Opinion
CV 21-84-M-DLC
08-23-2023
ALLIANCE FOR THE WILD ROCKIES; and NATIVE ECOSYSTEMS COUNCIL, Plaintiffs, v. UNITED STATES FOREST SERVICE; RANDY MOORE, Chief of the U.S. Forest Service; MARY ERICKSON, Forest Supervisor for the Custer Gallatin National Forest; and LEANNE MARTEN, Regional Forester for Region 1 of the U.S. Forest Service, Defendants.
ORDER
Dana L. Christensen, District Judge United States District Court
Before the Court is Plaintiffs Alliance for the Wild Rockies and Native Ecosystems Council's (collectively “Plaintiffs”) Motion for Summary Judgment (Doc. 17) and Defendants Randy Moor, Mary Erickson, Leanne Marten, and the United States Forest Service's (collectively “Defendants”) Cross-Motion for Summary Judgment (Doc. 26). For the reasons discussed below, the Court grants in part and denies in part both motions.
BACKGROUND
This action pertains to the Greater Red Lodge Area Vegetation and Habitat Management Project (the “Project”). The Project authorizes logging and thinning activities on approximately 21,871 acres in Carbon County, Montana, on the Beartooth Ranger District of the Custer Gallatin National Forest (“CGNF”). FS039043. The purpose of the Project is “to reduce hazardous fuels, maintain and/or improve resiliency of forest vegetation and grasslands, enhance aspen habitat, and improve water quality.” FS039083; see also FS039044-45. The Project is within the wildland-urban interface as identified in the Pre-Disaster Mitigation and Carbon County Community Wildfire Protection Plan (the “Carbon County Plan”). FS039043. The Project is also located within the Rock Creek and Rosebud Lynx Analysis Units (“LAUs”). FS039096.
The Forest Service released its first Environmental Impact Statement (“EIS”) and two Records of Decision (“RODs”) for the Project in 2015. See FS039083; see also FS000442-1261; FS000001-72; FS000073-93. This Court preliminarily enjoined the Project in 2016 until the agency completed reconsultation with the U.S. Fish and Wildlife Service (“FWS”) that incorporated the Northern Rockies Lynx Management Direction (“NRLMD”). All. for the Wild Rockies v. Marten, No. CV-15-99-M-BMM, 2016 WL 6901264, at *6 (D. Mont. Nov. 22, 2016). The Forest Supervisor then withdrew the RODs in January 2017. FS039089 . After completing the required consultation, the Forest Service issued its Supplemental Environmental Impact Statement (“SEIS”) in June 2020, FS039079-187, followed by a new ROD in March 2021, FS039039-78. Plaintiffs now challenge the 2015 EIS and 2020 SEIS (collectively the “Project EIS”) and 2021 ROD under several environmental statutes. (Doc. 18 at 8-10.)
SUMMARY JUDGMENT STANDARD
This Court can resolve an issue summarily if “there is no genuine dispute as to any material fact” and the prevailing party is “entitled to judgment as a matter of law.” FED. R. CIV. P. 56(a). Material facts are those which may affect the outcome of the case. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). A factual dispute is genuine when there is sufficient evidence for a reasonable factfinder to return a verdict for the other party. Id. If the moving party meets its initial responsibility, the burden then shifts to the opposing party to establish that a genuine issue of fact exists. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586 (1986).
STANDARDS OF REVIEW
I. Statutory Requirements
A. National Environmental Policy Act (“NEPA”)
NEPA requires federal agencies to prepare a detailed EIS for any “major Federal actions significantly affecting the quality of the human environment.” 42 U.S.C. § 4332(2)(C). Major Federal actions “include new and continuing activities, including projects and programs entirely or partly financed, assisted, conducted, regulated, or approved by Federal agencies; new or revised agency rules, regulations, plans, policies, or procedures; and legislative proposals.” 40 C.F.R. § 1508.18(a) (2020).Major Federal actions typically fall into one of four categories:
The Council on Environmental Quality adopted new NEPA regulations that became effective September 14, 2020, and were codified as of July 1, 2021. 85 Fed.Reg. 43,304, 43,357 (July 16, 2020) (codified at 40 C.F.R. § 1500-18 (2021)). Because the Forest Service's planning and decision in this case predate these changes, the Court applies the regulations in effect prior to those revisions.
(i) Adoption of official policy, such as rules, regulations, and interpretations adopted pursuant to the Administrative Procedure Act, 5 U.S.C. 551 et seq.; treaties and international conventions or agreements; formal documents establishing an agency's policies which will result in or substantially alter agency programs.
(ii) Adoption of formal plans, such as official documents prepared or approved by Federal agencies, which prescribe alternative uses of Federal resources, upon which future agency actions will be based.
(iii) Adoption of programs, such as a group of concerted actions to implement a specific policy or plan; systematic and connected agency decisions allocating agency resources to implement a specific statutory program or executive directive.
(iv) Approval of specific projects, such as construction or management activities located in a defined geographic area. Projects include actions approved by permit or other regulatory decision as well as Federal and federally assisted activities.Id. § 1508.18(b).
An EIS must provide a “full and fair discussion of significant environmental impacts,” and inform “decisionmakers and the public of the reasonable alternatives which would avoid or minimize adverse impacts or enhance the quality of the human environment.” Id. § 1502.1. NEPA does not, however, “mandat[e] that agencies achieve particular substantive environmental results.” Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 371 (1989). Instead, NEPA simply “ensures that the agency, in reaching its decision, will have available, and will carefully consider, detailed information concerning significant environmental impacts; it also guarantees that the relevant information will be made available to the larger audience that may also play a role in both the decision[-]making process and the implementation of that decision.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349 (1989). “If the adverse environmental effects of the proposed action are adequately identified and evaluated, the agency is not constrained by NEPA from deciding that other values outweigh the environmental costs.” Id. at 350.
B. Healthy Forests Restoration Act (“HFRA”)
HFRA directs the Forest Service “to reduce wildfire risk to communities, municipal water supplies, and other at-risk Federal land through a collaborative process of planning, prioritizing, and implementing hazardous fuel reduction projects” and to “enhance efforts to protect watersheds and address threats to forest and rangeland health, including catastrophic wildfire, across the landscape.” 16 U.S.C. § 6501(1), (3). To achieve these goals, the Forest Service is directed to “implement authorized hazardous fuel reduction projects” on “Federal land in wildland-urban interface areas.” Id. § 6512(a)(1). In satisfying HFRA, the Forest Service must also comply with NEPA. Id. § 6514(a)(1); see also WildWest Inst. v. Bull, 547 F.3d 1162, 1165 (9th Cir. 2008).
C. National Forest Management Act (“NFMA”)
NFMA mandates that the Forest Service “develop, maintain, and, as appropriate, revise land and resource management plans for units of the National Forest System, coordinated with the land and resource management planning processes of State and local governments and other Federal agencies.” 16 U.S.C. § 1604(a). Land and resource management plans-commonly referred to as forest plans-must “provide for multiple use and sustained yield of the products and services obtained” from individual forest units, and must “include coordination of outdoor recreation, range, timber, watershed, wildlife and fish, and wilderness.” Id. § 1604(e)(1). Forest plans must also “provide for diversity of plant and animal communities based on the suitability and capability of the specific land area in order to meet overall multiple-use objectives.” Id. § 1604(g)(3)(B). All projects planned within a forest unit must be consistent with the forest plan as well as any regulations in effect at the time of the decision. Native Ecosystems Council v. U.S. Forest Serv., 428 F.3d 1233, 1249 (9th Cir. 2005); see also 16 U.S.C. 1604(i).
II. Administrative Procedure Act (“APA”)
NEPA, HFRA, and NFMA claims are reviewed under the APA. Neighbors of Cuddy Mt. v. Alexander, 303 F.3d 1059, 1065 (9th Cir. 2002) (NEPA and NFMA); see also Alliance for the Wild Rockies v. Petrick, 68 F.4th 475, 491 (9th Cir. 2023) (explaining that where a statute, such as HFRA, “fails to provide a private right of action, judicial review of an agency action proceeds under the [APA]”) [hereinafter AWR v. Petrick]. Pursuant to the APA, the Court “shall . . . hold unlawful and set aside agency action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A).
“[R]eview under the arbitrary and capricious standard is narrow, and [the Court should] not substitute [its] judgment for that of the agency” whose decision is under review. Earth Island Inst. v. U.S. Forest Serv., 697 F.3d 1010, 1013 (9th Cir. 2012) (citations and internal quotation marks omitted). “[A]n agency's decision can be set aside only if the agency relied on factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Id. (citations and internal quotation marks omitted).
DISCUSSION
In their Motion for Summary Judgment (Doc. 17), Plaintiffs raise three claims: (1) Defendants violated NEPA by failing to take a hard look at the environmental effects of its revisions to the CGNF lynx habitat map, (Doc. 18 at 22); (2) Defendants violated HFRA, NEPA, and NFMA by relying on an erroneous delineation of the wildland-urban interface, (Id. at 29, 35); and (3) Defendants violated NEPA by failing to take a hard look at the cumulative impacts associated with the Palisades Timber Sale, (Id. at 38). Based on these claims, Plaintiffs ask that this Court vacate the Project EIS and ROD and enjoin the Project. (Id. at 4143.) Defendants cross-move for summary judgment on these claims and further argue that, should the issue of remedy arise, the Court should remand without vacatur. (Doc. 25 at 50.)
Plaintiffs' NFMA claim is separate from their HFRA and NEPA claim but relies on the same arguments regarding the wildland-urban interface. Therefore, the Court addresses them together.
I. Lynx Habitat Map
Plaintiffs argue that Defendants improperly tiered to Canfield (2016), thereby avoiding NEPA review of its decision to reduce mapped lynx habitat in the CGNF. (Doc. 18 at 22-29.) Defendants respond that Canfield (2016) was appropriately incorporated by reference and that the agency's decision to revise the CGNF lynx habitat map is consistent the NRLMD. (Doc. 25 at 21-29.)
In 2000, FWS listed the Canada lynx as a threatened species under the Endangered Species Act (“ESA”). FS041732. Following the listing, an interagency lynx biology team consisting of biologists from the Forest Service, FWS, Bureau of Land Management, and National Park Service developed the Lynx Conservation Assessment and Strategy (“LCAS”). FS041598. The LCAS recommended measures “intended to conserve the lynx, and to reduce or eliminate adverse effects from the spectrum of management activities on federal lands.” FS041681. These conservation measures “focuse[d] on areas where habitat could support resident populations and contribute to the long-term conservation of lynx.” FS041611.
The LCAS described the typical characteristics of lynx habitat but did not actually develop any maps of lynx habitat. See FS041610-13. Instead, the LCAS instructed that specific “national forests, BLM field offices, national parks, and wildlife refuges . . . should develop or refine maps of known lynx occurrence and potential lynx habitat.” FS041641. The LCAS also created LAUs to “provide analysis units of the appropriate scale with which to begin the analysis of potential direct and indirect effects of projects or activities on individual lynx, and to monitor habitat changes.” FS041678 . LAUs encompass both lynx habitat and non-lynx habitat, but the conservation measures generally only apply to lynx habitat within an LAU. FS041682. Using the LCAS criteria, the Forest Service developed lynx habitat maps for several national forests, including the Custer and Gallatin National Forests. FS040242 .
The LCAS described lynx habitat in the western United States as follows:
Most lynx occurrences (83%) were associated with Rocky Mountain Conifer Forest, and most (77%) were within the 1500-2000 m (4,920-6,560 ft) elevation zone (McKelvey et al 2000b). There is a gradient in the elevational distribution of lynx habitat from the northern to the southern Rocky Mountains, with lynx habitat occurring at 2,440-3,500 m (8,00011,500 ft) in the southern Rockies. Primary vegetation that contributes to lynx habitat is lodgepole pine, subalpine fir, and Engelmann spruce (Aubry et al. 2000). In extreme northern Idaho, northeastern Washington, and northwestern Montana, cedar-hemlock habitat types may be considered primary vegetation. In central Idaho, Douglas-fir on moist sites at higher elevations may be considered primary vegetation. Secondary vegetation that, when interspersed within subalpine forests, may also contribute to lynx habitat, includes cool, moist Douglas-fir, grand fir, western larch, and aspen forests. Dry forest types (e.g., ponderosa pine, climax lodgepole pine) do not provide lynx habitat.FS041611-12.
“LAUs do not depict actual lynx home ranges, but their scale should approximate the size of area used by an individual lynx” and are generally “6,500-10,000ha (16,000 - 25,000 acres or 25-50 square miles) in contiguous habitat,” or larger “in less contiguous, poorer quality, or naturally fragmented habitat.” FS041682-83.
At this time the Custer and Gallatin National Forests were two separate forests, but they have since been combined into the CGNF. FS040242.
In 2007, the NRLMD amended the forest plans of 18 national forests, including the CGNF, to add “management direction . . . [that] conserves and promotes recovery of Canada lynx, by reducing or eliminating adverse effects from land management activities on National Forest System lands, while preserving the overall multiple-use direction of existing plans.” FS041733. The NRLMD includes specific goals, objectives, standards, and guidelines that apply to “mapped lynx habitat on National Forest System land [in the Northern Rockies region] presently occupied by Canada lynx.” FS041732; see also FS041784-91.
The NRLMD describes lynx habitat as follows:
Lynx habitat occurs in mesic coniferous forest that experience cold, snowy winters and provide a prey base of snowshoe hare. In the northern Rockies, lynx habitat generally occurs between 3,500 and 8,000 feet of elevation, and primarily consists of lodgepole pine, subalpine fir, and Engelmann spruce. It may consist of cedar-hemlock in extreme northern Idaho, northeastern Washington and northwestern Montana, or of Douglas-fir on moist sites at higher elevations in central Idaho. It may also consist of cool, moist Douglas-fir, grand fir, western larch and aspen when interspersed in subalpine forests. Dry forests do not provide lynx habitat.FS041795. The NRLMD also explains that all lynx habitat in a national forest is considered “occupied” by lynx when: (1) “[t]here are at least two verified lynx observations or records since 1999 on the national forest unless they are verified to be transient individuals;” or (2) “[t]here is evidence of lynx reproduction on [the] national forest.” FS041817-18. The lynx habitat in the CGNF is considered occupied. See FS042267 .
The NRLMD includes a map of all occupied and unoccupied lynx habitat in the “Northern Rockies Lynx Planning Area.” FS042539. This map was created using the previously developed forest-level lynx habitat maps, such as those developed for the Custer and Gallatin National Forests. See FS040242 . However, the NRLMD discusses the expectation that “[d]uring site-specific project analysis, maps of lynx habitat [will] be reviewed and updated based on local information.” FS041913. The NRLMD also requires that changes to LAU boundaries “be based on site-specific habitat information” and be reviewed by the Forest Service Regional Officer. FS041784. There is no requirement that changes to mapped lynx habitat within an LAU be approved by the Regional Officer.
The LCAS was updated in 2013 and acknowledged that early mapping efforts that relied on the 2000 LCAS “mis-classified areas, either mapping areas that do not provide habitat for lynx as lynx habitat . . . or failing to identify areas that actually provide habitat for lynx.” FS042641. The 2013 LCAS also noted that “[i]n some areas, better information on identifying lynx habitat is available” and encouraged updating maps where “new vegetation databases will improve identification of lynx habitat.” FS042631. The 2013 LCAS also specified that “[l]ynx habitat mapping and the delineation of LAUs should be completed using criteria specific to each geographic area.” Id.
Consistent with the direction in the NRLMD and 2013 LCAS, the Forest Service has undertaken various efforts to revise its lynx habitat maps, including in 2013 and 2015. See FS040248 . Most relevant to this case, in 2016, the Forest Service published a white paper titled “Updating the lynx habitat map layer using the latest corporate standardized data and state-of-the-art GIS technology” (hereinafter “Canfield (2016)”). FS040241-58. Canfield (2016) used “new information and databases” to identify “potential lynx habitat consistent with the definition of lynx habitat as set forth in the [NRLMD].” FS040245. However, Canfield (2016) used a narrower elevation band-6,000 to 8,800 feet-than that included in the NRLMD's definition of lynx habitat-3,500-8,000 feet. FS040243 .
Canfield (2016)'s methodology produced a revised lynx habitat map that reduced lynx habitat in the Custer portion of the CGNF by 88,000 acres and increased lynx habitat on the Gallatin portion by 69,000 acres as compared to the original maps for these forests. See FS040248 . Thus, the Canfield (2016) map resulted in a net reduction of 19,000 acres of lynx habitat on the CGNF. Canfield (2016) explained that the primary difference “stems from eliminating the dry habitat types (dry Douglas fir, limber pine, whitebark pine) previously identified as lynx habitat on the Beartooth District,” which are not deemed to be suitable lynx habitat under the NRLMD and 2013 LCAS definitions. FS040255. Canfield (2016)'s lynx habitat map was further refined in 2018. FS039097 .
These revisions are not in the record before the Court but are referenced in the Project SEIS.
To assess the Project's effects on lynx and lynx habitat, the Forest Service first had to determine the extent of the Rosebud and Rock Creek LAUs and the extent and condition of lynx habitat within each LAU. For the 2020 SEIS, the agency relied on Canfield (2016)'s lynx habitat map and field verification that had occurred in 2016 to map the LAU boundaries and lynx habitat. FS039097; see also FS039100-01. The agency determined that the Rock Creek LAU encompasses 151,336 total acres with 32,561 acres of lynx habitat and the Rosebud LAU encompasses 160,101 total acres with 30,763 acres of lynx habitat. FS039100 .
Based on this delineation of lynx habitat, the agency went on to assess: (1) “[p]otential displacement of lynx from suitable habitat due to [P]roject implementation;” (2) “[a]cres of foraging habitat modified;” (3) “[a]cres of denning habitat modified;” and (4) “[c]onsistency with the NRLMD objectives, standards, and guidelines for vegetation management and connectivity.” FS039098. The agency concluded that: (1) “[p]otential displacement of lynx [from suitable habitat] would be temporary and occur in small areas along the eastern boundary of the LAUs annually,” FS039103; (2) lynx foraging would occur “unimpeded by project effects” in the majority of the LAU habitat, FS039104; (3) “[m]itigation measures . . . would minimize effects to current and future lynx denning habitat,” FS039107; and (4) the Project action is “consistent with the NRLMD's objectives, standards, and guidelines,” FS039107; see also FS039144-53.
NEPA's implementing regulations encourage agencies to “tier” to a prior EIS to “eliminate repetitive discussions of the same issues and to focus on the actual issues ripe for decision.” 40 C.F.R. § 1502.20 (2020). Tiering to an earlier EIS is appropriate where the subsequent EIS is either of “lesser scope” or is “at a later stage.” Cal. ex rel. Imperial Cnty. Air Pollution Control Dist. v. U.S. Dep't of Interior, 767 F.3d 781, 792 (9th Cir. 2014); see also 40 C.F.R. § 1508.28. NEPA's implementing regulations also direct agencies to incorporate by reference both NEPA and non-NEPA documents in order to “cut down on bulk without impeding agency and public review of the action.” 40 C.F.R. § 1502.21 (2020). The incorporated material must be cited, briefly described, and “made reasonably available for inspection by potentially interested persons within the time allowed for comment.” Id. “Ultimately, when reviewing for NEPA compliance, [the Court] look[s] to whether the agency performed the NEPA analysis on the subject action.” Alliance for the Wild Rockies v. U.S. Forest Serv., 907 F.3d 1105, 1119 (9th Cir. 2018) [hereinafter “AWR v. USFS”]; see also Cal. ex rel. Imperial Cnty. Air Pollution Control Dist., 767 F.3d at 792-95 (holding that the agency had merely incorporated an environmental report by reference because the necessary analysis was in the project EIS).
Plaintiffs argue that Canfield (2016) is an official policy or formal plan that has not undergone NEPA review and, therefore, is improperly relied upon in the Project EIS to delineate lynx habitat within the Project area. (Doc. 18 at 27.) Plaintiffs take the position that because Canfield (2016) used new criteria and data to map lynx habitat than was previously used, which resulted in a decrease in the acreage of protected lynx habitat in the Rock Creek and Rosebud LAUs, and the agency relied on this delineation of lynx habitat to approve the project, the agency must first subject Canfield (2016) to NEPA review. (Id. at 27-28.) The Court finds Kern v. U.S. Bureau of Land Mgmt., 284 F.3d 1062, 1066 (9th Cir. 2002), Native Ecosystems Council v. U.S. Forest Serv. ex rel. Davey, 866 F.Supp.2d 1209 (D. Idaho 2012) [hereinafter “Davey”], and AWR v. USFS, instructive on this issue.
In Kern, plaintiffs challenged a project EIS that referenced guidelines that had not been analyzed under NEPA, claiming that this constituted improper “tiering.” 284 F.3d at 1073. The Ninth Circuit had previously held that those guidelines were not subject to NEPA review at the time they were adopted because they “neither propose[d] any site-specific activity nor d[id] they call for specific actions directly impacting the physical environment,” but were more akin to “research, development, and information-gathering . . . intended to lay the groundwork for later decision making.” Northcoast Envtl. Ctr. v. Glickman, 236 F.3d 660, 667, 670 (9th Cir 1998). However, the Ninth Circuit had noted that once the guidelines were incorporated into a specific agency action, they would be subject to NEPA review in the context of that action. Id. at 670. That specific agency action was before the court in Kern and the Ninth Circuit held that the agency could not refer to or rely on-i.e., tier to-the guidelines in the project EIS because they had never been subjected to NEPA review. 284 F.3d at 1069, 1073.
In Davey, the District of Idaho addressed an issue nearly identical to the one now before this Court. Relying on Kern, the court held that the Forest Service improperly tiered to a revised lynx habitat map that removed eight LAUs from the forest without first reviewing the revised map under NEPA. 866 F.Supp.2d at 1227. The court declined to address whether the revised map should have been subjected to NEPA review at the time it was first adopted by the agency. Id. at 1227 n.10. Instead, the court held that “under Kern the Forest Service was required to conduct NEPA review of the [revised] map before using the map as a basis for approving the [project]” because the map constituted a formal plan that would “guide or prescribe alternative uses of Federal resources, upon which future agency actions will be based.” Id. at 1225. As the court explained, “[w]ithout the adoption of the [revised] map . . . the [p]roject area would have been subject to the restrictions contained in the [NRLMD].” Id. Because “the map was never subjected to independent NEPA review, which would have required an analysis of the potential affects the removal of the LAUs would have on the lynx, its habitat, and the habitat of the snowshoe hare,” the agency violated NEPA's procedural requirements. Id. at 1227.
In AWR v. USFS, the Ninth Circuit discussed both Davey and Kern as examples of improper tiering. 907 F.3d at 1120. In that case, plaintiffs claimed that a project EIS improperly tiered to a prior agency policy decision, which had not undergone full NEPA review, by “incorporat[ing] the [decision's] science and updated data,” which was “new and/or different science, or interpretation of science” than had been previously used. Id. at 1119-20. However, the Ninth Circuit ultimately held that the document had not been improperly tiered because the necessary NEPA analysis had occurred in the project EIS. Id. at 1120-21.
Because Canfield (2016) has not undergone NEPA review and is not an EIS, tiering to this document would be categorically improper under NEPA's implementing regulations; thus, Kern provides the relevant framework. Under Kern, the Forest Service could not rely on Canfield (2016)'s lynx habitat map without first reviewing Canfield (2016) under NEPA-either separately or as part of the Project EIS. In other words, the agency could still tier to Canfield (2016) if the agency were to provide the relevant NEPA review as part of the Project EIS.
In this case, like in Davey, the Forest Service adopted and relied on a revised lynx habitat map that removed areas within the project boundary from the NRLMD's protections. And, similar to AWR v. USFS, the science and data utilized in Canfield (2016) was new and different from that in the NRLMD. However, unlike AWR v. USFS, here, the Forest Service did not include the necessary analysis within the Project EIS itself. Because that analysis never occurred, the Court concludes that Canfield (2016) was improperly tiered.
Defendants attempt to distinguish this case from Davey on the basis that this case does not involve the removal of any LAUs, only a reduction in lynx habitat within LAUs resulting from improved data and methodology. (Doc. 25 at 29.) However, there is no meaningful distinction between revising a map to remove LAUs and revising a map in a manner that removes lynx habitat within an LAU because the NRLMD's protections only apply to those areas of an LAU that are occupied lynx habitat. The court's focus in Davey was on the removal of NRLMD protections for previously protected areas, which is what happened here as well.
Defendants also rely on this Court's decision in Alliance for the Wild Rockies v. Austin, 55 F.Supp.3d 1294 (D. Mont. 2014), for the premise that revisions to a forest's lynx habitat map are not subject to NEPA review. In that case, this Court upheld the agency's revisions to a lynx habitat map against challenges to the agency's mapping methodology. Id. at 1303-07. Here, Plaintiffs challenge the Project EIS on the basis that the remapping decision was improperly tiered, not that the methodology itself was flawed. Therefore, Austin does not support Defendants' position.
Defendants violated NEPA by failing to take a hard look at the environmental effects of its revisions to the CGNF lynx habitat map. Defendants argue that because the remapping was done pursuant to the NRLMD's directive, and the NRLMD underwent NEPA review, this is sufficient to satisfy NEPA's requirements. (Doc. 25 at 27.) The Court acknowledges that the NRLMD and 2013 LCAS specifically contemplate that lynx habitat maps should be revised at the project level using more accurate data as it becomes available. The Court also recognizes that the NRLMD does not require modifications to mapped lynx habitat to be approved by the Regional Officer, whereas such requirements exist for modifications to LAUs. However, following Defendants' logic, the Forest Service could continuously revise its lynx habitat maps from project to project, removing or adding lynx habitat with each revision, without subjecting those changes to NEPA review. To allow this outcome would create a loophole that allows the agency to circumvent the requirements of NEPA-the specific concern raised by the Ninth Circuit in Kern. Regardless of the directives found in the NRLMD and LCAS, the Court finds that this outcome is antithetical to the policy behind NEPA.
Because the Forest Service improperly tiered to Canfield (2016), the agency violated NEPA's procedural requirements. Plaintiffs' motion for summary judgment is granted on this claim.
II. Wildland-Urban Interface
Next, Plaintiffs argue that “[t]he Project EIS and ROD's claim that all of the proposed treatment units of the Project reside within the ‘wildland[-]urban interface' is based on misleading, inadequate, and incorrect mapping and information regarding the wildland urban interface,” because the agency improperly relied exclusively on the delineation of the wildland-urban interface in the Carbon County Plan in violation of HFRA and NEPA. (Doc. 18 at 29). Plaintiffs further claim that “the project will violate the [NRLMD] unless the agency can demonstrate that the Project occurs within the [wildland-urban interface],” and, because the CGNF Forest Plan incorporates the NRLMD's requirements, the agency thereby violated NFMA. (Id. at 37.) Defendants respond that “[a]ccording to the plain language of HFRA, legislative history, and in-District precedent, the [wildland-urban interface] is defined by the Community Wildfire Protection Plan.” (Doc. 25 at 29.) Defendants also argue that the Project meets the NRLMD's requirements regardless of whether the Project falls within the wildland-urban interface. (Id. at 42.)
A. Reliance on the Wildland-Urban Interface to Satisfy the NRLMD
The NRLMD includes exemptions for projects that fall within the wildland-urban interface. Specifically, modifications to potential lynx habitat are exempted from NRLMD vegetation standards VEG S1, VEG S2, VEG S5, and VEG S6 for up to 7,775 acres or six percent of lynx habitat on the national forest. FS039107; see also FS041752 . The Forest Service argues that the Project complied with the NRLMD's requirements without relying on the wildland-urban interface exemptions. (Doc. 25 at 42.) However, this argument is not supported by the record.
In Appendix C to the 2020 SEIS, the agency addresses the Project's compliance with each NRLMD requirement. The Forest Service's discussion in Appendix C makes clear that VEG S1, VEG S2, and VEG S5 are met without reliance on the wildland-urban interface exemption. VEG S1 requires that “[i]f more than 30 percent of the lynx habitat in an LAU is currently in a stand initiation structural stage that does not yet provide winter snowshoe hare habitat, no additional habitat may be regenerated by vegetation management projects.” FS041786 (footnotes omitted). The agency explains that this standard is met because only “15% of the Rosebud LAU and 4% of the Rock Creek LAU are in a stand initiation structural stage that does not yet provide winter snowshoe hare habitat.” FS039145. VEG S2 requires that “[t]imber management projects shall not regenerate more than 15 percent of lynx habitat on NFS lands within an LAU in a ten-year period.” FS041786 (footnotes omitted). The agency explains that this standard is met because “[n]one of the alternatives would regenerate more than 15 percent of potential lynx habitat on NFS lands within an LAU in a 10-year period.” FS039146. VEG S6 provides:
Vegetation management projects that reduce snowshoe hare habitat in multi-story mature or late successional forests may occur only:
1. Within 200 feet of administrative sites, dwellings, outbuildings, recreation sites, and special use permit improvements, including infrastructure within permitted ski area boundaries; or
2. For research studiesor genetic tree tests evaluating genetically improved reforestation stock; or
3. For incidental removal during salvage harvest(e.g. removal due to location of skid trails).
Exceptions 2 and 3 shall only be utilized in LAUs where Standard VEG S1 is met.FS041787 (footnotes omitted). VEG S6 also notes that “[t]imber harvest is allowed in areas that have potential to improve winter snowshoe hare habitat but presently have poorly developed understories that lack dense horizontal cover [e.g. neven age management systems could be used to create openings where there is little understory so that new forage can grow].” Id. The agency explains:
The intermediate thinning treatments in stem exclusion habitat would accelerate the development of a multi-story mature structural stage by allowing sunlight into the stand which facilitates the establishment of new tree seedlings under the overstory trees, approximately >40 years post project. The regeneration harvest would create openings where there is little understory so that new forage can grow 15-40 years post project.FS039147.
However, the agency's discussion of VEG S5 appears to rely on the wildland-urban interface exemption. VEG S5 states:
Precommercial thinning projectsthat reduce snowshoe hare habitat may occur from the stand initiation structural stage until the stands no longer provide winter snowshoe hare habitat only:
1. Within 200 feet of administrative sites, dwellings, or outbuildings; or
2. For research studies or genetic tree tests evaluating genetically improved reforestation stock; or
3. Based on new information that is peer reviewed and accepted by the regional level of the Forest Service, and state level of FWS, where a written determination states:
a. that a projectis not likely to adversely affect lynx; or b. that a projectis likely to have short term adverse effects on lynx or its habitat, but would result in long-term benefits to lynx and its habitat; or
4. For conifer removal in aspen, or daylight thinning around individual aspen trees, where aspen is in decline; or
5. For daylight thinning of planted rust-resistant white pine where 80% of the winter snowshoe hare habitatis retained; or
6. To restore whitebark pine.
Exceptions 2 through 6 shall only be utilized in LAUs where Standard VEG S1 is met.FS041786-87 (footnotes omitted). The agency's discussion does not explain how this standard is met other than to state:
All treatment units, including all precommercial thinning, are located in [the wildland-urban interface]. The mapping query identified 0.16 acres of stand initiation habitat in Unit 20T. The prescription for Unit 20T is a small patch clearcut that leaves approximately two-thirds of the unit untreated. Small trees would be left untreated. Up to 153 acres in the Red [sic] Creek LAU and 11 acres in the Rosebud LAU would be pre-commercially thinned. This is well within the 7,775 acres allowed in [wildland-urban interface] for the Custer portion of the Custer Gallatin National Forest.FS039146-47. This discussion does not address how the Project would meet VEG S5 other than to rely on the exemption. Moreover, the 2020 SEIS emphasizes the fact that the Project is within the wildland-urban interface and that “[n]o other vegetation project using the six percent exemption . . . has been implemented on the Custer portion of the CGNF.” FS039107.
The Court notes that the record contains a letter from Jodi L. Bush, Office Supervisor for the FWS Montana Field Office, dated November 24, 2020, in which she states that FWS “has reviewed the biological assessment and concurs with [the Forest Service's] determinations that the proposed action is not likely to adversely affect the threatened . . . Canada lynx, or designated critical habitat for Canada lynx.” FS039256-59. This determination may support the Forest Service's conclusion that VEG S5 is met under exception 3a listed above, without relying on the wildland-urban interface exemption. However, the letter was received several months after the final SEIS was published (July 2020) and its conclusion is not discussed in the SEIS or its Errata. Defendants did not address this in their briefing, but merely referred the Court to the discussion contained in Appendix C. Based on that discussion, the Court cannot conclude that VEG S5 is met without the wildland-urban interface exemption.
Based on the record before the Court, it appears that the Forest Service did rely on the Project's presence within the wildland-urban interface to meet NRLMD standard VEG S5. The agency explains that its determination of the wildland-urban interface is based solely on Carbon County's definition and determination of the wildland-urban interface found in the Carbon County Plan. FS039132. Therefore, the Court must address whether the agency erroneously determined that the Project falls within the wildland-urban interface as that term is defined by HFRA.
B. Designation of the Wildland-Urban Interface
Under HFRA, the wildland-urban interface is “an area within or adjacent to an at-risk community that is identified in recommendations to the Secretary in a community wildfire protection plan.” 16 U.S.C. § 6511(16)(A). Alternatively, where there is no community wildfire protection plan in effect, HFRA defines the wildland-urban interface as the area generally ranging from .5 to 1.5 miles from the boundary of an-risk community or the area adjacent to an evacuation route for an at-risk community. See id. § 6511(16)(B). HFRA defines “at-risk community” as either “an area . . . that is comprised of . . . an interface community” as defined by federal regulation, or “a group of homes and other structures with basic infrastructure and services . . . within or adjacent to Federal land.” Id. § 6511(1)(A)(i), (ii). Federal regulation defines an “urban wildland interface community” as an area “where humans and their development meet or intermix with wildland fuel.” 66 Fed.Reg. 751, 753 (Jan. 4, 2001) (emphasis added). The notice goes on to state that “[t]here are three categories of communities that meet this description:” (1) Interface Community: “where structures directly abut wildland fuels” with a development density of “usually 3 or more structures per acre” or “population density of 250 or more people per square mile;” (2) Intermix Community: “where structures are scattered throughout a wildland area” with a development density “rang[ing] from structures very close together to one structure per 40 acres” or “population density of between 28-250 people per square mile;” or (3) Occluded Community: “where structures abut an island of wildland fuels” with a development density “similar to those found in the interface community, but the occluded area is usually less than 1,000 acres in size.” Id.
The Ninth Circuit recently held that “[c]ommunity plans may well inform the Forest Service's analysis under HFRA. But reliance on a plainly overinclusive wildland-urban interface, without more, is the sort of ‘clear error of judgment' that arbitrary or capricious review is meant to prevent.” AWR v. Petrick, 68 F.4th at 494. While a community plan's definition of the wildland-urban interface need not “parrot HFRA's,” if a “community plan's definition of its wildland-urban interface-on its face-deviates from HFRA and likely results in a covered area beyond what Congress authorized . . . the Forest Service cannot properly rely on the . . . community plan-alone-to justify the categorical exclusion.” Id. (emphasis added).
In AWR v. Petrick, the challenged community wildfire protection plan “lacked any discussion of the HFRA definition of wildland-urban interface or ‘at-risk communities'” and also lacked “any discussion of interface communities or the relative location of Federal lands-i.e., the metrics for determining at-risk communities under HFRA.” 68 F.4th at 494. Rather, the plan provided its own “broader definition unmoored from the specifics of HFRA.” Id. The Ninth Circuit concluded that, “[u]ntethered from HFRA's more limited definitions,” the county's “broader definition may well sweep in more land than its HFRA counterpart.” Id. Thus, the Forest Service could not rely on the county's delineation of the wildland-urban interface, by itself, to justify use of HFRA's categorical exclusion. Id. at 495. However, the court left the door open for the agency to demonstrate that the project was, “in fact, within the wildland-urban interface as defined by HFRA.” Id.
In this case, Carbon County adopted a Community Wildfire Protection Plan and the Project falls within the Carbon County Plan's wildland-urban interface. FS039043. However, the Carbon County Plan's definition of the wildland-urban interface substantially deviates from HFRA's definition, such that it likely resulted in a covered area beyond what Congress authorized. Like the plan in AWR v. Petrick, the Carbon County Plan does not include HFRA's definitions of wildland-urban interface or at-risk community. Rather, the Carbon County Plan relies on the definition of wildland-urban interface found in the Montana code annotated, which is the “line, area or zone where structures and other human development meet or intermingle with undeveloped wildland or vegetative fuels.” FS010615. Notably absent from this definition is the qualifier that the wildland-urban interface be “within or adjacent to at-risk communities.”
The Carbon County Plan does include the definition of “interface community” as provided by federal regulation-“where humans and their development meet or intermix with wildland fuel”-and even included brief descriptions of the three identified categories of interface communities-interface, intermix, and occluded. FS010610. However, the Carbon County Plan's methodology does not follow the same development density standards found in the federal regulations for those community categories. Instead, the Plan measures and classifies development concentrations using much lower structural densities: low (1-5 structures/sqmi.); medium (5-25 structures/sqmi.); and high (>25 structures/sqmi.). FS010616. It is important to note that the Carbon County Plan measures development density using structures per square mile, as opposed to structures per acre, as measured in the federal regulation. So, only those areas with greater than approximately 16 structures per square mile would actually qualify as intermix (at minimum 1 structure per 40 acres),which is only a fraction of the communities that Carbon County included when developing the wildland-urban interface boundary. See FS010616 . The Carbon County Plan acknowledges this inconsistency with the regulations, stating that Carbon County “most closely resembles the Intermix Community category although most areas have a structure density less than one per 40 acres.” FS010610.
There 640 acres per square mile, therefore: (IMAGE OMITTED)
For the foregoing reasons, the Court concludes that the Carbon County Plan's wildland-urban interface is plainly overinclusive, and therefore, the Forest Service cannot rely exclusively on Carbon County's wildland-urban interface to justify the Project area's categorical exclusion under HFRA. However, the Court must also determine whether, under HFRA's standards, the Project would nonetheless fall within the wildland-urban interface. In other words, while the Carbon County Plan may be overinclusive, as a whole, the Court must take a closer look at that portion most relevant to the Project.
The Project area is near the city of Red Lodge, Montana. Red Lodge was identified in the notice entitled “Wildland Urban Interface Communities Within the Vicinity of Federal Lands that Are at High Risk From Wildfire” as an urban-wildland interface community at high risk from wildfire. 66 Fed.Reg. at 767. Plaintiffs argue that inclusion on this list, alone, is insufficient to conclude that Red Lodge is an at-risk community under HFRA. (Doc. 31 at 14.) However, Defendants need not rely on that listing to determine that Red Lodge fits HFRA's definition of an at-risk community. According to the Carbon County Plan's assessment of development density, Red Lodge, and the area around it extending inside the boundary with federal lands, is clearly shown to have a density in the high category, meaning greater than 25 structures per square mile. FS010617. As discussed above, this would qualify Red Lodge as an intermix community, which is one of the identified categories of interface communities recognized under HFRA. Moreover, the broader Red Lodge community is also “a group of homes and other structures with basic infrastructure and services . . . within or adjacent to Federal Lands;” therefore, Red Lodge would qualify as an at-risk community under 16 U.S.C. § 6511(1)(A)(ii).
Next, plaintiffs argue that the Project cannot fall within HFRA's wildland-urban interface because the Project “is not even remotely close to the City of Red Lodge.” (Doc. 31 at 13.) The Ninth Circuit again provides guidance for this stage of the analysis. In AWR v. Petrick, the court clarified that a project's boundaries need not directly abut an at-risk community. 68 F.4th at 496. Rather, HFRA only requires that the project fall within the wildland-urban interface, and that the wildland-urban interface abut an at-risk community. Id. Moreover, the court explained that while HFRA “creates a baseline protection of at least 0.5 or 1.5 miles around at-risk communities,” it also “permits communities with plans to identify wildland-urban interfaces that extend beyond those strict limitations to meet their communities' needs....[Thus,] giv[ing] communities with plans more-not less-flexibility.” Id. Thus, the court rejected the position that a project, itself, must border or abut an at-risk community and also acknowledged that a community plan's wildland-urban interface may encompass more area around an at-risk community than would be encompassed under HFRA's default protections.
Red Lodge, and its surround area, qualifies as an at-risk community under HFRA and the wildland-urban interface in the Carbon County plan abuts Red Lodge. Therefore, although Carbon County's wildland-urban interface is overinclusive as a whole, the Court finds that the portion of the wildland-urban interface that includes the Project is not. Because the Project falls within HFRA's definition of a wildland-urban interface, the Forest Service can rely on HFRA's exemption from the NRLMD's VEG S5 standard. Therefore, Defendants did not violate HFRA, NEPA, or the NFMA and the Court grants Defendants' cross-motion for summary judgment on this claim.
III. Palisades Timber Sale
Finally, Plaintiffs argue that Defendants violated NEPA by failing to consider the cumulative effects of a state-authorized timber sale adjacent to the Project boundary-the Palisades Timber Sale. (Doc. 18 at 40-41.) Specifically, Plaintiffs argue that the Forest Service “incorrectly assume[d] that the wildlife travel corridors that were part of [the] plan for the Palisades Timber Sale remain and function as productive wildlife habitat.” (Id. at 41.) Plaintiffs assert that those wildlife corridors disappeared during wind events, and consequently, “large acreages of land immediately adjacent to [the Project] no longer function as productive wildlife habitat” for grizzly, lynx, elk, and moose. (Id. at 40.) Defendants contend that the required cumulative effects analysis did occur. (Doc. 25 at 44.)
“NEPA requires that where several actions have a cumulative . . . environmental effect, this consequence must be considered in an EIS.” City of Tenakee Springs v. Clough, 915 F.2d 1308, 1312 (9th Cir. 1990). NEPA's implementing regulations define “cumulative impact” as “the impact on the environment which results from the incremental impact of the action when added to other past, present, and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person undertakes such other actions.” 40 C.F.R. § 1508.7 (2020).
The Forest Service first considered the cumulative effects of the projects in the 2015 EIS. The agency assessed the projects' cumulative effects on several big game species, including elk, mule deer, white-tailed deer, moose, and black bears, as well as providing separate analysis for grizzly bear and lynx. The agency considered the cumulative effects on secure habitat, forest cover, road density, bear denning habitat, understory forage, hunting security areas, noise disturbance, and movement through winter ranges. FS000973-74; FS001028-34. With regard to grizzly, the agency ultimately concluded that “disturbance and displacement would be limited and temporary . . . [and] [l]ong-term cumulative effects to grizzly bears are not expected.” FS000974. The agency also concluded that elk may be “temporarily displaced from the actively managed areas during project implementation,” FS001030, but that elk habitat would improve overall as a result of vegetative management on state and federal lands, FS001028. Regarding moose, the agency concluded that with some modifications to the proposed Alternative 2, the effects on moose are effectively mitigated. FS001033. Overall, the agency found that “[t]reatments that open up the forest canopy will allow increased sunlight to reach the forest floor and stimulate growth of grasses, forbs, and shrubs that provide forage for elk, deer, moose, and black bear, particularly in aspen.” FS001034. The agency supplemented its cumulative effects analysis in the 2020 SEIS, which was issued after the completion of the Palisades Timber Sale. See FS039315 .
Plaintiffs' specific concerns regarding wildlife corridors were raised by Montana Fish Wildlife and Parks (“MT FWP”) during the public comment period for the Draft SEIS. FS039168. In response, the Forest Service met with MT FWP's biologist to address the concern. FS039936; see also FS039052 . As a result of that conversation, the Forest Service agreed to investigate the status of the area surrounding the Palisades Timber Sale to assess potential impacts on moose habitat-the primary concern raised by the MT FWP biologist-and any future actions that may be necessary to ameliorate those impacts. Id. The Forest Service states that, as a result of those discussions, the agency agreed to remove certain units that function as winter habitat for moose. (Doc. 33 at 24.)
The record before the Court demonstrates that the Forest Service did consider the cumulative effects of the Project and the Palisades Timber Sale, including Plaintiffs' concerns regarding wildlife corridors. Accordingly, the Court finds that the Forest Service has not committed a “clear error of judgment” that would render its action “arbitrary and capricious,” Lands Council v. McNair, 537 F.3d 981, 991 (9th Cir. 2008), and grants Defendants' cross-motion for summary judgment on this claim.
IV. Remedy
Having found that the Forest Service violated NEPA by improperly tiering to Canfield (2016), the Court must determine appropriate relief. Plaintiff asks that this Court vacate the Project EIS and ROD and enjoin the implementation of the Project. (Doc. 18 at 41-43.) Alternatively, if the Court “finds an injunction unnecessary or inappropriate at this time, Plaintiffs request that the [C]ourt issue an order directing the Forest Service to provide 60 days notice to Plaintiffs of its intent [to] implement [the] Project” so that Plaintiffs may move to enjoin the action at that time. (Id. at 43.) Defendants argue that, should the Court reach the issue of remedy by finding for Plaintiffs on any of their claims, the Court should remand to the agency without vacatur and should deny Plaintiffs request for an injunction. (Doc. 25 at 48-52.)
The APA provides that the Court “shall . . . hold unlawful and set aside agency action, findings, and conclusion found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). “[V]acatur of an unlawful agency action normally accompanies remand” because “ordinarily when a regulation is not promulgated in compliance with the APA, the regulation is invalid.” AWR v. USFS, 907 F.3d at 1121 (cleaned up). However, where equity demands, the presumption in favor of vacatur may be overcome and the Court may remand without vacatur. Id.; see also Nat'l Wildlife Fed'n v. Epsy, 45 F.3d 1337, 1343 (9th Cir. 1995) (explaining that a reviewing court “is not required to set aside every unlawful agency action,” and that “the decision to grant or deny injunctive or declaratory relief under APA is controlled by principles of equity”); Idaho Farm Bureau Fed'n v. Babbitt, 58 F.3d 1392, 1405 (9th Cir. 1995) (“[W]hen equity demands, the regulation can be left in place while the agency follows the necessary procedures.”).
In determining whether to vacate an unlawful agency action, the Court must weigh the “competing claims of injury . . . and the effect on each party.” Nat'l Wildlife Fed'n, 45 F.3d at 1343; see also Cal. Cmtys. Against Toxics v. EPA, 688 F.3d 989, 992 (9th Cir. 2012) (weighing the seriousness of the agency's error against “the disruptive consequences of an interim change that may itself be changed”). The Court may also consider “whether the agency would likely be able to offer better reasoning or whether by complying with procedural rules, it could adopt the same rule on remand, or whether such fundamental flaws in the agency's decision make it unlikely that the same rule would be adopted on remand.” Pollinator Stewardship Council v. EPA, 806 F.3d 520, 532 (9th Cir. 2015).
The Court finds that this is not one of the limited circumstances where vacatur is inequitable. Defendants argue that vacatur “would significantly delay critical wildfire mitigation . . . [and] could therefore cause significant environmental harm.” (Doc. 25 at 51.) Weighing against this harm is the potential for harm to lynx and lynx habitat, a threatened species, over an area that extends beyond the Project's boundaries. Remand without vacatur would allow timber harvest and precommercial thinning activities to begin under the Forest Service's current delineation of lynx habitat while the agency revisits its lynx habitat map revisions under the appropriate NEPA analysis. As discussed above, the agency's assessment of the Project's impacts on lynx and compliance with the NRLMD's requirements is founded on the delineation of lynx habitat in the Rock Creek and Rosebud LAUs. To allow the Project to move forward while such a core component of the EIS is reconsidered would trivialize the statutory directives of NEPA and the APA.
Accordingly, the Court will both vacate and remand the Project for the Forest Service to remedy the errors identified above. Because the Court orders that the Project EIS and ROD be set aside, Defendants are enjoined from implementing the Project.
Conclusion
IT IS ORDERED that the cross-motions for summary judgment (Docs. 17 and 26) are GRANTED in part and DENIED in part as follows:
1. Claim 1: Plaintiffs are entitled to summary judgment on Plaintiffs' claim that the Project EIS improperly tiers to Canfield (2016) in violation of NEPA.
2. Claim 2: Defendants are entitled to summary judgment on Plaintiffs' claim that the Project ROD and EIS violate NEPA, HFRA, and NFMA by improperly designating the Project as within the wildland-urban interface.
3. Claim 3: Defendants are entitled to summary judgment on Plaintiffs' claim that the Project EIS failed to take a hard look at the cumulative effects of the Palisades Timber Sale.
IT IS FURTHER ORDERED that Defendants are ENJOINED from implementing the Project and this matter is REMANDED to the Forest Service to address the deficiencies identified in this Order.
The Clerk of Court is directed to (1) enter judgment in accordance with this Order and (2) close this case.