ORDER

In 2008, the United States Forest Service adopted a “Public Wheeled Motorized Travel Management Decision” for the Eldorado National Forest (“Travel Management Decision” and “ENF”). This decision designates specific roads and trails within the forest as opeh to public motor vehicle use and correspondingly prohibits cross-country motorized travel. Plaintiffs, three non-profit organizations dedicated to their perception of environmental protection, challenge the Travel Management Decision. Although plaintiffs invoke a variety of statutes and legal theories, their general position is that the decision leaves too many routes open to vehicle use. Four groups representing the interests of recreational vehicle users have intervened as defendants.

The parties have filed cross motions for summary judgment. The court resolves these motions on the administrative record, the parties’ briefing, and after oral argument. For the reasons stated below, each motion is granted in part. The Forest Service violated its obligation under section 7(a)(2) of the Endangered Species Act to consult with the Fish and Wildlife Service, and the decision designates routes through meadows in apparent violation of provisions of the governing forest plan, thereby violating the National Forest Management Act.

I. BACKGROUND

A. The Eldorado National Forest

The Eldorado National Forest lies west of Lake Tahoe and east of Sacramento, in the Sierra Nevada mountain range. The ENF contains over 789,994 acres of diverse topography, soil types, vegetation, and habitat types.

Plaintiffs assert that the ENF “provides habitat for numerous endangered, threatened, and sensitive wildlife species, species of concern, and management indicator species.” Pis.’ Br. 7 (Dkt. 52-1). The only particular species at issue in this order is the California red-legged frog, Rana aurora draytonii, as plaintiffs have not provided arguments regarding any other species.

B. The National Environmental Policy Act

Plaintiffs’ arguments turn on three statutes. The first is the National Environmental Policy Act, 42 U.S.C. § 4321 et seq. The Forest Service’s approval of the challenged Travel Management Decision was a “major Federal action[] significantly affecting the quality of the human environment.” 42 U.S.C. § 4332 (C). Accordingly, the NEPA required the Forest Service to prepare an Environmental Impact Statement (“EIS”) for the project. The Ninth Circuit recently summarized the structure and purpose of this requirement:

Or. Natural Desert Ass’n v. Bureau of Land Mgmt., 625 F.3d 1092, 1099-1100 (9th Cir.2010) (some internal citations, quotations, and modifications omitted).

NEPA specifies various information that must be included in an EIS. The “heart” of the EIS is an examination of reasonable alternatives to the proposed action. Id. at 1100 (citing 42 U.S.C. § 4332 (C); 40 C.F.R. § 1502.14 ). “[T]he agency must ‘[rjigorously explore and objectively evaluate all reasonable alternatives,’ and for alternatives which were eliminated from detailed study, briefly discuss the reasons for their having been eliminated.” Id. (quoting 40 C.F.R. § 1502.14 ).

C. The National Forest Management Act and the Multiple-Use Sustained-Yield Act

The second statute plaintiffs invoke is the National Forest Management Act (“NFMA”), 16 U.S.C. §§ 1600-1614 . NFMA, together with the Multiple-Use Sustained-Yield Act (“MUYSA”), 16 U.S.C. §§ 528-531 , provides the primary guidance to the management of the National Forests. Substantively, MUYSA directs the Forest Service to administer the national forests “for outdoor recreation, range, timber, watershed, and wildlife and fish purposes.” 16 U.S.C. § 528 . NFMA adds “wilderness” as an additional purpose, reflecting the passage of the Wilderness Act of 1964. 16 U.S.C. §§ 1604 (e), 1311 et seq.

The Forest Service applies NFMA through several layers of management. At a broad level, NFMA requires the Forest Service to develop a forest plan for each forest. 1 Once a forest plan is adopted, individual management actions within that forest must comply with the plan. 16 U.S.C. § 1604 (i); Lands Council v. McNair, 537 F.3d 981, 989 (9th Cir.2008) (en banc). The Forest Service evaluates a proposed project’s compliance with the applicable forest plan during the NEPA process. Inland Empire Pub. Lands Council v. U.S. Forest Serv., 88 F.3d 754 , 757 (9th Cir.1996). NFMA therefore injects a substantive component into the NEPA’s otherwise procedural requirements. In addition to the standard NEPA documents, the Forest service must prepare a “Biological Evaluation” for proposed actions “to determine their potential effect on sensitive species.” Forest Service Manual (“FSM”) 2670.32 ¶ 2. This document discusses species protected under the Endangered Species Act and those species the Forest Service has itself designated as sensitive.

The Forest Service has promulgated a forest plan for the Eldorado National Forest. Discussion of this plan is complicated by the fact that the plan is not codified in a single document. In 1989, the Forest Service adopted what was then a comprehensive forest plan for the Eldorado National Forest. Admin. Record (“AR”) 3396. In 2004, the Forest Service promulgated the Sierra Nevada Forest Plan Amendments (“SNFPA”), which amended the forest plans for eleven national forests in the Sierra Nevada region, including Eldorado National Forest. AR 10,938. 2 The *1143 SNFPA did not supersede or modify any pertinent provisions of the ENF Forest Plan, but the SNFPA did add various supplemental provisions. For convenience, the court refers to the provisions adopted in 1989 as the “ENF Forest Plan” and to the provisions added by the SNFPA in 2004 as the “Sierra Nevada Forest Plan.” See Earth Island Inst. v. U.S. Forest Serv., 351 F.3d 1291 , 1296 (9th Cir.2003) (adopting similar terminology). The court recognizes, however, that this terminology is technically inaccurate, in that both are aspects of a single forest plan.

D. The Endangered Species Act

Yet another statute at issue in this case is the Endangered Species Act (“ESA”), 16 U.S.C. 1531 et seq. The ESA “reflects ‘a conscious decision by Congress to give endangered species priority over the “primary missions” of federal agencies.’ ” Cal. ex rel. Lockyer v. U.S. Dep’t of Agrie., 575 F.3d 999, 1018 (9th Cir.2009) (quoting Tenn. Valley Auth. v. Hill, 437 U.S. 153, 185 , 98 S.Ct. 2279 , 57 L.Ed.2d 117 (1978)). The ESA protects species that have been “listed” as “threatened” or “endangered.” ESA § 4(c), 16 U.S.C. § 1533 (c); 50 C.F.R. § 402.01 . In this case, plaintiffs’ arguments pertain to a single listed species, the California red-legged frog, which is listed as “threatened.” 61 Fed. Reg. 25813 (May 23,1996).

Federal agencies must ensure that their actions “[are] not likely to jeopardize the continued existence of any endangered species or threatened species or result in the destruction or adverse modification of’ habitat that has been designated as critical to the species. ESA § 7(a)(2), 16 U.S.C. § 1536 (a)(2). The ESA provides a framework for cooperation between agencies to achieve this goal. The Fish and Wildlife Service (“FWS”) has primary expertise regarding and authority over the California red-legged frog, and over non-marine species generally. FWS is referred to as a “Service” under the ESA. Other federal agencies, including the Forest Service, must satisfy their section 7 obligations “in consultation with and with the assistance of’ FWS. Id. In this process, the Forest Service is referred to as the “action agency.”

The ESA and implementing regulations provide a procedural framework for consultation regarding proposed actions. In this case, the Forest Service was first required to prepare a Biological Assessment. ESA § 7(c)(1), 16 U.S.C. § 1536 (c)(1); 50 C.F.R. § 402.12 . The Biological Assessment determines, inter alia, whether the proposed action “may affect” listed species. 50 C.F.R. § 402.14 . If the action agency determines that the action “may affect” listed species, the agency must consult with the appropriate Service. ESA § 7(a)(2), 16 U.S.C. § 1536 (a)(2); 50 C.F.R. § 402.14 (a); see also Cal. ex rel. Lockyer, 575 F.3d at 1018 (“consultation is required whenever a federal action ‘may affect listed species.’ ”). Consultation may be formal or informal. 50 C.F.R. § 402.13 (a). During informal consultation, FWS determines whether the proposed action is “not likely to adversely affect” the listed species. Id. If the Service determines that the proposal is not likely to adversely affect any species or critical habitat, then “the consultation process is terminated, and no further action is necessary.” Id. If FWS cannot reach this conclusion, then formal consultation is necessary, during which the Service prepares a Biological Opinion. Ground Zero Ctr. for Non-Violent Action v. U.S. Dep’t of Navy, 383 F.3d 1082 , 1091-92 (9th Cir.2004). If FWS determines that the proposed action would violate Section 7’s “jeopardy” standard, FWS must set forth one or more reasonable and prudent alternatives that would avoid jeopardy. ESA § 7(b)(3)(A), 16 U.S.C. § 1536 (b)(3)(A); 50 C.F.R. § 402.14 .

*1144 A further permutation is that action agencies may engage in “programmatic consultation.” Ordinarily, the consultation process looks to a site-specific action. Programmatic consultation instead concerns planning documents and other scenarios in which an agency is preparing to undertake a number of later, similar actions, the specifics of which have not yet been defined. See Gifford Pinchot Task Force v. U.S. Fish and Wildlife Serv., 378 F.3d 1059 , 1062 (9th Cir.2004); Pac. Coast Fed’n of Fishermen’s Ass’ns v. Nat’l Marine Fisheries Serv., 482 F.Supp.2d 1248, 1267 (W.D.Wash.2007); Buckeye Forest Council v. U.S. Forest Serv., 378 F.Supp.2d 835, 843-44 (S.D.Ohio 2005).

In 2006, the Forest Service engaged in informal programmatic consultation regarding route designation in the Sierra Nevada. AR 1, 31-33. This consultation acknowledged that route designations “may affect” listed species. FWS agreed, however, that if route designations adhered to a list of six “design criteria,” then the designated routes would not be likely to adversely affect listed species, such that no further consultation would be necessary. AR 1, 3-4. Because plaintiffs do not challenge the validity of the programmatic consultation or these design criteria, the court assumes that this type of consultation complies with the ESA.

Section 7’s prohibition on federal actions that would jeopardize species or their critical habitat is not the only method by which the ESA seeks to protect and recover listed species. Notably, the Services must develop and implement recovery plans for listed species, which identify measures that will allow the species to be delisted. ESA § 4(f), 16 U.S.C. § 1533 (f). FWS adopted a recovery plan for the California red-legged frog in 2002. AR 15,843.

E. The Travel Management Rule

In addition to the above statutes, plaintiffs rely on the Forest Service’s Travel Management Rule, a regulation codified at 36 C.F.R. part 215. Plaintiffs invoke sub-parts A and B of this rule.

What is now Subpart A of the Travel Management Rule was first promulgated in 2001. Administration of Forest Transportation System, 66 Fed. Reg. 3206 (Jan 12, 2001), codified at 36 C.F.R. §§ 212.1 to 212.21. Subpart A requires the Forest Service to determine, for each National Forest, the “minimum road system needed for safe and efficient travel and for utilization, and protection of National Forest System lands.” 36 C.F.R. § 212.5 (b)(1). The notice of final rulemaking explained that this system is the minimum that will serve “forest health, emergency access, and public access needs,” and that the system must “eomplfy] with resource objectives, ... reflect likely funding, and ... minimize adverse environmental effects associated with road construction, reconstruction, and maintenance.” 66 Fed. Reg. at 3208, 3207. Subpart A further obliges the Forest Service to concurrently “identify the roads ... that are no longer needed to meet forest resource management objectives and that, therefore, should be decommissioned or considered for other uses, such as for trails.” 36 C.F.R. § 212.5 (b)(2). The requirement to identify roads for decommissioning is “[ejqually important” as the overall identification of the minimum road system. 66 Fed. Reg. at 3207. Both decisions must be based on “a science-based roads analysis at the appropriate scale.” 36 C.F.R. § 212.5 (b)(2).

Subpart B of the Travel Management Rule was promulgated four years later, in 2005. Travel Management; Designated Routes and Areas for Motor Vehicle Use, 70 Fed. Reg. 68,264 (Nov. 9, 2005), codified at 36 C.F.R. §§ 212.50-212.57 . Underlying *1145 the pertinent portions of Subpart B, however, are executive orders dating back to 1972. In 1972, President Nixon issued Executive Order No. 11,644, which recognized the potential for conflict between off-highway vehicle (“OHV”) 3 use and other land management goals. To limit these conflicts, Executive Order 11,644 directed federal land management agencies, including the Forest Service, to adopt regulations providing for administrative designation of areas and trails open and closed to motor vehicle use. Exec. Order No. 11,-644, §§ 1, 3; 37 Fed. Reg. 2877 (Feb. 9, 1972). These designations must “be based upon the protection of the resources of the public lands, promotion of the safety of all users of those lands, and minimization of conflicts among the various uses of those lands.” Id. § 3. The executive order further provides four “minimization criteria:”

Id. This executive order was amended in 1977 to further direct land management agencies to immediately close areas or trails that “caus[ed] considerable adverse effects” upon protected resources. Exec. Order No. 11,989, § 2, 42 Fed. Reg. 26,959 (May 24,1977).

In adopting Subpart B of the Travel Management Rule, the Forest Service explained that “the magnitude and intensity of motor vehicle use have increased to the point that the intent of [Executive Orders] 11644 and [] 11989 cannot be met while still allowing unrestricted cross-country travel.” 70 Fed. Reg. at 68,265. Subpart B addresses this changed circumstance by eliminating unrestricted cross-country recreational OHV use and restricting motorized vehicle use to roads and trails designated for that purpose. 36 C.F.R. § 212.55 (a). Under Subpart B, the Forest Service’s decision of which routes and areas to open to motor vehicle use must consider:

36 C.F.R. § 212.55 (a). Subpart B further codifies the Executive Order’s four “mini *1146 mization criteria,” providing that the Forest Service:

36 C.F.R. § 212.55 (b). In this litigation Federal defendants argue that the “shall consider ... with the objective of minimizing” language “does not require the Forest Service to minimize these impacts.” Fed. Defs.’ Br. 21 (Dkt. 57-2). The court rejects this interpretation of the regulation. Accord Idaho Conservation League v. Guzman, 766 F.Supp.2d 1056, 1074 (D.Idaho 2011); see also Part 111(A)(2) below. The language itself does not support the Forest Service’s analysis. More importantly, this aspect of Subpart B codifies Executive Order 11,644, and the executive order plainly states that the land management agencies “shall ... minimize” the four types of impacts. Exec. Order 11,644 § 3. Accordingly, in this regard, Subpart B is equivalent to the Bureau of Land Management’s corresponding regulation interpreting Executive Order 11,644. Guzman, 766 F.Supp.2d at 1074, n. 6 . See also Ctr. for Biological Diversity v. Bureau of Land Mgmt., 746 F.Supp.2d 1055, 1075-76 (N.D.Cal.2009) (“CBD v. BLM”) (BLM’s regulation, 43 C.F.R. § 8342.1 , compels BLM to minimize impacts).

Finally, under Subpart B the Forest Service must consider the “[cjompatibility of motor vehicle use with existing conditions in populated areas, taking into account sound, emissions, and other factors.” 36 C.F.R. § 212.55 (b)(5).

F. The Challenged Travel Management Decision

Plaintiffs’ suit is only the most recent chapter in two decades of litigation regarding vehicle use in the ENF. The Forest Service first promulgated an OHV plan for the ENF in 1990. That plan was challenged in court, culminating in a 2005 order by the undersigned directing the Forest Service to withdraw the 1990 plan. See Ctr. for Sierra Nevada Conservation v. Berry, No. 2:02-cv-00325-LKK-JFM (E.D.Cal. Aug. 16, 2005) (Order on Remedies (Dkt. 163)) (“Berry order”). The Berry order required the Forest Service to, among other things, withdraw the 1990 Plan and to “issue a Final Environmental Impact Statement and Record of Decision on a new ENF OHV Plan (or site-specific area plans)” by December 31, 2007. Berry Order at 1. This deadline was subsequently extended to April 2, 2008. The Travel Management Decision challenged in this suit was prepared in order to comply both with Subpart B of the Travel Management Rule and the Berry order. See AR 2440 (EIS at 1-9).

At the time of the Berry order, the ENF contained 2,342 miles of official National Forest System roads and trails open to public motorized use. AR 2396 (EIS at xi). These included roads at maintenance levels (ML) 1 through 5. ML-3, ML-4, and ML-5 roads are “managed for standard four wheel passenger vehicles.” Id. at 2395 . ML-2 roads are “maintained for high clearance vehicles” and “are generally not suitable for standard four wheel passenger vehicles.” Id. ML-1 roads in the ENF “were designed to be intermittently used service roads and were intended to be closed to public wheeled motor vehicle use, although a majority of them are no longer *1147 physically closed.” Id. at 2396 . Trails, which do not receive an “ML” classification, are distinct from roads in that trails are narrower and typically maintained only for motorcycle or all-terrain vehicle use.

In addition to these official routes, the ENF contained 526 miles of “unauthorized routes” where “use [was] continuing to occur.” Id. Unauthorized routes are generally user-created routes, e.g., routes that develop as a result of repeated cross-country travel by the public. Although these routes had not received official approval, designation, or maintenance, they were not necessarily “illegal” prior to the Berry order, because off-road and cross-country travel was previously permitted in the ENF. See Guzman, 766 F.Supp.2d at 1063 n. 3 (explaining the creation of unauthorized routes in national forests). The Berry order prohibited motorized use of unauthorized routes pending release of a new travel management decision.

The Forest Service released a draft EIS for the Travel Management Decision on July 20, 2007. AR 1476. The draft EIS discussed trails, ML-1 and ML-2 roads, and unauthorized routes. AR 1481. 4 The draft EIS examined a “no action” alternative, alternative A, and four action alternatives, B through E. The “no-action” alternative would allow use to continue on all existing routes and would allow cross-country travel. AR 1487. Alternatives B through E would prohibit cross-country travel and open decreasing numbers of roads and trails to vehicle use, with B authorizing the most miles of routes and E authorizing the least. AR 1491. Each of these alternatives permitted vehicle use on some previously unauthorized routes, ranging from 17.7 to 45.8 miles of such routes. Id. The preferred alternative was D, which authorized motorized travel on 1,061 miles of routes, 34.8 of which were previously unauthorized. Id.

In conjunction with the draft EIS, the Forest Service released a draft Biological Evaluation (under NFMA) addressing each alternative’s impacts on sensitive species, AR 12,638, and a Biological Assessment (under the ESA) assessing alternative D’s impacts on listed species, including the California red-legged frog. AR 12,869. The Forest Service submitted the draft Biological Assessment and draft EIS to FWS for informal consultation. AR 13. FWS concluded that alternative D was “not likely to adversely affect” listed species. AR 2354-55. In its concurrence, FWS stated that the ESA would require no further action “[u]nless new information reveals effects of the proposed action that may affect federally listed species in a manner or to an extent not considered, or a new species or critical habitat is designated that may be affected by the proposed action.” Id.

The Forest Service released its final EIS and Record of Decision (“ROD”) in March of 2008. AR 2387, 3270. Rather than adopt the previously preferred alternative, the ROD adopted a modified form of alternative B. Like the original alternative B, Modified B authorized a greater number of routes than any of the other action alternatives. AR 2389. Modified B differed from B as to the particular routes authorized, however, incorporating changes intended to protect environmental resources. Id. In total, Modified B authorized 1,212 miles of trails, ML-1, and ML-2 roads, of which roughly 23 miles were previously unauthorized. AR 2475. The ROD selected alternative Modified B “because it provides a balanced response to the public comments by satisfying many recreation and social benefit criteria while providing increased protection for resources.” AR 3277. The ROD also adopted a “non-significant” amendment to *1148 three provisions of the ENF Forest Plan, exempting twenty enumerated routes from compliance with these provisions. AR 3273-35.

In connection with the final EIS, the Forest Service also issued an updated Biological Evaluation and Biological Assessment, The new Biological Assessment again concluded that the proposal was “not likely to adversely affect” any listed species. The Forest Service relied on the 2006 programmatic consultation and did not further consult with FWS.

Plaintiffs filed an administrative appeal challenging the ROD and FEIS. On June 27, 2008, this appeal was denied. AR 321, 804. On September 4, 2009, plaintiffs submitted a notice of intent to sue under the ESA to both the Forest Service and FWS. This suit followed.

II. STANDARD

Plaintiffs argue that the Travel Management Decision violated NEPA, NFMA, the Travel Management Rule, and the ESA. Plaintiffs’ NEPA, NFMA, and Travel Management Rule claims are brought under section 706(2) of the Administrative Procedure Act. Ecology Ctr., Inc. v. Austin, 430 F.3d 1057 , 1062 (9th Cir.2005) (NFMA and NEPA claims are brought under 5 U.S.C. § 706 (2)); Skranak v. Castenada, 425 F.3d 1213, 1218-19 (9th Cir.2005) (claim that Forest Service violated its own regulation is brought under 5 U.S.C. § 706 ). Under section 706(2), the court will set aside agency action that is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law” or that is undertaken “without observance of procedure required by law.” 5 U.S.C. §§ 706 (2)(A), (2)(D). Plaintiffs’ ESA claim, which argues that the Forest Service violated an obligation to consult with FWS, is brought under the ESA’s citizen suit provision, 16 U.S.C. § 1540 (g)(1). W. Watersheds Project v. Kraayenbrink, 632 F.3d 472 , 495 (9th Cir.2011); Wash. Toxics Coal. v. Envtl. Prot. Agency, 413 F.3d 1024, 1034 (9th Cir.2005). The ESA claim is nonetheless reviewed under the same APA standards of review. W. Watersheds Project, 632 F.3d at 495-96. 5

Ml parties have moved for summary judgment. Under APA section 706(2) review, the court does not employ the usual summary judgment standard. Conservation Cong. v. U.S. Forest Serv., 555 F.Supp.2d 1093, 1100 (E.D.Cal.2008). This is because the court is not generally called upon to resolve facts in reviewing agency action. Occidental Eng’g Co. v. Immigration and Naturalization Serv., 753 F.2d 766, 769-70 (9th Cir.1985). Instead, the court’s function is to determine whether or not, as a matter of law, the evidence in the administrative record permitted the agency to make the decision it did. Id.

An agency decision is arbitrary and capricious where the agency “relied on factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Lands Council, 537 F.3d at 987 (quotations omitted). The agency “must articulate a rational connection between the facts found and the conclusions reached.” *1149 Earth Island Inst. v. U.S. Forest Service, 442 F.3d 1147, 1157 (9th Cir.2006) (citing Midwater Trawlers Co-op. v. Dep’t of Commerce, 282 F.3d 710 , 716 (9th Cir.2002)).

Especially in a case such as this one, where the agency must comply with a multitude of obligations, many of which pull the agency in competing directions, and which collectively lead to a record of tens of thousands of pages, this standard extends beyond mere deference to the agency’s considered judgment. The court will additionally overlook minor gaffes in the record. This added deference is demonstrated by the instruction to “uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned.” Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 , 103 S.Ct. 2856 , 77 L.Ed.2d 443 (1983). Under NEPA in particular, a court “may not ‘fly-speck [the EIS] and hold it insufficient on the basis of inconsequential, technical deficiencies.’” Friends of the Southeast’s Future v. Morrison, 153 F.3d 1059 , 1063 (9th Cir.1998) (quoting Swanson v. U.S. Forest Serv., 87 F.3d 339 , 343 (9th Cir.1996)). The inquiry is “ ‘whether claimed deficiencies in a FEIS are merely flyspecks, or are significant enough to defeat the goals of informed decision making and informed public comment.’ ” Nevada v. Dep’t of Energy, 457 F.3d 78, 93 (D.C.Cir.2006) (quoting Fuel Safe Wash. v. Fed. Energy Regulatory Comm’n, 389 F.3d 1313, 1323 (10th Cir.2004)).

III. ANALYSIS

The parties seek summary judgment on five claims. Plaintiffs argue (1) that the Forest Service was required to identify a minimum road system under Subpart A of the travel management rule before reaching a travel management decision under Subpart B; that the Forest Service violated NEPA by (2) failing to consider an adequate range of alternatives and (3) failing to include sufficient site-specific information; (4) that the travel management decision designates routes in violation of the governing forest plan, violating NFMA; and (5) that the Forest Service failed to adequately consult with FWS regarding the California red-legged frog, violating the ESA. 6

Defendants do not challenge plaintiffs’ standing to bring these claims, and the court has reviewed plaintiffs’ standing declarations and determined standing to be proper.

A. Plaintiffs’ “Subpart-A” Claim

Plaintiffs’ primary argument in this suit is that the Travel Management Rule required the Forest Service to complete the Subpart A identification of a minimum road system before the Subpart B designation of roads for public use. This claim turns primarily on a question of regulatory interpretation. The Ninth Circuit has held that an agency’s interpretation of the agency’s own regulation is ordinarily controlling, such that courts will defer to the agency interpretation “unless an alternative reading is compelled by the regulation’s plain language or by other indications of the agency’s intent at the time of the regulation’s promulgation.” Bassiri v. Xerox Corp., 463 F.3d 927, 930-31 (9th Cir.2006) (citing Auer v. Robbins, 519 U.S. 452, 461-63 , 117 S.Ct. 905 , 137 L.Ed.2d 79 *1150 (1997)). This raises the question of what counts as an agency interpretation. In this case, many of the purported “interpretations” cited by defendants are not of the types that have previously received Auer deference. Nonetheless, as the court explains below, the Forest Service Manual suggests that the Forest Service may address Subparts A and B in any order, and plaintiffs have not shown that the language of the Travel Management Rule “compels” plaintiffs’ interpretation. It follows that the Forest Service’s decision not to first complete the Subpart A analysis was neither arbitrary nor in violation of the procedures required by law. 7

1. The Travel Management Rule Is Ambiguous

The Ninth Circuit has articulated two different frameworks for regulatory interpretation. Bassiri used a two-step process, beginning with the agency interpretation and then looking to whether an alternative interpretation was “compelled.” Other courts begin with the plain language, looking to agency interpretation only if the regulation is ambiguous, and then looking to whether the interpretation is unreasonable. See, e.g., Bayview Hunters Point Cmty. Advocates v. Metro. Transp. Comm’n, 366 F.3d 692 , 698 (9th Cir.2004); Hells Canyon Alliance v. U.S. Forest Serv., 227 F.3d 1170, 1180 (9th Cir.2000). These two frameworks are presumably equivalent. Here, the court concludes that the Travel Management Rule is ambiguous, because no language in the rule specifically addresses the timing of the Forest Service’s obligations under Subparts A and B. Plaintiffs’ contention that the structure of the rule compels plaintiffs’ interpretation is better discussed through the prism of reasonableness.

2. The Agency’s Purported Interpretations of The Travel Management Rule

The Forest Service argues that it has interpreted the Travel Management Rule as permitting the Forest Service to begin with Subpart B. The Forest Service appears to argue that the EIS, the Forest Service Manual, and the Forest Service’s own litigation position are all interpretations warranting deference. Plaintiffs argue that the Forest Service overstates the degree to which it is entitled to deference. More significantly, plaintiffs cite some of the same language in the record as indicating that the Forest Service had itself adopted plaintiffs’ interpretation of the regulation. Below, the court reviews the law underlying deference to agency interpretation of regulations and then examines the potential interpretive documents identified by the parties. The EIS neither meaningfully interpreted the Travel Management Rule nor conceded that Subpart A must precede Subpart B. The agency’s litigation position is not entitled to deference. The Forest Service Manual is the type of agency statement entitled to deference, and although this manual does not address the sequencing question in the most explicit of terns, the manual provides implicit support for the Forest Service’s position.

a. Deference to Agencies under Auer

The law regarding agency interpretation of regulations may be best understood in juxtaposition with the law regarding agency interpretation of statutes. As to statutes, courts have recognized two *1151 distinct forms of deference to agency interpretations of statutes. Courts defer to agencies under Chevron U.S.A. v. Natural Res. Def. Council, 467 U.S. 837, 842-45 , 104 S.Ct. 2778 , 81 L.Ed.2d 694 (1984) where Congress has delegated lawmaking authority to the agency and the agency has interpreted a statute in the exercise of this authority. United States v. Mead Corp., 533 U.S. 218, 229-30 , 121 S.Ct. 2164 , 150 L.Ed.2d 292 (2001). If the statute is ambiguous, the agency properly exercised its delegated authority in reaching the interpretation, and the agency interpretation is reasonable, then courts must adhere to the agency interpretation. Wilderness Soc’y v. U.S. Fish and Wildlife Serv., 353 F.3d 1051, 1060 (9th Cir.2003) (en banc), amended by 360 F.3d 1374 (2004). Numerous cases have addressed which agency actions warrant Chevron deference. For example, in the Ninth Circuit, an agency interpretation only receives Chevron deference when it is adopted in a document that has a precedential effect that binds third parties. MarmolejoCampos v. Holder, 558 F.3d 903, 909 (9th Cir.2009) (en banc).

A second form of deference to agency interpretation of statutes exists under, 323 U.S. 134 , 65 S.Ct. 161 , 89 L.Ed. 124 (1944). While Chevron deference takes its force from an assumed Congressional delegation of authority to the agency, Skidmore deference reflects the fact that the agency has experience with the statute and is likely to reach a reasoned interpretation regardless of whether Congress intended the agency’s interpretation to be binding. Mead, 533 U.S. at 230 , 121 S.Ct. 2164 . In contrast with Chevron deference, Skidmore deference is not all-or-nothing. “The weight [accorded to an administrative] judgment in a particular case will depend upon the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control.” Skidmore, 323 U.S. at 140 , 65 S.Ct. 161 . An agency need not adhere to any particular procedures in order to receive Skid-more deference.

Returning to agency interpretation of regulations, it appears that -with regard to underlying justifications, Auer deference is more akin to Chevron than to Skidmore . Tautologically, where an agency is interpreting its own regulation, the agency is speaking on a subject for which “Congress [has] delegated authority to the agency generally to make rules carrying the force of law.” Mead, 533 U.S. at 226-27 , 121 S.Ct. 2164 . Additionally, Auer and Chevron provide essentially the same de gree of deference: under each, where the agency has satisfied the procedural prerequisites for deference, agency interpretation of an ambiguous law will be upheld so long as the interpretation is reasonable. 8 But see Dep’t of Health and Human Servs. v. Chater, 163 F.3d 1129, 1135 (9th Cir.1998) (explaining, as an additional ground for deferring to an agency’s interpretation of its own regulation, that the agency’s “expertise makes it well-suited to interpret its own language.”).

Auer may share Chevron’s justification, but it differs with regard to procedural prerequisites. As noted, an agency interpretation will only receive Chevron deference if it was adopted through procedures with some formality. Mead, 533 U.S. at 229-30 , 121 S.Ct. 2164 ; Barnhart v. Walton, 535 U.S. 212, 222 , 122 S.Ct. 1265 , 152 *1152 L.Ed.2d 330 (2002). Agency interpretation of a regulation may receive Auer deference even when the interpretation was adopted through informal procedures that would not suffice under Chevron. Bassiri, 463 F.3d at 930 .

Nonetheless, it cannot be the case that courts must defer to any interpretation of a regulation uttered by any agency official. As the Seventh Circuit has recognized:

Joseph v. Holder, 579 F.3d 827, 832 (7th Cir.2009). Although “formal” procedures are not required in the Ninth Circuit, Joseph’s observation that some agency employees lack the power to speak for the agency cannot be denied.

This principle is further illustrated by Auer itself. Auer looked to the circumstances surrounding the particular agency document at issue before concluding that deference was warranted. Auer concerned a regulation promulgated under the Fair Labor Standards Act. The Secretary of Labor had filed an amicus brief before the Supreme Court offering the Secretary’s interpretation of the regulation. The Court held that this briefs interpretation of the regulation was entitled to deference because it was “in no sense a post hoc rationalization advanced by an agency seeking to defend past agency action against attack” and there was “no reason to suspect that the interpretation [did] not reflect the agency’s fair and considered judgment on the matter in question.” Auer, 519 U.S. at 462 , 117 S.Ct. 905 . More recently, the Court applied a similar analysis in deferring to an amicus brief filed by the Federal Reserve Board, resting on these factors and the fact that “[t]he board [was] not a party to [the] case.” Chase Bank USA, N.A. v. McCoy, — U.S. -, 131 S.Ct. 871, 880 , 178 L.Ed.2d 716 (2011). In Chase Bank, the Court specified that it deferred to the particular amicus brief “in light of ‘the circumstances of this case,’ ” and that the Court was not holding that all informal interpretations were entitled to Auer deference. Id. at 884, citing Auer, 519 U.S. at 462 , 117 S.Ct. 905 .

Thus, although the Ninth Circuit has stated that an agency’s use of informal procedures does not itself preclude Auer deference, the Supreme Court has clearly indicated that not every document or statement issued by an agency representative is entitled to Auer deference. Neither court has articulated more specific guidance as to when deference is warranted. The Ninth Circuit’s practice, however, is to extend Auer deference to general statements of policy offered outside the context of litigation against the agency. The Ninth Circuit has granted Auer deference to: agency opinion letters, Bassiri, 463 F.3d at 930 ; interpretations published in the federal register, Strom v. U.S., 641 F.3d 1051, 1063-64 (9th Cir.2011), Silvas v. E*Trade Mortg. Corp., 514 F.3d 1001, 1005 (9th Cir.2008) and S.E.C. v. Phan, 500 F.3d 895, 903-04 (9th Cir.2007); 9 amicus briefs filed before the circuit court, Dreiling v. *1153 Am. Exp. Co., 458 F.3d 942 , 953 n. 11 (9th Cir.2006) and Zurich Am. Ins. Co. v. Whittier Props. Inc., 356 F.3d 1132 , 1137 n. 27 (9th Cir.2004); 10 and an agency’s “internal memorandum,” L.A Closeout, Inc. v. Dept. of Homeland Sec., 513 F.3d 940, 941-12 (9th Cir.2008).

b. The Forest Service’s Litigation Position

Auer and Chase Bank held that the fact that the agencies were not parties to the litigation indicated that the agencies’ amicus briefs “reflected] the agencies’] fair and considered judgment.” Auer, 519 U.S. at 462 , 117 S.Ct. 905 , Chase Bank, 131 S.Ct. at 884 . Mirroring this reasoning, other courts have conversely declined to defer to interpretations offered in the course of litigation against the agency. Robinson Knife Mfg. Co., Inc. v. Comm’r of Internal Revenue, 600 F.3d 121 , 134 n. 11 (2nd Cir.2010) (quoting Auer, 519 U.S. at 462 , 117 S.Ct. 905 ). See also Pierre v. Comm’r of Internal Revenue, 2009 WL 2591625 , *12-13, 133 T.C. No. 2 , 2009 U.S. Tax Ct. LEXIS 21 , *34-*36 (2009) (Cohen, J., concurring). The circumstances of this case demonstrate that the Forest Service’s litigation position is not itself entitled to Auer deference.

c. The EIS and Related Documents

As noted above, each party purports to support its interpretation of the Travel Management Rule by citing statements made in the record. As the court explains, the record’s limited discussion of this rule does not support either party’s position.

The record discusses the relationship between Subparts A and B in only two passages, both of which are responses to plaintiffs’ sequencing argument. The first is the Forest Service’s response to a comment plaintiffs had submitted regarding the draft EIS. AR 2898 (final EIS C-14). The response stated that “[t]he analysis of effects from implementing each of the alternatives presented in Chapter 3 of the FEIS informs the Forest Supervisor in making a decision regarding the minimum system ... In making his determination, the Forest Supervisor will consider the direction provided in the ENF LRMP.” Id.

The second passage occurs in the response to plaintiffs’ administrative appeal. 11 Plaintiffs’ administrative appeal asserted their minimum road system argument. AR 822. The appeal deciding officer first stated that the EIS’s alternatives analysis “informs the Forest Supervisor in *1154 making a decision regarding the minimum road system,” repeating the Forest Service’s earlier response to comments. Id. The officer went on to state, however, that “The Forestwide Roads Analysis completed in 2003 also helped identify this minimum system____I find that the FEIS and ROD succeed in designating a minimum system as required by the Travel Rule ....” Id. (emphases added).

Plaintiffs argue that these two passages demonstrate that the Forest Service’s decision to adopt the Travel Management Decision was based on the belief that the Decision satisfied the Forest Service’s obligations under both Subparts A and B. All parties now agree, however, that the Travel Management Decision did not identify a minimum road system. Plaintiffs argue that this warrants reversal, because “an agency’s action must be upheld, if at all, on the basis articulated by the agency itself.” Motor Vehicle Mfrs. Ass’n, 463 U.S. at 50 , 103 S.Ct. 2856 . The EIS’s response to comments defies plaintiffs’ characterization, since the response refers to subpart A in the future tense. The resolution of the administrative appeal presents a more difficult question. The decision stated that a minimum road system had already been identified. The denial of the administrative appeal is the final agency action, 36 C.F.R. § 215.1 (b), and in light of Federal Defendants’ insistence on the importance of administrative exhaustion, there would be a “poetic justice” in holding the Forest Service to statements made in the administrative appeal. Int’l Union of Operating Eng’rs v. County of Plumas, 559 F.3d 1041, 1045 (9th Cir.2009). Nonetheless, it is clear that, as Federal Defendants argue, the appeal deciding officer simply wrote in error. Not a single statement in the Record of Decision or EIS indicates that a minimum road system had been or was being identified. Cf. Guzman, 766 F.Supp.2d. at 1078-79. In light of the facts that plaintiffs’ argument is supported by a sole statement, that the error of this statement should have been plain to all readers, and that this statement therefore did not meaningfully interfere with public comment, judicial review, or the other purposes of NEPA and the APA, the court declines to hold that this statement represents the “basis articulated by the agency” for purposes of Motor Vehicle Mfrs. Ass’n . The Forest Service has not conceded that Subpart A must precede Subpart B.

On the other hand, the court also rejects defendants’ contention that these two passages meaningfully interpreted the Travel Management Rule as permitting the Forest Service to begin with Subpart B. Although the FEIS’s response to comments implies the interpretation now advocated by the Forest Service, the final agency action, insofar as it interpreted the regulation at all, implies the opposite. The Forest Service cannot point to any articulation of its purported interpretation of the Travel Management Rule in the record. These fleeting statements on the issue do not bear the hallmarks of the agency’s “fair and considered judgment” and are not entitled to Auer deference. 12

Ultimately, the Forest Service appears to argue that by acting, the Forest Service implicitly interpreted all applicable regulations as permitting the agency’s choice of action. This is a post-hoc rationalization, not the type of considered exercise of delegated lawmaking authority that underlies Auer . At the very least, if the agency wishes for its interpretations to receive *1155 Auer deference, the agency must articulate those interpretations.

d. The Forest Service Manual

After the 2005 promulgation of the Travel Management Rule, the Forest Service updated the Forest Service Manual. The Forest Service Manual, a document published and updated through notice-and-comment proceedings and which guides the entire agency in applying the regulation, is plainly the type of document entitled to Auer deference.

The Forest Service Manual requires “responsible officials to use travel analysis to consider the criteria in 36 C.F.R. § 212.55 [Subpart B] and contribute towards identification of the minimum road system needed for safe and efficient travel and for administration, utilization, and protection of [national forest] lands ( 36 C.F.R. § 212.5 (b) [Subpart A].”). 72 Fed. Reg. 10,632 , 10,635 (March 9, 2007). “Travel analysis for purposes of identification of the minimum road system is separate from travel analysis for purposes of designation of roads, trails, and areas for motor vehicle use. Travel analysis for both purposes may be conducted concurrently or separately.” FSM 7712. ¶ 2. The same section of the manual states that “[a]ny proposals resulting from travel analysis for either purpose may be addressed in the same or different environmental analyses.” Id. ¶ 3 .

These sections of the manual do not explicitly address the timing of Subparts A and B. Nonetheless, the manual’s statement that the travel analysis underlying Subpart B may occur “separately” from the travel analysis for Subpart A suggests that Subpart B itself may precede Subpart A. Accordingly, the court concludes that for purposes of Auer , the manual interprets the Travel Management Rule as allowing the agency to begin with Subpart B. The court must defer to this interpretation if it is reasonable.

3. Reasonableness of the Forest Service Manual’s Interpretation of the Rule

The Travel Management Rule contains no explicit statement regarding the timing of Subparts A and B. Plaintiffs argue that the structure of the regulation is such that Subpart A is the “essential prerequisite” to Subpart B, and that prior completion of Subpart A is “logically necessary.” Plaintiffs argue that if B is done first it will be impossible to do A properly, and that A provides information necessary to B. Plaintiffs additionally discuss the ENF’s road maintenance backlog. None of these arguments compel plaintiffs’ interpretation of the rule.

First, an essential aspect of the Subpart A analysis is the identification of roads to be decommissioned. Plaintiffs contend that if the agency first designates roads under Subpart B, the agency will have prejudged whether those roads should be decommissioned under Subpart A. The court agrees that during the Subpart A analysis the Forest Service will need to evaluate all roads, including any roads previously designated as open under subpart B, for decommissioning. Plaintiffs have not shown that prior completion of Sub-part B gives rise to dangers of “[b]ureaucratic rationalization and bureaucratic momentum” so great that this re-examination will be meaningless. Northern Cheyenne Tribe v. Hodel, 851 F.2d 1152, 1157 (9th Cir.1988).

Second, plaintiffs argue that Subpart A provides information essential to Subpart B. Although the chain of reasoning underlying plaintiffs’ “logical necessity” argument is unclear, plaintiffs appear to argue that Subpart B obliges the Forest Service to minimize the adverse effects of roads, that the effects of roads are correlated with the amount of roads, and that Subpart A identifies the minimum amount of roads *1156 necessary to satisfy the National Forest’s purposes. Of course, the Forest Service may consider minimizing the amount of roads during the Subpart B analysis as well. Plaintiffs apparently contend that Subpart A facilitates this analysis by revealing potential multiple-use roads. Whereas Subpart B looks only to recreation, Subpart A considers the needs of recreation as well as forest administration, fire protection, and other national forest purposes. 36 C.F.R. § 212.5 (b)(1); 66 Fed. Reg. at 3208. Plaintiffs argue that considering these purposes all at once will enable the Forest Service to evaluate a comprehensive system and thereby more fully minimize road impacts. 13 At most, this argument suggests that beginning with Subpart A may be prudent. This is too slender a reed to support plaintiffs’ assertion that the structure of the Travel Management Rule compels completion of Subpart A before Subpart B.

In an apparent extension of the preceding argument, plaintiffs emphasize the ENF’s road maintenance backlog. The record demonstrates a maintenance backlog in the ENF. Subpart A presumes that this backlog will be addressed in part by decommissioning roads. Plaintiffs argue that the Subpart B analysis must be informed by knowledge of which roads will be removed. Again, this argument merely provides a reason why the agency might be better off completing Subpart A first.

Finally, in an argument not obviously connected to Subpart A, plaintiffs argue that designating roads that the Forest Service cannot afford is itself arbitrary and capricious. When roads do not receive the required level of maintenance their environmental impact increases. See, e.g., AR 2522-26 (EIS 3-25 to 3-29). Preservation of the environment, however, is only one of many goals the Forest Service must consider in managing the national forest, and on this issue, plaintiffs have not demonstrated that the Forest Service’s balancing of these goals was arbitrary. Plaintiffs offer an illustrative car analogy. Plaintiffs contend that road maintenance is akin to getting an oil change every 5,000 miles, and that deferring an oil change until 12,000 miles would be arbitrary and capricious. Plaintiffs contend that the Forest Service must either change the oil more often or stop driving. One can easily imagine, however, circumstances in which driving 12,000 miles before changing oil is the best available option. A driver who cannot afford more frequent maintenance, but who needs to drive in order to work, might defer the oil change despite the knowledge that it is not the best of maintenance. Similarly, the Forest Service recognized the importance of maintenance and the chronic problems with funding maintenance, but the Forest Service concluded that in light of its budget and the multiple mandates the agency faced, designating roads in excess of the maintenance budget was the best option. 14

In summary, although plaintiffs’ arguments have identified various reasons why it might be preferable to conclude Subpart A before Subpart B, plaintiffs have not shown that the language of the regulation compels this ordering. On arbitrary and capricious review, the court will not reverse the agency merely based on the *1157 court’s perceptions of prudence. The court therefore grants summary judgment to the defendants on plaintiffs’ Travel Management Rule claim.

B. Plaintiffs’ NEPA Alternatives Analysis Claim

Plaintiffs argue that the Forest Service should have fully examined an alternative that would not have opened any previously unauthorized routes to public motor vehicle travel.

As noted above, examination of alternatives “is the heart of the environmental impact statement.” 40 C.F.R. § 1502.14 . The agency must “[rjigorously explore and objectively evaluate all reasonable alternatives [to the proposed action], and for alternatives which were eliminated from detailed study, briefly discuss the reasons for their having been eliminated.” 40 C.F.R. § 1502.14 (a). The “touchstone” is “whether [the] selection and discussion of alternatives fosters informed decision-making and informed public participation.” Westlands Water Dist. v. U.S. Dep’t of Interior, 876 F.Sd 858, 872 (9th Cir.2004) (quoting California v. Block, 690 F.2d 753, 767 (9th Cir.1982)). An EIS “cannot be found wanting simply because the agency failed to include every alternative device thought conceivable by the mind of man.” Vt. Yankee Nuclear Power Corp. v. Natural Res. Def. Council, 435 U.S. 519, 551 , 98 S.Ct. 1197 , 55 L.Ed.2d 460 (1978). Still, “[t]he existence of a viable but unexamined alternative renders an environmental impact statement inadequate.” Westlands Water Dist., 376 F.3d at 868 (quoting Morongo Band of Mission Indians v. Fed. Aviation Admin., 161 F.3d 569, 575 (9th Cir.1998)).

The scope of “viable” or “reasonable” alternatives is determined by the “the purpose and need statement articulated by that agency.” ‘Ilio‘ulaokalani Coal. v. Rumsfeld, 464 F.3d 1083 , 1097 (9th Cir.2006). This statement inevitably narrows the scope of alternatives, but the agency cannot “define its objectives in unreasonably narrow terms.” Muckleshoot Indian Tribe v. U.S. Forest Service, 177 F.3d 800, 813 (9th Cir.1999) (quoting City of Carmel-by-The-Sea v. U.S. Dept. of Transp., 123 F.3d 1142 , 1155 (9th Cir.1997)). Here, the statement of purpose largely recounts the goals and obligations imposed on the agency by statute, and is therefore reasonable. AR 2436-37. These competing goals include conservation and various types of recreation, as explained in part 1(C) supra.

In balancing competing uses the Forest Service faces a spectrum of choices, with one end representing greater motorized vehicle access and the other representing greater conservation and non-motorized recreation. The Ninth Circuit has twice invalidated agency decisions where the EIS “did not consider alternatives that sufficiently explored the ‘trade-off between wilderness use and development’ ” in similar land management dilemmas. Or. Natural Desert Ass’n, 625 F.3d at 1123 (quoting Block, 690 F.2d at 767 ). In these cases a broad range of options were left unexplored. Oregon Natural Desert Association concerned, in pertinent part, off-highway vehicle use on lands managed by BLM. BLM “did not consider any alternative that would have closed more than 0.77% of the planning area to [OHVs].” 625 F.3d at 1124 . The Ninth Circuit rejected the EIS, holding that BLM “must consider closures of significant portions of the land it manages.” Id. Block concerned allocation of roadless National Forest lands into three management categories. 690 F.2d at 758 . The Forest Service did not “seriously consider[ ]” allocating more than a third of the acreage as wilderness; conversely, all alternatives contemplated significant development. Id. at 765, 767 . This impermissibly “uncritically assume[d] that a substantial portion of the ... areas should be *1158 developed and considered] only those alternatives with that end result.” Id. at 767 . The Northern District of California recently followed these cases in invalidating an EIS in which every alternative allowed some vehicle use on every mile of the existing 5,098 mile route network. CBD v. BLM, 746 F.Supp.2d at 1087-88 .

This case is distinguished from Block and Oregon Natural Desert Association by the range of options explored together with the agency’s explanation as to why a “no unauthorized routes” alternative would be unreasonable. The Forest Service considered five action alternatives, which would designate between 5% and 9% of the existing unauthorized road network as open to vehicle use. AR 2408, 2410. Thus, the Forest Service considered closing “significant portions” of the unauthorized road network, Natural Desert Ass’n. The 5% of the unauthorized road network that all alternatives leave open is less than the “substantial portion” of land the Forest Service assumed should be developed in Block . 15

Furthermore, the Forest Service explained why these routes were necessary. The Forest Service held that the project’s multiple-use purpose would not be met by a “no previously unauthorized routes” alternative. AR 2469. As the EIS explains,

AR at 2397. Elsewhere, the EIS explains the benefits afforded by some of the specific previously-unauthorized routes that would be designated under Modified B. The added routes “will enhance motorized recreation in the Gold Note area” and will “provide[ ] a number of dispersed camping opportunities in the Capps Crossing area.” AR 2794. As a whole, this explanation demonstrates that assumption. 16 C.f. *1159 Block, 690 F.2d at 767 ; accord Guzman, 766 F.Supp.2d at 1070-71 (although no alternative considered closing all routes in recommended wilderness areas, the Forest Service had adequately explained that such a closure would fragment transportation network and therefore be unreasonable).

Accordingly, the Forest Sendee properly explained why a “no unauthorized routes” alternative would be unreasonable, and the exclusion of such an alternative from analysis did not violate NEPA. The court therefore grants summary judgment to the defendants on plaintiffs’ NEPA Alternatives Analysis Claim.

C. Plaintiffs’ NEPA Site-Specific Data Claim

Plaintiffs’ third claim is that the EIS failed to include adequate site-specific data. Plaintiffs argue that the data the EIS used was inappropriate and that the EIS lacked site-specific data regarding a number of particular issues. The court rejects both arguments, and accordingly grants summary judgment to the defendants on plaintiffs NEPA Site-Specific Data Claim.

1. Scope of NEPA’s ‘Hard Look’ Requirement

NEPA “requir[es] that agencies take a ‘hard look’ at how the choices before them affect the environment.” Or. Natural Desert Ass’n, 625 F.3d at 1100 . This ‘hard look’ requires a “reasonably thorough discussion of the significant aspects of the probable environmental consequences. This standard is not susceptible to refined calibration.... [A] reviewing court [must] make a pragmatic judgment whether the EIS’s form, content and preparation foster both informed decision-making and informed public participation.” Block, 690 F.2d at 761 (internal quotation omitted). “The detail that NEPA requires in an EIS depends upon the nature and scope of the proposed action.” ‘Ilio'ulaokalani Coal., 464 F.3d at 1095 (quoting Block, 690 F.2d 753 ). The agency’s determination as to whether particular evidence is, or is not, necessary to support the EIS’s conclusions will be upheld unless this determination is itself arbitrary and capricious. McNair, 537 F.3d at 992 . In this case, all parties agree that NEPA required site-specific analysis, ie., analysis of the impacts of the particular roads under consideration, informed by data regarding those specific roads.

2. Types of Site-Specific Data Used by the EIS

The EIS relied on three types of site specific data: GIS data, route evaluation forms, and additional field assessments. AR 2498-99.

The “primary” data source was GIS data. AR 2498. Plaintiffs argue that the GIS data was of too gross a scale to support meaningful analysis. This argument is based on a misinterpretation of one passage in the EIS, which discussed certain geological phenomena at a 1:24,000 scale. AR 2517. Plaintiffs mistakenly interpret this statement as indicating that all GIS data used in every aspect of the EIS was at this coarse scale. The record demonstrates that the EIS uses GIS data at a range of scales, including data with spatial resolutions in the tens of feet. See AR 2658-59 (EIS 3-161 to 3-162), 12,960-62 (Terrestrial Management Indicator Species Report 7-9), 13,138 (Biological Evaluation for Botanical Resources 20). More generally, the GIS data was itself derived “from past field surveys and inventories.” AR 2498 (EIS 3-1). This data was site-specific, and the agency’s decision to rely on it was proper.

*1160 The second data source consisted of “route evaluation[] forms” completed by Forest Service employees. Plaintiffs assert that these forms were not completed for roads presently designated as ML-2, but the page plaintiffs cite explicitly states that these forms were completed for “all ML-1 and ML-2 roads, as well as some unauthorized routes being considered for designation in the action alternatives.” AR 2498-99 (emphasis added). These completed forms are provided in the record, further refuting plaintiffs’ assertion. AR 13370-15834. 17

The third data source, which plaintiffs do not acknowledge, consisted of “field assessments and photo documentation” collected by Forest Service employees. AR 2499. These assessments were performed for all unauthorized routes proposed for designation as ML-2 roads or NFS trails in the action alternatives, as well as for specific other routes of concern. Id.

Thus, all ML-1 and ML-2 roads received a route evaluation form, all unauthorized routes proposed for designation received a field assessment, and some routes received both. Plaintiffs’ challenges to the adequacy of this type of data are meritless. 18

3. Data that Plaintiffs Contend Was Missing

Plaintiffs separately argue that the EIS lacked site-specific data regarding various specific road issues and impacts. For most of these, plaintiffs argue that the EIS itself concluded that this information were “necessary to support” the EIS’s analysis, McNair, 537 F.3d at 992 , but that the information was lacking. As the court explains, plaintiffs have generally misinterpreted the portions of the EIS they rely upon. In those cases where data was absent, this absence did not prevent the EIS from sufficiently “foster[ing] both informed decision-making and informed public participation.” Block, 690 F.2d at 761 . 19 The court therefore rejects this argument.

a. Geological Hazards

In surveying the geology of the ENF, the EIS discusses three types of geological hazards resulting from “soil failures:” landslides, debris flows, and soil creep. AR 2515. Landslides are large soil movements, typically involving tens to hundreds of acres moving with a “failure zone”, that is greater than ten feet deep. Id. Moreover, the individual landslides are generally nested together in a larger complex. Id. The EIS does not define debris flows except to state that they are generally naturally occurring as the result of rainfall, although “a few roads and trails have the potential for debris flows initiating from fillslope areas,” usually caused by “[flailed or plugged culverts.” Id. Similarly, the EIS does not define soil creep.

*1161 The EIS then addresses the alternatives’ potential impacts on geology. The EIS’s discussion is poorly worded, and this opacity gives rise to plaintiffs’ primary site-specific data argument. The EIS states that:

AR 2517. Plaintiffs interpret the first sentence of this passage to mean that the Forest Service chose to evaluate impacts at the 1:24000 scale and that this evaluation did not reveal any impacts. From this interpretation, plaintiffs launch the obvious retort: that the Forest Service should have used a finer scale evaluation.

The Forest Service argues that the 1:24000 scale represents the scale at which geological hazards occur, rather than simply the scale at which the Forest Service analyzed the geological impacts. The remainder of the quoted passage supports the Forest Service’s interpretation. The EIS’s explanation as to why roads will not meaningfully influence landslides and debris flows is not dependent on the scale of analysis. Indeed, the fact that the EIS is able to quantify routes’ effects on shallow landslides suggests that if there were effects on other geological hazards, the analysis would have been able to detect them. Thus, the EIS concludes that geological hazards occur at a large scale, that roads are largely incapable of meaningfully influencing these large-scale hazards, and that the roads’ effects are therefore “not measurable.” Accordingly, although the first sentence of this passage represents another poor choice of wording by the Forest Service, this passage does not concede the inadequacy of the Forest Service’s data, and the court does not conclude that this passage is so opaque as to preclude informed public participation or agency decisionmaking.

b. Hillslope Instability

Plaintiffs next argue that the EIS failed to include site-specific data as to which routes traverse unstable hill slopes. As explained above, the EIS concludes that routes would have only limited effects on hill stability or other geologic features. Looking in the other direction, the EIS concludes that geological features, especially hill stability, could potentially impact roads, impacting road maintenance and potentially presenting risks to road users. AR 2516. The EIS explains that there is a greater risk of hill slope instability when the slope gradient exceeds 57% and a spring is present. Id. The EIS includes a table listing, for alternatives A through E, the percentage of routes within each alternative that occur in areas where these two risk factors are present. AR 2518. This is precisely the data that plaintiffs contend was absent. This table does not, however, include an analysis of alternative Modified B, and the Federal Defendants have not explained this omission. Nonetheless, in light of the table’s conclusion that alternatives B through E present highly similar amounts of routes on unstable slopes, the similarity between Modified B and the oth *1162 er action alternatives, and the fact that the remainder of the EIS provides separate analyses for Modified B, the court concludes that this omission did not violate NEPA. 20

c. Soil Erosion and Steep Routes

Although the EIS concludes that routes would not affect soil on the geological scale, the EIS acknowledges that routes can cause smaller-scale soil erosion. The EIS explains that existing routes (whether official or unauthorized) can cause soil loss by concentrating water which then creates a gully in the adjacent soil where the water flows off the route. 21 AR 2523. The EIS analyzes each alternative’s potential impact on soil erosion by examining “soils susceptible to gully erosion, total miles of routes open by alternative, ... condition of native surfaced roads based on field assessments,” and the degree to which each alternative would impose seasonal route closures. Id.

In evaluating the extent to which soils were susceptible to erosion, the EIS looks to both the soil and to whether the route was on a hill slope with a gradient of over 30%. AR 2524. The EIS acknowledges that the gradient of the route itself, as opposed to the hill gradient, also contributes to routes’ erosion potentials. AR 2523. The Forest Service attempted to use the GIS to identify which routes included the type of “sustained, steep [road] gradients” that indicated a “risk of erosion,” but the GIS was unable to identify these routes. AR 2523.

Plaintiffs argue that if the GIS was unable to identify routes with sustained steep gradients then the Forest Service was required to identify these routes through other means. The court rejects this argument, because plaintiffs have not demonstrated that the other information was insufficient to provide a “reasonably thorough” discussion of the routes’ potential for soil erosion. Block, 690 F.2d at 761 . The EIS explains why the information regarding route gradients was not available. Neighbors of Cuddy Mt. v. U.S. Forest Serv., 137 F.3d 1372 , 1380 (9th Cir.1998). Athough the absence of this information might have precluded the EIS from determining “the precise extent” of the routes’ effects on erosion, the EIS “estimate[s] what those effects might be.” Conner v. Burford, 848 F.2d 1441, 1450 (9th Cir.1988). The Forest Service concluded that the available information was sufficient to inform this estimate, and plaintiffs have not shown that this conclusion was arbitrary or capricious. McNair, 537 F.3d at 992 .

d. Drainage on Trails

Plaintiffs next argue that the EIS concedes that there was inadequate information regarding the drainage characteristics of OHV trails. Plaintiffs have misinterpreted the EIS. The EIS explains that OHV trails that were constructed specifically for OHV use typically have good drainage characteristics. AR 2522. The EIS goes on to state that “[m]any OHV trails, however, were not originally designed and constructed for OHV use, but were converted from roads. Road prisms *1163 are well-compacted and provide a firm running surface, but the compacted surface makes installing OHV rolling dips difficult. The long sustained gradients of trails converted from roads demand more attention to drainage. ” Id. (emphasis added). Plaintiffs argue that the phrase “demand more attention” is an admission that the EIS required additional data and study on this issue. Federal defendants contend that “attention” refers to maintenance rather than data. The context demonstrates that this is a reasonable, and perhaps more natural, interpretation. This phrase therefore does not acknowledge a data deficiency.

e. Condition of Drainage Structures

The EIS states that “ ‘Maintaining drainage structures is the key to minimizing erosion on system roads. Drainage structures are particularly susceptible to damage during the wet season.’ ” AR 2522. Plaintiffs argue that the EIS therefore “should have noted which routes are suffering damage to drainage structures during the wet season.” Pis.’ Br. 19 (Dkt. 52-1). Although the EIS does not contain site-specific analysis as to which particular routes were presently suffering wet season damage to drainage structures, it does not ignore the issue. The EIS instead uses the extent to which the alternatives would close routes during the wet season as a proxy for the extent to which the alternatives would prevent this damage. AR 2527-28. The EIS further incorporates site-specific information regarding the condition of roads, which included the roads’ drainage. AR 2525. These factors provide a “reasonably complete” analysis of the impact of wet season damage to drainage structures on erosion.

f. Causes of Poor Road Conditions

As noted above, one type of information the EIS uses to assess routes’ impacts on soil erosion is field assessments and surveys. One way in which the EIS uses the field assessments is to look to whether roads are in poor condition. AR 2525. “Road condition was recorded as the percent of each route that was rutted, washed out, eroded, or slumped; had poor drainage; or was too steep.” Id. The EIS uses road condition as an indicator of the extent to which the alternatives “avoid problem areas.” AR 2526. The EIS reasons that if a road is in poor condition, this may be because the road is in a problem area, requiring more maintenance and presenting a greater potential for erosion. Id. The EIS acknowledges that this is an imperfect heuristic, however, because it could be that a route is in poor condition simply because it has not received ordinary maintenance. Id.

Plaintiffs argue that the Forest Service should have improved the quality of this heuristic by recording the causes of poor conditions on individual roads, rather than merely assuming a general correlation between poor conditions and problem areas. The EIS was “reasonably complete” without this information.

g.Type and Conditions of Stream Crossings

The draft aquatic species Biological Evaluation noted that “[t]he type and condition of stream crossings has not been spatially identified. Therefore, site-specific analysis of these characteristics is generally lacking. As a result, site-specific and localized adverse effects may be understated.” AR 12,656. Federal defendants argue that in light of the extensive other information regarding aquatic habitat, this information was not necessary to inform the agency or the public of the action’s consequences. Notably, the Aquatic Management Indicator Species Report specifies the total number of motorized route stream crossings under each alternative, AR 12,913. The Aquatic Species Biological Evaluation further breaks this figure down by the sizes of the *1164 streams crossed. AR 12,778. This document also identified particular routes which cross streams. AR 12,848-53 (tables A-8 and A-9).

Plaintiffs argue that, as the draft Biological Evaluation acknowledged, identifying the specific location of crossing might have increased the precision of the aquatic habitat analysis. The mere fact that a more thorough analysis was possible does not demonstrate that the analysis actually performed was inadequate. Plaintiffs have provided no other argument as to why it was necessary to know the location of stream crossings. Accordingly, the court does not hold that this information was required.

D. Plaintiffs’ National Forest Management Act Claim

Individual projects within the ENF, including the Travel Management Decision, must be consistent with the ENF Forest Plan and the Sierra Nevada Forest Plan. Plaintiffs’ NFMA claim argues that provisions of both the ENF and Sierra Nevada Forest Plans prohibit vehicle use in meadows, and that the Travel Management Decision designates routes in violation of these prohibitions.

Because this argument raises complicated issues, the court summarizes its conclusions here. The Forest Service addressed the two Forest Plans separately, adopting different strategies to reconcile each with the Travel Management Decision. 22 As to the ENF Forest Plan, the Travel Management Decision incorporates amendments to the plan, specifically exempting an enumerated list of routes from three ENF Forest Plan provisions regarding meadows. NFMA appears to permit this approach. 23 In this case, however, the Forest Service’s implementation of this strategy was incomplete. The record indicates that the Travel Management Decision designates 42 routes through meadows, but the incorporated amendments only refer to 20 routes. In discussing the Sierra Nevada Forest Plan, the Forest Service did not rely on the amendments to the ENF Forest Plan, instead concluding that the Travel Management Decision was consistent with the Sierra Nevada Forest Plan’s provisions as they stood. For the reasons stated below, the Forest Service failed to provide a rational explanation for this conclusion.

The Forest Service presumably could have adopted an amendment that encompassed all 42 routes and exempted these routes from the provisions of both ENF and Sierra Nevada Forest Plans. It appears that such an amendment would have reconciled the conflicts between the Travel Management Decision and the forest plans. 24 This is not what the Forest Service did. The court cannot uphold agency *1165 action on the ground that the agency might have been able to support its action had the agency acted differently. Accordingly, the Travel Management Decision conflicts with the governing Forest Plans, and thereby violates NFMA.

I. The Quantity of Routes through Meadows

The Travel Management Decision designated routes through meadows. While the record is consistent on this point, it provides conflicting statements regarding the quantity of such routes. The Aquatic Species Biological Evaluation indicates that alternative Modified B includes 42 routes through meadows. AR 12,845 (BE A-22, Table A-7). The discussion of the ENF Forest Plan identifies 20 routes through meadows totaling 4.8 miles. AR 3274 (ROD at 5). In responding to comments, the EIS states that “In Modified B, the preferred alternative, 21 routes with a total of 4.8 miles through meadows are proposed to allow public wheeled motor vehicle use.” AR 2980 (EIS C-96). The Forest Service’s briefing has not explained the discrepancy between these three figures. Instead, where plaintiffs have cited these divergent numbers, Federal Defendants have criticized plaintiffs for the inconsistency. Fed. Defs.’ Reply Br. 35 n. 20 (Dkt. 66). It may be that the 21 routes identified in the response to comments are meant to be the same 20 routes identified in the discussion of the ENF Forest Plan. Another explanation would be that the comment identified the remaining 22 routes, although it appears unlikely that the two separate groups of routes would each amount to exactly 4.8 miles. Absent explanation to the contrary, the court assumes that all 42 routes identified in Table 7 pass “through” meadows in the pertinent sense, and that the other figures merely refer to subsets of these routes. 25

2. The ENF Forest Plan’s Meadow Provisions

The EIS discussed the ENF and Sierra Nevada Forest Plans separately. Because the court looks to whether the reasoning offered by the agency supports the conclusion reached, the court similarly separates the two plans here.

The ENF Forest Plan includes three directives relating to meadow protection. The Forest Service must “consider closing and obliterating existing roads” throughout the ENF in order to protect meadows, AR 4995 (ENF LRMP 4-90), “[prohibit motor vehicle use on meadows,” AR 5183 (ENF LRMP 4-278), and “[c]lose roads to and across meadows,” AR 5187 (ENF LRMP 4-282). 26

*1166 The EIS acknowledges the conflict between these three prohibitions and the Travel Management Decision’s designation of routes through meadows. AR 2457(EIS), 3274(ROD). The Forest Service purported to address this conflict by amending the ENF Forest Plan. Id. For example, the second prohibition was amended to read “[prohibit motor vehicle use on meadows, except for the route segments identified in [an attached] table.” Id. This table enumerates 20 route segments through meadows, totaling 4.8 miles. Id. The EIS explains that the amendment is warranted because “these specific routes provide a unique recreation opportunity (such as a high elevation trail experience), enhance the recreation experience by connecting routes or areas, provide access to an area of interest, or allow access to dispersed camping.” AR 2455 (EIS 2-9).

By incorporating this amendment into the Travel Management Decision, the Forest Service validly determined that the 20 enumerated routes would not violate the ENF Forest Plan. The problem for the Forest Service is that the Travel Management Decision designates an additional 22 route segments through meadows. The record does not appear to discuss the relationship between these additional 22 routes and the prohibitions found in the ENF Forest Plan. At oral argument the court directly asked counsel for the Forest Service about this disparity, and counsel was unable to provide an explanation. Therefore, notwithstanding the amendment to the ENF Forest Plan, the plan apparently prohibits 22 of the routes designated by the Travel Management Decision. Alternatively, the record fails to provide a rational connection between the fact that the decision designates 42 routes through meadows and the conclusion that those routes do not violate the ENF Forest Plan.

3. The Sierra Nevada Forest Plan’s Meadow Provisions

The Sierra Nevada Forest Plan adds a parallel and more complex framework for protecting meadows and other aquatic resources. The Sierra Nevada Forest Plan imposes various “Riparian Conservation Objectives.” See AR 10,970 (SNFPA ROD-31 (Appendix A: Management Direction)). In preparing for the Travel Management Decision, the Forest Service concluded that these objectives required that “routes ... not bisect or go through meadows.” AR 12,828-29 (BE at A-4 to A-5). The Forest Service has failed to reconcile its own conclusion with the fact that Modified B designates 42 routes through meadows. Accordingly, the Forest Service’s conclusion that the Travel Management Decision complied with the Sierra Nevada Forest Plan was arbitrary and capricious.

The BE interprets the Sierra Nevada Forest Plan’s riparian conservation objectives as prohibiting routes through meadows. The riparian conservation objectives themselves, together with the concurrently issued standards and guidelines, are complex. 27 The BE adopts simplified criteria *1167 to be used in evaluating compliance with these objectives. The compliance criterion for objectives # 2 and 5 is simply “routes do not bisect or go through meadows.” AR 12,829.

The court rejects the Forest Service’s present attempts to retreat from this criterion. The Forest Service first argues that it is the EIS, rather than the BE, that presents the agency’s final word, and that the EIS does not use this criterion. This argument fails because the EIS does not independently analyze the objectives, instead referring to the BE. The EIS notes that alternative Modified B designates routes through meadows, albeit fewer of such routes than any other alternative save Alternative E, and then asserts in a footnote that “[tjhese routes were not considered inconsistent with the Riparian Conservation Objectives under the criteria established for the analysis.” AR 2673, 2675 (EIS 3-176, 3-178 n. 10). The EIS then refers the reader to the BE and provides no further discussion on this issue. AR 2675. The Forest Service alternatively argues that it is the objectives themselves, rather than the compliance criteria, that are binding. This is true, and the Forest Service presumably could have adopted different compliance criteria. 28 This argument is nonetheless unavailing, because “an agency’s action must be upheld, if at all, on the basis articulated by the agency itself,” and the only analysis the Forest Service offered for riparian conservation objectives 2 and 5 rested on the “routes do not bisect or go through meadows” criterion. Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43 , 103 S.Ct. 2856 .

Having concluded that, in this case, the Forest Service is bound by this criterion, the court turns to the agency’s application of it. 29 The BE concludes that alternative Modified B is “[l]ikely to meet” riparian conservation objectives 2 and 5. AR 12830 (BE at A-6). The BE offers two explanations for this conclusion. First, the BE excludes from analysis existing ML-2 roads that retain their ML-2 designation under alternative Modified B. AR 12,826. Of the forty-two routes Modified B designates through meadows, twenty-seven meet this description. AR 12,845 (BE A- *1168 22, Table A-7). Second, the BE observes that under alternative Modified B, no roads through meadows that were previously designated as ML-1 are designated for public use. AR 12,830-31. Because this fails to discuss the 15 trails and non-excluded ML-2 roads through meadows, the Forest Service’s conclusion that the decision complied with the riparian conservation objectives was arbitrary and capricious.

Two further issues warrant discussion. First, the discussion of the conservation objectives, and of the Sierra Nevada Forest Plan in general, did not rely on the amendment to the ENF Forest Plan. The portions of the record discussing the riparian conservation objectives are separate from those portions discussing the ENF Forest Plan. The BE juxtaposes alternative Modified B (the chosen alternative) against alternatives B, C, and D by stating that the latter three would violate the riparian conservation objectives and thereby require an amendment to the forest plan, indicating that the Forest Service concluded that Modified B satisfied the riparian conservation objectives without recourse to an amendment. AR 12789 (BE at 39). The amendments specifically pertain to the three provisions of the ENF Forest Plan, rather than exempting the 20 enumerated routes from the forest plan requirements in general. In this litigation, the Forest Service has not argued that the amendment to the ENF Forest Plan exempts any routes from compliance with the riparian conservation objectives. 30

Second, the decision to exclude twenty-seven ML-2 routes from analysis was itself arbitrary and capricious. The BE excludes ML-2 roads that would remain designated as ML-2 roads. AR 12,826. The BE concludes that this is proper “because [these] roads will not be subjected to new management activities or a new type of use.” AR 12,826 (BE at A-2). The BE further explains that this determination was “based on [Sierra Nevada Forest Plan] Standard and Guideline 92[,] which states ‘Evaluate new proposed management activities within ... [riparian conservation areas] during environmental analysis to determine consistency with the riparian conservation objectives at the project level ...’” Id. 31 exclusion conflicts with the Forest Service’s court-imposed obligation to reconsider vehicle travel. The Travel Management Decision does not represent a change from a permissible status quo. Rather, this court previously held that the Forest Service had operated without a valid plan for management of vehicle travel since 1990. Berry, Order filed February 15, 2005 (pages 52-54). The court remanded to the Forest Service with instructions to consider vehicle travel anew. Berry, Order filed August 16, 2005. The Forest Service properly interpreted these orders in stating that the “purpose and need” for the Travel Management Decision was to “reconsider whether motorized use should be allowed to continue on [national forest] roads maintained for high clearance vehicles (NFS ML-2) and [national forest] trails managed for [off-highway vehicle] use.” AR 2436 (EIS at 1-5) (emphasis added). 32

*1169 For all of these reasons, the Court concludes that the Forest Service’s conclusion that the Travel Management Decision complied with the ENF and Sierra Nevada Forest Plans was arbitrary and capricious, in violation of NFMA.

The question of remedy is another matter, however. It appears that out of a forest encompassing 789,994 acres, that is, more than 1,234 square miles, the offending roads apparently span about 10 miles of meadows. On the other hand, the total span of meadows in the forest is about 10,416 acres, or more than 16 square miles. The requested remedy would set aside years of decision-making by the Forest Service, and an administrative record spanning thousands of pages, because the Forest Service failed to properly consider these 10 miles of roads. Accordingly, the court directs the parties to brief what remedy is appropriate in this situation. The parties should give particular attention to whether the Ninth Circuit’s caution that the courts not “flyspeck” the Environmental Impact Statement, see Nevada v. DOE, 457 F.3d at 93 , has application in the context of determining a remedy based upon such an apparently insignificant impact.

E. Plaintiffs’ Endangered Species Act Claim

Plaintiffs argue that the Forest Service violated the Endangered Species Act by failing to adequately consult with FWS regarding the designation of previously unauthorized routes. 33 The Forest Service contends that alternative Modified B, which was adopted by the ROD, complied with the design criteria established during the 2006 programmatic consultation, such that no further consultation was necessary. Plaintiffs argue that the Forest Service’s decision not to separately consult was arbitrary and capricious for four reasons: (1) in the record, the Forest Service itself stated that alternative Modified B violated the design criteria, (2) because the Forest Service did not evaluate habitat using “protocol surveys,” the Forest Service’s determinations of habitat unsuitability are inadequate, (3) the Forest Service misapplied the criteria by excluding areas with ephemeral streams from its analysis, and (4) the Forest Service was aware of additional information, primarily contained in the recovery plan for the frogs, that called into question whether compliance with the design criteria alone was sufficient to protect the frogs.

*1170 1. Whether the Forest Service Admitted that the Decision Violates the Design Criteria, and the “Protocol Survey” Argument

Plaintiffs’ first two ESA arguments are best addressed jointly. The Forest Service did not specifically consult with the FWS Service regarding alternative Modified B, relying instead on the 2006 programmatic consultation. Plaintiffs’ first ESA theory argues that the Forest Service’s Biological Assessment (“BA”) concedes that alternative Modified B violated the programmatic consultation’s design criteria, and that the Forest Service was therefore required to seek a separate consultation. Plaintiffs’ second ESA theory is that in evaluating whether habitat was suitable for frogs, the Forest Service was required to use surveys conducted according to a FWS protocol, but that the Forest Service instead relied on “field reconnaissance” and other non-protocol data. The court rejects both arguments on the merits. 34

As noted above, the Forest Service must consult with FWS regarding any project that “may affect” listed species. Prior to the Travel Management Decision, the Forest Service engaged in programmatic consultation with FWS. FWS agreed that future designation of unauthorized routes would not be likely to adversely affect listed species if the routes conformed to various design criteria. AR 1, 3-4. Plaintiffs’ ESA claims primarily rest on the second and fifth of these criteria. Design criterion #2 provides: “In suitable California red-legged frog habitat, routes avoid Riparian Reserve and Riparian Conservation Areas except where necessary to cross streams.” AR 12,891. 35 Criterion # 5 is functionally equivalent, but pertains to “areas” rather than “routes.” Id. Thus, a route or area violates these design criteria when it is both within suitable habitat and within a Riparian Reserve or Conservation Area.

The Forest Service’s application of the design criteria is illustrated by the differences between the draft and final BAs. Both assessments use a two step process for determining whether routes within riparian conservation areas were also within suitable frog habitat. In the first step, the Forest Service uses the GIS database to identify areas that were potentially suitable. The draft BA, which evaluated alternative D, identifies 2.5 miles of routes in potentially suitable habitat. AR 12,881. In the second step, the Forest Service uses “field reconnaissance” to evaluate whether these areas were in fact suitable habitat. The draft BA concludes, on the basis of this reconnaissance, that only 1.4 miles of routes designated by alternative D were in actually suitable habitat. Id.

Because these 1.4 miles were in suitable frog habitat, they did “not meet the programmatic Project Design Criteria.” AR 12,884. The Forest Service concluded that *1171 these routes were nonetheless not likely to adversely affect the frogs, AR 12885, because “field surveys” revealed that no frogs were actually present in areas adjacent to those 1.4 miles. AR 12,881. Because the “not likely to adversely affect” conclusion rested on reasoning other than that approved by the 2006 programmatic consultation, the Forest Service was required to consult with FWS.

The Forest Service did so, and FWS concurred that alternative D was “not likely to adversely affect” the frogs. FWS explained that:

AR 28. In this analysis, FWS concurred with the Forest Service’s determination as to which areas were and were not suitable. For those areas that were suitable, FWS relied on past “survey” data to determine that no frogs were actually present. Thus, both the draft BA and the 2007 Concurrence demonstrate that whether habitat is suitable and whether frogs are actually present are distinct questions. The second question only needs to be asked once the habitat is determined to be suitable, and FWS only used surveys in addressing this second question. 36

After the 2007 concurrence, the Forest Service prepared a final BA regarding alternative Modified B. Modified B designates six unauthorized routes, totaling 2.8 miles, that were not included in alternative D, although Modified B also excludes many of the unauthorized routes that would have been included under D. AR 12889 (Modified B includes 23.2 miles of unauthorized routes, 11.6 fewer miles than D). The final BA again uses the two step process to determine whether routes or areas within riparian conservation areas were also within suitable habitat. The GIS database identified four route segments, amounting to 0.41 miles, that were “in suitable habitat ... as defined using the GIS model.” AR 12,899. 37 The Forest Service then looked to additional data sources to decide whether or not the habitat surrounding these four route segments was actually suitable. For two of the four routes, the Forest Service conducted its own “field reconnaissance,” whereas for the other two, the Forest Service relied on “surveys ... conducted along these streams during hydroelectric re-licensing processes.” Id. Based on this additional data, the Forest Service concluded that none of the routes were in habitat with conditions suitable for frogs. Id. The For *1172 est Service applied a similar analysis with regard to three “staging areas.” AR 12, 900. 38

From there, the final BA’s explanation of its reasoning becomes unclear, giving rise to plaintiffs’ first ESA argument. The BA explicitly states that “[t]he best available information indicates that 0.41 miles of unauthorized routes in the Preferred Alternative and the location of two staging areas do not meet the programmatic Project Design Criteria. ” AR 12, 902 (emphasis added). This passage goes on to state that none of these unauthorized routes were in habitat that was suitable for the listed frogs, so the decision was not likely to adversely affect the frogs. Id. As explained above, the ESA does not permit the Forest Service to make a “not likely to adversely affect” determination absent FWS’s concurrence. If the routes violated the design criteria, the Forest Service could not rest on the prior concurrence in the programmatic consultation, and the Forest Service instead needed to initiate a separate consultation. The court concludes, however, that the quoted language was merely another example of mistaken wording by the Forest Service. The Forest Service’s conclusion that none of the unauthorized routes or areas were in suitable habitat entails the conclusion that none of the routes or areas violated the pertinent aspects of design criteria 2 and 5. The apparently intended meaning of this statement is that although the Travel Management Decision designates routes in riparian conservation areas that the GIS identified as potentially suitable habitat, the habitat is in fact unsuitable, rendering the decision compliant with the design eriteria. This interpretation is further supported by the fact-that the first page of the BA states that “Modified Alternative B implements route designations as described in the Route Designation Project Design criteria for ... the California red-legged frog.... The project is therefore not likely to adversely affect these species and no further consultation is required with the Fish and Wildlife Service based upon the USFWS memo dated Dec. 27, 2006).” AR 12889. Finally, plaintiffs have not identified any potential violation of the design criteria other than the designation of routes in suitable habitat. Accordingly, the court rejects plaintiffs’ contention that the Forest Service itself conceded that the decision violated the design criteria.

Plaintiffs’ second argument challenges the way in which the Forest Service determined habitat suitability. Plaintiffs argue that “field reconnaissance” and the other data was insufficient to demonstrate unsuitability, and that the Forest Service should have instead “surveyed” the areas according to a FWS protocol, which would determine whether frogs were actually present in the areas. Plaintiffs attempt to support this argument by citing the language used by FWS in its 2007 concurrence regarding alternative D. Pis.’ Br. 26 (Dkt. 52-1) (citing AR 28). The 2007 concurrence, however, does not suggest that surveys (whether conducted to protocol or otherwise) are necessary for determining whether habitat is suitable. Instead, the concurrence indicates that if routes are in suitable habitat, then surveys may reveal that the routes are nonetheless unlikely to adversely affect listed frogs. Plaintiffs of *1173 fered no other purported authority demonstrating that reconnaissance was inadequate to determine habitat suitability. Accordingly, plaintiffs have not shown that the Forest Service’s suitability determinations were arbitrary or capricious. 39

2. The “Ephemeral Streams” Argument

Riparian conservation areas are defined to include ephemeral streams. Plaintiffs’ next argument is that the Forest Service’s application of the design criteria improperly excluded ephemeral streams from its analysis. Defendants argue that this argument is barred by ESA § 11(g)(2) because it falls outside the scope of plaintiffs’ notice of intent to sue 40 and that this argument fails on the merits. The court concludes that plaintiffs provided adequate notice and that by excluding ephemeral streams the Forest Service failed to demonstrate compliance with the design criteria.

a. Notice of the Ephemeral Stream Argument

The ESA requires a would-be private plaintiff to notify both the party alleged to be in violation of the ESA and the Secretary of the pertinent Service of the plaintiffs intent to sue at least 60 days prior to filing suit. ESA § ll(g)(2)(A)(i), 16 U.S.C. § 1540 (g)(2)(A)(i). This requirement is jurisdictional. Sw. Ctr. For Biological Diversity v. Bureau of Reclamation, 143 F.3d 515 , 520 (9th Cir.1998). “A failure to strictly comply with the notice requirement acts as an absolute bar to bringing suit under the ESA.” Id. at 522 (citing Hallstrom v. Tillamook County, 493 U.S. 20, 26-28 , 110 S.Ct. 304 , 107 L.Ed.2d 237 (1989) and Lone Rock Timber Co. v. U.S. Dept. of Interior, 842 F.Supp. 433, 440 (D.Or.1994)).

Here, defendants concede that plaintiffs properly submitted a timely notice to the proper parties. Defendants’ argument is that plaintiffs’ claim regarding suitability of ephemeral streams falls outside the scope of this notice. The notice must “provide sufficient information of a violation so that the Secretary or [alleged violator] could identify and attempt to abate the violation.” Sw. Ctr. for Biological Diversity, 143 F.3d at 522.

Two Ninth Circuit cases have addressed whether ESA notice letters were sufficiently specific, and the letter at issue in this suit surpasses the thresholds set by these cases. In Southwest Center, plaintiffs’ complaint argued in pertinent part that the Bureau of Reclamation was violating the ESA “by allowing its operations to jeopardize the [Southwestern Willow] Flycatcher, and ... by ‘taking’ Flycatchers in the Lake Mead delta without a valid incidental take statement.” 143 F.3d at 519. Although plaintiffs had timely submitted three notice letters, none of these letters indicated that plaintiffs “had a grievance about the Flycatcher habitat at the Lake Mead delta.” Id. at 521. Without this information, the letters failed to “provide sufficient information of a violation so that the Secretary [of the Interior] or the [Bureau of] Reclamation could identify and attempt to abate the violation.” Id. at 522. *1174 Instead, these letters merely demonstrated that plaintiffs desired consultation regarding the Bureau’s operations in a different geographical area and that plaintiffs believed that a memorandum of agreement regarding management of that area should be invalidated. Id. at 521. The Northern District of California recently relied on Southwest Center to hold that where the notice of intent to sue identified the location of the asserted violation, the provisions of the ESA at issue, and the agency action at issue, the notice was sufficiently specific. Ctr. for Biological Diversity v. Chertoff No. C-08-2999, 2009 WL 889042 , *4 (N.D.Cal. Mar. 30, 2009).

In Marbled Murrelet v. Babbitt, 83 F.3d 1068 (9th Cir.1996), the court held that plaintiffs had provided adequate notice regarding their ESA § 7(a)(2) claim notwithstanding the fact that the notice letter primarily focused on ESA § 9. 83 F.3d at 1073 . Because the letter included a sentence reciting the pertinent section 7 obligations and asserting a violation thereof, “the letter as a whole provided notice sufficient to afford the opportunity to rectify the asserted ESA violations.” Id.

In this case, plaintiffs’ notice letter argued that the Forest Service violated the ESA by designating routes within the ENF without adequately consulting with FWS, thereby violating ESA § 7(a)(2). Fed. Defs.’ Supp. Br. Ex. A, 10-11 (Dkt. 74-1). The notice letter highlighted the importance of “suitable breeding and non-breeding habitats,” id. at 4, and argued that the decision designates routes through streams and other aquatic resources, which will harm the California red-legged frog, id. at 9-10. Thus, plaintiffs provided all of the information that was found lacking in Southwest Center. Under Marbled Murrelet , it is irrelevant that the notice letter also asserted violations of ESA §§ 2(c), 7(a)(1), 9, or that the letter’s primary argument regarding consultation was not the ephemeral streams argument at issue here.

Defendants argue that further specificity was required, relying on ONRC Action v. Columbia Plywood, Inc., 286 F.3d 1137 (9th Cir.2002). ONRC Action is inapplicable here because it rested on a Clean Water Act (“CWA”) regulation with no analogue under the ESA and on particular facts not present here. Beginning with the first issue, the language of the ESA and CWA notice requirements is similar. 16 U.S.C. § 1540 (g)(2); 33 U.S.C. § 1365 (b)(1)(A). “[C]ourts generally interpret similar language in different statutes in a like manner when the two statutes address a similar subject matter.” U.S. v. Novak, 476 F.3d 1041, 1051 (9th Cir.2007) (citing Metro. Life Ins. Co. v. Taylor, 481 U.S. 58, 65 , 107 S.Ct. 1542 , 95 L.Ed.2d 55 (1987); Northernss v. Bd. of Educ., 412 U.S. 427, 428 , 93 S.Ct. 2201 , 37 L.Ed.2d 48 (1973)). However, where regulations have been promulgated under one such statute but not the other, cases interpreting those regulations do not apply to the latter statute. Glenbrook Homeowners Ass’n v. Tahoe Reg’l Planning Agency, 425 F.3d 611, 615-16 (9th Cir.2005). An Environmental Protection Agency regulation interprets the CWA notice provision, and this regulation includes an extensive list of information that must be included in the notice. 40 C.F.R. § 135.3 (a). No analogous regulation exists under the ESA. ONRC Action rested in part on this regulation. 286 F.3d at 1143 (quoting 40 C.F.R. § 135.3 (a)). Thus, ONRC Action provides limited guidance, if any, to interpretation of the ESA’s notice provision.

ONRC Action also relied on an unusual factual situation not present here. The complaint in that case offered three theories as to why a permit was invalid. The notice letter had “in quite explicit language ... put forward a particular theory on which the permit was invalid,” without dis *1175 cussing the other two theories. Id. at 1143 . The court held that notice was inadequate as to these other theories. “[W]hen the 60-day notice so specifically identified only one attack,” the defendant “was not required to speculate as to all possible attacks.” Id. As other courts have observed, ONRC Action “was silent as to whether a more generalized notice (i.e., one merely stating that the permit was invalid) would have been sufficient.” N. Calif. River Watch v. Lake County Sanitation Dist., No. C 03-4552, 2004 WL 3154580 , *3 (N.D.Cal. Sept. 2, 2004) (interpreting the CWA’s notice provision). In this case, plaintiffs’ notice did not specifically argue that the Forest Service’s failure to evaluate the suitability of ephemeral streams precluded reliance on the programmatic consultation. Nonetheless, the notice broadly argued that possible effects on frogs precluded reliance on the programmatic consultation. Fed. Defs.’ Supp. Br. Ex. A, 10 (Dkt. 74-1). Although the notice advanced some arguments as to which effects had not been considered or as to why the programmatic consultation could not be used, these arguments, when read in the context of the notice as a whole, were not so “particular” and “specific” as to implicitly disclaim reliance on other theories supporting plaintiffs’ broader argument.

Accordingly, plaintiffs provided adequate notice of their ephemeral stream argument.

b. Merits of the Ephemeral Stream Argument

Plaintiffs argue that because riparian conservation areas include ephemeral streams, the Forest Service was required to analyze whether routes near ephemeral streams intruded on suitable habitat. The court concludes that the Forest Service improperly excluded ephemeral streams from its analysis, and thereby failed to determine whether the routes actually complied with the design criteria.

Ephemeral streams are those that “are not connected to the subsurface water system and flow only in response to intense rainstorms that exceed the infiltrative capacity of the soil or during snowmelt .... Since ephemeral streams do not have a sustainable source of flow, they are not able to support riparian plant species.” Dkt. 74-2 (SNFPA FEIS page 203). 41

Riparian conservation areas include areas surrounding ephemeral streams. As explained in the final EIS:

AR 2530 (FEIS 3-33 n. 2) (emphasis added). The Record of Decision for the SNFPA, cited by the EIS, states that the term “seasonally flowing streams” “includes intermittent and ephemeral streams.” AR 10979.

The final EIS explicitly excluded ephemeral streams from its discussion of riparian conservation areas. AR 2534; see also AR 2424 n. 3, 2676 n. 11, 2680 n. 14. The parties agree that BA similarly excluded ephemeral streams from the analysis. Thus, under the multi-step suitability analysis discussed above, the first step (using the GIS) never identified ephemeral streams as potentially suitable habitat.

The parties dispute whether ephemeral streams can provide suitable habitat, but *1176 the preliminary question is whether the Forest Service considered this issue. If the Forest Service properly determined that ephemeral streams could not provide suitable habitat, then no further discussion of ephemeral streams was required in the analysis of the design criteria, and this theory of ESA liability fails.

All parties’ arguments primarily rely on the same passages in the record, which define the elements of suitable frog habitat:

AR 12,895-96 (BA) (citations omitted). 42 Except for the last sentence, the BA quotes this language from FWS’s rule designating critical habitat for the California red-legged frog. 43

Plaintiffs argue that ephemeral streams may constitute both aquatic breeding and non-breeding habitat. As to breeding habitat, plaintiffs rely on the statement that this habitat can include “ephemeral ... water bodies.” The term “ephemeral water bodies” appears not to be a simple synonym for “ephemeral stream.” Rather, “water bodies” include streams, ponds, and perhaps other features., See Fed. Defs.’ Supp. Brief at 8 n. 5. “Ephemeral,” as used in the SNFPA, only means “not connected to the subsurface water system.” Id. Ex. B. This explains why an ephemeral pond, for example, might catch and retain rainwater for a period of 20 weeks or more, but an ephemeral stream will dry up after the rain stops. Thus, despite the reference to ephemeral water bodies in the definition of aquatic breeding habitat, the BA implicitly but adequately demonstrates that ephemeral streams cannot provide such habitat.

*1177 The BA does not, however, demonstrate that ephemeral streams cannot provide suitable non-breeding habitat. As plaintiffs observe, non-breeding habitat includes “ephemeral ... water bodies” and is not limited to bodies that hold water for at least 20 weeks. The Forest Service’s brief argues that ephemeral streams cannot constitute non-breeding habitat because “ephemeral streams do not provide wetland areas or riparian vegetation.” Fed. Defs.’ Supp. Br. 7. Although the record indicates that ephemeral streams do not support riparian vegetation, id. Ex. B, defendants have not cited any explanation as to why riparian vegetation is necessary. The BA merely states that non-breeding habitat must “provide for shelter, foraging, predator avoidance, and aquatic dispersal.” AR 12,896. The BA indicates that these needs can be met without riparian vegetation. In discussing upland habitat, the BA states that “various vegetational series,” including grasslands and woodlands, can provide shelter, forage, and predator avoidance. Id. Common sense suggests that “aquatic dispersal” may occur when water is present in low gradient ephemeral streams, regardless of whether riparian vegetation is present, and defendants have not argued otherwise. Moreover, the BA specifically concluded that low-gradient seasonal streams provided non-breeding habitat; insofar as seasonal streams include ephemeral streams, this suggests that low gradient ephemeral streams can provide non-breeding habitat.

The only statement in the record approaching an explicit discussion of whether ephemeral streams may provide suitable habitat occurs in the EIS’s explanation as to why ephemeral streams were excluded from the overall analysis. The EIS states that ephemeral streams “generally do not contain aquatic habitat that is considered necessary for the survival and reproduction” of any threatened, endangered, and sensitive species. AR 2534. It is unclear whether the Forest Service uses “necessary” as a synonym for “suitable.” If it does, then this blanket assertion of unsuitability is inadequate because the Forest Service has failed to explain its basis. On the other hand, if “necessary” and “suitable” have distinct meanings, the Forest Service has overstepped its role under the ESA. The Forest Service can only rely on the 2006 programmatic concurrence by strictly adhering to its design criteria. A conclusion that routes were in habitat that was suitable but unnecessary would not adhere to these criteria, and would therefore require a separate concurrence from FWS. 44

Finally, the court rejects defendants’ contention that FWS’s 2007 concurrence approved of the exclusion of ephemeral streams. Neither the draft BA nor the draft EIS stated that ephemeral streams were excluded from the analysis. Instead, the draft EIS explicitly stated that riparian conservation areas included “150 feet [on] each side of ephemeral and seasonal streams.” AR 1582 (DEIS 76 n. 2). If the draft BA excluded ephemeral streams from analysis, it appears that nothing in the record communicated that fact to *1178 FWS. Accordingly, the Forest Service cannot rely on an implicit approval from FWS to overcome the issues discussed above.

In summary, the programmatic consultation required the Forest Service to evaluate whether routes in riparian conservation areas were also in suitable habitat. Ephemeral streams and their surroundings are riparian conservation areas, but the Forest Service did not determine whether routes were near ephemeral streams or, if so, whether these areas provided suitable habitat. Accordingly, the Forest Service’s conclusion that alternative Modified B satisfied the design criteria was arbitrary and capricious.

Once again, the question of remedy is another matter. The violation here deals with the Forest Service’s failure to properly address whether roads were near “ephemeral streams,” which by definition, can exist for 20 weeks or less. The parties are directed to brief what remedy is appropriate for this failure, in light of what would appear to be the great difficulty of even identifying such ephemeral streams, in a forest of 789,994 acres.

3. Core Recovery Areas

Finally, plaintiffs argue that “the Forest Service cannot ignore impacts to frog Core Recovery Areas [designated in the Recovery Plan] ... by blindly deferring to the FWS’s four-year-old programmatic concurrence; the agency itself remains obligated to ensure its actions comply with the ESA.” Pis.’ Response Br. 23 (Dkt. 60). The Forest Service was entitled to rely on FWS’s concurrence “if a challenging party can point to no ‘new5 information-i.e., information that [FWS] did not take into account-which challenges the [concurrence’s] conclusions.” Pyramid Lake Paiute Tribe of Indians v. U.S. Dept. of Navy, 898 F.2d 1410, 1415 (9th Cir.1990). Here, the recovery plan was promulgated by FWS four years before FWS issued the programmatic concurrence. Accordingly, the recovery plan, and the designations of Core Recovery Areas therein, are not “new” information that precludes the Forest Service from relying on the programmatic concurrence.

IV. CONCLUSION

For the reasons stated above, the parties’ motions for summary judgment (Dkt. Nos. 52, 57, and 58) are GRANTED IN PART AND DENIED IN PART.

IT IS SO ORDERED.

. These plans are also referred to as "Land Resource Management Plans” or "LRMPs.”

. An earlier round of Sierra Nevada Forest Plan Amendments was adopted in 2001. AR 12,530. The 2001 amendments were superseded by the 2004 amendments, and as such, the 2001 amendments are not at issue here. AR 10,938.

. Exec. Order Nos. 11,644 and 11,989 use the terms "Off-Road Vehicles” and “ORVs,” whereas more recent documents generally use the terms "Off-Highway Vehicles” and "OHVs.” For purposes of this case, the two terms are interchangeable; both refer to vehicles designed to travel cross-country and on roads and trails not suitable for street-legal passenger vehicles. To conform to the Administrative Record, the court uses "OHV.”

. Plaintiffs have not challenged the exclusion of ML-3 to -5 roads from the analysis.

. Western Watersheds Project held that the APA standard of review applied, but that review of ESA citizen suits, unlike review of suits brought under 5 U.S.C. § 706 (2), is not confined to the administrative record. W. Watersheds Project, 632 F.3d at 495-96. See also Wash. Toxics Coal., 413 F.3d at 1034 . In this case, no party relies on this broader scope of review.

. Plaintiffs’ complaint enumerates eight claims. Nonetheless, the parties agree that resolution of the five claims articulated above will fully resolve this suit. Specifically, plaintiffs abandoned their seventh claim at oral argument; plaintiffs’ NFMA argument encompasses what were previously enumerated as the third and fourth claims; and plaintiffs’ NEPA argument regarding site-specific data encompasses what were previously the first and sixth claims.

. Other than to argue that Subpart A must precede Subpart B, plaintiffs have not challenged the Forest Service’s failure to meet its Subpart A obligations. For example, plaintiffs have not argued that the Forest Service has "unlawfully withheld or unreasonably delayed” compliance with Subpart A. 5 U.S.C. § 706 (1).

. The Court recognized an additional limit on Auer deference in Christensen v. Harris County, 529 U.S. 576 , 120 S.Ct. 1655 , 146 L.Ed.2d 621 (2000). Christensen held that an agency cannot "under the guise of interpreting a regulation ... create de facto a new regulation.” Id. at 588 , 120 S.Ct. 1655 . Christensen’s limit on Auer is not at issue in this case.

. Notably, Strom and Phan summarized the Auer rule by stating that courts "owe substantial deference to an agency's published interpretation of its own regulations.” Phan, 500 F.3d at 904 (emphasis added); accord Strom, 641 F.3d at 1063-64 .

. The court observes that the Sixth and Seventh Circuits have "expressed some skepticism about the applicability of Auer deference to amicus briefs at the circuit court level” because " ‘agency briefs, at least below the Supreme Court level, normally are not reviewed by the members of the agency itself; and it is odd to think of Congress delegating lawmaking power to unreviewed staff decisions.’ " Michigan Bell Telephone Co. v. Covad Communications Co., 597 F.3d 370 , 375 n. 6 (6th Cir.2010) (quoting Keys v. Barnhart, 347 F.3d 990, 993-94 (7th Cir.2003)). This court has no knowledge of what procedures agencies use before submitting amicus briefs to circuit courts, but the court shares the underlying skepticism about the propriety of extending Auer deference to "unreviewed staff decisions.” As to the specific question of whether amicus briefs warrant deference, this court would of course be bound by the Ninth Circuit decisions cited above.

. This portion of the record may be better understood in light of the process used to evaluate the appeal. The "responsible official” for the challenged Travel Management Decision itself was the forest supervisor for the Eldorado National Forest. AR 3272 (ROD at 3). The administrative appeal was first reviewed by the forest supervisor for the separate Sequoia National Forest. AR 831; 36 C.F.R. § 215.19 . This reviewing officer's recommendations were then adopted in full by the "appeal deciding officer,” in this case, the regional forester. AR 805; 36 C.F.R. §§ 215.8 , 215.18. Because the deciding officer incorporated the recommendation into his decision, the court attributes the language of the decision to the deciding officer.

. The court further observes that an EIS is unlike the types of documents to which the Ninth Circuit has deferred under Auer . Because the EIS in this case did not articulate a coherent interpretation of the regulation, the court need not decide whether, in another case, an EIS may be entitled to Auer deference.

. The court must confess that plaintiffs' argument on this issue is opaque. The above may not represent the precise chain of reasoning plaintiffs intended, but if plaintiffs intended something else, they failed to coherently articulate it.

. The court emphasizes that in making this argument, plaintiffs appear to solely rely on the broad purposes of the National Forest Management Act, which inherently require compromise. The court’s analysis would potentially differ if the plaintiffs had identified a particular obligation to maintain roads or to avoid the consequences of road degradation.

. Although the overall contrast between this case and Oregon Natural Desert Association, Block, and CBD v. BLM is stark, the specific figures cannot be compared on an "apples to apples” basis. One issue is a denominator problem. As the Forest Service points out, all alternatives considered closing the majority of the unauthorized road network. The picture changes if the court looks to the total road network. The status quo (i.e., the no action alternative) involved 2,188 miles of roads open to public use. AR 2410 (EIS at xxiv). Alternative E, the most restrictive, permits motorized use on 845 miles of routes, or 35% of the prior road network. Id. These figures exclude the ML-3 + roads, which were outside the scope of the Travel Management Decision. Although this raw number approaches the percentages found inappropriate in Block , these percentages also measure different things. Block concerned acres of hitherto roadless area to be opened to development, whereas this case concerns existing (albeit sometimes unauthorized) routes which may be closed to public use. Because these are fundamentally different problems, the simple percentages do not tell the whole story.

The court focuses on the percentage of unauthorized routes, rather than the percentage of total routes, because this is how plaintiffs have structured their argument. Plaintiffs have not argued that the Forest Service should have considered an alternative that would have closed more roads overall.

. The mere fact that these roads provide specific recreation opportunities is not itself sufficient to demonstrate that closing them would be unreasonable. Presumably, every unauthorized road exists because some user decided that it would provide a worthwhile recreational opportunity: every road leads *1159 somewhere. The court is satisfied with the Forest Service’s broader explanation that the unauthorized routes at issue here play needed roles in the overall road network.

. Federal Defendants also seek to distinguish these forms from those at issue in California v. Bergland, 483 F.Supp. 465, 483-87 (E.D.Cal.1980), aff'd in part, rev'd in part by Block, 690 F.2d 753 . In the instant motions, plaintiffs have not challenged the forms on this ground. Nonetheless, the court agrees that the agency’s use of route evaluation forms in this case was proper.

. No party has addressed the extent to which existing NFS trails received route evaluations or field assessments.

.In presenting this argument, plaintiffs' opening memo can fairly be described as throwing against the wall everything available in order to see what would stick, including numerous brief arguments raised only in footnotes. Here, the court discusses arguments raised in the body of plaintiffs’ memo and those arguments originally presented in footnotes but reiterated in plaintiffs’ reply. As to the other arguments, the court has considered them and found them lacking for reasons similar to those articulated here.

. Plaintiffs additionally substantively challenge the EIS's decision to designate any routes through areas that meet both of these hillslope instability criteria, arguing in particular that it was unreasonable to designate Hunter's Trail, route 14E09, as open to motorized use. Pis.' Response 12 (Dkt. 59). NEPA's requirements are procedural rather than substantive. The EIS acknowledged the number of routes that were on unstable slopes and analyzed the effects of these routes. NEPA does not require more.

. The EIS did not examine the effects of route creation on soils or erosion, because although each alternative would designate some existing unauthorized routes, no alternative would create new routes.

. As explained in part 1(c) above, what the court refers to as the ENF and Sierra Nevada Forest Plans are in fact part of the same Forest Plan.

. In any event, plaintiffs have waived any challenge to the adoption of these amendments (separate from their challenges to the EIS and ROD as a whole). If an amendment to a Forest Plan will "significantly affect[] the quality of the human environment,” then an EIS is required for the amendment itself. In this case, the Forest Service determined that the amendments to the ENF Forest Plan were non-significant. At oral argument, plaintiffs conceded that they have waived their challenge to the determination of non-significance, or to the procedures used to adopt these amendments. See, e.g., Koerner v. Grigas, 328 F.3d 1039, 1048-49 (9th Cir.2003) (failure to raise an issue in an opening brief waives the issue).

.The court does not decide whether such a broader amendment would have been non-significant or compliant with NFMA; the court merely observes that such an amendment would eliminate the conflict between the forest plans and the travel management decision.

. The record provides similarly varied estimates of the total mileage of route segments within meadows. At several places, the EIS and Aquatic Species Biological Evaluation state that alternative Modified B designates a total of 4.1 miles of routes within meadows. AR 2572 (EIS 3-75), AR 2488 (EIS 2-41), AR 12,781 (BE at 31). These estimates do not identify the particular routes counted. Other portions of the EIS state that 20 or 21 segments through meadows amount to 4.8 miles. AR 2980. Presumably the 42 segments amount to a greater mileage. See also AR 2581 (FEIS at 3-84) ("approximately five miles” of routes designated by Modified B are "within meadows.”).

. Although these prohibitions contain several apparent caveats, these caveats are not discussed by the parties' briefing or the cited portions of the Travel Management Decision EIS and documents. The first prohibition only requires that the Forest Service "consider” closing roads. Nonetheless, the Forest Service concluded that routes through meadows would need to be exempted from this obligation to "consider” their closure. AR 3274.

The second and third of these prohibitions apply only within "Management Area 28.” The court infers that only a fraction of the meadows are included in this area. There are roughly 10,416 acres of meadows in the ENF. AR 2531 (EIS 3-34), 12,769 (BE at 19). *1166 Management Area 28, however, contains only 2,937 acres. AR 5182 (ENF LRMP 4-277).

. For example, Riparian Conservation Objective # 2 sets the objective of:

AR 10,970. The "standards and guidelines" for this objective include:

AR 11,000.

. Indeed, the draft BE adopted three different criteria for objective # 2: "[riparian conservation areas] in 7th field watersheds do not exceed route densities of more than 5 mi/sq mile,” "Watersheds do not have more than 30 crossings per mile of [riparian conservation areas],” and “Routes in [riparian conservation areas] are not in poor condition." AR 12693 (Draft BE A-55). The Forest Service similarly could have adopted a more tempered criterion, perhaps putting a cap on the number of routes crossing meadows rather than flatly prohibiting such crossings.

Plaintiffs argue that the failure to explain the change in criteria between the draft and final BE itself rendered the decision arbitrary and capricious. Because the court grants plaintiffs’ NFMA claim on other grounds, the court does not address this argument.

. In this litigation, the Forest Service has rested solely on the argument that this criterion was inapplicable; the Forest Service has not attempted to argue that the alternative Modified B actually complied with this criterion. Nonetheless, before granting summary judgment to the plaintiffs on this issue, it appears appropriate for the court to look to whether the record adequately explained its findings regarding this criterion.

. Even if the 20 routes enumerated by the amendments to the ENF Forest Plan were also excluded from the riparian conservation objective analysis, there would remain routes through meadows in apparent violation of the compliance criteria, such as routes 17E16 and 17E18.

. The BE further explains that "[gjuidance for this determination was provided by the Regional Office of the Forest Service (USFS, Region 5) in February 2008.” AR 12926. No party has identified this guidance document in the record, and the court has been unable to locate it.

.Plaintiffs also argue that the exclusion is arbitrary because it is inconsistent with other positions taken by the Forest Service. The draft BE did not exclude these roads, and *1169 even the final EIS included these roads in all other aspects of its analysis. Where an agency reverses its own position the agency must offer some explanation for the change. Sierra Club v. U.S. Army Corps of Eng’rs, 772 F.2d 1043 , 1053 (2nd Cir.1985) (failure to justify change between draft and final EISs rendered final EIS deficient); Nw. Envtl. Def. Ctr. v. Bonneville Power Admin., 477 F.3d 668, 687 (9th Cir.2007) ("an agency changing its course must supply a reasoned analysis indicating that prior policies and standards are being deliberately changed, not casually ignored.”) (quotation omitted). Because the court holds that the exclusion was improper in light of the prior orders in Berry, the court does not address whether the inconsistency between the Forest Service's own statements would be independently sufficient to invalidate this exclusion.

. It appears that the Biological Assessment solely looked to the designation of unauthorized routes. AR 12889. Similarly, the design criteria specify that they are to be "used for designation of unauthorized or unclassified routes and areas for recreational wheeled motorized vehicle use. System roads, trails, and areas are not subject to these criteria or consultation.” AR 1 (Project Design Criteria, Oct. 2006). The Travel Management Decision, in addition to designating previously unauthorized routes, reconsidered whether existing system routes should be open to public use. It is not clear to the court why this reconsideration was not also subjected to ESA analysis. Plaintiffs have not raised this issue, so the court does not further address it.

. Defendants also argue that plaintiffs failed to present these two arguments during their administrative appeal, such that these arguments are barred for failure to exhaust pursuant to 7 U.S.C. § 6912 (e). Because the court rejects these arguments on the merits, the court does not decide whether section 6912(e) applies to ESA claims or whether, if it does, plaintiffs exhausted .these arguments. McBride Cotton & Cattle Corp. v. Veneman, 290 F.3d 973, 976 (9th Cir.2002) ( 7 U.S.C. § 6912 (e) is non-jurisdictional); Wash. Toxics Coal., 413 F.3d at 1033-34 (plaintiffs not required to exhaust administrative remedies provided by the Federal Insecticide, Fungicide, and Rodenticide Act prior to bringing ESA claim regarding EPA's treatment of pesticides).

. This criterion also requires that "Crossing approaches get the riders in and out of the stream channel and riparian area in the shortest distance possible while meeting the gradient and approach length standards,” AR 12,891, but plaintiffs do not invoke this additional requirement.

. Obviously, the presence of frogs in an area would indicate that the area was suitable for frogs. Neither the agencies nor the court holds that surveys are irrelevant to the question of suitability; the court merely defers to the agencies’ conclusion that suitability can be determined without recourse to surveys.

. The BA’s analysis identifies these four routes, totaling 0.41 miles, and plaintiffs' arguments concern that discussion. A subsequent table in the BA, titled "Summary of findings states that 1.3 miles of routes occur "in suitable California red-legged frog habitat that occur in RCAs, but are not necessary to cross streams.” AR 12902. Plaintiffs mention this table in passing, but plaintiffs’ arguments discuss only the 0.41 miles of routes discussed above. Accordingly, although the meaning of this table, and its consistency with the conclusions reached elsewhere in the BA, are unclear to the court, the court does not further address the issue.

. In discussing the staging areas, plaintiffs solely discuss the Hunter's Staging Area, which is on the Rubicon River. Pis.' Br. 26, 28 (Dkt. 52-2). The BA determined that this area did not provide suitable habitat. AR 12900. For. the other two staging areas within riparian conservation areas, the BA appears to conclude that habitat is suitable but that no frogs are present. Although it is unclear how this conclusion squares with the determination that these areas satisfy the design criteria, this is another issue that plaintiffs have not raised, and which the court therefore does not address.

. Plaintiffs make several tangential arguments regarding survey methodology. Because the agencies appear to agree that surveys are only required in suitable habitat, and because plaintiffs have not refuted the determination that no habitat was suitable, no surveys were required, and the question of survey methodology is irrelevant.

. Defendants have not argued that plaintiffs failed to provide notice with regard to any of plaintiffs’ other theories of ESA liability. Moreover, although defendants contend that the administrative exhaustion requirement of 7 U.S.C. § 6912 (e) applies to ESA claims against the Forest Service, defendants do not argue that plaintiffs’ failed to administratively exhaust their ephemeral stream argument.

. The administrative record provided to the court included the index to the SNFPA EIS, but not the EIS itself. The court concludes that consideration of this document is appropriate, and no party has objected to its inclusion.

. The record further defines "upland habitats” as areas bordering aquatic and riparian habitat. AR 12,896. The BA indicates that riparian conservation areas encompass upland habitat by including areas within 300 feet of perennial streams. AR 12,896 n. 8. Because upland habitat is defined by proximity to aquatic habitat, it appears that the only pertinent question here is whether ephemeral streams may constitute aquatic breeding or non-breeding habitat.

. Specifically, the EIS cited the FWS’s 2006 rule designating critical habitat for the California Red-Legged Frog, 71 Fed. Reg. 19244 , 19262-63 (April 13, 2006); AR 1735 (citing this rule). This rule was later revised, although the revision did not alter this language. The parties note that subsequent to the travel management decision, the FWS repeated this language in its Revised Declaration of Critical Habitat for the California Red-Legged Frog, 75 Fed. Reg. 12,816 , 12835 (March 17, 2010).

. Defendants further argue that evaluating effects on ephemeral streams would be unreasonably difficult, in light of the difficulty in finding these streams using the GIS. This was one reason given by the EIS for the blanket exclusion of ephemeral streams from analysis. AR 2534. Defendants provide no authority indicating that this difficulty excused the Forest Service from the consultation requirement. If the Forest Service believes that ephemeral streams may provide suitable habitat, but that this habitat is too unimportant to warrant a difficult analysis, the Forest Service should reinitiate consultation-either programmatic or site-specific-and seek concurrence as to whether analysis of ephemeral streams is necessary.