Center for Biological Diversity, et al. v. U.S. Forest Service, et al.

• Court: District Court, D. Montana
• Decided: December 11, 2025
• Docket: 9:23-cv-00110
• Status: Unknown

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MONTANA MISSOULA DIVISION

CENTER FOR BIOLOGICAL Lead Case No. DIVERSITY, et al., CV 23–110–M–DWM

Plaintiffs, Member Case No. CV 23–154–M–DWM vs.

U.S. FOREST SERVICE, et al., ORDER Defendants,

and

SUN MOUNTAIN LUMBER, INC., a Montana Corporation,

Defendant-Intervenor.

The plaintiffs are environmental organizations challenging the United States Forest Service’s (the “Forest Service”) and the United States Fish and Wildlife Service’s (the “Fish and Wildlife Service”) approval of a forest treatment project in the Custer Gallatin National Forest (the “South Plateau Landscape Area Treatment Project,” the “South Plateau Project,” or the “Project”) under the National Environmental Policy Act (“NEPA”), the National Forest Management Act (“NFMA”), and the Endangered Species Act (“ESA”). (Doc. 12.)1 Center for Biological Diversity, Alliance for the Wild Rockies, and Council on Wildlife and

Fish (collectively “Center for Biological Diversity”) are plaintiffs in the lead case, CV 23–110–M–DWM. Gallatin Wildlife Association, Native Ecosystems Council, and WildEarth Guardians (collectively “Gallatin Wildlife Association”) are

plaintiffs in the member case, CV 23–154–M–DWM. Collectively they are referred to as “Plaintiffs.” The defendants are all officials or agencies within the United States Department of Agriculture (“Federal Defendants”) and Sun Mountain Lumber, Inc. (“Defendant-Intervenor”) (collectively, “Defendants”).

The parties’ filed cross-motions for summary judgment, (Docs. 45, 47, 54, 56), and the Center for Biological Diversity moved to supplement the administrative record, (Doc. 43). A motion hearing was held before United States

Magistrate Judge Kathleen L. DeSoto on March 11, 2025, (see Doc. 68 (Min. Entry)), and on March 31, 2025, Judge DeSoto entered Findings and Recommendations, recommending that the Center for Biological Diversity’s motion to supplement be denied and that summary judgment be granted in favor of

Defendants, (Doc. 69), Plaintiffs filed objections, (see Docs. 72, 73), to which Defendants responded, (Docs. 75, 76). Plaintiffs’ objections are reviewed de novo, 28 U.S.C. § 636(b)(1), and addressed individually below. The Findings and

1 Docket citations are to the lead case, CV 23–110–M–DWM. Recommendations are otherwise reviewed for clear error. See Thomas v. Arn, 474 U.S. 140, 154 (1985); United States v. Syrax, 235 F.3d 422, 427 (9th Cir. 2000)

(defining “clear error”). Because Judge DeSoto provided a complete background of the Project, (Doc. 69 at 4–7), it is not restated here. The Forest Service’s compliance with NEPA, NFMA, and the ESA is

reviewed under the Administrative Procedure Act (“APA”). See Native Ecosystems Council v. Marten, 883 F.3d 783, 788 (9th Cir. 2018). The APA authorizes a court to “hold unlawful and set aside agency action, findings and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or

otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). Agency action is arbitrary and capricious if the administrative record demonstrates that “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., Inc. v. State Farm Mut. Auto. Ins.

Co., 463 U.S. 29, 43 (1983). Where an agency’s administrative record is complete and constitutes the whole and undisputed facts underlying agency decisionmaking, summary judgment is appropriate. See City & Cnty. of San Francisco v. United

States, 130 F.3d 873, 877 (9th Cir. 1997). ANALYSIS Although Plaintiffs argue specific components of the South Plateau Project

fail to comply with NEPA, NFMA, and the ESA, their primary challenge concerns the Project’s condition-based management approach. “Condition-based management involves developing proposed treatments based on pre-identified

management requirements but deferring specific decisions about which treatments will be applied in particular locations until the Forest Service conducts pre- implementation field reviews.” N. Cascades Conserv. Council v. U.S. Forest Serv., 136 F.4th 816, 829 (9th Cir. 2025). Consistent with this approach, the Forest

Service “has identified areas as preliminarily suitable for treatment actions” within the Project area without identifying the precise location and size of the treatment units or the location and configuration of associated roads. See SP_004339–40,

4343. Rather, the Decision Notice and Final EA assess the potential “maximum” effects of possible treatments and outline several “design features” that future treatment activities must comply with before they may proceed. See SP_004340, 4343. For example, “treatment actions will only be implemented after [an]

interdisciplinary team of resource specialists verifies that the action is consistent with the Treatment Matrix (Appendix A), meets the Design Features (Appendix B), and completes Resource Review Checklists (Appendix C). Then, the

Responsible Official will review the plan and complete the Concurrence Checklist (Appendix C) to approve the action.” SP_004343. According to the Forest Service, this approach “increase[s] [its] flexibility to respond to any change in

conditions in the project area during the time between [the] decision and implementation.” SP_004490. Plaintiffs’ skepticism of whether a condition-based management approach—

i.e., the deferral of all specific decisions regarding project activities—meets the requirements of NEPA, NFMA, or the ESA permeates their objections. Plaintiffs argue that this approach conflates a promise of future statutory compliance with actual compliance. Plaintiffs insist that it is impossible to assess the true

environmental impacts of the Project under NEPA, or determine whether it meets statutory or Forest Plan protections for grizzly bears and lynx under the ESA and NFMA, without identifying the specific project activities that will occur. More

specifically, Plaintiffs argue that: (1) the Forest Service failed to take the requisite “hard look” at the environmental impacts on grizzly bears and climate change and failed to prepare an environmental impact statement (“EIS”) in violation of NEPA; (2) the Forest Service failed to show that the Project complies with specific Forest

Plan Standards governing grizzly bear secure habitat and regenerative habitat for lynx in violation of NFMA; and (3) the Fish and Wildlife Service improperly defined “patch” size for grizzly bear secure habitat, failed to adequately consider

the Project’s effects on grizzly bears in its “no jeopardy” determination, and improperly relied on the Project’s Design Features to mitigate potential harm in violation of the ESA. Having reviewed those objections de novo, 28 U.S.C.

§ 636(b)(1), and the remainder of the Findings for clear error, Thomas, 474 U.S. at 154, the Findings and Recommendation are adopted in part, rejected in part, and modified as reflected below.

I. NEPA “[NEPA] is a procedural statute that requires federal agencies to take a ‘hard look’ at the environmental consequences of their actions.” N. Cascades, 136 F.4th at 821 (internal quotation marks omitted). To satisfy the “hard look” requirement,

an agency must provide “a reasonably thorough discussion of the significant aspects of the probable environmental consequences.” Ctr. for Biological Diversity v. Nat’l Highway Traffic Safety Admin., 538 F.3d 1172, 1194 (9th Cir.

2008). NEPA requires agencies to prepare an EIS for “major Federal actions significantly affecting the quality of the human environment.” 42 U.S.C. § 4332(2)(C). To determine whether an EIS is necessary, an agency “may prepare an Environmental Assessment (‘EA’) first.” N. Cascades, 136 F.4th at 821 (citing

40 C.F.R. § 1508.1(j) (2024)).2 “[Courts] examine the EA with two purposes in

2 NEPA’s implementing regulations have since been rescinded. See Removal of Nat’l Envtl. Policy Act Implementing Regs., 90 Fed. Reg. 10610 (Feb. 25, 2025) (to be codified at 40 C.F.R. pts. 1500–08). Because the Forest Service proceeded under the 2020 regulations, see SP_004502, those regulations are applied here. mind: to determine whether it has adequately considered and elaborated the possible consequences of the proposed agency action when concluding that it will

have no significant impact on the environment, and whether its determination that no EIS is required is a reasonable conclusion.” Ctr. for Biological Diversity, 538 F.3d at 1215. “An EIS must be prepared if substantial questions are raised as to

whether a project may cause significant degradation of some human environmental factor.” Blue Mtns. Biodiversity Project v. Blackwood, 161 F.3d 1208, 1212 (9th Cir. 1998) (internal quotation marks and alteration omitted). If the EA reveals no significant effect, the agency may issue a Finding of No Significant Impact, or a

“FONSI.” See Ocean Advocates v. U.S. Army Corps of Eng’rs, 402 F.3d 846, 864 (9th Cir. 2005). In their objections, Plaintiffs argue that the Forest Service failed to take the

requisite “hard look” at the Project’s impacts on grizzly bears and climate change and that it was necessary to prepare an EIS. One of Plaintiff’s objections regarding grizzly bears has merit. A. Grizzly Bears

Plaintiffs argue that by relying on condition-based management, which does not identify the location of the 56.8 miles of temporary roads associated with the Project, the Forest Service failed to adequately consider the environmental impact

the Project will have on grizzly bear secure habitat. Plaintiffs also argue that the Forest Service arbitrarily limited the geographic scope of its NEPA cumulative effects analysis, omitting consideration of the nearby Yale Creek Project.

Plaintiffs’ first argument has merit. 1. Temporary Roads Plaintiffs first argue that the Forest Service’s reliance on condition-based

management and a maximum effect analysis violates NEPA’s requirement that the agency take a “hard look” at the impact temporary roads will have on grizzly bear secure habitat. “Grizzly bear secure habitat is defined as areas greater than 10 acres that are more than 500 meters from motorized routes.” SP_004350. “The

most important predictors of survival . . . [a]re the amount of secure habitat within a bear’s home range and road densities outside of secure habitat.” FWS_002869. Roads therefore inherently pose a threat to grizzly bear abundance. FWS_000025–

32, 2868. “Up to 56.8 miles of temporary project roads may be constructed to a standard appropriate for their intended use to support project actions.” SP_004343. It is undisputed that the EA does not identify the precise locations of those roads.

See SP_004343 (“The exact locations of temporary roads are not yet known . . . .”). However, Defendants maintain that the Forest Service adequately considered the potential maximum impact these roads would have on grizzly bear secure habitat

by analyzing the “worst case scenario” of all roads being in existence at one time, SP_041089, and mapping their probable location based on “GIS data, current road system locations, and field surveys,” SP_004344, 041205, 041208. The Forest

Service also provided proposed maps for the first two timber sale contracts. See SP_006168, 4344. Plaintiffs argue that this analysis is inherently insufficient because road location, not total mileage, determines how roads will affect secure

habitat. Defendants maintain that the Forest Service’s identification of potential road location and its “maximum effects” analysis adequately considered those impacts. On this record, Plaintiffs have the better argument. A condition-based management approach has been historically limited to

situations involving inherent uncertainties in project activities, such as projects involving mineral exploration. See Te-Moak Tribe of W. Shoshone of Nev. v. U.S. Dep’t of Interior, 608 F.3d 592, 600 (9th Cir. 2010) (“An exploration

project . . . inherently involves uncertainties; if mining companies knew the precise location of mineral deposits before drilling, exploration would not be required. In approving mineral exploration projects, the [agency] must balance these uncertainties with its duty under NEPA to analyze possible environmental

impacts.”). In those situations, NEPA requires that the agency “analyze the [project] activities in all parts of the project area and impose effective avoidance and mitigation measures to account for unknown impacts.” Id.; see also N. Alaska

Envt’l Ctr. v. Kempthorne, 457 F.3d 969, 977 (9th Cir. 2006) (stating that the agency did not have to perform a “parcel by parcel examination of potential environmental effects” of oil and gas leases in the ocean “because the parcels

likely to be affected are not yet known”). But, as argued by Plaintiffs, the South Plateau Project involves no such uncertainty. Nonetheless, in its recent Northern Cascades decision, the Ninth Circuit

upheld a condition-based management approach in the context of a timber treatment project much like the one at issue here. 136 F.4th 816 (9th Cir. 2025). In that case, the Forest Service performed a “maximum effects analysis” for various treatment activities on 21,149 acres without specifically identifying the

treatment units. Id. at 829. But because the Forest Service “identified specific methods of understory thinning, overstory treatments, and fuels reduction and provided unit-by-unit maps of the maximum effects of each treatment,” the Ninth

Circuit ultimately held that the agency complied with NEPA, though it was a “close question.” Id. at 829–30. The narrowness of that holding is reflected in a recent district court decision, Alliance for the Wild Rockies v. United States Forest Service, where the court found an agency’s use of condition-based management

was “too vague” because the agency failed to provide a “unit-by-unit map[] of the maximum effect of each treatment type. Instead, the maps developed by the Service only identif[ied] locations for the two different restoration types— commercial and noncommercial.” 2025 WL 2655984, at *8 (E.D. Wash. Sept. 16, 2025).

This case falls somewhere between Northern Cascades and Alliance. Consistent with Northern Cascades, the Forest Service has met its NEPA mandate here as it relates to the maximum possible effect the treatment activities may have

across the identified units. See SP_004339–43. But what makes this case unique is the enduring dilemma of grizzly bears and roads. As stated above, the Forest Service evaluated the impacts on the landscape if all 56.8 miles of temporary road were implemented at the same time and explained how it identified the likely

location of those roads. See SP_004344–52. But it is the precise location of those roads, not their total mileage, that determines the impact they will have on grizzly bear secure habitat. And the Forest Service does not purport to analyze a “worst

case” scenario based on road placement, as opposed to mileage. See SP_041089 (explaining that its mapping of “temporary routes is generally considered the ‘worst case’ scenario, or the maximum mileage of temporary routes that would be constructed under the project”). Accordingly, the current NEPA analysis fails to

address an important aspect of the problem, Motor Vehicle, 463 U.S. at 43, “detract[ing] from the decisionmaker’s or public participant’s ability to assess a proposed action’s environmental consequences,” N. Cascades, 136 F.4th at 829. The Supreme Court has recently reaffirmed that “NEPA does not require the agency to weigh environmental consequences in any particular way. Rather, an

agency may weigh environmental consequences as the agency reasonably sees fit under its governing statute and any relevant substantive environmental laws.” Seven Cnty. Infrastructure Coal v. Eagle Cnty., Co., 605 U.S. 168, 173 (2025).

“Simply stated, NEPA is a procedural cross-check, not a substantive roadblock. The goal of the law is to inform agency decisionmaking, not to paralyze it.” Id. Here, however, in the context of grizzly bear secure habitat, the lack of specific road mapping prevents such informed decisionmaking. See Neighbors of Cuddy

Mtn. v. U.S. Forest Serv., 137 F.3d 1372, 1380 (9th Cir. 1998) (“NEPA requires consideration of the potential impact of an action before the action takes place.” (internal quotation marks omitted)). Accordingly, the Forest Service has failed to

meet its NEPA mandate. 2. Cumulative Effects – Yale Creek Project Plaintiffs’ second grizzly-bear based NEPA objection focuses on the Forest Service’s cumulative effects analysis, specifically its failure to consider the nearby

Yale Creek Project, which falls within the same bear management unit (“BMU”), but not subunit, as the South Plateau Project. It is well-recognized that an agency has discretion to define the scope of its cumulative effects analysis area so long as

it “show[s] that it considered the relevant factors,” “provide[s] support for,” and “justif[ies] its decision.” Selkirk Conserv. All. v. Forsgren, 336 F.3d 944, 959 (9th Cir. 2003). According to the Forest Service, because “[s]ubunits provide the

optimal scale for evaluation of seasonal feeding opportunities and landscape patterns of food availability for grizzly bears,” and “[t]he project area lies within the Madison #2, Henry’s #2, and Plateau #1 Subunits,” those subunits dictate the

“spatial boundary” of analysis for the Project. SP_041072; see also SP_004403. As a result, the Forest Service did not consider the cumulative effects of Yale Creek Project, which is located in the Henry’s BMU but outside Henry’s #2 subunit and the Project area. See SP_004403.

There is no question that NEPA does not mandate that analysis occur at any specific scale or the BMU level. But Plaintiffs argue that the cumulative effects analysis should have included the entire Henry’s BMU because certain forest plan

standards apply at the BMU, not subunit, level. Specifically, Plaintiffs cite to FW- STD-WLGB(3)(b), or the “1% Standard,” which addresses the total amount of secure habitat within a BMU: Total acreage of secure habitat below baseline values within a given [BMU] shall not exceed 1 percent of the acreage in the largest subunit within that [BMU]. The acreage of a project that counts against the 1 percent limit (for example the amount of secure habitat affected) is measured as the acreage within the 500-meter buffer around any temporary motorized access route or low-level helicopter flight line that intrudes into existing secure habitat.

SP_000065. In determining the scope of its cumulative analysis, “the agency has to draw the line somewhere” and “offer[] a reasonable justification for why it drew the line

where it did.” Friends of the Wild Swan v. Weber, 767 F.3d 936, 944 (9th Cir. 2014). Here, the Forest Service explained that the BMU subunit was the optimal scope of analysis based on grizzly bear foraging needs, SP_004403, and that even

at that scale, the action area is over 300,000 acres, SP_004402. See Friends of the Wild Swan, 767 F.3d at 945 (indicating that larger action areas may “dilute” effects analysis). But while the Yale Creek Project does not fall within those 300,000+ acres, it was not ignored. Indeed, the Forest Service specifically recognized that

“[i]mplementation of the South Plateau Project (Henry’s Lake #2 Subunit) depends on the Yale Creek Project’s impact to the 1%, as the allowed temporary reduction below baseline is assessed at the scale of the entire Bear Management Unit, not the

Subunit scale.” SP_004432, 004511–12. Thus, the very concern identified by Plaintiffs, i.e., the Yale Creek Project’s cumulative impact on the 1% Standard, was addressed. Accordingly, the Forest Service’s selection of the boundaries for analyzing cumulative effects on the grizzly bear was neither arbitrary nor

capricious. B. Climate Change Plaintiffs further argue that the Forest Service failed to adequately address

the cumulative impacts of climate change. This Court has previously explained that “merely discussing carbon impacts and concluding that they will be minor does not equate to a ‘hard look.’” Ctr. for Biological Diversity v. U.S. Forest

Serv., 687 F. Supp. 3d 1053, 1076–77 (D. Mont. 2023), overruled on other grounds 2025 WL 586358 (9th Cir. Feb. 24, 2025). Indeed, “NEPA requires more than a statement of platitudes, it requires appraisal to the public of the actual

impacts of an individual project.” Id. at 1077. Here, the Forest Service determined that the Project would result in a short-term release of carbon but, over the long term, was “likely to increase carbon storage and reduce emissions by reducing disturbance risk and storing carbon in wood products, as well as recapturing carbon

as forests regrow.” SP_004371–72. In reaching this conclusion, the Forest Service “tiered” the carbon analysis for the Project “to the programmatic analysis of carbon sequestration in the [2022] Land Management Plan EIS” for the Custer Gallatin

National Forest. SP_004372; see SP_001121–796. According to the 2022 Forest Plan EIS, potential management actions, including projects such as the South Plateau Project, would “affect up to less than 0.25 percent of the forested area” and result in “less than 1 teragram[] of carbon annually.” SP_001440. While the Ninth

Circuit has made clear that “agencies cannot ‘tier’ their environmental review under NEPA to assessments of similar projects that do not actually discuss the impacts of the project at issue,” Envt’l Def. Ctr. v. Bureau of Ocean Energy Mgm’t,

36 F.4th 850, 874 (9th Cir. 2022) (internal quotation marks omitted), the 2022 Forest Plan EIS directly considered this type of management activity in the Project area. Given the nature of this Project and the recency and scope of the tiered

analysis, the Forest Service adequately considered the Project’s cumulative effects on climate change. This objection lacks merit. C. EIS

Plaintiffs’ final objection under NEPA is that an EIS should have been prepared because the South Plateau Project is a “major Federal action[] significantly affecting the quality of the human environment.” 42 U.S.C. § 4332(2)(C). More specifically, Plaintiffs argue that an EIS was required because

they have raised a substantial question as to five of the ten “intensity” factors identified in 40 C.F.R. § 1508.27(b). However, other than disagreeing with Judge DeSoto’s conclusion as to this issue, Plaintiffs fail to specifically object to her

conclusions. Plaintiffs’ objection, (see Doc. 73 at 30–34), is almost identical to their summary judgment argument, (see Doc. 48 at 65–69). Finding no clear error in Judge DeSoto’s analysis, it is adopted here. Nonetheless, given the issues identified in this Order, the agency should consider whether an EIS is required

following any additional analysis performed on remand. II. NFMA “NFMA creates a two-step process for the management of our national

forests.” Native Ecosystems Council v. U.S. Forest Serv., 428 F.3d 1233, 1249 (9th Cir. 2005). First, the Forest Service must develop, maintain, and, as appropriate, revise Forest Plans, which provide a framework for where and how certain

activities can occur in national forests. Id.; see also 16 U.S.C. § 1604(f). Second, the Forest Service must ensure that all individual projects within a forest are “consistent with each forest’s overall management plan.” Native Ecosystems

Council, 428 F.3d at 1249; see also 16 U.S.C. § 1604(i). “It is well-settled that the Forest Service’s failure to comply with the provisions of a Forest Plan is a violation of NFMA.” Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 961 (9th Cir. 2005).

Here, Plaintiffs argue that the Project violates NFMA because it fails to comply with the Forest Plan standards for grizzly bear secure habitat and limiting timber harvest within lynx habitat. Plaintiffs’ primary objection is that the promise

of future compliance with Forest Plan standards does not satisfy NFMA. Plaintiffs further argue that the Project fails to comply with the Forest Plan Standard VEG- S2 governing lynx or FW-STD-WLGB(03)(b) (the “1% Standard”) and FW-STD- WLGB(03)(c) (the “Four-Year Standard”) governing grizzly bears. Because

Plaintiffs are correct on all fronts, the Project fails to comply with NFMA. A. Future Compliance Plaintiffs’ primary objection focuses on the Forest Service’s use of “Design

Features” to meet its NFMA obligations. Because the Project is anticipated to occur across 16,462 acres over the next 15 years, it does not identify the precise location of the proposed treatment activities or Project roads. See SP_004338–40.

Rather, the Forest Service “has identified areas as preliminary suitable for treatment actions” and set maximum acreage footprints for treatment units with “[t]he precise location and size of the treatment units [to] be determined by

applying the Design Features” that appear in Appendix B in the EA. SP_004340. According to the Forest Service, this approach is intended “to account for potential changes in on the ground conditions over the 15-year period of implementation.” SP_004340. Under this approach, all future treatment activities, including fuels

reduction and road construction, must comply with the identified Design Features prior to implementation. See SP_004340–43. More specifically, After the decision, treatment actions will only be implemented after the interdisciplinary team of resources specialists verifies that the action is consistent with the Treatment Matrix (Appendix A), meets the Design Features (Appendix B), and complete Resource Review Checklists (Appendix C). Then, the Responsible Official will review the plan and complete the Concurrent Checklist (Appendix C) to approve the action.

SP_004343; see SP_004475 (stating the Decision “includes and makes binding the application of the Treatment Matrix, Design Features, Resource Review Checklists, and the Monitoring Plan”). Relevant here, the Design Features incorporate the Forest Plan Standards for lynx, see SP_004429–31, and grizzly bears, see SP_004432–34. Plaintiffs argue that because these Design Features are mere “promise[s] to comply with forest plan standards” in the future, (Doc. 73 at 20), they are

insufficient to show NFMA compliance in the present. In response, Defendants argue that because all project activities must comply with the Design Features, such compliance is not optional but rather “[is] the Decision framework” even if

the substantive review will occur in the future. (Doc. 75 at 19–20.) Plaintiffs have the better argument. The Ninth Circuit has previously rejected an agency’s “decision to authorize the Project first and verify [its compliance with NFMA] later.” Ecology Ctr. v. Austin, 430 F.3d 1057, 1071 (9th Cir. 2005). In that case,

for example, the Forest Service “plan[ned] to verify soil conditions in the activity areas after authorizing the Project, but before actually commencing harvest activities.” Id. The Ninth Circuit held that that approach “violates NFMA.” Id.

The situation here is no different. The EA does not show that the Project meets the Forest Plan Standards but rather that such compliance will be assessed after the specific treatment actions are identified. But that assurance is merely a truism of NFMA itself: “NFMA requires that the proposed site-specific actions be consistent

with the governing Forest Plan.” Native Ecosystems Council v. Weldon, 697 F.3d 1043, 1056 (9th Cir. 2012); see All. for the Wild Rockies v. U.S. Forest Serv., 907 F.3d 1105, 1110 (9th Cir. 2018) (“A project is consistent [with NFMA] if it

conforms to the applicable ‘components’ of the forest plan, including the standards, guidelines, and desired conditions that are set forth in the forest plan and that collectively establish the details of forest management.”). The Forest Service

avowing that future activities will comply with NFMA falls far short of showing that those activities do so. “Under NFMA, the Forest Service calculations need not be perfect. However, [courts] must still be able to reasonably ascertain from the

record that the Forest Service is in compliance with the [Forest] Plan standard[s].” Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 963 (9th Cir. 2005). That is not possible here. Accordingly, the Forest Service failed to meet its obligations under NFMA

by relying on future compliance with its provisions. See WildEarth Guardians v. Jeffries, 370 F. Supp. 3d 1208, 1234 (D. Or. 2019) (“The duty to demonstrate Forest Plan consistency applies at the time of the decision, not at a speculative

future date.”). The problems posed by a condition-based management approach to NFMA compliance are only emphasized by the fact that the record indicates that future activities may violate the standards identified by Plaintiffs. See Save Our Cabinets v. U.S. Dep’t of Agr., 254 F. Supp. 3d 1241, 1259 (D. Mont. 2017)

(finding a NFMA violation because the authorized action was expected to violate forest plan standards in the future and therefore the decision was not supported by the record).

B. VEG-S2: Lynx Regeneration Standard Pursuant to the VEG S2 Forest Plan Standard, “[t]imber management projects shall not regenerate more than 15 percent of lynx habitat on [National

Forest system] lands within a[ Lynx Analysis Unit] in a ten-year period.” SP_000567. Here, the Forest Service calculated that while 14,572 acres of lynx habitat would be affected in the South Madison Lynx Analysis Unit, SP_004676, to

comply with VEG S2, no more than 4,696 acres of lynx habitat may be regenerated within the Project area, SP_004508. Accordingly, one of the Design Features of the Project limits regeneration to “no more than 4,600 acres of lynx habitat to comply with this standard.” SP_004508. But because the exact nature of the

treatments is unknown and the Decision Notice authorizes 5,551 acres of clearcuts, SP_004340, 4477, the current record does not show that Standard will be met. More specifically, the record does not show where the regeneration harvest units

are located in relation to lynx habitat, nor the size of those units. To the contrary, the record shows an intent to clearcut areas greater than that permitted under VEG S2 Standard. NFMA compliance demands more than the agency’s “word” that it will comply. See Alliance for the Wild Rockies v. Marten, 2021 WL 4551496, at *4

(D. Mont. Oct. 5, 2021). C. FW-STD-WLGB(3)(b): The 1% Standard FW-STD-WLGB(03)(b), or the 1% Standard, addresses the total amount of

grizzly bear secure habitat within a BMU: Total acreage of secure habitat below baseline values within a given [BMU] shall not exceed 1 percent of the acreage in the largest subunit within that [BMU]. The acreage of a project that counts against the 1 percent limit (for example the amount of secure habitat affected) is measured as the acreage within the 500-meter buffer around any temporary motorized access route or low-level helicopter flight line that intrudes into existing secure habitat.

SP_000065. The Forest Service interpreted this standard as assessing the total impacts to secure habitat at a given moment in time, which would permit the “staging” of project activities over time. Plaintiffs argue that the Forest Service’s interpretation is neither permissible nor supported by the Project record. Ultimately, while the Forest Service’s interpretation of the Standard is reasonable, the condition-based approach the Forest Service has adopted here prevents it from showing the Standard is met. The Forest’s Service’s temporal interpretation is permissible. “Agencies are entitled to deference to their interpretation of their own regulations, including Forest Plans.” Native Ecosystems Council v. U.S. Forest Serv., 418 F.3d 953, 960 (9th Cir. 2005). No such deference is due, however, if “it is plainly inconsistent with the regulation at issue,” “is without a substantial basis in fact,” or is “based on

a clear error of judgment.” Id. (internal quotation marks omitted). Plaintiffs insist that the 1% Standard applies over the life of the Project and that the Project’s “staging” of activities would, in aggregate over the course of years, reduce secure

habitat below the 1% limit in violation of the Forest Plan. But that reading is contrary to the plain language of the provision. The 1% Standard, consistent with the Forest Service’s interpretation, falls under the heading of “temporary [c]hanges

in secure habitat” or “temporary reductions” and applies to the “total acreage of secure habitat below baseline values” in the present, not in the aggregate. See SP_000065 (emphasis added). It is also consistent with the Grizzly Bear

Conservation Strategy’s use of the term “at any one time” to describe the limitation. See SP_045858 (“at any one time”). Thus, the Forest Service’s interpretation is permissible. Nevertheless, the Project record does not show that this Standard will be met

here. Analyzing this issue requires a disassociation between the numerical calculations provided and the facts upon which they are based. Essentially, while the Forest Service’s math checks out, it is unclear how the Forest Service can

calculate the amount of secure habitat that will be lost without first determining where the temporary roads will be routed. For example, the Forest Service determined that in the Madison #2 Subunit, project activities will reduce secure habitat by 248 acres below baseline, which equates to a reduction in secure habitat

of 0.2% of the acreage of the largest subunit in that BMU. SP_041106. Similarly, while the Project authorizes a total reduction in secure habitat below baseline in Henry’s #2 of 2,426 acres over the lifetime of the Project (a 1.9% reduction), it

permits only 1,276 acres at a time. SP_004511–12, SP_041091; see also SP_041205. But the Project authorizes the construction of 56.8 miles of temporary roads without committing to exact routes or locations. See SP_004343 (“The exact

locations of temporary roads are not yet known . . . .”). Plaintiffs insist it is impossible to assess compliance with the 1% Standard without knowing the road locations within each of the three BMUs. Indeed, “secure habitat” is defined as

blocks of habitat of at least 10 acres that are at least 500 meters from a road. SP_004406. Accordingly, road location, not total mileage, is critical to determine compliance with this Standard. While the Forest Service has provided a potential map of those routes, see SP_041205, 4339, the EA clearly states that these routes

“have not been vetted and are entirely subject to change,” SP_004344. Where even a ten-foot shift in a single road could affect compliance, the Forest Service has failed to show that the 1% Standard has been met here.

D. FW-STD-WLGB(3)(c): The Four-Year Standard FW-STD-WLGB(3)(c), or the Four-Year Standard, also addresses grizzly bear secure habitat: New temporary roads shall be limited to administrative purposes associated with project activities. Project activities shall not reduce secure habitat below baseline levels for more than four consecutive years. The collective set of temporary roads that affect secure habitat below baseline levels shall be closed to all motorized use after three years. Temporary roads shall be decommissioned such that secure habitat is restored within one year after closure. SP_000065. The Forest Service concluded that this standard was met here because the Project will be split into 4 to 6 separate timber sale or other contracts,

SP_004343, 041092, and each individual timber sale its own “project” with its own “project activities.” Thus, according to the agency, it is the individual timber contracts, not the South Plateau Project as a whole, that must comply with the

Four-Year Standard. Plaintiffs argue that this “interpretation conflicts with the standard’s plain language, undermines its purpose, and is contradicted by the agency’s own interpretation in the record.” (Doc. 73 at 20.) Plaintiffs are correct. The Forest Plan does not define the term “project activities.” And while the

Forest Plan often lists both “projects and activities” or “projects or activities,” see, e.g., SP_000012, 18–19, it only uses the term “project activities” seven times, all but once in reference to the grizzly bear standards and guidelines at issue here, see

SP_000065, 67, 68, 107. Nevertheless, the Forest Plan refers to “projects” in terms of the site-specific decisions that must be authorized through NEPA. See, e.g., SP_000010. Thus, the term “project,” acting as a noun adjunct, means that “project activities” refers to the collective whole of those activities undertaken as

part of the NEPA-approved decision, not the component parts of that decision. As a result, the Forest Service’s attempt to fragment the South Plateau Project into four to six different sets of “project activities” is counter to the plain language of the Forest Plan: all activities associated with the NEPA-approved project may not reduce secure habitat below the requisite levels for more than four years.

Review of the Forest Plan’s purpose only strengthens this conclusion. Defendants argue that aggregating “project activities” would prevent “long-term” projects from occurring. But that appears to be by design. The Forest Plan itself

states that it is only meant to “provide guidance for project and activity-level decision making on the national forest for approximately the next 15 years.” SP_000010. This indicates that, counter to Defendants’ position, the Forest Plan was not authored with the purpose of approving projects that have the same

lifespan as the Forest Plan itself. But even if the Forest Plan’s use of the term “project activities” could be read to apply to the segregated actions within a single project, this record shows

that was not how the Forest Service used these terms in this case. The term “project” is used over a thousand times in the EA for the South Plateau Project, SP_004326–4469, and over four hundred times in the Decision Notice, SP_004470–4551. Every single one of those references is to the “South Plateau

Project” as a whole, not to the individual timber sales or other contracts underlying the Project. The EA also uses the term “project activities” multiple times, but once again, always in reference to the South Plateau Project as a whole. See, e.g.,

SP_004380 (“Project activities will be spread out in time and space, so unaffected portions of the project area will be available while treatments are ongoing.”); SP_004403 (“Project activities will not reduce the presence or abundance of moose

and elk (ungulates) in the area, but the proposed activities could alter their distribution.”); SP_004407 (“[P]roject activities will be implemented in stages.”); SP_004413 (“[P]roject activities will be spread out in time and space so that

undisturbed habitat will be available for nesting and foraging for migratory birds in any given year in the project area.”); SP_004446 (“Project activities include all proposed treatment actions and associated activities.”). The same can be said of the references to “project activities” in the Decision Notice. See, e.g., SP_004490

(“Based on effects analyses conducted for this project, I have determined that the effects likely to result from project activities do not rise to the level of significance.”); SP_004493 (“In the short-term, (the period of project

implementation) adverse effects will result from disturbance caused by project activities.”); SP_004525 (“Project activities include all proposed treatment actions and associated activities.”). To be sure, the Fish and Wildlife Service states in the Project Biological

Opinion that “[e]ach individual sale or stage is considered as an individual project under the standard and application rules.” SP_005220. But this interpretation is not reflected by the Forest Service. To the contrary, in interpreting the Forest

Plan’s “one project” standard—which states that “[o]nly one project affecting secure habitat below baseline values may be active within a given bear management subunit at any one time,” SP_000065—the Forest Service considered

the South Plateau Project to be that “one project,” not the individual treatments or timber contracts. See SP_004432 (referencing the nearby Yale Creek Project and affirming that “[o]nly one project may reduce secure habitat below baseline in a

unit at a time”); see also SP_004401, 4409. Because neither the plain language of the Four-Year Standard nor this record supports the Forest Service’s anomalous interpretation of “project activities” in this context, the Forest Service’s decision is arbitrary and capricious in violation of the

APA and NFMA. III. ESA Section 7 of the ESA requires agencies proposing actions to engage in

consultation with the Fish and Wildlife Service to ensure the action “is not likely to jeopardize the continued existence of any endangered species or threatened species or result in the destruction or adverse modification of habitat of such species[.]” 16 U.S.C. § 1536(a)(2). An action will result in “jeopardy” if it will “reduce

appreciably the likelihood of both the survival and recovery of a listed species.” 50 C.F.R. § 402.02. If formal consultation is deemed necessary, the Fish and Wildlife Service issues a biological opinion. 50 C.F.R. § 402.14(g)–(h). Under

this consultation process, the Fish and Wildlife Service is required to, inter alia, “[e]valuate the effects of the action and cumulative effects on the listed species or critical habitat.” 50 C.F.R. § 402.14(g)(3). This analysis must consider the effects

on the species from the proposed action as well as the “consequences of other activities that are caused by the proposed action.” 50 C.F.R. § 402.02. The “environmental baseline includes the past and present impacts of all Federal, State,

or private actions and other human activities in the action area, the anticipated impacts of all proposed Federal projects in the action area that have already undergone formal or early section 7 consultation, and the impact of State or private actions which are contemporaneous with the consultation process.” Id.

Plaintiffs challenge the Fish and Wildlife Service’s ESA analysis for grizzly bears as it relates to patch size, secure habitat, and the agency’s no jeopardy finding. With limited exception, these challenges have merit.

A. Patch Size In the biological opinions for both the 2022 Forest Plan and for the present Project, the Fish and Wildlife Service relied on metrics associated with grizzly bear secure habitat to assess the effects of roads on grizzly bears. See SP_005199–201.

“Secure habitat is defined in the 2016 [Grizzly Bear] conservation strategy as areas more than 500 meters from an open or gated motorized access route or reoccurring helicopter flight line that are greater than or equal to 10 acres in size.” SP_005200

(citations omitted). Consistently, both biological opinions determined the overall quantity of secure habitat using a 10-acre “patch size.” According to Plaintiffs, there is no scientific support for defining “secure habitat” as patches of land as

small as 10 acres. That objection has merit. The ESA requires an agency to use “the best available scientific and commercial data available” when formulating a biological opinion. 16 U.S.C.

§ 1536(a)(2). “An agency complies with the best available science standard so long as it does not ignore available studies, even if it disagrees with or discredits them.” San Luis & Delta-Mendota Water Auth. v. Locke, 776 F.3d 971, 995 (9th Cir. 2014); see Kern Cnty. Farm Bureau v. Allen, 450 F.3d 1072, 1081 (9th Cir.

2006) (“Without any evidence in the record that [the Service] ignored relevant information, we hold that [the Service] satisfied its duty to base its listing determinations on the best available data.”). Courts “must defer to the agency’s

interpretation of complex scientific data” so long as the agency “has articulated a rational basis for its conclusion.” Nw. Ecosystem All. v. U.S. Fish & Wildlife Serv., 475 F.3d 1136, 1150 (9th Cir. 2007). Here, in defending the Fish and Wildlife Service’s reliance on a 10-acres

patch size, Defendants argue that: (1) this has been the patch size used in the Greater Yellowstone Ecosystem since 2003; (2) the science supports any patch size, however small; and (3) a small patch size is more sensitive in assessing loss

of secure habitat than a larger patch size. Because none of these arguments provide scientific support for the decision to adopt a 10-acre patch size, they do not support a finding that the agency used the “best available science” as required under the

ESA. 1. Metric for the Greater Yellowstone Ecosystem Since the late 1980s, grizzly bears within the Greater Yellowstone

Ecosystem have been “managed under standards and guidelines established by the Interagency Grizzly Bear Committee.” SP_044366; see FWS000815. “Acknowledging that humans are the primary agent of grizzly bear mortalities, a principal objective of the [Committee] was to improve survival rates by

implementing management strategies that minimized anthropogenic influences and grizzly bear-human conflicts.” SP_044366. That includes the recognition “that secure habitat (areas that are free of motorized traffic, also referred to as core

areas) is an important component of grizzly bear habitat,” SP_044341, and that “[a]dequate secure habitat is essential to the survival and reproductive success of grizzly bears,” SP_044372. Accordingly, “[i]t is the goal of habitat management agencies to maintain or improve habitat conditions . . . at or above 1998 levels,”

SP_044372, as the “conditions in 1998 are believed to have supported and contributed to the population growth observed during 1983–2001[,] . . . establish[ing] a clear benchmark against which future improvement and impacts of

habitat can be measured,” SP_044368. It is less clear, however, how the Committee determined what qualifies as “secure habitat” as that term is used above. For example, the 2016 Interagency

Grizzly Bear Conservation Strategy merely states that “[s]ecure habitat is defined as any contiguous area ≥ 10 acres in size and more than 500 m[eters] from an open or gated motorized access route (road or trail) or recurring low level helicopter line

during the non-denning period (March 1 – November 30),” SP_044373, and “secure habitat is correlated with key habitat components, such as foods, cover, space, arrangement of habitat types, and ability of grizzly bears to move between them,” SP_044374. But neither reference explains the adoption of a 10-acre

standard. Nor do the 2003 or 2007 conservation strategies provide any further information regarding how the 10-acre metric was chosen, other than to state that “[h]abitat security allows a population to increase in numbers and distribution as

lowered mortality results in more reproduction and cub recruitment into the adult population. This results in an increasing population.” SUPP-FWS_8090; see SUPP-FWS_008075, 8109, 8215 (2003 Strategy); SUPP-FWS_007208, 7223, 7244 (2007 Strategy).3

3 Notably, these strategies do explain the portion of the “secure habitat” definition regarding proximity to roads, stating that “[g]rizzly bear researchers and managers generally agree that secure habitat, defined as those area more than 500 meters (550 yards) from a motorized access route during the non-denning period, are especially important to the survival and reproductive success of grizzly bears, especially adult female grizzly bears.” SUPP-FWS_007244, 8111. Thus, contrary to Defendants’ argument, the Interagency Grizzly Bear Conservation Strategy documents do not provide a scientific basis for the Fish and

Wildlife Service’s adoption of the 10-acre patch size for secure habitat. 2. Any Patch Size In the 2022 Species Status Assessment for the Grizzly Bear in the Lower-48

States, the Fish and Wildlife Service explained that its “definition of secure habitat includes areas as small as 10 acres . . . in size because both the [Interagency Grizzly Bear Study Team] and the [Yellowstone Ecosystem Subcommittee] concluded that all secure habitats are important for grizzly bears in the [Greater

Yellowstone Ecosystem], regardless of size, particularly in the peripheral areas.” FWS_000819, 1042. But that statement does not cite a specific document by either organization. While the agency states that “[r]esearch by Schwartz et al. (2010 p.

661) supported this conclusion and demonstrated a direct link between this definition and grizzly bear survival in the [Greater Yellowstone Ecosystem],” FWS_000819, Schwartz (2010) merely concluded that “[t]he most important predictors of survival . . . [a]re the amount of secure habitat within a bear’s home

range and road densities outside of secure habitat,” SUPP-FWS_006747. Contrary to the Fish and Wildlife Service’s characterization, Schwartz (2010) does not provide any independent justification for a 10-acre patch size, as it merely adopted

the “secure habitat” definition used by the Interagency Grizzly Bear Committee with no further discussion. See SUPP-FWS_006743. Thus, while the record supports the Fish and Wildlife Service’s general conclusion that some amount of

roadless acreage is better for bears than none, Defendants identify no specific explanation of why 10 acres, as opposed to 1 acre, 5 acres, or 50 acres, is an appropriate metric.

That omission is only more telling when one considers the secure habitat standard for the Northern Continental Divide Ecosystem. In that Ecosystem, “secure habitat” is defined as “areas on Federal lands within the analysis area more than 500 m[eters] (1,650 f[ee]t) from a motorized access route and at least 2,500

acres . . . in size, and in place for 10 years.” FWS_001043. In that case, the size of the secure area is explicitly tied to foraging range: “The 2,500 acre . . . minimum size for secure core habitat is based on the 1994 [Interagency Grizzly Bear

Conservation] Guidelines that state minimum size will be recovery zone specific and ‘the minimum size for the core area(s) be that area necessary to support a female grizzly bear for 24 hours of foraging.’” FWS001043. While grizzly bears in different Ecosystems may have different habitat needs, see SP_005401–02, the

10-acre patch size here remains untethered to any identified grizzly bear survival metric. Thus, even if Defendants are correct that “any size” of secure habitat will do, that lends no specific support for a 10-acre patch size.

3. Sensitivity Defendants’ primary argument in support of the Fish and Wildlife Service’s adoption of a 10-acre patch size, and the one most clearly presented in the

biological opinions for the 2022 Forest Plan and this Project, is that a smaller patch size is more sensitive to habitat loss than a large patch size. In its 2022 programmatic biological opinion for the Forest Plan, the Fish and Wildlife Service

explicitly recognized that grizzly bears in different Ecosystems have different habitat requirements, SP_005402, and that there are pros and cons to both large and small patch sizes, explaining: Patch size of secure habitat is also an important consideration in the effectiveness of secure habitat. Although small patches of secure habitat may provide stepping stones to facilitate movement by bears, Gibeau et al. (2001, p.[ ]124) estimated the minimum daily foraging requirements for grizzly bears in Banff National Park, Canada is . . . []2,224 acres[]. Small patches of secure habitat may also be less effective if they are within a matrix of high road density . . . . However, a small patch size provides greater sensitivity for determining loss of secure habitat. Existing secure habitat calculated using a minimum patch size of 2,500 acres would allow many additional roads to occur in patches <2,500 acres without indicating a net loss in secure habitat. Comparatively, a minimum patch size of 10 acres would be much more sensitive to net losses in secure habitat from additional roads.

SP_005403. This is then consistent with the Final EIS issued by the Forest Service for the 2022 Forest Plan, which states: [t]he 10-acre minimum patch size for secure habitat is a reasonably sized area that is useable by an individual grizzly bear, while avoiding disturbance associated with motorized use on roads and trails. The 10- acre minimum patch size represents the minimum size area that would be protected by plan components for secure habitat. If the minimum patch size for secure habitat were much larger, it would mean that larger patches of secure habitat could be eliminated by road building, and still comply with plan components to maintain secure habitat.

SP_001537 (citation omitted). But as argued by Plaintiffs, neither explanation fills the analytical gap identified above, i.e., how or why does a 10-acre area represent the minimum area needed for grizzly bear habitat? In the absence of a connection between the size of the patch and the habitat needs of a grizzly bear, a 10-acre metric “cannot be more

‘sensitive’ to loss of secure habitat if the prescribed patch size is insufficient to provide grizzly bears truly secure habitat in the first place.” (Doc. 73 at 37.) Indeed, the record shows that the 1994 Interagency Grizzly Bear Committee recommended secure areas “be that area necessary to support a female grizzly bear

for 24 hours of foraging,” FWS_003694, and based on similar metrics, grizzly bears in other Ecosystems have been found to need upwards of 2,000 acres of secure habitat, SP_005403 (recognizing estimated need for 2,224 acres of range in

Banff National Park, Canada based on daily foraging requirements); FWS001043 (defining secure habitat as 2,500 acres for the Northern Continental Divide Ecosystem). Even Schwartz (2010), heavily relied on by Defendants here, explicitly states that smaller secure areas can create isolated “islands” of habitat

that actually increase grizzly bear mortality by “fragmenting the landscape.” See FWS002869. 4. Conclusion In relying on a 10-acre patch size to define grizzly bear secure habitat in the absence of any scientific evidence showing that such acreage provides adequate

habitat, the Fish and Wildlife Service’s failed to use the “best available science” in violation of the ESA. B. Effects Analysis

During formal consultation, the Fish and Wildlife Service is, among other things, required to “[e]valuate the effects of the action and cumulative effects on the listed species,” and formulate its biological opinion as to whether the action, taken together with cumulative effects, “is likely to jeopardize the continued

existence of listed species.” 50 C.F.R. § 402.14(g); see also 16 U.S.C. § 1536(a)(2). Here, the Fish and Wildlife Service determined that while motorized access associated with the Project will temporarily disturb or displace grizzly bears

in the Project area, SP_005222, the Project is “not likely to jeopardize the continued existence of” the grizzly bear, SP_005233, 5240. Plaintiffs raise three objections to the Fish and Wildlife Service’s analysis of the Project’s effects on grizzly bears: (1) the agency’s speculation regarding

unknown road locations and project timing, (2) the agency’s reliance on the condition-based management features discussed above, and (3) the agency’s determination of the action area boundaries for the Project. Plaintiffs’ first

argument has merit. 1. Road Locations and Project Timing Plaintiffs first argue that the Fish and Wildlife Service failed to use the best

available science or consider an important aspect of the problem when it analyzed the putative effects the Project will have on grizzly bear secure habitat without knowing the location of the temporary project roads or the timing of the project

activities. In response, Defendants argue that the Fish and Wildlife Service’s decision to focus on secure habitat, not road location, is entitled to deference and that, regardless, the agency considered possible road locations, the maximum possible impact on secure habitat, and seasonal bear activity. Akin to the NEPA

analysis discussed above, Plaintiffs are ultimately correct that it was arbitrary and capricious for the agency to fail to consider road placement in its assessment of grizzly bear secure habitat.

As an initial matter, deference on the issue of secure habitat is not dispositive here. In Gifford Pinchot Task Force v. United States Fish and Wildlife Service, the plaintiffs challenged the Fish and Wildlife Service’s use of habitat as a proxy for its jeopardy analysis for the spotted owl. 378 F.3d 1059, 1065–66 (9th

Cir. 2004), superseded by regulation 81 Fed. Reg. 7214 (Feb. 11, 2016). The Ninth Circuit ultimately upheld the agency’s analysis because “[a]n agency’s scientific methodology is owed substantial deference,” and the Fish and Wildlife

Service demonstrated that its “habitat proxy reasonably correlate[d] to the actual population of owls.” Id. at 1066. But Plaintiffs’ challenge here is fundamentally different. Plaintiffs do not argue that secure habitat does not accurately reflect

grizzly bear abundance or that the Fish and Wildlife Service should not have relied on secure habitat as a proxy for jeopardy. Rather, Plaintiffs argue that the Fish and Wildlife Service could not accurately analyze secure habitat without knowing the

temporary route locations because secure habitat is defined by distance from a motorized route. Thus, Plaintiffs do not challenge the agency’s scientific judgment, but rather its execution, arguing that it was arbitrary and capricious to fail to consider a critical component of secure habitat, i.e. road placement. That

argument has merit. The Fish and Wildlife Service recognizes that “[r]esearch has confirmed adverse impacts of roads on grizzly bears,” SP_005211, and that “secure habitat

and road densities outside of secure habitat [a]re important predictors of grizzly bear survival,” SP_005215. Consistently, in its 2022 programmatic Biological Opinion prepared in conjunction with the Forest Plan, the agency “anticipated that some level of adverse effects to female grizzly bears with home ranges impacted

by such [motorized access] routes may occur in some situations in the future[,]” although “[t]he level of effects would depend on such things as location of the routes, length of the routes, and the frequency and intensity of use,” SP_005218;

see also SP_005222 (“The amount of disturbance and/or displacement would vary across the action area, depending on site-specific conditions and actions.”). Thus, any project-specific analysis would require consideration of those features. But, as

recognized by Defendants, the Fish and Wildlife Service’s secure habitat analysis here was based on its consideration of “the maximum proposed miles of temporary road construction and use (56.8 miles),” and the acreage information provided by

the Forest Service in its Biological Assessment, see SP_004707–08, which found that “approximately 2,417 acres or 5.2 percent of the secure habitat in the Henry’s Lake #2 subunit would be temporarily affected; approximately 228 acres or 0.4 percent of secure habitat in the Madison #2 subunit would be temporarily affected;

and approximately 2,328 acres or 1.9 percent of secure habitat in the Plateau #1 subunit would be temporarily affected,” SP_005219. Even assuming that these acreages accurately reflect the probable temporary

routes associated with the Project,4 the Fish and Wildlife Service failed to consider the potential impact to secure habitat if actual road placement differs in any way from the Forest Service’s projections. See SP_004741 (Figure 8). Despite acknowledging the importance of road placement in this analysis, see SP_005218,

the negative impacts roads will have on grizzly bear secure area in the Project area,

4 It is not clear how these acreages were calculated. The Biological Assessment states: “[a]creages were calculated using model runs performed for this project on 10 January 2022. Proposed Action secure habitat acreage was subtracted from the existing condition of secure habitat to get acres of secure affected. Results of this model run are included in the project record.” SP_004708. see SP_005222, and the uncertainty surrounding road placement under the condition-based management approach utilized here, see SP_004343, the Fish and

Wildlife Service provides no discussion of this issue. As argued by Plaintiffs, the worst-case scenario here is not that the Project may have 56.8 total miles of temporary roads, but that those temporary roads may be located in such a way as to

eviscerate secure habitat. The failure to consider that possibility is arbitrary and capricious. 2. Conditional Management/Mitigation Plaintiffs further argue that it was arbitrary and capricious for the Fish and

Wildlife Service to rely on the Project’s Design Features to mitigate effects on grizzly bears. Specifically, the Fish and Wildlife Service relied on the potential future dropping or staging of the Project in Henry’s #2 as a “key factor” in

concluding that the Project would not jeopardize grizzly bears. See SP_005234– 35. In response, Defendants argue that the Design Features are components of the Project, not mitigation measures, and that even if they were considered mitigation, they could be appropriately relied upon here. Defendants have the better

argument. As argued by Defendants, the Design Features are an inherent part of the Project and, as such, do not qualify as “mitigation” measures under the ESA. But

even if they did, their vagueness is not fatal in this context. To be sure, for a mitigation measure to be relied on, it must represent a “clear, definite commitment of resources” that is “reasonably certain to occur.” See Ctr. for Biological

Diversity v. Bernhardt, 982 F.3d 723, 743 (9th Cir. 2020) (internal quotation marks omitted) (“Binding mitigation measures cannot refer only to generalized contingencies or gesture at hopeful plans; they must describe, in detail, the action

agency’s plan to offset the environmental damage caused by the project.”). Because the Design Features themselves contain no details or deadlines, they fail to meet that requirement. But the Ninth Circuit recognizes that mitigation “measures can be made enforceable in a variety of ways, including by

incorporation into the terms and conditions of an incidental take statement.” Id. at 744. Because the Fish and Wildlife Service did that here, see SP_005245–46, its reliance on the Design Features in making its “no jeopardy” finding was not

arbitrary and capricious. That said, these features remain problematic for the other reasons identified herein. 3. Action Area – Yale Creek Project Plaintiffs are also critical of the Fish and Wildlife Service’s failure to

consider the effects of the nearby Yale Creek Project in its analysis. “When evaluating the effects of the action under the ESA, the agency is to consider direct and indirect effects to the species and/or critical habitat, together with other

activities added to the environmental baseline, which includes ‘the past and present impacts of all Federal activities in the action area, and the anticipated impacts of all proposed Federal projects in the action area.’” Friends of the Wild Swan v. Weber,

767 F.3d 936, 950 (9th Cir. 2014) (quoting 50 C.F.R. § 402.02) (alterations omitted). “The choice of appropriate action areas requires application of scientific methodology and, as such, is within the agency’s discretion.” Id. (internal

quotation marks omitted). Here, as discussed above, the Yale Creek Project is located in the Henry’s Lake BMU and therefore is relevant to assessing the South Plateau Project’s compliance with the 1% Forest Plan Standard, which is calculated at the BMU

level. See SP_000065. Nevertheless, the Fish and Wildlife Service performed its effects analysis at the BMU subunit level because “[s]ubunits approximate the annual home range of adult female grizzly bears and provide the optimal scale for

evaluation of seasonal feeding opportunities and landscape patterns of food availability for grizzly bears.” SP_005199. According to the agency, this level of analysis also prevents the action area from being “so large as to dilute the potential effects of the action,” SP_005199, and the Forest Plan Standards prevent

concurrent activity, SP_005221. Because the agency provided a reasonable explanation for its decision to define the action area at the subunit level, which excludes the Yale Creek Project from that area, that decision is due deference.

Friends of the Wild Swan, 767 F.3d at 950. This objection lacks merit. IV. Remedy The APA directs that “[t]he reviewing court shall . . . set aside agency

action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). Vacatur is the presumed remedy where an agency has acted unlawfully, All. for the

Wild Rockies v. U.S. Forest Serv., 907 F.3d 1105, 1121 (9th Cir. 2018), but the district court “is not required to set aside every unlawful agency action,” Nat’l Wildlife Fed’n v. Espy, 45 F.3d 1337, 1343 (9th Cir. 1995); see Cal. Cmtys. Against Toxics v. U.S. E.P.A., 688 F.3d 989, 992 (9th Cir. 2012) (per curiam) (“A

flawed rule need not be vacated.”). “Whether agency action should be vacated depends on how serious the agency’s errors are and the disruptive consequences of an interim change that may itself be changed.” Cal. Cmtys., 688 F.3d at 992

(internal quotation marks omitted). In assessing the seriousness of the error, courts “consider whether vacating a faulty [decision] could result in possible environmental harm.” Pollinator Stewardship Council v. Env’t Prot. Agency, 806 F.3d 520, 532 (9th Cir. 2015).

Another consideration is “whether the agency would likely be able to offer better reasoning or whether by complying with procedural rules, it could adopt the same [decision] on remand, or whether such fundamental flaws in the agency’s decision

make it unlikely that the same [decision] would be adopted on remand.” Id. Additionally, courts consider whether the errors are “limited in scope.” All. for the Wild Rockies v. Savage, 375 F. Supp. 3d 1152, 1156 (D. Mont. 2019).

Here, while the Forest Service may be able to offer additional reasoning on remand and approve a variation of the Project, the errors identified above undermine the ability to use a condition-based management approach in this

context. Because that approach is the very foundation of the Project, vacatur is the appropriate remedy. III. Conclusion Based on the foregoing and finding no clear error in the portions of Judge

DeSoto’s Findings and Recommendations that were not objected to, IT IS ORDERED that: (1) The Findings and Recommendations, (Doc. 69), is ADOPTED IN

PART, REJECTED IN PART, and MODIFIED IN PART as reflected above. (2) The Center for Biological Diversity’s motion to supplement, (Doc. 43), is DENIED. (3) The parties’ cross-motions for summary judgment, (Docs. 45, 47, 54,

56), are GRANTED IN PART and DENIED IN PART. Summary judgment is GRANTED in favor of Plaintiffs for the following claims: a. Plaintiffs’ NEPA claim based on temporary roads;

b. Plaintiffs’ NFMA claims; and c. Plaintiffs’ ESA claims based on patch size and road locations. Summary judgment is otherwise granted in favor of Defendants. (5) The South Plateau Project EA and Decision Notice are VACATED and REMANDED to the agency to address the deficiencies identified above. (6) The Clerk is directed to enter judgment consistent with this Order and close the case file. DATED this 11" day of December, 2025. United States [District Court