Southeast Alaska Conservation Council v. United States Forest Service

• Court: District Court, D. Alaska
• Decided: March 11, 2020
• Docket: 1:19-cv-00006
• Status: Unknown

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF ALASKA

SOUTHEAST ALASKA CONSERVATION COUNCIL, et al.,

Plaintiffs,

v.

UNITED STATES FOREST Case No. 1:19-cv-00006-SLG SERVICE, et al., Defendants.

DECISION AND ORDER This is an action in which Plaintiffs Southeast Alaska Conservation Council, Alaska Rainforest Defenders, Center for Biological Diversity, Sierra Club, Defenders of Wildlife, Alaska Wilderness League, National Audubon Society, and Natural Resources Defense Council seek the invalidation of portions of the 2018 Environmental Impact Statement and 2019 Record of Decision for the Forest Service’s Prince of Wales Landscape Level Analysis Project for the Tongass National Forest. Briefing on the merits concluded on August 23, 2019.1 Oral argument was held on February 7, 2020.2

1 Pursuant to the briefing schedule for administrative agency appeals, see L. R. Civ. P. 16.3(c), Plaintiffs’ Opening Brief is at Docket 10, Defendants’ Brief in Opposition is at Docket 12, and Plaintiffs’ Reply Brief is at Docket 19. Plaintiffs subsequently filed a Notice of Errata and Corrected Opening Brief at Docket 22. 2 Docket 37. BACKGROUND The factual background for this case was set out at some length in the Court’s prior order granting Plaintiffs’ motion for a preliminary injunction.3 It is

repeated here to a certain extent with some additions. The Tongass National Forest (“Tongass”) is a 16.7 million-acre forest in Southeast Alaska.4 The nation’s largest national forest,5 the Tongass has seen timber harvesting of varying intensity over the past 100 years.6 In the 1950s, the Forest Service awarded several 50-year timber sale contracts in the forest to

“provide a sound economic base in Alaska through establishment of a permanent year-round pulp industry.”7 But logging in the Tongass began to slow in the 1980s and 1990s, when several of these long-term contracts were terminated due to market fluctuation, litigation, and other factors.8 Prince of Wales Island, a large island in the Alexander Archipelago, lies

within the Tongass.9 Two large pulp mills once operated on the island, where industrial scale logging occurred in the second half of the 20th century, but both

3 Docket 27 at 2–7 4 Administrative Record (“AR”) 833_0404 at 063052, 063054. 5 AR 833_0404 at 063407. 6 AR 833_2077 at 069553–55. 7 AR 833_2077 at 069553. 8 AR 833_2077 at 069553–55. 9 AR 833_0404 at 063054.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. mills closed in the 1990s.10 There are 12 communities on the island with a total of approximately 4,300 residents, many of whom are Alaska Native.11 Tourism and sport and commercial fishing are important to the local economy,12 and many

residents rely to some degree on subsistence hunting, fishing, and gathering.13 Pursuant to the National Forest Management Act (“NFMA”) and its implementing regulations, the Forest Service has developed land and resource management plans, also called forest plans, to govern its management of the Tongass.14 Forest plans “operate like zoning ordinances, defining broadly the uses

allowed in various forest regions, setting goals and limits on various uses . . . , but do not directly compel specific actions, such as cutting of trees in a particular area or construction of a specific road.”15 Any activity occurring within a national forest must comply with the governing forest plan,16 which the Forest Service is required to revise at least every 15 years.17 The current forest plan for the Tongass was

issued in 2016, following the completion of an environmental impact statement

10 AR 833_2167 at 01750. 11 AR 833_2167 at 01753; see also AR 833_2167 at 01751, tbl. 70 (showing population change). 12 AR 833_2167 at 001750 13 See AR 833_2167 at 001753–57 (describing different communities on the island). 14 See 16 U.S.C. § 1604. 15 Citizens for Better Forestry v. U.S. Dept. of Agric., 341 F.3d 961, 966 (9th Cir. 2003). 16 16 U.S.C. § 1604(i). 17 36 C.F.R. § 219.7(a).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. (“EIS”).18 The Forest Plan provides that “[t]imber harvest unit cards will document resource concerns and protection measures,” and requires that these “unit cards, including a map with relevant resource features, . . . be provided electronically

when Draft or Final NEPA documents and decisions are published.”19 In late 2016, the Forest Service initiated environmental planning for a proposed project within the Tongass: the Prince of Wales Landscape Level Analysis Project (“Project”).20 The agency describes the Project as “a large landscape-scale NEPA analysis that will result in a decision whether or not to

authorize integrated resource management activities on Prince of Wales Island over the next 15 years.”21 The Forest Service released a final EIS for the Project on October 19, 201822 and issued a Record of Decision (“ROD”) selecting the preferred alternative from the EIS on March 16, 2019.23 The Project encompasses all of the land within the national forest system on

Prince of Wales Island, consisting of roughly 1.8 million acres.24 It authorizes four

18 AR 833_0404 at 063039–063554 (2016 Tongass Forest Plan); AR 833_2079 at 071034– 072626 (EIS for Tongass Forest Plan). 19 AR 833_0404 at 063265. 20 AR 833_2167 at 001468. 21 AR 833_2167 at 001459. 22 AR 833_2167 at 001437–001863 (Final EIS). 23 AR 833_2427 at 000776–001118; see also 42 U.S.C. § 4332(2)(C) (requiring agencies to prepare a “detailed statement” for actions with significant environmental impacts). 24 AR 833_2167 at 001460–61; see also AR 833_2427 at 000781 (map showing distribution of national forest system land on Prince of Wales Island).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. categories of activities within this area: vegetation management, including timber harvesting; watershed improvement and restoration; sustainable recreation management; and “associated actions.”25 The EIS for the Project does not specify

when and where individual activities will occur within the Project Area. Rather, the Project is designed to be a flexible planning framework intended to allow the Forest Service to tailor resource management to changing conditions on the ground over the course of the Project’s 15-year term. The Forest Service appended to the EIS what it terms an Activity Card for

each of the 46 activities included in the four activity categories.26 “The Activity Cards describe each potential activity and the related resource considerations,” and include “[p]roject-specific design criteria and mitigation measures.”27 The Activity Cards were designed using “on-the-ground inventories, computer (GIS) data, and aerial photographs to assess project area conditions and resource-

specific concerns.”28 The Activity Cards describe and govern activities at the project level, but they do not identify the specific geographic areas within the Project Area where each activity will occur.29 Unlike prior sales, the Project EIS

25 AR 833_2167 at 001443. 26 AR 833_2427 at 000848–001030. 27 AR 833_2167 at 001492; see, e.g., AR 833_2427 at 000848–52 (Activity Card 01). 28 AR 833_2167 at 001492. 29 See, e.g., AR 833_2427 at 000848 (Activity Card 01 explaining that “rotational harvest of young growth using even-aged management” may occur “within the suitable land base based on legal and technical factors” and “within potential harvest units as shown in the [LSTA]”); AR

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. was not accompanied by timber harvest unit cards with maps detailing specific harvest configurations.30 In preparing the Project EIS, the Forest Service also developed a Logging

System Transportation Analysis (“LSTA”) to “identif[y] potential stands for timber harvest and the associated transportation network that would be needed.”31 The LSTA “was developed for National Forest System lands within the project area layer using information from the Forest GIS library, aerial photos, and the Forest Service Activity Tracking System database.”32 The LSTA identified 125,529 acres

of potential timber harvest in the Project Area: 48,140 old-growth acres and 77,389 young-growth acres.33 The LSTA also identified 643 miles of new roads, 505 of them temporary and 138 permanent.34 The Forest Service represented this

833_2167 at 001459 (“The Activity Cards are used to identify and analyze a suite of possible activities that could be implemented in the project area. With this approach, activities can vary in magnitude and intensity to respond to resource conditions.”). 30 Cf. AR 833_2084 at 061279–80 (card for timber harvest unit 101 in 2007 Kuiu Timber Sale EIS). 31 AR 833_2167 at 001561. 32 AR 833_2167 at 001561; see also AR 833_2167 at 1561–71 (describing Forest Service’s methodology and integration of LSTA into Project EIS); AR 833_1369 at 074744 (providing path for GIS data .zip file for the draft EIS); AR 833_2115 at 074758 (providing path for GIS data .zip file for the final EIS); AR 833_2403 at 074762–074854 (Jan. 1, 2015 spreadsheet documenting raw GIS numbers for entire Tongass National Forest). 33 AR 833_2167 at 001481. The Forest Service further subdivided the Project Area into 18 Timber Analysis Areas (“TAA”) “[t]o allow for more site-specific analysis.” Id. at 001562; see also id. at 001563 (map showing distribution of TAAs); id. at 001569–70 (tables showing distribution of old- and young-growth acreage across TAAs). 34 AR 833_2167 at 001481.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. information in a Commercial Vegetation Management map,35 which it appended to the ROD, and to which it provided a link in the EIS.36 The Project EIS addresses four alternatives in detail, including a no-action

alternative.37 Each activity under each alternative must be consistent with the applicable Activity Card and certain alternative-specific features, and each activity must also occur within the areas identified for that activity in the LSTA.38 Each action alternative establishes a maximum potential amount of timber harvest and road construction.39 Focusing on timber harvest, Alternative 2, the preferred

alternative, allows a maximum of 23,269 acres of old-growth harvest and 19,366 acres of young-growth harvest, or roughly 34 percent of the total potential acreage in the LSTA.40 However, the EIS does not identify where the harvest authorized

35 833_2178. This map is attached as an appendix to this decision. See Docket 40-1. 36 AR 833_2167 at 001480. 37 AR 833_2167 at 001479–80. 38 See AR 833_2167 at 001480 (describing role of LSTA and Activity Cards in comparison of alternatives); id. at 001485–92 (describing features specific to the action alternatives). 39 AR 833_2167 at 001480 (“The total acreage and maximum miles of road construction under any one alternative is expected to change based on the logging systems used and where harvest occurs on the landscape, but would not exceed the amount identified within that alternative.”). 40 AR 833_2167 at 001481. Alternative 3 allows a maximum of 13,014 acres of old-growth and 36,670 acres of young-growth harvest, or roughly 40 percent of the total potential acreage in the LSTA; and Alternative 5 allows a maximum of 6,365 acres of old-growth and 36,670 acres of young-growth harvest, or roughly 34 percent of the total potential acreage in the LSTA. Id.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. by each alternative would occur within the potential acreage identified in the LSTA.41 In order to capture the “maximum effects” of the Project, the Project EIS

makes several assumptions in addressing each alternative.42 First, in analyzing each alternative, the Forest Service indicates that it assumed that all acres of potential harvest in the LSTA would be harvested and all roads proposed by the alternative would be built.43 Second, the Forest Service assumed that all acres would be harvested using clear-cut methods.44 Third, the Forest Service assumed

that each Wildlife Analysis Area—a land division used by the Alaska Department of Fish and Game—would be harvested to the maximum acreage available.45 As noted above, the alternatives do not provide the specific locations or configurations of harvest or roadbuilding within the LSTA. Instead, the Project EIS

41 Instead, each alternative considers the Project Area as a whole and “describe[s] the conditions being targeted for treatments and what conditions cannot be exceeded in an area, or place[s] limits on the intensity of specific activities.” AR 833_2167 at 001459. 42 AR 833_2167 at 001639. 43 See, e.g., AR 833_2167 at 001629 (“[A]ssumptions include that all harvest stands from the LSTA would be harvested . . . .”); AR 833_2167 at 001789–90 (discussing road construction by alternative); see also Docket 12 at 31 (describing analytical approach). It is unclear from the record whether the Forest Service assumed that the maximum potential harvest in the LSTA would be harvested or the maximum acreage under each alternative would be harvested; it appears that the agency made different assumptions about harvest intensity depending on the impact under consideration. See Docket 39 at 13:23–16:3 (Tr. of Feb. 7, 2020 Oral Arg.) (discussing Forest Service’s methodology). 44 AR 833_2167 at 001450. 45 See AR 833_2167 at 001500. There are 32 Wildlife Analysis Areas in the Project Area. Id. at 001514.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. provides that “site-specific locations and methods” for activities such as timber harvest “will be determined during implementation” over the 15-year lifespan of the Project.46 It explains that siting decisions and the parameters of actual timber

sales will be determined pursuant to an Implementation Plan, in a way that is consistent with the alternative selected by the ROD and the Activity Cards developed for the EIS.47 However, the EIS makes clear that these subsequent, site-specific decisions will not be subject to additional NEPA review.48 The Forest Service terms this approach “condition-based analysis.”49

The Implementation Plan published with the ROD sets out a nine-step process for making site-specific determinations.50 This process includes checking the action against the relevant Activity Card, the final EIS, and the ROD, as well as engaging in “workshops and other public involvement techniques.”51 It is during

46 AR 833_2167 at 001459. 47 AR 833_2167 at 001459; see also 833_2169 at 002076–002126 (Implementation Plan attached to Project EIS). 48 See AR 833_2169 at 002078. At oral argument, the Forest Service maintained that the Implementation Plan could require supplemental NEPA analysis if new information arose indicating that an “activity is no longer going to have an effect that was disclosed or analyzed” in the Project EIS. Docket 39 at 21:18–25. The Court has been unable to locate a specific provision in the Implementation Plan that mandates supplemental review, but the plan does require the Forest Service to “verify that the effects of the proposed activity [are] within the effects analyzed for the Selected Alternative” in the EIS. AR 833_2427 at 001039. 49 AR 833_2167 at 001443; see also AR 833_2427 at 000802 (“The POWLLA FEIS uses a condition-based approach where specific harvest units and roads will be determined during implementation through a collaborative public process and interdisciplinary review.”). 50 See AR 833_2427 at 001037 (graphically describing Implementation Plan). 51 AR 833_2427 at 001037. The ROD explains that a proposed activity “will take several

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. this process, also, that site-specific unit cards will be developed that describe particular harvest configurations.52 The ROD explains that the Implementation Plan “is integral to the analysis of effects in the [Project EIS] and the Selected

Alternative in the [ROD],” and was “developed . . . to provide a linkage from the [Project EIS] to the project-specific work without the need for additional NEPA analysis.”53 That said, the ROD describes the Implementation Plan as a “living document” that “may need to be adjusted.”54 The Forest Service began implementing the Project shortly after issuing the

ROD. It held a public workshop on April 6, 201955 and published an “Out-Year Plan” for fiscal year 2019 that included a proposed timber sale of 1,156.34 acres, known as the Twin Mountain Timber Sale.56 The Forest Service also published

months to a year to go through all steps of the implementation process.” Id. 52 AR 833_2427 at 000826. 53 AR 833_2427 at 001034. The ROD expanded on the relationship between the Activity Cards and the Implementation Plan, stating that the “Activity Cards provide activity-specific design criteria, best management practices, and mitigation measures and the implementation plan includes site conditions, triggers, or other requirements for each type of activity to inform future data needs and field visits to develop treatment scenarios that are consistent with the NEPA analysis.” Id. 54 AR 833_2427 at 001034. The ROD explains that “[a]s activities are designed, the process will likely be smoother and new technology or expertise may be used.” Id. At oral argument, the Forest Service maintained that the Implementation Plan was “not necessary for NEPA compliance.” Docket 39 at 18:18–19; cf. Docket 12 at 19 (“Implementation is an integral part of the Project’s design.”). 55 See, U.S. Forest Serv., Prince of Wales Landscape Level Analysis Project, Dear Planning and Implementation Participant Letter, https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fseprd622020.pdf (last visited March 5, 2020). 56 See, U.S. Forest Serv., Prince of Wales Landscape Level Analysis Project, Out-Year Plan,

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. draft unit cards for the sale, which identify the specific locations and method of timber harvest in graphical and narrative form.57 Plaintiffs initiated this case on May 7, 2019.58 The Complaint is brought

pursuant to the Administrative Procedure Act (“APA”), 5 U.S.C. §§ 702–06, and alleges that the Project EIS violates three federal laws: (1) the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4332; (2) the Alaska National Interest Lands Conservation Act (“ANILCA”), 16 U.S.C. § 3120; and (3) the National Forest Management Act (“NFMA”), 16 U.S.C. § 1604.59 The Complaint

seeks declaratory judgment, vacatur of the ROD “or portions of it deemed not in compliance with law,” and “preliminary and permanent injunctive relief as needed to prevent irreparable harm from implementation of the [Project].”60 Plaintiffs refine their claim in their merits briefing, requesting vacatur of the “portions of the ROD authorizing vegetation management and road construction.”61

https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fseprd622075.pdf (last visited March 5, 2020); see also Docket 21-1 at 2-3, ¶ 6 (providing size of sale). 57 See, U.S. Forest Serv., Twin Mountain Sale Draft Unit Cards, https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fseprd641767.pdf (last visited March 5, 2020). Plaintiffs have produced an area map of the proposed timber activities. See Docket 10-2 at 5 (Ex. A). 58 Docket 1. 59 Docket 1 at 15–19, ¶¶ 47, 51, 58. 60 Docket 1 at 19, ¶¶ 1–5. Plaintiffs also seek “the costs of this action, including reasonable attorneys’ fees pursuant to the Equal Access to Justice Act, 28 U.S.C. § 2412.” Id. at 19, ¶ 6. 61 Docket 22-1 at 43–44. In the alternative, Plaintiffs seek a permanent injunction of “the vegetation management activities and road construction in the ROD.” Id. at 44–47.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. On September 23, 2019, the Court entered a preliminary injunction, which prohibited the Forest Service from awarding a contract or authorizing ground- disturbing activities associated with the Twin Mountain Timber Sale during the

pendency of this case.62 In the preliminary injunction order, the Court found that Plaintiffs had raised serious questions going to the merits of their NEPA claim.63 The order also informed the parties that the Court intended to issue a final decision on the merits no later than March 31, 2020.64 JURISDICTION

The Court has subject matter jurisdiction pursuant to 28 U.S.C. § 1331, which “confer[s] jurisdiction on federal courts to review agency action, regardless of whether the APA of its own force may serve as a jurisdictional predicate.”65 LEGAL STANDARD Plaintiffs’ claims arise under the APA.66 Under that statute, a reviewing court

shall not set aside an agency's decision unless it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”67 Agency action is arbitrary and capricious if it

62 Docket 27 at 25. 63 Docket 27 at 20. 64 Docket 27 at 8. 65 Califano v. Sanders, 430 U.S. 99, 105 (1977). 66 Docket 1 at 15–19, ¶¶ 47, 51, 58. 67 5 U.S.C. § 706(2)(A).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. relie[s] on factors which Congress has not intended it to consider, entirely fail[s] to consider an important aspect of the problem, offer[s] an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it c[an]not be ascribed to a difference in view or the product of agency expertise.68 A court’s review of whether an agency action is arbitrary and capricious should be “searching and careful,” but “narrow,” as a court may not substitute its judgment for that of the administrative agency.69 Courts will generally “uphold agency decisions so long as the agencies have ‘considered the relevant factors and articulated a rational connection between the factors found and the choices made.’”70 “Agency action is ‘not in accordance with the law’ when it is in conflict with the language of the statute relied upon by the agency.”71 “Whether agency action is ‘not in accordance with law’ is a question of statutory interpretation, rather than an assessment of reasonableness in the instant case.”72

68 Prot. Our Cmtys. Found. v. LaCounte, 939 F.3d 1029, 1034 (9th Cir. 2019) (alterations in original) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto Ins. Co., 463 U.S. 29, 43 (1983)). 69 Marsh v. Or. Nat. Res. Council, 490 U.S. 360, 378 (1989) (quoting Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971)). 70 Prot. Our Cmtys. Found., 939 F.3d at 1034 (quoting City of Sausalito v. O’Neill, 386 F.3d 1186, 1206 (9th Cir. 1206)). 71 City of Cleveland v. Ohio, 508 F.3d 827, 838 (9th Cir. 2007). 72 Singh v. Clinton, 618 F.3d 1085, 1088 (9th Cir. 2010) (citing Nw. Envtl. Advocates v. U.S. Envtl. Prot. Agency, 537 F.3d 1006, 1014 (9th Cir. 2008)).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. DISCUSSION The Court will address Plaintiffs’ NEPA, ANILCA, and NFMA claims in turn. I. National Environmental Policy Act

Pursuant to NEPA, agencies must prepare an EIS before taking an action “significantly affecting the quality of the human environment.”73 Regulations issued by the Council on Environmental Quality require an EIS to include discussion of the direct and indirect effects of the action, as well as “[t]he environmental effects of alternatives.”74 After completing an EIS, “the agency must select a course of

action within the range of alternatives analyzed and issue an ROD,” which “explains why the agency chose a particular alternative, whether all practical means for avoiding or minimizing environmental harm have been adopted, and, if not, why not.”75 “An EIS must ‘reasonably set forth sufficient information to enable the

decisionmaker to consider the environmental factors and make a reasoned decision.’”76 The agency meets this obligation if its EIS “contains a reasonably

73 42 U.S.C. § 4332(2)(C) (requiring a “detailed statement” analyzing “the environmental impact of the proposed action,” among other things). 74 40 C.F.R. § 1502.16. 75 Protect Our Cmtys. Found., 939 F.3d at 1035 (citing 40 C.F.R. §§ 1505.1, 1505.2). 76 Alaska Ctr. for Env’t v. Armbrister, 131 F.3d 1285, 1289 (9th Cir. 1997) (quoting Or. Envtl. Council v. Kunzman, 817 F.2d 484, 493 (9th Cir. 1987)); see also City of Tenakee Springs v. Block, 778 F.2d 1402, 1407 (9th Cir. 1985) (explaining that EIS must be “sufficient ‘to give to decision makers . . . removed from the initial decision sufficient data from which to draw their own conclusions.’” (omissions in original) (quoting Coal. for Canyon Preservation v. Bowers, 632 F.2d 774, 782 (9th Cir. 1980))).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. thorough discussion of the significant aspects of the probable environmental consequences.”77 The NEPA process “serves two fundamental objectives”: “First, it ‘ensures that the agency, in reaching its decision, will have available, and will

carefully consider, detailed information concerning significant environmental impacts,” and “second, it requires ‘that the relevant information will be made available to the larger audience that may also play a role in both the decisionmaking process and the implementation of that decision.’”78 “NEPA requires . . . procedural steps but does not require an agency to

reach any particular result.”79 Indeed, agencies retain significant discretion over their “methodology and planning strategy” when engaging in environmental review, and “NEPA’s ‘requisite “hard look” does not require adherence to a particular analytic protocol.’”80 The act “merely prohibits uninformed—rather than unwise—

77 WildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d 920, 924 (9th Cir. 2015) (quoting City of Sausalito v. O’Neill, 386 F.3d 1186, 1206 (9th Cir. 2004)). 78 Id. (quoting Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349 (1989). Put another way: The requirement of an EIS serves two ends. A properly prepared EIS ensures that federal agencies have sufficiently detailed information to decide whether to proceed with an action in light of potential environmental consequences, and it provides the public with information on the environmental impact of a proposed action and encourages public participation in the development of that information. Or. Envtl. Council v. Kunzman, 817 F.2d at 492. 79 Protect Our Cmtys. Found., 939 F.3d at 1035. 80 Or. Nat. Desert Ass’n v. Bureau of Land Mgmt., 625 F.3d 1092, 1121 (9th Cir. 2010) (quoting Ass’n of Pub. Agency Customers, Inc. v. Bonneville Power Admin., 126 F.3d 1158, 1188 (9th Cir. 1997)).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. agency action.”81 To determine whether an agency has complied with NEPA’s requirements, courts apply a rule of reason, which involves “a pragmatic judgment whether the EIS’s form, content and preparation foster both informed decision-

making and informed public participation.”82 One method available for agencies engaging in long-term planning is to first prepare a programmatic EIS that considers the broad, program-level effects of a coordinated series of actions, and then conduct subsequent site-specific NEPA analysis for each action as it occurs, tiering back to the programmatic EIS.83

Regarding this approach, the Ninth Circuit has explained that “[t]he detail that NEPA requires in an EIS depends on the nature and scope of the proposed action,” and that “[t]he critical inquiry in considering the adequacy of an EIS prepared for a large scale, multi-step project is not whether the project’s site-specific impact should be evaluated in detail, but when such detailed evaluation should occur.”84

81 Methow Valley Citizens Council, 490 U.S. at 351. 82 Churchill Cty. v. Norton, 276 F.3d 1060, 1071 (9th Cir. 2001) (quoting California v. Block, 690 F.2d 753, 761 (9th Cir. 1982)). 83 See, e.g., W. Watersheds Proj. v. Abbey, 719 F.3d 1035, 1049 (9th Cir. 2013) (noting that agencies can plan at programmatic or site-specific levels and that “[w]hen an agency develops an EIS for a programmatic plan . . . , the EIS ‘must provide “sufficient detail to foster informed decision-making,” but “site-specific impacts need not be fully evaluated until a critical decision has been made to act on site development”’” (quoting Friends of Yosemite Valley v. Norton, 348 F.3d 789, 800 (9th Cir. 2003))); see also 40 C.F.R. § 1502.20 (NEPA regulation encouraging tiering to “eliminate repetitive discussions of the same issues and to focus on the actual issues ripe for decision at each level of environmental review”). 84 California v. Block, 690 F.2d at 761. The Circuit added that “[w]hen a programmatic EIS has already been prepared, . . . site-specific impacts need not be fully evaluated until a ‘critical decision’ has been made to act on site development.” Id. (quoting Sierra Club v. Hathaway, 579 F.2d 1162, 1168 (9th Cir. 1978)); see also City of Tenakee Springs v. Block, 778 F.2d 1402, 1407

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. The Forest Service chose not to take that approach here. Instead of preparing a programmatic EIS to be followed by site-specific NEPA analyses for individual timber sales as they occur, the agency compressed its NEPA review for

the entire 15-year Project into a single document. The Forest Service maintains that its “landscape-scale NEPA analysis” enables informed decision-making about integrated resource management at the programmatic level and contains sufficient site-specific information and analysis to proceed with individual timber sales over the 15-year Project period without additional NEPA review.85

The Court will not here decide whether the nature of the Project required the Forest Service to complete a programmatic EIS to which later planning documents would tier.86 Nor do Plaintiffs request such a decision.87 Instead, the Court will evaluate the analytical method that the agency employed and determine whether the Project EIS, a landscape-level long-term planning document, adequately

evaluated the site-specific impacts of as-yet undefined timber sales that could

(9th Cir. 1985) (“Where there are large scale plans for regional development, NEPA requires both a programmatic and a site-specific EIS.” (citing 40 C.F.R. § 1508.28, 1502.20; Kleppe v. Sierra Club, 427 U.S. 390, 409–14 (1976))); cf. Prot. Our Cmtys. Found., 939 F.3d at 1039 (explaining that “a site-specific project demands site-specific analysis” and that “[a]gencies cannot rely on a general discussion in a programmatic EIS or other document to satisfy [their] NEPA obligations for a site-specific action”). 85 Docket 12 at 24–35; see also AR 833_2167 at 001459 (describing purpose of EIS); AR 833_2169 at 002078 (stating that there is no “need for additional NEPA analysis” under the Forest Service’s approach). 86 See Churchill Cty., 276 F.3d at 1074–79 (cataloguing cases regarding the necessity of programmatic EISs). 87 Docket 22-1 at 22–36.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. potentially occur on certain acreage on Prince of Wales Island over the Project’s term. Plaintiffs contend that the Project EIS does not provide sufficient site-specific

information or analysis to comply with NEPA.88 They argue that this case is governed by the Ninth Circuit’s decision in City of Tenakee Springs v. Block.89 In that case, the Circuit reversed a district court’s decision not to enjoin “construction of an 11-mile road through the Kadashan watershed” in the Tongass.90 The plaintiffs had challenged the adequacy of an EIS for a five-year operating plan that

“specified no timber harvesting in the Kadashan Watershed in the years 1981 through 1986,” but did authorize the construction of the road for future harvest activity.91 The Circuit ordered the entry of a preliminary injunction, in part due to its conclusion that the plaintiffs had raised serious questions about the merits of their NEPA claim.92 It explained that the challenged EIS did not “g[ive] any

indication of its overall plan for timber harvesting” in the designated area and that “it [was] impossible to determine where and when harvesting will occur on the

88 Docket 22-1 at 22–36; see also Docket 1 at 16–17, ¶¶ 44–47 (Count I of Compl.). The Court has already considered this argument in its order granting Plaintiffs’ request for a preliminary injunction, where it found that Plaintiffs had raised serious questions on the merits on their NEPA claim. See Docket 27 at 14–20. The Court will necessarily repeat that analysis to a certain degree here. 89 778 F.2d 1402 (9th Cir. 1985). 90 Id. at 1403. 91 Id. at 1408. 92 Id. at 1407–08.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. 750,000 acres of land.”93 The Circuit held that the EIS was inadequate, reasoning that the location and timing of logging would affect “the locating, routing, construction techniques, and other aspects of the road, or even the need for its

construction.”94 In City of Tenakee Springs, the Ninth Circuit separately rejected the trial court’s conclusion that the Forest Service had discretion to determine the specificity of its environmental review.95 Instead, it held that “[a]lthough the agency does have discretion to define the scope of its actions, such discretion does not

allow the agency to determine the specificity required by NEPA.”96 The Circuit explained that “[w]here there are large-scale plans for regional development, NEPA requires both a programmatic and a site-specific EIS.”97 Here, Plaintiffs argue that the Project EIS, with its condition-based analysis, is similarly deficient, and that the Forest Service impermissibly limited the

specificity of its environmental review.98 The Forest Service maintains that it has

93 Id. at 1408. 94 Id. (quoting Thomas v. Peterson, 753 F.2d 754, 760 (9th Cir. 1985), abrogated on other grounds by Cottonwood Envtl. Law Ctr. v. U.S. Forest Serv., 789 F.3d 1075 (9th Cir. 2015)). 95 Id. at 1407. 96 Id. (citing California v. Block, 690 F.2d 753, 765 (9th Cir. 1982)). 97 Id. 98 Docket 22-1 at 23–25.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. complied with NEPA by creating a project-level map that “provide[s] information on where timber harvest and road construction activities may take place.”99 The Forest Service used the LSTA to identify “potential stands for timber

harvest as well as the transportation network needed to access those stands” within the roughly 1.8-million-acre Project Area over a 15-year period.100 Through this process, the agency identified 125,529 acres of timber for potential harvest: 48,140 acres of old growth and 77,389 acres of young growth.101 The Project EIS links to a Commercial Vegetation Management map, which portrays this potential

acreage graphically.102 However, the EIS expressly leaves site-specific determinations about the actual location of timber harvest within this potential acreage for future determination.103 For example, Alternative 2—the selected alternative—allows 23,269 acres of old-growth harvest, but does not specify where this harvest will be located within the 48,140 acres of old growth identified as

suitable for harvest in the Project Area.104 The Activity Cards likewise do not

99 Docket 12 at 26–27, 28–30. 100 AR 833_2167 at 001480. 101 See AR 833_2167 at 001481 tbl. 2. The Forest Service also determined that a maximum of 505 miles of temporary roads and 138 miles of permanent roads would be necessary to access the potential timber harvest. Id.; see also AR 833_2178 (showing potential locations of temporary and permanent road construction). 102 AR 833_2167 at 001480; see also AR 833_2178. 103 AR 833_2167 at 1459 (“The site-specific locations and methods [of activities authorized by the EIS] will be determined during implementation based on defined conditions in the alternative selected in the [ROD] in conjunction with the Activity Cards . . . and Implementation Plan.”). 104 AR 833_2167 at 001481 tbl. 2. Similarly, the Project EIS does not specify where the 129

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. identify with specificity where harvesting will occur, although they do contain mitigation measures that could constrain future siting decisions.105 The Forest Service maintains that it properly exercised its discretion to

determine the scope of the Project while at the same time providing the specificity required by NEPA.106 The agency argues that through Project EIS, it “has provided information on where timber harvest and road construction may take place and is not ‘attempt[ing] to justify’ any lack of information, or to opt-out of any of NEPA’s requirements.”107 However, similar to the EIS found inadequate by the Ninth

Circuit in City of Tenakee Springs, the Project EIS does not include a determination—or even an estimate—of when and where the harvest activities or road construction authorized by each alternative will actually occur.108 Rather, it reserves actual siting decisions for the future, as individual timber sales are offered.

miles of temporary roads and 35 miles of permanent roads would be located. See id. The EIS explains that for each alternative, “[t]he total road miles needed will be determined by the specific harvest units offered and the needed transportation network.” AR 833_2167 at 001789. 105 See, e.g., AR 833_2168 at 001935–36 (Activity Card for old-growth clearcutting requiring Forest Service to, among other things, “limit the size of even-age openings to 100 acres with certain exceptions” and “locate activities outside of required nest/den buffers” where feasible); but see Docket 12 at 29 (Forest Service asserting that the “Activity Cards are site-specific to the Project Area in that they reflect analyses by resource specialists who used on-the-ground inventories, GIS data, and aerial photographs to develop them”). 106 Docket 12 at 26–28 (distinguishing City of Tenakee Springs v. Block, 778 F.2d 1402, 1407 (9th Cir. 1985)). 107 Docket 12 at 26–27 (alteration in original) (quoting Docket 22-1 at 24). 108 See, e.g., AR 833_2167 at 001789 (“The total road miles needed will be determined by the specific harvest units offered [for sale] and the needed transportation network.”).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. The Forest Service contends that the phrase from City of Tenakee Springs on which Plaintiffs rely—that it “is impossible to determine where and when harvesting will occur on the 750,000 acres of land”—was factually inaccurate,

citing the district court’s decision on remand.109 But regardless of that statement’s factual accuracy, the Circuit’s reasoning is still binding precedent: NEPA requires that environmental analysis be specific enough to ensure informed decision- making and meaningful public participation.110 The Project EIS’s omission of the actual location of proposed timber harvest and road construction within the Project

Area falls short of that mandate. The Forest Service maintains that Plaintiffs are demanding more detail than is required by NEPA. The agency relies on Stein v. Barton, which concerned a challenge to the EIS for a five-year operating plan for an area of the Tongass that was then being logged pursuant to a long-term contract held by the Ketchikan Pulp

109 Docket 12 at 27; see City of Tenakee Springs v. Courtright, No. J86-024 CIV., 1987 WL 90272, at *3 (D. Alaska June 26, 1987) (“The [Ninth Circuit] opinion also contains puzzling language suggesting that the EIS did not state where and when harvesting would take place in the APC contract area. This may be an improperly-drafted allusion to the fact that the final EIS did not reveal where and when logging would take place along the Kadashan Road . . . .”). 110 City of Tenakee Springs v. Block, 778 F.2d 1402, 1407–08 (9th Cir. 1985); see also Alaska Ctr. for Env’t v. Armbrister, 131 F.3d 1285, 1289 (9th Cir. 1997) (“An EIS must ‘reasonably set forth sufficient information to enable the decisionmaker to consider the environmental factors and make a reasoned decision.’” (quoting Or. Envtl. Council v. Kunzman, 817 F.2d 484, 493 (9th Cir. 1987))). Also, the Court notes that the district court in City of Tenakee Springs v. Courtright invalidated an EIS that did not provide sufficient site-specific detail for “information [to] . . . sufficiently correlat[e] with environmental factors, such as distribution of fish and wildlife populations, to facilitate public discussion of the [four alternative road and harvest] configurations.” 1987 WL 90272, at *4.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. Company.111 Among other issues, the plaintiffs in that case argued the “harvesting plans for each area” were not sufficiently site-specific, “object[ing] that they cannot determine from the FEIS ‘where, when, and how logging and roading activities will

occur on the 812,477 acres of land.’”112 The district court rejected this argument, concluding that the EIS “contain[ed] comprehensive, detailed quantitative and qualitative descriptions of the logging and roading plans for each harvest unit.”113 The court noted that “the only details that the FEIS does not disclose are exact timetables and locations on the ground for planned harvesting activities in each

harvest unit,” but that this was not a fatal omission since “the Forest Service does not develop these details at the pre-implementation stage.”114 The Forest Service contends that Plaintiffs’ arguments are similar here and that NEPA does not require disclosure of the exact location of actual timber harvest in the Project Area.115 But the EIS that survived review in Stein contained

significantly greater site-specificity than the Project EIS at issue here. In Stein, the district court described a “nine-volume FEIS [that] employ[ed] a combination of annotated topographic maps, textual, and tabular data to describe the project

111 Docket 12 at 26 (citing 740 F. Supp. 743, 746 (D. Alaska 1990)). 112 Stein, 740 F. Supp. at 749 (citing plaintiffs’ briefing). 113 Id. 114 Id. 115 Docket 12 at 26. Plaintiffs note in their reply that they “do not fault the [Project EIS] for lack of timetables.” Docket 19 at 13.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. alternatives and their impacts on cognizable values within the affected areas” and contained “comprehensive, detailed quantitative and qualitative descriptions of the logging and roading plans for each harvest unit.”116 Similarly, in Alliance for the

Wild Rockies v. Weber, another case the Forest Service cites,117 the district court upheld an environmental analysis for a timber sale that “identif[ied] the project boundaries down to the township and range level” and contained maps that would “allow the Plaintiffs to identify where those activities will take place in relation to bull trout critical habitat,” a resource value the plaintiffs had claimed was

inadequately addressed.118 The Project EIS at issue here does not approach this level of specificity; it does not delineate harvest units, let alone identify planned activities within them and describe their impacts on localized cognizable values. Nor does the Project EIS allow the public to identify where specific harvest activities will occur in relation

to various cognizable values on Prince of Wales Island.119 Far from “unwarranted

116 Stein, 740 F. Supp. at 749. 117 Docket 12 at 25. 118 979 F. Supp. 2d 1118, 1125–28 (D. Mont. 2013). Moreover, the district court in Alliance for the Wild Rockies held that the Forest Service reasonably concluded that the sale at issue fell within a categorical exclusion, such that no NEPA analysis was required at all. Id. at 1124–25; see also 40 C.F.R. § 1508.4 (authorizing agencies to categorically exclude activities from NEPA’s analysis requirements). 119 As Plaintiffs maintained at oral argument: “The island complex is 130 miles long and 60 miles wide. So if you live in one of the 12 subsistence communities on this island, the timber sales that have been approved in this project might be right next door to you, or they might be a hundred miles away.” Docket 39 at 8:24–9:3.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. ‘fly-specking,’”120 Plaintiffs’ objections identify serious shortcomings in the sufficiency of the Project EIS’s environmental analysis. Moreover, the district court in Stein rejected the plaintiffs’ site-specificity

claims because they had not asserted or “show[n] why disclosure of more details regarding site-specific impacts [was] necessary in order to ‘foster both informed decision-making and informed public participation.’”121 Here, Plaintiffs maintain that more detailed information about the specific location of timber harvest under the Project is necessary to properly assess its ecological and subsistence

impacts.122 The Forest Service contends, however, that the EIS satisfies NEPA because it analyzes the Project’s maximum potential impacts.123 In its briefing before this Court, the agency describes the Project EIS as assuming that “the entire . . . Project Area would be harvested by clear-cut methods and that every

120 Stein, 740 F. Supp. at 749. 121 Id. (quoting Or. Envtl. Council v. Kunzman, 817 F.2d 484, 492 (9th Cir. 1987)). 122 See Docket 22-1 at 28–34. For example, Plaintiffs argue: [W]hile all of [the species occurring within the Project Area] depend to some degree on old growth, the extent of that dependence varies, and they have different needs respecting forest structure, elevation, proximity to beaches and streams, proximity to roads, prey availability, and fragmentation of their habitat. For all these reasons . . . the specific locations proposed for new logging and road construction matter a great deal for wildlife, for hunters, and for other people who use and enjoy the forest. Docket 22-1 at 34. 123 Docket 12 at 30–32.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. mile of road would be constructed up to the maximum number of harvest acres and miles of road authorized under each alternative.”124 The Forest Service maintains that as a result of this worst-case-scenario analysis, “whatever units [it]

ultimately selects within the constraints outlined in the alternatives, Activity Cards, and Implementation Plan, the Project will produce environmental effects that fall within those already disclosed and analyzed in the EIS.”125 It is not entirely clear from the record, but for at least some of the impacts analyzed, the Project EIS assumes that considerably more timber would be

harvested than is actually authorized under each alternative. For example, to determine the effects of each alternative on wildlife and subsistence, the Project EIS focused on the 32 Wildlife Analysis Areas (“WAA”)—“land divisions used by [the Alaska Department of Fish and Game] for wildlife analysis and regulating wildlife populations”—situated in the Project Area.126 In its briefing, the Forest

Service explained that “for each WAA, the Service assumed [timber] harvest would be concentrated in that WAA and would occur at the maximum level.”127 At oral argument, the agency confirmed that this approach could cause the EIS to assume that more harvest would occur under each alternative than actually allowed by that

124 Docket 12 at 31 (citing AR 833_2167 at 001449, 001634). 125 Docket 12 at 32 (citing WildEarth Guardians v. Conner, 920 F.3d 1245, 1258 (10th Cir. 2019)). 126 AR 833_2167 at 001514. 127 Docket 12 at 37.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. alternative.128 And language in the Project EIS itself indicates that this approach assumed that all potential harvest stands identified in the LSTA were cut, regardless of the acreage authorized by the alternative in question.129 The Project

EIS makes these assumptions despite its recognition that “[t]he specific location and amount of harvest in each WAA would be determined during implementation and vary by alternative.”130 The Forest Service cites WildEarth Guardians v. Conner to defend its approach.131 There, the Tenth Circuit upheld an Environmental Assessment (“EA”)

128 Docket 39 at 19:1–18. 129 See, e.g., AR 833_2167 at 001629 (“For purposes of analysis [of impacts to wildlife], assumptions include that all harvest stands from the LSTA would be harvested and the only harvest method would be clearcut.”); id. at 001634 (“The effects of the POW LLA Project [on wildlife] are similar between all alternatives because all alternatives assume that all acres proposed for timber harvest will be harvested.”). 130 AR 833_2167 at 001639. 131 Docket 12 at 17 n.5, 31, 32 (citing 920 F.3d 1245 (10th Cir. 2019)). The Forest Service also cites Protect Our Communities Foundation v. Jewell, No. 13CV575, 2014 WL 1364453, at *9 (S.D. Cal. Mar. 25, 2014), aff’d 825 F.3d 571 (9th Cir. 2016), to support its assertion that its maximum-impacts analysis satisfied NEPA. Docket 12 at 31. However, the narrow holding in that case is inapplicable to the Project EIS. There, the Bureau of Land Management (“BLM”) “adopted a cautious and conservative approach to measuring turbine noise” in completing the EIS for a proposed wind facility. 2014 WL 1364453, at *9. This analysis led BLM to conclude that the facility would result in adverse noise impacts and caused it to develop a “site-specific noise mitigation plan.” Id. The district court rejected the plaintiffs’ argument that BLM should have modelled turbine noise using larger turbines and that it had therefore “underestimate[ed] overall noise levels,” citing the agency’s scientific expertise. Id.; see also Native Ecosys. Council v. Weldon, 697 F.3d 1043, 1051 (9th Cir. 2012) (“A court generally must be ‘at its most deferential’ when reviewing scientific judgments and technical analyses within the agency’s expertise under NEPA.” (quoting N. Plains Res. Council, Inc. v. Surface Transp. Bd., 668 F.3d 1067, 1075 (9th Cir. 2011))). This issue was not raised on appeal, but the Ninth Circuit did address the BLM’s consideration of inaudible noise impacts and similarly deferred to the agency’s judgment. Prot. Our Cmtys. Found. v. Jewell, 825 F.3d 571, 584 (9th Cir. 2016). The present case does not concern a discrete “evaluation of complex scientific data” such as the proper method to model noise impacts from a wind turbine, id.; rather, it presents a broad legal question about the general structure of the Project EIS. Moreover, as Plaintiffs note, BLM in

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. for a tree-thinning project designed to address a beetle infestation in two national forests.132 The EA “evaluat[ed] the Project’s effects on lynx in a worst-case scenario in which all the mapped lynx habitat in the Project area is treated.”133 Due

to the listing of the Canada lynx as a threatened species, the Forest Service had previously amended the forest plans for the two forests to prohibit the clearcutting of more than 15% of lynx habitat or precommercial thinning of more than 1% of lynx habitat in a given analysis area after analyzing the impact of these changes in an EIS.134 In the subsequent EA for the tree-thinning project, the Forest Service

took “the conservative approach of assuming that all lynx habitat in the Project area w[ould] be treated.”135 Using that approach, the Forest Service found that only 6% of lynx habitat would be subject to clearcutting, and no more than 0.2% subject to precommercial thinning—well below the percentages prohibited by the governing forest plans.136 The Tenth Circuit held that because the project’s

impacts to lynx habitat were below the caps established in the amended forest

Protect our Communities Foundation “was not substituting the worst-case analysis for site- specific decisions,” but was instead using the analysis to assess the impact produced by a wind facility in a specific location. Docket 19 at 15. 132 920 F.3d at 1251. 133 Id. at 1258. 134 Id.at 1252–53, 1256. Precommercial thinning is “a process of thinning stands that were clear-cut 20-30 years earlier, so that growth can be concentrated on the more commercially valuable trees.” Id. at 1255. 135 Id. at 1255 (emphasis in original). 136 Id. at 1255–56.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. plans, which had been adopted after a full analysis of the lynx-habitat impacts in the EIS, the Forest Service “could reasonably assess the maximum impact that the Project could have on the lynx and conclude it was unlikely to adversely affect

them.”137 Although in the instant case the Forest Service applied an analytical framework similar to the one it used in WildEarth Guardians v. Conner, the difference between an EA and an EIS renders that case inapplicable. An EA is meant to determine whether a proposed action will have a significant impact on

the environment, such that an EIS is necessary.138 In contrast, an EIS must compare the environmental impacts of different alternatives, not just determine whether environmental impacts will occur.139 While an agency’s analysis of a proposed action’s maximum potential impacts may be appropriate for an EA, the Forest Service’s analytical framework in this case is not sufficient to meet the

requirements for an EIS.140

137 Id. at 1257. 138 See id. at 1251 (describing purpose of EA). 139 See id.; see also 40 C.F.R. § 1502.16(d) (requiring discussion of “[t]he environmental effects of alternatives including the proposed action”); 40 C.F.R. § 1502.14(a) (requiring EIS to “[r]igorously explore and objectively evaluate all reasonable alternatives”). 140 Indeed, the Tenth Circuit explained in WildEarth Guardians v. Conner that it had earlier held an EIS inadequate that did not base its analysis of impacts to vegetation and wildlife on the actual location of the project, “reasoning that ‘the location of development greatly influences the likelihood and extent of habitat preservation’” because “’[d]isturbances on the same total surface acreage may produce wildly different impacts on plants and wildlife depending on the amount of contiguous habitat between them.’” 920 F.3d at 1257–58 (quoting Richardson v. Bureau of Land Mgmt., 565 F.3d 683, 706–07 (10th Cir. 2009)); see also Docket 17 at 14 (Plaintiffs arguing that “[w]here . . . significant environmental impacts demand an EIS for a site-specific project, the

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. The Forest Service candidly acknowledges in the EIS that its analytical framework overestimates the Project’s impacts and is unlikely to reflect the actual extent and nature of activities under each of the proposed alternatives within the

Project Area.141 By focusing on the Project’s maximum potential impacts for all alternatives rather than its actual or foreseeable impacts for each alternative, the EIS falls short of NEPA’s directive to “contain[] a reasonably thorough discussion of the significant aspects of the probable environmental consequences” for each alternative.142 This approach, coupled with the lack of site-specific information in

the Project EIS, detracts from a decisionmaker’s or public participant’s ability to conduct a meaningful comparison of the probable environmental impacts among the various alternatives. For example, in the introduction to the section discussing the Project’s impacts to wildlife habitat, the EIS states that “[t]he effects . . . are similar between all alternatives because all alternatives assume that all acres

agency must disclose with specificity where it intends to undertake the project, what the impacts will be, and what alternatives are available.”). Yet the Tenth Circuit was careful to note that its holding in Richardson was not “that an agency’s EA or EIS always must specify the precise locations within a project area that will be affected.” WildEarth Guardians v. Conner, 920 F.3d at 1258 (emphasis added). 141 See, e.g., AR 833_2167 at 1634 (“The total acres estimated to be needed to meet timber needs are likely over-estimated and therefore the effects are likely over-estimated as well.”). The Forest Service’s briefing admits this as well, stating that the Project EIS “analyz[es] the Project’s maximum potential impacts, . . . even though in reality impacts would be less.” Docket 12 at 31; see also id. at 31 n.10 (noting that “under Alternative 2, about fifty-seven percent of the old-growth acreage proposed for harvest would be harvested using uneven-aged harvest prescriptions,” even though the EIS’s analysis assumed that all acres would be clear-cut). 142 WildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d 920, 924 (9th Cir. 2015) (emphasis added) (quoting City of Sausalito v. O’Neill, 386 F.3d 1186, 1206 (9th Cir. 2004)).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. proposed for timber harvest will be harvested. The analysis also assumes that all acres will be harvested by even-aged harvest methods.”143 Due to these identical assumptions for each alternative, it appears that at least with respect to wildlife

impacts, the Project EIS only meaningfully analyzed the different mitigation measures contained in each alternative, not the harvest limits.144 And where the Project EIS does differentiate between alternatives, it does so in partly conditional terms due to its lack of site-specific information. For example, when discussing impacts to high productive old growth habitat, the EIS concludes that Alternative 2

“may result in two [Wildlife Analysis Areas] dropping below 50% habitat remaining.”145 The Project EIS identified a total acreage of potential timber harvest, but not the distribution of the specific acreage authorized by each alternative within these areas. This omission is meaningful given the duration and scale of the project.146

Despite “additional parameters that limit the ultimate selection of units and

143 AR 833_2167 at 001634; see also supra note 43 and accompanying text. 144 See, AR 833_2167 at 001642–43 (differentiating impacts of alternatives to deep snow habitat purely in terms of mitigation measures). 145 AR 833_2167 at 001500 (emphasis added). 146 For example, Alternative 2 authorizes a maximum 42,635 acres of timber harvest—23,269 acres of old growth and 19,366 acres of young growth—which is roughly 34 percent of the 125,529 acres identified for potential old- and young-growth harvest in the Project Area. AR 833_2167 at 001481.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. activities,”147 such as mitigation measures contained in the Activity Cards,148 the Project EIS’s structure creates ambiguity about the actual location, concentration, and timing of timber harvest and road construction on Prince of Wales Island.149

By doing so, the Project EIS fails to provide a meaningful comparison of alternatives. By authorizing an integrated resource management plan but deferring siting decisions to the future with no additional NEPA review,150 the Project EIS violates NEPA. The Forest Service has not yet taken the requisite hard look at the

environmental impact of site-specific timber sales on Prince of Wales over the next 15 years. The Forest Service’s plan for condition-based analysis may very well streamline management of the Tongass and decrease the amount of falldown acreage associated with each timber sale;151 however, it does not comply with the

147 Docket 12 at 29. 148 See, e.g., AR 833_2168 at 001935–39 (Activity Card for “Rotational Harvest of Old Growth Using Even-aged Management”). 149 See Docket 19 at 9 (“Even the maximum level of logging . . . and road-building . . . could be distributed in countless combinations around the 125,529 acres allowed in the Project.”) 150 The ROD states that the Implementation Plan is to “provide a linkage from the [Project EIS] to the project-specific work without the need for additional NEPA analysis.” AR 833_2427 at 001034. As Plaintiffs note, the Forest Service’s “approach might be permissible if the agency were going to prepare a subsequent EIS” for decisions at the site-specific level. Docket 22-1 at 26; see also City of Tenakee Springs v. Block, 778 F.2d 1402, 1407 (9th Cir. 1985) (“Where there are large-scale plans for regional development, NEPA requires both a programmatic and a site-specific EIS.” (citing 40 C.F.R. §§ 1508.28, 1502.20)); California v. Block, 690 F.2d 753, 761 (9th Cir. 1982) (“The detail that NEPA requires in an EIS depends upon the nature and scope of the proposed action. . . . The critical inquiry in considering the adequacy of an EIS prepared for a large scale, multi-step project is not whether the project’s site-specific impact should be evaluated in detail, but when such detailed evaluation should occur.”). 151 Docket 12 at 10 (describing condition-based management as “superior method to analyze

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. procedural requirements of NEPA, which are binding on the agency.152 “NEPA favors ‘coherent and comprehensive up-front environmental analysis to ensure . . . that the agency will not act on incomplete information, only to regret its decision

after it is too late to correct.”153 Plaintiffs have therefore established Count I of the Complaint; the Project EIS violates NEPA and is therefore not in accordance with law.154 II. Alaska National Interest Lands Conservation Act Congress enacted ANILCA to “cause the least adverse impact possible on

rural residents who depend upon subsistence uses of the resources of [the public lands in Alaska].”155 To achieve this purpose, § 810 of ANILCA imposes

the environmental effects of multi-year, landscape level analyses such as this one”). At oral argument, the Forest Service explained that one motivation behind the decision to adopt condition-based analysis was to minimize downward reductions of harvest volume as sales are finalized, known as “falldown.” Docket 39 at 19:20–20:8; see also AR 833_2167 at 001561 (describing falldown). The Court notes that while the EIS does discuss falldown, neither it nor the ROD identifies falldown as a motivating factor in the Project’s design. 152 See Protect Our Cmtys. Found. v. LaCounte, 939 F.3d 1029, 1035 (9th Cir. 2019) (“NEPA requires . . . procedural steps but does not require an agency to reach any particular result.”). In its briefing, the Forest Service maintained that “resource concerns and mitigations measures [would be] further refined” through the Implementation Plan “after treatment units are identified.” Docket 12 at 29–30. But the Implementation Plan is not a valid substitute for NEPA. For example, NEPA provides for a minimum 45-day comment period on draft EISs, 40 C.F.R. § 1506.10(c), while the Implementation Plan allows for only a 30-day comment period. AR 833_2427 at 001035. The ROD notes that the Implementation Plan may be altered by the Forest Service, see AR 833_2427 at 000811, and the agency maintained at oral argument that the Project would be NEPA-compliant even if no Implementation Plan existed. Docket 39 at 18:6–13. 153 Protect Our Cmtys. Found., 939 F.3d at 1035 (alterations in original) (quoting Churchill Cty. v. Norton, 276 F.3d 1060, 1072–73 (9th Cir. 2001)). 154 See 5 U.S.C. § 706(2)(A). 155 16 U.S.C. § 3112(1); see also Amoco Prod. Co. v. Vill. of Gambell, 480 U.S. 531, 544 (1987) (“The purpose of ANILCA § 810 is to protect Alaskan subsistence resources from unnecessary

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. procedural requirements upon federal decisionmakers; pursuant to its terms, an agency proposing an action resulting in the “use, occupancy, or disposition of public lands” must evaluate that action’s effects on “subsistence uses and needs,”

the availability of other lands for the same purpose, and “other alternatives” that would reduce the impacts to subsistence uses.156 Upon determining that the action “would significantly restrict subsistence uses,” the agency must provide notice to the affected communities, hold public hearings, and make three findings: [T]hat (A) such a significant restriction of subsistence uses is necessary, consistent with sound management principles for the utilization of the public lands, (B) the proposed activity will involve the minimal amount of public lands necessary to accomplish the purposes of such use, occupancy, or other disposition, and (C) reasonable steps will be taken to minimize adverse impacts upon subsistence uses and resources resulting from such actions.157 When the proposed action requires completion of an EIS, the agency “shall provide the notice and hearing and include the findings required by [ANILCA § 810(a)] as part of such [EIS].”158 The Project EIS contains a section discussing the impacts to subsistence activities.159 This section recognizes that “[s]ubsistence hunting, fishing, trapping, and gathering activities are a major focus of life for many residents on Prince of

destruction.”). 156 16 U.S.C. § 3120(a). 157 16 U.S.C. § 3120(a)(1)–(3). 158 16 U.S.C. § 3120(b). 159 AR 833_2167 at 001540–1559.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. Wales Island.”160 The EIS notes that commenters were particularly concerned about the Project’s impacts to Sitka black-tailed deer, which the EIS “considered an ‘indicator’ for potential subsistence resource effects concerning the resources

associated with old-growth forest habitat.”161 Using the maximum-impact methodology described in the previous section of this order,162 the Forest Service determined that the Project presents “a significant possibility of a significant restriction for the use of deer,” largely “due to the loss of deep snow habitat in some WAAs [Wildlife Analysis Areas].”163 The Project EIS identified “five WAAs of

concern . . . [:] 1214, 1315, 1317, 1318, and 1420,” which are “near the communities of Thorne Bay, Coffman Cove, Hollis and Klawock.”164 In light of this determination, the Forest Service held seven subsistence hearings in Prince of Wales Island communities after issuing the draft EIS.165 In

160 AR 833_2167 at 001540. 161 AR 833_2167 at 001540. 162 See supra 26–27; see also Docket 12 at 37 (“In other words, for each WAA [Wildlife Analysis Area], the Service assumed [timber] harvest would be concentrated in that WAA and would occur at the maximum level.”) 163 AR 833_2167 at 1557. Regarding other subsistence resources, the Forest Service concluded that “[n]one of the project alternatives would present ‘a significant possibility of a significant restriction’ of subsistence uses for most subsistence resources (food plants, personal use timber, upland game birds and waterfowl, furbearers, and marine mammals).” AR 833_2167 001545; see also AR 833_2167 at 001548 (finding same for aquatic resources). 164 AR 833_2167 at 001550. The Project EIS noted that “[t]he Forest Plan estimates that with full implementation of the Forest Plan, WAAs 1420 and 1421 in the project area may retain 50 percent or less of the estimated deer habitat capability.” AR 833_2167 at 001554; see also AR 833_2167 at 1500–01 (showing impacts of various alternatives to different habitats). 165 AR 833_2167 at 001470. The Forest Service held subsistence hearings in Whale Pass,

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. response to concerns raised at two of these hearings, the agency decided “not to authorize commercial harvest of old-growth stands in the area ‘North of the 20 Road, and in VCU 5280.”166 In the final Project EIS, the Forest Service made the

three findings required by ANILCA § 810(a)(3).167 Plaintiffs contend that “[t]he lack of site-specific information in the [Project EIS] violates not only NEPA, but also section 810 of ANILCA.”168 They maintain that “analysis of impacts on subsistence uses under section 810 should be at least as site-specific as that for the environmental impacts under NEPA.”169 As such,

Plaintiffs also argue that the Forest Service’s § 810(a)(3) findings were premature since they were made “before deciding the specific location or extent of logging or road construction over the next 15 years.”170 The Forest Service argues that Plaintiffs have simply repackaged their NEPA claim.171 However, due to the similarities between the procedural

requirements of NEPA and ANILCA, courts have evaluated the two statutes under

Klawock (twice), Hydaburg, Point Baker, Kasaan, and Naukati. Id. 166 AR 833_2427 at 000787. The “area North of 20 Road” is on the northern tip of Prince of Wales Island, to the East of the communities of Point Baker and Port Protection. See AR 833_2178. 167 AR 833_2167 at 001558–59. 168 Docket 22-1 at 36; see also Docket 1 at 17–18, ¶¶ 48–51 (Count II in Complaint). 169 Docket 22-1 at 37. 170 Docket 22-1 at 39. 171 Docket 12 at 35.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. similar standards.172 Although there is little case law on the issue, at least one court has looked to NEPA decisions to determine the site-specificity required by ANILCA § 810. In City of Tenakee Springs v. Clough, the district court rejected

the plaintiffs’ claims that supplemental EISs were “deficient because they discuss site-specific impacts of proposed harvesting on subsistence resources but do not correlate those impacts with the specific subsistence needs of each affected community.”173 The court cited Stein v. Barton’s description of NEPA site- specificity as the relevant standard, stating that “[p]roposed activities must be

sufficiently correlated with environmental factors in each affected area to facilitate public discussion of the project.”174 But the court found that no additional site- specificity was necessary to comply with ANILCA because the “EIS’s identify site- specific impacts on subsistence resources, incorporate maps identifying which sites are important for subsistence use generally, and catalog how and to what

extent each community utilizes subsistence resources, including data on per capita consumption.”175

172 See, e.g., Alaska Wilderness Recreation & Tourism Ass’n v. Morrison, 67 F.3d 723, 731 (9th Cir. 1995) (evaluating adequacy of EIS’s consideration of alternatives under NEPA and ANILCA together); City of Tenakee Springs v. Clough, 915 F.2d 1308, 1310–12 (9th Cir. 1990) (same); Sierra Club v. Penfold, 664 F. Supp. 1299, 1307 (D. Alaska 1987) (“NEPA case law is helpful in interpreting § 810.”). 173 750 F. Supp. 1406, 1422 (D. Alaska 1990), rev’d on other grounds, 915 F.2d 1308 (9th Cir. 1990). 174 Id. at 1422 (quoting Stein v. Barton, 740 F. Supp. 743, 749 (D. Alaska 1990)). 175 Id.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. The Court has already determined that the Project EIS does not contain the level of site-specificity required by NEPA. For the same reasons, the Court finds that despite the public hearings it held and the findings it made, the Forest Service

has failed to comply with ANILCA § 810. The purpose of that section is to promote informed decision-making, such that the impacts of an action to subsistence activities are considered; those actions “which would significantly restrict subsistence uses can only be undertaken if they are necessary and if the adverse effects are minimized.”176 This purpose can only be fulfilled if specific information

about the actual, not the potential, proposed action is available. This is made clear by the statute’s command to consider site-specific aspects of a proposed action, such as its effect on local “subsistence uses and needs.”177 The Forest Service contends that it has fulfilled its duty under ANILCA by “identifying where Project activities would occur, evaluating the maximum impacts

of the Project on subsistence uses, and by taking actions to benefit subsistence uses and reduce the adverse effects of the project on . . . those uses.”178 However, as discussed above, the Commercial Vegetation Management map only identifies the potential, and not the actual, locations of timber harvest and road building within the 1.8-million acre Project Area. The Forest Service’s evaluation of the

176 Amoco Prod. Co. v. Village of Gambell, 480 U.S. 531, 544 (1987). 177 16 U.S.C. § 3120(a); see also supra note 119. 178 Docket 12 at 35–36.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. Project’s maximum impacts does not evaluate the Project’s actual expected impacts.179 By not developing actual site-specific information, the Forest Service limited its ability to make informed decisions regarding impacts to subsistence uses

and presented local communities with vague, hypothetical, and over-inclusive representations of the Project’s effects over a 15-year period.180 The Implementation Plan does envision additional information-gathering and public participation before site-specific decisions are made.181 However, as Plaintiffs note, this process does not require the Forest Service to make additional

findings under ANILCA § 810(a)(3) or provide the public with a right of appeal.182 Moreover, the Implementation Plan is subject to change by the Forest Service, and there is no certainty that its public participation provisions will last for the Project’s 15-year duration.183 Without either an up-front discussion of actual site-specific impacts or future ANILCA analysis when siting decisions are made, the Project

EIS and ROD are inconsistent with ANILCA § 810. Plaintiffs have therefore

179 See supra 30–32. As noted previously, see supra note 146, Alternative 2 authorized 42,635 acres of timber harvest, or 34 percent of the 125,529 acres identified for potential harvest by the LSTA, but gave no indication of where that harvest would occur within the total acreage. AR 833_2167 at 001481. 180 As Plaintiffs note, “subsistence activities are inherently location-specific,” and “[p]eople care about the places they use for subsistence and how the action will affect those places and nearby habitat.” Docket 19 at 18. 181 See AR 833_2427 at 001037 (describing implementation process). 182 Docket 19 at 19. 183 See AR 833_2427 at 000811 (“The plan is meant to be a ‘living’ document and may need to be adjusted . . . .”).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. established Count II of the Complaint; the Project EIS violates ANILCA and is therefore not in accordance with law.184 III. National Forest Management Act

Pursuant to NFMA, the Forest Service must prepare a land and resource management plan, also called a “forest plan,” for each forest it manages.185 Projects occurring in a national forest must comply with that forest’s management plan.186 Standard and Guideline TIM3.I.C in the 2016 Forest Plan that currently governs management of the Tongass provides:

Timber harvest unit cards will document resource concerns and protection measures. The unit cards, including a map with relevant resource features, will be provided electronically when Draft or Final NEPA documents and decisions are published. (Consult Tongass National Forest Supplement 1909.15-2015-1.)187 Tongass National Forest Supplement 1909.15-2015-1, which “[e]stablishe[d] procedures for producing and distributing unit and road cards associated with NEPA documents,”188 was rescinded by the Forest Service in October 2018.189

184 See 5 U.S.C. § 706(2)(A). 185 See 16 U.S.C. § 1604(a). 186 16 U.S.C. § 1604(i); see also Friends of Se.’s Future v. Morrison, 153 F.3d 1059, 1068 (9th Cir. 1998) (“[P]ursuant to the NFMA, the Forest Service must demonstrate that a site-specific project would be consistent with the land resource management plan of the entire forest.” (quoting Neighbors of Cuddy Mountain v. U.S. Forest Serv., 137 F.3d 1372, 1377 (9th Cir. 1998))). 187 AR 833_0404 at 63265. Forest plans for the Tongass have included a unit-card requirement since at least 1997. AR 833_2076 at 068765 (1997 Tongass Forest Plan). 188 AR 833_2526 at 074726. 189 AR 833_2525 at 074720–25.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. The Project EIS does not contain unit cards, and the ROD explains in the section describing Activity Cards that “[u]nit cards would be developed for any timber sales when site-specific locations are determined” through the

Implementation Plan.190 Plaintiffs contend that the “Forest Service violated the forest plan requirement to include unit cards with the draft or final EISs for the Prince of Wales Project.”191 Responding to comments arguing that the Project EIS had failed to comply with the Tongass Forest Plan, the Forest Service explained that it understood the

rescission of Tongass National Forest Supplement 1909.15-2015-1 to have rendered “the timing for when electronic unit cards are provided . . . no longer applicable.”192 The Forest Service added that the Implementation Plan’s “opportunity for public comment on the maps, and unit and road cards meets the intent of Forest Plan TIM3.I.C.”193

In its briefing to this Court, the Forest Service advances a different argument. It now contends that TIM3.I.C is ambiguous regarding “the contents

190 AR 833_2427 at 000826; see also id. at 000802 (“The Implementation Plan identifies that unit and road-specific cards will be developed when specific harvest units and road locations are determined as part of the Implementation Plan process.”) 191 Docket 22-1 at 39; see also Docket 1 at 18–20, ¶¶ 52–58 (Count III in Complaint). 192 AR 833_2440 at 000020 (Responses to Comments on Draft ROD); see also AR 833_2427 at 000802 (ROD determining that publishing unit cards during Implementation Plan “is an alternative way to fully comply with Forest Plan Standard TIM3.I.C, which is no longer applicable in terms of timing when unit cards are provided”). 193 AR 833_2171 at 002149 (Responses to Comments on Draft EIS).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. and format of unit cards, except that the cards ‘will document resource concerns and protection measures’ and ‘include[e] [sic] a map with relevant resource features.’”194 The Forest Service maintains that the Activity Cards and Commercial

Vegetation Management map provided with the Project EIS comply with this requirement. 195 The Forest Service argues that its interpretation of TIM3.I.C is entitled to Auer deference, which accords “defer[ence] to agencies’ reasonable readings of genuinely ambiguous regulations.”196 But Auer deference is not automatic; the

Supreme Court has explained that “before concluding that a rule is genuinely ambiguous, a court must exhaust all the ‘traditional tools’ of construction,” meaning it “must ‘carefully consider[]’ the text, structure, history, and purpose of a regulation.”197 Addressing first the Forest Service’s position in the administrative record,

the Court finds no ambiguity in TIM3.I.C about when the agency is required to provide unit cards, notwithstanding the rescission of the supplement cited therein. The Standard and Guideline clearly states that “[t]he unit cards . . . will be provided

194 Docket 12 at 42 (emphasis and alterations in original) (quoting AR 833_0404 at 063265). 195 Docket 12 at 41–45. 196 Kisor v. Wilkie, ___ U.S. ___, 139 S. Ct. 2400, 2408 (2019) (citing Auer v. Robbins, 519 U.S. 452 (1997)). 197 Id. at 2415 (2019) (second alteration in original) (first quoting Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 843 n.9 (1984), then quoting Pauley v. BethEnergy Mines, Inc., 501 U.S. 680, 707 (1991) (Scalia, J., dissenting))).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. electronically when Draft or Final NEPA documents and decisions are published.”198 Under the plain language of the Forest Plan, the Forest Service must provide unit cards when the relevant NEPA document is published.

Regardless of the provision’s intent,199 the Forest Service departed from the unambiguous directive of TIM3.I.C. The agency’s decision to delay the publication of unit cards until the Implementation Plan, after NEPA review was completed for the Project, is therefore inconsistent with the Forest Plan. The Court finds the Forest Service’s position in its briefing to be no more

convincing. While TIM3.I.C does not fully explain what a unit card should contain, it is clear from the Standard and Guideline’s language that each card must relate to a discrete geographic area, or “[t]imber harvest unit,” within the Project Area.200 The Commercial Vegetation Management map does not identify specific timber harvest units,201 and the Activity Cards “document resource concerns and

protection measures at the Project level.”202 They are not unit cards within the meaning of TIM3.I.C. 203

198 AR 833_0404 at 63265 (emphasis added). 199 See AR 833_2171 at 002149 (stating that Implementation Plan’s “opportunity for public comment on the maps, and unit and road cards meets the intent of the Forest Plan TIM3.I.C”). 200 AR 833_0404 at 63265. 201 AR 833_2178. 202 Docket 12 at 44; see also AR 833_2427 at 000826 (describing Activity Cards). 203 Friends of Se.’s Future v. Morrison, 153 F.3d 1059, 1069 (9th Cir. 1998) (“However, an agency’s interpretation does not control, where, as here, it is plainly inconsistent with the regulation at issue.”). The Forest Service has previously used the term “activity cards”

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. Moreover, even if TIM3.I.C were ambiguous, the current interpretation advanced in the Forest Service’s briefing would not be entitled to Auer deference. Throughout the administrative record, the Forest Service’s position is consistent:

provision of unit cards after the completion of NEPA review complies with TIM3.I.C due to the rescission of Tongass National Forest Supplement 1909.15-2015-1. It is only in its briefing to this Court that the agency asserts that the Activity Cards and the Commercial Vegetation Management map themselves constituted the unit cards required by TIM.3.I.C.204 “[A] court should decline to defer to a merely

‘convenient litigating position’ or ‘post hoc rationalization[n] advanced’ to ‘defend past agency action against attack.’”205 And the administrative record itself belies the Forest Service’s litigation position; in the ROD, the agency differentiated

collectively to refer to cards “used to explain site-specific proposed activities,” including “timber harvest units and proposed and existing roads.” AR 833_2084 at 061267 (EIS for Kuiu Timber Sale). Unlike the Activity Cards in this case, the Kuiu Timber Sale unit cards described harvest activities in specific geographic locations, including both graphical and narrative information for each harvest unit. See AR 833_2084 at 061279–80 (card for unit 101); see also id. at 061268– 76 (describing activity cards, as defined by that project, and locating individual harvest units and roads on map of Project Area). Although Forest Supplement 1909.15-2015-1 was still effective when the Kuiu Timber Sale EIS was prepared, the Court finds a comparison between the timber harvest unit cards in that case and the Activity Cards used here to be instructive. 204 At oral argument, the Forest Service maintained that the agency had twice asserted in the administrative record that the Activity Cards and Commercial Vegetation Management map themselves complied with TIM3.I.C. Docket 39 at 24:24–26:19 (citing AR 833_2427 at 000802 and AR 833_2171 at 002149). Not so. As discussed above, the two pages of the record cited by the Forest Service concern the timing of unit card publication. 205 Kisor v. Wilkie, ___ U.S. ___, 139 S. Ct. 2400, 2417 (2019) (alteration and emphasis in original) (quoting Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 155 (2012)).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. between Activity Cards and unit cards and explained that the latter would be provided through the Implementation Plan in an effort to comply with TIM3.I.C.206 Due to their failure to include timber harvest unit cards corresponding to

discrete geographic locations, the Project EIS and ROD are inconsistent with the 2016 Tongass Forest Plan. Plaintiffs have therefore established Count III of the Complaint; by not complying with the applicable forest plan, the Project violated NFMA and is therefore not in accordance with law.207

206 The ROD lays out the difference between Activity Cards and unit cards, explaining that the former inform but do not replace the latter: The POWLLA FEIS uses a condition-based approach where specific harvest units and roads will be determined during implementation through a collaborative public process and interdisciplinary review. Activity Cards and maps were included with the FEIS and Draft ROD and will also be part of this Final ROD. The Activity Cards were designed to honor the public process developed at the community level during the POW LLA NEPA process, to be a resource for the public and Forest Service resource specialists, to assist in alternative development, and to accompany the EIS to provide clarity for environmental effects analysis and guide implementation . . . . The Implementation Plan identifies that unit and road -specific cards will be developed when specific harvest units and road locations are determined as part of the Implementation Plan process. This Implementation Plan allows for more collaboration during implementation, and responsiveness to dynamic on-the-ground conditions, new science information, and public input. This process provides for publishing unit and road cards online and providing an opportunity for public review and comment before final line officer decisions on specific project activities are made. This has been clarified in the Implementation Plan, ROD Appendix 2, under step 4. I have determined this process is an alternative way to fully comply with Forest Plan Standard TIM3.1.C, which is no longer applicable in terms of the timing of when unit cards are provided. AR 833_2427 at 000802. 207 See 5 U.S.C. § 706(2)(A).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. Nevertheless, the Forest Service asserts that this error was not prejudicial.208 “Relief is available under the APA only for ‘prejudicial error.’”209 Plaintiffs challenging agency action bear the burden of showing prejudice, but this

is not “a particularly onerous requirement.”210 Plaintiffs have met their burden here. They argue that they “were prejudiced by the lack of unit cards that would have given them the opportunity to provide meaningful input on logging locations, impacts, and alternatives.”211 The Ninth Circuit has consistently held that an error that affects the public’s ability to meaningfully participate in the NEPA review

process is prejudicial.212 The Forest Service contends that Plaintiffs had the opportunity to comment on the Activity Cards and Commercial Vegetation Management map, which contain “proposed locations of timber harvest.”213 However, as discussed above,

208 Docket 12 at 45–47. 209 Drakes Bay Oyster Co. v. Jewell, 747 F.3d 1073, 1090–91 (9th Cir. 2014) (quoting 5 U.S.C. § 706). 210 Shinseki v. Sanders, 556 U.S. 396, 410 (2009). 211 Docket 19 at 24; see also supra note 119. 212 See, e.g., Or. Nat. Desert Ass’n v. Jewell, 840 F.3d 562, 571 (9th Cir. 2016) (holding error to be prejudicial where “the public was not able to tailor its comments to address concerns regarding the potential winter presence of sage grouse” because “baseline conditions [were] inadequately established”); Friends of Se.’s Future v. Morrison, 153 F.3d 1059, 1068 (9th Cir. 1998); cf. Cal. Wilderness Coal. v. U.S. Dep’t of Energy, 631 F.3d 1072, 1091–92 (9th Cir. 2011) (reaffirming Circuit’s “consistent case law holding that ‘harmless error’ requires a determination that the error ‘had no bearing on the procedure used or the substance of [the] decision reached’” (alteration in original) (quoting Paulsen v. Daniels, 413 F.3d 999, 1006 (9th Cir. 2005))). 213 Docket 12 at 46.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. these documents do not contain the level of site-specificity required by TIM3.I.C— or by NEPA for that matter—and did not allow Plaintiffs to meaningfully comment on the specific harvest activities that would have been identified on timber harvest

unit cards had they been published with the draft EIS.214 The Court therefore finds that the NFMA violation was prejudicial and that relief is warranted under the APA. IV. Proper Remedy Having determined that Plaintiffs prevail on all three counts in their Complaint, the Court turns to the question of the proper remedy. Plaintiffs request

a judgment declaring that the Project EIS “violates NEPA, section 810 of ANILCA, and NFMA,” and “vacating those portions of the ROD authorizing vegetation management and new road construction.”215 The Forest Service contends that remand without vacatur may be appropriate, and requests the Court to allow supplemental briefing to address the proper remedy.216

Vacatur is the default remedy under the APA, which directs reviewing courts to “hold unlawful and set aside” unlawful agency action.217 The Ninth Circuit has

214 Further, the ability to comment on draft unit cards produced pursuant to the Implementation Plan does not render the error harmless, see Docket 12 at 47, since the Implementation Plan is subject to change and does not provide for the same public participation and review process as NEPA. See supra note 152. 215 Docket 22-1 at 43. 216 Docket 12 at 47–49. 217 5 U.S.C. § 706(2)(A). Both partial and complete vacatur of the offending agency action are acceptable forms of relief under the APA. See Monsanto Co. v. Geertson Seed Farms, 561 U.S. 139, 165–66 (2010) (“If a less drastic remedy (such as partial or complete vacatur of APHIS’s deregulation decision) was sufficient to redress respondents’ injury, no recourse to the additional

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. explained that a court should “order remand without vacatur only in ‘limited circumstances,’” and “leave an invalid rule in place only ‘when equity demands.’”218 To determine whether to remand an action without vacatur, a court is to “weigh the

seriousness of the agency’s errors against ‘the disruptive consequences of an interim change that may itself be changed.’”219 The Forest Service maintains that “[b]oth considerations are informed by the Court’s decision on the merits,” and the agency “believes the consequences of vacatur would be extremely disruptive due to the recent low and uncertain supply

of timber in Southeast Alaska, which threatens businesses in the region.”220 The Forest Service maintains that “[t]he economic need for Project timber and the harm that would be caused by delaying timber harvesting activities authorized by the Project cannot be adequately addressed within the page limits for [its] merits brief.”221 In response, Plaintiffs note that the Forest Service “made a strategic

choice to devote more pages to the merits of the claims, with no basis to assume it was entitled to supplemental briefing on the remedy.”222 Since Plaintiffs did

and extraordinary relief of an injunction was warranted.”). 218 Pollinator Stewardship Council v. U.S. Envtl. Prot. Agency, 806 F.3d 520, 532 (9th Cir. 2015) (first quoting Cal. Cmtys. Against Toxics v. U.S. Envtl. Prot. Agency, 688 F.3d 989, 994 (9th Cir. 2012) then quoting Idaho Farm Bureau Fed’n v. Babbitt, 58 F.3d 1392, 1405 (9th Cir. 1995)). 219 Id. (quoting Cal. Cmtys. Against Toxics, 688 F.3d at 992). 220 Docket 12 at 48. 221 Docket 12 at 49. 222 Docket 19 at 26.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. choose to devote a section of their briefing to the remedy issue, they maintain that allowing supplemental briefing would unfairly prejudice them.223 The Court finds that supplemental briefing on the appropriate remedy could

be helpful. Any prejudice to Plaintiffs as a result of this delay will be eliminated because the Court will keep the preliminary injunction in effect until the appropriate remedy is determined. // //

// // // // //

//

223 Docket 19 at 26.

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al. CONCLUSION In light of the foregoing, IT IS HEREBY ORDERED that: Plaintiffs’ request at Docket 10 for declaratory relief is GRANTED. The

Project EIS violates NEPA, 42 U.S.C. § 4332(2)(C), ANILCA, 16 U.S.C. § 3120(a), and NFMA, 16 U.S.C. § 1604(i). It is therefore “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”224 Each party may file a supplemental brief addressing the proper remedy in this case within 21 days of the date of this order; such brief shall not exceed

15 pages. Each party will then have an additional 14 days to file a response, not to exceed 8 pages. The preliminary injunction at Docket 27 shall remain in place until the Court enters a final judgment.

DATED this 11th day of March, 2020 at Anchorage, Alaska.

/s/ Sharon L. Gleason UNITED STATES DISTRICT JUDGE

224 5 U.S.C. § 706(2)(A).

Case No. 1:19-cv-00006-SLG, SEACC, et al. v. U.S. Forest Serv., et al.