Alliance for the Wild Rockies v. Lyder

728 F. Supp. 2d 1126 (2010)

ALLIANCE FOR the WILD ROCKIES, Native Ecosystems Council, Center for Native Ecosystems, and Sierra Club, Plaintiffs,
v.
Jane LYDER, Assistant Deputy Secretary of the Department of Interior, Kenneth Salazar, Secretary of the Department of Interior, and U.S. Fish and Wildlife Service, Defendants.

No. CV 09-73-M-DWM.

United States District Court, D. Montana.

July 28, 2010.

Eric E. Huber, Sierra Club, Boulder, CO, Timothy M. Bechtold, Bechtold Law Firm, Rebecca Kay Smith, Missoula, MT, for Plaintiffs.

John H. Martin, U.S. Department of Justice, Denver, CO, Mark Steger Smith, Office of the U.S. Attorney, Billings, MT, for Defendants.

ORDER

DONALD W. MOLLOY, District Judge.

I. Introduction

On February 25, 2009, the United States Fish and Wildlife Service (the "Service") issued a revised designation of critical habitat for the contiguous United States distinct population segment of the Canada lynx (Lynx canadensis). The designation identified approximately 39,000 square miles of habitat in the States of Maine, Minnesota, Montana, Wyoming, Idaho, and Washington, The acres are designated as critical habitat for the recovery of the Canada lynx, Plaintiffs, four environmental organizations, seek declaratory and injunctive relief, challenging the designation in two broad ways. First, they argue the Service arbitrarily failed to designate occupied and unoccupied critical habitat as required by the Endangered Species Act ("ESA"), and second, they argue the Service failed to base its decision on the best scientific data available also as required by the ESA, For the reasons that follow, I find the challenge is well taken in part.

II. Background

A. The Canada Lynx

Lynx canadensis, the Canada lynx ("lynx"), is a medium-sized cat similar in size and appearance to a bobcat. 74 Fed. Reg. 8616, 8616 (Feb. 25, 2009), Unlike the bobcat, lynx have long legs and large paws making them well-adapted for hunting and surviving in areas that experience cold winters with "deep, fluffy snow." Id.

The lynx is a specialized predator of snowshoe hare, which comprises a majority of its diet. Id. at 8616-17. With its adaptations for snowy conditions and diet heavily based on snowshoe hares, lynx habitat consists of "moist boreal forests that have cold, snowy winters and a snowshoe hare prey base." Id. at 8616. The boreal forest landscape must be large enough to ensure adequate snowshoe hares are available. Id. The home range of an individual lynx varies based on the abundance of prey. As snowshoe hare numbers decline, lynx require a broader landscape to survive and reproduce. Id.

The contiguous United States is at the southern edge of the boreal forest range, resulting in limited and patchy forests that can support snowshoe hare and lynx populations. Canada on the other hand, with an expansive boreal forest and fewer snowshoe hare predators and competitors such as bobcats, has higher lynx densities and an overall greater lynx population than found in the contiguous United States. Id. at 8717.

B. The ESA and Critical Habitat

When enacting the ESA, Congress recognized the destruction of "natural habitat" to be the main threat to species. Tennessee Valley Authority v. Hill, 437 U.S. 153, 179, 98 S. Ct. 2279, 57 L. Ed. 2d 117 (1978). The principled conceptual basis of the ESA provides "a means whereby the ecosystems upon which endangered species and threatened species depend may be conserved." 16 U.S.C. § 1531(b). Such protections require a species first to be listed as endangered or threatened. 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).

When listing a species as endangered or threatened, "to the maximum extent prudent and determinable," the Service is required to "concurrently . . . designate any habitat of such species which is then considered to be critical habitat." Id. § 1533(a)(3). The ESA defines critical habitat as follows:

(i) the specific areas within the geographical area occupied by the species, at the time it is listed . . . on which are found those physical or biological features (I) essential to the conservation of the species and (II) which may require special management considerations or protection; and (ii) specific areas outside the geographical area occupied by the species at the time it is listed . . . upon a determination by the Secretary that such areas are essential for the conservation of the species.

Id. § 1532(5)(A). By Congressional mandate conservation means "to use and the use of all methods and procedures which are necessary to bring any endangered species or threatened species to the point at which the measures provided pursuant to this chapter are no longer necessary." Id. § 1532(3). Once designated, the Service "may, from time-to-time thereafter as appropriate, revise" its critical habitat designation. Id. § 1533(a)(3)(A)(ii).

Designating an area as critical habitat is important under § 7 of the ESA. Section 7(a)(2) of the statute requires all federal agencies to consult with the Service to ensure any actions they fund, authorize, or carry out will not "jeopardize" or "result in the destruction or adverse modification of" critical habitat. Id. § 1536(a)(2). The Service must issue a biological opinion on the effects any federal agency action will have on the critical habitat, and it must provide reasonable and prudent alternatives to avoid any agency action resulting in the adverse modification of critical habitat. 50 C.F.R. § 402.14; 16 U.S.C. § 1536(b)(3)(A).

C. The Critical Habitat Designation for the Canada Lynx

The path to the current designation of lynx critical habitat has proven to be a bumpy way. On March 24, 2000, the Service listed the entire contiguous United States distinct population segment of the lynx (the "DPS") as "threatened." 65 Fed. Reg. 16,052. When it did so it did not designate critical habitat for the lynx. This failure was challenged in court, and resulted in a district court order that the Service "undertake prompt rulemaking to designate [l]ynx critical habitat." Defenders of Wildlife v. Norton, 239 F. Supp. 2d 9, 26 (D.D.C.2002). The Service complied with the order and then published a final rule designating 1,841 square miles of critical habitat for the lynx on November 9, 2006. 71 Fed. Reg. 66,008. The critical habitat designation was short lived. In 2007, the Service withdrew the first designation because of concerns that it was not supported by the record or adequately explained. 74 Fed. Reg. 8618. The District Court for the District of Columbia then ordered the Service to submit a final rule designating lynx critical habitat no later than February 15, 2009. Defenders of Wildlife v. Kempthorne, nos. 00-2996, 04-1230 (D.D.C Jan. 15, 2008). The Service published the final revised critical habitat for the lynx (the "Final Rule") on February 25, 2009. 74 Fed. Reg. 8616.

The Final Rule designates approximately 39,000 square miles as critical habitat in Maine, Minnesota, Montana, Idaho, Washington, and Wyoming. Id. This designation consists solely of geographic areas occupied by lynx with the necessary physical and biological features essential to its conservation. The Service determined such areas were enough to ensure the conservation of the species, meaning no unoccupied areas had to be or were designated.

III. Standard of Review

A. Administrative Procedure Act

Judicial review of an agency's compliance with the ESA is governed by the APA. Oregon 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, 416, 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. Oregon 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, 42, 103 S. Ct. 2856, 77 L. Ed. 2d 443 (1983). When presented with a challenge like the current case, the 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, 361 F.3d 1108, 1119 (9th Cir.2004) (internal citations and quotations omitted). Nevertheless, there is no room for a court to substitute its judgment for that of the agency or merely determine it would have decided an issue differently. Oregon Natural Res. Council, 476 F.3d at 1035.

B. Summary Judgment Standard

Summary judgment is proper if "the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law." Fed. R.Civ.P. 56(c). Summary judgment is a particularly appropriate tool for resolving claims challenging agency action. See Occidental Eng'g Co. v. INS, 753 F.2d 766, 770 (9th Cir.1985). Summary judgment is appropriate here because the issues presented address the legality of Defendants' actions based on the administrative record and do not require resolution of factual disputes.

IV. Analysis

A. Occupied Habitat

Critical habitat for a listed species includes "the specific areas within the geographical area occupied by the species, at the time it is listed . . . on which are found those physical or biological features (I) essential to the conservation of the species and (II) which may require special management considerations or protection." 16 U.S.C. § 1532(5)(A) (emphasis added).

To identify that habitat for lynx, the Service conducted a two-step analysis. First, it identified the geographic area occupied by lynx when it was listed. 74 Fed. Reg. 8640. Then, it refined the occupied geographic area to include only portions containing the physical and biological features essential to the conservation of the lynx. Id.

Plaintiffs insist the Service's designation of lynx habitat does not comport with the statute because (1) the Service did not identify the "specific areas within the geographical area occupied by the species," and (2) the Service improperly identified the primary constituent elements and failed to consider areas with such features.^[1]

1. Definition of Occupied

Whether a species occupies an area under the ESA "is a highly contextual and fact-dependent inquiry." Arizona Cattle Growers' Ass'n v. Salazar, 606 F.3d 1160, 1164 (9th Cir.2010). Here, the Service "utilized data providing verified evidence of the occurrence of lynx and evidence of the presence of breeding lynx populations as represented by records of lynx reproduction" to make that determination. 74 Fed. Reg. 8640. Only data since 1995 was considered. Id. "Verified evidence" of lynx occurrence consists of a lynx observed by someone knowledgeable in lynx identification, genetic confirmation or location data from radio-collared lynx. Id. at 8641. Verified evidence of lynx reproduction included observation of lynx kittens or snow tracks for a family of lynx by someone knowledgeable in such matters. Id.

Plaintiffs interpret the fact that the Service "utilized data providing verified evidence of the occurrence of lynx and evidence. . . of breeding" to mean the Service required evidence of occurrence and breeding before an area was considered occupied by lynx. From this view they argue the Final Rule's definition of occupied area is inconsistent with the plain language of the statute. They also postulate this conjunctive standard is arbitrary and capricious because it contradicts previous standards used to identify areas occupied by lynx. See AR_L_4283^[2] (defining lynx habitat as "occupied" when "[t]here are at least 2 verified lynx observations or records since 1999 . . . or [t]here is evidence of lynx reproduction on the national forest").

Plaintiffs' argument on this issue is flawed. The Service "utilized data providing verified evidence of the occurrence of lynx and evidence of the presence of breeding" to determine the area occupied by lynx. 74 Fed. Reg. 8640. It did not require evidence of both occurrence and breeding for an area to qualify as occupied. The record shows it considered and relied upon data of lynx occurrences independent of evidence of lynx breeding. See AR_C_1441-73 (requesting information of reproducing populations or occurrences of the lynx); AR_B_2916-73 (detailing lynx distribution without reference to breeding). It then analyzed data of lynx occurrences to determine occupied areas independent of the data for evidence of reproduction. See, e.g., AR_D_4415 (describing the process to propose critical habitat by focusing on areas with verified records of lynx occurrences). Plaintiffs offer no examples of areas with evidence of lynx occurrence but not breeding that the Service did not deem occupied.^[3] The record and the Rule show the Service did not define "occupied" as Plaintiffs contend. Nor do Plaintiffs show the Service's utilizing evidence of lynx occurrence and lynx breeding is not "entitled to standard deference afforded such agency determinations." Arizona Cattle Growers' Ass'n, 606 F.3d at 1165. Plaintiffs' argument that the Service utilized an improper definition of occupied, or arbitrarily applied a different "occupancy" standard in designating lynx critical habitat is ill conceived.

The next claim is that the Service inexplicably required evidence of a self-sustaining population before an area in Colorado could qualify as "occupied." This argument is unsupported by the record. The Service found areas in Colorado occupied by lynx. 73 Fed. Reg. 10,860, 10,871 (Feb. 28, 2008) ("[T]he area occupied by the lynx. . . [includes] portions of Colorado."). It chose not to designate those Colorado areas as critical habitat because of uncertainty that a self-sustaining lynx population will establish there, not because it deemed the areas to be unoccupied. 74 Fed. Reg. 8620. The Service did not define "occupied" improperly.

2. Physical or Biological Features Essential to the Lynx

After identifying the geographic area occupied by the species, the ESA then requires the Service to determine whether "those physical or biological features (I) essential to the conservation of the species and (II) which may require special management consideration or protection" are "found" within that area. 16 U.S.C. § 1532(5)(A)(i). Regulations guide the identification of "those physical or biological" features:

When considering the designation of critical habitat, the Secretary shall focus on the principal biological or physical constituent elements within the defined area that are essential to the conservation of the species. Known primary constituent elements shall be listed with the critical habitat description. Primary constituent elements may include, but are not limited to, the following: roost sites, nesting grounds, spawning sites, feeding sites, seasonal wetland or dryland, water quality or quantity, host species or plant pollinator, geological formation, vegetation type, tide, and specific soil types.

50 C.F.R. § 424.12(b)(5) (emphasis added). The Service discusses the physical and biological features essential to the conservation of lynx in terms of "primary constituent elements," based on the regulatory guidance.

Here, the Service identified a single— albeit compound—primary constituent element for lynx: "boreal forest landscapes supporting a mosaic of differing successional forest stages" and containing snowshoe hares for prey, abundant range, woody debris piles for denning, and deep, fluffy snow for extended periods of time. 74 Fed. Reg. 8638. This primary constituent element's purpose is to locate "boreal forest landscapes of sufficient size to encompass the temporal and spatial changes in habitat and snowshoe hare populations to support interbreeding lynx populations or metapopulations over time." Id. at 8640. The presence of the primary constituent element was identified through review of verified lynx presence and through evidence of lynx reproduction, as well consideration of boreal forest characteristics, connectivity to lynx populations in Canada and other data such as snow depth and prevalence of bobcats. Id. at 8641. The Rule does not quantify how much boreal forest landscape or how many snowshoe hares are necessary to support an interbreeding lynx population. Instead, the Service used evidence of lynx breeding to identify areas with the primary constituent element because that evidence confirmed an area contained the physical and biological features essential to lynx in sufficient quantity and spatial arrangement. Id. at 8640. Lynx critical habitat only consists of areas with the necessary physical and biological features essential to lynx conservation, and that, in terms of critical habitat, only exists where lynx have reproduced.

The Service justifies its reliance on evidence of reproduction as follows:

[A]reas with [lynx presence and reproduction] represent resiliency during population lows, which is key to the species' survival. Areas that meet these criteria contrast with areas that may serve as temporary habitat for unsuccessful dispersers during population highs, but do not support lynx reproduction, and therefore are not likely to play a role in lynx conservation.

Id.

The conception of this primary constituent element was informed by the Service's Recovery Outline for the contiguous United States DPS of the Canada lynx. See 74 Fed. Reg. 8619. In 2005, the Service designed the Recovery Outline to serve as an "interim strategy to guide recovery efforts and inform the critical habitat designation process" for the lynx. AR_B_3670. The Outline established three categories of lynx habitat: core areas, secondary areas, and peripheral areas. AR_B_3673, Core area lynx habitat contained evidence of (1) past and present lynx populations and (2) recent reproduction, (3) boreal forest vegetation types of the quality and quantity to support lynx and snowshoe hare populations, and (4) fluffy, deep snow to allow lynx a competitive advantage. AR_B_3674. As for quality and quantity of boreal forest vegetation types, the Outline explained "the best available information suggests" at least .5 hares per hectare and "a minimum of 1250 square kilometers of boreal forest habitat as part of a larger landscape" are necessary to support a lynx population. Secondary areas are distinguished from core areas because they (1) have no documentation of reproduction and (2) the "quality and quantity of lynx habitat . . . is less clear." AR_B_3675. Peripheral areas are those with "questionable" habitat to support hare or lynx populations. AR_B_3676.

Plaintiffs challenge the Service's failure to designate critical habitat in certain national forests in Montana and Idaho, as well as its failure to designate any critical lynx habitat in Colorado. They insist the Service inappropriately failed to designate these areas and do so on two grounds. First, the Service arbitrarily and improperly excluded the areas as lacking the primary constituent element because there was no evidence of reproduction or a self-sustaining population, and second, the record shows such areas in fact contain the primary constituent element.

a. Montana and Idaho

Plaintiffs take exception to the Service's failure to designate the Beaverhead-Deerlodge, Bitterroot, Helena, and Lolo National Forests in Montana, and the Clearwater and Nez Pearce National Forests in Idaho, as lynx critical habitat.^[4] The Service informs that these national forests, to the extent they were not designated,^[5] are identified as secondary areas in the recovery outline. 74 Fed. Reg. 8642. As secondary areas, there was no "evidence of reproducing lynx populations" or "the ability to maintain and produce lynx during population lows" in these forests. Id. In the absence of reproductive evidence the Service concluded these areas only "provide habitat to dispersing lynx, especially when populations are at a cyclic high" and so the areas lack the physical and biological features to support lynx populations over time. Id.

The first challenge to this position is the claim that requiring evidence of lynx reproduction is arbitrary. The record demonstrates that "evidence of reproduction is difficult to obtain" and "many survey methods that detect lynx do not produce it." AR_L_4644; see also AR_B_2934 ("[R]eliable data on lynx reproduction in Montana are scarce."). The Recovery Outline acknowledges some areas may be secondary habitat in part because there are "no recent surveys to document the presence of lynx and/or reproduction." AR_B_3673. Plaintiffs take this to mean areas were excluded based on an absence of "survey effort rather than any biological or ecological principles." AR_L_4644. The second contest on this issue is the argument that the record shows these areas contain sufficient quantity and spatial arrangement of the primary constituent element to be designated as habitat. Plaintiffs suggest the argument is bolstered by the lynx Recovery Outline, which defines secondary areas as core areas but for an absence of evidence of lynx presence and/or reproduction. AR_B_3673.^[6]

In response, the Service does not dispute that it relied on a lack of evidence of reproduction to exclude these specific areas. Instead, it cites the Final Rule and its purported reasoning for relying on such evidence: evidence of reproduction distinguishes the areas that have the characteristics for "resiliency during population lows" from "areas that may serve as temporary habitat for unsuccessful dispersers during population highs, but do not support lynx reproduction." 74 Fed. Reg. 8640.

On this point Plaintiffs have the stronger argument. By relying on reproduction to identify areas that contain the primary constituent element, the Service arbitrarily excluded areas based on flawed logic. The exclusion is arbitrary because the Service set a requirement for critical habitat that it knew many areas could not possibly meet—not because the areas lacked the biological or physical elements but—because there simply was no data to meet the Service's requirement. See AR_B_3675 ("Some of the secondary areas have not been surveyed following any survey protocol"). In light of this fact, it is disingenuous for the Service to now argue it was appropriate for these forests to not be designated because there is no "persuasive proof that they contain the primary constituent element. Def.s' Opening Br. 32. There is no persuasive proof because the Service would only consider evidence of reproduction to suffice, yet many of the areas in question have not been surveyed to detect such proof.

There is recognition that evidence of breeding provides evidence that habitat contains the necessary elements for lynx conservation. The problem is the Service misapplies this conclusion. The Rule finds that "evidence of breeding populations is the best way to verify that the physical and biological features essential to lynx are present in sufficient quantity and spatial configuration to meet the needs of the species." 74 Fed. Reg. 8640 (emphasis added). This means evidence of reproduction is a sufficient condition that assures the existence of the primary constituent element. It does not, however, mean that only areas with such evidence contain the element. The Recovery Outline properly captures this distinction. The Recovery Outline acknowledges an absence of evidence of reproduction and lynx abundance "may" stem from the "quality and quantity of the habitat but does not necessitate such a conclusion. AR_B_3675. While it is rational to conclude areas with evidence of reproduction contain the primary constituent elements and should be designated as critical habitat, the Service could not flip that logic so it means critical habitat only exists where there is evidence of reproduction. Such a proposition alleviates the need to further consider the actual physical and biological features of the occupied area. The Service arbitrarily treated evidence of reproduction as a litmus test rather than as a relevant factor to consider if the challenged national forests in Montana and Idaho contain the primary constituent element.

The Service tries to justify its position by noting the best available science does not allow it to define the primary constituent element in a more "quantitative or direct fashion." Def.s' Reply 8. For example, a sufficient abundance of snowshoe hares is necessary for lynx survival in an area, but there are no broad-scale snowshoe hare density estimates the Service could rely upon to incorporate this element. 73 Fed. Reg. 10,866. While the science might not allow the Service to easily identify areas with the primary constituent element, this does not justify the Service using evidence of reproduction, (where there is also insufficient data), as a proxy for the primary constituent element. It is a fact to consider,^[7] not the cornerstone of designation.

Plaintiffs' second argument—that the Recovery Outline proves these national forests, as secondary areas, contain the primary constituent element—does not fare so well. The Recovery Outline does distinguish secondary areas from core areas because of the absence of evidence of lynx reproduction. AR_B_3673. However, the Recovery Outline also notes the "habitat in secondary areas may be patchier, drier, and/or more maritime resulting in snow or habitat conditions that are not favorable to lynx." AR_B_3675. The question thus becomes do the secondary areas, in fact, lack the requisite primary constituent element. This issue is unanswered by both the Recovery Outline and the Final Rule.

On this question the Service improperly and arbitrarily excluded areas occupied by lynx in Idaho and Montana when it considered the absence of reproductive proof but did not consider the actual physical and biological features of the areas. On remand, the Service must consider the physical and biological features of the occupied areas to determine whether they should be designated as critical habitat under the ESA.

b. Colorado

Colorado is a different story. It is undisputed that Colorado contains a breeding lynx population. Even so, the Service designated no critical habitat in Colorado. Its reasoning is predicated on the uncertainty that Colorado habitat can support a viable lynx population. The Service concluded Colorado does not possess a sufficient primary constituent element for designation.

Lynx are indigenous to Colorado. AR_B_3334. The Final Rule states this indigenous species was extirpated in Colorado at the time of listing. 74 Fed. Reg. 8641. This conclusion is predicated on a study conducted in 2000 that found 1974 to be the last verified lynx record in Colorado, "despite large-scale snow-tracking efforts." AR_B_2940. On the other hand, the Service co-authored with three other federal agencies the Lynx Conservation Assessment Plan, a plan that cites the same study from 2000. The second plan offered a different view. In it lynx were confirmed in Colorado in 1988 and 1989, a confirmation verified through tracks and corroborated with hair and scat samples. AR_B_3335. The Assessment Plan also noted that survey efforts since the 1974 lynx record "have not provided the systematic statewide coverage and intensity necessary to make conclusions about population persistence or numbers (large tracts of terrain have never been surveyed)," compelling the federal agencies to conclude—given "the level of effort"—that "lynx are apparently rare in" Colorado but not extirpated. Id.

The lynx population was very low at the time of listing and did not increase after the 1960s and 1970s "despite the removal of certain key suppressing factors, including commercial trapping and indiscriminate predator control." Id. By the 1990s Colorado biologists considered the extant lynx population too small to be self sustaining. Id. This led the State to initiate a program in 1999 to re-introduce lynx. As of 2007, the State had released 218 lynx, 98 of which have died. AR_B_4087. Nearly 30% of the deaths were the result of either vehicle collision or gunshot. Another 20% died due to starvation and disease. Id. At the same time, the released lynx gave birth to 116 kittens. Id.

The 2005 Recovery Outline listed Colorado as a "provisional core area" for lynx habitat.^[8] It deemed the area provisional because of the introduced population and the fact that "it is too early to determine whether a self-sustaining lynx population will result." AR_B_3673. Though not clearly stated, part of Colorado, otherwise qualifying as a core area, implicitly has a historical and current lynx population with recent evidence of reproduction, the necessary snow conditions, and boreal forest of the quality and quantity to support both lynx and snowshoe hare populations. AR_B_3674.

The Final Rule designates no areas in Colorado as critical habitat. It states "the marginal habitat in [Colorado] . . . [is] not essential to the conservation of lynx because [the areas] likely lack the quantity and spatial arrangement of [the primary constituent element]." 74 Fed. Reg. 8641. The conclusion is supported by the Service's explanation that it is too early to determine whether Colorado's introduced lynx will be a "self-sustaining population." Id. The viability of Colorado's lynx population is uncertain in light of (1) the limited historical presence of lynx in the area, (2) the introduced population's low reproduction rate, and (3) the inability for the population to be sustained by ingress from Canadian lynx due to distance. Id.

Plaintiffs insist the Service's failure to designate any critical habitat in Colorado is wrong because the record establishes the requisite primary constituent element is found there. A variety of sources are cited to show Colorado has the snow, forest type and prey necessary for lynx habitat. See, e.g., AR_B_3336-38. But they fail to cite evidence that these characteristics are meaningful in Colorado. The Service, on the other hand, argues the record shows Colorado does not possess the primary constituent element in sufficient quantity and quality. It cites the record to show it is uncertain whether Colorado can support a lynx population. AR_B_3673. But this reference does not link this uncertainty to the lack of physical and biological features of Colorado habitat. Thus, the record is unclear whether Colorado possesses the primary constituent element to qualify as critical habitat under the ESA.^[9]

Plaintiffs also challenge the Service's decision to exclude Colorado from the designation of critical habitat based on the rate of lynx reproduction and failure of existing lynx to demonstrate a self-sustaining population. The claim is the Rule arbitrarily excludes Colorado because it never defines what level of lynx reproduction would show when the requisite primary constituent element is present. The Service concedes it made no attempt to define what rate of reproduction would show a viable population. Instead, it argues it did not have to do so because the Colorado lynx population has yet to be deemed self sustaining. The government's position is not strong. The Rule makes clear that "evidence of breeding populations is the best way to verify that the [primary constituent element is] present in sufficient quantity and spatial configuration to meet the needs of the species, and qualify as critical habitat" Id. at 8626. For several years Colorado has had a lynx population with evidence of reproduction. AR_B_4087. The Rule does not explain how this is not a breeding population, nor does it establish what reproduction rate over a specific time would satisfy its murky metric. Whether this will become a viable population over time does not relieve the agency of its analytical responsibility. As with evidence of reproduction, evidence of a self-sustaining population indicates the necessary physical and biological features for viability are present, but its absence does not—on its own— mean the habitat features are missing.

Requiring Colorado to prove a self-sustaining population before the Service will conclude it has the requisite primary constituent element is a more stringent requirement than the ESA demands. A self-sustaining population means an area has the habitat features necessary for conservation. However, the absence of such a population only means there is something holding the population back, which may— but not necessarily—stem from the lack of the primary constituent element By way of example, human-caused mortalities or a lack of connectivity might be the problem, not a lack of snow conditions or snowshoe hares.

The purpose of critical habitat designation is to promote the recovery of the species. The Service is required to designate the geographical areas with the features necessary to promote that goal. By requiring proof that an area already hosts a recovered viable population before it can be designated, the Service created a metric more stringent than, and contrary to, what the ESA dictates.

Plaintiffs also argue the exclusion of Colorado is arbitrary considering the Service's decision to designate habitat in the Greater Yellowstone area. That habitat, which is similar to the areas in Colorado, consists of "naturally marginal lynx habitat with highly fragmented foraging habitat" 74 Fed. Reg. 8643-44. The Service takes the position that the Greater Yellowstone area's proximity to Canada allows for connectivity between the two populations, unlike the Colorado situation. Proximity to Canada suggests that the Greater Yellowstone area may need less of the primary constituent element necessary for lynx conservation. Because the Service fails to identify whether Colorado has the requisite element, relying instead on the lack of a viable lynx population, it is arbitrary to use proximity to another population as a basis to exclude Colorado as critical habitat lacking the necessary physical and biological features for inclusion in the designated area.^[10]

The problem with the Service's identification of occupied areas with the primary constituent element is that it has designated habitat that ensures the survival of lynx but not the conservation of the lynx. The ESA requires the Service to designate areas with the physical or biological features "essential to the conservation of the species." 16 U.S.C. § 1532(5)(A), Conservation means "to use and the use of all methods and procedures which are necessary to bring any endangered species or threatened species to the point at which the measures provided pursuant to this chapter are no longer necessary." Id. 1532(3). Critical habitat is thus defined and designated "in relation to areas necessary for the conservation of the species, not merely to ensure its survival." Arizona Cattle Growers' Ass'n, 606 F.3d at 1166 (emphasis in original); see also Gifford Pinchot Task Force v. U.S. Fish & Wildlife Serv., 378 F.3d 1059, 1070 (discussing the difference between conservation and survival). The Service excluded Colorado because it was an introduced population and the service was unsure whether it would become a "stable . . . population structure." 74 Fed. Reg. 8626. Designating habitat for a threatened species on such grounds limits critical habitat to areas of stable populations. This concept promotes the survival of the species. It does not comport with the statutory requirement to identify areas that promote the conservation of the species. Here, the Service failed to determine whether areas occupied by lynx in Colorado possess the physical or biological features essential to the conservation of the species. As such it did not comply with the law.

B. Unoccupied Habitat

Critical habitat also includes "specific areas outside the geographical area occupied by the species upon a determination by the Secretary that such areas are essential for the conservation of the species." 16 U.S.C. § 1532(5)(A)(ii). Compared to occupied areas, the ESA imposes "a more onerous procedure on the designation of unoccupied areas by requiring the Secretary to make a showing that unoccupied areas are essential for the conservation of the species." Arizona Cattle Growers' Ass'n, 606 F.3d at 1163; see also 50 C.F.R. § 424.12(e). Thus, unlike with occupied habitat, the unoccupied area itself must be essential to the species. Cape Hatteras Access Preservation Alliance v. U.S. Dep't of the Interior, 344 F. Supp. 2d 108, 119 (D.D.C.2004).

In this case, the Service determined that the critical habitat designated from the presently occupied areas of the species is sufficient for the conservation of the lynx. This determination is based on the large amount of high-quality habitat designated across a broad geographic area. 73 Fed. Reg. 10,871. The designation of such habitat theoretically allows the species to recover from periodic disturbances. Because the designation consists of 5 units spread across a broad geographic area there is a redundancy factor. This redundancy allows the species to withstand catastrophic events in light of the fact that all 5 units are unlikely to be subjected to such an event simultaneously. Id.

Plaintiffs challenge the Service's decision not to designate any unoccupied habitat on two principal grounds. They argue the Service wrongly excluded unoccupied linkage or travel corridors necessary for the conservation of the lynx. Then they argue the Service failed to consider the need for unoccupied habitat to account for future habitat loss due to climate change.

1. Travel Corridors

Suitable lynx habitat in the United States is naturally fragmented and disjunct. 65 Fed. Reg. 16,079. This habitat structure creates subpopulations of lynx that rely on connectivity and travel corridors to other suitable habitat (and subpopulations) to allow the species to expand and colonize new areas, provide population support to shrinking populations, and to effect dispersal and find a mate. AR_B_3341. With this in mind, the Service designated habitats at the landscape level that "encompass multiple home ranges" and "provide connectivity among patches of suitable habitat." 74 Fed. Reg. 8640. The Service explains its primary constituent element designation "provides habitat connectivity for travel within home ranges, and exploratory movements and dispersal within critical habitat units." Id. at 8627. No unoccupied habitat to protect travel corridors and promote connectivity outside of the five critical habitat units is designated. 74 Fed. Reg. 8626.

Plaintiffs contend travel corridors consist of "non-lynx habitat such as basins, valleys, [and] agricultural lands." AR_K_820. These lynx habitat, were not designated, but they are important for maintaining lynx populations over time. The Service did not protect a travel corridor between lynx in the Greater Yellowstone area with lynx in Colorado. The government responds by insisting its critical habitat designation provides the necessary connectivity for populations within the designated habitat, and it argues connectivity to marginal or sink populations is unnecessary to conserve the species. The Service also determined that lynx dispersal is not hampered by human-caused barriers, providing another reason not to protect unoccupied travel corridors. The Service has the better argument on these points.

The problem with Plaintiffs' position is that it focuses too much on establishing that linkages are important, an issue not in dispute, and not enough proof that the Rule fails to protect the necessary linkages. The science shows connectivity within a lynx's home range, AR_K_784, and with the core population found in Canada is key to maintaining a lynx population over the long term. AR_L_2753. The Final Rule addresses both forms of connectivity.

The Final Rule designates critical habitat that is a matrix habitat-areas that support snowshoes closely juxtaposed with areas that do not. This designation provides connectivity throughout a lynx's large home range. 74 Fed. Reg. 8638. As such, Plaintiffs' argument that unoccupied habitat is necessary to protect these linkages is not well taken.

As for connectivity with Canada, Units 1 through 4 all provide direct connectivity with Canada, see id. at 8662, so that designating unoccupied areas is unnecessary to ensure connectivity for those units. Unit 5, the Greater Yellowstone area, is a different story. Unoccupied habitat separates it from Canada. The record shows lynx are able to maintain connectivity without the presence of specifically set aside corridors so that connectivity is not an issue. There is "no evidence that human-caused factors [such as roads] have significantly reduced the ability of lynx to disperse or have resulted in the loss of genetic interchange." 65 Fed. Reg. 16,079. At the same time, lynx are capable of crossing large expanses of unforested terrain, such as valley floors. 68 Fed. Reg. 40,079. Accordingly, the Service's finding that unoccupied habitat need not be designated to ensure connectivity with Canada is reasonable.

Plaintiffs take issue with this conclusion by pointing to a portion of the Final Rule where the Service notes, in discussing when ESA section 7 consultations are needed, that increases in traffic could reduce connectivity within the critical habitat units. 74 Fed. Reg. 8644-45. Because traffic may negatively impact intra-habitat connectivity does not undermine the Service's conclusion that human-caused factors should not impact genetic exchange and migration across habitat units in a negative way.

Plaintiffs also object that the Rule does not ensure connectivity to Colorado or secondary habitat areas. They argue the Service must designate unoccupied habitat to protect such linkages to allow for colonization and expansion into new areas.^[11] There are two problems with this argument. To the extent Colorado lacks the physical and biological features essential for the conservation of lynx, it makes sense for the Service to find it unnecessary to protect corridors to the area; linkages to sink populations are not essential to the conservation of the species. The Service has determined the chances of lynx colonizing new, marginal habitat to be "minimal," and unoccupied areas do not materially contribute to the persistence of the lynx. 68 Fed. Reg. 40,077. Furthermore, as discussed above, the Service reasonably found that lynx can disperse across and connect between areas of unsuitable habitat. The Service did not act unreasonably in failing to designate corridors to secondary and unoccupied areas as Plaintiffs wish.

2. Climate Change

Next, Plaintiffs fault the Service for failing to designate unoccupied habitat to account for the future loss of occupied habitat due to climate change. This is essentially a repeat of Plaintiffs' argument that the Service failed to consider the best available science on climate change and lynx habitat As discussed in greater detail below, the Service considered the issue. The Service found the available science did not allow for climate predictions at the appropriate scale to enable it to designate unoccupied habitat. This determination is reasonable. The science referenced in the record discusses changes to lynx habitat occurring at the regional scale, see AR_B_0514, and being dependent upon "traits and environmental conditions at landscape and local scales." AR_B_0479. Because the science does not provide the specificity needed to identify the location of lynx habitat in the future, the Service did not act arbitrarily or unreasonably in not designating unoccupied lynx habitat to account for climate change.^[12]

C. Best Available Science

Plaintiffs argue the Service failed to base its critical habitat designation on various sources of data, both in and extraneous to the record. The Service must designate critical habitat "on the basis of the best scientific data available." 16 U.S.C. § 1533(b)(2). This requirement ensures the ESA is not "implemented haphazardly, on the basis of speculation or surmise." Bennett v. Spear, 520 U.S. 154, 176, 117 S. Ct. 1154, 137 L. Ed. 2d 281 (1994). The statutory requirement "merely prohibits the [Service] from disregarding available scientific evidence that is in some way better than the evidence [it] relies on." Kern County Farm Bureau v. Allen, 450 F.3d 1072, 1080 (9th Cir.2006) (quoting Sw. Ctr. for Biological Diversity v. Babbitt, 215 F.3d 58 (D.C.Cir.2000)). In other words, the Service "cannot ignore available [scientific data]." Conner v. Burford, 848 F.2d 1441, 1454 (9th Cir.1988).

1. Climate Change Data

Plaintiffs' principal argument here is the Service ignored data in the administrative record on the impact climate change will have on lynx habitat in the future. The information consists of scientific papers on climate change and lynx habitat. In one study, the Gonzalez report, scientists analyzed potential changes in snow cover and vegetation and found potential lynx habitat may shift northward and up mountain slopes due to climate change. AR_B_518. Based on its analysis, the report concluded "potential [lynx] habitat could decrease by up to two-thirds" in the contiguous United States by the year 2100. The paper identifies parts of the Teton-Bridger National Forest in Wyoming and Superior National Forest in Minnesota as potential refugia. AR_B_517, 521. In another report, the Knowles paper, scientists evaluated past trends of snow accumulation and snowmelt across western North America, and found warming will have a negative effect on both. Based on this analysis, the report concluded climate change will decrease the persistence and volume of snowpacks. AR_B_961. In a third study, the Danby paper, scientists looked at changes in tree lines—the boundary between tundra and forest zones—in northern Canada during the past century, and concluded such changes "vary significantly at regional, landscape and even local scales and are partially contingent on terrain variability, landscape setting and existing vegetation." AR_B_489.

Plaintiffs challenge the Final Rule because it is not based on this climate change science and because the Service improperly (a) put off considering the issue, and (b) it failed to utilize this data to designate critical habitat in response to climate change.

a. Did the Service Delay Consideration of Climate Change?

The ESA requires the Service to base its determination on the "best scientific data available." 16 U.S.C. 1533(b)(2) (emphasis added). The Rule states the reports on climate change need "to be evaluated further to determine how climate change might affect lynx and lynx habitat," 74 Fed. Reg. 7617. Plaintiffs surmise that because the Rule finds the data needs to be "evaluated further," the Service did not consider the data but instead improperly delayed designating critical habitat based on climate change. The argument is not well taken.

The Service did not "ignore" climate change data in designating lynx habitat. The proposed Rule discusses how increased temperatures will lead to decreased snow cover, resulting in a "likely shift upward in elevation and northward in latitude." 73 Fed. Reg. 10,867. The Final Rule also analyzes the future of lynx habitat and how climate change "may be an issue of concern for the future of conservation of lynx." 74 Fed. Reg. 8617. The Rule then goes on to note that the revised critical habitat designation includes "higher elevation habitats that lynx would be able to continue to use if lynx distribution or habitat shifted upward in elevation." Id. Further evaluation is necessary not because the Service put off analyzing the data, but rather because revisions "may be necessary in the future to accommodate shifts in the occupied range of the lynx" due to climate change. Id. As such, the Service did not delay considering the data.

b. Did the Service Fail to Base Its Determination on the Data?

Plaintiffs' real bone of contention here is that the Service did not designate additional habitat based on the data. While the Service considered the data, it found "reliable projections of future climate in lynx habitat in the contiguous United States" not available at this time. 74 Fed. Reg. 8621. Plaintiffs contend this conclusion is inaccurate by pointing to the Gonzalez report, which maps potential lynx habitat in the year 2100. AR_B_532. Plaintiffs insist the Service should have designated habitat based on the Gonzalez report "so that it will be available for occupancy in the future as habitat is lost and shifts due to climate change." Pl.s' Br. 11. The argument fails because it ignores that the Gonzalez report is useful for "regional planning." AR_B_0521. It is reasonable for the Service to consider the scale of this data as inappropriate to predict the location of future critical habitat.

Moreover, Plaintiffs' reliance on the Gonzalez report is misplaced. The Gonzalez report identifies no gain in potential lynx habitat in the contiguous United States due to climate change. AR_B_521, 532. Rather, the report identifies two pockets of potential refugia of lynx habitat in the contiguous United States—Teton-Bridger National Forest in Wyoming and Superior National Forest in Minnesota. Id. The Rule already designates those portions of Wyoming and Minnesota as critical habitat. See 74 Fed. Reg. 8642-44. Unlike the Gonzalez report, the Service did not designate critical habitat based solely on the presence of snow and forest type. The Service also considered factors like the density of snowshoe hare populations and the scale of the landscape. Id. at 8638, 8640. Plaintiffs' argument also ignores the science in the record that notes changes in the location of forests will depend on a variety of "species-specific traits and environmental conditions at . . . local scales." AR_B_0479, The Gonzalez report does not provide the specificity to identify lynx habitat—which requires the presence of boreal forests—at that scale.

Plaintiffs specifically argue that portions of Colorado should be designated as critical habitat since Colorado contains high elevation terrain that will maintain the necessary snow conditions despite climate change, and thus could serve as a refugium for lynx. The Service looked to factors beyond snow conditions in designating lynx critical habitat. Thus, it would be illogical for the Service to designate critical habitat solely on the basis of favorable future snow conditions if the area lacks the other necessary physical and biological features, Furthermore, the Gonzalez report, the very climate change data Plaintiffs rely upon, found suitable snow conditions to all but disappear in Colorado due to climate change, resulting in no noticeable potential lynx habitat there by the year 2100. AR_B_521, 532.

The data does not show potential lynx habitat emerging in the contiguous United States that would justify additional habitat designations. This leaves Plaintiffs' climate change argument as little more than an attempt to force the Service to designate backup habitat in the hope it will someday become useful to the lynx. The Service, however, "may not statutorily cast a net over tracts of land with the mere hope that they will" someday acquire the potential to be critical habitat. Cape Hatteras, 344 F.Supp.2d at 122. To do so would run afoul of the axiom that an agency's decision rationally relate to the facts in the record. Id. at 123.

The Service based its decision on the climate change data in the record, and it did not make its decision related to this data in an arbitrary or capricious manner.

2. The Southern Rockies Lynx Amendment

On February 28, 2008, the Service issued the proposed rule designating lynx critical habitat. The Final Rule was then published on February 25, 2009. 74 Fed. Reg. 8616. Between the issuing of the proposed rule and publication of the Final Rule, the Forest Service released several documents pertaining to lynx in the Southern Rockies. On April 25, 2008, the Forest Service issued its Southern Rockies Lynx Amendment Biological Assessment. (Dkt. # 19-6.) The Fish & Wildlife Service then issued its Biological Opinion to the Southern Rockies Lynx Amendment on July 28, 2008. (Dkt. # 19-2.) The Forest Service subsequently issued its Final Environmental Impact Statement ("FEIS") and Record of Decision ("ROD") for the Southern Rockies Lynx Management Direction on October 28, 2008. (Dkt. 19-4, 19-5.) None of these documents were included as part of the administrative record. Instead, the record included the Forest Service's draft EIS to the Southern Rockies Canada Lynx Amendment. AR_K_1226.

Plaintiffs argue the failure to include these documents in the record shows the Service failed to base its decision on the best scientific data. To bolster its position, Plaintiffs cite the FEIS where it states its chosen alternative "should maintain lynx habitat and connectivity well enough to maintain lynx population persistence long term." (Dkt. # 19-4 at 153.) In addition, they cite the ROD, which concludes "[b]ased on the best scientific information available" that the Forest Service's "management direction will provide habitat to support persistence of lynx in the Southern Rockies in the long term." (Dkt. # 19-5 at 33.) Plaintiffs also argue that these documents should be considered as extra-record evidence to determine whether the Service considered all relevant factors in deciding Colorado's lynx habitat is too marginal to be designated as critical habitat.

The Service counters that it did not ignore any best available science related to these documents because the documents Plaintiffs seek to introduce contain no data the Service did not consider, as well as the fact that the record contained the draft EIS. The Service is correct on this issue.

Generally the Court reviews an agency decision based on the administrative record in existence at the time of the decision. Lands Council v. Powell, 395 F.3d 1019, 1030 (9th Cir.2005). The Ninth Circuit identified four narrow exceptions under which a reviewing court may consider extra-record evidence:

(1) if admission is necessary to determine "whether the agency has considered all relevant factors and has explained its decision," (2) if "the agency has relied on documents not in the record," (3) "when supplementing the record is necessary to explain technical terms or complex subject matter," or (4) when plaintiffs make a showing of agency bad faith.

Id. (internal quotations omitted).

Plaintiffs try to introduce these documents into the record to show the Service ignored the best scientific data. That is not what the documents show, Plaintiffs have failed to identify any data in these documents that was not covered elsewhere in the record. Instead, they seek to introduce conclusions contrary to the one the Service reached.^[13] This is not a best available scientific data argument. These documents (dkt. 19-2, 9-4, 19-5 and 19-6) do not fit within any narrow exception for the Court to consider extra-record evidence and are thus stricken from the record.^[14]

3. Pre-1995 Data

Lastly, Plaintiffs argue the Service failed to base its decision on the best available science because it did not consider data of lynx occurrence and reproduction prior to 1995. The Service excluded evidence from prior than 1995 "to ensure that this critical habitat designation is based on the data that most closely represents the current status of lynx in the contiguous United States and the geographical area known to be occupied by the species at the time of listing." 74 Fed. Reg. 8640. This cutoff line was chosen because the Service "assumed that a lynx born in 1995 would have been alive in 2000 or 2003, when the final listing rule and the clarification of findings were published." Id.

Plaintiffs argue the best available science requirement means the Service must consider the available science, even if some of it is dated. They point out that the Service created this cutoff even though it knew some areas lacked more recent data.^[15] The Service responds that the ESA defines critical habitat as the "area occupied by the species, at the time it was listed." Id. § 1532(5)(A). The Service used records since 1995 because those lynx would possibly have been alive in 2000 or 2003, when the species was listed. The Service has the better argument.

In determining the "area occupied by the species, at the time it was listed," it is reasonable for the Service to establish a cutoff to identify what data is relevant to accomplish the task. Here, a cutoff of eight years based on the lifespan of a lynx is reasonable. This cutoff, to the extent it precludes outdated data as to where lynx occurred at the time of listing, does not run afoul of the Service's requirement to base its decision on the best available data. Nor is the Service's rationale for the cutoff so implausible that the Court should not defer to it.

V. Remedy

After finding the Final Rule unlawful, the question becomes what is the appropriate remedy. Plaintiffs seek an injunction that keeps the current Rule in place while it is remanded back to the agency.

Ordinarily an agency rule not promulgated in accordance with the law is invalid. Paulsen v. Daniels, 413 F.3d 999, 1008 (9th Cir.2005). "The effect of invalidating an agency rule is to reinstate the rule previously in force." Id. The Court, however, has discretion to shape an equitable remedy, and sometimes equity requires an invalid rule to stay in place while the agency revisits the issue. Western Oil and Gas Ass'n v. EPA, 633 F.2d 803, 813 (9th Cir. 1980). This appears to be such a case. To invalidate the Final Rule and reinstate the prior one would put in place a rule that designated only 1,841 squares miles of lynx critical habitat that the Service withdrew over concerns that the rule was not supported by the record or adequately explained. 74 Fed. Reg. 8618. To reinstate that rule would unnecessarily deprive lynx of critical habitat based on a rule the Service found indefensible. Thus, the Final Rule must stay in effect while the Service revisits the issue.

VI. Conclusion

For the foregoing reasons, IT IS HEREBY ORDERED that Plaintiffs' motion for summary judgment (dkt. # 17) is GRANTED IN PART and DENIED IN PART. The Motion is GRANTED as to Count II: the Service designated occupied critical habitat contrary to the ESA and APA. The Motion is DENIED in all other aspects.

IT IS FURTHER ORDERED that the Defendants' motion for summary judgment (dkt. # 29) is GRANTED IN PART and DENIED IN PART. It is GRANTED as to Counts I, III and IV of Plaintiffs' Complaint and DENTED in all other aspects.

IT IS FURTHER ORDERED that the existing Final Rule is remanded to the U.S. Fish & Wildlife Service for further action consistent with the reasoning provided herein. The current critical habitat designation shall remain in place until the Service issues a new final rule on lynx critical habitat, at which time the current, invalidated Final Rule (74 Fed. Reg. 8616) will be superceded.

NOTES

[1] Plaintiffs actually argue that the Service did not define "occupy" correctly and then it failed to designate areas with the proper definition. In doing so, the argument conflates the definition of "specific areas within the geographical area occupied by the species," which alone does not dictate the designation of critical habitat, with the identification of the physical and biological features "essential to the conservation of the species." The substance of the argument is dealt with but in the context of the statute.

[2] Citations to the Administrative Record refer to the Bates-stamped number in the lower right-hand corner of each page.

[3] In reply, Plaintiffs contend the Service failed to designate occupied areas as critical habitat because the areas lacked evidence of reproduction, and thus it is beyond dispute that the Service applied an inappropriate standard for occupied areas in the Final Rule. The proposition seems to conflate the Service's definition of "occupied" with the Service's designation of critical habitat. The latter is discussed below.

[4] Whether these forests contain the primary constituent element was identified by Plaintiffs subsequent to their opening Brief, In their opening, Plaintiffs argue the administrative record shows these Montana and Idaho national forests are in fact occupied habitat, yet the Service failed to designate these forests as critical habitat. The problem with this argument is occupancy is only one requirement for an area to be considered critical habitat. In response, the government acknowledges the record shows such forests to be occupied, but responds the areas were not designated as critical habitat because they lacked the primary constituent element.

[5] Plaintiffs' challenge is too broad as portions of the Lolo and Helena National Forests are designated as critical habitat. See 74 Fed. Reg. 8687. Thus, it is presumed Plaintiffs mean to challenge the failure to designate the entirety of these national forests.

[6] Plaintiffs also point to a Northern Rockies Final Environmental Impact Statement written by the U.S. Forest Service. While that Statement identifies hundreds of thousands of acres of "lynx habitat" in those forests, AR_K_607, Plaintiffs do not discuss how the Forest Service's definition of "lynx habitat" includes only areas that possess the necessary primary constituent element.

[7] Not only is the Service's over dependence on reproduction improper, it appears the Service also acted arbitrarily when relying on such evidence to designate critical habitat. At the hearing, Plaintiffs raised the issue that there is no evidence of lynx breeding to support the designation of critical habitat in the parts of Montana that were designated. Because this was raised for the first time at the hearing and Plaintiffs ultimately prevail on this claim, the Court reserves making a final determination on the issue at this time.

[8] The Recovery Outline listed the Southern Rockies—consisting of southern Wyoming and all of Colorado—as a provisional core area. The Final Rule discusses the Southern Rockies and Colorado interchangeably. Since the Plaintiffs challenge the failure to designate critical habitat in Colorado, this opinion refers to Colorado interchangeably with the Southern Rockies when not misleading to do so.

[9] Plaintiffs' also cite the Recovery Outline, which classifies Colorado as provisional core habitat "because it contains a reintroduced population." AR_B_3673. It is unclear if this means the Service classified Colorado habitat as such because it met the core area criteria but had a reintroduced population, or it did not meet the criteria but was considered provisionally core habitat in light of its reintroduced population.

[10] Plaintiffs also argue the Service applied the statutory standard for unoccupied habitat to exclude Colorado. This argument is based on a response to a comment where the Service stated "habitat in Colorado is not essential to the conservation of the species." 74 Fed. Reg. 8619. The sentence came after the Service described Colorado's habitat as marginal and not proven to support a breeding population, id., to say nothing of the Rule's discussion of criteria to identify critical habitat. See id. at 8641. Read as a whole, the Service did not exclude Colorado by applying the unoccupied standard.

[11] Plaintiffs also cite the Recovery Outline for the proposition that the Service previously recognized the importance of maintaining corridors between core habitat and secondary habitat, such as population sinks. The recovery outline, however, states such linkages should be available to "adjacent populations in Canada or secondary areas in the United States." AR_C_1684. The Service has done that.

[12] To the extent the science shows where future lynx habitat will occur, it shows remaining and new habitat will be found to the north and altitudinally above existing habitat. These areas, where not in Canada, are already protected. 73 Fed. Reg. 10,867.

[13] The documents and conclusions Plaintiffs seek to introduce are not relevant to the Service's designation of critical habitat as challenged here. The documents were created to incorporate management direction for lynx habitat to contribute to the species' conservation. (Dkt. # 19-4 at 7.) The conclusions Plaintiffs cite are not the result of a detailed scientific and technical analysis on whether Colorado's habitat possesses the biological and physical features to support such a population—the question the Service faced in the Final Rule—but instead whether the Forest Service's management direction will contribute to that goal. For example, the FEIS discusses how timber harvest, thinning and fire management can be harmonized with lynx denning and prey densities. It does not discuss the prevalence of denning or snowshoe habitat necessary for Colorado to sustain a lynx population. (Dkt. # 19-4 at 58.)

[14] Plaintiffs also argue the Service ignored various sources of other data in reaching its decision. It points to maps of lynx habitat and linkages, and occurrence data from the Montana Department of Fish, Wildlife and Parks and the Idaho Department of Fish and Game that were not a part of the record. Plaintiffs have failed to show how this information is not duplicative of data the Service considered or otherwise fills any gaps in the administrative record. Accordingly, the Court finds the extra-record evidence does not show the Service failed to consider the best available science and the documents (dkt. 19-7, 19-10 and 19-12) are stricken from the record.

[15] Plaintiffs also cite an Eleventh Circuit case where the EPA was found to have acted arbitrarily by failing to consider data older than 7.5 years when identifying water bodies that do not meet water quality standards under the Clean Water Act. Sierra Club, Inc. v. Leavitt 488 F.3d 904, 913 (11th Cir.2007). In that case, the statutory and regulatory language required the agency to "evaluate all existing and readily available" data. Id. at 913 (citing 40 C.F.R. § 130.7(b)(5)). There is no parallel language here.