OPINION
PER CURIAM:This appeal concerns whether the process of establishing management guidelines governing 11.5 million acres of federal land in the Sierra Nevada region complied with both the procedural requirements of the National Environmental Policy Act (NEPA) and the substantive restrictions of the National Forest Management Act (NFMA). Sierra Forest Legacy, the Center for Biological Diversity, the Natural Resources Defense Council, the Sierra Club and the Wilderness Society (collectively “Sierra Forest”) appeal a largely unfavorable summary judgment against them and a favorable but limited remedial order in their NEPA and NFMA suit challenging the 2004 Sierra Nevada Forest Plan Amendment (“the 2004 Framework”) and the Basin Project, a timber harvesting project approved under the 2004 Framework. The State of California also appeals a summary judgment against it and a limited remedial order in a related NEPA action. The district court found that the U.S. Forest Service and related federal defendants (collectively “the Forest Service”) violated NEPA by failing to consider alternative actions using the same modeling techniques and management priorities, but the court rejected several other NEPA and NFMA claims. The district court ordered the Forest Service to prepare a supplemental environmental impact statement (SEIS) to remedy the NEPA error and denied Sierra Forest and California’s requests to enjoin implementation of the 2004 Framework in the interim.
*1169Sierra Forest and California argue that the Forest Service violated NEPA both by failing to consider short-term impacts of the 2004 Framework and by failing to disclose and rebut expert opposition. Sierra Forest separately contends that the Forest Service violated NEPA when approving the Basin Project by failing to analyze cumulative impacts to sensitive species. Sierra Forest also argues that the 2004 Framework violates NFMA by failing to maintain viable populations of old forest wildlife. Sierra Forest further argues that the Basin Project specifically violates NFMA by failing to comply with the 2004 Framework’s management indicator species monitoring requirement, despite a 2007 Amendment to the 2004 Framework that purports retroactively to eliminate the monitoring requirement. Both Sierra Forest and California also contend that the district court abused its discretion when considering the equitable factors governing entry of a permanent injunction. The Forest Service and numerous intervenors contest these assertions and assert several procedural bars to relief.
For the reasons that follow, a majority affirms the district court’s decision on the merits of Sierra Forest and California’s NEPA claim. Specifically, we hold that Sierra Forest and California have standing to assert a facial NEPA claim against the 2004 Framework but that the Framework SEIS adequately addressed short-term impacts to old forest wildlife and disclosed and rebutted public opposition. Similarly, we hold that the Forest Service did not violate NEPA when approving the Basin Project because the Forest Service adequately addressed cumulative impacts of the proposed management action. And we hold that the Forest Service violated NEPA by failing to update the alternatives from the 2001 Framework SEIS to reflect new modeling techniques used in the 2004 Framework SEIS. We vacate, however, the district court’s orders granting a limited remedy and remand for reconsideration of the equities of a “substantive” injunction without giving undue deference to government experts. Judge Noonan dissents for the reasons explained in his concurrence in Sierra Forest Legacy v. Rey, 577 F.3d 1015, 1024-26 (9th Cir.2009) (Noonan, J., concurring).
We remand also because we reverse the district court’s decision on Sierra Forest’s NFMA claim. A majority agrees to reverse, but for differing reasons. Judge Fisher would affirm. Judges Reinhardt and Noonan would reverse. Judge Noonan would reverse for the reasons stated in his previous concurrence in Rey. Judge Reinhardt’s holding is narrower, and therefore controls the disposition of this case.1
Judge Reinhardt holds that the Forest Service lacks power retroactively to amend forest plans, so the 2007 Amendment to the 2004 Framework did not change the population monitoring requirements for management indicator species applicable to the Basin Project. We therefore remand for the district court to determine in the first instance whether, when it approved the Project, the Forest Service had complied with the 2004 Framework’s popu*1170lation monitoring requirements. The district court should consider the 2004 Framework’s requirements as they were at the time the Project was approved, not as the Forest Service represented them to be following the 2007 Amendment. Sierra Forest’s challenge to the 2004 Framework itself is not presently ripe for judicial consideration because the district court has yet to consider whether the Basin Project complied with the 2004 Framework as it existed at the time of the Project’s approval. Until the district court decides whether the Project complies with the 2004 Framework without the 2007 Amendment, it should not consider Sierra Forest’s facial challenge to that Framework.
Judge Fisher would affirm the district court on the NFMA claim. He would hold that the Forest Service had power retroactively to amend the 2004 Framework and thus that the species monitoring claim related to the Basin Project is moot. Having concluded that the Basin Project complied with the amended 2004 Framework, Judge Fisher would hold Sierra Forest’s NFMA challenge to the 2004 Framework is ripe, as applied in the Basin Project. He would further hold, however, that the adaptive management provisions of the 2004 Framework applied in the Basin Project do not violate NFMA.
There are thus four separate opinions in this case. First, a NEPA opinion written by Judge Fisher and joined by Judge Reinhardt appears as Parts I-VI of the decision. Second, a NFMA opinion written by Judge Reinhardt appears as Part VII. Third, a dissent by Judge Fisher on the NFMA issue follows Part VII. Fourth, an opinion by Judge Noonan concurring in the result on the NFMA issue, and dissenting from the NEPA opinion, concludes the decision.
FISHER, Circuit Judge:I. Background
A. The 2004 Framework
In January 2001, the U.S. Forest Service completed the Sierra Nevada Forest Plan Amendment and an accompanying Rule of Decision (collectively “the 2001 Framework”), the conclusion of a 10-year comprehensive review process. The 2001 Framework significantly altered guidelines for management of 10 national forests and one management unit, which collectively include 11.5 million acres in the Sierra Nevada region of California. The 2001 Framework restricted logging based on overlapping guidelines concerning tree size, canopy closure and the presence of sensitive species including the California spotted owl, the northern goshawk (a bird of prey), the Pacific fisher (a small carnivorous mammal), the pine marten (another small carnivorous mammal), the willow flycatcher (a small bird) and the Yosemite toad. Less than a year later, the Forest Service began a review of the 2001 Framework to address six new policy priorities: fuel treatments, compatibility with the National Fire Plan, implementation of pilot projects outlined in the Herger-Feinstein Quincy Library Group Forest Recovery Act, 16 U.S.C. § 2104 note (HFQLG Act), impact on grazing permit holders, impact on recreational users and impact on local communities.
In 2003, the Forest Service released a draft SEIS explaining proposed changes to the 2001 Framework. The Forest Service sought internal review from both its Watershed, Fish, Wildlife, Air and Rare Plants staff and a science consistency review team, as well as interagency review from the Environmental Protection Agency and Fish and Wildlife Service. The State of California also submitted comments on the draft SEIS. The Forest Service received over 50,000 public com*1171ments, including approximately 1,300 individual letters. Numerous experts presented vigorous critiques of the plan for its lack of emphasis on species preservation and purported assumptions concerning fire ecology.
In 2004, the Forest Service released the 2004 Sierra Nevada Forest Plan Amendment and final SEIS. The 2004 Framework significantly liberalizes management restrictions, most notably by emphasizing mechanical thinning over controlled burns and increasing the maximum size of trees subject to logging from either six or 20 inches diameter at breast height to 30 inches, subject to minimum canopy retention levels and maintenance of specified percentages of existing tree volume. The 2004 Framework maintains specific protections for California spotted owls both in protected activity centers around nests and in home range core areas. On the other hand, the 2004 Framework permits broad implementation of the HFQLG Act, which loosens logging restrictions in specified areas in order to experiment with fire suppression techniques such as Community Defense Zones and Defensible Fuel Profile Zones. Finally, the 2004 Framework alters grazing limitations to permit pasture use outside of breeding periods in areas known to contain willow flycatchers — a small bird — and Yosemite toads and would allow for site-specific waivers of any limitation subject to development of local management plans. The SEIS included over 130 pages of responses to public comments.
After the Regional Forester decided to select the 2004 Framework, members of the public submitted 6,241 administrative appeals. The Chief of the Forest Service denied the appeals, with instructions that the regional forester provide supplemental information concerning adaptive monitoring, a system under which the Forest Service will continuously assess the effects of management on sensitive species and adjust practices accordingly. The Under Secretary for Natural Resources and the Environment affirmed the appeal decision a few months later.
B. Basin Project
Also in 2004, the Forest Service released an environmental assessment (EA) for the Basin Project, a timber harvesting project designed to implement the 2004 Framework. The Basin Project would harvest timber in a 40,000-acre area of the Plumas National Forest, including limited individual tree selection and group selection (removal of most trees in one-half- to two-acre clusters). The EA notes both direct and cumulative effects on the California spotted owl, northern goshawk, willow flycatcher and forest carnivores and concludes that “none of [the Basin Project’s] anticipated direct, indirect, or cumulative effects, considering both context and intensity, is expected to constitute a significant environmental effect, as that term is defined in the NEPA regulations.” (Emphasis in original.)
C. Procedural History
Sierra Forest Legacy, the Center for Biological Diversity, the Natural Resources Defense Council, the Sierra Club and the Wilderness Society are membership organizations dedicated to the protection and restoration of the environment. Their individual members use and enjoy the Sierra Nevada mountains, including searching for and observing rare wildlife. The State of California contains numerous national forests and owns large tracts of land in the Sierra Nevada region.
Sierra Forest and California filed separate actions in 2005, both challenging the adequacy of the 2004 Framework under NEPA. Sierra Forest additionally asserted violations of NFMA and attacked specific *1172projects. The federal government moved to dismiss California’s complaint for lack of standing, and the district court granted the motion with leave to file an amended complaint. See California ex rel. Lockyer v. U.S. Dep’t of Agric., No. 2:05-cv-0211, 2005 WL 1719892 (E.D.Cal. July 18, 2005) (“California v. USDA I ”). California filed an amended complaint soon thereafter, and the Forest Service did not renew its motion to dismiss. The district court granted numerous entities defendant-intervenor status in both cases, including the Quincy Library Group, the American Forest and Paper Association and the California Cattlemen’s Association.
The parties in both suits cross-moved for summary judgment. While the motions remained pending, Sierra Forest filed a motion for a preliminary injunction against implementation of the 2004 Framework in three specific projects, which the district court denied from the bench. See Sierra Nevada Forest Protection Campaign v. Rey, No. 2:05-cv-0205, 2007 WL 3034931, at *1 (E.D.Cal. Oct. 16, 2007) (“Sierra Forest I ”), rev’d sub nom. Sierra Forest Legacy v. Rey, 577 F.3d 1015 (9th Cir.2009). In a subsequent written opinion, the district court found that Sierra Forest was unlikely to succeed on the merits of its claims, that the three specific projects Sierra Forest sought to enjoin were unlikely to harm forests or species irreparably and that the balance of hardships and public interest favored permitting the government to carry out logging. See id. at *7-*ll.
On appeal, we reversed and remanded. See Sierra Forest Legacy v. Rey, 526 F.3d 1228 (9th Cir.2008) (“Sierra Forest II”), withdrawn and superseded, 577 F.3d 1015 (9th Cir.2009). Specifically, we held that the district court had abused its discretion by concluding that “[the Forest Service] complied with NEPA’s requirement to ‘[rigorously explore and objectively evaluate all reasonable alternatives.’ ” Id. at 1233 (quoting 40 C.F.R. § 1502.14(a) (2000)). We additionally concluded that at an interim stage the equities favored Sierra Forest and instructed the district court “to grant immediately a preliminary injunction on the three proposed projects to the extent that they are inconsistent with the 2001 [Framework].” Id. at 1233-34.
In response to a petition for panel rehearing and the Supreme Court’s opinion in Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008), we withdrew Sierra Forest II and issued a superseding opinion. See Sierra Forest Legacy v. Rey, 577 F.3d 1015 (9th Cir.2009) (“Sierra Forest IV”). We reiterated that Sierra Forest was likely to succeed on the merits of its NEPA claim and specified that the Forest Service had introduced new management objectives and modeling techniques in the 2004 SEIS without applying them to the full range of alternatives. See id. at 1021-22. We then remanded for application of proper legal standards in the first instance and expressed “ ‘no opinion as to whether an injunction should issue.’ ” Id. at 1024 (quoting Arcamuzi v. Continental Air Lines, Inc., 819 F.2d 935, 939 (9th Cir.1987)).
After we decided Sierra Forest II but before we granted the Forest Service’s petition for review and issued Sierra Forest IV, the district court resolved the parties’ cross-motions for summary judgment. See Sierra Nevada Forest Protection Campaign v. Rey, 573 F.Supp.2d 1316 (E.D.Cal.2008) (“Sierra Forest III”); California ex rel. Lockyer v. U.S. Dep’t of Agric., No. 2:05-cv-0211, 2008 WL 3863479 (E.D.Cal. Sept. 3, 2008) (“California v. USDA II”). In both cases, the district court granted summary judgment in favor of defendants, with the exception of the NEPA alternatives claim, on which *1173it granted summary judgment in favor of the plaintiffs. See Sierra Forest III, 573 F.Supp.2d at 1353; California v. USDA II, 2008 WL 3863479, at *28.
In Sierra Forest III, the district court first found that the 1982 regulations implementing NFMA, rather than the 2005 regulations, applied to the 2004 Framework and that the NFMA claim was ripe for adjudication “to the extent its provisions are implicated within [Sierra Forest’s] challenge to the site-specific Basin Project.” 573 F.Supp.2d at 1327-29. However, on the merits the court found that the Forest Service “did not act arbitrarily or capriciously in finding that the [Basin] Project would maintain species viability concurrently with meeting other multiple-use objectives.” Id. at 1333. The district court also found no NFMA violation due to the lack of monitoring, concluding that monitoring requirements under NFMA regulations did not dictate the particular requirements contained in the 2004 Framework and that those requirements had been removed by the 2007 Amendment to the 2004 Framework. See id. at 1333-1337. The district court similarly denied Sierra Forest’s NEPA claims, finding that “the SEIS does recognize the importance of addressing short term impacts,” which were further mitigated by adaptive management strategies, id. at 1338-42, and that the SEIS considered opposing scientific viewpoints and acknowledged scientific uncertainty, see id. at 1342-45. The court also found that the Forest Service did not act arbitrarily in violation of NEPA by concentrating its 2004 Framework cumulative effects analysis on projected reductions in stand-replacing fires resulting from intensified management or by relying on the 2004 Framework cumulative effects analysis in the Basin EA. See id. at 1346-47,1352-53. The district court next found that the Forest Service permitted adequate public participation, as required by NEPA, when preparing the Basin EA. See id. at 1348-51. Finally, the district court found that the Forest Service did violate NEPA by failing to consider an adequate range of alternatives, because the Forest Service had not updated alternatives set out in the 2001 Framework SEIS to reflect new modeling techniques used in the 2004 Framework SEIS. See id. at 1347-48.
In California v. USDA II, the district court first concluded that the State had standing to challenge the 2004 Framework based on its “ ‘responsibilities, powers, and assets’ ” concerning “wildlife, water, state-owned land, and public trust lands in and around the Sierra Nevada.” 2008 WL 3863479, at *5-*6 (quoting City of Sausalito v. O’Neill, 386 F.3d 1186, 1197 (9th Cir.2004)). The court then rejected California’s broad Administrative Procedure Act challenge, finding that the Forest Service had adequately articulated “ ‘reasoned analysis’ in adopting the provisions of the 2004 Framework” concerning fuels management, California spotted owl impacts and grazing. Id. at *6-*13 (quoting Motor Vehicle Mfrs. Ass’n v. State Farm, Mut. Auto. Ins. Co., 463 U.S. 29, 42, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)). The district court next found that the Forest Service had undertaken the analysis required by NEPA concerning old forest conditions and species including the California spotted owl, willow flycatcher and Yosemite toad. See id. at *14-*23. The court also found that the Forest Service had adequately considered opposing scientific viewpoints and uncertainty and had explained the concept and utility of adaptive management. See id. at *23-*27. Again, however, the district court found that the Forest Service did violate NEPA by failing to consider an adequate range of alternatives that all relied on the same modeling techniques. See id. at *27-*28.
*1174In a separate opinion addressing both cases, the district court granted a limited remedy for inadequacies in the NEPA alternatives analysis. See Sierra Forest Legacy v. Rey, 670 F.Supp.2d 1106 (E.D.Cal.2009) (“Sierra Forest V”). Rejecting Sierra Forest’s and California’s requests to enjoin implementation of the 2004 Framework, the court first found that it lacked jurisdiction to impose a “substantive” injunction for a “procedural” NEPA violation. Id at 1110-11. The district court also found that traditional injunction standards — specifically the public interest and the balance of hardships — did not favor an injunction setting aside the 2004 Framework. See id at 1111-13. Rather, the court ordered the Forest Service “to prepare another supplemental EIS on the[2004] Framework, one that meets the range of alternatives and analytical consistency identified by the Ninth Circuit in its decision on the preliminary injunction portion of this case” by May 1, 2010. Id at 1113.
The district court entered judgment, and Sierra Forest and California timely appealed. The Forest Service also entered a notice of appeal in Sierra Forest’s appeal, but we granted the Forest Service’s subsequent motion for voluntary dismissal. See Order, Sierra Forest Legacy v. Sherman, No. 10-15376 (9th Cir. May 10, 2010).
Sierra Forest moved for an injunction pending appeal, which the district court denied. See Sierra Forest Legacy v. Rey, 691 F.Supp.2d 1204, 1207-14 (E.D.Cal.2010) (“Sierra Forest VI ”). However, the district court granted Sierra Forest’s unopposed request to delay completion of a SEIS integrating new alternatives analysis until after resolution of the merits appeal. See id at 1214. Sierra Forest then requested that we enjoin implementation of the 2004 Framework pending appeal. We denied the motion with leave to refile because Sierra Forest had presented no evidence that ground would be broken in projects named in its complaint — Empire, Basin and Slapjack — during the pendency of its appeal. See Order, Sierra Forest Legacy v. Sherman, No. 09-17796 (9th Cir. Apr. 29, 2010). After argument Sierra Forest filed another motion for an injunction pending appeal, presenting evidence that logging was imminent in Empire, Basin and Slapjack. We granted the motion in part and enjoined logging in those three projects “that is inconsistent with the 2001 Framework, except such logging as may occur within the wildland urban intermix defense and threat zones, as defined in the” 2004 Framework SEIS. Order, Sierra Forest Legacy v. Sherman, No. 09-17796 (9th Cir. July 23, 2010). That injunction remains in place.
II. Jurisdiction
The district court had jurisdiction over Sierra Forest Legacy v. Rey under 28 U.S.C. § 1331 and entered judgment on December 18, 2009. Sierra Forest filed a notice of appeal two weeks before judgment, which we “treat[] as filed on the date of and after the entry” of judgment. Fed. R.App. P. 4(a)(2). The district court had jurisdiction over California v. U.S. Department of Agriculture under 28 U.S.C. § 1331 and entered judgment on December 7, 2009. California filed a timely notice of appeal.
The Forest Service contends that we nevertheless lack jurisdiction because the orders from which Sierra Forest and California appeal are not final. The district court ordered an agency remand for the Forest Service to prepare another SEIS to correct defects in the NEPA alternatives analysis, and an agency remand is ordinarily final only for purposes of a government appeal. See Alsea Valley Alliance v. Dep’t of Commerce, 358 F.3d 1181, 1184 (9th Cir.2004). Sierra Forest *1175argues that the judgment is final for practical purposes, given that the draft SEIS already released by the Forest Service demonstrates that “the remand will not address even the NEPA violation identified by” Sierra Forest IV, “much less the additional violations of NEPA and NFMA” that Sierra Forest alleges on appeal.
28 U.S.C. § 1291 provides us with jurisdiction over “appeals from all final decisions of the district courts of the United States,” subject to exceptions inapplicable here. “[T]he requirement of finality is to be given a ‘practical rather than a technical construction.’ ” Gillespie v. U.S. Steel Corp., 379 U.S. 148, 152, 85 S.Ct. 308, 13 L.Ed.2d 199 (1964) (quoting Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 546, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949)). We have held that
a remand order is final where (1) the district court conclusively resolves a separable legal issue, (2) the remand order forces the agency to apply a potentially erroneous rule which may result in a wasted proceeding, and (3) review would, as a practical matter, be foreclosed if an immediate appeal were unavailable.
Collord v. U.S. Dep’t of the Interior, 154 F.3d 933, 935 (9th Cir.1998). Because we apply a practical construction to the finality requirement, however, these are considerations, rather than strict prerequisites. See, e.g., Skagit County Pub. Hosp. Dist. No. 2 v. Shalala, 80 F.3d 379, 384 (9th Cir.1996) (holding that an agency remand that would not foreclose a later appeal was nevertheless “final and appealable”).
In Alsea Valley Alliance, we addressed an appeal from a district court order invalidating the listing of a particular population of salmon as “threatened” under the Endangered Species Act (ESA) and ordering the National Marine Fisheries Service to conduct further analysis consistent with the opinion. 358 F.3d at 1183. We dismissed an appeal by a group of intervenors for lack of jurisdiction, based on the absence of a final judgment, and held that “only agencies compelled to refashion then-own rules face the unique prospect of being deprived of review altogether.” Id. The decision depended on the possibility “that the action taken by the Service on remand will provide the [appellant] with all the relief it seeks,” in which case any decision by the court of appeals “could prove entirely unnecessary.” Id. at 1185.
On other hand, in Skagit County we held that a remand order may be final when the broad relief sought could not be achieved through the action the district court directed the agency to undertake, in that case a “meaningless remand” for recalculation of a portion of the claim, in other words “a party with no cake.” 80 F.3d at 384. We similarly held in Pauly v. U.S. Department of Agriculture that a narrow partial remand for a “mechanical recalculation” does not preclude appellate review of the underlying claims because the district court’s opinion is “practically final.” 348 F.3d 1143, 1148 (9th Cir.2003); see also Pit River Tribe v. U.S. Forest Serv., 615 F.3d 1069, 1075-76 (9th Cir.2010) (“Pit River II”) (“Alsea did not announce a hard-and-fast rule prohibiting a non-agency litigant from appealing a remand order.”).
It is theoretically possible that on remand the Forest Service could reconsider alternatives from the 2001 Framework in light of modeling and policy changes reflected in the 2004 Framework SEIS and conclude that one of those alternatives is preferable to the 2004 Framework. However, the final judgment rule deals in practice, not theory. The narrow injunction both left the 2004 Framework in place and placed a judicial imprimatur on the vast majority of the challenged SEIS. Moreover, the Forest Service has already re*1176leased a draft SEIS that updates the NEPA alternatives analysis, and that SEIS concludes without detailed analysis that the modified alternatives would not “fulfill the purpose and need for the proposed action.” As a practical matter, the work of both the district court and the agency is complete.
Moreover, the three considerations we articulated in Collord illustrate the practical finality of the district court’s decision. There is no question that the district court decided numerous legal issues distinct from those to be addressed in the agency remand. Although the district court’s order would not “force[ ] the agency to apply a potentially erroneous rule,” it permits adherence to rules that plaintiffs continue to challenge and the Forest Service continues to defend on appeal, “which may result in a wasted proceeding.” Collord, 154 F.3d at 935. The Forest Service recognized the inefficiency of such procedures by acquiescing to Sierra Forest’s request to stay the remedy pending appeal. Finally, although review would not be foreclosed after further administrative proceedings, we have ignored this requirement in the face of a “cakeless” remand. See Skagit County, 80 F.3d at 384.
Our recent decision in Pit River II provides a useful counterpoint. There we held that an agency remand for an entirely new EIS, along with mandatory consultation with the plaintiff Indian tribe, did not constitute a final judgment subject to review. See 615 F.3d at 1073-78. Unlike the partially erroneous NEPA analysis found by the district court in this case, the agencies in Pit River II had failed entirely to engage in the required NEPA analysis prior to a disputed lease extension. See id. at 1072-73. We also noted that the Pit River Tribe would “have an opportunity to participate in the agencies’ processes on remand” and that it was “possible that the agencies may decide that no geothermal power production should occur on the land,” obviating the need for appellate review. Id. at 1076. This broad remand stands in stark contrast to the correction ordered in this case.
For the foregoing reasons, the district court’s judgments are final and therefore subject to review under 28 U.S.C. § 1291. We therefore have jurisdiction over Sierra Forest’s and California’s appeals.
III. Standard of Review
Standing, ripeness and mootness are questions of law that we review de novo. See Mayfield v. United States, 599 F.3d 964, 970 (9th Cir.2010) (standing); California ex rel. Lockyer v. U.S. Dep’t of Agric., 575 F.3d 999, 1010 (9th Cir.2009) (ripeness); Siskiyou Regional Educ. Project v. U.S. Forest Serv., 565 F.3d 545, 559 (9th Cir.2009) (mootness). We also “review the district court’s grant of summary judgment de novo.” California ex rel. Lockyer, 575 F.3d at 1011. “Agency decisions that allegedly violate ... NEPA and[the] NFMA are reviewed under the Administrative Procedure Act (‘APA’), and may be set aside only if they are'arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’ ” Or. Natural Res. Council Fund v. Goodman, 505 F.3d 884, 889 (9th Cir.2007) (quoting 5 U.S.C. § 706(2)(A)).
Review under the arbitrary and capricious standard is narrow, and [federal courts do] not substitute [their] judgment for that of the agency. Rather, [courts] will reverse a decision as arbitrary and capricious only if the agency relied on factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be as*1177cribed to a difference in view or the product of agency expertise.
Lands Council v. McNair, 537 F.3d 981, 987 (9th Cir.2008) (en banc) (internal quotation marks and citations omitted), abrogated on other grounds by Winter, 129 S.Ct. at 375.
“Although we review the district court’s decision to grant a permanent injunction for an abuse of discretion, we review the rulings of law relied upon by the district court in awarding injunctive relief de novo.” Biodiversity Legal Found, v. Badgley, 309 F.3d 1166, 1176 (9th Cir.2002) (internal citations omitted).
IY. National Environmental Policy Act
The National Environmental Policy Act is “our basic national charter for protection of the environment.” 40 C.F.R. § 1500.1(a). NEPA requires that
all agencies of the Federal Government shall include in every recommendation or report on proposals for legislation and other major Federal actions significantly affecting the quality of the human environment, a detailed statement by the responsible official on (i) the environmental impact of the proposed action, (ii) any adverse environmental effects which cannot be avoided should the proposal be implemented, (iii) alternatives to the proposed action, (iv) the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity, and (v) any irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented.
42 U.S.C. § 4332(2)(C). Agencies must also “study, develop, and describe appropriate alternatives to recommended courses of action in any proposal which involves unresolved conflicts concerning alternative uses of available resources.” Id. § 4332(2)(E). When an agency produces an environmental impact statement (EIS), it must “provide full and fair discussion of significant environmental impacts and shall inform decisionmakers and the public of the reasonable alternatives which would avoid or minimize adverse impacts or enhance the quality of the human environment.” 40 C.F.R. § 1502.1.
“NEPA ... does not impose any substantive requirements on federal agencies — it ‘exists to ensure a process.’ ” Lands Council, 537 F.3d at 1000 (quoting Inland Empire Pub. Lands Council v. U.S. Forest Serv., 88 F.3d 754, 758 (9th Cir.1996)). So long as “the adverse environmental effects of the proposed action are adequately identified and evaluated, the agency is not constrained by NEPA from deciding that.other values outweigh the environmental costs.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989).
A. Standing
As a preliminary matter, the Forest Service contends that the State of California lacks standing to challenge the 2004 Framework under NEPA. Several intervenors also argue that Sierra Forest lacks standing to challenge the 2004 Framework under NEPA, asserting that no person or entity may ever have standing for “a facial challenge to a first-level NEPA document.”
Constitutional standing requires a plaintiff to demonstrate:
(1) it has suffered an “injury in fact” that is (a) concrete and particularized and (b) actual or imminent, not conjectural or hypothetical; (2) the injury is fairly traceable to the challenged action of the defendant; and (3) it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.
*1178Friends of the Earth, Inc. v. Laidlaw Envtl. Servs., 528 U.S. 167, 180-81, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000).
[Deprivation of a procedural right without some concrete interest that is affected by the deprivation — -a procedural right in vacuo — is insufficient to create Article III standing. Only “a person who has been accorded a procedural right to protect his concrete interests can assert that right without meeting all the normal standards for redressability and immediacy.”
Summers v. Earth Island Inst., 555 U.S. 488, 129 S.Ct. 1142, 1151, 173 L.Ed.2d 1 (2009) (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 572 n. 7, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992)) (emphasis in original).
1. California
California, like all states, “does not have standing as parens patriae to bring an action against the Federal Government.” Alfred L. Snapp & Son, Inc. v. Puerto Rico, 458 U.S. 592, 610 n. 16, 102 S.Ct. 3260, 73 L.Ed.2d 995 (1982). States are also not “normal litigants for the purposes of invoking federal jurisdiction.” Massachusetts v. EPA 549 U.S. 497, 518, 127 S.Ct. 1438, 167 L.Ed.2d 248 (2007). On the other hand, “well-founded desire to preserve [a state’s] sovereign territory” “support[s] federal jurisdiction,” which may be further reinforced by ownership of “a great deal of the ‘territory alleged to be affected’” by a challenged federal action. Id. at 519, 127 S.Ct. 1438 (quoting Georgia v. Tenn. Copper Co., 206 U.S. 230, 237, 27 S.Ct. 618, 51 L.Ed. 1038 (1907)). A political body may also uniquely “sue to protect its own ‘proprietary interests’ that might be ‘congruent’ with those of its citizens,” including “responsibilities, powers, and assets.” City of Sausalito v. O’Neill, 386 F.3d 1186, 1197 (9th Cir.2004) (quoting Colorado River Indian Tribes v. Town of Parker, 776 F.2d 846, 848 (9th Cir.1985)); see also id. at 1198 (including among such interests the “ability to enforce land-use and health regulations” and “protecting its natural resources from harm”). California has unquestionably asserted a well-founded desire to protect both its territory and its proprietary interests both from direct harm and from spill-over effects resulting from action on federal land, including ownership and trusteeship over “wildlife, water, State-owned land, and public trust lands in and around the Sierra Nevada.” Therefore, the State of California has concrete and particularized interests protected by the application of NEPA to the 2004 Framework.
Because of California’s protected interests, this case is distinguishable from Summers v. Earth Island Institute. In Summers, the plaintiff environmental organizations had challenged a failure to apply particular regulations to a logging project, the Burnt Ridge Project, but they settled the site-specific dispute during litigation. See 129 S.Ct. at 1148. The environmental organizations had submitted only two affidavits to establish standing: one addressing the Burnt Ridge Project and one, the Bensman affidavit, that did not address the challenged regulation, a particular site or a future injury. See id. at 1150. The Court held that a “vague desire” to visit locations that might be harmed by the challenged regulations was insufficient to establish a particularized interest. Id. at 1150-51. The difference here is that California maintains concrete interests spanning its entire territory. Unlike the Bensman affidavit, California’s affidavit addresses the challenged regulation, the entirety of the Sierra Nevada and future injury that the State claims will result from any logging under the 2004 Framework. Thus, California’s unique proprietary interests will invariably be affected by the 2004 Framework because logging will occur soon somewhere in the *1179State. The potential injury is neither vague nor speculative.
California has also asserted actual harm to its procedural interest in federal management decisions made under the deliberation-forcing requirements of NEPA. The State’s standing is not defeated by its not having submitted affidavits establishing approval of specific logging projects under the 2004 Framework. A land resource and management plan (LRMP), such as the 2004 Framework, “sets logging goals, selects the areas of the forest that are suited to timber production, and determines which probable methods of timber harvest are appropriate [but] does not itself authorize the cutting of any trees.” Ohio Forestry Ass’n v. Sierra Club, 523 U.S. 726, 729, 118 S.Ct. 1665, 140 L.Ed.2d 921 (1998) (internal quotation marks and citations omitted). Upon approval of a particular logging project, environmental “harm is more imminent and more certain.” Id. at 734, 118 S.Ct. 1665. However a procedural injury is complete after an LRMP has been adopted, so long as is it is fairly traceable to some action that will affect the plaintiff’s interests. See id. at 737, 118 S.Ct. 1665 (acknowledging the possibility of “a person with standing” “at the time the [NEPA violation] takes place” in an LRMP and before anticipated implementation decisions in a site-specific project); see also Pit River Tribe v. U.S. Forest Serv., 469 F.3d 768, 785 (9th Cir.2006) (“Pit River I ”) (“The purpose of an EIS is to apprise decisionmakers of the disruptive environmental effects that may flow from their decisions at a time when they retain a maximum range of options.” (emphasis added and internal quotation marks and citation omitted)). The 2004 Framework permits the Forest Service to implement forest management projects in California, and there is no real possibility that the Forest Service will then decline to adopt any management projects under the framework governing over 10 million acres of federal land. Cf. Rapid Transit Advocates, Inc. v. Southern California Rapid Transit District, 752 F.2d 373, 378 (9th Cir.1985) (per curiam) (holding, after an agency decision to fund design and engineering work, that harm was not imminent and that plaintiffs therefore lacked standing). Even if a specific application were required for standing, we would take judicial notice of the Basin EA, contained within Sierra Forest’s record on appeal, which approves intensive management within California. Therefore, we hold that California has standing to assert a facial NEPA claim against the 2004 Framework.2
2. Sierra Forest
Sierra Forest similarly has standing to bring a facial NEPA challenge to the 2004 Framework, independent from specific implementing projects. Sierra Forest has challenged the 2004 Framework as an imminent source of harm to its *1180members’ interests in the area encompassed by the Basin, Slapjack and Empire projects within the Plumas National Forest, and there is no question that affidavits establish members’ interests in those areas. For the reasons discussed above, a procedural NEPA violation is complete even before an implementing project is approved. Nor could the Forest Service cure flaws in an LRMP in the EIS for a site-specific project. See Pit River I, 469 F.3d at 785 (“[Dilatory or ex post facto environmental review cannot cure an initial failure to undertake environmental review.”). We have never held that an LRMP is not subject to facial attack based on an alleged NEPA violation.
B. 2004 Framework: Short-Term Harm
Sierra Forest and California first argue that the Forest Service violated NEPA by focusing on uncertain long-term impacts in the 2004 Framework SEIS, at the expense of known near-term harm. “The sweeping policy goals announced in § 101 of NEPA are ... realized through a set of ‘action-forcing’ procedures that require that agencies take a ‘hard look’ at environmental consequences.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989) (internal quotation marks and citations omitted). “A ‘hard look’ includes considering all foreseeable direct and indirect impacts. Furthermore, a ‘hard look’ should involve a discussion of adverse impacts that does not improperly minimize negative side effects.” N. Alaska Envtl. Ctr. v. Kempthorne, 457 F.3d 969, 975 (9th Cir.2006) (internal quotation marks and citations omitted); see also 42 U.S.C. § 4332(2)(C)(iv) (requiring analysis of “the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity”); Or. Natural Res. Council Fund v. Brong, 492 F.3d 1120, 1133 (9th Cir.2007) (“[G]eneral statements about possible effects and some risk do not constitute a hard look absent a justification regarding why more definitive information could not be provided.” (internal quotation marks and citations omitted)). Nevertheless, where agency experts have analyzed the immediate harm of a proposed action, the Forest Service may conclude that long-term benefits outweigh short-term costs. See Native Ecosystems Council v. U.S. Forest Serv., 428 F.3d 1233, 1251 (9th Cir.2005). Proper analysis may also rely on long-term modeling, despite the inherent uncertainty of projections. See Nev. Land Action Ass’n v. U.S. Forest Serv., 8 F.3d 713, 718 (9th Cir.1993).
The 2004 Framework final SEIS addresses short-term effects concerning the California spotted owl, fisher and marten — old forest species at the core of Sierra Forest’s NEPA claim. Concerning the California spotted owl, the SEIS acknowledges that implementing the HFQLG Act is likely to reduce owl habitat, widen gaps between habitat patches and create stand openings. Similarly, the SEIS indicates that implementation of the 2004 Framework will eliminate some canopy cover, which affects both California spotted owl habitat and habitat for owl prey. In sum, the SEIS states, “there is some risk of negatively affecting California spotted owls in the short term because of the uncertainty associated with the effects of using mechanical treatment,” potentially affecting five percent of protected activity centers. In order to mitigate potential harm and enhance understanding of the effects of forest management on the California spotted owl, the 2004 Framework adopts specific owl monitoring programs.
The SEIS provides similar details with regard to the fisher and marten. It acknowledges that treatments “may increase fragmentation and create barriers to fisher *1181movement” and yield “short-term trade offs in current habitat quality” through reductions in canopy closure and logging of large trees. The SEIS also notes that implementation of the 2004 Framework will harm marten habitat as compared to the 2001 Framework in the short term by eliminating one percent of canopy closure and five percent of old growth forests. Group selection logging in the HFQLG pilot project area will affect 0.57 percent of the project area per year for the first five years, reducing marten habitat, but would not render large areas “unsuitable for foraging or dispersal.”
California also raises concerns about short-term effects on the willow flycatcher and the Yosemite toad. The SEIS explains that the 2004 Framework creates some short-term risk to the willow flycatcher by permitting grazing in nine sites where flycatchers have not been observed since 1982 or have been observed only after breeding season. Similarly, the SEIS explains that during the development of site-specific management plans, permitting cattle to graze increases the risk that Yosemite toads will be trampled. California does not specify what other types of short-term impacts it anticipates.
California also points out that a particular chart found in the 2001 Framework SEIS was omitted from the 2004 Framework SEIS. This chart attached a degree of certainty to whether old forest habitats would be preserved by each alternative. Although NEPA requires us to determine “whether the EIS’s ‘form, content and preparation foster both informed decisionmaking and informal public participation,’ ” City of Sausalito, 386 F.3d at 1207 (quoting Idaho Conservation League v. Mumma, 956 F.2d 1508, 1519 (9th Cir.1992)), we may apply only “pragmatic judgment,” Churchill County v. Norton, 276 F.3d 1060, 1071 (9th Cir.2001) (internal quotation marks and citations omitted), and may not impose “upon the agency [our] own notion of which procedures are ‘best’ or most likely to further some vague, undefined public good,” Vt. Yankee Nuclear Power Corp. v. Natural Res. Def. Council, Inc., 435 U.S. 519, 549, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978). We cannot mandate that a SEIS include a particular graph, no matter how helpful. Moreover, the SEIS states in text that the 2001 and 2004 Frameworks “are indistinguishable in their short-run outcomes” and presents in graphical form short-term acreage of old growth forest projected to exist during implementation of the alternatives under serious consideration. Collectively, the substantial discussion of potential harm to old forest species and projections of the size of old growth forests adequately addresses the relative short-term effects of these two alternatives.
There is no question that implementation of the 2004 Framework will destroy some owl habitat, a consequence the SEIS plainly acknowledges. Annualizing anticipated effects would not have furthered the deliberation-forcing goals of NEPA. Moreover, the projected loss of habitat is small compared to both the scale of the Sierra Nevada and earlier management practices insensitive to the needs of old forest species. For example, the 2004 Framework SEIS states, “Within the HFQLG project area ... [the 2004 Framework] is projected to result in roughly 65,000 fewer acres of suitable owl habitat in year 20 than [the 2001 Framework].” Less than 20 years ago, the federal courts reviewed practices that raised the question “whether the owl can survive the near-term loss of another half-million acres of its habitat.” Seattle Audubon Soc’y v. Moseley, 798 F.Supp. 1473, 1478 (W.D.Wash.1992). It is the prerogative of the Forest Service to determine that long-term effects — even long-term effects *1182subject to uncertainty — remain desirable despite short-term harm. The Forest Service, when promulgating the 2004 Framework, disclosed and focused adequately on short-term effects of intensified management and therefore complied with NEPA.
C. 2004 Framework: Dissenting Views
Sierra Forest and California also both argue that the Forest Service violated NEPA when approving the 2004 Framework by failing to disclose and to respond to the views of experts opposed to intensified management. “In preparing the final EIS, the agency must ‘discuss at appropriate points ... any responsible opposing view which was not adequately discussed in the draft statement and [must] indicate the agency’s response to the issue raised.’ ” Robertson, 490 U.S. at 350 n. 13, 109 S.Ct. 1835 (quoting 40 C.F.R. § 1502.9(b)) (alterations in original); see also Lands Council, 537 F.3d at 1001 (“[T]he Forest Service must acknowledge and respond to comments by outside parties that raise significant scientific uncertainties and reasonably support that such uncertainties exist.”); 40 C.F.R. § 1503.4(a) (“An agency preparing a final environmental impact statement shall assess and consider comments both individually and collectively, and shall respond ... in the final statement.”). The mere presence of expert disagreement does not violate NEPA because “experts in every scientific field routinely disagree.” Lands Council, 537 F.3d at 1001. NEPA also does not require an agency to publish “every comment ... in the final EIS. Nor must an agency set forth at full length the views with which it disagrees.” California v. Block, 690 F.2d 753, 773 (9th Cir.1982) (internal citations omitted).
Sierra Forest and California present an array of experts who submitted comments to the 2004 Framework draft SEIS. As a general matter, the final SEIS incorporates a science consistency review that raises conflicting perspectives. It also acknowledges and responds to general critiques concerning the use of science. Plaintiffs’ experts’ more specific criticisms can be broken down into five categories.
Most critiques concerned the California spotted owl. The final SEIS, however, expressed uncertainty concerning California spotted owl analysis, noted submission of concerns “about the reliability of habitat projections” for the California spotted owl and disclosed “conflicting science about the effects of canopy cover reductions from fuel treatments.” Most importantly, the SEIS dedicates nearly 12 pages to airing concerns about California spotted owl management and providing agency responses.
Second, the experts expressed concerns regarding the uncertainty inherent in long-term modeling. The final SEIS acknowledges that “[e]oncerns have been expressed about the reliability of habitat projections used in this analysis and the deterministic nature of the models underlying those projections,” but explains the importance and inherent flaws of modeling. The SEIS also includes modeling appendices, which describe modeling assumptions and “sensitivity analysis to address questions about uncertainty in modeling outcomes.” Moreover, the Regional Forester acknowledged the validity of some critiques and chose not to rely on 120-year projections when deciding to adopt the 2004 Framework.
Third, the experts argued that the 2004 Framework will lead to further decline of fisher and marten populations. The final SEIS acknowledges uncertainty concerning marten and fisher habitat use and the effect of management on persistence in the Sierra Nevada. It also recognizes that concerns have been expressed “that treatments ... may increase fragmentation and *1183create barriers to fisher movement,” that reductions to canopy that will harm fisher habitat and about “effects of the [HFQLG] pilot project, particularly on marten in eastside pine habitats.” The SEIS also airs and responds to three pages of additional concerns regarding fisher and marten management.
Fourth, the experts raised concerns regarding meadow species, such as the willow flycatcher and Yosemite toad. The final SEIS acknowledges uncertainty concerning the effects of grazing on these species and accepts one of the willow flycatcher working group’s suggestions concerning development of a conservation strategy. More importantly, the SEIS raises and addresses a host of public concerns regarding both meadow species in the volume dedicated to responding to public comments.
Fifth, the experts critique the fire ecology underpinning the core management analysis. The final SEIS notes uncertainty “whether unaltered wildfires would have a greater or lesser impact ... on ecosystem integrity and habitat” compared to fires in treated areas. Again, the SEIS acknowledges and responds to a substantial number of critics addressing fire and fuels management, including critiques of the scope and methods of treatment.
California specifically argues that “the agency did not bring attention to ... critical expert comments but rather mixed them into the stack of all public comments .... ” Similarly, Sierra Forest contends that the “SEIS does not disclose that these ‘other’ viewpoints were expressed by the country’s leading spotted owl experts, including the retired Forest Service owl expert ... and the agency’s own wildlife office.” However, NEPA does not require that a final SEIS prioritize the concern of scientific experts or disclose their identities amongst public critiques. The practical concerns of individual landholders or hikers may be just as important — and just as trenchant — as the formal submissions of academic experts. So long as an EIS addresses the substance of public comments, it need not single out the authors.
In sum, the SEIS dedicates over 120 pages to raising and meaningfully responding to public critiques. That is all NEPA requires. Sierra Forest and California do not argue that the Forest Service’s decision not to adopt critiques was arbitrary, capricious or contrary to law. Therefore, the Forest Service did not violate NEPA by failing to disclose conflicting scientific opinion.
D. Basin Project: Cumulative Impacts
Sierra Forest separately challenges approval of the Basin Project under NEPA, arguing that the Basin EA failed to assess the cumulative impact of the Project. To comply with a NEPA alternatives analysis,
the Forest Service must consider, among other things, the “cumulative impacts” of the proposed action, which NEPA’s implementing regulations define as the “impact on the environment which results from the incremental impact of the action when added to other past, present, and reasonably foreseeable future actions.... ”
League of Wilderness Defenders v. US. Forest Serv., 549 F.3d 1211, 1216 (9th Cir.2008) (quoting 40 C.F.R. § 1508.7). A cumulative impact analysis must describe related projects, “enumerate the environmental effects of those projects” and “consider the interaction of multiple activities.” Or. Natural Res. Council, 492 F.3d at 1133.
The Basin EA provides detailed cumulative analysis of soil and watershed effects and incorporates a substantial cumulative analysis concerning fish and wildlife in the *1184area. A section of the EA is titled “Potential Cumulative Impacts” and summarizes cumulative effects of the Basin Project alongside both prior management and other planned projects in the vicinity. Sierra Forest’s argument that the EA fails “to disclose the cumulative impact that these projects have had and will have on old forest wildlife” or to provide “any explanation” of “cumulative effects” is largely conclusory and is belied by the record. The EA cumulative impact analysis covers a variety of factors that could contribute to an overall decline in old forest wildlife, including group logging, HFQLG pilot projects and barred owl range expansion.
The EA is supplemented by the extensive discussions of cumulative impact in the 2004 Framework SEIS, which is the cumulative assessment of planned management throughout the Sierra Nevada. Sierra Forest argues that it would be improper for the Basin Project to rely on the 2004 Framework SEIS. Although the 2004 Framework commits to “[djetailed cumulative effects analysis at the ... project level” concerning soil and watershed effects, this does not preclude reliance on “cumulative effects ... addressed pro-grammatically in the [2004 Framework] SEIS.” Between the EA and SEIS analysis, the Forest Service has conducted and disclosed a substantial assessment of cumulative impacts. Therefore, the Forest Service did not violate NEPA when approving the Basin Project.
V. Injunctive Relief on the NEPA Claim
Although we agree with the district court’s conclusions concerning the substantive application of NEPA to the 2004 Framework and the Basin Project, we part ways concerning the appropriate remedy for the NEPA violation that occurred when the Forest Service established the 2004 Framework. Before an award of permanent injunctive relief, a plaintiff must meet four well-established requirements:
(1) that it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction.
eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388, 391, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006). Even in NEPA cases, “[a]n injunction should issue only if the traditional four-factor test is satisfied”; no “thumb on the scales is warranted.” Monsanto Co. v. Geertson Seed Farms, — U.S. —, 130 S.Ct. 2743, 2757, 177 L.Ed.2d 461 (2010).
A. Jurisdiction
The first issue we face is the district court’s conclusion that it lacked “jurisdiction over Plaintiffs’ substantive claims against the programmatic 2004 Framework,” based on the legal premise that “[o]n a programmatic Framework basis ... [courts] are limited to providing procedural relief.” Sierra Forest V, 670 F.Supp.2d at 1110. This is plainly erroneous. As noted above, Sierra Forest and California have standing to assert a facial NEPA challenge to the 2004 Framework, and their claim is ripe. The APA requires that a reviewing court shall “hold unlawful and set aside agency action, findings, and conclusions found to be ... without observance of procedure required by law.” 5 U.S.C. § 706(2)0» (emphasis added). We have directed or upheld setting aside agency action pending NEPA compliance on numerous occasions. See, e.g., Klamath Siskiyou Wildlands Ctr. v. Boody, 468 F.3d 549, 562 (9th Cir.2006) (enjoining timber sales premised on a policy change that *1185violated NEPA). If courts could not stop the federal government from applying a substantive rule promulgated without adherence to required procedures, regardless of the equities, both NEPA and the APA would be toothless.3 Therefore, the district court abused its discretion by finding that it lacked jurisdiction to bar implementation of the 2004 Framework during a remand for analysis required by NEPA.
B. Equitable Analysis: Reliance on Government Experts
As an alternative ground for its limited remedial order, the district court engaged in a traditional equitable analysis and concluded that the appropriate remedy was to leave the 2004 Framework in place and to order the Forest Service “to prepare another supplemental EIS on the Framework, one that meets the range of alternatives and analytical consistency identified by the Ninth Circuit in its decision on the preliminary injunction portion of this case.” Sierra Forest V, 670 F.Supp.2d at 1113. Sierra Forest challenges this decision on numerous grounds.
When assessing the four prerequisites for a permanent injunction, the district court correctly noted that “the Forest Service is entitled to rely on the reasoned opinions of its experts.” Sierra Forest V, 670 F.Supp.2d at 1111. However, the court then deferred to those experts in its own equitable analysis. In so doing, the district court improperly conflated deference in the context of judicial review of an agency decision, see Lands Council, 537 F.3d at 993, with deference in consideration of the equities after a violation of law has been found. Although the federal government is undoubtedly permitted to follow its own experts when making a decision, see Marsh v. Or. Natural Res. Council, 490 U.S. 360, 385, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989), federal experts are not always entitled to deference outside of administrative action. See Seattle Audubon Soc’y v. Evans, 771 F.Supp. 1081, 1096 (W.D.Wash.), aff'd 952 F.2d 297 (9th Cir.1991) (“This is not the usual situation in which the court reviews an administrative decision and, in doing so, gives deference to agency expertise.”); see also Seattle Audubon Soc’y v. Evans, 952 F.2d 297, 302 (9th Cir.1991) (holding that district court’s remedy order was not an abuse of discretion).
Winter v. Natural Resources Defense Council is illustrative of the circumstances in which deference is appropriate when considering a broad equitable question. In Winter, the Supreme Court held that “lower courts failed properly to defer to senior Navy officers’ specific, predictive judgments about how the preliminary injunction would reduce the effectiveness of the Navy’s ... training exercises.” 129 S.Ct. at 379. It is reasonable that courts would defer to particular experts when the government has unique expertise, in fields such as national security or the internal functioning of the military. However, Winter applied no such deference concerning the possibility that sonar testing would irreparably harm whales. See id. at 383-84. Ecology is not a field within the *1186unique expertise of the federal government. See Lands Council, 537 F.3d at 1004-05 (applying balance of harms analysis without deference to agency views); see also Massachusetts v. EPA, 549 U.S. at 519-20, 127 S.Ct. 1438 (noting “special solicitude” given to states concerning interests in health and welfare). Deference to agency experts is particularly inappropriate when their conclusions rest on a foundation tainted by procedural error.
If the federal government’s experts were always entitled to deference concerning the equities of an injunction, substantive relief against federal government policies would be nearly unattainable, as government experts will likely attest that the public interest favors the federal government’s preferred policy, regardless of procedural failures. We hold that the district court abused its discretion by deferring to agency views concerning the equitable prerequisites for an injunction. We therefore vacate the district court’s narrow permanent injunction and remand for analysis of the requirements of a permanent injunction without deference to the Forest Service’s experts simply because of their relationship with the agency. Our interim injunction will remain in place until the district court has addressed these cases on remand and crafted its own injunctive order.
VI. NEPA Conclusion
We hold that the district court properly granted the Forest Service summary judgment on Sierra Forest’s and California’s NEPA claims. The Forest Service did not violate NEPA when promulgating the 2004 Framework or approving the Basin Project. We vacate, however, the district court’s permanent injunction and remand for analysis without unwarranted deference to Forest Service experts.
REINHARDT, Circuit Judge:VII. National Forest Management Act
A.
Sierra Forest challenges the United States Forest Service’s forest management plan for the Sierra Nevada (“the 2004 Framework” or “the Framework”) and the Forest Service’s decision to approve the Basin Group Selection Project (“the Project”), a timber harvesting project designed to implement the 2004 Framework. Sierra Forest contends that the Forest Service violated the National Forest Management Act (the NFMA) by approving the Project without complying with the population monitoring requirements for management indicator species that were set forth in the 2004 Framework. Sierra Forest also contends that the 2004 Framework itself violates the NFMA because it will not sufficiently maintain species viability.
First, the district court granted summary judgment for the Forest Service on the claim that the Project had been approved without required population trend data. It did so because it concluded that the 2007 Amendment to the forest plan removed the monitoring requirements contained in the 2004 Framework, and that the 2007 Amendment applied retroactively to the Basin Project. Sierra Nevada Forest Protection Campaign v. Rey, 573 F.Supp.2d 1316, 1335 (E.D.Cal.2008). I disagree. The 2007 Amendment cannot retroactively eliminate the 2004 Framework’s population monitoring requirement. A remand is therefore required for the district court to determine in the first instance whether, when it approved the Project, the Forest Service complied with the 2004 Framework’s population monitoring requirements. In making that determination, the district court must consider the 2004 Framework’s requirements as they were at the time the Project was approved, *1187and not as the Forest Service represented them to be following the 2007 Amendment.
Second, the district court granted summary judgment for the Forest Service on Sierra Forest’s claim that the 2004 Framework violates the NFMA because it will not maintain species viability. However, as discussed above, the district court did not determine whether the Project complied with the 2004 Framework as it existed at the time of the Project’s approval. Because on remand the district court must answer that question, Sierra Forest’s challenge to the 2004 Framework is not presently ripe for judicial consideration. Until the district court decides whether the Project complies with the 2004 Framework without the 2007 Amendment, it should not consider Sierra Forest’s facial challenge to that Framework.
B.
The Sierra Nevada is an 11.5 million acre area in the western United States that contains eleven national forests. It is managed by the United States Forest Service pursuant to the National Forest Management Act, which “specifies the procedural and substantive requirements under which the Forest Service is to manage National Forest System lands.” Lands Council v. McNair, 537 F.3d 981, 988 (9th Cir.2008) (en banc).
The NFMA directs the Forest Service to create a “forest management plan” to govern its activities in each national forest “unit,” including the Sierra Nevada. 16 U.S.C. § 1604(a). The current forest management plan for the Sierra Nevada is referred to by the parties as the 2004 Framework. The content and history of the 2004 Framework are discussed in detail in Judge Fisher’s opinion, including the Amendment in 2007, which the Forest Service purported to make retroactive.
After adopting a forest management plan such as the 2004 Framework, the Forest Service implements the plan through discrete projects. See Citizens for Better Forestry v. U.S. Dept. of Agriculture, 341 F.3d 961, 966 (9th Cir.2003). The Project here is a timber harvesting project located on a 38,893 acre area in the Plumas National Forest, which is one of the 11 forests within the Sierra Nevada. The Project will allow for the logging of about 800 “groups” of trees, each group about Vk acres in size, as well as the logging of a number of individually selected trees with diameters of up to 30 inches.
It is an aim of the Project to “contribut[e] to the economic stability of rural communities” by creating jobs in the saw timber industry. The Project also raises money for the federal government; the expected net income from the Project is $700,000. In addition, the Project is part of a long-term strategy to “eventually achiev[e] desired future conditions of an all-aged, multistory, fire resilient forest.” The Project will contribute to this strategy by reducing fuel loads in some areas. But the Project is “not ... designed to meet a purpose and need for hazardous fuel reduction per se” and “[t]he group selection units are too small and widely scattered to provide much strategic benefit to wildland fire suppression.”4 The Forest Service approved the Project in 2004 after concluding that it would have “[n]o significant adverse direct or indirect effects on the environment.”
C.
Sierra Forest’s first contention under the NFMA is that the Project does *1188not comply with the NFMA because it was approved in violation of the population monitoring requirements set forth in Appendix E of the 2004 Framework. Under the NFMA, “after a forest plan is developed, all subsequent agency action, including site-specific plans such as the [Project], must ... be consistent with the governing[forest management] plan.” Lands Council, 537 F.3d at 989; see also 16 U.S.C. § 1604(i). The Project was approved in 2004. At that time, the governing forest management plan was the 2004 Framework, the same forest management plan that, with one significant amendment, addressed below, remains in effect today.
The Forest Service contends that Sierra Forest’s claim was rendered moot by the 2007 Amendment to the 2004 Framework (“the 2007 Amendment”). It asserts that the 2007 Amendment eliminated the obligation to comply with the monitoring requirements in Appendix E of the 2004 Framework. It further asserts that the 2007 Amendment’s elimination of Appendix E’s monitoring requirements applies retroactively so as to remove the obligation that the Forest Service had in 2004 to comply with those monitoring requirements. The Forest Service relies on the language in the 2007 Amendment that provides that, for projects approved prior to the effective date of the 2007 Amendment, including the Project, “obligations relating to MIS [Management Indicator Species] will have been met if the project record discloses impacts the project may have on MIS habitat or populations.... No other project-level analysis or disclosure requirements shall apply to these projects, including any particular requirements related to MIS set forth in Appendix E.”
The parties contest whether the Forest Service had the power to make a forest plan amendment that applies retroactively. Judge Fisher agrees with the district court that the Forest Service had that power, and, thus, that the 2007 Amendment eliminated the Forest Service’s pre-existing monitoring requirements for the Project. Therefore, he would hold that the district court was correct to grant summary judgment on Sierra Forest’s first NFMA claim. I disagree.
As Judge Fisher points out, the 2007 Amendment purports, by its own terms, to be retroactive. But the Amendment cannot apply retroactively without statutory authority in the NFMA. As we held in Friends of Southeasts’s Future v. Morrison, 153 F.3d 1059, 1070 (9th Cir.1998), the Forest Service only has the authority to “change the legal consequences of completed acts ... if Congress conveys such authority in an express statutory grant.” Our holding in Morrison was clear; we wrote that the “Forest Service contends that retroactive application of the area analysis provisions of the amended Plan (or, more accurately, the lack thereof) is authorized by NFMA. We disagree.” Id. (emphasis added).
In his opinion, Judge Fisher contends that even though Morrison addresses the very issue before us, whether the NFMA contains an express statutory grant that allows for retroactive application of plan amendments, the opinion does not control our decision in this case. He asserts that Morrison considered only whether one section of the NFMA, 16 U.S.C. § 1604(i), provides authority for the Forest Service to promulgate retroactive amendments. His opinion further states that a different provision of the NFMA, § 1604(f)(4), does provide the “necessary express statutory grant” to enable the Forest Service to promulgate a retroactive amendment. The Forest Service made this argument in its brief on appeal in this case, but it does not appear to have done so in Morrison. Having failed to convince us in Morrison that § 1604(i) provides the express statu*1189tory grant in the NFMA necessary for the adoption of retroactive amendments, the Forest Service now claims it has found a new express statutory grant in the same statute.
Although Morrison did not explicitly examine § 1604(f)(4), we did not limit our holding to whether § 1604(i) was retroactive; instead we held that the NFMA does not provide the Forest Service with retroactive amending authority. Moreover, it is irrelevant whether we examined the whole statute in Morrison. “One three-judge panel of this court cannot reconsider or overrule the decision of a prior panel.” United States v. Gay, 967 F.2d 322, 327 (9th Cir.1992). The only exception to this rule is when “the reasoning or theory of our prior circuit authority is clearly irreconcilable with the reasoning or theory of intervening higher authority.” Miller v. Gammie, 335 F.3d 889, 893 (9th Cir.2003) (en banc).5
Here, there is no intervening higher authority; indeed there is no intervening authority of any kind: the relevant portions of the NFMA are the same today as they were when we decided Morrison, and the Forest Service could have argued in that case that § 1604(f)(4) expressly granted retroactive amending power. Thus, we have no authority to overrule our previous decision, and are bound by Morrison. This rule makes sense because we cannot continually re-litigate issues that our court has already decided simply because a party puts forth a new argument about why we should rule differently. The rule makes even more sense when the same person or organization is a party in both cases, and nothing prevented him from raising the issue in the first case.
In United States v. Contreras (Contreras II), 593 F.3d 1135 (9th Cir.2010) (en banc), we went en banc to withdraw a three-judge panel opinion, United States v. Contreras (Contreras I), 581 F.3d 1163 (9th Cir.2009) that purported to overrule several previous panel opinions. The en banc court vacated the part of Contreras I dealing with the earlier opinions but adopted the substantive portions, thus reaching the same result as the panel. Because it was an en banc court it was free to, and did, overrule the earlier opinions itself. The Contreras I panel had sought to correct prior decisions that had erroneously failed to consider a 1993 Amendment to the Sentencing Guidelines and had relied instead on a line of cases that pre-dated the 1993 Amendment. Even though all parties agreed that the posU-1993 panels should have considered the 1993 Amendment, only the en banc court had the power to overrule those panel decisions. Id. at 1136. If it was improper for the Contreras I panel to overrule previous panel decisions, it is certainly improper here where the relevant text of the NFMA has not changed subsequent to Morrison. Moreover, even if the retroactivity question was not foreclosed by our decision in Morrison, § 1604(f)(4) does not provide an express statutory grant that allows for retroactive amendments. Judge Fisher acknowledges that prospectivity is the default rule in statutory interpretation. “[Cjlear congressional intent” is required in order to establish retroactive application. Landgraf v. USI Film Prods., 511 U.S. 244, 272-73, 280, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). It is well established that a clear statement is required in order to show this Congressional intent. See, e.g., INS v. St. Cyr, 533 U.S. 289, 317, 121 S.Ct. 2271, 150 L.Ed.2d 347 (2001). The Supreme Court has ex*1190plained the rationale behind the presumption of prospective application as follows:
Because it accords with widely held intuitions about how statutes ordinarily operate, a presumption against retroactivity will generally coincide with legislative and public expectations. Requiring clear intent assures that Congress itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits. Such a requirement allocates to Congress responsibility for fundamental policy judgments concerning the proper temporal reach of statutes, and has the additional virtue of giving legislators a predictable background rule against which to legislate.
Landgraf, 511 U.S. at 272-73, 114 S.Ct. 1483.
The relevant subsection in this case, 16 U.S.C. § 1604(f)(4), provides that:
Plans developed in accordance with this section shall be amended in any manner whatsoever after final adoption after public notice, and, if such amendment would result in a significant change in such plan, in accordance with the provisions of subsections (e) and (f) of this section and public involvement comparable to that required by subsection (d) of this section.
Judge Fisher concludes that the words “amended in any manner whatsoever” provide the authority for the Forest Service to make amendments that have a retroactive, effect. However, this sentence is far from a clear statement authorizing such amendments; in fact, I do not believe that Judge Fisher’s interpretation is even a plausible construction of that sentence. The plain meaning of the words demonstrates that the sentence is not referring to the effect of an amendment, but to how an amendment may be made — the manner by which a change may be made to the original plan.6
Context is critical for statutory interpretation: here, when examined in the context of the rest of the subsection, it is even clearer that “amended in any manner whatsoever” is not a clear statement about whether amendments can be given retroactive effect. The subsection deals with the procedures for amending a final plan. In fact, we have previously described the subsection’s requirements as procedural rather than substantive. Lands Council v. Martin, 529 F.3d 1219, 1227 (9th Cir.2008).
The subsection first establishes that insignificant amendments to a final plan may *1191be adopted through any procedure whatsoever.7 It goes on to state that significant amendments must be adopted through the procedures set forth in subsections (d)-(f). The only cases that have examined 16 U.S.C. § 1604(f)(4) considered whether an amendment resulted in a significant change in the plan such that the procedural requirements for amending the plans set forth in subsections (d), (e), and (f) apply. See, e.g., id. Under Judge Fisher’s interpretation, Congress wrote one set of words that was intended to address two critical but entirely unrelated matters 1) that a final plan can be amended to have a retroactive effect; and 2) that any significant changes must be adopted through the use of the principal procedures applicable to the adoption of the plan itself, including even preparation of the amendment by an interdisciplinary team in a publicly available document. See 16 U.S.C. § 1604(f)(1), (3). Although opaque statutes are far too common, Judge Fisher’s interpretation stretches the bounds of reason, and I “decline to take such a dim view of the legislative endeavor.” See In re Lorillard Tobacco Co., 370 F.3d 982, 985 (9th Cir.2004). At the absolute least, the provision, with its omission of any reference to retroactivity, does not by any means demonstrate “a clear Congressional intent” to authorize amendments that have a retroactive effect. See Landgraf, 511 U.S. at 272-73, 114 S.Ct. 1483.
Moreover, when engaging in statutory interpretation, we must examine the whole section in which the language is located. See U.S. v. van den Berg, 5 F.3d 439 (9th Cir.1993). Subsection (f) relates to procedures for developing the forest plans. Subsection (f) is entitled “Required Provisions.” See Florida Dept. of Revenue v. Piccadilly Cafeterias, Inc., 554 U.S. 33, 47, 128 S.Ct. 2326, 171 L.Ed.2d 203 (2008) (noting that “statutory titles and section headings are tools available for the resolution of a doubt about the meaning of a statute.”) (internal quotation marks omitted). Subsection (f)(1) requires the Plan to be embodied in one publicly available document; Subsection (f)(2) requires the Plan to be written and to include maps where necessary; Subsection (f)(3) requires that the Plan “be prepared by an interdisciplinary team;” Subsection (f)(5) requires the Plan to be amended at least every fifteen years following public participation and the processes laid out in (d) through (f). Subsection (f)(4), the subsection at issue here, requires that certain procedures be used if significant amendments are made; if the amendments are insignificant, then any procedure is appropriate. Nowhere in this subsection is there a statement, let alone a clear statement, about the retroactive effect of plan amendments.
Although “magic words” need not be invoked to establish a retroactive application, see Alaska v. EEOC, 564 F.3d 1062, 1066-67 (9th Cir.2009), the clear statement rule requires more than a vaguely plausible statement. Therefore, even were we able to revisit our holding in Morrison that the NFMA does not provide for the retroactive effect of amendments to forest plans, a fresh look at the statute would reinforce our earlier holding. Accordingly, I conclude that the 2007 Amendment does not retroactively free the Forest Service of its obligation to ensure that the Project complies with the 2004 Framework’s population monitoring requirements as they stood at the time the Project was approved.
*1192The Forest Service also contends that, even if the 2007 Amendment did not retroactively eliminate the Forest Service’s obligation to comply with the 2004 Framework’s population monitoring requirements in approving the Project, the 2007 Amendment rendered Sierra Forest’s claim moot by eliminating any effective relief for its failure to do so. Northwest Environmental Defense Center v. Gordon, 849 F.2d 1241,1244 (9th Cir.1988). The defendant’s “burden of demonstrating mootness is a heavy one.” Northwest Envm’t Def Ctr. v. Gordon, 849 F.2d 1241, 1244 (9th Cir.1988). Moreover, “[t]he basic question in determining mootness is whether there is a present controversy as to which effective relief can be granted.” Therefore, “in deciding a mootness issue, ‘the question is not whether the precise relief sought at the time the application for an injunction was filed is still available. The question is whether there can be any effective relief.’ ” Id. (quoting Garcia v. Lawn, 805 F.2d 1400, 1403 (9th Cir.1986)).
In this case, there is effective relief for the Forest Service’s alleged failure to comply with the NFMA in approving the Project. That relief would consist of vacating the approval of the Project. At that point, the Forest Service could once again conduct its process for approving the Basin Project. In conducting that approval process, the Forest Service would need to ensure that the Project complies with the then-applicable Forest Plan, which may be the 2007 Amendment, or perhaps a subsequent Amendment. Regardless, vacating the approval of the Project would provide effective relief, although perhaps not the precise relief sought by Sierra Forest. The need to go through a new approval process alone is the effective relief that prevents the Forest Service from meeting its heavy burden of demonstrating mootness. Accordingly, this matter is remanded to the district court so that it may now determine whether the Project complies with the 2004 Framework as it existed prior to the 2007 Amendment.8
D.
Sierra Forest’s second claim is that the 2004 Framework violates the NFMA because it will not maintain species viability. Sierra Forest contends that the 2004 Framework depends on an “adaptive management strategy” to maintain species viability, and that the adaptive management strategy is insufficient for that purpose.
The 1982 Rule requires the Forest Service to develop forest management plans that will “maintain viable populations of existing ... species in the planning area.” 36 C.F.R. § 219.19 (2000). The 1982 Rule allows the Service to designate certain species as “management indicator species,” which serve as bellwethers or “class representative^]” for related species. Inland Empire Public Lands Council v. U.S. Forest Service, 88 F.3d 754, 762 n. 11 (9th Cir.1996). The Forest Service must *1193monitor the population trends of these management indicator species in order to determine how the Forest Service’s “management activities” — such as timber sales and grazing allocations — are affecting species. Id. at § 219.19(a)(l)-(7).
The 2004 Framework’s adaptive management strategy is, in theory at least, quite straightforward: the Forest Service will monitor projects that implement the 2004 Framework, assess whether those projects are threatening species viability and, if so, modify the Framework. The specifics of the adaptive management strategy were originally laid out in Appendix E of the 2001 Framework, which was adopted by the 2004 Framework. We described the monitoring component of the strategy in our decision in Earth Island Institute v. United States Forest Service, 442 F.3d 1147, 1173-76 (9th Cir.2006). To satisfy the requirements of the 1982 Rule, the Forest Service committed in the 2004 Framework to a “management approach ... [that] will provide the fish and wildlife habitat and other ecological conditions necessary to maintain well-distributed viable populations of vertebrate species in the planning area, and maintain the diversity of plants and animals.”
Sierra Forest does not dispute the efficacy of adaptive management as a general matter. Rather, it contends that the adaptive management strategy in the 2004 Framework is too vague. It asserts that the Framework’s strategy will not maintain species viability because it does not include an effective monitoring component, or guidelines explaining “when or how the 2004 Framework will be altered if monitoring reveals that the plan is impacting old forest wildlife.” Therefore, Sierra Forest contends, the 2004 Framework violates the NFMA.
I do not, however, address the merits of Sierra Forest’s facial challenge to the 2004 Framework, because that claim is not ripe for adjudication. The “ripeness doctrine is drawn both from Article III limitations on judicial power and from prudential reasons for refusing to exercise jurisdiction.” Nat’l Park Hospitality Ass’n v. Dep’t of the Interior, 538 U.S. 803, 808, 123 S.Ct. 2026, 155 L.Ed.2d 1017 (2003) (internal quotation marks omitted). A basic rationale for that doctrine is to “prevent the courts, through avoidance of premature adjudication, from entangling themselves in abstract disagreements over administrative policies.” Abbott Laboratories v. Gardner, 387 U.S. 136, 148, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967), abrogated on other grounds by Califano v. Sanders, 430 U.S. 99, 97 S.Ct. 980, 51 L.Ed.2d 192 (1977).
In Ohio Forestry Ass’n v. Sierra Club, 523 U.S. 726, 734, 118 S.Ct. 1665, 140 L.Ed.2d 921 (1998), the Supreme Court held that an NFMA challenge to a forest management plan is not ripe for adjudication until the “Plan then matters, i.e., if the Plan plays a causal role with respect to the ... imminent harm from logging.” As Judge Fisher recognizes in his opinion, we may only consider the facial challenge to the extent that the “Basin Project threatens species viability precisely because of flaws in the 2004 Framework.”
In this case, the district court must first determine whether the Project complied with the 2004 Framework before a facial challenge to the 2004 Framework based on that Project is ripe for adjudication. As discussed above, Sierra Forest alleges that, in granting approval of the Project, the Forest Service did not comply with the population monitoring provisions of the 2004 Framework then in effect. If, as Sierra Forest alleges, the approval of the Project did not comply with the 2004 Framework, then there is no showing that there is a causal relationship between any alleged deficiencies in the Project and any alleged defect in the Framework itself. *1194Indeed, deficiencies in the Project may stem from a failure to comply with the Framework, rather than a failure of the Framework to comply with the NFMA. Because we do not know whether the Project’s approval complied with the Framework, the Framework’s alleged defects under the NFMA do not currently “matter;” they may play no causal role with respect to any imminent logging. The very speculation about the cause of the harm is the reason why this claim is not ripe.9
I recognize that the Sierra Club’s claims in Ohio Forestry were more abstract than Sierra Forest’s claims in this case, because in Ohio Forestry there was no Project at issue that the court could use to evaluate the 2004 Framework. Here, Sierra Forest contends that there is a Project “at issue,” but unless that Project’s approval complied with the 2004 Framework, it is not a project we can use to evaluate the legality of the Framework. To do so would run counter to our admonition in Wilderness Society v. Thomas, 188 F.3d 1130, 1133 (9th Cir.1999), that we must evaluate the harm when it is “tangible, rather than theoretical.”
Therefore, as discussed above, this matter is remanded to the district court to determine in the first instance whether the Forest Service complied with the 2004 Framework’s population monitoring requirements when it approved the Project. It is not appropriate to decide whether the 2004 Framework complied with the NFMA before the district court considers that threshold issue. Because the district court only evaluated the Project under the 2004 Framework with the 2007 Amendment, the question of whether the Project’s approval complied with the Framework as it existed at the time of the approval is properly one for the district court now to decide in the first instance.
The district court has already purported to decide Sierra Forest’s facial challenge to the 2004 Framework. However, it should not have done so because that claim was not ripe, and will not be at least until after the district court properly decides whether the Project’s approval complied with the 2004 Framework. The district court should therefore strike the part of its prior decision which evaluated the facial challenge. On remand, if it decides that the Project’s approval did not comply with the 2004 Framework then in effect, it should not reach Sierra Forest’s facial challenge. On the other hand, if the district court decides that the Project’s approval did comply with the then applicable 2004 Framework, the district court should then re-examine Sierra’s Forest challenge, which would then be ripe for review.
E.
The district court erred in granting summary judgment for the Forest Service on Sierra Forest’s first NFMA claim because it applied the 2007 Amendment retroactively. Moreover, Sierra Forest’s second NFMA claim, a facial challenge to the 2004 Framework, is not ripe for review until after the district court decides the first claim under the 2004 Framework without the 2007 Amendment. Accordingly, the district court’s decision on those issues is vacated and the case is remanded for further proceedings in light of the opinions on this appeal.
Judge NOONAN concurs in the result.AFFIRMED in part and VACATED and REMANDED in part. The parties shall bear their own costs.
. For future cases, however, the holding is not binding authority on NFMA because there is no common ground between Judges Reinhardt and Noonan. See Hayes v. Ayers, 632 F.3d 500, 519-20 (9th Cir.2011) (concluding that a fragmented en banc decision of this court had no binding authority because the determinative vote relied on a completely distinct rationale) (citing Marks v. United States, 430 U.S. 188, 193, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977)); FTC v. Whole Foods Market, Inc., 548 F.3d 1028, 1061 n. 8 (D.C.Cir.2008) (Kavanaugh, J., dissenting) (discussing the rarity of a decision with no precedential value under the Marks principle).
. The Forest Service also asserts in a footnote that “failure to identify and challenge projects that will injure it highlights that the State has not challenged a proper agency action under the APA.” This argument attacks the ripeness of California’s claim, rather than the imminence of the asserted injury. However, the Forest Service relies on cases denying substantive review of an LRMP prior to application in a logging project, rather than a procedural NEPA challenge. See, e.g., Ecology Ctr. v. Castaneda, 574 F.3d 652, 658 (9th Cir. 2009). The Supreme Court has specifically excepted NEPA claims from an application requirement because “NEPA, unlike the NFMA, simply guarantees a particular procedure, not a particular result.” Ohio Forestry Ass’n, 523 U.S. at 737, 118 S.Ct. 1665; see also id. (noting that "the claim can never get riper” than when “a failure to comply with the NEPA procedure” has taken place). Therefore California’s facial NEPA challenge to the 2004 Framework is ripe.
. None of the cases cited by the district court or the intervenors supports a contrary rule. See Northern Cheyenne Tribe v. Hodel, 851 F.2d 1152, 1157-58 (9th Cir.1988) (merely holding that a court is not “compelled to issue an injunction without a balancing of the equities”); see also Lujan v. Nat'l Wildlife Fed'n, 497 U.S. 871, 890, 110 S.Ct. 3177, 111 L.Ed.2d 695 (1990) (holding that plaintiffs challenged an evolving series of agency operations rather than a final agency action); Ohio Forestry Assn, 523 U.S. at 737, 118 S.Ct. 1665 (holding that although a facial challenge under NFMA is not ripe, a challenge under NEPA may be); Sierra Club v. Peterson, 228 F.3d at 561 (addressing a substantive NFMA challenge rather than a procedural NEPA challenge).
. That distinguishes this case from our recent opinion in League of Wilderness v. Allen, 615 F.3d 1122 (9th Cir.2010). In Allen, we noted that the commercial benefit of the project was "merely attendant to the primary goal of lessening” the risk from large scale fires. Id. at 1126 n. 1.
. We are not reviewing a previous panel decision in the same case and thus exceptions to the law of the case doctrine cannot apply. Cf. Gonzalez v. Arizona, 624 F.3d 1162, 1191 (9th Cir.2010); Jeffries v. Wood, 114 F.3d 1484, 1489 (9th Cir.1997) (en banc).
. Judge Fisher provides examples of previous uses of the phrase "retroactive manner,” but the relevant question is not whether the phrase "retroactive manner” can be used, it is whether one speaks of amending a statute or a plan in a retroactive manner, and thus whether "amend in any manner whatsoever” clearly authorizes amendments with retroactive effects. It seems clear that "amend in any manner whatsoever” refers to how the amendment may be made and not the effect that the amendment may have. Although Judge Fisher says his examples show that "actions can be taken” in a retroactive manner, all of his examples are of something applying or operating in a retroactive manner. That is not an issue that is in dispute. There is no doubt that Congress may provide that a statute or plan may apply retroactively if it makes a clear statement that the statute or plan will or may so apply. Judge Fisher provides no examples, however, of amendments being made in a retroactive manner, and more important, he fails to provide a single example of a case that uses the term "amended in a retroactive manner.” Moreover, a search of all of the federal reporters fails to turn up any such case. It is beyond dispute that laws can have retroactive effects and can apply retroactively, but the fact that a law can apply in a retroactive manner provides no support for the interpretation that "amend in any manner whatsoever” is a clear statement authorizing amendments that have a retroactive effect.
. In full, the relevant clause reads, “amended in any manner whatsoever after final adoption after public notice;” the reference to public notice further establishes that this clause and subsection refer to procedural requirements and do not clearly authorize amendments that have a retroactive effect.
. The district court found that the Forest Service had not performed "actual quantitative population monitoring" in approving the Project, but had instead relied on habitat analysis. Sierra Forest, 573 F.Supp.2d at 1334. It then questioned whether the habitat analysis that the Service had performed satisfied the requirements of the 2004 Framework. It also questioned whether the habitat analysis served as a reliable proxy for changes in population given the Service’s admitted lack of knowledge regarding the connection between habitat and population trends for many management indicator species. Id. at 1334. However, because the district court concluded that the 2007 Amendment had mooted Sierra Forest’s claim, it cut short its review of the sufficiency of the Forest Service's monitoring. The questions it raised with respect to the unamended 2004 Framework, and those it didn’t even reach, are serious indeed and require thorough examination upon remand.
. The plaintiffs have established standing; therefore, the imminent harm, were there any, would be to the plaintiffs and their interests.