FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

FLATHEAD-LOLO-BITTERROOT No. 23-3754

CITIZEN TASK

D.C. No.

FORCE; WILDEARTH

9:23-cv-00101-

GUARDIANS,

DWM

Plaintiffs - Appellees, v. OPINION

STATE OF MONTANA; LESLEY

ROBINSON; GREG GIANFORTE,

Defendants - Appellants.

Appeal from the United States District Court for the District of Montana Donald W. Molloy, District Judge, Presiding

Argued and Submitted January 12, 2024 Pasadena, California

Filed April 23, 2024

Before: Richard C. Tallman and Mark J. Bennett, Circuit Judges, and Robert S. Lasnik, District Judge. *

* The Honorable Robert S. Lasnik, United States District Judge for the Western District of Washington, sitting by designation. 2 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

Opinion by Judge Bennett; Partial Concurrence and Partial Dissent by Judge Tallman

SUMMARY **

Environmental Law / Preliminary Injunction

In an action brought under the Endangered Species Act, the panel affirmed in part and vacated in part the district court’s preliminary injunction limiting wolf trapping and snaring in certain parts of Montana to January 1, 2024, through February 15, 2024, and remanded. Plaintiffs alleged that Montana’s laws authorizing recreational wolf and coyote trapping and snaring, including regulations approved by the Montana Fish and Wildlife Commission, allowed the unlawful “take” of grizzly bears, a threatened species, in violation of § 9 of the Endangered Species Act. The district court granted plaintiffs’ motion for a preliminary injunction as to wolf trapping and snaring only. The panel held that the district court did not abuse its discretion by considering new arguments and new materials, submitted with plaintiffs’ reply brief in support of their motion for a preliminary injunction, because the record showed that defendants had an opportunity to respond to plaintiffs’ submissions.

** This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 3

The panel held that the district court applied the proper preliminary injunction standard by requiring plaintiffs to show only a serious question going to the merits instead of a likelihood of success on the merits. Because plaintiffs sought a preliminary injunction under the Endangered Species Act, the last two factors of the “serious questions” test—balance of hardships and the public interest—tipped sharply in favor of the protected species. Plaintiffs were required to demonstrate that they presented serious questions as to the merits of their claim and that absent an injunction, irreparable harm was likely. The panel held that the district court did not abuse its discretion in finding that there were serious questions going to the merits of plaintiffs’ claim that Montana’s 2023 recreational wolf trapping and snaring regulations would cause the unlawful take of grizzly bears in violation of § 9 of the Endangered Species Act. Fairly read, the record reflects a genuine scientific and factual debate over this question. Reviewing deferentially, the panel also affirmed the district court’s finding of a reasonably certain threat of imminent harm to grizzly bears had Montana’s wolf trapping and snaring season proceeded as planned. Addressing the scope of the preliminary injunction, the panel affirmed the temporal scope of the injunction, but held that the injunction was geographically overbroad, and remanded for the district court to expeditiously reconsider the geographic scope. The panel also held that the injunction was overbroad because it prevents the State of Montana from trapping and snaring wolves for research, vacated that part of the injunction, and remanded for the district court to make proper modifications to the scope of its order. 4 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

Concurring in part and dissenting in part, Judge Tallman would vacate the district court’s preliminary injunction in toto. He concurred in the majority’s decision to remand the district court’s preliminary injunction as geographically overbroad and not sufficiently clear in preventing any takings so as to accommodate the State of Montana’s necessary scientific research activities. He dissented from the majority’s conclusion that plaintiffs established a reasonably certain threat of imminent harm to the increasing grizzly bear population through scientifically driven wolf trapping regulations sufficient to warrant a preliminary injunction.

COUNSEL

Timothy M. Bechtold (argued), Bechtold Law Firm PLLC, Missoula, Montana, for Plaintiffs-Appellees. Sarah M. Clerget (argued), Assistant Attorney General; Alexander R. Scolavino, III, Attorney; Kevin Rechkoff, Agency Counsel; Montana Department of Fish and Wildlife Parks, Helena, Montana; for Defendants-Appellants. Matthew G. Monforton, Monforton Law Offices PLLC, Bozeman, Montana; Gary R. Leistico, Leistico & Esch PLLC, Clear Lake, Minnesota; for Amicus Curiae Montana Trappers Association, National Trappers Association, and Fur Takers of America, Inc..

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 5

OPINION

BENNETT, Circuit Judge:

In this case, we review the district court’s order granting a preliminary injunction. The order limited wolf trapping and snaring 1 in certain parts of Montana to January 1, 2024 through February 15, 2024—when, as the district court found, it is reasonably certain that almost all grizzly bears will be in dens. Thus, under the injunction, Montana cannot authorize any wolf trapping and snaring (in the specified areas) anytime outside that period. The Flathead-Lolo-Bitterroot Citizen Task Force, a nonprofit public interest organization, together with WildEarth Guardians (collectively, “Plaintiffs”) sued the State of Montana (“State”); the Chair of the Montana Fish and Wildlife Commission (“Commission”) 2 , Lesley Robinson; and Governor Greg Gianforte (collectively, “Defendants”). Plaintiffs allege that Montana’s laws authorizing recreational wolf and coyote trapping and snaring, including the regulations approved by the Commission, allow the unlawful taking of grizzly bears in violation of § 9 of the Endangered Species Act (“ESA”), 16

U.S.C. § 1538.

1 Snares are cable devices that are designed to noose around an animal’s neck or foot. 2 The Commission establishes the hunting, fishing, and trapping rules for Montana’s Department of Fish, Wildlife and Parks (“MFWP”). Mont. Code Ann. § 87-1-301(b). The MFWP’s stated mission is to “[s]teward the fish, wildlife, parks, and recreational resources for the public, now and into the future.” Montana Fish, Wildlife & Parks, About FWP, https://fwp.mt.gov/aboutfwp/at-a-glance. 6 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

The district court granted Plaintiffs’ motion for a preliminary injunction as to wolf trapping and snaring only.3 Defendants filed an expedited interlocutory appeal. We have jurisdiction under 28 U.S.C. § 1292(a)(1). Under our limited and deferential standard of review, we affirm the district court’s grant of injunctive relief. But the injunction is overbroad in two respects. It is geographically overbroad, and thus we remand for the district court to expeditiously reconsider the geographic scope. But to prevent harm to Plaintiffs, the current geographic scope remains in place until the district court reconsiders the geographic scope. The injunction is also overbroad as to wolf trapping and snaring related to the State’s research activities. We therefore vacate that part of the injunction and remand for the district court to make proper modifications to the scope of its order consistent with this opinion. I. A. In 1973, Congress passed the ESA, 16 U.S.C. §§ 1531– 1544. “[T]he ESA sets forth a comprehensive program to limit harm to endangered species within the United States.” Or. Nat. Res. Council v. Allen, 476 F.3d 1031, 1033 (9th Cir. 2007). “Section 9 of the ESA establishes a blanket prohibition on the taking of any member of a listed endangered species,” id. (footnote omitted) (citing 16 U.S.C. § 1538(a)(1)(B)), unless the “take” is authorized by the relevant federal agency, 16 U.S.C. § 1539. A “take” means

3 Plaintiffs also sought to enjoin Montana’s recreational coyote trapping and snaring laws. But the district court declined to enjoin such laws because Plaintiffs did not “identif[y] a remedy appropriately tailored to the specific harm caused by coyote trapping and snaring in Montana.” Because Plaintiffs do not challenge that denial, we do not address it.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 7

to “harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect, or to attempt to engage in any such conduct.” Id. § 1532(19). Species listed as either “endangered” or “threatened” are protected under the ESA. 4 See 50 C.F.R. § 17.31(a); Crow Indian Tribe v. United States, 965 F.3d 662, 671 (9th Cir. 2020). The ESA allows private parties to sue to enjoin ESA violations. 16 U.S.C. § 1540(g). In 1975, the grizzly bear was listed as “threatened” under the ESA, and it is still listed as “threatened.” See Amendment Listing the Grizzly Bear of the 48 Conterminous States as a Threatened Species, 40 Fed. Reg. 31734 (July 28, 1975); U.S. Fish & Wildlife Serv., Grizzly Bear, https://www.fws.gov/species/grizzly-bear-ursus- arctos-horribilis (last visited Jan. 14, 2024). In 1975, about 700 to 800 grizzlies were in the 48 contiguous states, living primarily in areas of Wyoming, Washington, Montana, and Idaho—the same areas where they are found today. U.S. Fish & Wildlife Serv., Grizzly Bear, supra. Most lived in Montana or Yellowstone National Park in areas now called the Northern Continental Divide Ecosystem and the Greater Yellowstone Ecosystem, respectively. These areas are still home to most of the grizzlies in the 48 contiguous states. Id. About 2,000 grizzlies currently live in these two areas and in the Cabinet Yaak Ecosystem, an area of northwestern Montana and northern Idaho. Id.

4 “The term ‘endangered species’ means any species which is in danger of extinction throughout all or a significant portion of its range . . . .” 16 U.S.C. § 1532(6). “The term ‘threatened species’ means 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). 8 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

Montana has allowed recreational wolf trapping since 2012. In 2020, the wolf trapping season opened on December 15 and closed on February 28, statewide. In 2021, the Montana Legislature required the Commission to also authorize snaring of wolves and extended the length of the wolf trapping and snaring season (“trapping season”) to the first Monday after Thanksgiving through March 15. Mont. Code Ann. § 87-1-304(8) (2021); id. § 87-1-901(1) (2021). But since 2021, in the “estimated occupied grizzly bear range” (“occupied grizzly range”), 5 also called the “occupied grizzly bear habitat,” the default start date “floats,” meaning that instead of a fixed date, the start date can be any time from the Monday after Thanksgiving until December 31. See Montana Fish, Wildlife & Parks, 2023 Furbearer Trapping and Hunting Regulations 15, https://fwp.mt.gov/binaries/content/assets/fwp/hunt/regulati ons/2023/2023-wolf-and-furbearer-final-for-web.pdf (“2023 Regulations”); Montana Fish, Wildlife & Parks, 2022 Furbearer Trapping and Hunting Regulations 15, https://fwp.mt.gov/binaries/content/assets/fwp/hunt/regulati ons/2022/wolf-and-furbearer-final-for-web.pdf (“2022 Regulations”); Montana Fish, Wildlife & Parks, 2021 Montana FWP Wolf Hunting and Trapping Regulations 6, https://fwp.mt.gov/binaries/content/assets/fwp/hunt/regulati ons/2021/2021-wolf-final-for-web.pdf. The floating start date is determined by the Commission based on its

5 Collaborating biologists, including from the MFWP, the U.S. Fish and Wildlife Service, and the U.S. Geological Survey Interagency Grizzly Bear Study Team, have defined this area as “an estimate of the roughly contiguous, minimum area within which grizzly bears have established residency or have demonstrated habitat use. It does not include occasional forays outside the estimated range or low-density peripheral areas and therefore does not represent the total known extent of occurrences.”

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 9

assessment of real-time, on the ground observations of grizzly bear activities. 2023 Regulations, supra, at 15. So essentially, in those areas where it is estimated that grizzlies reside or show habitat use, the default start date is December 31 unless the Commission announces an earlier start date. The Commission is also permitted under the regulations to adjust the trapping season end date if there is a “non-target capture” 6 of a grizzly bear. Id. at 2, 15. For the 2021 and 2022 trapping seasons, Montana opted to use easily identifiable landmarks, such as roads or creeks to delineate the boundaries of the occupied grizzly range. This adjustment resulted in a larger area than the actual occupied grizzly range, and so during the 2021 and 2022 seasons, Montana applied floating start dates to areas outside the actual occupied grizzly range. The floating start dates for 2021 varied between December 15 and 27, and for 2022, between December 12 and 24. In August 2023, the Commission adopted the 2023 Regulations. As in past years, it set the wolf trapping season from the first Monday after Thanksgiving through March 15, 2024, but with floating start dates for the occupied grizzly range. However, due to criticism that its prior use of landmarks to delineate the boundaries of the occupied grizzly range included areas with no grizzlies, Montana’s map of the occupied grizzly range for the 2023 trapping season used the actual occupied grizzly range boundaries. 7 For the 2023 season, Montana also used a new mapping

6 Montana’s regulations define this term as “[c]apture of any animal that cannot be lawfully trapped, including domestic animals.” 2023 Regulations, supra, at 2. 7 According to Defendants, this was made possible due to the availability and proliferation of specific GPS technologies. 10 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

method, which purportedly resulted in a more precise map of the occupied grizzly range. These changes resulted in a smaller occupied grizzly range than in prior years, and consequently, the early default start date—the first Monday after Thanksgiving—would apply to more areas than before. B. On September 11, 2023, Plaintiffs filed this suit in the United States District Court for the District of Montana. The operative amended complaint alleges that Montana’s recreational wolf and coyote trapping and snaring laws are causing and will cause the unlawful “take” of grizzly bears in violation of § 9 of the ESA. Plaintiffs seek a declaration that Montana’s laws violate the ESA and an injunction barring Defendants from violating the ESA. On September 22, 2023, Plaintiffs moved for a preliminary injunction, asking the district court to “enjoin the State of Montana from authorizing wolf trapping and snaring in occupied grizzly bear habitat in Montana to avoid irreparable harm until a ruling on the merits in this matter.” Plaintiffs’ brief in support of its motion referenced a bulletin issued by the MFWP, which stated that “Grizzly bears have the potential to be found anywhere in the western two-thirds of Montana (west of Billings).” Montana Fish, Wildlife & Parks, Hunters Must Expect to See Bears (Aug. 30, 2023, 8:41 AM), https://fwp.mt.gov/homepage/news/2023/aug/0830--- hunters-must-expect-to-see-bears. The bulletin warns the public about the presence of grizzlies and provides a list of precautions for hunters. Id. Defendants requested and received an extension to respond to the motion and eventually responded on October 23, 2023. On November 3, 2023, Plaintiffs submitted their

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 11

reply brief, in which they more specifically asked the district court to “enjoin the 2023 regulations and to require Montana to establish regulations that are reasonably certain not to lead to the take of grizzly bears” and “to limit wolf . . . trapping and snaring in all areas ‘west of Billings’ to the time period when it is reasonably certain that almost all grizzly bears will be in dens: January 1 to February 15.” Plaintiffs submitted six new declarations and four exhibits with their reply. They also submitted two additional declarations before the hearing on the motion. Defendants never objected to this additional evidence. At the November 20, 2023 hearing, all parties had the opportunity to offer testimony and arguments. Plaintiffs reiterated that, based on their evidence, grizzlies would most likely be in their dens from January 1 to February 15 and so the court should limit the trapping season to that period. During Plaintiffs’ opening argument, the court also asked about the evidence that grizzlies may be found anywhere west of Billings, signaling that it was considering whether the “west of Billings” geographical scope would be proper. 8 Defendants did not address this evidence. Nor did they address Plaintiffs’ argument that the grizzlies would most likely be in their dens from January 1 to February 15. On November 21, 2023, the district court granted the motion for a preliminary injunction as to the wolf trapping and snaring regulations. It found that Plaintiffs had raised serious questions going to the merits and had shown a reasonably certain threat of imminent harm to grizzly bears

8 The court asked Plaintiffs’ counsel: “What do you make of the announcement by the powers that be to hunters that anybody west of Billings has got a—there is a likelihood—not likelihood. There is a potential of encountering a grizzly bear?” 12 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

should the 2023 wolf trapping season proceed as planned. The district court’s order limited the wolf trapping season “in all areas included in wolf regions one through five, plus Hill, Blaine, and Phillips counties” to January 1 through February 15, “the time period when it is reasonably certain that almost all grizzly bears will be in dens.” Although Plaintiffs sought to enjoin only Montana’s recreational wolf trapping and snaring regulations, the district court’s order appears to apply more broadly because it enjoins all “authoriz[ed] wolf trapping and snaring” in the identified areas, including the State’s trapping and snaring for research purposes. Defendants filed this interlocutory appeal the same day the district court issued its order. They argue that the district court erred in granting the preliminary injunction because: (1) it improperly considered new arguments and new materials submitted with Plaintiffs’ reply brief; (2) it applied the wrong preliminary injunction standard; and (3) there is no reasonable certainty of irreparable harm. For the reasons discussed below, we reject these arguments and affirm the issuance of injunctive relief. But Defendants also argue that the injunction is temporally and geographically overbroad, and that it is also overbroad with respect to wolf trapping and snaring related to the State’s research activities. While we affirm the injunction’s temporal scope, we agree that it is overbroad geographically and with respect to the State’s research activities.

II.

Our review of a grant of a preliminary injunction is “limited and deferential.” Sw. Voter Registration Educ. Project v. Shelley, 344 F.3d 914, 918 (9th Cir. 2003) (en banc). “Appellate review of a decision to grant . . . a

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 13

preliminary injunction is restricted to determining whether the district court abused its discretion or based its decision on an erroneous legal standard or clearly erroneous findings of fact.” Nat’l Wildlife Fed’n v. Burlington N. R.R. Inc., 23 F.3d 1508, 1510 (9th Cir. 1994). A district court’s finding of a likelihood of future harm is subject to clear error review. Id. at 1512; see also Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 422 F.3d 782, 795 (9th Cir. 2005). A finding of fact is clearly erroneous “if it is implausible in light of the record, viewed in its entirety, or if the record contains no evidence to support it.” Nat’l Wildlife Fed’n, 422 F.3d at 794 (citation omitted). “Mere disagreement with the district court’s conclusions is not sufficient reason for us to reverse the district court’s decision regarding a preliminary injunction.” Id. at 793. We “review the scope of an injunction for abuse of discretion.” Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 886 F.3d 803, 823 (9th Cir. 2018). It is an abuse of discretion to issue an overly broad injunction. Id. We review for abuse of discretion whether it was proper for the district court to consider new arguments and new materials submitted with Plaintiffs’ reply brief. See Getz v. Boeing Co., 654 F.3d 852, 868 (9th Cir. 2011) (reviewing for abuse of discretion a district court’s decision to consider evidence submitted for the first time in a reply brief); El Pollo Loco, Inc. v. Hashim, 316 F.3d 1032, 1040–41 (9th Cir. 2003) (reviewing for abuse of discretion the district court’s consideration of an argument raised for the first time in a reply brief). 14 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

III.

A. We first address Defendants’ argument that the district court abused its discretion by considering new arguments and new materials submitted with Plaintiffs’ reply brief. Because the record shows that Defendants had an opportunity to respond to Plaintiffs’ submissions, we find no abuse of discretion. As mentioned above, for the first time in their reply brief, Plaintiffs asked the district court “to limit wolf . . . trapping and snaring in all areas ‘west of Billings’ to the time period when it is reasonably certain that almost all grizzly bears will be in dens: January 1 to February 15.” They also submitted six new declarations and four exhibits with their reply brief and submitted two additional declarations after they filed their reply brief. All were submitted at least five days before the hearing on the motion. A district court does not abuse its discretion in considering new arguments or evidence if the opposing party had an opportunity to respond. See El Pollo Loco, 316 F.3d at 1040–41 (holding that the district court did not abuse its discretion in considering an argument raised for the first time in a reply brief because the adverse party had an opportunity to rebut the argument at the hearing and “the district court listened to, considered, and rejected” the rebuttal argument); id. at 1041 (citing Provenz v. Miller, 102 F.3d 1478, 1483 (9th Cir. 1996), for the proposition that “a district court may consider new evidence presented in a reply brief if the district court gives the adverse party an opportunity to respond”).

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 15

All of Plaintiffs’ documents were submitted at least five days before the hearing, and nothing in the record suggests that Defendants were prevented from responding to the documents. 9 And Defendants had an opportunity to respond at the hearing, as the district court gave both parties the chance to offer testimony and arguments. Further, based on Plaintiffs’ opening argument and the court’s questions to Plaintiffs’ counsel, Defendants were on notice that the court was considering whether Plaintiffs’ requested remedy raised for the first time in its reply brief was proper. During the hearing, the court also referenced some of Plaintiffs’ new evidence, signaling that it was considering such evidence.10 Despite being put on notice that the court was considering Plaintiffs’ new arguments and evidence and having the opportunity to respond, Defendants failed to do so. Under these circumstances, the district court did not abuse its discretion in considering Plaintiffs’ new arguments and new materials. 11

9 Defendants also do not suggest that they were barred from responding to Plaintiffs’ new arguments or evidence before the hearing. 10 For example, Plaintiffs submitted as an exhibit to their reply brief a letter from the Director of the United States Department of the Interior to the Director of the MFWP. During the hearing, the court questioned Defendants’ counsel about the letter, and counsel raised no objection to its submission. 11 Defendants never objected to the submission of Plaintiffs’ new arguments or evidence. Although Plaintiffs do not argue waiver, we note that “[i]f a party does not object to or challenge the improper submission of new evidence before the district court, the party who fails to object has ‘waived any challenge on the admissibility of [the] evidence.’” Dutta v. State Farm Mut. Auto. Ins. Co., 895 F.3d 1166, 1172 (9th Cir. 2018) (quoting Getz v. Boeing Co., 654 F.3d 852, 868 (9th Cir. 2011)). 16 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

Defendants also argue that it was improper for the district court to consider a news article that “came out the day of oral argument” and “was never corroborated by any witness.” The article stated that an adult grizzly bear had been photographed on a ranch in October 2023 “along the Missouri and Judith rivers, the farthest east a bear has been seen in Montana in more than 100 years.” Brett French, Grizzly Bear Photographed in Upper Missouri River Breaks, Missoulian (Nov. 20, 2023), https://perma.cc/F2D7-LHGC. But it appears that during the hearing, Plaintiffs’ counsel referenced the very same facts, stating that recently “a grizzly was seen on a ranch at the junction of the Missouri and . . . Judith,” and that “[i]t’s been a long time since a bear has been out on the Judith.” Defendants made no contemporaneous objection to counsel’s statements. Further, the Federal Rules of Evidence do not strictly apply in the preliminary injunction context. Given that the “purpose of a preliminary injunction is merely to preserve the relative positions of the parties until a trial on the merits can be held,” and “given the haste that is often necessary if those positions are to be preserved, a preliminary injunction is customarily granted on the basis of procedures that are less formal and evidence that is less complete than in a trial on the merits.” Univ. of Tex. v. Camenisch, 451 U.S. 390, 395 (1981). Indeed, “[t]he trial court may give even inadmissible evidence some weight, when to do so serves the purpose of preventing irreparable harm before trial.” Flynt Distrib. Co. v. Harvey, 734 F.2d 1389, 1394 (9th Cir. 1984). Under these principles and based on the record, we see no error in the district court’s consideration of the news article.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 17

B. Defendants next argue that the district court applied the wrong preliminary injunction standard. They claim that the district court erred by requiring Plaintiffs to show only a serious question going to the merits instead of a likelihood of success on the merits. Our case law does not support Defendants’ position, as we have affirmed the viability of the “serious questions” test. See All. for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1134 (9th Cir. 2011) (“[T]he ‘serious questions’ version of the sliding scale test for preliminary injunctions remains viable after the Supreme Court’s decision in Winter [v. Natural Resources Defense Council, Inc., 555 U.S. 7 (2008)].”). 12 Under the Winter test, a party is entitled to a preliminary injunction if it demonstrates (1) “that [it] is likely to succeed

12 Our circuit along with the Second and Seventh Circuits have confirmed the viability of a sliding scale test post-Winter. See Cottrell, 632 F.3d at 1134 (“[W]e join the Seventh and the Second Circuits in concluding that the ‘serious questions’ version of the sliding scale test for preliminary injunctions remains viable after the Supreme Court’s decision in Winter.”). We note, however, that since our decision in Cottrell, the Tenth Circuit has joined the Fourth Circuit in rejecting such test as invalid under Winter. See Real Truth About Obama, Inc. v. Fed. Election Comm’n, 575 F.3d 342, 347 (4th Cir. 2009) (“Because of its differences with the Winter test, [our] balance-of-hardship test may no longer be applied in granting or denying preliminary injunctions in the Fourth Circuit, as the standard articulated in Winter governs the issuance of preliminary injunctions not only in the Fourth Circuit but in all federal courts.”), vacated on other grounds, 559 U.S. 1089 (2010), and reinstated in relevant part sub nom. The Real Truth About Obama, Inc. v. F.E.C., 607 F.3d 355 (4th Cir. 2010); Diné Citizens Against Ruining Our Env’t v. Jewell, 839 F.3d 1276, 1282 (10th Cir. 2016) (“Under Winter’s rationale, any modified test which relaxes one of the prongs for preliminary relief and thus deviates from the standard test is impermissible.”). 18 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

on the merits,” (2) “that [it] is likely to suffer irreparable harm in the absence of preliminary relief,” (3) “that the balance of equities tips in [its] favor,” and (4) “that an injunction is in the public interest.” Cottrell, 632 F.3d at 1131 (quoting Winter, 555 U.S. at 20). But our court has adopted the “serious questions” test—a “sliding scale” variant of the Winter test—under which a party is entitled to a preliminary injunction if it demonstrates (1) “serious questions going to the merits,” (2) “a likelihood of irreparable injury,” (3) “a balance of hardships that tips sharply towards the plaintiff,” and (4) “the injunction is in the public interest.” Id. at 1135. As to the first factor, the serious questions standard is “a lesser showing than likelihood of success on the merits.” All. for the Wild Rockies v. Pena, 865 F.3d 1211, 1217 (9th Cir. 2017). “In cases involving the ESA, Congress removed from the courts their traditional equitable discretion in injunction proceedings of balancing the parties’ competing interests.” Burlington N. R.R., 23 F.3d at 1511. So when applying the “serious questions” test to ESA claims, the last two factors— balance of hardships and the public interest—always “tip[] heavily in favor of protected species.” Id. In practice, then, only the first two factors—serious questions on the merits and likelihood of irreparable injury—are at issue when analyzing ESA claims under the “serious questions” test. See Cottrell, 632 F.3d at 1135. The district court applied the proper preliminary injunction standard. It correctly determined that our circuit employs the “serious questions” test; that because Plaintiffs sought an injunction under the ESA, the last two factors of the “serious questions” test tip sharply in Plaintiffs’ favor; and that, therefore, in order to issue an injunction, “Plaintiffs in this case must demonstrate they have presented serious

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 19

questions as to the merits of their ESA claim and that absent an injunction, irreparable harm is not only possible, but likely.” Defendants also argue that the district court should have used “the higher standard for issuance of [a] mandatory injunction.” See Hernandez v. Sessions, 872 F.3d 976, 997– 98 (9th Cir. 2017) (explaining that our circuit continues to apply a heightened standard to mandatory injunctions). “A mandatory injunction ‘orders a responsible party to take action.’” Marlyn Nutraceuticals, Inc. v. Mucos Pharma GmbH & Co., 571 F.3d 873, 879 (9th Cir. 2009) (quotation marks omitted) (quoting Meghrig v. KFC W., Inc., 516 U.S. 479, 484 (1996)). “A prohibitory injunction prohibits a party from taking action and ‘preserves the status quo pending a determination of the action on the merits.’” Id. at 878 (brackets omitted) (quoting Chalk v. U.S. Dist. Ct., 840 F.2d 701, 704 (9th Cir. 1988)). Defendants also argue that, even if the injunction is prohibitory rather than mandatory, the status quo would be a return to the 2022 Regulations, “not a complete cessation of all wolf trapping.” We are unpersuaded. The injunction prevents the State from authorizing any wolf trapping and snaring in certain areas outside January 1, 2024 to February 15, 2024. That is a prohibitory injunction because it prevents the State from enforcing its regulations that authorize wolf trapping. See, e.g., Bay Area Addiction Rsch. & Treatment, Inc. v. City of Antioch, 179 F.3d 725, 727–28, 732 n.13 (9th Cir. 1999) (holding that an injunction preventing enforcement of a local ordinance was prohibitory rather than mandatory); Hernandez, 872 F.3d at 998 (holding that an injunction that prohibits the government from engaging in likely unconstitutional conduct is a “classic form of prohibitory injunction”). 20 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

The injunction also maintains the status quo. The “status quo ante litem” for preliminary injunction purposes “refers not simply to any situation before the filing of a lawsuit, but instead to ‘the last uncontested status which preceded the pending controversy.’” GoTo.com, Inc. v. Walt Disney Co., 202 F.3d 1199, 1210 (9th Cir. 2000) (quoting Tanner Motor Livery, Ltd. v. Avis, Inc., 316 F.2d 804, 809 (9th Cir. 1963)). Plaintiffs filed suit in September 2023. Thus, the last uncontested status preceding the lawsuit was no authorized recreational wolf trapping or snaring, as the 2022 wolf trapping season had ended on March 15, 2023, and wolf trapping was not scheduled to start again until November 27, 2023. See 2022 Regulations, supra, at 15; 2023 Regulations, supra, at 15. For these reasons, the district court applied the proper preliminary injunction standard, and its injunction, which was prohibitory, preserved the status quo. We next address whether the district court abused its discretion in analyzing the two preliminary injunction factors at issue: serious questions on the merits and likelihood of irreparable injury. C. 1. Defendants do not appear to meaningfully challenge the district court’s conclusion that there are serious questions going to the merits of the ESA claim. Instead, their challenge seems to hinge entirely on their claim (that we rejected above) that the district court had to find that Plaintiffs were likely to succeed on the merits before it could issue a preliminary injunction. But generously construing Defendants’ argument as one challenging the district court’s serious questions determination, we reject it.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 21

“Serious questions” are ones “that ‘cannot be resolved one way or the other at the hearing on the injunction’ because they require ‘more deliberative investigation.’” Manrique v. Kolc, 65 F.4th 1037, 1041 (9th Cir. 2023) (quoting Republic of the Philippines v. Marcos, 862 F.2d 1355, 1362 (9th Cir. 1988) (en banc)). They “need not promise a certainty of success, nor even present a probability of success, but must involve a ‘fair chance of success on the merits.’” Marcos, 862 F.2d at 1362 (quoting Nat’l Wildlife Fed’n v. Coston, 773 F.2d 1513, 1517 (9th Cir. 1985)). The question is whether Montana’s 2023 recreational wolf trapping and snaring regulations will cause the unlawful “take” of grizzly bears in violation of § 9 of the ESA. Plaintiffs provided the district court with many declarations from qualified experts 13 as well as research materials 14 , which show that the regulations permit wolf trapping and snaring when a substantial number of grizzlies are active outside their dens, and that the grizzlies will be attracted to and get caught in wolf traps, resulting in

13 For example, Plaintiffs’ experts included a scientist and retired wildlife management professional with over forty years of experience in grizzly bear research and conservation. This expert has a B.S. in Forest Resource Management, an M.S. in Plant Ecology, and a Ph.D. in Wildlife Resource Management, has been consulted by grizzly bear researchers from around the world, and has authored more than 130 scientific articles and reports, many of which address the ecology and behavior of grizzly bears. Another retired wildlife biologist, with a B.A. in Environmental Sciences and an M.S. in Fish and Wildlife Management, has been involved in grizzly and black bear research since the 1970s and has authored or co-authored dozens of journal articles, scientific reports, and book chapters on grizzly bears. 14 Plaintiffs’ research materials included articles concluding that multiple grizzly bears in southeast British Columbia suffered toe amputations from wolf traps. 22 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

unlawful “take” under the ESA. Defendants also presented evidence from qualified experts as well as research materials relied on by those experts. Defendants’ evidence supports that the State’s mitigation measures—including the floating start date based on a comprehensive assessment of grizzly bear activity, use of trap devices that would allow grizzly bears to escape unharmed, and trapper education courses— ensure that the recreational wolf trapping regulations will not cause the unlawful “take” of grizzlies. Defendants’ position is also supported by the absence of any verified reports of a grizzly bear being caught in a public wolf trap during the wolf trapping season following implementation of the floating start date. Both sides presented evidence that unlawful grizzly bear killings are underreported. Fairly read, the record reflects a genuine scientific and factual debate over whether Montana’s 2023 recreational wolf trapping and snaring regulations will cause the unlawful “take” of grizzly bears. The issue therefore “cannot be resolved one way or the other” and requires “more deliberative investigation.” Manrique, 65 F.4th at 1041 (quoting Marcos, 862 F.2d at 1362). Given Plaintiffs’ substantial proffered evidence, which could be credited by a fact finder even in light of Defendants’ evidence, the district court did not abuse its discretion in determining that Plaintiffs have at least a “fair chance of success on the merits,” Coston, 773 F.2d at 1517, and thus raise a serious question on the merits. 2. Defendants challenge the district court’s finding of a reasonably certain threat of imminent harm to grizzly bears had Montana’s wolf trapping and snaring season proceeded as planned. As described above, our review of this finding

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 23

is limited and “very deferential.” Nat’l Wildlife Fed’n, 422 F.3d at 794; see also Sw. Voter Registration Educ. Project, 344 F.3d at 918. We may only reverse if the finding is clearly erroneous, meaning it must be “implausible in light of the record, viewed in its entirety,” or nothing in the record supports it. Nat’l Wildlife Fed’n, 422 F.3d at 794. There was no reversible error under this deferential standard. “A reasonably certain threat of imminent harm to a protected species is sufficient for issuance of an injunction under section 9 of the ESA.” Marbled Murrelet v. Babbitt, 83 F.3d 1060, 1066 (9th Cir. 1996), as amended on denial of reh’g (June 26, 1996). “[W]e do not require that future harm be shown with certainty before an injunction may issue, [but] we do require that a future injury be sufficiently likely.” Burlington N. R.R., 23 F.3d at 1512. This means “we require . . . a definitive threat of future harm to protected species, not mere speculation.” Id. at 1512 n.8. Plaintiffs’ evidence showed that grizzly bears have wide home ranges, are not limited to geographical borders, are active in areas frequented by wolves, are highly attracted to scented and baited wolf traps, and often get caught in those traps. For example, one of Plaintiffs’ experts stated in his declaration that the current “spatial and temporal overlap of trapping for . . . wolves in Montana with places and times that grizzly bears are also active results in widespread exposure of bears to risks posed by non-target injuries from snares and body-hold traps set to capture other species.” (emphasis added). Another expert stated in her declaration that “[s]cented and baited traps set for wol[ves] . . . are likely to attract grizzly bears from long distances,” and that “[i]n those situations, grizzly bears can be highly vulnerable to being caught in traps and suffer injuries to extremities or even be killed.” As the district court pointed out, 24 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

Defendants’ own evidence showed that, in Montana, grizzly bears have been caught in traps targeted at other animals— the MFWP has recorded twenty-one reports of such instances since 1988. 15 Plaintiffs also presented evidence that grizzlies have been seen in Montana with injuries— missing claws and toes and missing lower leg—likely caused by traps. Plaintiffs also showed that a large percentage of grizzly bears would be active outside their dens during Montana’s proposed wolf trapping season. One of their experts stated in his declaration that “[d]epending on the ecosystem, nearly 40% of grizzly bears in Montana have historically been active outside their dens either after November 27th or before March 15th.” 16 He added that this trend is likely to continue and increase due to increasing climate temperatures: “Grizzly bears in the Northern Rockies will almost certainly enter dens later and exit dens earlier as

15 Defendants argue that not all twenty-one reports are relevant for various reasons, including because some involved non-recreational traps. But the evidence is relevant, as it tends to show that grizzly bears are attracted to and get caught in traps generally, which supports the conclusion that they will get caught in recreational wolf traps. And this conclusion was laid out by Plaintiffs’ experts in their declarations. 16 Defendants criticize this evidence, arguing that it is based on data that is not entirely Montana-specific, that is outdated, and that does not account for the floating start date. But these criticisms go toward the weight of the evidence, which is left to the trial court’s discretion. See 11A Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 2949 (3d ed. Apr. 2023 update) (“Once received, the question of how much weight an affidavit will be given is left to the trial court’s discretion and the quality of the affidavit will have a significant effect on this determination.”).

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 25

annual temperatures continue to warm and vegetal foods become available earlier and later in the year.” As a whole, Plaintiffs’ evidence supported that a large percentage—nearly 40%—of grizzly bears in Montana would be active during the proposed wolf trapping season in the same areas as wolves and would be highly attracted to the wolf traps, which would likely result in grizzlies being caught in those traps. Given all the evidence, it was plausible for the district court to find a reasonably certain threat of imminent harm to grizzly bears should Montana’s wolf trapping and snaring season proceed as planned. In other words, the district court’s finding was not implausible given Plaintiffs’ evidence and the evidence as a whole. 17 Defendants mainly challenge the district court’s finding by identifying evidence that conflicts with or could undermine Plaintiffs’ evidence. For example, Defendants point to evidence that grizzly bears would not be active during the wolf trapping season, and that Montana’s

17 Respectfully, we disagree with our dissenting colleague. In arguing that Plaintiffs’ evidence of harm is too speculative, the dissent fails to take into consideration and give appropriate weight to the evidence that supports the district court’s determination. Instead, the dissent focuses on and gives more weight to the evidence that could undermine the district court’s determination. Dissent 34–37. Such an analysis conflicts with our standard of review. We also disagree with the dissent’s nexus analysis. Plaintiffs’ expert declared that “nearly 40% of grizzly bears in Montana have historically been active outside their dens either after November 27th or before March 15th.” Because Montana’s laws permit wolf trapping during the same time period when a large percentage of grizzlies will be active outside their dens, it was reasonable for the district court to find a sufficient nexus. We believe that the dissent concludes otherwise based on conflicting evidence. Dissent 42–43, 42 n.4. But again, the standard of review does not permit us to reweigh the evidence. 26 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

regulatory structure, including its mandatory trapper education program, significantly reduces the chance that a grizzly bear would be recreationally trapped during the wolf trapping season. But “[c]lear error is not demonstrated by pointing to conflicting evidence in the record.” United States v. Frank, 956 F.2d 872, 875 (9th Cir. 1991). Defendants also make much of the evidence reflecting the lack of any verified reports of grizzly bears being caught in recreational wolf traps in Montana since 2013. But the district court considered that evidence, and the court was not required to give it controlling weight because other evidence undercut its persuasiveness. As the district court pointed out, one of Plaintiffs’ experts declared that “only 12% of unpermitted grizzly bear killings are actually reported,” and that the “data shows that trappers who find grizzly bears in their traps are highly unlikely to call a government agent.” 18 Even Defendants’ evidence showed that more than 25% of grizzly bear killings go unreported. Given this evidence, the district court could appropriately infer that the verified reports significantly underrepresented the number of

18 Defendants argue that these statements are based on a study of “non- hunting human-caused mortalities” in “Flathead Valley, British Columbia” and thus these statements cannot support the underreporting of grizzlies caught by traps in Montana. See Bruce N. McLellan et al., Estimating Unrecorded Human-Caused Mortalities of Grizzly Bears in the Flathead Valley, British Columbia, Canada, PeerJ (2018), https://peerj.com/articles/5781/. We disagree. While the study did not purport to cover the exact geographical regions at issue here, it was reasonable for the district court to infer that it nevertheless supported significant underreporting generally, particularly in light of Defendants’ own evidence showing underreporting of grizzly bear killings in Montana.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 27

grizzlies caught in recreational wolf traps and thus the actual number was likely more. The outcome here is controlled by the standard of review. While we might have reached a different conclusion than the district court, that is not the test. “Mere disagreement with the district court’s conclusions is not sufficient reason for us to reverse the district court’s decision regarding a preliminary injunction.” Nat’l Wildlife Fed’n, 422 F.3d at 793. “[O]ur review in the preliminary injunction context is very deferential.” Id. at 794 (emphasis added). Because the district court’s finding of a reasonably certain threat of imminent harm was not implausible on the record, we must affirm that finding. 19 D. Finally, we address the scope of the injunction. Defendants appear to challenge the geographical and

19 We reject Defendants’ argument that this case is controlled by Burlington Northern Railroad in which we upheld the district court’s denial of a preliminary injunction. There, the district court found that there was no reasonable likelihood of future harm to grizzlies, and we held that such finding was supported by the evidence in that case. Burlington N. R.R., 23 F.3d at 1511–12. We did not, as Defendants suggest, hold as a matter of law that evidence of “no incidental takes of grizzly bears for three years br[eaks] the chain of likelihood” of harm. We also reject Defendants’ contention that the district court should have followed Center for Biological Diversity v. Little, 622 F. Supp. 3d 997 (D. Idaho 2022). The district court was not bound by Little. See Camreta v. Greene, 563 U.S. 692, 709 n.7 (2011) (“A decision of a federal district court judge is not binding precedent in either a different judicial district, the same judicial district, or even upon the same judge in a different case.” (quoting 18 J. Moore et al., Moore’s Federal Practice § 134.02[1][d] (3d ed. 2011))). And regardless, we hold that the district court’s finding of a reasonably certain threat of imminent harm was not clearly erroneous based on the record here. 28 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

temporal scope of the injunction. Although “[a] district court has considerable discretion in fashioning suitable relief and defining the terms of an injunction,” injunctive relief “must be tailored to remedy the specific harm alleged.” Lamb-Weston, Inc. v. McCain Foods, Ltd., 941 F.2d 970, 974 (9th Cir. 1991). “The scope of the [injunction] must be no broader and no narrower than necessary to redress the injury shown by the plaintiff[s].” California v. Azar, 911 F.3d 558, 584 (9th Cir. 2018). As to the injunction’s temporal scope, Defendants have failed to persuade us that the district court abused its discretion. Defendants attack the injunction’s time limitations based on the premise that the district court erred by considering the new arguments and new materials submitted with and after Plaintiffs’ reply brief. But we rejected that premise above. Further, evidence in the record supports that the time limitations were “no broader . . . than necessary to redress the injury shown by the plaintiff[s].” Id. After presenting information related to grizzlies’ denning habits, one of Plaintiffs’ experts declared that “[w]hile it is important to delay the start of the trapping and snaring season until at least January 1 to avoid catching grizzly bears, it is equally important to end the season by early February in low elevations and mid-February in higher elevations.” In support of his conclusion, this expert explained that “[g]rizzly bears in the Northern Rockies will almost certainly enter dens later and exit dens earlier as annual temperatures continue to warm and vegetal foods become available earlier and later in the year.” Defendants’ own default start date of December 31 in the occupied grizzly range also supports that it was reasonable for the district court to conclude that most grizzlies would be denned by

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 29

January 1. Defendants do not explain how all this evidence, if considered, could lead to the conclusion that the district court abused its discretion as to the injunction’s temporal scope. Thus, we affirm the temporal scope of the injunction, which limited wolf trapping and snaring to January 1, 2024 through February 15, 2024. We agree, however, with Defendants that the injunction is geographically overbroad. The district court enjoined wolf trapping and snaring “in all areas included in wolf regions one through five, plus Hill, Blaine, and Phillips counties.” 20 That comprises what appears to be more than half of the entire state of Montana and includes expansive areas outside the occupied grizzly range and even some areas east of Billings—areas that Plaintiffs did not even ask to be covered by the injunction. See 2023 Regulations, supra, at 14. On the current record, there is insufficient evidence to reasonably conclude that this expansive geographical scope is necessary to prevent the unlawful “take” of grizzly bears. The district court did not give an explanation supporting the geographic scope. But Plaintiffs identify what appear to be the strongest pieces of evidence in the record supporting the injunction’s geographic scope. First, the MFWP’s bulletin, which states (without any supporting information) that “[g]rizzly bears have the potential to be found anywhere in the western two-thirds of Montana (west of Billings), and their distribution is denser and more widespread than in previous years.” Montana Fish, Wildlife & Parks, Hunters Must Expect to See Bears, supra. Second, the news article that reported a single grizzly bear had been photographed “along the Missouri and Judith rivers, the farthest east a bear

20 For simplicity, we refer to this geographic scope as “all areas west of Billings” even though it also includes areas east of Billings. 30 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

has been seen in Montana in more than 100 years.” French, supra. The news article shows that one bear was seen in one area west of Billings, and that the occurrence was extremely rare—it hadn’t happened in the last 100 years. Thus, rather than support that grizzlies are likely to be found in all areas west of Billings, the article shows the opposite: grizzly bear presence is very rare in some areas west of Billings. The bulletin fares no better. It states that grizzlies could “potential[ly]” be found anywhere west of Billings. But “potential” means “existing in possibility” or only “capable of development into actuality.” Potential, Merriam- Webster, https://www.merriam- webster.com/dictionary/potential (last visited Feb. 1, 2024). The bulletin’s statement is thus couched in speculation and is too hypothetical to support the conclusion that grizzlies will likely be present in all areas west of Billings such that the injunction’s geographic scope is necessary to protect the grizzlies. On the current record, the injunction is geographically overbroad. But because Defendants have not presented an alternative geographic scope, we remand for the district court to reconsider the geographic scope and instruct that it do so expeditiously. See Baird v. Bonta, 81 F.4th 1036, 1047–48 (9th Cir. 2023) (instructing the district court to expeditiously decide the preliminary injunction motion on remand). To prevent harm to Plaintiffs, the injunction’s current geographic scope remains in place until the district court reconsiders the geographic scope. See Nat’l Org. for Reform of Marijuana L. (NORML) v. Mullen, 796 F.2d 276, 276 (9th Cir. 1986) (ordering that an injunction remain in place while the district court considers whether the injunction should be modified).

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 31

We also agree with Defendants that the injunction is overbroad for another reason: it prevents the State from trapping and snaring wolves for research, which usually occurs during the summer. Oral Arg. at 7:43–8:24. Plaintiffs’ counsel agreed that, even if we affirmed the injunction, Plaintiffs would have no objection to clarifying that the injunction does not apply to government research activities. Id. at 12:26–12:51. Given that concession, and because Plaintiffs’ suit does not challenge trapping or snaring related to government research, we agree that the injunction is also overbroad as to wolf trapping and snaring related to government research. We therefore vacate the injunction to the extent that it prevents the State from trapping and snaring wolves for research purposes and remand for the district court to modify the scope of the injunction. See Galvez v. Jaddou, 52 F.4th 821, 838 (9th Cir. 2022) (vacating an overbroad provision of an injunction and remanding for proper modification).

IV.

For the reasons stated above, we AFFIRM the district court’s issuance of injunctive relief. But because the injunction is geographically overbroad, we REMAND for the district court to reconsider the geographic scope and instruct that it do so expeditiously. The current geographic scope, however, remains in effect until the district court reconsiders the geographic scope. The injunction is also overbroad as to wolf trapping and snaring related to the State’s research activities. We therefore VACATE that part of the injunction and REMAND for the district court to make proper modifications to the scope of its order consistent with this opinion. The district court may conduct any further proceedings consistent with this opinion. The mandate shall issue forthwith. 32 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

AFFIRMED IN PART; VACATED AND

REMANDED IN PART. 21

TALLMAN, Circuit Judge, concurring in part and dissenting in part.

I concur in the majority’s decision to remand the district court’s injunction as geographically overbroad and not sufficiently clear in preventing any takings so as to accommodate the State of Montana’s necessary scientific research activities. However, I respectfully dissent from the majority’s conclusion that Plaintiffs established a reasonably certain threat of imminent harm to the increasing grizzly bear population through scientifically driven wolf trapping regulations sufficient to warrant a preliminary injunction. The district court erred in granting equitable relief for that reason as well. The purpose of a preliminary injunction is to preserve the status quo and prevent the “irreparable loss of rights” before a final judgment on the merits. See Sierra On-Line, Inc. v. Phoenix Software, Inc., 739 F.2d 1415, 1422 (9th Cir. 1984); Reno Air Racing Ass’n v. McCord, 452 F.3d 1126, 1130–31 (9th Cir. 2006). The remedy is considered extraordinary and should not be awarded as a matter of right but only “upon a clear showing that the plaintiff is entitled to such relief.” Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 22 (2008). In Winter we were reversed for not adhering to these black letter principles. For that reason, we need to tread cautiously

21 Each party shall bear its own costs on appeal.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 33

before blocking a state’s entire regulatory scheme for wildlife conservation. To justify a preliminary injunction, Plaintiffs had to not only establish a likelihood of success on the merits, but also that irreparable harm is likely, not just possible. Winter, 555 U.S. at 22. The evidence falls woefully short of meeting that standard. While I agree with the majority that Plaintiffs established a serious question on the merits, the evidence of record establishes that Plaintiffs failed to show a reasonably certain threat of imminent harm to grizzly bears should Montana’s wolf regulations remain in force. See Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 886 F.3d 803, 819 (9th Cir. 2018). The reason for this is twofold: first, the evidence of harm presented is too speculative to warrant a preliminary injunction under Winter; second, the evidence presented is not sufficiently tied to the challenged revised Montana regulations. I Plaintiffs who seek to enjoin a violation of the Endangered Species Act (“ESA”) must show a “definitive threat of future harm to protected species.” Nat’l Wildlife Fed’n, 886 F.3d at 818–19 (internal quotation marks omitted). My colleagues correctly recite that this harm cannot be based on “mere speculation.” Nat’l Wildlife Fed’n v. Burlington N. R.R., 23 F.3d 1508, 1512 n.8 (9th Cir. 1994); Ctr. for Biological Diversity v. Otter, No. 1:14-CV- 258-BLW, 2018 WL 539329, at *3 (D. Idaho Jan. 24, 2018). To merit the extraordinary award of a preliminary injunction, the supporting materials presented to the court “must be based on evidence not speculation.” Otter, 2018 WL 539329, at *3 (“There is always the danger that lynx captures are going unreported. But injunctions and 34 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

declaratory relief must be based on evidence not speculation.”). A “showing of a mere possibility of irreparable harm is not sufficient under Winter.” Earth Island Inst. v. Carlton, 626 F.3d 462, 468 (9th Cir. 2010); Nat’l Wildlife Fed’n, 886 F.3d at 819 (“A ‘possibility’ of irreparable harm cannot support an injunction.”). The record viewed in totality does not support a finding that irreparable harm is likely, other than just possible. Of the four exhibits, ten declarations, and nine affidavits submitted by the Plaintiffs, the bulk of the information provided is speculative in nature, offering theories about what could happen in the face of climate change or food scarcity, instead of offering any actual evidence that the harm is likely to occur. For example, Plaintiffs in the district court challenged the new regulations primarily on the grounds that the floating start date and increase of permitted wolf trapping will result in greater unreported grizzly bear bycatch. However, while the State of Montana offers actual data that no recreational wolf trapper has had accidental grizzly bear bycatch since 2013, Plaintiffs argue that “[t]he absence of evidence is not the evidence of absence;” “even if there’s no verified reports of grizzly bears stepping in [wolf] traps, it doesn’t lessen the likelihood that future captures are [going to] harm affected bears;” and that “conversations with various members of the trapper community” lead them to believe that “trappers will not report incidental capture.” D. Mont. Case No. 9:23-cv-00101-DWM, ECF No. 6-3, at 17 (hereinafter “D. Mont.”). This is speculation, not sufficient to overcome the data presented by the State that no bycatch is known to have occurred in more than a decade.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 35

Moreover, Plaintiffs in the district court challenged the new regulations reducing the grizzly bear “occupied range,” stating it could result in more takes as grizzly bears “may be present” outside of the calculated and monitored territory. But, while the State of Montana offers a specific map based on enhanced technology showing where grizzly bears are known to be from GPS tracking data and scientific monitoring, 1 Plaintiffs advance only varying, inconsistent, and speculative theories of where grizzlies may be. They include a purported territory map noting all possible bear sightings from the previous decade, and multiple declarations which state that with current weather patterns grizzly bears could be “anywhere west of Billings.” D. Mont. ECF No. 6-2, at 3; ECF No. 6-5, at 3. Neither of these theories are persuasive. In fact, my colleagues agree that the declaration that grizzly bears could be anywhere west of Billings is nothing but unsupported speculation, clarifying in the opinion that “[o]n the current record, there is insufficient evidence to reasonably conclude that this expansive geographical scope is necessary.” Plaintiffs also assert that the permitted 9-inch foothold wolf traps specifically harm grizzly bears as they are “large enough to capture grizzly bears and all furbearers” and “can cause toe fractures and toe amputations in grizzly bears,” speculating that “[s]maller bears may not be able to break a neck snare set for wolves.” But Defendants directly refute Plaintiffs’ contention that the changes will result in more takes. Defendants explain that actual data reflects no

1 Notably, every declaration and affidavit which challenges the effectiveness of the floating start date uses the same one-paragraph text, arguing that the sample size of GPS tagged and tracked bears is not big enough. D. Mont. ECF No. 21, at 2; ECF No. 22, at 2; ECF No. 24, at 3. 36 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

increase in bear mortality from July 1, 2021, to December 31, 2021, after the floating start date rule changes went into effect. Moreover, “[n]o known grizzly bears have been caught in a public, legally set wolf trap of any kind (including snare or foot traps) since 2013, and none since implementing the floating start date in the ‘estimated occupied range of grizzly bears’ in 2021.” Instead of providing actual data or concrete examples of harm, Plaintiffs submit declarations with “anecdotal eyewitness reports of grizzly bears with traps on their feet,” D. Mont. ECF No. 6-7, at 3, and an instance where the declarant “doe[sn’t] know with certainty what caused the injury to her leg, but it . . . could have been caught in a trap.” D. Mont. ECF No. 6-6, at 2. Even more surprising, Plaintiffs offered two exhibits to the district court that directly support the State’s current floating start date regulation as the cure to avoid trapping bycatch, stating that the solution is “[r]educing the overlap between the bear active season and trapping season,” and waiting until “bears were denning in the fall.” 2 D. Mont. ECF No. 27-3, at 8; ECF 27-4, at 8. Viewing the record in its entirety, the evidence presented strongly supports the State’s contention that there has not been any recorded grizzly bear bycatch during the recently implemented seasons under the revised regulations.

2 This peer reviewed article submitted by Plaintiffs as an exhibit “found that the median den entry date . . . was November 06, and the 95th quartile of den entry was November 22.” This supports the State’s floating start date which can only begin, at the earliest, on the Monday after Thanksgiving (November 27, 2023). This same article concludes that researchers “found no overlap between den emergence and the end of the trapping season.” D. Mont. ECF No. 27-3, at 8; ECF 27-4, at 8. See also D. Mont. ECF No. 6-3, at 22 (declaring that “[t]raps and snares should only be allowed during the non-denning season for grizzly bears”).

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 37

Plaintiffs’ speculative arguments, anecdotal statements, and inconsistent criticism of the wildlife regulations without tangible evidence falls well short of the bar to warrant a preliminary injunction preventing its implementation. Speculative harm is simply not enough to show a possibility that grizzly bears will be taken as a consequence of compliance with Montana’s revised laws and new regulations pertaining to the lawful trapping and snaring of wolves. See Friends of the Clearwater v. Higgins, 472 F. Supp. 3d 859, 876 (D. Idaho 2020), aff’d, Friends of Clearwater, 847 F. App’x at 394 (2021) (finding the plaintiffs’ claims of harm speculative when no bears or lynx were identified in the project area). It is true that courts have found a preliminary injunction appropriate based on pure scientific evidence when that evidence is “uncontroverted” in its showing of future harm. See Loggerhead Turtle v. Cnty. Council of Volusia Cnty., Fla., 896 F. Supp. 1170, 1180 (M.D. Fla. 1995) (finding the evidence uncontroverted that driving a vehicle over sea turtle hatching grounds would result in the take of endangered species eggs). Such evidence of a direct threat to the species is not the case here. Instead, Plaintiffs supporting documents look, taste, and smell of the “mere speculation” directly prohibited by Winter. Plaintiffs’ statements of harm are generalized allegations tied to their theory that snares and traps are inherently harmful to grizzly bears, who may or may not wander outside of their previous known territory, and who may or may not get stuck in traps specifically baited and regulated to avoid their take. My colleagues justify the lack of definitive evidence by declaring they are simply relying on our standard of review, invoking the general rule that we may only overturn the granting of a preliminary injunction if the district court 38 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

abused its discretion and committed clear error in weighing the evidence of future harm. But as the opinion acknowledges, a “finding of fact is clearly erroneous ‘if it is implausible in light of the record, viewed in its entirety, or if the record contains no evidence to support it.’” Nat’l Wildlife Fed’n v. Nat’l Marine Fisheries Serv., 422 F.3d 782, 794 (9th Cir. 2005) (quoting Serv. Employees Int’l Union v. Fair Political Practices Comm’n, 955 F.2d 1312, 1317 n.7 (9th Cir. 1992)); see also Or. Nat. Res. Council v. Marsh, 52 F.3d 1485, 1492 (9th Cir. 1995). With the greatest respect for the district judge here, the trial court’s finding is implausible in light of the record viewed in its entirety. While the State provides concrete proof of no past harm by recreational wolf trappers—as well as a plan to negate any future harm—Plaintiffs instead offer the court 19 declarations and affidavits that speculate and theorize that harm could occur. This does not meet their burden of proof. Based on this record, a “reasonably certain” threat of future harm is simply implausible. Federal judges are not wildlife biologists. And while our review may be “limited and deferential” to the district court, Earth Island Inst., 626 F.3d at 468, both appellate and district courts must “grant considerable discretion to agencies on matters ‘requir[ing] a high level of technical expertise.’” Ecology Ctr. v. Castaneda, 574 F.3d 652, 658– 59 (9th Cir. 2009) (quoting Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 377 (1989)); see also Selkirk Conservation Alliance v. Forsgren, 336 F.3d 944, 954 (9th Cir. 2003). We are to be “most deferential” when the agency is “making predictions, within its special expertise, at the frontiers of science.” Forest Guardians v. U.S. Forest Serv., 329 F.3d 1089, 1099 (9th Cir. 2003). As we emphasized in the landmark environmental case of Lands Council v. McNair—

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 39

en banc with all judges concurring—it is not our job to instruct the agency on “how to validate its hypotheses regarding wildlife viability” or order “the agency to explain every possible scientific uncertainty.” 537 F.3d 981, 988 (9th Cir. 2008) (en banc). Moreover, courts “are not free to ‘impose on the agency [our] own notion of which procedures are ‘best’ or most likely to further some vague, undefined public good.’” Churchill Cnty. v. Norton, 276 F.3d 1060, 1072 (9th Cir. 2001) (quoting Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 549 (1978)). Instead, “[t]o determine whether deference is warranted, we look to the sufficiency of the evidence.” Lands Council, 537 F.3d at 995. It is legal error to grant a preliminary injunction invalidating Montana’s updated wolf trapping regulations based on largely speculative evidence that is contradicted by the evidence in the administrative record compiled by the wildlife agency. That runs counter to the goals of Winter, and casts a blind eye to “the deference we owe to agencies and their methodological choices.” See Lands Council, 537 F.3d at 991. Simply put, courts can not, and should not, grant the extraordinary award of a preliminary injunction to plaintiffs who are crying wolf. II Second, the evidence presented here was not sufficiently related to the challenged regulation to warrant the issuance of a preliminary injunction. To support a preliminary injunction, the moving party must show that future ESA violations are likely if Montana’s existing laws and new regulations remain in place. See, e.g., Winter, 207 U.S. at 564 (finding the scientific theory that sonar could hurt sea mammals insufficient for an injunction); see Otter, 2018 WL 40 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

539329 at *3 (holding that a single violation of the ESA did not sufficiently “show that future ESA violations are likely in Idaho if the existing regulations remain in place”); Nat’l Wildlife Fed’n v. Burlington N. R.R., 23 F.3d 1508, 1511 (9th Cir. 1994) (finding past ESA violations instructive but not an indicator of future violations as the risks had been “substantially minimized” by changed policy). As the District of Idaho emphasized in a remarkably similar case, Ctr. for Biological Diversity v. Little, 622 F. Supp. 3d 997, 1009 (D. Idaho 2022), there must be a causal link between the evidence presented and the actual challenged regulation. In Little, the court found that evidence of two instances of wolf trapping accidentally resulting in grizzly bear bycatch was insufficient for an injunction, as they both happened outside of the scope of Idaho’s permitted hunting regulations—“Plaintiffs have no evidence that either bear was snared by a lawful wolf snare or trap set in compliance with Idaho’s regulatory scheme.” Id. 3 Proof of harm to bears elsewhere does not automatically guarantee protection for bears in Montana, as Canada and other states have different regulatory and monitoring

3 Here, the district court attempts to distinguish Little by stating it is “non-precedential,” and contained less record evidence than this case. Opinion 27. Not so. Plaintiffs in Little offered concrete evidence of two recreational wolf trappers taking grizzly bears at an unpermitted time, both outside the scope of the regulation. Plaintiffs here, as discussed above, offer evidence of only one accidental recreational bycatch, occurring over a decade ago under less protective regulations. Excluding speculative evidence of future harm, a direct comparison of the cases shows Little to be more persuasive than the record relied upon by the district court to support the relief it ordered here. See Little, 622 F. Supp. 3d at 1008–10.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 41

schemes in place. Theoretical or general fear of harm to grizzly bears is insufficient to impose an injunction on the State’s wolf trapping regulations when there is little to no proof that the State’s activities are causing that harm. Nat’l Wildlife Fed’n, 23 F.3d at 1511. The evidence here simply does not show Montana’s current wolf trapping regulations will result in grizzly bear take—just that grizzly bear takes have generally happened under different regulatory schemes. Plaintiffs argue, for example, that there have been twenty-one documented grizzly bear takes since 1988, and both the district court and my colleagues heavily rely on these past takes to show future harm. On superficial review past grizzly bear takings may “plausibly show that it is likely that additional takings may occur unless further regulations are implemented.” Ctr. for Biological Diversity v. Strommen, No. 20-CV-2554 (ECT/JFD), 2023 WL 2136650, at *2 (D. Minn. Feb. 21, 2023). However, the evidence presented here of the twenty-one bear injuries is categorically not related to the current Montana regulations. Out of the twenty-one listed grizzly bears from 1988 to present, two were unverified, and sixteen were trapped outside of wolf trapping season under old regulations. The majority of these documented catches were caught by researchers, not recreational trappers. Only a single bear on the provided list was caught by a recreational wolf trapper in 2013 during what is now wolf hunting season—a capture that predated Montana’s introduction of the “floating start date,” designed specifically to guarantee that incidental bycatch such as this take does not occur. The district court and my colleagues have applied this evidence to “show that grizzly bears are attracted to and get caught in traps generally.” Yet, if we look below the surface at this evidence through the required lens comparing Montana’s 42 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

past and current regulations, instead of taking a generalist approach, it specifically shows that the regulations are working to avoid accidental bycatch by recreational wolf trappers during wolf trapping season. Not a single known catch has occurred by recreational trappers after the floating start date took effect in 2021. This same analysis applies to Plaintiff’s assertion, relied on by the Majority, that “[n]early 40% of grizzly bears in Montana have historically been active outside their dens either after November 27th or before March 15th and this trend is likely to increase.” 4 As discussed above, the State’s regulation is designed to avoid bycatch by monitoring grizzly bear activity outside of their dens during the hunting and trapping dates. If—as Plaintiffs ask us to speculate— 40% of grizzly bears did not den during this season, the regulation as designed would not allow trapping to occur in occupied grizzly bear range; the floating start date would be adjusted to accommodate such facts. This assertion does not actually challenge the regulation in place, as the regulation is designed to accommodate this exact hypothetical. This absence of a nexus between the evidence and the challenged Montana regulation is consistent throughout Plaintiffs’ materials. One declaration before the district court, for example, discusses how “baited traps set for wolf, coyote, and furbearers are likely to attract grizzly bears.” D. Mont. ECF No. 6-2, at 3. But this is directly redressed by

4 As discussed above, Plaintiff’s own researchers “found no overlap between den emergence and the end of the trapping season.” D. Mont. ECF No. 27-3, at 8. I point this out not to “reweigh the evidence,” as my colleagues suggest, but show that the evidence presented by the Plaintiffs here is inherently speculative in nature and therefore inconsistent, failing to both meet their burden of proof and establish a clear nexus between their evidence and Montana’s regulations.

FLATHEAD-LOLO-BITTERROOT V. STATE OF MT 43

the current challenged regulations, which require traps to be baited at least 10 feet away and breakable by a grizzly-size pull. The Ninth Circuit has explicitly stated that to succeed on a motion for a preliminary injunction, the Plaintiff must establish a relationship between the injury claimed in the motion and the conduct asserted in the complaint. See generally, Pac. Radiation Oncology, LLC v. Queen's Med. Ctr., 810 F.3d 631, 637–38 (9th Cir. 2015). As discussed in Winter, we need to “have before us at least some estimate of the harm likely avoided by [the preliminary injunction]. Without such evidence, it is difficult to assess the relevant harm—that is, the environmental harm likely caused by the . . . measures[] in place.” 555 U.S. at 34 (J. Breyer, concurring in part). While my colleagues recite that “[c]lear error is not demonstrated by pointing to conflicting evidence in the record,” the error here is not one of conflicting evidence; it is the improper reliance on insufficient, speculative, and non-causational evidence in the first place. While Plaintiffs offer some direct challenges to the new regulations, including a challenge to the slight reduction in size of the known grizzly bear zone through improved data collection, the evidence presented is simply too generalized. A preliminary injunction cannot, at its core, cure harm that is not causally connected to the challenged regulations. Enjoining Montana’s regulatory scheme is not only inherently speculative, but it violates long held principles of redressability. Plaintiffs have not established that irreparable harm is likely to occur prior to a decision on the merits. Thus, the granting of this preliminary injunction relying on generalized harm and speculation not tied to Montana’s conduct was error. 44 FLATHEAD-LOLO-BITTERROOT V. STATE OF MT

III

Accordingly, I would hold that the district court erred by not viewing the evidence in its totality, focusing instead on insufficient evidence that in context amounts to mere speculation and conjecture. Nor did Plaintiffs establish that the lawful placement of wolf traps and snares under Montana’s prior and current laws and regulations is reasonably likely to result in future take of grizzly bears. Because Plaintiffs did not establish a likelihood of future harm, the preliminary injunction should be vacated in toto. I also agree that the district court’s unilateral line drawing covering virtually the entire State geographic area west of Billings, Montana, was overly broad and too generalized to stand.