed designating a northern Rocky Mountain gray wolf distinct population segment and removing it from the ESA’s threatened and endangered species list. 72 Fed.Reg. 6106. Even so, the Service stated unless and until Wyoming developed an adequate wolf management plan, wolves in Wyoming outside the national parks would remain on the endangered species list. Id. at 6117. Wyoming subsequently revised its management plan, and the Service announced it was satisfied with Wyoming’s new plan. First Harbine Decl. Ex. 12.

On February 27, 2008, the Fish & Wildlife Service issued a final rule designating a northern Rocky Mountain gray wolf distinct population segment and removing it from the list of threatened and endangered species. 73 Fed.Reg. 10,514 (Feb. 27, 2008). The distinct population segment takes in all of Idaho, Montana, and Wyoming. It also includes eastern Washington, eastern Oregon, and northern Utah. Id. at 10,518. Wolf packs are known to occupy only a portion of this area. The areas with known wolf packs include northwestern Montana, central Idaho, and the Greater Yellowstone Area. Id.

III. Statutory Background

A. Endangered Species Act

The ESA is meant to conserve the ecosystems upon which endangered and threatened species depend and to provide a program for the conservation of such species. 16 U.S.C. § 1531(b). To this end, Congress enacted Section 4 of the Act which requires species that are in danger of extinction to be listed as endangered or threatened after public notice and comment. Id. § 1533. The ESA defines “species” to include “any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature.” Id. § 1532(16). An endangered species is “any species which is in danger of extinction throughout all or a significant portion of its range.” Id. § 1532(6). A threatened species is “any species which is likely to become an endangered species within the foreseeable future throughout all or a significant portion of its range.” Id. § 1532(20).

Through the ESA, Congress requires the Secretary to examine five factors when determining whether a species is threatened or endangered. Id. § 1533(a)(1). The factors include:

(A) the present or threatened destruction, modification, or curtailment of its habitat or range;

(B) overutilization for commercial, recreational, scientific, or educational purposes;

(C) disease or predation;

(D) the inadequacy of existing regulatory mechanisms; [and]

(E) other natural or manmade factors affecting its continued existence.

Id. Any one of the factors is sufficient to support a listing determination if that factor causes the species to be in danger of extinction or likely to become an endangered species in the foreseeable future throughout all or a significant portion of its range. See id. Listing decisions must be made “solely on the basis of the best scientific and commercial data available.” Id. § 1533(b)(1)(A); 50 C.F.R. § 424.11(b). Listings must be made without reference to possible economic or other impacts of such a determination. 16 U.S.C. § 1533(b)(1)(A); 50 C.F.R. § 424.11(b).

If a species is listed as endangered or threatened, all Federal departments and agencies must seek to conserve the species. 16 U.S.C. § 1531(c).

B. Administrative Procedure Act

Judicial review of an agency’s compliance with the ESA is governed by the judicial review provisions of the APA. Or. Natural Res. Council v. Allen, 476 F.3d 1031, 1035 (9th Cir.2007). Agency decisions can only be set aside under the APA if they are “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971) (quoting 5 U.S.C. § 706(2)(A), overruled on other grounds by Califano v. Sanders, 430 U.S. 99, 97 S.Ct. 980, 51 L.Ed.2d 192 (1977)). Review under the arbitrary and capricious standard is “narrow,” but “searching and careful.” Marsh v. Or. Natural Res. Council, 490 U.S. 360, 378, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989). Agency action can be set aside “if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Motor Vehicle Mfrs. Ass’n of U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). When an agency action is challenged, a court must ask “whether the [agency’s] decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment ... [The court] also must determine whether the [agency] articulated a rational connection between the facts found and the choice made. [The] review must not rubber-stamp ... administrative decisions that [the court deems] inconsistent with a statutory mandate or that frustrate the congressional policy underlying a statute.” Ocean Advocates v. U.S. Army Corps of Eng’rs, 402 F.3d 846, 859 (9th Cir.2005) (internal citations and quotations omitted). Although the court’s review must be searching, the court may not substitute its judgment for that of the agency or merely determine it would have decided an issue differently. Or. Natural Res. Council, 476 F.3d at 1035.

IV. Preliminary Injunction Standard

The parties disagree about which standard the Court should apply in ruling on Plaintiffs’ motion for a preliminary injunction. The Ninth Circuit has articulated three different standards for issuance of a preliminary injunction. “Under the traditional test, a plaintiff must show: (1) a strong likelihood of success on the merits, (2) the possibility of irreparable injury to plaintiff if preliminary relief is not granted, (3) a balance of hardships favoring the plaintiff, and (4) advancement of the public interest (in certain cases).” Ranchers Cattlemen Action Legal Fund v. USDA, 415 F.3d 1078, 1092 (9th Cir.2005) (quotation omitted). “The alternative test requires that a plaintiff demonstrate either a combination of probable success on the merits and the possibility of irreparable injury or that serious questions are raised and the balance of hardships tips sharply in his favor.” Id. “The[] two formulations [in the alternative test] represent two points on a sliding scale in which the required degree of irreparable harm increases as the probability of success decreases.” Id. at 1092-93; see also Lands Council v. McNair, 537 F.3d 981, 2008 WL 2640001, *3 (9th Cir. July 2, 2008).

The third test applies to ESA injunctions. Applying this test, the court must still measure the likelihood of success on the merits as well as the possibility of irreparable injury. See Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 422 F.3d 782, 795 (9th Cir.2005). Unlike the two standards previously described, however, once likelihood of success and irreparable injury are shown, the court may not fine-tune its analysis by weighing the hardships of the parties. Id. at 793-94. This legal principal is a direct acknowledgment of congressional intent. Under the ESA, “Congress has spoken in the plainest of words, making it abundantly clear that the balance [of hardships] has been struck in favor of affording endangered species the highest of priorities,” and courts “may not use equity’s scales to strike a different balance.” Id. at 794. What this means is that if a plaintiff is likely to succeed on its ESA claim and irreparable injury is possible, then the court should issue an injunction when it is necessary to effectuate the purpose of the ESA. 1 Id. at 796.

The Intervenors in this case claim the third test only applies when a species is listed as threatened or endangered at the time a preliminary injunction is sought. The minor premise of this syllogism is that because the northern Rocky Mountain gray wolf is no longer listed, the traditional or alternative test should be applied. Even though the third test has only been applied in cases involving species that were listed at the time a preliminary injunction was sought, the cases, nonetheless, speak of the ESA and its purposes in broad terms and thus do not preclude applying the third test to delisting decisions. If this were not the case, the intent of Congress to grant listed species the highest priority in the balance of concerns could be readily circumvented. Moreover, a mistaken decision to delist a species that still qualifies for protection under the ESA is no less harmful to the species than a decision to implement a project without consideration of its effects on an already listed species. In the context of requests for permanent injunctive relief in ESA cases, the Ninth Circuit instructs district courts may not balance the hardships regardless of whether a plaintiffs challenge involves Section 7 (formal consultation) or Section 4 (listing/delisting) of the ESA. Biodiversity Legal Found. v. Badgley, 309 F.3d 1166, 1177 (9th Cir.2002) (refusing to balance hardships in case challenging government’s failure to timely rule on petitions to list certain species as threatened or endangered). The same reasoning applies in the context of a preliminary injunction. Because Congress has, in the plainest of words, made a policy determination to give endangered or threatened species the highest priority, the third test will be applied in this case.

V. Analysis

A. Lack of Connectivity

Plaintiffs claim the Fish & Wildlife Service violated the Endangered Species Act by delisting the wolf while knowing there was a lack of genetic exchange between populations in the three northern Rocky Mountain core recovery areas. The legal deficiency claimed is that in doing so the Service ignored its own recovery criteria and ignored the best science available. In the 1994 EIS, the Fish & Wildlife Service identified specific recovery criteria of “[tjhirty or more breeding pairs comprising some 300 + wolves in a metapopulation (a population that exists as partially isolated sets of subpopulations) with genetic exchange between subpopulations.” 1994 EIS, App. 9, at 42; 73 Fed.Reg. at 10,521. Plaintiffs point to a 2007 genetics study commissioned by the Fish & Wildlife Service (the “VonHoldt Study”) that confirmed a metapopulation does not yet exist because wolves in Yellowstone National Park have remained genetically isolated from wolves in the northwestern Montana and central Idaho core recovery areas since their reintroduction in 1995. First Harbine Decl. Ex. 9.

The argument is that without genetic exchange between core recovery areas, wolves faces serious threats to survival. In support of this claim, Plaintiffs rely on the VonHoldt Study as well as the Fish & Wildlife Service’s own observations. The VonHoldt Study concluded “if the Yellowstone wolf population remains relatively constant at 170 individuals (estimated to be Yellowstone’s carrying capacity), the population will demonstrate substantial inbreeding effects within 60 years,” resulting in an “increase in juvenile mortality from an average of 23 to 40%, an effect equivalent to losing an additional pup in each litter.” 2 First Harbine Decl. Ex. 9, at 19. The contention is bolstered by Plaintiffs’ reasoning that the Fish & Wildlife Service itself observed that without ongoing genetic exchange, isolated subpopulations of merely 100 individuals and 10 breeding pairs will not exhibit genetic diversity sufficient to withstand environmental variability and stochastic events.

The Fish & Wildlife Service acknowledges there is no evidence of genetic exchange between wolves in Yellowstone National Park and the northwestern Montana or central Idaho core recovery areas. 3 Nevertheless, the Service now takes the position that documented proof of DNA exchange is not required to achieve a me-tapopulation under the 1994 EIS. According to the Service and its recent interpretation, the 1994 EIS emphasizes spacial distribution of wolves and the potential for genetic exchange. The Fish & Wildlife Service argues that the 1994 EIS does not define the term “genetic exchange” and it does not discuss the need for proof of DNA exchange. The Service concludes the potential for genetic exchange contemplated by the 1994 EIS has been achieved because of documented cases of dispersal between the Greater Yellowstone core recovery area and the other two core recovery areas.

Even if the 1994 EIS requires documented proof of DNA exchange, the Fish & Wildlife Service maintains its decision to delist the wolf was not arbitrary and capricious. The Service now claims that meeting the criteria in a recovery plan is not dispositive of whether a species should be delisted. This argument is in turn predicated on the proposition that the Service’s decision was properly based on its consideration of the five criteria set forth in the ESA regardless of the 1994 EIS’s goals. See 16 U.S.C. § 1533(a)(1). In particular, the Service stresses its determination that the lack of genetic connectivity between wolves in Yellowstone National Park and the wolves in the rest of the northern Rocky Mountains is not a threat to survival of the species because -there is a high level of genetic diversity within the Yellowstone National Park population and evidence of documented wolf dispersals. The Service refers to much smaller wolf populations with lower levels of genetic diversity where the wolves have still persisted for decades. The Service takes the view that the VonHoldt Study’s dire predictions of inbreeding and increased juvenile mortality are not based on the best available science. If this determination is accurate, the balance of the argument is that the Service’s experts were justified in rejecting the Study’s predictions. 4

The Fish & Wildlife Service’s assertion that the 1994 EIS requires only the potential for genetic exchange, not actual genetic exchange, is disingenuous. The 1994 EIS clearly requires “a metapopulation ... with genetic exchange between subpo-pulations.” 1994 EIS, App. 9, at 42. The 1994 EIS does not define the term “genetic exchange” because the term can only mean one thing: exchange of genetic material between subpopulations.

Realizing the weakness of its argument regarding the interpretation of the term “genetic exchange,” the Service next suggests genetic exchange has occurred, but it was not detected by the VonHoldt Study because of the Study’s limitations. The Study sampled 30 percent of the wolves in Yellowstone National Park and only conducted sampling until 2004. Bangs Decl. ¶ 18C. In the absence of proof, the Service surmises wolves not sampled, or wolves born after 2004, may have a genetic link with wolves in other core recovery areas. Bangs Decl. ¶ 18C. The argument continues with the observation that the Von-Holdt Study looked only at wolves in Yellowstone National Park, not wolves in the Greater Yellowstone core recovery area. Bangs Decl. ¶ 18C. The Service then opines genetic exchange is more likely to have occurred among wolves in the Greater Yellowstone core recovery area because of documented dispersals into that area of wolves from northwestern Montana and central Idaho. Mech Decl. ¶ 17; Bangs Decl. ¶ 18C.

The Fish & Wildlife Service’s speculation about genetic exchange is not convincing. The VonHoldt Study did not collect DNA samples from every wolf in Yellowstone National Park. This fact, however, does not render the Study, or its findings, useless. There is no question about the adequacy of the Study’s sample size or its statistical significance. Nor does the Service contend the testing methods used in the Study were flawed. None of the wolves tested in the Study showed a genetic link to wolves in northwestern Montana or to wolves in central Idaho. Additionally, in the thirteen years since wolves were reintroduced in Yellowstone National Park, there have been only four to twelve documented cases of wolves dispersing from or into other core recovery areas. 73 Fed.Reg. 10,553. This limited number of dispersals, along with the VonHoldt Study’s finding that a sample of 30 percent of the wolves in Yellowstone National Park showed no genetic connection with wolves in the other core recovery areas, is strong evidence that genetic exchange has not occurred.

In a final attempt to justify its decision to delist the wolf, the Fish & Wildlife Service rejects its own recovery criteria. The Service argues, even if the 1994 EIS recovery criteria required genetic exchange and such exchange has not occurred, the Service is entitled to change course and conclude the continued existence of the species is not threatened by a lack of genetic connectivity. The road less traveled may make.all the difference but the difference must be measured against the law’s requirements.

While the Service is entitled to change its recovery criteria, it must provide a “reasoned analysis” for doing so. See Motor Vehicle Mfrs. Ass'n, 463 U.S. at 42, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). In this case, the Service has not sufficiently justified or explained its change of course. The obvious shift focuses exclusively on the wolves’ success in meeting the recovery criterion of 30 breeding pairs and 300 wolves. The genetic diversity requirement for viability is pushed to the back burner of consideration with no explanation of its precipitous drop in importance. The Service instead suggests the 30/300 criterion is the magic tipping point at which the wolves will no longer be endangered. Yet, in 1994, the Service expressly rejected this numerical criterion in favor of recovery criteria that required not only numerical abundance, but also genetic exchange. The Service’s original conclusion that “[tjhirty or more breeding pairs comprising some 300 + wolves in a metapopulation (a population that exists as partially isolated sets of subpopulations) with genetic exchange between subpopulations should have a high probability of long-term persistence” is created in a way left unexplained. 1994 EIS, App. 9, at 42; see also id. (“It is fairly clear that ten breeding pairs in isolation will not comprise a ‘viable’ population (i.e., have a high probability of survival for a long period without human intervention)”); id. (“The importance of movement of individuals between sub-populations cannot be overemphasized.”). Although the Service now says genetic exchange is unnecessary, it provides no persuasive reasons for this change of course that were not known in 1994, when the new criteria were established, or in 2001 and 2002, when the criteria were reaffirmed.

In an effort to justify its change of course, the Service focuses on the likelihood that genetic exchange will occur in the near future. The Service points to evidence of wolf dispersals, 73 Fed.Reg. at 10,552-10,554; Bangs'Ded. ¶ 18C, and the inherent mobility of wolves, Mech Decl. ¶¶ 17, 19. Evidence of four to twelve dispersers between Yellowstone National Park and the northwestern Montana and central Idaho core recovery areas in the last thirteen years is left hanging as a promising predictor of future genetic exchange. Moreover, as the Service itself acknowledges, the chance of future genetic exchange is lessened considerably because more wolves will be killed under state management plans than under the ESA. It stands to reason that fewer wolves means less opportunity for dispersal and hence less chance for genetic exchange, further undermining the argument that the switch is not arbitrary.

The argument boils down to a contention that the original requirement for genetic exchange now makes no difference for wolf viability. The Service now argues wolves are not likely to become endangered absent genetic exchange. In rejecting the VonHoldt Study’s predictions of inbreeding and increased juvenile mortality among Yellowstone National Park wolves, the Service contends the Yellowstone National Park wolf population was founded with high genetic diversity and its diversity continues to be high. 73 Fed.Reg. at 10,552-10,554; Mech Decl. ¶ 14. It resorts to the observation that much smaller, more isolated wolf populations with lower levels of genetic diversity have persisted for decades. 73 Fed.Reg. at 10,552-10,554; Mech Deck ¶ 14. For example, the Service points to a wolf population on Isle Royale National Park in Michigan that began with two founders in 1949 and has remained very small (less than 50 wolves) and isolated, but persisted until the present time. 73 Fed.Reg. at 10,552-10,554; Mech Deck ¶ 14.

The rationale for rejecting the VonHoldt Study’s predictions is not convincing nor well explained. When the Service established the recovery criteria in the 1994 EIS, it knew that the Yellowstone National Park population would be founded with high genetic diversity. It also knew, or should have known, of other isolated wolf populations that had survived despite lower levels of genetic diversity. The Isle Royale National Park population, for example, had been in existence since 1949. Nevertheless, despite this knowledge, the Service concluded genetic exchange was necessary to maintain a viable wolf population. The Service provides no new evidence or research to support its change of course. Congress does not intend agency decision making to be fickle. When it is, the line separating rationality from arbitrariness and capriciousness is crossed.

Although the Service’s recovery criteria are not binding, the Service must provide adequate reasons for rejecting those criteria. Here, the Service continues to stand behind one component of the recovery criteria- — 30 breeding pairs and 300 wolves— but rejects another component — genetic exchange — as unnecessary. In doing so, the Service provides no new evidence or research that did not exist when the recovery criteria were established. The Service cannot change course without reason. The change of course is especially problematic in this case because delisting will undeniably reduce the chances for future genetic exchange. At oral argument, it was acknowledged that once delisting takes place, it is nearly impossible to reverse course. Although the Service finds comfort in the fact that Idaho, Montana, and Wyoming have committed to managing for at least 150 wolves in 15 breeding pairs, the record in the case demonstrates genet ic exchange is not likely to occur with these numbers. At the time of delisting, there were approximately 1,513 wolves in 106 breeding pairs in the northern Rocky Mountains. Jimenez Deck ¶ 3. Genetic exchange that did not occur under these conditions is not likely to occur with fewer wolves and fewer breeding pairs. Because the wolf does not meet the 1994 EIS recovery criteria and the Fish & Wildlife Service has not provided adequate reasons for rejecting those criteria, Plaintiffs are likely to succeed on their lack of connectivity claim.

B. State Regulatory Mechanisms

1. Wyoming’s 2007 Wolf Management Plan

Plaintiffs allege the Fish & Wildlife Service also violated the ESA by delisting the northern Rocky Mountain gray wolf even though provisions in Wyoming’s 2007 wolf management plan that the Service had earlier rejected and found to be inadequate to protect wolves were still a part of the plan now approved. In particular, the Service rejected Wyoming’s earlier 2003 wolf management plan because, among other things, (1) the plan failed to clearly commit to managing for at least 15 wolf packs in Wyoming; and (2) the predatory status of wolves under the plan did not “provide sufficient management controls to assure the Service that the wolf population [would] remain above recovery levels.” First Harbine Deck Ex. 11. The argument here is that Wyoming’s 2007 plan suffers from the same defects.

Plaintiffs maintain that just like the 2003 plan, the 2007 plan fails to clearly commit the State to managing for at least 15 breading pairs within the state. They claim Wyoming’s commitment to maintaining only seven (7) breading pairs outside the National Park Units is insufficient to sustain the wolf population in Wyoming because it does not guarantee fifteen (15) breeding pairs in the state if the National Park Units’ population drops below eight (8) breeding pairs. Plaintiffs pointedly show that under the 2007 plan, wolves in ninety (90) percent of the State of Wyoming remain classified as predators and the remaining ten (10) percent of the geographical area of the state is classified as trophy game area. That is a small, malleable area.

The Fish & Wildlife Service deals with this contention by asserting that the 2007 plan cures both deficiencies found in the 2003 plan. The Service maintains Wyoming has clearly committed to maintaining 15 breeding pairs in the state regardless of the number of breeding pairs in the National Parks Units. The Service further contends the designated trophy game area in the 2007 plan is sufficient to sustain Wyoming’s share of the wolf population. While the Service acknowledges Wyoming’s 2007 plan still classifies the wolf as predatory in ninety (90) percent of the state, it notes that even so the majority of this area is unsuitable wolf habitat. In the Service’s view, the trophy game area under the 2007 plan encompasses seventy (70) percent of the suitable wolf habitat in Wyoming.

Wyoming argues Plaintiffs’ claims regarding the inadequacies in the 2007 plan should fail because they are speculative. Wyoming claims Plaintiffs, like Cassandra, are concerned about issues that may only arise in the future. If this is the case, Wyoming asserts, then such concerns are not relevant to this analysis.

Plaintiffs are likely to succeed on this claim as well. Wyoming’s 2003 plan classified wolves in the State as predatory so long as there were seven (7) wolf packs outside the National Park Units or fifteen (15) wolf packs in the entire state (the “seven or fifteen criteria”). 71 FecLReg. 43,410, 43,428-43,430 (Aug. 1, 2006). If wolves failed to satisfy the seven or fifteen criteria, they were reclassified as trophy game animals. Id. The Fish & Wildlife Service rejected this plan because, under the seven or fifteen criteria, the more protective trophy game animal designation was not necessarily activated when the population of wolves in the state dropped below 15 breeding pairs. Id. The Service gave the example of a situation where there were 3 packs inside the National Park Units and 10 packs outside the National Park Units. 5 Id. Although the seven or fifteen criteria would be met in this example — thus allowing for predatory treatment of wolves — there would be less than 15 packs in Wyoming. Id. The Service found Wyoming’s reliance on the seven or fifteen criteria unacceptable because it was unrealistic to expect the National Park Units to consistently maintain 8 breeding pairs. Id. The Service thus concluded, ‘Wyoming state law must clearly commit to managing for at least 15 wolf packs in Wyoming.” First Harbine Deck Ex. 11.

The Service’s contention that the 2007 plan corrects this deficiency by clearly committing the State of Wyoming to manage for 15 breeding pairs is once again disingenuous. The Service looks to the 2007 plan in support of its assertion. The 2007 plan states, “According to [Wyo. Stat. § 23-1-304] and interpretation of said statute by the Wyoming Attorney Generals Office, upon delisting, Wyoming will maintain a minimum of 15 breeding pairs within the State including [the National Park Units]. Seven of the 15 breeding pairs will be maintained in northwestern Wyoming but outside [the National Park Units].” Gustafson Decl. Ex. 21, at 10. Although this statement seems to limply support the Service’s position, the next sentence of the plan, and Wyo. Stat. § 23-1-304 — which was enacted to implement the plan under Wyoming state law — tell a different story. When the entire statutory and regulatory scheme in Wyoming is considered, the provisions demonstrate there is nothing clear about Wyoming’s commitment to maintaining fifteen (15) breeding pairs in its geographical area.

The 2007 Wyoming plan states “[s]ince the [Wyoming Fish & Wildlife Commission] does not have the legal authority to actively manage wolves within the National Parks, its management emphasis will be applied to maintaining seven (7) breeding pairs that inhabit primarily areas outside the Parks.” Gustafson Deck Ex. 21, at 10. Further, Wyo. Stat. § 23-l-304(a) allows the Wyoming Fish & Wildlife Commission to set seasons and bag limits annually in areas where the wolf is classified as trophy game “only as necessary to reasonably ensure at least seven (7) breeding pairs of gray wolves are located in [the] state and primarily outside of [the National Park Units].” Subsection (j) of the statute provides the State with the authority to take any action necessary to protect big and trophy game populations in the state from predation by wolves, if the number of breeding pairs exceeds seven (7) outside the National Park Units. Wyo. Stat. § 23 — 1—304(j); see also id. § 23-1-304(n). What these state laws show is that Wyoming is not committed to maintaining fifteen (15) breeding pairs of wolves within the state; rather, Wyoming intends to rely on the National Park Units’ ability to maintain eight (8) breeding pairs of wolves to satisfy Wyoming’s obligation to preserve fifteen (15) breeding pairs as its share of the required wolf population. It was precisely this reliance that was previously rejected by the Service as unrealistic. 71 Fed.Reg. at 43,429. Because the Fish & Wildlife Service rejected Wyoming’s 2003 plan because it failed to clearly commit Wyoming to managing for fifteen (15) breeding pairs, the Service’s acceptance of the 2007 plan despite the same deficiency was probably arbitrary and capricious.

The Fish & Wildlife Service’s approval of Wyoming’s 2007 plan despite its classification of the wolf as predatory throughout much of the state is also problematic and once again represents an agency change of course unsupported by adequate reasoning. In rejecting Wyoming’s 2003 plan, the Service indicated the wolf should be designated as trophy game statewide. Gustafson Decl. Ex. 11. The Service stated:

[t]he designation of wolves as ‘trophy game’ statewide would allow Wyoming to devise a management strategy that provides for self-sustaining populations above recovery goals, regulated harvest and adequate monitoring of that harvest. As is the case with other trophy game in Wyoming, the state could establish management areas, season dates, and quota limits to control populations in a regulated manner. In addition, Wyoming could address wolf depredation concerns through regulations that exist for currently classified trophy game animals.

Gustafson Decl. Ex. 11. Now, in approving Wyoming’s 2007 plan, the Service has compromised its earlier thinking and accepted less than statewide trophy game designation for the wolf. The Service seeks to justify this additional change of course by arguing the area designated as predatory in the 2007 plan is mostly unsuitable wolf habitat and Wyoming could maintain its share of the wolf population in the trophy game area alone. 73 Fed.Reg. at 10,549-10,550. This reasoning is an unexplained surrender of the agency’s rational rejection of the 2003 plan. The Fish & Wildlife Service presumably knew of the areas of suitable wolf habitat in Wyoming and the carrying capacity of those areas in 2003. In spite of this knowledge, the Service rejected Wyoming’s 2003 plan in favor of statewide trophy game designation. The Service has failed to provide any rationale for accepting something less now. Left unexplained, the capitulation is arbitrary and capricious.

The Service’s approval of the 2007 plan despite the malleable nature of the trophy game area is even more problematic. This aspect of the Wyoming plan presents a metaphorical moving target. In its briefing, the Fish & Wildlife Service cited to a map in the 2007 plan to assure the Court that the trophy game area is fixed. At oral argument, the Service clarified that the trophy game area is “fixed, but not permanent.” Clever wording aside, it is clear that Wyoming law permits the Wyoming Fish & Wildlife Commission to alter the parameters of the trophy game area. Specifically, the Wyoming Fish & Wildlife Commission may diminish the trophy game area if it “determines the diminution does not impede the delisting of gray wolves and will facilitate Wyoming’s management of wolves.” Wyo. Stat. § 23-1-101(a)(xii)(B)(I). The Service here too decides without explanation.

The State of Wyoming takes the position that the Court should not consider speculative future actions. The authority the State cites for this proposition, however, is not persuasive here. See Arizona Cattle Growers’ Ass’n v. U.S. Fish & Wildlife, 273 F.3d 1229, 1244 (9th Cir.2001) (addressing speculation regarding whether an endangered species actually occupies a particular area in addressing a Section 7 ESA claim). States have flexibility to manage unlisted species within their borders so long as that management does not threaten the continued viability of the species. Here, however, Wyoming submitted a plan that classifies the wolf as a predatory animal in approximately 90 percent of the state. The Service acknowledged wolves in this area will not likely persist because wolves cannot survive unregulated human-caused mortality. 73 Fed.Reg. at 10,549-10,550. The remaining 10 percent of Wyoming, which provides some protection for the wolf, can be altered by the Wyoming Fish & Wildlife Commission at any time. These facts, when coupled with Wyoming’s failure to firmly commit to managing to preserve at least 15 breeding pairs in the state, show the continued existence of the wolf in Wyoming and outside of the National Park Units is in serious jeopardy. The Service’s approval of the 2007 plan despite these deficiencies at this stage appears to be arbitrary and capricious.

2. State Depredation Control Laws

Plaintiffs next argue the Fish & Wildlife Service failed to consider an important aspect of delisting wolves because the Service did not assess state laws in Idaho, Montana, and Wyoming that authorize the unregulated killing of wolves in defense of property. Plaintiffs claim the unregulated killing exacerbates a threat to the wolves’ viability.

The Fish & Wildlife Service responds to the argument by observing that the states’ depredation control laws are similar to laws that authorized the removal of problem wolves under federal management. The Service also notes lethal control of wolves in the states will be stopped if recovery is ever at risk because Idaho, Montana, and Wyoming have each committed to maintaining at least 150 wolves and at least 15 breeding pairs.

Before delisting, the experimental wolf populations in the Greater Yellowstone and central Idaho core recovery areas were managed under § 10(j) of the ESA. The § 10(j) regulations allowed individuals to kill wolves that were “in the act of attacking.” 50 C.F.R. § 17.84(n)(4)(iii)(A). “In the act of attacking” was defined as “actual biting, wounding, grasping, or killing of livestock or dogs, or chasing, molesting, or harassing by wolves that would indicate to a reasonable person that such [activities] are likely to occur at any moment.” 50 C.F.R. § 17.84(n)(3). The specificity of this regulation is not repeated in the state depredation laws.

Idaho law permits the killing of wolves without a permit when they are “molesting or attacking livestock or domestic animals.” Idaho Code § 36-1107(c). “Molesting” is defined as “the actions of a wolf that are annoying, disturbing or persecuting, especially with hostile intent or injurious effect, or chasing, driving, flushing, worrying, following after or on the trail of, or stalking or lying in wait for, livestock or domestic animals.” Id. Unlike the Idaho depredation control law, the § 10(j) regulations required a wolfs molesting or harassing to indicate to a reasonable person that the wolf was likely to attack. Idaho law only requires “annoying ... especially with hostile intent or injurious effect.” Although these two standards differ, Idaho law is sufficiently similar to the § 10(j) regulations to provide assurance that Idaho’s depredation control law will not likely threaten the continued existence of the wolf in Idaho. The killing of wolves pursuant to Idaho’s depredation control law must be reported to government officials within seventy-two hours. Id. The number of wolves killed will then become part of the mortality limit for each wolf zone in Idaho, such that any wolf killed under Idaho’s depredation control law will reduce the take allowed by hunters. Nadeau Deck ¶ 20. Once the total mortality limit for a zone or the state has been reached, all wolf hunting in Idaho will be closed. Nadeau Decl. ¶ 20. Because Idaho has committed to managing for at least 15 breeding pairs, its depredation control law is not likely to threaten the continued existence of the wolf in Idaho. Additionally, the Fish & Wildlife Service considered Idaho’s depredation control law in its de-listing decision and thus did not fail to address an important aspect of the problem. See 73 Fed.Reg. at 10,548.

Montana’s depredation control law is not likely to threaten the continued existence of the wolf for the same reasons. Montana law permits the killing of wolves who are attacking, killing, or threatening to kill a person or livestock or who are attacking or killing a domestic dog. Mont.Code Ann. § 87-3-130(1). The statute requires notice to Montana Fish & Game of any take within seventy-two hours. Id. The § 10(j) regulations only applied to experimental wolf populations. Consequently, they did not govern wolves in the northwestern Montana core recovery area prior to de-listing. Nevertheless, the Service adequately considered Montana’s depredation control law, see 73 Fed.Reg. at 10,548, and concluded it is not likely to threaten the continued existence of the wolf because the law is similar to the § 10(j) regulations and those regulations did not threaten the continued existence of the experimental wolf populations in central Idaho and the Greater Yellowstone core recovery areas. Additionally, Montana has committed to managing for at least fifteen (15) breeding wolf pairs in the state and has indicated licensed public hunting of wolves will not occur unless this minimum standard is satisfied. Sime Deck ¶ 35.

Wyoming’s depredation control law is more problematic. It permits the killing of wolves “doing damage to private property.” Wyo. Stat. § 23-3-115(a). Unlike the § 10(j) regulations, Wyoming law does not limit its reach to wolves that are attacking or harassing persons, livestock, or domestic animals. Instead, it reaches wolves that are, in someone’s subjective view, damaging property. Because Wyoming’s depredation control law is significantly more expansive than the § 10(j) regulations, it is unclear whether a viable wolf population can be sustained under the law. This uncertainty is particularly problematic in light of Wyoming’s failure to clearly commit to managing for fifteen (15) breeding wolf pairs within its borders. Moreover, although the Service noted Wyoming law permits killing wolves in defense of property in the trophy game area, it did not discuss whether Wyoming’s law was similar to the § 10(j) regulations. Thus, it appears the Service did not consider this aspect of the delisting issue. For these reasons, Plaintiffs have raised serious questions on their claim regarding Wyoming’s depredation control law.

C. Possibility of Irreparable Injury

Plaintiffs claim a preliminary injunction is necessary because delisting has and will continue to result in irreparable injury to individual wolves, wolf packs, the entire wolf population in the northern Rocky Mountains, and members of Plaintiff organizations. They show at least 37 wolves have been killed since delisting on March 28, 2008. Some of the killings are justified, others are not. The contention is that the killing of even a small number of wolves that have been unlawfully removed from the endangered species list is sufficient to demonstrate irreparable harm.

The Fish & Wildlife Service acknowledges some individual wolves will be killed under state management that would not have otherwise been killed absent de-listing. The Service nevertheless takes the position such loss does not constitute irreparable harm. Part of its position is based on the notion that lethal take of wolves in the Greater Yellowstone and central Idaho core recovery areas was authorized under the § 10(j) regulations. 73 Fed.Reg. at 10,545. Despite wolves being killed under these regulations, the Service contends the wolf population in the northern Rocky Mountains continued to grow at a rate of twenty-four (24) percent annually. Id. The Service concludes human-caused mortality could increase post-delisting to remove an additional 24 percent of the wolf population without decreasing the total population of wolves because holding a wolf population static requires a total take of 28 to 50 percent each year. Id.

The Intervenors generally argue Plaintiffs’ showing of irreparable injury cannot be based on the killing of individual wolves. Instead, they claim Plaintiffs must show irreparable harm to the species as a whole. This cannot be done, as the syllogism goes, because wolves are an incredibly fecund species. They note the northern Rocky Mountain wolf population has increased at a rate of 24 percent annually even while experiencing 26 percent mortality. Id. The Intervenors conclude that the killing of individual wolves is not likely to threaten the species during the pendency of this case.

Plaintiffs cite Humane Society of the United States v. Kempthorne, 481 F.Supp.2d 53 (D.D.C.2006), for the proposition that harm to a small number of animals is sufficient to demonstrate irreparable harm to an endangered or threatened species. Citing several cases in support of their position, the Intervenors argue Plaintiffs’ reliance on this case is unwarranted because it dealt with a species that was listed under the ESA and the wolf is no longer listed. Even if Plaintiffs must show a possibility of irreparable harm to the species, as opposed to individual wolves, to obtain a preliminary injunction, they have done so here.

The Service and the Intervenors argue irreparable harm is not possible because wolf mortality would have to increase 24 percent before the total wolf population will decrease. The Service and the Inter-venors once again focus on the numerical recovery criterion of 30 breeding pairs and 300 wolves, suggesting that as long as this numerical goal is met, wolves are not endangered. This ignores the underlying premise of the 1994 EIS. When the Service established its recovery criteria in 1994, it determined wolves were not likely to survive environmental variability and stochastic events based on numeric abundance alone. Rather, the Service concluded genetic exchange between subpopula-tions was necessary for long-term viability. See 1994 EIS, App. 9. Genetic exchange has not yet taken place. Moreover, dispersal of wolves between the Greater Yellowstone core recovery area and the northwestern Montana and central Idaho core recovery areas — a precondition to genetic exchange — is rare. Only four to twelve wolves have dispersed between the core recovery areas in the thirteen years since wolves were reintroduced. 73 Fed.Reg. 10,553. The reduction in the wolf population that will occur as a result of public wolf hunts 6 and state depredation control laws in Idaho, Montana, and Wyoming is more than likely to eliminate any chance for genetic exchange to occur between sub-populations. Genetic exchange that has not taken place between larger subpopula-tions under ESA protections is not likely to occur with fewer wolves under state management. Absent genetic exchange, the wolf will not likely be able to withstand future environmental variability and stochastic events. 1994 EIS, App. 9. Plaintiffs therefore have demonstrated a possibility of irreparable harm.

VI. Conclusion

For these reasons, IT IS HEREBY ORDERED that Plaintiffs’ Motion for Preliminary Injunction (dkt. # 2) is GRANTED. Endangered Species Act protections are hereby reinstated for the northern Rocky Mountain gray wolf pending final resolution of this matter on the merits.

1 . Plaintiffs interpret the third test as requiring only a showing of likely success on the merits. In National Wildlife Federation v. National Marine Fisheries Service, 422 F.3d 782, 795 (9th Cir.2005), however, the Ninth Circuit assessed both the likelihood of success and the possibility of irreparable harm. Thus, the third test requires a showing of possible irreparable harm. Nat'l Wildlife Fed’n v. Burlington N. R.R., 23 F.3d 1508, 1511 (9th Cir.1994) ("[T]hese cases do not stand for the proposition that courts no longer must look at the likelihood of future harm before deciding whether to grant an injunction under the ESA”). What the test prevents is the court's consideration of hardships to the Fish & Wildlife Service and the Intervenors in determining whether to issue a preliminary injunction.

2 . The Delisting Rule identified 100 years as the appropriate time frame for assessing genetic threats. 73 Fed.Reg. 10,531.

3 . The northwestern Montana and central Idaho core recovery areas are well connected to each other, and to wolf populations in Canada, through regular dispersals. These subpo-pulations have established genetic and demographic linkages.

4 . The Intervenors raise arguments similar to those asserted by the Service. They focus particularly on the deference owed to the Service and its decision.

5 . The Service's discussion of the 2003 plan spoke of packs because Wyoming had not yet committed to managing for breeding pairs. In the 2007 plan, Wyoming made such a commitment and Plaintiffs do not challenge this aspect of the plan.

6 . The Idaho Fish & Game Commission recently finalized its wolf hunting regulations for this fall. It established an annual mortality quota of 428 wolves. Third Harbine Decl. Ex. 1. This represents over half of the approximately 732 wolves living in Idaho prior to delisting. Jimenez Decl. ¶ 3. Montana and Wyoming have not yet finalized their hunting regulations. Both states, however, plan to permit public wolf hunts this fall. Bangs Decl. ¶ 15.