FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

CASCADIA WILDLANDS, an No. 24-4542 Oregon non-profit corporation; D.C. No. OREGON WILD, an Oregon non- 6:23-cv-01358- profit corporation, MC Plaintiffs - Appellants, v.

OPINION

UNITED STATES BUREAU OF

LAND MANAGEMENT, a federal agency,

Defendant - Appellee.

Appeal from the United States District Court for the District of Oregon Michael J. McShane, Chief District Judge, Presiding Argued and Submitted June 10, 2025 San Francisco, California Filed August 27, 2025

Before: SIDNEY R. THOMAS and MILAN D. SMITH, JR., Circuit Judges, and DOUGLAS L. RAYES, District Judge. *

* The Honorable Douglas L. Rayes, United States District Judge for the District of Arizona, sitting by designation. Opinion by Judge Milan D. Smith, Jr.

SUMMARY **

Environmental Law

The panel affirmed the district court’s summary judgment in favor of the U.S. Bureau of Land Management (BLM) in an action brought by environmental groups challenging BLM’s approval of the Big Weekly Elk Forest Management Project (BWE Project). Plaintiffs alleged that (1) the BWE Project violated an earlier resource management plan governing coastal Oregon forests, so its approval violated the Federal Land Policy and Management Act (FLPMA); and (2) BLM failed to take a “hard look” at the BWE Project’s environmental impacts as required by the National Environmental Policy Act (NEPA). The Northwestern and Coastal Oregon Resource Management Plan (the RMP) covers over 1.2 million acres of BLM-administered lands. The RMP provides that all habitat for the threatened murrelet bird that had been designated as occupied under the earlier Northwest Forest Plan would be designated as “Late-Successional Reserve” (LSR), which is managed with the objective of maintaining existing nesting habitat for the marbled murrelet and the northern spotted owl, as well as promoting the development of additional nesting habitat for those

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

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 3

species. Approximately 65% of the BWE Project lands fall within the LSR. The RMP sets forth resource programs that include management directions for the marbled murrelet (the Murrelet Management Direction). The panel held that Kisor v. Wilkie, 588 U.S. 558 (2019), provides the applicable framework for when to defer to an agency’s construction of a land use plan, and therefore analyzed the RMP under its strictures. To defer to an agency’s interpretation of its own regulations, three criteria must be met: the regulation is genuinely ambiguous, the interpretation is reasonable, and the interpretation is entitled to controlling weight. The panel held that the term “modifying nesting habitat” in the Murrelet Management Direction was genuinely ambiguous. The panel also held that BLM’s narrow interpretation was reasonable. Finally, the character and context of BLM’s interpretation indicated that it was entitled to deference where BLM’s interpretation represents its official position, BLM’s interpretation depends on its substantive expertise, and BLM’s interpretation represents its fair, considered judgment. Applying BLM’s narrow interpretation of “modifying nesting habitat” in the Murrelet Management Direction to the BWE Project, the panel concluded that the BWE Project fully conformed to the RMP. The approval of the BWE Project was thus not arbitrary and capricious, and there was no violation of FLPMA. Concerning whether BLM took a “hard look” at the environmental consequences of the BWE Project as required by NEPA, the panel concluded that BLM did not act arbitrarily or capriciously in preparing the Environmental Assessment (EA) and Finding of No Significant Impact (FONSI) for the BWE Project. In an appendix to the EA, BLM explained the possible impacts of the project on murrelets, including how thinning the forest pursuant to the project could expose murrelets to possibly deleterious edge effects. It concluded that the project was likely to ultimately benefit the murrelet and that certain aspects of the project would minimize the costs of edge effects. BLM concluded that it was not required to assess the impacts to murrelets in more detail because, inter alia, the possible impacts to murrelets had already been thoroughly addressed in the Environmental Impact Statement (EIS) for the RMP—a document incorporated by reference and to which the EA tiered. The EA also incorporated by reference the extensive discussions in the BWE Project biological assessment (BA) and in the correspondence between the Fish and Wildlife Service and BLM. The panel held that this constituted a “hard look” at the environmental impacts of the BWE Project on murrelets, and rejected plaintiffs’ challenges to the EA.

COUNSEL

Nicholas S. Cady (argued), Cascadia Wildlands Project, Eugene, Oregon; John S. Persell, Oregon Wild, Portland, Oregon; for Plaintiffs-Appellants. Kyle Glynn (argued), Shannon Boylan, and Joan Pepin, Attorneys, Environment & Natural Resources Division; Todd Kim, Assistant Attorney General; United States Department of Justice, Washington, D.C.; Brian Perron, Attorney, United States Department of the Interior, Portland, Oregon; for Defendant-Appellant.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 5

OPINION

M. SMITH, Circuit Judge:

Environmental groups Cascadia Wildlands and Oregon Wild (collectively, the Plaintiffs) appeal the district court’s grant of summary judgment in favor of the U.S. Bureau of Land Management (BLM). The Plaintiffs brought a civil action challenging BLM’s approval of the Big Weekly Elk Forest Management Project (BWE Project), raising claims pursuant to the Federal Land Policy and Management Act (FLPMA), 43 U.S.C. §§ 1701–1787, and the National Environmental Policy Act (NEPA), 42 U.S.C. §§ 4321– 4370m-12. Specifically, they alleged that (1) the BWE Project violated an earlier resource management plan governing coastal Oregon forests, so its approval violated FLPMA; and (2) BLM failed to take a “hard look” at the BWE Project’s environmental impacts as required by NEPA. The parties filed cross-motions for summary judgment, and the district court granted summary judgment in favor of BLM. We affirm. For the reasons given below, we conclude that the BWE Project does not violate the resource management plan, so there is no violation of FLPMA. We also conclude that BLM took the “hard look” required by

NEPA.

STATUTORY BACKGROUND

The issues that we address are buried amidst a tangle of overlapping statutes and regulations. We begin our trek by discussing four areas of law that are crucial to understanding the Plaintiffs’ claims, as well as the factual and procedural history of this case. This case involves a claim that the BWE Project violates FLPMA. FLPMA “establishes requirements for land use planning on public land” and requires that BLM “‘develop, maintain, and when appropriate, revise land use plans’ to ensure that land management be conducted ‘on the basis of multiple use and sustained yield.’” Or. Nat. Res. Council Fund v. Brong, 492 F.3d 1120, 1125 (9th Cir. 2007) (quoting 43 U.S.C. §§ 1701(a)(7), 1712(a)). 1 “[L]and use plans are a

1 “‘Multiple use’ and ‘sustained yield’ are both technical terms.” Or. Nat. Desert Ass’n v. BLM, 625 F.3d 1092, 1096 n.2 (9th Cir. 2010). “The term ‘sustained yield’ means the achievement and maintenance in perpetuity of a high-level annual or regular periodic output of the various renewable resources of the public lands consistent with multiple use.” Id. (quoting 43 U.S.C. § 1702(h)). For its part, multiple use is defined to mean: the management of the public lands and their various resource values so that they are utilized in the combination that will best meet the present and future needs of the American people; making the most judicious use of the land for some or all of these resources or related services over areas large enough to provide sufficient latitude for periodic adjustments in use to conform to changing needs and conditions; the use of some land for less than all of the resources; a combination of balanced and diverse resource uses that takes into account the long-term needs of future generations for renewable and nonrenewable resources, including, but not limited to, recreation, range, timber, minerals, watershed, wildlife and fish, and natural scenic, scientific and historical values; and harmonious and coordinated management of the various resources without permanent impairment of the productivity of the land and the quality of the environment with consideration being given to the relative values of the resources and not necessarily to

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 7

preliminary step in the overall process of managing public lands—‘designed to guide and control future management actions and the development of subsequent, more detailed and limited scope plans for resources and uses.” Norton v. S. Utah Wilderness All., 542 U.S. 55, 69 (2004) (quoting 43 C.F.R. § 1601.0–2 (2003)). Thus, a land use plan—such as a resource management plan—generally does not authorize on-the-ground actions; rather, such actions are authorized through site-specific implementations. See 43 U.S.C. § 1712(e); 43 C.F.R. § 1601.0-5(n). In keeping with this structure, “FLPMA requires the government to ‘manage the public lands . . . in accordance with the land use plans . . . when they are available.” Mont. Wildlife Fed’n v. Haaland, 127 F.4th 1, 42 (9th Cir. 2025) (alterations in original) (quoting 43 U.S.C. § 1732(a)). “Once a plan is adopted, ‘[a]ll future resource management authorizations and actions . . . and subsequent more detailed or specific planning, [must] conform to the approved plan.’” Id. (alterations in original) (quoting 43 C.F.R. § 1610.5- 3(a)). “‘The statutory directive that BLM manage “in accordance with” land use plans, and the regulatory requirement that authorizations and actions “conform to” those plans, prevent BLM from taking actions inconsistent with the provisions of a land use plan.’” Id. (quoting Norton, 542 U.S. at 69). Accordingly, “[a]ny BLM land use decision contrary to the plan ‘can be set aside as contrary to law pursuant to 5 U.S.C. § 706(2).’” Id. (quoting Norton, 542 U.S. at 69); see also Brong, 492 F.3d at 1135 (setting aside a site-specific project because it was at odds with an earlier

the combination of uses that will give the greatest economic return or the greatest unit output. 43 U.S.C. § 1702(c). resource management plan). That is the crux of the Plaintiffs’ first claim—that the BWE Project violates an earlier resource management plan and thus cannot stand.

II. NEPA

The Plaintiffs’ second claim arises under section 102 of NEPA, see 42 U.S.C. § 4332. A. Principles of NEPA Review “NEPA has ‘twin aims.’” Ctr. for Biological Diversity v. BLM, 141 F.4th 976, 993 (9th Cir. 2025) (quoting Kern v. BLM, 284 F.3d 1062, 1066 (9th Cir. 2002)). “It first requires that a federal agency ‘consider every significant aspect of the environmental impact of a proposed action.’” Id. (quoting Kern, 284 F.3d at 1066). “It then ensures that the agency will ‘inform the public that it has indeed considered environmental concerns in its decisionmaking process.’” Id. (quoting Kern, 284 F.3d at 1066). In short, NEPA requires the preparation of reports that inform the agency and the public of the environmental consequences of proposed projects. See Seven Cnty. Infrastructure Coal. v. Eagle Cnty., 605 U.S. ----, 145 S. Ct. 1497, 1510 (2025). NEPA “‘establishes “action-forcing” procedures that require agencies to take a “hard look” at environmental consequences.’” Kern, 284 F.3d at 1066 (quoting Metcalf v. Daley, 214 F.3d 1135, 1141 (9th Cir. 2000)). When “[p]roperly applied, NEPA helps agencies to make better decisions and to ensure good project management.” Seven Cnty. Infrastructure Coal., 145 S. Ct. at 1510. It is important, though, to acknowledge what NEPA does not do. As the Supreme Court recently emphasized, unlike other environmental laws such as the Endangered Species

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 9

Act or the Clean Water Act, NEPA “is a purely procedural statute” that “imposes no substantive environmental obligations or restrictions.” Id. at 1507. In other words, although NEPA requires the agency to analyze environmental impacts and prepare documents and make such analyses available for public inspection, “NEPA does not require the agency to weigh environmental consequences in any particular way.” Id. “Rather, an agency may weigh environmental consequences as the agency reasonably sees fit under its governing statute and any relevant substantive environmental laws.” Id. At bottom, then, NEPA does not require the agency to prioritize environmental concerns over other concerns in determining whether to proceed with a project; instead, all it requires is that the agency consider and disclose environmental impacts. “NEPA requires no more.” Id. at 1511 (quoting Strycker’s Bay Neighborhood Council, Inc. v. Karlen, 444 U.S. 223, 228 (1980)). The limited scope of NEPA also circumscribes the scope of judicial review. As the Supreme Court emphasized in Seven County Infrastructure Coalition, “the central principle of judicial review in NEPA cases is deference.” Id. “The ‘role of a court in reviewing the sufficiency of an agency’s consideration of environmental factors is a limited one.’” Id. at 1514–15 (quoting Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 555 (1978)). Courts cannot substitute their judgment for that of the agency. See id. at 1514 (“NEPA’s procedural mandate helps ‘to insure a fully informed and well-considered decision, not necessarily a decision . . . judges . . . would have reached had they been members of the decisionmaking unit of the agency.” (quoting Vt. Yankee, 435 U.S. at 558)). B. NEPA Procedure In terms of the procedure prescribed by NEPA, “[a]ny agency undertaking a ‘major Federal action[] significantly affecting the quality of the human environment’ must prepare an environmental impact statement (EIS).” Mont. Wildlife Fed’n, 127 F.4th at 20 (second alteration in original) (quoting 42 U.S.C. § 4332(C)). “To determine whether an EIS is required, an agency may first prepare an Environmental Assessment (EA).” Id. “If, after preparing the [EA], an agency determines that the action ‘will not have a significant effect on the human environment,’ then the agency makes a ‘Finding of No Significant Impact,’” also called a FONSI, and it “need not prepare an EIS.” Id. (quoting 40 C.F.R. § 1508.13 (1978)). For the sake of efficiency, a later EA can “tier” to an existing EIS in the right circumstances. See All. for the Wild Rockies v. USFS, 907 F.3d 1105, 1118–19 (9th Cir. 2018). ‘“Tiering’ is defined as ‘avoiding detailed discussion by referring to another document containing the required discussion,’ and, under [applicable] regulations, it is expressly permitted . . . .” Id. (citation omitted) (quoting Kern, 284 F.3d at 1073). “Tiering is appropriate when the sequence of statements or analyses is . . . [f]rom a program, plan, or policy environmental impact statement to a program, plan, or policy statement of lesser scope or to a site-specific statement or analysis.” 40 C.F.R. § 1508.28(a) (2020). 2 But

2 The Executive Branch issued an interim final rule removing the NEPA implementing regulations issued by the Council on Environmental Quality (CEQ) from the code of federal regulations. See Removal of National Environmental Policy Act Implementing Regulations, 90 Fed. Reg. 10610 (Feb. 25, 2025). We agree with the Plaintiffs that this has no bearing on the issues before us because BLM applied the regulations

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 11

tiering is inappropriate if the earlier documents contain only “general statements” about the disputed issues. See Klamath-Siskiyou Wildlands Ctr. v. BLM, 387 F.3d 989, 997–98 (9th Cir. 2004). In this case, these rules are applied in the context of NEPA review for a resource management plan and a subsequent implementing action. “Under BLM regulations, the agency must prepare an EIS when adopting a Resource Management Plan.” Mont. Wildlife Fed’n, 127 F.4th at 20 (citing 43 C.F.R. § 1601.0-6) (1983)). “It must also prepare an [EA] ‘for all proposed Federal actions’ not categorically excluded or ‘covered sufficiently by an earlier environmental document.’” Id. (quoting 43 C.F.R. § 46.300(a) (2008)). “‘When available,’ BLM ‘should use existing NEPA analyses for assessing the impacts of a proposed action and any alternatives.’” Id. (quoting 43 C.F.R. § 46.120(a) (2008)). In other words, BLM should tier to earlier NEPA review documents when feasible. See id.

III. Endangered Species Act Consultation

This case involves consultation between BLM and the U.S. Fish & Wildlife Service (FWS) pursuant to the Endangered Species Act (ESA), 16 U.S.C. §§ 1531–1544, although the Plaintiffs do not directly raise an ESA claim. Under section 7(a)(2) of the ESA, see 16 U.S.C. § 1536(a)(2), “agencies contemplating certain kinds of federal action are required to insure [sic] that the action they take ‘is not likely to jeopardize the continued existence’ or ‘result in the destruction or adverse modification of [critical] habitat’ of an endangered or threatened species.”

as they existed at the time in approving the BWE Project and resource management plan. Conservation Cong. v. Finley, 774 F.3d 611, 615 (9th Cir. 2014) (second alteration in original) (quoting Conservation Cong. v. USFS, 720 F.3d 1048, 1051 (9th Cir. 2013)). The action agency (here, BLM) “must prepare a ‘biological assessment’ to determine whether” species listed under the ESA or critical habitat for such species “‘are likely to be adversely affected’ by the proposed action.” Ctr. for Biological Diversity v. BLM, 698 F.3d 1101, 1107 (9th Cir. 2012) (quoting 50 C.F.R. § 402.12 (2012)). If so, BLM is required to consult with the appropriate wildlife agency (here, FWS) “to determine the likely effects of [BLM’s] proposed actions on endangered or threatened species.” Finley, 774 F.3d at 615. FWS must “prepar[e] a ‘biological opinion’ stating whether the proposed action, ‘taken together with cumulative effects, is likely to jeopardize the continued existence of listed species or result in the destruction or adverse modification of critical habitat.’” Id. (quoting 50 C.F.R. § 402.14(g)(4)). In short, the biological assessment (BA) involved in this case was prepared by BLM whereas the two biological opinions (BiOps) were prepared by FWS.

IV. O&C Act

Finally, much of the land at issue is subject to the Oregon and California Railroad and Coos Bay Wagon Road Grant Lands Act (O&C Act), 43 U.S.C. §§ 2601–34. The O&C Act directs BLM “to determine which portions of the land [granted by that act] should be set aside for logging and which should be reserved.” Murphy Co. v. Biden, 65 F.4th 1122, 1134 (9th Cir. 2023). The statute’s primary goal is “providing a permanent source of timber supply,” but the statute “delineates a number of purposes for the” granted lands. Id. (quoting 43 U.S.C. § 2601). Concordant with the

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 13

goals of the O&C Act, BLM has established the “allowable sale quantity” (ASQ) of timber, which is “an estimate of the volume of O&C timber that can be cut and sold in a given year without depleting the timberland.” Am. Forest Res. Council v. United States, 77 F.4th 787, 791 (D.C. Cir. 2023).

FACTUAL AND PROCEDURAL BACKGROUND

I. Factual Background

This case involves a number of administrative actions centering around timber harvesting and wildlife protection in coastal Oregon forests. From a bird’s-eye view, the Plaintiffs challenge BLM’s approval of the BWE Project, a land management action authorizing certain forms of logging in federally owned Oregon coastal forests. According to the Plaintiffs, the BWE Project runs afoul of the provisions of an earlier resource management plan that protects a species of seabird called the marbled murrelet. They also claim that the NEPA review for the BWE Project was insufficient because BLM failed to take a “hard look” at the project’s impacts on marbled murrelets. This dispute has its roots in administrative actions going back decades. To fully situate our analysis of this case, we discuss (A) marbled murrelets, (B) the Northwest Forest Plan, (C) the 2016 Northwestern and Coastal Oregon Resource Management Plan, (D) correspondence between BLM and FWS over the meaning of a provision in that resource management plan, and then (E) the BWE Project. A. Marbled Murrelets Marbled murrelets are small seabirds that can be found along the northern Pacific coast. They are listed as “threatened” under the ESA. FWS has also designated “critical habitat” for the marbled murrelet in Oregon. Unlike many other seabirds, marbled murrelets nest in inland forests. Marbled murrelets do not build traditional nests and instead lay their eggs on thick, flat moss-covered branches known as “platforms.” The requisite platforms are mostly found on trees in mature and old-growth forests. Murrelets have high “site fidelity,” which means that they tend to return to the same forest stand (and even the same tree) in different years. 3 According to BLM, marbled murrelet nesting habitat is “generally characterized by coniferous forest 80 years of age or older within 50 miles of the coast with multi-storied canopies, moderate canopy closure, containing large trees that have sufficient limb size and substrate [such as moss] to support nesting, flight accessibility, and protective cover from ambient conditions and potential avian predators.” Generally, suitable nesting habitat for marbled murrelets overlaps with nesting habitat for another ESA-listed species, the northern spotted owl. B. The Northwest Forest Plan From the mid-1990s through 2016, the relevant land use plan governing the lands at issue was the Northwest Forest Plan. The Northwest Forest Plan made certain forest stands available for timber harvest and allocated other forest stands to reserves, including the late-successional reserve. Generally, lands in the late-successional reserve were entitled to “heightened environmental protection,” Brong, 492 F.3d at 1123, and they were managed with the goal of conserving and enhancing habitat for endangered species, including the marbled murrelet and the northern spotted owl.

3 In general, a forest “stand” refers to “an aggregation of trees occupying a specific area managed as a discrete operational or management unit.”

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 15

The Northwest Forest Plan required pre-project surveying in suitable murrelet habitat that was proposed for timber harvest; if the surveys showed murrelet activity, the site would be designated as “occupied” and “all contiguous existing and recruitment habitat for marbled murrelets (i.e., stands that are capable of becoming marbled murrelet habitat within 25 years) within a 0.5-mile radius [would] be protected.” This land would be designated to the late-successional reserve, and timber harvest would generally be prohibited. See id. at 1126 (“Pursuant to these goals, the [Northwest Forest Plan] makes programmed ‘stand management’ activities, such as logging, impermissible in [late-successional reserves].”). Not all forests allocated to the late-successional reserves were ideal murrelet or northern spotted owl habitat; some needed silvicultural treatment and management to qualify as suitable nesting habitat. Thus, to meet long-term species goals, thinning of non-habitat was permitted around occupied sites as needed to protect or enhance existing or recruitment habitat. C. The Resource Management Plan In 2016, BLM adopted new resource management plans to govern the management of public lands in western Oregon consistent with FLPMA’s multiple-use and sustained-yield mandate. 43 U.S.C. § 1701(a)(7). At issue here is one specific plan, the Northwestern and Coastal Oregon Resource Management Plan (the RMP), which covers over 1.2 million acres of BLM-administered lands. The RMP has several purposes, including to “[p]rovide a sustained yield of timber” and to “[c]ontribute to the conservation and recovery of threatened and endangered species,” including the marbled murrelet. The RMP offers more sustained-yield timber production than the earlier Northwest Forest Plan. The RMP includes management objectives and management directions for specific land use allocations as well as for resource programs (which generally apply across land use allocations). The BLM defines management objectives as “descriptions of desired outcomes for BLM- administered lands and resources in [a resource management plan.]” A management direction, in turn, “identifies where future actions may or may not be allowed and what restrictions or requirements may be placed on those future actions to achieve the objectives set for the BLM- administered lands and resources.” 1. Land Use Allocations The RMP sets forth different land use allocations to accomplish its goals of, inter alia, timber production and endangered species conservation. Specifically, as relevant here, it allocates (1) 247,045 acres of land to the “Harvest Land Base” (HLB), and (2) 576,714 acres of land to the “Late-Successional Reserve” (LSR). Turning first to the HLB, land in that allocation is managed with the primary goal of sustainably producing timber for sale and contributing to the ASQ. Commercial thinning is permitted in the HLB for a variety of reasons, including improving merchantability and stand health, and heavier regeneration harvesting is also permitted. Whereas the HLB is managed primarily for timber production, the LSR is managed with more wildlife-oriented goals in mind. Specifically, land in the LSR is managed with the objectives of, inter alia, maintaining existing nesting habitat for the marbled murrelet and the northern spotted

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 17

owl, as well as promoting the development of additional nesting habitat for those species. To meet those objectives, management directions include: (1) protecting stands of older, structurally complex coniferous forest; (2) protecting marbled murrelet occupied stands; 4 and (3) applying silvicultural treatments to foster the growth of complex forests, especially for northern spotted owls. “[P]rotect marbled murrelet occupied stands means to prohibit activities in the occupied stand,” at least generally. 5 But certain activities “needed to protect the overall health of the [occupied] stand or adjacent stands” are permitted, “as long as the occupied stand continues to support marbled murrelet nesting.” So are “felling and removal of trees for habitat restoration” and “the construction or maintenance of linear and nonlinear rights-of-way, spur roads, yarding corridors, or other facilities, as long as the occupied stand continues to support marbled murrelet nesting.” Thinning within the LSR to promote the development of habitat (particularly for the northern spotted owl) is one of the goals of the RMP. The RMP provides that all murrelet habitat that had been designated as occupied under the earlier Northwest Forest Plan would be designated as LSR. Additionally, in the relevant portion of the RMP lands, occupied stands

4 “Marbled murrelet occupied stand refers to all forest stands, regardless of age or structure, within ¼ mile . . . of the location of marbled murrelet behavior indicating occupancy and not separated from the location of marbled murrelet behavior indicating occupancy by more than 328 feet of non-forest.” 5 The Plaintiffs assert that “protect marbled murrelet occupied stands” includes prohibiting activities in “adjacent stands.” But the Plaintiffs misread the management objective and associated directions, which focus on protecting the occupied stand itself. discovered after the RMP’s effective date would also be allocated to the LSR. Only timber harvested from the HLB contributes to the ASQ, although timber may also be harvested from the reserves. As to the timber to be harvested from the reserves, the RMP provides that “[t]he BLM will consider through monitoring and plan evaluation whether the implementation of management actions within the reserve allocations that produce non-ASQ timber volume is consistent” with the management objectives for the reserves. 2. Murrelet Management Direction The RMP also sets forth resource programs that apply to all types of land use allocations. These protect, among other things, wildlife resources and accordingly set forth specific management directions for certain species. At the heart of this case are a portion of the management directions for the marbled murrelet (the Murrelet Management Direction). The relevant language of the Murrelet Management Direction 6 provides that “[b]efore modifying nesting habitat or removing nesting structure in . . . all land use allocations within 35 miles of the Pacific Coast,” 7 BLM is required to “assess the analysis area for marbled murrelet nesting structure”—that is, assess the “analysis area” for trees with certain characteristics such as the existence of a platform

6 There are other portions of the Murrelet Management Direction not at issue here. For instance, a separate portion prohibits activities that “disrupt” marbled murrelet nesting at occupied sites.” No party contends that this portion is relevant here. 7 This direction also applies to certain land use allocations farther from the coast, but that is not relevant to this appeal, as the BWE Project takes place close to the coast.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 19

(regardless of actual occupancy). “The analysis area consists of the proposed project and lands within 726 feet of the project boundary” in consideration of potential edge effects, and it “includes all nesting structures that could be affected by habitat modification.” If the analysis area contains no such structure, no further consideration of murrelet habitat is required. If enough nesting structures are located in the analysis area, additional steps would need to be taken. Specifically, “[b]efore modifying forest stands in any 5-acre portion . . . of the analysis area that contains at least 6 trees with nesting structure,” BLM must choose from a menu of additional protective procedures to ensure murrelets are not harmed. Only one of these options is relevant here: Option One, which involves surveying and buffering. Under Option One, BLM must “[s]urvey for the marbled murrelet” in the analysis area. If no occupancy is found from the survey, no further consideration of marbled murrelet habitat is necessary. If, however, occupancy is determined, BLM cannot conduct activities within the “occupied stand”—defined as all forest stands within a quarter mile of the marbled murrelet behavior indicating occupancy, see supra n.4—and all forest within 300 feet of the occupied stand. There are limited exceptions to Option One for certain actions, such as felling hazard trees or constructing roads, so long as the stand continues to support nesting. 3. EIS BLM authored a comprehensive EIS for the RMP; this EIS included a whole section on marbled murrelets. The EIS concluded that the RMP would result in an increase in the amount of murrelet nesting habitat over the long term, even if there would be a decline in the short term. The EIS provided that surveys would be required prior to timber harvest in “nesting habitat in all land use allocations” within 35 miles of the Pacific Coast. The EIS further explained that the RMP “would protect lands within 300 feet . . . of forecasted, occupied site delineations.” It expressly explained that this coverage would extend to “forecasted” marbled murrelet sites. The EIS also considered habitat fragmentation and “edge effects”— deleterious impacts to nesting habitat (such as microclimate changes) arising from the creation of an “edge” by logging in adjacent forest. 4. ESA Consultation Pursuant to the ESA consultation procedure, FWS prepared a BiOp analyzing the impacts of the RMP on the marbled murrelet. FWS ultimately concluded that the RMP “is not likely to jeopardize the continued existence of the murrelet, and is not likely to destroy or adversely modify murrelet critical habitat.” In relevant part, FWS reasoned that “[a]lthough there are likely to be some adverse effects to murrelets and murrelet critical habitat . . ., the overall outcome of [RMP] implementation will be the protection of the vast majority of extant murrelet nesting habitat, and a large long-term net increase in total area and amount of murrelet habitat during the life of the plan.” It further noted that the RMP would provide for the survival and recovery of the murrelet in part because the RMP “will significantly minimize habitat modification by applying protective measures to activities in [the zone within 35 miles of the Pacific Coast] and to activities in the [LSR]” further inland. With respect to the Murrelet Management Direction, FWS stated that “BLM protection measures apply for

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 21

occupied stands (including stands with an unknown occupancy status) in all” coastal land use allocations. FWS specifically interpreted the “modify nesting habitat or remove nesting structure” language in a manner that would encompass edge effects from activities in adjacent stands:

To determine if protective measures apply, BLM will conduct an analysis of all activities that modify nesting habitat or remove nesting structure. Modify nesting habitat includes affecting adjacent stands that would modify the nesting structure[’]s wind firmness, microclimate and/or predation risks. . . . Although impacts from treating adjacent stands without nesting structure (referred to as buffer habitat) is expected within 300 to 600 feet from nesting structure, the analysis of habitat goes out 726 feet from the treatment boundary to properly identify low density nesting structure in younger stands . . . .

As to Option One, FWS explained that this option would ensure that timber harvest would not “remove occupied habitat” because “[i]f occupancy is detected,” a circle of protection and buffer is applied. FWS reiterated this reasoning later on, concluding that any timber harvest within the LSR “will be subject to protection of occupied habitat under the Wildlife Resource Program’s Management Direction for murrelets and will only be harvested if no adverse effects are anticipated to occupied murrelet habitat either within or adjacent to the harvest unit.” FWS also stated that “[b]ased on the [Murrelet Management Direction], timber harvest will not negatively impact murrelets in any [land use allocations] within 35 miles of the Pacific Coast.” 5. Record of Decision The RMP was approved in a Record of Decision issued by BLM. In large part, the Record of Decision reiterated what was already provided in the RMP and the accompanying EIS. As relevant here, it again explained that marbled murrelet occupied sites found after the issuance of the RMP would be reallocated to the LSR. The Record of Decision concluded that the RMP would contribute to the conservation and recovery of the marbled murrelet because it would protect older, structurally complex forests. It further provided that the RMP “will require pre-project surveys for marbled murrelets and the protection of occupied sites” up to 35 miles inland, but it did not delineate specifically when those surveys would be required. 8 D. Interpretive Memoranda As memorialized by agency memoranda, it soon became clear that BLM and FWS interpreted the Murrelet Management Direction differently. Specifically, the agencies disagreed about when the 300-foot buffer applied and when the direction’s protective requirement would be triggered by modifications that occurred outside of occupied sites. In June 2018, BLM explained that the term “modifying nesting habitat” in the Murrelet Management Direction

8 A panel of our court upheld the EIS for the RMP over a challenge brought by environmental groups based on NEPA and the ESA. See Pac. Rivers v. BLM, 815 F. App’x 107, 109 (9th Cir. 2020) (unpublished). The D.C. Circuit also upheld the RMP over a challenge by the logging industry. See Am. Forest Res. Council, 77 F.4th at 802.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 23

“refers to direct alteration of nesting habitat; that is, habitat- altering activities occurring within a polygon that BLM identifies as marbled murrelet habitat.” BLM explained that its “delineation of nesting habitat may contain an entire stand, multiple stands, or a portion of a stand, and is based on a site-specific evaluation of forest conditions.” Thus, “[t]he requirement to ‘assess the analysis area for marbled murrelet nesting structure’ is not triggered by actions outside of nesting habitat unless the action would remove nesting structure.” According to BLM, FWS’s BiOp for the RMP “mischaracterized” the Murrelet Management Direction insofar as it construed the direction as being triggered by modification to adjacent non-habitat. BLM admitted that “[i]ndirect effects on nesting habitat of actions outside of nesting habitat may still need to be addressed in NEPA compliance and ESA consultation.” FWS then responded with its own memo specifically addressing whether its murrelet BiOp conclusion “would be the same if the BLM conducts harvest activities (thinning and regeneration harvest) directly adjacent to stands mapped as occupied under the Northwest Forest Plan” and later incorporated in the LSR by the RMP. FWS first addressed the interpretive differences, indicating that although it had originally interpreted the Murrelet Management Direction in the RMP to provide for a 300-foot buffer around all occupied stands, BLM’s interpretation was that this direction applied only to “‘newly’ delineated occupied sites.” As indicated above, FWS in the RMP BiOp indicated that “modifying nesting habitat” would encompass affecting adjacent stands through edge effects and that it expected BLM to include a 300-foot buffer on all occupied habitat. Revisiting this interpretation, FWS acknowledged that BLM’s own modeling of ASQ harvest did not account for a 300-foot buffer around all occupied stands. It also acknowledged that BLM “intended modification” in the Murrelet Management Direction “to refer only to direct impacts on murrelet habitat” and not to “actions adjacent to an occupied stand [that] would modify habitat” through increased edge effects. FWS then analyzed impacts to murrelets under BLM’s interpretation. It concluded that marbled murrelets would be adversely impacted by activities on adjacent stands within 300 feet of occupied stands but that BLM’s interpretation would not change the outcome of the jeopardy analysis in its BiOp. FWS ultimately recommended that BLM adhere to FWS’s initial understanding (that the 300-foot buffer would apply for all occupied marbled murrelet sites, whether newly discovered or designated as LSR in the RMP and thus preclude action in stands adjacent to occupied sites), but it made clear that either interpretation would not change the no-jeopardy determination in FWS’s BiOp for the RMP. Specifically, FWS did a fresh effects analysis and section 7 ESA consultation under BLM’s understanding that logging could occur up to the edge of occupied sites identified under the Northwest Forest Plan without triggering requirements for surveying or buffering. FWS concluded that this new interpretation would lead to greater harvesting of forest adjacent to marbled murrelet habitat, which would have impacts on marbled murrelets by increasing edge effects, such as increasing predation levels and wind throw risks, and changing microclimates. Although such increased edge effects might adversely affect murrelets, FWS noted that there was some disagreement about the strength of the true impact of such effects. Ultimately, it concluded that its original no-jeopardy determination would still stand, as the thinning of adjacent forests could actually increase the total

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 25

habitat for marbled murrelets in the long term and because the thinned forests would be expected to maintain “some buffer function.” E. The BWE Project 1. Overview The BWE Project proposes forest management activities on BLM-administered lands in Coos County, Oregon, within 28 miles of the Pacific Coast—land that is already subject to the RMP. In a nutshell, the site-specific BWE Project authorizes thinning of forest in the LSR and heavier regeneration harvesting in the HLB, as well as “transportation management actions,” such as road construction. The BWE Project is supposed to be carried out through various sales between 2021 and 2026. Approximately 65% of the BWE Project lands fall within the LSR. BLM concluded that certain portions of the LSR need to be thinned to speed habitat development for the northern spotted owl (which, again, occupies similar habitat to the murrelet), as such treatment would improve tree diversity and canopy structure as well as reduce tree density. BLM thus targeted forest stands within the LSR that do not meet the desired habitat conditions for northern spotted owls. Generally, these stands are highly dense, plantation- style groupings of trees that BLM identified through various methods. Under the BWE Project, BLM will authorize heavy commercial thinning in certain of these stands, light commercial thinning elsewhere, and non-commercial thinning for older growth-forests without the requisite canopy features. Importantly, no treatment is proposed for LSR stands within the well-developed habitat for northern spotted owls or murrelets or within murrelet occupied sites. But although BLM does not propose LSR thinning “in stands with suitable murrelet habitat or in stands delineated as occupied habitat,” it does propose thinning “adjacent to these stands” with the stated goal of developing, in the long term, large contiguous areas of murrelet and northern spotted owl nesting habitat. In the HLB, the BWE Project proposes thinning and regeneration harvest. This harvest was authorized in part to meet the ASQ. Some of this harvest will take place in suitable murrelet habitat within the HLB, but BLM indicates that it will assess an analysis area and apply Option One in the Murrelet Management Direction when doing so. BLM will not harvest in areas that have been previously surveyed as occupied; instead, it will “assume[] that occupied sites continue to be occupied and [will] not re-survey[] those areas.” In other words, murrelet survey protocols under Option One of the Murrelet Management Direction will be conducted prior to the removal of trees with murrelet nesting platforms or the direct modification of murrelet nesting habitat. If the surveys indicate occupancy, the stand will be delineated as occupied per the RMP, and no treatments will occur in the occupied stand or within a 300-foot buffer. However, BLM will not assess an analysis area or apply Option One prior to thinning in non-habitat adjacent to existing nesting habitat or occupied sites. The BWE Project thus does not incorporate a 300-foot buffer around all occupied sites. The number of acres to be harvested or thinned adjacent to murrelet habitat and occupied sites is substantial. Indeed, if a 300-foot buffer were applied around all occupied sites, many of the LSR units in the BWE Project would be

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 27

eliminated from consideration. And some of the harvesting within the HLB also occurs within 300 feet of occupied, protected habitat. To mitigate impacts, the BWE Project includes project design features (PDFs) that will help to protect murrelets from edge effects and disturbances, such as seasonal and timing restrictions during murrelet breeding season and smaller, voluntary buffers not required by the Murrelet Management Direction. It bears noting that the parties frame the BWE Project in starkly different terms. According to BLM, the thinning in the LSRs will ultimately benefit the murrelet because it will, in the long term, result in the creation of diverse forests that support murrelet nesting. From the Plaintiffs’ perspective, the BWE Project is a thinly veiled attempt to maximize timber production and circumvent murrelet protections. 2. ESA Consultation a. BLM Biological Assessment As part of the ESA consultation process, BLM drafted a BA for the BWE Project. Again, BLM made clear that (1) “all suitable habitat within the HLB is in the process of being surveyed”; (2) it was not proposing thinning within the LSR “in stands with suitable murrelet habitat or in stands delineated as occupied habitat” but (3) that it would propose thinning adjacent to such stands. The BA drew a distinction between the RMP’s surveying and buffering requirements and the broader concept of alterations to adjacent stands that could possibly harm murrelets through edge effects. Specifically, the BA indicated that BLM interpreted the Murrelet Management Direction to be triggered only by activity that would directly modify nesting habitat or remove nesting structure. However, BLM used a broader definition including edge effects flowing from actions in adjacent stands when it evaluated whether the proposed timber sales would cause an adverse impact to the murrelet for purposes of section 7 of the ESA. Thus, for the purpose of deciding whether there would be an impact on murrelets and whether the PDFs would mitigate that impact (but not for delineating when surveys were required), BLM used a 300-foot edge effects calculation. The BA expressly referenced and “tier[ed] to” the interpretive memoranda between FWS and BLM. 9 In line with the other BWE Project documents, the BA indicated that BLM would not conduct regeneration harvests within occupied stands or within 300 feet of newly occupied stands, although no such buffer would be required adjacent to older occupied stands or suitable nesting habitat. However, to minimize edge effects, BLM indicated that it would apply the voluntary-buffer PDFs. b. FWS Biological Opinion FWS issued a BiOp stating that although the BWE Project would adversely impact listed species, including murrelets and northern spotted owls, the project was (1) not likely to jeopardize the continued existence of the murrelet or the northern spotted owl and (2) not likely to destroy or

9 Still elsewhere, the BA defined “modify habitat” as “to alter the conditions around murrelet nesting habitat reducing the nesting function of the habitat stand.” BLM indicated that it “refine[d] this definition into direct and indirect effects.” With respect to the latter, BLM observed that “[i]ndirect effects to adjacent stands can occur when the stand surrounding the nesting platforms is altered, which may increase predation and alter wind-firmness, temperature, and affect moss on branches.”

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 29

adversely modify critical habitat for the murrelet or the northern spotted owl. The BiOp incorporated BLM’s understanding of the Murrelet Management Direction as memorialized in the interpretive memoranda—that is, that the requirement to assess an analysis area and possibly implement Option One was not triggered by actions merely adjacent to nesting habitat or occupied stands. The BWE Project BiOp defined “[m]odify or treat/maintain habitat” as “to alter the conditions around murrelet nesting habitat reducing the nesting function of the habitat stand,” but noted that BLM had refined this definition based on “direct and indirect effects to habitat.” See supra n.9. The BWE Project BiOp acknowledged that some of the “assumptions and terminology” in the earlier BiOp for the RMP would be “applied differently within the [BWE Project].” Specifically, “[t]he proposed action [would] not apply 300-foot buffers to occupied sites in HLB known to occur prior to 2016.” Instead, BLM would voluntarily apply smaller PDF buffers based on the age and makeup of the harvest unit. FWS followed BLM’s methodology and divided modifications into “indirect” and “direct” effects. FWS acknowledged that the proposed actions could lead to increased edge effects that would be harmful for the murrelet, partially due to the lack of required buffers. But it concluded that there would be beneficial long-term effects through the development of stagnated adjacent stands into suitable nesting habitat. 3. EA for the BWE Project BLM produced an EA for the BWE Project that tiered to the EIS for the RMP. It analyzed a variety of alternatives, including the alternative ultimately chosen, which included road construction not included in the other alternatives. In general, the EA discussed the contours of the BWE Project, described which areas would be thinned and harvested, and comprehensively addressed the impacts on northern spotted owls. In an appendix to the EA, BLM addressed a number of issues that it concluded were not necessary to analyze in detail. This is where we find the primary analysis of the impact of the BWE Project on marbled murrelets. First, BLM indicated in the appendix that it was not required to address in detail the issue of how “the proposed management activities” would “affect marbled murrelet nesting habitat and/or marbled murrelet nesting structures” going forward. It reasoned that a detailed analysis was not necessary because it was not relevant to the “Purpose and Need” of the BWE Project and because the impacts would be the same as those addressed in the EIS prepared for the RMP. BLM reasoned that the analysis it conducted with respect to the impacts on spotted owls applied largely to murrelets, so it is likely that murrelets would benefit. Additionally, negative impacts would be minimal because of the PDFs it was applying, including buffers. It also explained that impacts would be minimal in the HLB because, where required, it was surveying for murrelets under Option One. Second, BLM indicated that it was not required to analyze in detail the impacts of “direct vegetation modification activities” on marbled murrelet occupied sites because that issue was already analyzed in the EIS prepared for the RMP. Additionally, no harvest was proposed within any occupied murrelet sites, and surveys were being conducted in the HLB when logging was conducted in units that qualified as suitable, but unoccupied, nesting habitat.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 31

BLM did not re-survey the occupied sites because it assumed that they remained occupied. Third, BLM indicated that it was not required to analyze in detail the question of how activities in adjacent stands would indirectly impact murrelet occupied sites and suitable nesting habitat. BLM admitted that although newly identified occupied murrelet sites would be designated and buffered per Option One, harvest activities would occur directly adjacent to stands that were designated as occupied at the time the RMP was issued and in unoccupied suitable nesting habitat. BLM ultimately concluded that, for several reasons, it was not required to analyze edge effects in detail. It reasoned that the BWE Project was within the scope of the RMP EIS already prepared and there would be no significant impacts beyond that. BLM also pointed to its correspondence with FWS in the interpretive memoranda, which indicated that even if logging occurred within 300 feet of occupied sites without surveying, there would be no jeopardy to the continued existence to the murrelet—and, in fact, implementation of the RMP would still result in an increased murrelet population. Moreover, BLM reasoned that the science around edge effects did not show a consistently strong impact to murrelets. Finally, BLM further noted that it was applying voluntary buffers and other PDFs to minimize edge effects “unless protocol surveys determine that the area is not occupied or if a biologist determines through field review that” edge effects would be unlikely due to forest structure. 4. FONSI Finally, BLM issued a FONSI for the BWE Project, which incorporated the EA by reference. The FONSI indicated that “there would not be a significant impact on the quality of the human environment from the no action alternative or the implementation of either of the action alternatives.” It also concluded that an EIS was not required and that “the effects of the proposed activities would be in conformance with the 2016 [RMP].” The FONSI itself did not mention murrelets, but an appendix thereto contained responses to questions about the project’s impacts on murrelets—some of which were filed by the Plaintiffs. In large part, BLM’s responses involved citations to the EA.

II. Procedural History

The Plaintiffs filed the operative First Amended Complaint against BLM on December 5, 2023. The First Amended Complaint challenged the approval of the BWE Project as well as the issuance of the EA and FONSI under NEPA and FLPMA. The parties filed cross-motions for summary judgment. The Plaintiffs argued, inter alia, that the BWE Project does not adhere to the Murrelet Management Direction because BLM refused to buffer occupied sites designated prior to the issuance of the RMP and because BLM failed to survey for marbled murrelets. According to the Plaintiffs, BLM’s interpretation of “modifying nesting habitat” runs counter to the language of the RMP, BLM’s own contemporaneous understanding, and the understanding of FWS. The Plaintiffs further argued that they were entitled to summary judgment because BLM violated NEPA by not preparing an EIS or taking a hard look at the effects of the BWE Project on marbled murrelets. In its cross-motion, BLM admitted that the BWE Project provided for timber harvest within 300 feet of sites known to be occupied but argued that the buffering responsibilities

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 33

under the Murrelet Management Direction apply only to newly discovered sites. As in the BWE Project documents and the interpretive memoranda, it interpreted “modifying nesting habitat” as applying only to actions that directly modify suitable nesting habitat, such as the logging of nesting habitat itself. BLM also argued that it had satisfied NEPA because the EA for the BWE Project was tiered to the RMP and because it took a hard look at impacts to murrelets. After a hearing, the district court granted BLM’s cross- motion for summary judgment and denied the Plaintiffs’ motion. First, the district court rejected the claim that BLM violated FLPMA because the BWE Project did not conform to the RMP. It reasoned that “BLM’s interpretation of ‘modifying nesting habitat’ is, at worst, a reasonable interpretation of an ambiguous term” and thus entitled to deference under the framework of Kisor v. Wilkie, 588 U.S. 558 (2019). It further found that, under BLM’s interpretation of the Murrelet Management Direction, the BWE Project did not violate the RMP because the BWE Project did not involve any actions “modifying nesting habitat or removing nesting structure,” so additional protective steps, like Option One, need not be taken. Second, the district court also rejected the Plaintiffs’ NEPA claims. As relevant here, it reasoned that the decision not to prepare an EIS was reasonable because the BWE Project was consistent with, and tiered to, the RMP. Likewise, BLM took the requisite “hard look” at the environmental impacts of the BWE Project and did not ignore edge effects. The district court entered judgment on July 3, 2024, and the Plaintiffs timely appealed. Because this case involves purported violations of NEPA and FLPMA, the district court had original jurisdiction pursuant to 28 U.S.C. § 1331. We have appellate jurisdiction—including over the denial of the Plaintiffs’ motion for summary judgment—pursuant to 28 U.S.C. § 1291. See McGregor v. Nat’l R.R. Passenger Corp., 187 F.3d 1113, 1114 (9th Cir. 1999). “We review de novo a district court’s grant or denial of summary judgment.” Siino v. Foresters Life Ins. & Annuity Co., 133 F.4th 936, 943 (9th Cir. 2025) (quoting Botosan v. Paul McNally Realty, 216 F.3d 827, 830 (9th Cir. 2000)). “Thus, on appellate review, we employ the same standard used by the trial court under Federal Rule of Civil Procedure 56(c).” Id. (quoting Animal Legal Def. Fund v. U.S. FDA, 836 F.3d 987, 988 (9th Cir. 2016) (en banc) (per curiam)). “Under that standard, ‘[s]ummary judgment is appropriate only if, taking the evidence and all reasonable inferences drawn therefrom in the light most favorable to the non- moving party, there are no genuine issues of material fact and the moving party is entitled to judgment as a matter of law.’” Id. (alteration in original) (quoting Furnace v. Sullivan, 705 F.3d 1021, 1026 (9th Cir. 2013)). “We review the BLM’s compliance with FLPMA and NEPA de novo.” Brong, 492 F.3d at 1124. “Decisions that allegedly violate NEPA and FLPMA are reviewed under the Administrative Procedure Act (‘APA’), which ‘dictates that we should “hold unlawful and set aside any agency action . . . [that is] arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”’” Id. at 1124–25 (alterations in original) (quoting Nat. Res. Def. Council v. Nat’l Marine Fisheries Serv., 421 F.3d 872, 877

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 35

(9th Cir. 2005)). Although the scope of this review is “narrow and a court is not to substitute its judgment for that of the agency,” the agency must nevertheless “examine the relevant data and articulate a satisfactory explanation for its action.” All. for the Wild Rockies, 907 F.3d at 1112 (first quoting Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983); then quoting Turtle Island Restoration Network v. U.S. Dep’t of Com., 878 F.3d 725, 732 (9th Cir. 2017)). “We will strike down an agency action as arbitrary and capricious ‘if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or if the agency’s decision is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.’” Id. (quoting Turtle Island, 878 F.3d at 732–33). Additionally, “‘[u]nexplained inconsistency’ between agency actions is ‘a reason for holding an interpretation to be an arbitrary and capricious change.’” Mont. Wildlife Fed’n, 127 F.4th at 36 (quoting Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981 (2005)).

ANALYSIS

This case presents two main questions: (1) whether the contours of the BWE Project, including its authorization of logging in stands adjacent to murrelet nesting habitat and occupied stands, conforms to the terms of the RMP, and (2) whether BLM took the “hard look” required by NEPA when it prepared the BWE Project EA. For the reasons given below, the answer to both questions is “yes,” and we affirm. I. Whether the BWE Project Violates the RMP The crux of the Plaintiffs’ first claim is that the BWE Project fails to conform to the RMP (and thus violates FLPMA, see Brong, 492 F.3d at 1125). According to the Plaintiffs, logging in stands adjacent to suitable marbled murrelet habitat—including logging in stands adjacent to occupied habitat—constitutes “modifying nesting habitat,” thus triggering the requirements of the Murrelet Management Direction to survey for murrelets and implement Option One. The Plaintiffs insist that because the BWE Project permits thinning in such adjacent stands without calling for application of the Murrelet Management Direction’s protective measures, the BWE Project violates the RMP. 10 In contrast, BLM insists that the BWE Project fully comports with the RMP because logging in adjacent stands is not “modifying nesting habitat” that would trigger the Murrelet Management Direction. BLM insists that only “direct alteration of nesting habitat” counts as “modifying nesting habitat,” and thus the obligations to assess the analysis area and survey for murrelets “is not triggered by actions outside of nesting habitat unless the action would remove nesting structure.” BLM does admit that, under the

10 In their reply brief, the Plaintiffs shift position and argue that the BWE Project authorizes thinning within murrelet nesting habitat because “nesting habitat” should be broadly interpreted to encompass younger stands. We conclude that this argument is forfeited because it was raised neither in the opening brief nor before the district court. See Lui v. DeJoy, 129 F.4th 770, 780 (9th Cir. 2025). Contrary to the Plaintiffs’ argument, this represents a substantial shift in position and fundamentally alters the contours of our analysis; it is not simply a response to an argument made by BLM. We thus decline to reach it and will consider only the Plaintiffs’ original argument.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 37

BWE Project, it is directly modifying murrelet habitat in the HLB but insists that it is surveying and buffering as required in that circumstance. Our analysis proceeds in two steps. First, we must (A) interpret the RMP to determine when the protective requirements of the Murrelet Management Direction must be carried out. See Brong, 492 F.3d at 1125 (“Because the [land use plan] embodies the substantive management directives with which the BLM must comply under FLPMA, our review must start with, and remain anchored in, an understanding of the [land use plan].”). Then we must (B) determine whether, under the proper interpretation, the BWE Project violates the requirements of the RMP. See id. at 1127–32. A. Interpreting the RMP In interpreting the RMP, the district court turned to the framework for deference provided by Kisor v. Wilkie, 588 U.S. 558 (2019). On appeal, the Plaintiffs do the same. We agree that Kisor provides the applicable framework for when we can defer to an agency’s construction of a land use plan such as a resource management plan, so we analyze the RMP under its strictures. 11

11 We have not always applied this precise framework for agency deference in the context of land use plans. For example, in Brong, we explained that “[t]hough we normally afford deference to an administrative agency’s interpretation of its own regulations, ‘an agency’s interpretation “does not control, where . . . it is plainly inconsistent with the regulation at issue.”’” 492 F.3d at 1125 (omission in original) (quoting Native Ecosystems Council v. USFS, 418 F.3d 953, 960 (9th Cir. 2005)). But this articulation of agency deference, which is derived from the Supreme Court’s early decision in Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 414 (1945), is imprecise after Kisor. In Kisor, the Supreme Court articulated the circumstances in which a court may defer to an administrative agency’s interpretation of its regulation. 12 Often, we do not need to resort to deference to interpret a regulation. “‘If [a] regulation is unambiguous and “there is only one reasonable construction of [the] regulation,” then we’ simply apply that meaning.” United States v. Cal. Stem Cell Treatment Ctr., Inc., 117 F.4th 1213, 1222 (9th Cir. 2024) (alterations in original) (quoting Mountain Cmtys. for Fire Safety v. Elliott, 25 F.4th 667, 675 (9th Cir. 2022)). “If the text seems to have more than one plausible meaning, then we must try to resolve the ambiguity by ‘carefully consider[ing] the text, structure, history, and purpose of [the] regulation.’” Id. (alterations in original) (quoting Kisor, 588 U.S. at 575). “If, after ‘exhaust[ing] all the “traditional tools” of construction,’ we determine that ‘the interpretive question still has no single right answer,’ then we consider whether the agency’s interpretation is reasonable, and, if so,

See 588 U.S. at 568, 574; see also United States v. Castillo, 69 F.4th 648, 655–56 (9th Cir. 2023). 12 To be sure, Kisor was simply a refinement of an existing deference doctrine, commonly called Auer deference or Seminole Rock deference, which governed when courts defer to agency interpretations of their own regulations. See 588 U.S. at 563 (discussing Auer v. Robbins, 519 U.S. 452 (1997), and Seminole Rock & Sand Co., 325 U.S. 410 (1945)). For the sake of simplicity, and consistent with current practice, see, e.g., United States v. Trumbull, 114 F.4th 1114, 1117 (9th Cir. 2024), we refer to the currently governing scheme as “Kisor deference.” We note also that “[t]he Supreme Court’s decision in Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), overruling Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), ‘did not call Kisor into question . . ., so we continue to apply it.’” United States v. Yafa, 136 F.4th 1194, 1197 n.4 (9th Cir. 2025) (omission in original) (quoting Trumbull, 114 F.4th at 1118 n.2).

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 39

whether it is entitled to deference . . . .” Id. (alteration in original) (quoting Kisor, 588 U.S. at 575–76). In sum, for us to defer to an agency’s interpretation of its own regulation, three criteria must be met: “(1) the regulation is ‘genuinely ambiguous,’ (2) the interpretation is ‘reasonable,’ and (3) the interpretation is entitled to ‘controlling weight.’” United States v. Yafa, 136 F.4th 1194, 1197 (9th Cir. 2025) (quoting Kisor, 588 U.S. at 574–79). Applying this framework, we agree that this is one of the circumstances in which we defer to the agency’s construction of its own regulation. 1. Ambiguity We begin with ambiguity. For an agency interpretation to be entitled to deference, the regulation at issue must be “genuinely ambiguous.” Kisor, 588 U.S. at 574. “‘[B]efore concluding that a rule is genuinely ambiguous, a court must exhaust all the “traditional tools” of construction.’” United States v. Castillo, 69 F.4th 648, 655 (9th Cir. 2023) (quoting Kisor, 588 U.S. at 575). The Murrelet Management Direction provides that “[b]efore modifying nesting habitat or removing nesting structure in . . . all land use allocations within 35 miles of the Pacific Coast,” BLM is required to “assess the analysis area for marbled murrelet nesting structure” and, if enough structures are present, implement protective measures, including Option One. Everyone agrees that this case turns on the meaning of “modifying nesting habitat,” which is not expressly defined in the RMP. Perhaps unsurprisingly, both parties argue that the Murrelet Management Direction is unambiguous and supports only their reading. According to BLM, “‘modifying nesting habitat’ refers to direct alteration of nesting habitat; that is, habitat-altering activities occurring within a polygon that BLM identifies as marbled murrelet habitat.” Put differently, BLM views direct alterations to suitable nesting habitat, including occupied sites, to be the trigger that activates the rest of the protective measures in the Murrelet Management Direction. In contrast, the Plaintiffs take a much more expansive view of “modifying nesting habitat.” According to the Plaintiffs, “modifying nesting habitat” includes activities that take place outside the borders of suitable nesting habitat that would have indirect impacts—through edge effects—on the suitable nesting habitat. The Plaintiffs insist that this definition of “modifying nesting habitat” flows from the structure of the Murrelet Management Direction and its delineation of the relevant “analysis area.” We conclude that the Murrelet Management Direction is “genuinely ambiguous” on this point. See Kisor, 588 U.S. at 574. This is not a scenario where “there is only one reasonable construction of a regulation.” Id. at 575. After considering the “text, structure, history, and purpose,” id., of the RMP—all of the tools in our interpretive toolbox—we conclude that the Murrelet Management Direction is susceptible to more than one reasonable reading. Beginning with the text, the RMP does not expressly explain whether “modifying nesting habitat” encompasses only direct alterations of suitable nesting habitat—such as thinning of nesting habitat itself—or whether it sweeps more broadly to encompass indirect modifications flowing from projects taking place nearby. Despite how the Plaintiffs view it, we do not see the portion of the Murrelet Management Direction explaining the contours of the

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 41

“analysis area” to be decisive on this point. It is unclear from the text alone whether the “analysis area” is meant to elucidate the scope of “modifying nesting habitat.” Additionally, there is no mention of “adjacent stands” in the Murrelet Management Direction itself even though, elsewhere in the RMP, BLM mentioned when “adjacent stands” should be considered. And although the inclusion of the 300-foot buffer around occupied sites in the Murrelet Management Direction does indicate that BLM considered edge effect impacts to be a relevant concern, that does not answer the question of whether the term “modifying” was meant to encompass indirect edge effects flowing from thinning activities in adjacent stands. Shifting our focus to the RMP as a whole, we still find mixed signals about how far “modifying nesting habitat” extends. There are portions of the RMP favoring BLM’s narrow interpretation. That reading facilitates thinning in non-habitat adjacent to habitat, which accords with the management objectives for the LSR—namely, that BLM “[p]romote the development of nesting habitat for the marbled murrelet in stands that do not currently meet nesting habitat criteria.” It likewise accords with the LSR management directions to “apply silvicultural treatments to speed the development of northern spotted owl nesting- roosting habitat” and to “utilize integrated vegetation management,” including “thinning,” to promote the development of structurally complex forest and restore species habitat. Moreover, reading the protective measures of the Murrelet Management Direction to apply only when nesting structures are removed or when nesting habitat itself is directly modified tracks the management objectives and directions for the HLB. The focus of the HLB is on generating a sustained yield in timber—an RMP goal that must be considered alongside the goal of protecting murrelets. In modeling the ASQ from the HLB, although BLM factored in a 300-foot buffer around newly discovered occupied murrelet sites, it, as FWS later acknowledged, “assum[ed] that harvest would occur up to the edge of occupied sites identified under the Northwest Forest Plan.” That assumption accords with a narrow interpretation of the Murrelet Management Direction because such an assumption would be invalid if BLM were required to assess an analysis area and survey for murrelets anytime it proposed logging adjacent to an occupied site. Indeed, FWS admitted in the interpretive memoranda that the ASQ modeling was consistent with a narrower interpretation of “modifying nesting habitat.” On the other hand, certain parts of the RMP point to a broader interpretation of “modifying nesting habitat”—one more protective of murrelets. For instance, FWS’s BiOp for the RMP contemporaneously defined “modifying nesting structure” in the context of the Murrelet Management Direction broadly to encompass even indirect impacts from edge effects. Additionally, the RMP was designed to be more protective of murrelets than the Northwest Forest Plan, which protected only isolated fragments of habitat. Moreover, there are indications that, in the RMP, expansive surveying was expected. The structure of the RMP thus does not resolve the interpretive quandary. History, too, points in multiple directions. A review of the record reveals that the proper interpretation of this provision engendered disagreement even between federal agencies. As detailed above, FWS originally interpreted “modifying nesting habitat” to include modifications flowing from edge effects because, in part, it viewed the 300-

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 43

foot buffer to expand outwards from sites previously designated as occupied under the Northwest Forest Plan. Although FWS later hewed to BLM’s interpretation and conducted a fresh jeopardy analysis, the initial disagreement about the meaning of the phrase—as demonstrated in the agencies’ interpretive memoranda—is yet another indication of ambiguity. Finally, we look to the purpose of the Murrelet Management Direction. Again, we discern no clear guiding light. To be sure, the provision—and a good portion of the RMP in general—was designed to help protect habitat for the marbled murrelet and contribute to the recovery and growth of that species. But a land use plan, such as the RMP, is an exercise in balancing various goals. Those goals include ecological and environmental goals, such as helping to ensure the recovery of the marbled murrelet, as well as commercial and productivity goals—including the requirement to consistently produce timber that would meet the demands of the O&C Act. We thus do not see purpose as dissolving the ambiguity that hangs over the Murrelet Management Direction and the term “modifying nesting habitat” specifically. We are unpersuaded by the Plaintiffs’ arguments as to why the Murrelet Management Direction is unambiguous. We disagree with their position that the plain language and purpose of the Murrelet Management Direction indicates only one permissible reading. Additionally, the fact that FWS interpreted the Murrelet Management Direction broadly does not indicate that the RMP is unambiguous in the manner that the Plaintiffs insist—although it does deal a sharp blow to BLM’s position that the Murrelet Management Direction unambiguously requires a narrow reading. Their remaining arguments also fail. For example, they argue that there are indications in the record that BLM itself has applied a broader definition of “modify habitat” for purposes of biological assessments. But, as BLM explains, a BA aims to assess the distinct question of whether a listed species would be “adversely affected” by a project for purposes of the ESA; it does not purport to set management objectives for public lands as a resource management plan does. Indeed, BLM specifically contemplated that a broader view encompassing edge effects would apply for purposes of FWS consultation but not for purposes of interpreting the Murrelet Management Direction. And BLM makes clear elsewhere that for purposes of interpreting the RMP language, the narrower understanding governs. In sum, after “‘exhaust[ing] all the “traditional tools” of construction,’” Castillo, 69 F.4th at 655 (alteration in original) (quoting Kisor, 588 U.S. at 575), we are left with the conviction that the Murrelet Management Direction is genuinely ambiguous. 2. Reasonableness Even if a regulation is genuinely ambiguous, “the agency’s reading must still be ‘reasonable’” for it to be entitled to deference. Kisor, 588 U.S. at 575 (quoting Thomas Jefferson Univ. v. Shalala, 512 U.S. 504, 515 (1994). “In other words, it must come within the zone of ambiguity the court has identified after employing all its interpretive tools.” Id. at 576. In conducting this step of the analysis, the same considerations that came into play at the first stage of the inquiry—text, structure, history, and purpose—remain relevant. See id. We find BLM’s narrower interpretation to be within the permissible zone of ambiguity. Although the reasonableness

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 45

analysis is not coextensive with the ambiguity analysis, see id., many of the circumstances indicating that the Murrelet Management Direction is ambiguous also lead us to conclude that BLM’s interpretation is reasonable. In other words, the tools in our interpretive toolkit led us to conclude that “modifying nesting habitat” could be read narrowly or broadly. It follows for many of those same reasons that BLM’s narrow reading is reasonable. The Plaintiffs raise several arguments as to why BLM’s interpretation is unreasonable. We are unpersuaded. First, the Plaintiffs insist that this interpretation results in surplusage because, under BLM’s view, “modifying nesting habitat” and “removing nesting structure” mean the same thing. Not so. Even if “modifying nesting habitat” is restricted to direct modifications, it would still maintain a unique meaning because it would be possible to “modify[] nesting habitat” without removing a nesting structure. Likewise, it would be possible to “remov[e] nesting structure” without modifying suitable habitat. Second, the Plaintiffs suggest that BLM’s interpretation essentially results in the enactment of an alternative that was rejected during the NEPA analysis. But the key difference between the RMP and Alternative C is that, under Alternative C, when surveying under Option One, BLM would need to survey only “in stands 120 years old or older” rather than in all land use allocations. This is a difference in the scope of the surveys, not how the survey requirement is triggered. Alternative C also had a larger HLB, which is not implicated by BLM’s interpretation, and it also contemplated that BLM could resurvey sites and remove them from the LSR if they were found to be unoccupied. It is hard to see how BLM’s interpretation could result in the sub silentio adoption of Alternative C. Third, the Plaintiffs argue that BLM’s reading is impermissible because, as applied, it results in the taking of more murrelets than the RMP initially intended. But although the RMP does include broad language about protecting occupied sites, a close reading of the portions pointed to by the Plaintiffs shows that it did not anticipate that there would be no impact to murrelets whatsoever; rather, it anticipated that there would be no loss of occupied sites within 35 miles of the coast. That is consistent with BLM’s interpretation. 13 Fourth, the Plaintiffs broadly suggest that BLM’s reading is unreasonable because it results in outcomes that are insufficiently protective of murrelets. Again, we are unpersuaded. The RMP does not protect murrelets at all costs; rather, the protections given the murrelet must be understood in the context of the RMP’s multi-use nature. Given this reality, we cannot say that the lesser protections given to murrelets under BLM’s interpretation makes that interpretation unreasonable. In sum, then, we conclude that BLM’s narrow interpretation of the “modifying nesting habitat” language in the Murrelet Management Provision is reasonable. 3. Entitled to Deference The fact that BLM has advanced a reasonable interpretation of an ambiguous provision still does not end the inquiry. We will defer to BLM’s interpretation of the

13 The Plaintiffs point out that the BiOp for the RMP indicates that timber harvest will not negatively impact murrelets within 35 miles of the coast. True enough. But the BiOp was prepared by FWS pursuant to their original interpretation of the Murrelet Management Direction. When FWS re-analyzed the impacts to murrelets under BLM’s interpretation, it did identify that there could be impacts to murrelets.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 47

RMP only if it is “entitled to ‘controlling weight.’” Yafa, 136 F.4th at 1197 (quoting Kisor, 588 U.S. at 576). At this step of the Kisor analysis, we consider “the character and context” of the agency’s interpretation. Id. at 1199. “While this inquiry ‘does not reduce to any exhaustive test,’ Kisor instructs courts to consider whether the interpretation (1) constitutes the agency’s ‘official position, rather than any more ad hoc statement not reflecting the agency’s views,’ (2) implicates the agency’s ‘substantive expertise,’ and (3) reflects the agency’s ‘fair and considered judgment.’” Id. (quoting Kisor, 588 U.S. at 576–79). These three considerations are “‘especially important markers for identifying’ when deference is appropriate.” United States v. Trumbull, 114 F.4th 1114, 1120 (9th Cir. 2024) (quoting Kisor, 588 U.S. at 576–77). The district court pretermitted its analysis and did not reach this portion of the Kisor test, at least not explicitly. That was improper. As the Supreme Court explained in Kisor, “not all reasonable agency constructions of . . . ambiguous rules are entitled to deference.” 588 U.S. at 573. Before deferring to BLM’s interpretation of the Murrelet Management Direction, the district court should have analyzed whether “the character and context of the agency interpretation entitles it to controlling weight.” Id. at 576. However, given that we are reviewing de novo and can affirm the grant of summary judgment for any reason supported by the record, see Brown v. Arizona, 82 F.4th 863, 874 (9th Cir. 2023) (en banc), we may overlook the truncated nature of the district court’s analysis and reach the third step of the Kisor framework ourselves. 14 See Omnicare, Inc. v.

14 Framed differently, under the circumstances of this case, our de novo review renders the district court’s analytical misstep harmless. See UnitedHealth Grp., Inc., 629 F.3d 697, 717 (7th Cir. 2011) (“If the district court did misstep . . ., our de novo review should go far toward rectifying any errors.”); see also Deltoro-Aguilera v. United States, 625 F.3d 434, 438 (8th Cir. 2010) (“Because our review is de novo, it is of no moment that we analyze [the] claim without the benefit of full consideration by the court below.”). At bottom, the Kisor analysis seeks to answer a question of law— interpretation. And we are convinced that a remand to the district court to analyze this issue would be unnecessary and formalistic. Undertaking the analysis ourselves, we have little difficulty in concluding that the character and context of BLM’s interpretation indicates that it is entitled to deference. Indeed, all three of the “especially important markers” that an agency’s interpretation is entitled to controlling weight are present here. Trumbull, 114 F.4th at 1120 (quoting Kisor, 588 U.S. at 576). First, BLM’s interpretation represents its official position. The interpretation appears in public-facing documents related to the BWE Project—namely, the BWE Project BA and the BWE Project BiOp, as well as (at least implicitly) in the BWE Project EA. As explained above, the position is also, to at least some extent, baked into the structure of the RMP. Finally, the agency took this position in its interpretive memoranda that it exchanged with FWS,

Ratanasen v. Cal. Dep’t of Health Servs., 11 F.3d 1467, 1469 (9th Cir. 1993); cf. Singh v. Holder, 591 F.3d 1190, 1199 (9th Cir. 2010) (noting that any legal error by an immigration judge was rendered harmless by de novo review by the board of immigration appeals); see also Maydak v. SeaFirst, No. 96-35660, 108 F.3d 338, 1997 WL 75685, at *2 n.2 (9th Cir. Feb. 20, 1997) (unpublished).

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 49

which are referenced in the BWE Project documents. Considering these together, we are satisfied that the interpretation by BLM is the agency’s “‘authoritative’” position rather than an “ad hoc statement not reflecting the agency’s views.” Kisor, 588 U.S. at 577 (quoting United States v. Mead Corp., 533 U.S. 218, 257–59, 258 n.6 (2001) (Scalia, J., dissenting)). 15 Second, BLM’s interpretation depends on its substantive expertise. See id. at 577–78. As we explained at the outset, a resource management plan is prepared by BLM and balances multiple interests, including wildlife conservation. See Norton, 542 U.S. at 58–60; Brong, 492 F.3d at 1125. It is hard to conceive of a regulatory document that more clearly is derived from an agency’s substantive expertise. It is clear beyond peradventure that Congress would delegate interpretive power to BLM in this context. See Kisor, 588 U.S. at 578. The Plaintiffs suggest that it is FWS, not BLM, that has primary substantive expertise because this case involves murrelets. The Plaintiffs are mistaken. While FWS may indeed be the expert on murrelet conservation and what is necessary for murrelet recovery, this case involves the interpretation of a land use plan that seeks to balance many different uses, including murrelet conservation. This case does not fall outside the ambit of BLM’s expertise simply because the portion of the RMP at issue involves an endangered species that is also regulated by FWS. Third, BLM’s interpretation represents its fair, considered judgment. See id. at 579. This is not a situation

15 In their appellate briefing, the Plaintiffs do not contest that the interpretation represents BLM’s official position. where the agency presents a “‘post hoc rationalizatio[n] advanced’ to ‘defend past agency action against attack.’” Id. (alteration in original) (quoting Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 155 (2012)). To a meaningful degree, BLM’s interpretation is built into the structure of the RMP. And long before approval of the BWE Project, BLM made its interpretation clear in the memoranda between it and FWS. Nor is this a case where an agency provides an interpretation conflicting with its own previous interpretation. See id. Although BLM and FWS have disagreed, BLM has not altered its position on what the relevant language means. The Plaintiffs argue that BLM’s interpretation results in undue surprise because, until this litigation, it was unaware that BLM disagreed with FWS’s interpretation of the Murrelet Management Direction. We disagree. To be sure, that an agency’s interpretation creates unfair surprise can be a reason not to defer. See id. But that general principle is inapplicable here. A close inspection of the RMP and the BiOp would have revealed the interpretive disagreement lurking beneath the surface. Moreover, the “unfair surprise” that the Plaintiffs point to is of a different nature than the kind discussed in Kisor or its predecessors. See id. at 579 (discussing unfair surprise “to regulated parties”); Christopher, 567 U.S. at 155–56 (declining to defer to an interpretation that would provide “potentially massive” retroactive liability). We are not convinced that the fact that an interpretation is surprising to an interested party is a reason to decline to defer—particularly when, as here, there are numerous other indicia that the agency’s interpretation constitutes its fair, considered judgment and that the interpretation is entitled to controlling weight.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 51

In sum, then, after undertaking the Kisor analysis, we conclude that BLM’s narrow interpretation of “modifying nesting habitat” in the Murrelet Management Direction is a reasonable interpretation of a genuinely ambiguous term and that this interpretation is entitled to controlling weight. B. Application to the BWE Project The question now becomes whether the BWE Project conforms to the RMP, as properly understood. If not, we must set aside the BWE Project. See Mont. Wildlife Fed’n, 127 F.4th at 42; Brong, 492 F.3d at 1132. Once we view the Murrelet Management Direction through the lens of BLM’s interpretation, we need not linger long. The BWE Project fully conforms to the RMP. BLM is required to assess an analysis area and, if applicable, survey for murrelets if nesting structure would be removed or nesting habitat would be thinned (or otherwise modified). It is not required to assess an analysis area and survey when it merely modifies non-habitat adjacent to existing habitat and occupied sites. The actions that the BWE Project authorizes in both the LSR and HLB conform to this requirement. In the LSRs, no thinning will occur within any existing nesting habitat (or occupied sites). To be sure (and leaving aside the voluntary PDF buffers which serve to mitigate some edge effects), thinning will take place in stands adjacent to suitable habitat and sites previously designated as occupied. But, as explained, that is insufficient to trigger the obligation to survey and buffer under the Murrelet Management Direction because it does not constitute “modifying nesting habitat.” In the HLB, where logging occurs in suitable murrelet habitat, the required surveys are being conducted. And, BLM is not proposing to log in any sites that have been designated as occupied on the assumption that those sites remain occupied. 16 There is thus no violation of the terms of the Murrelet Management Direction by the thinning authorized by the BWE Project. The Plaintiffs suggest two other areas of conflict. First, they contend that the BWE Project authorizes the “take” of up to four murrelets, which they say is entirely at odds with the RMP. As noted above, we do not agree. As understood by BLM, the RMP required that there would be no loss of occupied sites in the coastal region—not that there would be no impact to murrelets whatsoever. That remains true under the BWE Project. Second, the Plaintiffs briefly assert that even if BLM’s interpretation is entitled to deference, the BWE Project is still at loggerheads with the RMP because the BWE Project authorizes the removal of murrelet nesting structures for purposes of road construction and yarding corridors. According to the Plaintiffs, this removal of nesting structure “triggers the murrelet management directions and requires the implementation of one of the protective options.” In

16 The Plaintiffs repeatedly assail BLM for assuming occupancy of areas previously designated as occupied as a means to bypass the strictures of the Murrelet Management Direction. We are unconvinced. Such an approach accords with the RMP and BLM’s interpretation. And we see BLM’s practice of assuming occupancy not as a means of avoiding surveying, but rather to delineate what sites are off-limits to harvesting. Despite the Plaintiffs’ fervor in arguing this point, we do not think it is of much importance in the overall scheme of the questions presented. In the same vein, the Plaintiffs conceded below and here that the 300-foot buffers contemplated in Option One do not necessarily apply to sites designated as occupied under the Northwest Forest Plan. We agree, although we note that is of limited utility in resolving this case.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 53

response, BLM contends that roadbuilding falls within an “exception” to Option One. The Plaintiffs disagree. We decline to reach this argument. “[A]n appellate court will not consider issues not properly raised before the district court.” Smith v. Marsh, 194 F.3d 1045, 1052 (9th Cir. 1999); see also In re Mercury Interactive Corp. Secs. Litig., 618 F.3d 988, 992 (9th Cir. 2010). This argument was not raised in any developed manner in the Plaintiffs’ motion for summary judgment. Tellingly, the district court did not pass on it. And, indeed, on appeal, the Plaintiffs’ opening brief never so much as mentions the relevant exception. We thus deem this argument waived and will not reach it. 17 See In re Mercury, 618 F.3d at 992 (“Although ‘no “bright line rule” exists to determine whether a matter [h]as been properly raised below,’ an issue will generally be deemed waived on appeal if the argument was not ‘“raised sufficiently for the trial court to rule on it.”’ ‘This principle accords to the district court the opportunity to reconsider its rulings and correct its errors.’” (citation omitted) (quoting Whittaker Corp. v. Execuair Corp., 953 F.2d 510, 515 (9th Cir. 1992))). Accordingly, the BWE Project is consistent with the RMP. The approval of the BWE Project is thus not arbitrary and capricious, and there was no violation of FLPMA. The district court properly granted summary judgment to BLM on the Plaintiffs’ FLPMA claims.

17 Even if the issue had been properly preserved, the Plaintiffs still fail to grapple with whether, under the Kisor framework, BLM’s interpretation of the exception to Option One represents a reasonable interpretation of an ambiguous regulation and would be entitled to controlling weight. II. NEPA Hard Look Claim With the FLPMA claim resolved, we turned to the Plaintiffs’ remaining claim—that BLM failed to take a “hard look” at the environmental consequences of the BWE Project as required by NEPA. As with the FLPMA claim, “[w]e review the BLM’s compliance with NEPA under the deferential ‘arbitrary and capricious’ standard of the Administrative Procedure Act.” Ctr. for Biological Diversity v. U.S. Dep’t of the Interior, 623 F.3d 633, 641 (9th Cir. 2010) (quoting 5 U.S.C. § 706(2)(A)). A. Contours of the “Hard Look” Analysis “We examine the EA with two purposes in 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.” 350 Mont. v. Haaland, 50 F.4th 1254, 1265 (9th Cir. 2022) (quoting Ctr. for Biological Diversity v. NHTSA, 538 F.3d 1172, 1215 (9th Cir. 2008)). “‘Federal agencies must undertake a “full and fair” analysis of the environmental impacts of their activities,’ and ‘NEPA imposes procedural requirements designed to force agencies to take a “hard look” at environmental consequences’ of their proposed actions.” Id. (quoting League of Wilderness Defs./Blue Mountains Biodiversity Project v. Connaughton, 752 F.3d 755, 762–63 (9th Cir. 2014)); see also Ctr. for Biological Diversity, 623 F.3d at 641 (“Under NEPA, ‘we must ensure that the agency has taken a “hard look” at the environmental consequences of its proposed action. . . . [W]e must defer to an agency’s decision that is “fully informed and well-considered.”’” (alterations in original)

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 55

(quoting Blue Mountains Biodiversity Project v. Blackwood, 161 F.3d 1208, 1211 (9th Cir. 1998))). “To satisfy the ‘hard look’ requirement, an agency must provide ‘a reasonably thorough discussion of the significant aspects of the probable environmental consequences.’” 350 Mont., 50 F.4th at 1265 (quoting NHTSA, 538 F.3d at 1194). “In reviewing an agency’s decision not to prepare an EIS, the arbitrary and capricious standard under the APA requires this court to determine whether the agency has taken a hard look at the consequences of its actions, based its decision on a consideration of the relevant factors, and provided a convincing statement of reasons to explain why a project’s impacts are insignificant.” Id. (citation modified) (quoting Barnes v. U.S. Dep’t of Transp., 655 F.3d 1124, 1132 (9th Cir. 2011)). “The statement of reasons is crucial to determining whether the agency took a hard look at the potential environmental impact of a project.” NHTSA, 538 F.3d at 1220 (citation modified) (quoting Blackwood, 161 F.3d at 1212). We find guidance in the Supreme Court’s recent decision in Seven County Infrastructure Coalition, 605 U.S. ----, 145 S. Ct. 1497. The Supreme Court emphasized that in determining whether a document “complied with NEPA, a court should afford substantial deference to the agency.” Id. at 1511–12. That includes deferring to the agency regarding what level of detail is required, what alternatives are feasible, and the scope of the environmental effects that the NEPA document will address. See id. at 1512–13. That is because “[w]hen assessing significant environmental effects and feasible alternatives for purposes of NEPA, an agency will invariably make a series of fact-dependent, context- specific, and policy-laden choices about the depth and breadth of its inquiry—and also about the length, content, and level of detail of the resulting” NEPA documents. Id. at 1513. “Courts should afford substantial deference and should not micromanage those agency choices so long as they fall within a broad zone of reasonableness.” Id. Accordingly, the role that the judicial branch plays in policing NEPA compliance is “a limited one.” Id. at 1515 (quoting Vt. Yankee, 435 U.S. at 555). Although Seven County Infrastructure Coalition involved a different NEPA posture from the case before us—it arose in the context of evaluating plaintiffs’ challenge to an EIS rather than an EA—we find its teachings fully applicable here. B. Application to the BWE Project EA Keeping those principles in mind, we conclude that BLM did not act arbitrarily or capriciously in preparing the EA and FONSI for the BWE Project. BLM took the required “hard look” at the BWE Project’s impacts on marbled murrelets. In an appendix to the EA, BLM explained the possible impacts of the project on murrelets, including how thinning pursuant to the project could expose murrelets to possibly deleterious edge effects. It concluded that the project was likely to ultimately benefit the murrelet and that certain aspects of it, such as the voluntary PDFs, would minimize the costs of edge effects. It also noted that in its judgment, “the science around edge effects does not show a consistent edge effect.” BLM concluded that it was not required to assess the impacts to murrelets in more detail because, inter alia, the possible impacts to murrelets had already been thoroughly addressed in the EIS for the RMP—a document incorporated by reference and to which the EA tiered. The EA also incorporated by reference the extensive discussions in the

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 57

BWE Project BA and in the correspondence between FWS and BLM. This was sufficient for BLM to discharge its requirement to take a “hard look” at the environmental impacts of the BWE Project on murrelets. The EIS for the RMP already addressed the possible impacts on murrelets at length. Documents discussing the possible edge effects of logging in stands adjacent to murrelet occupied sites were incorporated by reference. Particularly in light of the Supreme Court’s guidance in Seven County Infrastructure Coalition, we will not micromanage or unduly scrutinize the agency’s assessment as to the level of detail to provide on the murrelet issue. Some of the Plaintiffs’ arguments boil down to substantive disagreements with BLM’s ultimate conclusions. For instance, the Plaintiffs assert that it is a mischaracterization of the evidence to suggest that the thinning and regeneration harvesting would ultimately benefit the murrelet and outweigh the short-term impacts of edge effects. But NEPA does not set substantive environmental requirements; it requires only analysis and disclosure of potential environmental impacts. Moreover, whether the possible benefits to the murrelet from thinning will ultimately outweigh the increased edge effects is a quintessential matter of agency judgment that this Court cannot disturb on NEPA review so long as the agency considered the relevant information and based its conclusion on such information. See Ctr. for Biological Diversity v. Ilano, 928 F.3d 774, 783 (9th Cir. 2019) (“Plaintiffs take issue with the Forest Service’s conclusion. We conclude however, that the Forest Service considered relevant scientific data, engaged in a careful analysis, and reached its conclusion based on evidence supported by the record. Therefore, its decision was not arbitrary or capricious.”); Protect Our Cmtys. Found. v. Jewell, 825 F.3d 571, 584 (9th Cir. 2016) (observing that because NEPA is a procedural statute, the plaintiffs’ “underlying substantive disagreement” with the conclusions in an EIS was an unavailing argument). BLM met this standard here. The Plaintiffs point to several purported flaws with the EA that they contend indicate that BLM failed to take the requisite “hard look” at impacts to murrelets. 18 None of these arguments persuade. 1. Brevity of Discussion The Plaintiffs first point out that the BWE Project EA relegates its primary discussion of impacts on murrelets to an appendix. According to them, that is an indication that BLM failed to take a “hard look” at murrelet impacts. We disagree. The BWE Project EA adequately discusses the possible impacts on murrelets. The BWE Project EA tiered to the RMP EIS, which exhaustively discussed the impacts of the RMP and its land use allocations on murrelets. It incorporated by reference the extensive discussions in the BWE Project BA and in the correspondence between FWS and BLM. Additionally, the body of the EA substantially discusses impacts to northern spotted owls, which share

18 The Plaintiffs’ additional argument that the EA failed to address BLM’s novel “interpretive dance” to avoid the Murrelet Management Direction is unpersuasive and not worthy of extended discussion. As we indicated above, we conclude that what the Plaintiffs term an “interpretive dance” is a reasonable interpretation of an ambiguous regulation that is entitled to deference. Additionally, the EA incorporated the BA for the BWE Project, as well as the interpretive memoranda, which fully discussed this issue.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 59

habitat with the murrelet, and provides more detail about where logging will occur. Moreover, asserting that the EA discussed impacts to murrelets only in the appendix understates the detailed discussion that the appendix actually contains. “Brevity should not be mistaken for lack of detail. A relatively brief agency explanation can be reasoned and detailed . . . .” Seven Cnty. Infrastructure Coal., 145 S. Ct. at 1512. NEPA documents “need not meander on for hundreds or thousands of pages.” Id. An agency is not required to compose the Aeneid whenever it assesses a particular issue or decides that a particular issue does not warrant a full analysis. Nor are we convinced by the argument that because the BWE Project area is “particularly significant for marbled murrelets,” a more detailed analysis was required. The question of how detailed a report must be “requires the exercise of agency discretion—which should not be excessively second-guessed by a court.” Id. The agency acted within that discretion in concluding that although the BWE Project would doubtless have some impact on murrelets and it would consider those impacts, there did not need to be an exhaustively detailed discussion of the issue. That reality is particularly salient given the fact that EA tiers to the far more detailed EIS. 19

19 The Plaintiffs also fault BLM for not mentioning murrelets in its statements of reasons in the FONSI for the BWE Project. But the CEQ regulations provided that if the FONSI incorporates the EA, the FONSI “need not repeat any of the discussion in the assessment.” 40 C.F.R. § 1508.13 (2020). Moreover, the record shows that an appendix to the FONSI specifically responds to comments about the BWE Project’s potential impacts on the murrelets—some of which were filed by the Plaintiffs. 2. Site Specificity Second, and relatedly, the Plaintiffs argue that the discussion of murrelet impacts contained in the EA appendix fails the “hard look” test because its analysis is insufficiently site-specific. For example, the Plaintiffs take issue with the fact that the EA utilizes “rough acreage” rather than discussing each logging unit and that there is substantial discretion as to when and how BLM will apply the PDFs, including the voluntary buffers. This argument, too, fails. To be sure, NEPA documents should generally analyze impacts to imperiled species at a site-specific level rather than through an overly general lens. See Anderson v. Evans, 314 F.3d 1006, 1019–20 (9th Cir. 2002). Here, though, the EIS for the RMP contained a great deal of detail about impacts to murrelets. The EA built upon that and analyzed the impacts to murrelets in both the LSR and HLB. Reviewing the totality of the EA, the incorporated documents, and the EIS—and keeping in mind that the level of detail for a NEPA document is itself a matter within agency discretion—we are satisfied that BLM’s analysis was sufficiently site-specific. It was not required to conduct a parcel-by-parcel analysis of the issue or identify each individual stand where PDFs would be applied. 3. Potential Take of Murrelets The Plaintiffs next contend that BLM failed to take a “hard look” because the EA did not disclose that the BWE Project could result in the “take” of murrelet eggs or young at four sites. This is unpersuasive. The EA and the incorporated documents explained that there would be impacts on murrelets; they just did not list the specific number of sites at which eggs could be taken.

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 61

This was sufficient. The applicable regulations required only that BLM consider “[t]he degree to which the action may adversely affect an endangered or threatened species or its habitat” in an EA. 40 C.F.R. § 1508.27(b)(9) (2020). The Plaintiffs have pointed to no case law or regulation requiring the EA/FONSI to discuss the specific number of sites at which take could occur. Indeed, case law makes clear that “NEPA regulations direct the agency to consider the degree of adverse effect on a species, not the impact on individuals of that species.” Env’t Prot. Info. Ctr. v. USFS, 451 F.3d 1005, 1010 (9th Cir. 2006). That was certainly done here. 20 This is thus not a basis for concluding that BLM failed to take a hard look at impacts to the murrelet. 4. Alleged Misrepresentation of the Science of Edge Effects The Plaintiffs’ final NEPA argument is that BLM distorted the relevant science regarding edge effects when it discussed the impact of edge effects on the murrelet in the EA. The Plaintiffs specifically take issue with BLM’s discussion in the appendix to the EA about why it did not analyze in detail the edge effects that would be caused by thinning in stands adjacent to occupied sites. In the relevant portion of the EA appendix, BLM concluded that there were no significant edge effects beyond those evaluated in the RMP EIS and discussed by FWS in the interpretive memoranda. BLM then went on to say that, as discussed in the BA, “the science around edge effects does not show a consistent edge effect.” Both the relevant portion of the EA and the BA cited a number of scientific papers on the topic.

20 Additionally, as a factual matter, the EA does expressly disclose adverse impacts from regeneration harvest on some of the relevant sites even if it does not expressly use the word “take.” According to the Plaintiffs, BLM “gross[ly] mischaracteriz[ed]” the scientific literature relating to murrelet edge effects, and this runs afoul of its duty to present “high quality” data and “[a]ccurate scientific analysis.” See 40 C.F.R. § 1500.1(b) (2024). This argument fails. In essence, the Plaintiffs challenge BLM’s conclusions about the severity and existence of deleterious edge effects. This is a technical, scientific issue where deference to agency technical expertise is at its apogee. See Idaho Wool Growers Ass’n v. Vilsack, 816 F.3d 1095, 1107 (9th Cir. 2016); Native Ecosystems Council v. Weldon, 697 F.3d 1043, 1051 (9th Cir. 2012). That dooms the Plaintiffs’ claim. Although they nitpick issues with the studies cited by BLM—such as critiquing the methodology of one study and the geographic relevancy of another—they have not shown that BLM’s reliance on these studies was arbitrary and capricious. At a high level, the studies are relevant and generally support BLM’s conclusion about the uncertainty of the impacts of edge effects on murrelets. That is sufficient to end the inquiry, as we must “defer to agency decisions so long as those conclusions are supported by studies ‘that the agency deems reliable.’” Native Ecosystems Council, 697 F.3d at 1053 (emphasis omitted) (quoting N. Plains Res. Council, Inc. v. Surface Transp. Bd., 668 F.3d 1067, 1075 (9th Cir. 2011)). That is the case here. We have reviewed the record evidence alongside BLM’s conclusions and have concluded that BLM did not act arbitrarily or capriciously in reaching its conclusion regarding edge effects. “We are not to ‘act as a panel of scientists, instructing the agency, choosing among scientific studies, and ordering the agency to explain every possible scientific uncertainty.’”

CASCADIA WILDLANDS V. BUREAU OF LAND MGMT. 63

Lands Council v. McNair, 629 F.3d 1070, 1074 (9th Cir. 2010) (quoting Lands Council v. McNair, 537 F.3d 981, 988 (9th Cir. 2008) (en banc)). That is precisely what the Plaintiffs ask us to do here: don a lab coat, sharpen our peer- review pencils, and dissect BLM’s conclusion in the EA that the science does not show a consistent edge effect to murrelets from modifications to nearby stands. We decline that invitation. We may set aside agency action only if it is arbitrary and capricious—and that is an exacting standard that the Plaintiffs have not met here. Accordingly, we reject the Plaintiffs’ arguments that BLM failed to take the required “hard look” at impacts to murrelets in the EA for the BWE Project.

CONCLUSION

Emerging from the thicket of overlapping statutes, regulations, and administrative actions, we are satisfied that BLM did not act arbitrarily or capriciously. We defer to the agency’s reasonable interpretation of the ambiguous term “modifying nesting habitat” in its resource management plan. Under that interpretation, the BWE Project fully comports with the resource management plan. Additionally, we conclude that the EA and FONSI for the BWE Project satisfy NEPA. The district court thus properly granted summary judgment to BLM on the Plaintiffs’ FLPMA and NEPA claims, and it likewise properly denied the Plaintiffs’ motion for summary judgment.

AFFIRMED.