WILDEARTH GUARDIANS WESTREN WATERSHEDS PROJECT v. FISH AND WILDLIFE SERVICE (2025)

OPINION and ORDER

Plaintiffs WildEarth Guardians and Western Watersheds Project (together, “Plaintiffs”) are environmental organizations challenging the United States Fish and Wildlife Service's (the “Service”) issuance of five Cooperative Agriculture Agreements and Commercial Special Use Permits, and related actions,¹ authorizing commercial livestock grazing in the Red Rock Lakes National Wildlife Refuge under the National Environmental Policy Act (“NEPA”), the National Wildlife Refuge System Improvement Act of 1997 (the “Refuge Act”), and the Administrative Procedure Act (“APA”). (Doc. 1.) The Service now voluntarily moves to remand this matter without vacatur of the Commercial Special Use Permits. (Doc. 29.) Plaintiffs oppose. (Doc. 30.) As explained below, the Service's motion is granted. The matter is remanded without vacatur.

Background

I. Factual Background

a. United States National Wildlife Refuge System

The Refuge Act establishes the “National Wildlife Refuge System,” which “administer[s] a national network of lands and waters for the conservation, management, and where appropriate, restoration of the fish, wildlife, and plant resources and their habitats within the United States.” 16 U.S.C. § 668dd(a)(1), (2). The Act delegates the authority to manage refuges to the Service, id. § 668dd(a)(1), and requires that such management “ensure that the biological integrity, diversity, and environmental health of” refuges, id. § 668dd(a)(4)(B). Under the Act, the Service must “issue a final conservation plan” for each refuge and each plan must be revised “any time ․ conditions that affect the refuge ․ have changed significantly” and at least “every 15 years ․ as may be necessary.” Id. § 668dd(e)(1)(A)(iii)–(iv), (E). However, “[u]ses or activities consistent with th[e] Act may occur on any refuge ․ before existing plans are revised or new comprehensive conservation plans are issued.” Id. § 668dd(e)(1)(D).

b. Grazing on the Red Rocks Lakes National Wildlife Refuge

The Red Rock Lakes National Wildlife Refuge (the “Refuge”), located in southwestern Montana's Centennial Valley, FWS_01355, was established in 1935 “as a Refuge and breeding ground for wild birds and animals,” FWS_00982. The Refuge is comprised of tens of thousands of acres of protected federal lands including grasslands, sagebrush-steppe habitats, mid-elevation forested areas, and the largest wetland system in the Greater Yellowstone Ecosystem, which together provide important habitat for numerous birds, fish, and other wildlife, such as trumpeter swans, sage grouse, Arctic grayling, and grizzly bears. FWS_01329–30.

The Refuge “evolved with grazing by large native ungulates such as bison.” FWS_01360; see FWS_00998, 01000. Herds of cattle were first introduced to the area in 1876 when white settlers moved into the Centennial Valley. FWS_00982; see FWS_01349. Following the establishment of the Refuge, livestock grazing was discontinued by the Service due to overgrazing. FWS_01562. Grazing was then renewed to mitigate fire and rodent issues. FWS_01562. However, overgrazing again occurred, and the Service reduced the amount of grazing in 1968. FWS_01562. In 1994, the Service issued a Final Environmental Assessment (“1994 EA”) that reevaluated the management of habitat within the Refuge, including whether to allow grazing and, if so, under which specific parameters. FWS_00976–1072. The 1994 EA evaluated several grazing management alternatives, including, inter alia, Alternative A, which was a continuation of the then-existing rest-rotation system, FWS_00993–97, and Alternative E, which although closely related, “require[d] Refuge staff to plan, monitor, analyze, adjust, and replan for the following year rather than rely on a pre-determined rest-rotation schedule,” FWS_01002. In its 1994 Finding of No Significant Impact (“1994 FONSI”), the Service selected Alternative E as the preferred action. FWS_00977. This Alternative “became the basis for grazing permits issued from” then until present. (Doc. 29 at 13.)

As required by the Refuge Act, 16 U.S.C. § 668dd(e)(1)(A), (B), the Service issued the “Comprehensive Conservation Plan, Red Rocks Lakes National Wildlife Refuge, MT” in 2009 (“2009 Plan”), which determined that grazing “is a compatible use” on the Refuge, FWS_01481–82, and presented six “stipulations necessary to ensure [such] compatibility,” FWS_01482. The 2009 Plan also identified as a “key issue[ ]” the “inadequate monitoring of the current grazing program to determine its effectiveness as a management tool.” FWS_01331.

c. Authorization Process for Grazing

When authorizing commercial livestock grazing on refuge lands, the Service enters into cooperative agriculture agreements with private entities that are accompanied by the issuance of special use permits. See, e.g., FWS_00060–110, 00111–135, 00143–148. Such permits may be issued “only” after a determination that the use “contributes to the achievement of the national wildlife refuge purposes or the National Wildlife Refuge System mission.” 50 C.F.R. § 29.1.

a. 2023 Agreements and Permits

In January 2023, the Service issued five Cooperative Agriculture Agreements (“Agreements”), FWS_00060–110,² in conjunction with five Special Use Permits (“Permits”), FWS_00111–135, 00143–148, all of which are challenged here.

The Agreements are between the Service and the following five local ranching operations (“Ranching Operations”): Huntsman Ranch, FWS_00060–69, JbarL Ranch, FWS_00070–79, Lee Martinell Company, FWS_00080–90, Pete and Katie Mickelsen, FWS_00091–100, and Raffety Cattle Company, FWS_00101–110. Each Agreement's period of performance is June 1, 2023, through October 31, 2027. FWS_00064, 00073, 00083, 00093, 00104. Each Agreement Plan of Operations outlines the obligations of the Ranching Operation, the obligations of the Service, general terms and conditions, special conditions, and an Annual Work Plan, which sets a rotational grazing schedule for the Permit for the next five years and includes a map of the grazing area. FWS_00066, 00076, 00086, 00097, 00107. Consistently, each Permit is issued from the Service to the permitee Ranching Operation and contains general conditions and requirements, special conditions, and an Annual Work Plan. FWS_00111–135, 00143–148. Each Permit term is from June 1, 2023, to October 31, 2027, FWS_00111, 00117, 00123, 00130, 00143.

b. 2023 Compatibility Determination and 2023 EAS

Following the issuance of the Agreements and Permits, the Service issued a final Compatibility Determination (“2023 Compatibility Determination”), FWS_00241–51,³ and an Environmental Action Statement (“2023 EAS”), FWS_01073. Both the 2023 Compatibility Determination and the 2023 EAS relied on the 2009 Plan and the 1994 EA.

In the 2023 Compatibility Determination, the Service determined, inter alia, that the “benefits” of the grazing system “outweigh the[ ] negative impacts pursuant to findings in the Refuge's 1994 [EA],” and stated that the “monitoring program will provide Refuge management information needed to make informed wildlife and habitat management decisions pursuant to goals and objectives outlined in the [2009 Plan].” FWS_000245. In the 2023 EAS, the Service determined that the 1994 EA was “still ․ factual, relevant, and appropriate insofar as it relates to the grazing program” and that Alternative E “continues to be an appropriate management tool.” FWS_01073. The Service then concluded that livestock grazing on the Refuge did “not have significant environmental effects as determined by the [1994 EA] and [1994 FONSI]” and that the 2009 Plan “concurs with the EA's findings.” FWS_01073.

II. Procedural Background

On May 14, 2024, Plaintiffs challenged the Service's issuance of the Agreements and the Permits, the 2023 Compatibility Determination, and the 2023 EAS, and its failure to prepare an Environmental Impact Statement (“EIS”) under NEPA, the Refuge Act, and the APA. (Doc. 1.) Plaintiffs seek both vacatur of the Service's issuance and a Court order that the Service must complete a full EIS to correct deficiencies with the Service's permitting decision. (Id.) After Plaintiffs moved for summary judgment on November 8, 2024, (Doc. 18), the parties jointly moved to stay the matter pending settlement negotiations on three different occasions, (Docs. 21, 23, 25). Each of these motions was granted. (Docs. 22, 24, 26.) On March 25, 2025, the stay was lifted and a revised case management order was entered. (Doc. 28.) On April 18, 2025, the Service moved to voluntarily remand this matter without vacatur. (Doc. 29.) Plaintiffs opposed, (Doc. 30), and filed a reply in support of their summary judgment motion, (Doc. 31), despite the Service's failure to respond to that motion.

Legal Standard

“Voluntary remand is consistent with the principle that administrative agencies have inherent authority to reconsider their own decisions, since the power to decide in the first instance carries with it the power to reconsider.” Nat'l Res. Def. Council, Inc. v. U.S. Dep't of Interior, 275 F. Supp. 2d 1136, 1141 (C.D. Cal. 2002) (internal quotation and citation omitted); see Cal. Cmtys. Against Toxics v. U.S. Env't Prot. Agency, 688 F.3d 989, 992 (9th Cir. 2012) (per curiam) (“A federal agency may request remand in order to reconsider its initial action.”). When seeking voluntary remand, there are three “litigation positions an agency might take[:]” first, seeking voluntary “remand because of intervening events outside the agency's control[;]” second, seeking voluntary “remand [ ]without confessing error[ ] in order to reconsider its previous position[;]” and third, seeking voluntary “remand because it believes its original decision is incorrect on the merits and wishes to change the result.” Keltner v. United States, 148 Fed. Cl. 552, 560–61 (Fed. Cir. 2020) (quoting SKF USA Inc. v. United States, 254 F.3d 1022, 1027–29 (Fed. Cir. 2001)).⁴ Here, the Service's position falls into the second category. See, e.g., SFK USA Inc., 254 F.3d at 1029 (noting that an agency “might simply state that it had doubts about the correctness of its decision or that decision's relationship to the agency's other policies”). Thus, whether to grant a request for voluntary remand is within the broad discretion of the court. SKF USA Inc., 254 F.3d at 1029.

Such requests are ordinarily granted if an agency “profess[es] intention to reconsider, re-review, or modify the original agency decision that is the subject of the legal challenge[,]” In re Clean Water Act Rulemaking, 60 F.4th at 593 (quoting Limnia, Inc. v. U.S. Dep't of Energy, 857 F.3d 379, 387 (D.C. Cir. 2017)), and “the agency's concern is substantial and legitimate,” SKF USA Inc., 254 F.3d at 1029. Generally, such requests are only refused when “the agency's request is frivolous or made in bad faith.” Cal. Cmtys. Against Toxics, 688 F.3d at 992. However, this “broad discretion allows a court to deny a voluntary remand ․ if the risk of harm from indefinitely leaving an allegedly unlawful rule in place outweighs considerations of judicial and administrative efficiency.” In re Clean Water Act Rulemaking, 60 F.4th at 596. But, “that discretion does not include the power to vacate a rule without first holding it unlawful.” Id.; see also id. at 593–94 (explaining that because “federal courts do not have unlimited equitable authority[,] ․ permanent equitable remedies can be awarded against only illegal executive action”). If available, vacatur turns on “the seriousness of the [action's] deficiencies ․ and the disruptive consequences of an interim change that may itself be changed.” Allied-Signal, Inc. v. U.S. Nuclear Regul. Comm'n, 988 F.2d 146, 150–51 (D.C. Cir. 1993) (internal quotation marks omitted).

Analysis

The Service voluntarily moves to remand this matter without vacatur of the Permits recognizing that “it is appropriate to revisit the analysis underlying the issuance of grazing permits.” (Doc. 29 at 1.) Plaintiffs oppose. (Doc. 30 at 1.) As a threshold matter, Plaintiffs do not distinguish between the standards for voluntary remand and the standards for vacatur. Instead, Plaintiffs argue that remand without vacatur should be denied here because it would allow the Service to evade judicial review, Plaintiffs would be prejudiced, and both judicial efficiency and the need for finality would be undermined. As discussed above, voluntary remand is ordinarily only refused when the “request is frivolous or made in bad faith.” Cal. Cmtys. Against Toxics, 688 F.3d at 992. On the other hand, vacatur depends on how serious an agency's “deficiencies” and “the disruptive consequences of an interim change that may itself be changed.” Allied-Signal, Inc., 988 F.2d at 150–51. Applying these two distinct standards here, the Service's request for voluntary remand without vacatur is granted.

I. Voluntary Remand

Acknowledging Plaintiffs’ challenges here, the Service seeks to revisit the analysis underlying the issuance of the Permits to specifically address the substantial and legitimate concern of “whether grazing impacts were properly considered under both the Refuge Act and NEPA.” (Doc. 29 at 6.) The Service has also committed to modification or termination of the Permits consistent with that reconsideration. Accordingly, voluntary remand is appropriate here.

a. Concern and Reconsideration

The Service commits to implementation of “a multi-part planning effort on remand,” (Doc. 34-1 ¶ 4), to address the grazing-related concerns it has recognized since issuance of the Permits, (Doc. 29-1 ¶ 7.) These concerns regard the implementation of Alternative E and the Service's reliance on the 1994 EA and 2009 Plan in issuing the 2023 Compatibility Determination, the 2023 EAS, and the Permits. (Doc. 29-1 ¶ 7; see Doc. 34-1 ¶ 13.)

Service Regional Director Matthew Hogan ⁵ explains that the agency's remand effort begins with an “improved grazing monitoring protocol” (the “Protocol”). (Doc. 34-1 ¶ 5.) The Service initiated this new Protocol with a reevaluation of monitoring techniques and implemented it to gather data from “grazing unit vegetation points located all around the Refuge.” (Id. ¶¶ 6–7.) The Service will finalize the Protocol with any necessary improvements and then implement it in the summer of 2026 “to collect the first round of vegetation data under the finalized protocol” with its results provided to the public. (Id. ¶¶ 8–9.)

The Protocol “will provide essential data” to the Service's consideration of the impacts of livestock grazing in its new NEPA analysis. (Doc. 34-1 ¶ 5.) Indeed, the Service admits that since that 1994 EA was issued, it “has never been able to fully implement Alternative E as a result of staffing and resources shortages.” (Doc. 29 at 14.) These shortages, the Service continues, “have prevented [its] ability to monitor the impact of grazing on vegetation, wildlife, or nutrient level to the degree anticipated by Alternative E; nor has [the Service] been able to prepare site-specific prescriptive treatment plans prior to grazing.” (Doc. 29 at 14.) Thus, on remand, the Service will review Alternative E as actually administered since 1994 in consideration of the funding and staffing resource limitations. (Doc. 29-1 ¶¶ 6–9.) This reconsideration will be accomplished through undertaking an EA, and, following that process, the Service will either issue a notice of intent to prepare an EIS or issue a FONSI. (Doc. 29 at 19–20.)

As to the Refuge Act, the Service also plans to use data from the Protocol to draft a new Compatibility Determination to “determine whether grazing is compatible with Refuge purposes,” (Doc. 34-1 ¶ 16), and following completion of “one or more Habitat and Species Step-down Plan(s),” the Service will “amend the objectives and strategies” of the 2009 Plan “as necessary.” (Doc. 34-1 ¶ 13.)

More broadly, the Service seeks to revisit “cattle grazing in general on the Refuge.” (Doc. 34 at 4.) Indeed, the Service commits to “determin[ing] whether to modify or eliminate livestock grazing on the Refuge” all together. (Doc. 29 at 6.) Making this determination requires the new data and analyses described above. (Doc. 29-1 ¶ 7.)

The multi-part process planned on remand, shows the Service is not seeking “a second bite at the apple.” Am. Waterways Operators v. Wheeler, 427 F. Supp. 3d 95, 98 (D.D.C. 2019) (denying remand where the agency sought to revisit an “otherwise final decision based solely on its new-found desire” to “reconsider” certain factors that the agency had already considered in rendering its decision prior to litigation). To the contrary, the Service commits to more thorough monitoring of existing grazing through new methods, “improv[ing] the quality of available [grazing] data and information” through the collection and analysis of new data, (Doc. 29 at 20); (Doc. 34-1 ¶¶ 5–11), and performing both a new NEPA analysis and a new Compatibility Determination, as well as revising the 2009 Plan as necessary. Essentially, the Service seeks remand to implement new techniques and obtain new data that will allow it to address “potential deficiencies with the challenged decision.” (Doc. 29 at 18.)

Having explained its inability to implement Alternative E and its reliance on the 1994 EA and 2009 Plan, the Service professed its intention to reconsider its original action, In re Clean Water Act Rulemaking, 60 F.4th at 593, and articulated a substantial and legitimate concern, SKF USA Inc., 254 F.3d at 1029. Plaintiffs do not contest that this intention is professed nor do they dispute the Service's concern, but instead they argue that there is “no assurance” that the Service “will follow through on completing the processes it proposes.” (Doc. 30 at 6.) Plaintiffs insist that remand without vacatur is improper for numerous reasons that sit on the spectrum between the “relative extremes of bad faith and substantial and legitimate concerns.” (Doc. 30 at 9.) Plaintiffs’ argument is unpersuasive.

b. Bad Faith or Frivolity

As explained above, Plaintiffs do not distinguish between the standards for voluntary remand and vacatur. To the extent their arguments can be applied under the appropriate standards, it has been done so below. Ultimately, there is no basis to find that the Service's request is either made in bad faith or frivolous.

i. Evasion of Judicial Review

Relying on Limnia, Inc. v. United States Department of Energy, Plaintiffs argue that the request for voluntary remand here “is an improper tactic to evade judicial review” because the Service does not actually intend to reconsider its original decision subject to the instant legal challenge. (Doc. 30 at 10 (citing 857 F.3d 379, 387 (D.C. Cir. 2017)).) Specifically, Plaintiffs assert that “there is no assurance [the Service] will follow through on completing the processes it proposes on remand, and without this Court reaching the merits, Plaintiffs would face the risk that in future grazing permit approvals, [the Service] duplicates the same errors Plaintiffs seek to rectify in this case.” (Doc. 30 at 6.)

Limnia clarifies that although an agency does not need to “confess error or impropriety in order to obtain voluntary remand[,]” it “does at least need to profess intention to reconsider, re-review, or modify the original agency decision that is the subject of the legal challenge.” 857 F.3d at 387. For example, in Limnia, a company challenged the Department of Energy's rejections of its loan applications as unlawful under the APA. Id. at 381. The Department sought remand not to reconsider the applications, but to permit the company to submit new applications. Id. Remand was granted. Id. Such remand was then found to be in error because “the Department did not intend to revisit the original application decisions under review.” Id. at 388. That is not the case here.

The Permits at issue in this matter were challenged under NEPA ⁶ for the Service's reliance on the 1994 EA and failure to conduct an EIS, and under the Refuge Act for the Service's reliance on the 2023 Compatibility Determination and the 2009 Plan. (See Doc. 1.) The Service has professed an intention to perform both a new NEPA analysis and a new Compatibility Determination, and to revise the 2009 Plan as necessary. Further, the Service commits to “completing the NEPA and [Compatibility Determination] process prior to the expiration of the Permits[,]” (Doc. 29-1 ¶ 9), and because the Permits are revocable at any time, they “may well be modified or terminated,” (Doc. 34 at 12 (citing Doc. 34-1 ¶ 12).) Plaintiffs maintain that this is not enough. Plaintiffs insist that they cannot rely on the Service's commitments or timeline, and accordingly, may lose their opportunity to vindicate their rights to challenge the Permits and that such a “process could repeat itself.” (Doc. 30 at 12.) But that concern is contrary to the record in the case. The Service has professed its intention to revisit the analyses underlying the Permits by undertaking an EA and performing a new Compatibility Determination, and has committed to complete such analyses before the expiration of the Permits so that the Permits may be modified or terminated. Thus, this is not “ ‘remand’ in name only.” Limnia, Inc., 857 F.3d at 388.

Based on the current record, the Service's request for voluntary remand is not a bad faith or frivolous attempt to evade judicial review.

ii. Bolstering the Administrative Record

Plaintiffs assert that the Service seeks to bolster its rationale through voluntary remand so that its actions have a higher chance of withstanding subsequent judicial scrutiny. (Doc. 30 at 13 (citing Keltner, 148 Fed. Cl. at 564–67).) The Service responds that it is seeking to collect new data and conduct new analyses on remand, not simply “add documents to the administrative record without considering the underlying issues of grazing.” (Doc. 34 at 8.) That distinguishes this case from Keltner, where the government specifically sought to “further develop the record” on remand. 148 Fed. Cl. at 565. And, even more importantly, like Limnia, Keltner denied the government's request for voluntary remand in large part because the agency was not actually seeking to revisit its previous position. Id. (“[T]he government's primary goal on remand is to write a better decision for a predetermined outcome”). That is not the case here. As explained above, the Service has provided a multi-part process that consists of new data collection and analyses that it plans to implement upon remand, which will culminate in a new EA and Compatibility Determination, and which may result in the termination or modification of the Permits. Such a process demonstrates that the outcome is not predetermined. Accordingly, the Service's request for voluntary remand is not a bad faith or frivolous attempt to bolster the administrative record.

iii. Prejudice

Next, Plaintiffs argue that they will be prejudiced if voluntary remand is granted because the scope of the remand request made by the Service is narrower than the scope of the Plaintiffs’ challenges. Specifically, Plaintiffs assert that such remand “would not bear at all on [their] first claim for relief—that [the Service] violated the Refuge Act when it issued the 2023 Permits, which are not consistent with requirements from the 2009 [Plan],” (Doc. 30 at 17), and would “ignore[ ] numerous other issues” raised by Plaintiffs in support of their three other claims, (id. at 18). Although Plaintiffs are correct that remand may be denied when the scope of the request is inappropriate, see Keltner, 148 Fed. Cl. at 566–67, that is not the situation here.

As to the first claim, Plaintiffs insist that although the Service has admitted inconsistencies with the 2009 Plan requirements, “its remand motion does not propose doing anything about” it. (Doc. 30 at 17.) Plaintiffs are adamant that to properly address their challenge as articulated in their first claim, the Service “would need to modify the existing Permits and any future permits” to conform grazing with the 2009 Plan. (Id.) If Plaintiffs’ recounting of the Service's remand plans was on point, their argument would be persuasive. However, the Service commits to “[a]s necessary, ․ amend the objectives and strategies outline[d] in the 2009 [Plan].” (Doc. 34-1 ¶ 13.) Indeed, the Service explains that it will “complete one or more Habitat and Species Step-down Plan(s) ․ tiered from the 2009 [Plan], ․ and will include detailed evaluation and analysis of potential management alternatives for the Refuge's priority habitats and species.” (Id.) In so doing, the Service has professed an intention to complete at least one Habitat and Species Step-down Plan, and the specific evaluation and analysis in this Plan demonstrate the Service's substantial and legitimate concern as to the 2009 Plan. Further, the Service commits to performing a new Compatibility Determination “to determine if livestock grazing is compatible with the Refuge's purpose.” (Doc. 34 at 8; Doc. 29-1 ¶ 9.) Accordingly, the Service's request for voluntary remand is not so narrow as to prejudice Plaintiffs’ first claim.

As to the remaining claims, Plaintiffs argue that the scope of remand is too narrow because the Service “does not propose to address” the following key issues on remand: whether the Service “ignored its prior findings about the negative impacts of rest-rotation grazing, failed to consider impacts to arctic grayling and grizzly bear, misconstrued available science on sage-grouse impacts, and ignored climate change.” (Doc. 30 at 18.) That is simply not the case. The Service explains that “prior monitoring has not yielded results showing that grazing harmed any threatened or endangered species,” (Doc. 34 at 13), and the Protocol will “build on other recent studies that have found that grazing has not caused adverse impacts to endangered and threatened species or Arctic grayling populations found on the Refuge,” (Doc. 34-1 ¶ 11). Further, at least one Habitat and Species Step-down Plan will evaluate and analyze both grazing and non-grazing management strategies. (Doc. 34-1 ¶ 13.) Such new data and analyses will then be used by the Service to perform both a new NEPA analysis and a new Compatibility Determination, and to revise the 2009 Plan as necessary. And, as explained above, if the results show “grazing is failing to adequately manage wildlife habitat or harming endangered or threatened species,” the Permits will be modified or terminated. (Doc. 34-1 ¶ 12); FWS_00111–148. Accordingly, the Service's request for voluntary remand is not so narrow as to prejudice Plaintiffs as to their remaining claims.⁷

iv. Finality and Judicial Economy

Plaintiffs argue that remand would “delay ․ resolving with finality several of the issues” because many of their “claims would likely still require adjudication by this Court at a later date.” (Doc. 30 at 18.) Courts consider whether “the need for finality in th[e] matter clearly outweighs the government's justification for requesting a voluntary remand,” Keltner, 148 Fed. Cl. at 565, by balancing whether “the risk of harm from indefinitely leaving an allegedly unlawful ․ [agency action] in place outweighs considerations of judicial and administrative efficiency,” In re Clean Water Act Rulemaking, 60 F.4th at 596. The risk of harm here is evaluated in detail in regard to the disruptiveness of change in the vacatur analysis below. Ultimately, the risks posed by the Permits do not outweigh judicial and administrative efficiency given the Service's justification for its pending request: substantial and legitimate concern as to livestock grazing on the Refuge with articulated commitments to address these concerns on remand. Based on the current record, the Service's request for voluntary remand is not a bad faith or frivolous attempt to undermine finality or offend judicial or administrative efficiency.

c. Conclusion

Ultimately, because the Service has identified substantial and legitimate concerns, professed its intention to revisit the challenged action, and has not sought remand in bad faith or frivolously, remand is appropriate.

II. Vacatur

Because voluntary remand is appropriate, the next question is one of vacatur. The Service requests remand without vacatur, while Plaintiffs oppose remand of any kind and seek disposition on the merits. Balancing the seriousness of the deficiencies of the Service's action with the disruptiveness of interim change, Allied-Signal, Inc., 988 F.2d at 150–51, the Service has the better argument.

a. Seriousness of Deficiencies

The deficiencies identified by the Service itself are a lack of staff and resources to plan, monitor, and study grazing on the Refuge. (Doc. 29 at 14). To determine the seriousness of these deficiencies is precisely why the Service is requesting voluntary remand. Indeed, the Service's “analysis on remand will not assume the level of staffing and resources for range monitoring and management that was assumed in the 1994 EA, but rather will consider [Alternative E] as actually administered since 1994 with funding and staffing resource limitations.” (Id. at 19; Doc. 29-1 ¶ 6.) Through the Protocol, new data will be collected and new analyses conducted, which will be used to undertake a new EA, draft a new Compatibility Determination, and update the 2009 Plan as necessary. (Doc. 29-1 ¶ 9; Doc. 34-1 ¶¶ 5, 13–14, 16–17.) Thus, on this record, it cannot be said that it is “unlikely” the Permits, as they are or modified, comport with the statutes at issue. See Fox Television Stations, Inc. v. F.C.C., 280 F.3d 1027, 1049 (D.C. Cir. 2002) (explaining vacatur is inappropriate where a court “cannot say it is unlikely the [agency] will be able to justify a future decision to retain the [challenged action]”).

b. Disruptiveness of Interim Change

Plaintiffs argue that keeping the Permits in place would allow grazing to continue on the Refuge in a manner “incompatible with” and “threaten[ing] to the Refuge's” purposes and values, thereby unduly prejudicing them. (Doc. 30 at 6.)

In response, the Service argues that termination of the Permits pending further analysis would be harmful to the parties, non-parties, and the Refuge itself. The Service insists that the “sudden removal” of grazing would likely cause more harm to the Refuge than the continuation of such grazing. (Doc. 29 at 23.) According to the Service, this harm would be both ecological harm to the Refuge and economic harm to the permittee Ranching Operations. The ecological harm would likely result because grazing has existed on the Refuge for over two-hundred years. Thus, “immediately eliminating grazing without looking at impacts to vegetation structure and bird nesting may have long term impacts on species diversity on the Refuge,” and would remove “one of only a very select few tools available to reduce wildfire risk.” (Id. at 22–23). Further, the permittee Ranching Operations have made “significant and irreversible financial decisions” in reliance on the Permits, (Doc. 34 at 15), and, if immediate termination occurred, they would face substantial economic loss, and either an abrupt need to sell cattle, (Doc. 29-1 ¶ 15), or a need to find alternative grazing locations, which is not likely possible, (Doc. 34-1 ¶ 18).

Plaintiffs submit the Declaration of John G. Carter ⁸ in response to these arguments, specifically “to provide [an] additional perspective regarding points made in the Hogan declaration.” (Doc. 30-3 at 6.) Essentially, Carter “believe[s]” that removal of livestock grazing from the Refuge would “likely ․ benefit the native species and habitats” there. (Doc. 30-3 at 14.)

To determine the appropriateness of vacatur, the disruptiveness of the interim change is crucial. See Allied-Signal, Inc., 988 F.2d at 150–51. Carter presents well-supported arguments ⁹ to assert that the “removal of livestock from the ․ [Refuge] would not result in any immediate or short-term ecological harm.” (Doc. 30-3 at 14.) However, it is undisputed that livestock grazing was introduced to the Refuge area in the late 1800s, FWS_00982; see FWS_01349, and that such grazing has been ongoing in some form for the past nearly one-hundred years, FWS_01562. Indeed, grazing at its current levels has been occurring since the 1970s, FWS_00993, and in its current form since 1994, see FWS_00976–1072. Thus, the status quo for the parties, permittees, and the Refuge is grazing in its current form, and termination of the Permits would be disruptive. Moreover, the new data the Service intends to gather on remand requires that the Permits remain in effect and grazing continue. (Doc. 29-1 ¶ 16.)

c. Conclusion

Ultimately, balancing the seriousness of the Service's deficiencies with the disruptiveness of Permit termination, vacatur is not appropriate here.

III. Plaintiffs’ Request for Summary Judgment

In response to the Service's voluntary remand motion, Plaintiffs additionally request that summary judgment be granted in their favor. Because remand without vacatur is appropriate here, the merits are not reached and this request is denied. See In re Clean Water Act Rulemaking, 60 F.4th at 593–94.

Conclusion

Based on the foregoing, IT IS ORDERED that the Service's motion for voluntary remand without vacatur, (Doc. 29), is GRANTED. The matter is REMANDED WITHOUT VACATUR to the agency to perform both a new NEPA analysis and a new Compatibility Determination, and to revise the 2009 Plan as necessary.

IT IS FURTHED ORDERED that all pending motions are DENIED as MOOT and all deadlines are VACATED.

IT IS FURTHER ORDERED that within twenty-one (21) days of this Order, the Service must file a proposed schedule for remand.

DATED this 15th day of December, 2025.

FOOTNOTES

1.   Following the issuance of these agreements and permits, the Service issued a final Compatibility Determination, FWS_00241–51, and an Environmental Action Statement, FWS_01073, and did not prepare an Environmental Impact Statement.

2.   The record contains the Plan of Operations for each Cooperative Agriculture Agreement, which outlines “the terms and conditions of the [Agreement].” E.g., FWS_00060–69.

3.   The Service issued a draft Compatibility Determination in May 2023, FWS_00149–58, accepted public comments, FWS_00159–237, and responded to comments in the final Compatibility Determination, FWS_00241–51. Despite being titled “draft” on its cover page, the final Compatibility Determination is in the record. Compare FWS_00241–51, with FWS_00149–58.

4.   While there are actually five litigation positions, which are often referred to as “SKF USA situations,” only three address voluntary requests for remand.

5.   His “responsibilities include overseeing [Refuge] management.” (Doc. 29-1 ¶ 1.)

6.   The 2023 EAS was also challenged under NEPA for reliance on the 1994 EA.

7.   Plaintiffs also argue that voluntary remand is unduly prejudicial because it follows summary judgment briefing. (Doc. 30 at 10.) However, the Service filed the pending motion for voluntary remand before the deadline to do so, (Doc. 28), and only after this motion was filed did the Plaintiffs file their reply in support of summary judgment, (Doc. 31).

8.   Carter sits on the Board of Directors for Plaintiff Western Watersheds Project, holds a Bachelor's degree in Mechanical Engineering from the Georgia Institute of Technology and a Ph.D. in Ecology from Utah State University, and has engaged in environmental consulting and research for government and private clients since the late 1980s. (Doc. 30-3 at 2, 16.)

9.   Carter opines: Bison grazing and livestock grazing are distinct in their effects on ecosystems, (Doc. 30-3 at 6–7); short-duration grazing, as outlined in the 2023 Compatibility Determination, has negative consequences on these ecosystems, particularly as compared to rest-rotation, (id. at 8–9); livestock grazing is distinct from the other herbivore grazing meaning “there would be no lack of herbivory in the Refuge ․ without livestock,” (id. at 9); the Service's grazing program is not “a net ecological benefit at the Refuge,” (id. 9–10); the positive effect livestock grazing has on biodiversity is nuanced, for example, nest cover for certain native birds is reduced by grazing, (id. at 10); and there is research that shows livestock grazing fosters invasive species, (id. at 11), and wildfires are not more likely if livestock grazing ceases, (id. at 12–14).