Wilderness Watch, Inc. v. U.S. Fish & Wildlife Service

BYBEE, Circuit Judge,

dissenting:

This should not have been a hard case. For more than seventy years, “[t]he dominant wildlife and habitat management theme for the Kofa ... has been the preservation of the desert bighorn sheep species.” 1 For much of that period, the Fish and Wildlife Service (“Service”), together with the BLM and the Arizona Game and Fish Department (“AGFD”), has supplied water to bighorn herds in the Kofa. So it is no surprise that after observing a dramatic decline in the bighorn population in this region, the Service proposed a variety of non-mutually exclusive solutions, including redevelopment of the Yaqui and McPherson tanks — two existing but unreliable water sources considered to be critical to the survival of the bighorn in the Eastern Kofa.

Although “[tjhere is little question that improvements to the water supply likely will help the sheep recover,” Maj. Op. at 1039, and although “[t]he Service undoubtedly found that ... the development of the two water structures was necessary,” id. at 1037, and although the documents prepared by the Service “amply describe the reasons for the Service’s decision to construct these two particular water structures,” id. at 1038, the majority nevertheless holds that the agency acted arbitrarily and capriciously because it never explained whether there was a “need for additional water structures,” id., or even “whether *1041water structures were necessary at all.” Id. at 1037. In the process, the majority holds that the Service should have engaged in a formalized, side-by-side comparative analysis of the various factors affecting the bighorn’s decline, id. at 1038, even though we previously told the Service that the Wilderness Act “does not specify any particular form or content” for a finding that an action is “necessary.” High Sierra Hikers Ass’n. v. Blackwell, 390 F.3d 630, 647 (9th Cir.2004). In so holding, as I explain in Part I, the majority ignores our deferential standard of review under the APA and engrafts new procedural requirements onto the Wilderness Act.

Additionally, as I explain in Part II, the majority fundamentally misconstrues the appropriate remedy for cases where an agency provides insufficient reasons for its action. Rather than remanding this case to the agency for a fuller explanation of its reasoning — as both Supreme Court and circuit precedent require — the majority remands to the district court for an exercise in futility: determination of the appropriate remedy. Then, under the guise of “expressing] no opinion on whether the Service may be precluded, by waiver or otherwise, from objecting to th[e] form of relief’ requested by the Plaintiffs, the majority even suggests that dismantling the structures may be the only appropriate remedy in this case. Maj. Op. at 1040 n. 10.

I respectfully dissent.

I

There is nothing unusually difficult about this case. Like any other routine administrative law appeal, it presents a finding made by an agency — here, the Service’s finding that building additional water sources was necessary to meet minimum administrative requirements for the conservation of the bighorn in the Kofa— and a subsequent decision based upon this finding — here, the Service’s decision to redevelop the Yaqui and McPherson tanks. Our task is, simply, to determine whether the agency’s finding, and thus its subsequent decision, was “arbitrary” or “capricious,” 5 U.S.C. § 706(2)(A), in light of the administrative record as a whole.

We know the standard of review: We may not overturn the Service’s decision simply because we disagree with it, nor may we substitute our own judgment for that of the Service because we feel the Service’s decision was imprudent or unwise. See River Runners for Wilderness v. Martin, 593 F.3d 1064, 1070 (9th Cir.2010). Rather, we must simply determine whether “the [Service’s] decision is founded on a rational connection between the facts found and the choices made ... and whether [the Service] has committed a clear error of judgment.” Id. (citation and quotation marks omitted). “To do so, we look to the evidence the ... Service has provided to support its conclusions, along with other materials in the record, to ensure that the Service has not, for instance, relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or an explanation that is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” The Lands Council v. McNair, 537 F.3d 981, 993 (9th Cir.2008) (en banc) (quotation marks and alterations omitted). “Even when [the Service] explains its decision with less than ideal clarity, [we] will not upset the decision on that account if the [Service’s] path may reasonably be discerned.” Alaska Dept. of Envtl. Conservation v. EPA, 540 U.S. 461, 497, 124 S.Ct. 983, 157 L.Ed.2d 967 (2004) (quotation marks and citation omitted).

*1042Here, the Service has produced three primary documents to explain why redeveloping the Yaqui and McPherson water structures was necessary to meet minimum requirements for the conservation of the bighorn: (1) the Investigative Report and Recommendations for the Kofa Bighorn Sheep Herd (“Kofa Bighorn Investigative Report”), a 39-page report that found a need to supply bighorn sheep with water; (2) the Kofa National Wildlife Refuge Minimum Tool Analysis (“Kofa Minimum Tool Analysis”), a six-page document that concluded that without action, the bighorn population would continue to decline; and (3) the Kofa National Wildlife Refuge Minimum Requirements Analysis (“Kofa Minimum Requirements Analysis ”), a two-page decision document that found there were no other less intrusive actions that could be taken to reverse the decline of the bighorn.2 Together, these documents demonstrate that the Service considered all the important aspects of the problem and offered a reasonable explanation for the necessity of building additional water sources for the sheep. Accordingly, with a few citations to important sections of these various reports, the majority should uphold the Service’s decision in a rather unremarkable opinion. Instead, the majority holds the Service’s decision arbitrary and capricious not because of what the Service said in its reports, but because of what, allegedly, it did not say. Specifically, the majority claims the Service never stated that water structures were necessary and never explained why addressing factors other than water would be insufficient to reverse the bighorn’s decline. As I discuss below, the record reveals the Service thoroughly addressed the majority’s concerns.

A

First, the majority holds the Service acted arbitrarily and capriciously because the Service “never stated that the two water structures are necessary, either in the abstract or with respect to the statutory standard.” Maj. Op. at 1038. However, the Service did explicitly state “there[are no] other less intrusive actions,” other than redeveloping the two water tanks, that would help the sheep recover. See Kofa Minimum Requirements Analysis at 1. I give this statement its plain and ordinary meaning: “the two water structures are necessary, either in the abstract or with respect to the statutory standard.” Maj. Op. at 1038. To explain the Service’s statement, the majority conjures up a handy expedient: the Service’s statement is meaningless because, you see, the Service had already “assum[ed] that water structures are necessary and reasoned from that starting point.” Id. at 1038. Thus, according to the majority, the Service’s statement “likely carried forward [this] primary underlying assumption.” Id. As I explain below, a close reading of the Service’s reports reveals the Service never assumed water structures were necessary; rather, the Service arrived at this conclusion after carefully studying the bighorn’s decline.

*1043The Kofa Bighorn Investigative Report, conducted jointly by the Service and the AGFD, reveals both that bighorn sheep need reliable water sources to survive and that reliable water sources must be supplied to the bighorn. Specifically, the report found that “water distribution is a critical habitat variable for desert bighorn” and that the sheep are “dependent] on reliable water sources.” Kofa Bighorn Investigative Report at 8. The report found that in areas where the AGFD had maintained water for the sheep, the populations “remained fairly stable in comparison to the Kofa population.” Id. at 9. Yet despite the critical importance of reliable water sources, “[f]ew of the [existing] bighorn waters could be considered permanent, which means that they may go dry during the height of summer when water is needed most, or during drought years.” Id. Indeed, a water survey conducted during a drought year revealed that “very few of the known waterholes were not dry.” Id. This observation is significant first, because it shows sheep must be supplied with water, and second, because when “a water source dries up, bighorn may not move to new areas to find water.” Id. Thus, it does not matter much that the Kofa Wilderness “has a large number of water sources” because “their utility for sheep and their reliability varies greatly.” Id. In light of these observations, the report determined that

[i]n addition to a need for better monitoring and maintenance of existing waterholes, a better distribution of permanent water supplies is needed to provide water in all areas of suitable sheep habitat.... Some existing ephemeral waters can be improved or redeveloped to hold water longer, but in some areas new waters might need to be built.... New water developments can likely be constructed outside of wilderness, although construction in wilderness should remain an option if a wilderness location best meets wildlife management needs.

Id. Accordingly, the report recommended that the agency “[i]dentify existing waters that need to be redeveloped to improve water holding capacity and efficiency” to “[e]nsure year-round water availability for all bighorn sheep on the Kofa [National Wildlife Refuge].” Id. at 10.

The Kofa Bighorn Investigative Report was consistent with the Service’s comprehensive management plan, which likewise stressed that bighorn sheep must be supplied with water to survive. Specifically, the management plan explained that

[i]n the extremely dry Sonoran Desert ecosystem, water is the primary habitat component and variable. Over the years, wildlife managers have learned to manipulate the conservation of water in the desert for wildlife management purposes .... Kofa NWR has a long history of water hole development projects aimed at improving wildlife numbers and distribution throughout the refuge.

Kofa Management Plan at 31.3 The Kofa Management Plan further explained that “[e]ven with ... improvements some [water] tanks occasionally go dry.” Id. “To prevent large scale wildlife movement *1044away from these areas, or even worse, wildlife die offs, water is hauled to these drought susceptible tanks when needed.” Id. Moreover, the Service addressed the need for “artificial water catchments” to help preserve bighorn sheep in the Kofa. Id. at 36; see also id. at 38 (“The development of water sources for the bighorn sheep has been an important factor in species recovery since the 1950s.”); id. at 39 (“Strategic water development programs” for bighorn are a “key management tool[ ] in the restoration of moribund [bighorn] populations to historic carrying capacities.”) (internal quotation marks omitted). Thus, as obvious as the need for water is in the desert Southwest, the Service never “assum[ed] that water structures [were] necessary,” as the majority asserts. Maj. Op. at 1038. Rather, the Service arrived at this conclusion after carefully documenting the impact of water availability on the survival of the bighorn.

After explaining why — as a general proposition — water structures are necessary for the survival of the bighorn, the Service explained why the two particular water structures at issue here — the Yaqui and McPherson tanks — were necessary. Specifically, the Service noted that the Yaqui and McPherson tanks were “considered to be critical to bighorn sheep, based on them locations in sheep habitat and documentation of sheep use.” Kofa Bighorn Investigative Report at 9. The Service further explained that “permanent water sources for desert bighorn sheep are relatively clustered, leaving large areas of desert bighorn sheep habitat without permanent water” and that the planned Yaqui and McPherson tanks were “relatively distant” from other existing, permanent water sources.4 Kofa Minimum Tool Analysis at 1. Thus, the Yaqui and McPherson tanks “would fill gaps in the eastern portion of the Kofa Mountains and in the McPherson Pass area.” Id. Aside from highlighting the remote locations of the Yaqui and McPherson tanks, the Service also explained that redeveloping these two tanks “would provide reliable, year-round water for desert bighorn sheep at two locations ... where the existing Yaqui and McPherson Tanks have not always been reliable and have been very difficult, logistically, to maintain.” Id. The Service further stressed the unreliability of the existing water tanks when explaining why the project was a redevelopment: the project involved the installation of “a modern design requiring little water augmentation ... in the vicinity of an existing, nonfunctional or unreliable source.” Id. (emphasis added).

The Service then considered alternatives: (1) not to redevelop the Yaqui or McPherson tanks, (2) to redevelop the tanks using mechanized means, or (3) to redevelop the tanks through non-mechanized means. See Kofa Minimum Tool Analysis at 2-5. It rejected the first alternative — doing nothing — because “[wjithout improvements to the current distribution of permanent water, the desert bighorn sheep population could be expected to continue to decline,” while providing water would “assist[ ] lactating ewes in milk production” thereby “improving] lamb survival, which is critical to population recovery.” Id. at 3. The Service characterized the option of doing nothing as “failing to meet its responsibilities for wildlife and habitat management ... [and] its legal or policy requirements.” Id. at 4. *1045Of the two remaining alternatives, the Service recommended the second — mechanized redevelopment of the Yaqui and McPherson tanks — because using vehicles and mechanized tools would minimize the time its employees spent in the wilderness (three days versus two to three weeks for non-mechanized redevelopment).

Based on the documents discussed above, the Service ultimately determined that there were no other “less intrusive actions that can be taken or that should be tried first inside or outside wilderness” to reverse the decline of the bighorn and that the “activity [could not] be accomplished outside of [the] wilderness.” Kofa Minimum Requirements Analysis at 1. The Service then evaluated the potential adverse effects on the -wilderness character and concluded that the project should proceed because “[w]ildlife is a wilderness resource” and “[t]he proposed action is intended to restore and maintain wildlife and wildlife habitat.” Id. at 2. Significantly, the Service even explained how redeveloping the Yaqui and McPherson tanks would actually increase the wilderness character of the land: “Overall, the impact of maintenance activities in the Kofa Wilderness would decrease since very little water hauling after construction is expected to take place.” Kofa Categorical Exclusion at 5.5

The majority nevertheless argues that “nowhere in the record does the ... Service articulate why the action taken is necessary to meet the minimum requirements of the Act,” Maj. Op. at 1038-39 (alterations omitted), and that the Service did not address “the key question — whether water structures were necessary at all,” leaving that question “entirely unanswered and unexplained.” Id. at 1037. The record belies the majority’s claim. Looking at these documents as a whole, the Service articulates repeatedly that the bighorn sheep need water to survive and that the existing water sources are not adequate for sheep in the Kofa Mountains and the McPherson pass area. See, e.g., Kofa Bighorn Investigative Report at 8-9 (stating that “water distribution is a critical habitat variable for desert bighorn, ... [sheep are] dependent] on reliable water sources, ... a better distribution of permanent water supplies is needed to provide water in all areas of suitable sheep habitat, ... new waters might need to be built, [and] ... construction in wilderness should remain an option if a wilderness location best meets wildlife management needs”); Kofa Minimum Tools Analysis at 3 (stating that “[w]ithout improvements to the current distribution of permanent water, the desert bighorn sheep population could be expected to continue to decline”); Kofa Minimum Requirements Analysis at 1 (stating that “there [are no] other less intrusive actions,” other than redeveloping the two water tanks, that would help the sheep recover); Kofa Categorical Exclusion at 4 (stating that “[w]ithout improving the distribution and reliability of available *1046water, it is likely that the desert bighorn sheep will continue to decline”).

Admittedly, the Service could have been more didactic in its explanation, locating it in one document rather than several. But because the Wilderness Act “does not specify any particular form or content” for the Service’s finding of necessity, High Sierra Hikers, 390 F.3d at 647, the Service does not fail to make a finding of necessity simply because it does not explain its reasoning in a single document. The majority, however, accuses me of “cobbl[ing] up [my] own explanation [of why water structures were necessary] from disparate parts of the various reports ... [when] it is the Service’s explanation, or lack thereof, that we must review.” Maj. Op. at 1037 n. 8. With respect, the majority misses the point. The majority wants the Service to hold us by the hand and to slowly and exhaustively explain to us its actions at every junction along the way. But the Service’s reasoning and conclusions are readily discernable from the record, and so we are required to uphold the Service’s decision to redevelop the two water tanks. See Alaska Dept. of Envtl. Conservation v. EPA 540 U.S. 461, 497, 124 S.Ct. 983, 157 L.Ed.2d 967 (2004) (“Even when an agency explains its decision with less than ideal clarity, a reviewing court will not upset the decision on that account if the agency’s path may reasonably be discerned.”) (internal quotation marks omitted); see also River Runners, 593 F.3d at 1078 (holding that absence of a specific discussion of the amount of motorized traffic found necessary and appropriate does not require overturning the agency’s action if the agency’s consideration of the motorized traffic necessary “can reasonably be discerned” from the documents provided by the agency). We have no right to impose additional burdens on the agency because we think we can improve its handling of the matter. See Vermont Yankee Nuclear Power Corp. v. Natural Res. Def. Council, 435 U.S. 519, 546, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978) (“[I]f courts continually review agency proceedings to determine whether the agency employed procedures which were, in the court’s opinion, perfectly tailored to reach what the court perceives to be the ‘best’ or ‘correct’ result, judicial review would be totally unpredictable.”).

B

In addition to holding that the Service acted arbitrarily and capriciously because it never stated that water structures were necessary, the majority also holds the Service acted arbitrarily and capriciously because “nowhere in the record does the Service explain why [addressing other possible causes for the bighorn’s decline], alone or in combination, are insufficient to restore the population of the bighorn sheep.” Maj. Op. at 1038. But the Service did not consider the need for the Yaqui and McPherson tanks in isolation, as the majority suggests. Rather, these tanks — which service a remote area of the Kofa — are part of a comprehensive plan for promoting the survival of the bighorn. The Service also analyzed other possible causes of the bighorn’s decline and incorporated recommendations regarding these causes into its overall proposal. This comprehensive analysis discharges the Service’s obligation to consider other possible causes.

1

Nevertheless, the majority argues that redeveloping the water tanks was not “necessary” because “many factors other than access to water, either alone or in combination, might suffice to restore the population of bighorn sheep.” Maj. Op. at 1037. The majority points to four factors: translocation, human disturbance, mountain lion predation, and hunting. Id. at 31-32. But the majority is wrong. The *1047Service did consider these factors in the 2007 Kofa Bighorn Investigative Report. Specifically, the Service found that each factor contributed in some way to the decline in the bighorn sheep population from 813 to 390 between 2000 and 2006, but concluded that drought was the principal explanation for the decline.

For example, with regard to translocation, the Service observed that “[transplants of sheep from the Kofa were conducted nearly every year from 1979 through 1998 with no apparent decline in population.” Kofa Bighorn Investigative Report at 18. No sheep were transplanted from the Kofa in 2000, 2003, or 2004 “because of drought conditions.” Id. In 2005, thirty-one sheep were translocated, and the report acknowledged that “[w]hile not the ultimate cause of the population decline, the 2005 transplant may have contributed to the low numbers” in 2006. Id. In 2006, because the levels of sheep were “historically low,” the Service discontinued the transplant program until the bighorn sheep population increased to an average of 800. Id. at 18-19.

With respect to human disturbance, the investigative report acknowledged that a “study conducted from 1977 to 1984 documented strong reactions ... from Kofa sheep in response to 1 or 2 people” and that “[f]requent human disturbance of ewes may cause them to abandon the[ir] areas for less optimal habitat.” Id. at 17. To address these concerns, the report recommended conducting surveys to identify critical lambing areas and instituting seasonal closures if substantial recreational use is observed in these areas. Id. These measures would certainly have some positive effect on the bighorn’s recovery. But given that hikers have been present in the Kofa for decades and that their presence has not historically correlated with a decline in the bighorn’s population, the Service could reasonably conclude that addressing human disturbance alone would not solve the bighorn’s decline.

With respect to predators, particularly mountain lions, the investigative report found that “[i]t is unlikely that lion predation alone accounts for the decline observed.” Id. at 13. The report noted that “[tjhere are no verified records of mountain lions on the refuge between 1944 and 2001” but “a female lion and 2 kittens were spotted” in 2003 and “at least 5 lions [were documented] on the refuge in 2006.” Id. at 12. The report acknowledged that researchers did not know much about the threat mountain lions posed to the bighorn sheep population because “[l]ittle is known about the ... specific diet of mountain lions on Kofa NWR.” Id. The report found “evidence that some mountain lions in bighorn sheep habitat may kill multiple sheep within a year, some may kill only one sheep within a year, and some may kill no sheep at all. The key factor is that lion-sheep interactions must be studied.... ” Id. at 13. The report also stated that “predation is generally believed to rarely threaten population survival.” Id. at 11. The report proposed further study and recommended collaring lions and sheep to “[d]etermine [the] cause and extent of predation.” Id. at 14.

Notwithstanding the Service’s explanation of the need for further study, the majority speculates that addressing predation alone could be sufficient to restore the bighorn sheep population. See Maj. Op. at 1037-38. The majority places far too much stock in the explanatory power of the mountain lions. The report makes clear that the relatively recent appearance of mountain lions in the Kofa may contribute to, but cannot explain, the bighorn’s decline. The bighorn herds in the Kofa declined by more than 400 sheep — a more than fifty percent decline — between 2000 and 2006. No mountain lions were even spotted until 2003, and only five were ob*1048served in 2006 in the 600,000 acre refuge. Id. at 1029-30. Most importantly — and this is one of the reasons the report recommended further study and made radio collars for lions and sheep a priority — the report stated that mountain lions were believed to have killed only one bighorn sheep in 2005 and 2006. Id. The presence of mountain lions in the Kofa cannot account for a 400-sheep decline that began three years before any lions were even observed. I thus respectfully disagree with the majority’s claim that “[ending predation by mountain lions] could have met the goal of conserving bighorn sheep without having to construct additional structures within the wilderness area.” Maj. Op. at 1037. There is no support for such conclusion in the record.

Lastly, the investigative report addressed hunting. It observed that Arizona issues hunting permits in an auction to support bighorn sheep management and conservation and that anywhere from five to seventeen bighorn sheep permits are issued each year. Kofa Bighorn Investigative Report at 19. The majority comments that “the report recommends, without explanation, that hunting be continued in the Refuge.” Maj. Op. at 1038. The Service did explain this. The permits authorize hunting of mature rams only. The Service cited studies showing that “a healthy vigorous herd can be maintained by conservative harvest of mature rams.... ” Kofa Bighorn Investigative Report at 19 (emphasis added). The hunting of mature rams has little effect on the survival of the sheep.

After reviewing the Service’s various reports, it becomes apparent that the Service took a comprehensive look at the problem of declining bighorn populations and the various factors that may have contributed to that decline. Then, based on its scientific and technical expertise and drawing on decades of its own and Arizona’s experience in conserving the bighorn in the Kofa, the Service reasonably concluded that supplying water to the bighorn was necessary to ensure the survival of a critical herd. That other factors — predation, translocation, hunting, and human disturbance— might come into play does not discredit the Service’s determination that drought was the critical factor leading to the bighorn’s decline.6 The majority errs by failing to afford the Service’s scientific determination the special deference it is due. See, e.g., Arizona Cattle Growers’ Ass’n v. Salazar, 606 F.3d 1160, 1167 (9th Cir.2010) (“[W]hen an agency is acting within its expertise to make a scientific determination a reviewing court must generally be at its most deferential.”) (internal quotation marks omitted); Ctr. For Biological Diversity v. Kempthorne, 588 F.3d 701, 707 (9th Cir.2009) (“[High] deference is especially warranted when reviewing the agency’s technical analysis and judgments, based on an evaluation of complex scientific data within the agency’s technical expertise.”) (internal quotation marks and citation omitted).

*10492

The majority recognizes that the 2007 report was “a thorough, neutral, and sciem tifie assessment of the many factors that contribute to sheep mortality,” but faults it for not “assessing] the factors relative to each other.” Maj. Op. at 1037. Specifically, the majority argues that High Sierra Hikers requires the Service to conduct a side-by-side, comparative analysis of the factors contributing to the bighorn’s decline. Maj. Op. at 1038-39. But nothing in High Sierra Hikers, 16 U.S.C. § 1133(c), or the APA requires a comparison of this kind.

In High Sierra Hikers, we interpreted 16 U.S.C. § 1133(d)(5) in the context of the Forest Service’s decision to authorize commercial packers to take groups through a wilderness area.7 390 F.3d at 646. We held that the Forest Service’s determination that “packstock was needed to provide access to those people who would otherwise not be able to gain access for themselves or their gear” supported a finding of necessity, but that the Forest Service failed to meet the Act’s requirement of a “specialized” finding of necessity because the Service could authorize commercial services only “to the extent necessary.” Id. at 647 (emphasis in original). Because the Act required that commercial services could be authorized “to the extent necessary” to realize the recreational purposes of the area, the court found that the “Forest Service must show that the number of permits granted was no more than was necessary to achieve the goals of the Act” and that the Service must “articulate why the extent of such packstock services authorized by the permits is ‘necessary.’ ” Id. (emphasis in original). The court emphasized that the language “to the extent necessary” required a more specialized determination, one that balanced the purposes of the Wilderness Area with the antithetical purposes of commercial activity. Because it failed to conduct a specialized finding of necessity — that the licenses were issued “to the extent necessary” — the Service “[a]t best ... failed to balance the impact that the level of commercial activity was having on the wilderness character of the land [and a]t worst ... elevated recreational activity over the long-term preservation of the wilderness character of the land.” Id.

Unlike § 1133(d)(5), § 1133(c) does not require a specialized determination of necessity. Section 1133(c) prohibits structures “except as necessary to meet minimum requirements for the administration of the area for the purpose of this chapter.” Without a qualifier on the word “necessary,” I would not interpret the statute as requiring a specialized finding of absolute necessity, as the majority does. See Maj. Op. at 1039 (“Just because [the need for water] affects the sheep’s viability, the Service is not free to create structures addressing [the need for water] without regard to any other variables at play.... [T]he Service must, at the very least, explain why addressing [the need for water] is more important than addressing the other variables and must explain why addressing [the need for water] is even necessary at all.... ”).8

*1050I see no reason to require the Service to show that improving the bighorn’s access to water at the Yaqui and McPherson tanks is the only way to stop the decline of the bighorn sheep population. Indeed, courts have long rejected interpreting “necessary” to mean “absolutely necessary” or “indispensable;”9 as the D.C. Circuit has explained: “[Cjourts have frequently interpreted the word ‘necessary’ to mean less than absolutely essential, and have explicitly found that a measure may be ‘necessary’ even though acceptable alternatives have not been exhausted.” Natural Res. Def. Council, Inc. v. Thomas, 838 F.2d 1224, 1226 (D.C.Cir.1988) (citations omitted); see also Sierra Club v. Lyng, 663 F.Supp. 556, 560 (D.D.C.1987) (interpreting § 1133(d)(1) — which permits the agency to take such measures in the wilderness “as may be necessary in the control of ... insects” — to allow the agency “to use measures that fall short of full effectiveness so long as they are reasonably designed to restrain or limit the threatened [insect] infestation”). A determination of absolute necessity would be nearly impossible for scientists and researchers to make and would render meaningless § 1133(c)’s exception to the structure prohibition.

Here, the Service took a tried-and-true approach: it looked at the potential impacts on the bighorn population and proposed a comprehensive solution. Rather *1051than trying to isolate a one-size-fits-all solution, the Service moved forward on all fronts. It addressed predators, hunting, translocation, human disturbance, and water structures. The Service recognized that some of its proposals called for immediate action, while others required further study. Thus, for example, the Service did not recommend killing mountain lions, but studying them; it recommended seeking funding for GPS collars, but noted that such collars only “last 1-2 years, which would require frequent recollaring for long term monitoring efforts.” Kofa Bighorn Investigative Report at 14. By contrast, as any denizen of the Southwest knows, obtaining water requires short-term planning, because finding adequate water in the desert is both an immediate and a constant concern. It is thus far from clear that the Service would have had a good basis for comparing the efficacy of mountain lion control with rebuilding the Yaqui and McPherson tanks, and nothing in the Wilderness Act suggests the Service must ignore immediate needs to compare them with long-term proposals. See Lyng, 663 F.Supp. at 560 (“The degree of efficacy of various ... methods [of controlling insect infestation] is not to be debated between various scientists and resolved before this Court. The [agency’s] judgment that the control measures authorized are reasonably efficacious is entitled to respect under[the APA].”).

Although the Sex*vice did not do a side-by-side comparative analysis of the various causes of the bighorn’s decline, nothing in the Wilderness Act, the Service’s interpretation of the Act, the APA, or our decisions obligates the Service to conduct its review in that way. In fact, we have eschewed imposing such form on the Service. See High Sierra Hikers, 390 F.3d at 646-47 (“The Wilderness Act is framed in general terms and does not specify any particular form or content for [the Service’s finding of necessity.]”). I would thus decline to hold the Service to a particular form or content to determine necessity. As long as the explanations the Sexvice offered to support its finding of necessity are not implausible, do not omit an important aspect of the problem, and do not run counter to the evidence presented, I would uphold the Service’s finding under the APA. See The Lands Council, 537 F.3d at 993.

The majority may not think that the Yaqui and McPherson tanks are necessary, but whether these two water structures are necessary or not is at best debatable. The Plaintiffs may consider bighorn management contrary to the natural order and an affront to the idea of wilderness, but the Service has declared — and the majority today affirms — that preserving bighorn in the Kofa is consistent with the Service’s charge from Congress. See Maj. Op. at 1035-36. Where Congress has placed such decision-making power in the Service, the majority errs by failing to “defer to the informed discretion of the responsible federal agencfy].” Marsh v. Or. Natural Res. Council, 490 U.S. 360, 377, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989) (quotation marks omitted).

II

Ultimately, the majority concludes “the Service must provide enough evidence and explanation in the record to assure this court that it fully considered [alternative factors for the bighorn’s decline] and nevertheless rationally concluded that new water structures are, in fact, necessary.” Maj. Op. at 1040. I think the Service did just that. But even if I were to agree that the record in this case is “wholly inadequate,” I would not remand this case to the district court “with instructions to determine the appropriate remedy.” Id. When both Supreme Court and circuit precedent mandate a remand to the agen*1052cy for a more adequate explanation of the agency’s reasoning, an open-ended remand to the district court is but an exercise in futility. The only remedy the district court should consider is remand to the Service for further explanation. See Florida Power & Light Co. v. Lorion, 470 U.S. 729, 744, 105 S.Ct. 1598, 84 L.Ed.2d 643 (1985) (“[I]f the reviewing court simply cannot evaluate the challenged agency action on the basis of the record before it, the proper course, except in rare circumstances, is to remand to the agency for additional investigation or explanation.”); Humane Soc’y of the U.S. v. Locke, 626 F.3d 1040, 1053 (9th Cir.2010) (holding that when the agency’s “explanation is incomplete and inadequate to permit meaningful judicial review,” the proper remedy is a remand “to afford the agency the opportunity to ... articulate a reasoned explanation for its action”); see also Seavey v. Barnhart, 276 F.3d 1, 12 (1st Cir.2001) (“When an agency has not considered all relevant factors in taking action, or has provided insufficient explanation for its action, the reviewing court ordinarily should remand the case to the agency.”); Public Power Council v. Johnson, 674 F.2d 791, 794 (9th Cir.1982) (“When there is a need to supplement the record to explain agency action, the preferred procedure is to remand to the agency for its amplification.”) (citation omitted); 3 Charles H. Koch, Jr., Administrative Law and Practice § 8:32 (3d ed. 2010) (“Remand is particularly appropriate where the agency justification is inadequate ... [because] the agency should be given every opportunity to justify its decision.”).

For the reasons discussed above, the majority engrafts a new procedural requirement onto the Wilderness Act and the APA. Before building a permanent structure within wilderness, the Service must now make a thoroughly documented, formalized finding of necessity accompanied by the comparative, multi-factor, side-by-side analysis the majority dictates. But even more consequential is the majority’s holding that the Service’s failure to offer an explanation that satisfies the majority’s new criteria is fatal to its case: without creating this judicially-mandated record, the agency’s action is per se arbitrary and capricious. But

[t]here is a fine line between agency reasoning that is “so crippled as to be unlawful” and action that is potentially lawful but insufficiently or inappropriately explained. In the former circumstance, the court’s practice is to vacate the agency’s order, while in the later the court frequently remands for further explanation (including discussion of relevant factors and precedents) while withholding .judgment on the lawfulness of the agency’s proposed action.

Radio-Television News Directors Ass’n v. F.C.C., 184 F.3d 872, 888 (D.C.Cir.1999); see also Checkosky v. SEC, 23 F.3d 452, 463 (D.C.Cir.1994) (Silberman, J., concurring) (citing some of the “many instances where [the court has] remanded to an agency for a better explanation before finally deciding that the agency’s action was arbitrary and capricious ”) (emphasis added).

Nevertheless, the majority concludes the Service’s action was arbitrary and capricious because the Service failed to consider an important aspect of the problem— “namely, the failure to consider whether new water structures are necessary at all.” Maj. Op. at 1039. The majority’s holding rests on the following syllogism: (1) arbitrary and capricious agency actions must be set aside; (2) all inadequately explained agency actions are arbitrary and capricious; (3) therefore, inadequately explained agency actions must be set aside. I disagree with the second proposition of this syllogism; not all inadequately explained agency actions are arbitrary and *1053capricious. The majority’s approach sets aside agency actions

even when the reviewing court is unsure of the agency’s reasoning. This assertion, if accepted, would fundamentally alter the role of the judiciary vis-a-vis administrative agencies by forcing courts to decide that the agency’s action is either unlawful or lawful on the first pass, even when the judges are unsure as to the answer because they are not confident that they have discerned the agency’s full rationale. This assertion finds support in neither logic nor precedent.

Checkosky, 23 F.3d at 463 (Silberman J., concurring).

The majority deflects analysis of the appropriate remedy in this case by giving the district court a choice: on remand, the district court may either require the Service to dismantle the water structures or may remand the matter to the Service for a fuller explanation of the Service’s decision. The problem, of course, is that the majority errs by giving the district court this very choice. See 1 Richard J. Pierce Jr., Administrative Law Treatise § 8.5 (5th ed. 2010) (“If a court concludes that the explanation provided by the agency ... is too terse to allow the court to apply the arbitrary and capricious test, the only remedy authorized by [the Supreme Court] is a remand to the agency for further explanation.”) (emphasis added). Additionally, the majority even hints the Service may have waived its right to object to any remedy other than the dismantling of the structures. See Maj. Op. at 1040 n. 10. I offer just one comment in reply: I find it mind-boggling to think the Service can waive what courts must do.

* * * * * *

An inescapable theme runs through the Service’s various reports: water is, and historically has been, critical to conserving bighorn populations in the Kofa, and any proposal that does not take account of the herds’ need for additional water sources is unlikely to succeed. In my view, the record in this case clearly indicates the agency conducted a comprehensive analysis of the possible causes for the bighorn’s decline and reasonably decided to address the most pressing cause. Scattered over three reports, the Service’s finding of necessity is not as graceful or as elegant as it could have been. But, unlike judges at a pageant competition, our proper role is not to grade the Service on poise and style. Rather, as the Supreme Court has consistently held, even if the Service does not explain its decision with “ideal clarity,” we must uphold the decision “if the agency’s path may reasonably be discerned.” Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974). Here, the Service’s reasoning for redeveloping two critical but previously unreliable water sources can readily be discerned. Because the majority requires the Service to do more than it must do, I dissent.

. U.S. Fish & Wildlife Service, Kofa National Wildlife Refuge and Wilderness and New Water Mountains Wilderness Interagency Plan, Environmental Assessment, and Decision Record 35 (1996) [hereinafter Kofa Management Plan ].

. The Service has also discussed the bighorn in more comprehensive studies, such as the 1996 Kofa Management Plan, a more than 100-page management plan that considered, among other things, how the Service could "carry out a dual, but nonetheless interrelated, role of managing for bighorn sheep within the context of wilderness.” Kofa Management Plan at 37. See 16 U.S.C. § 1133(a) (The Wilderness Act is "within and supplemental to the purposes for which ... units of the ... natural wildlife refuge systems are established and administered.”) (emphasis added). Additionally, the Service described its proposed redevelopment of the two tanks in a five-page document, Categoncal Exclusion, Yaqui and McPherson Tanlcs Redevelopment Projects ("Kofa Categorical Exclusion ”).

. Water developments in the Refuge have been a longstanding conservation tool of the bighorn sheep in the Kofa area. As early as 1937, refuge managers were discussing water developments and predator control as methods to conserve the population. David Brown, Early History, in The Desert Bighorn Sheep in Arizona 7 (Raymond M. Lee, ed., 1993). Since the 1950s, the AGFD helped develop over 100 water sources for bighorn sheep, and in 1968 non-profit organizations began coordinating volunteers who offered to improve and maintain sheep water sources in cooperation with Arizona and the federal government. Kofa Management Plan at 38; see also William Werner, Water Development, in The Desert Bighorn Sheep in Arizona at 163.

. The location of these tanks was critical because the "optimum water distribution” for the bighorn is two miles. William Werner, Water Development, in The Desert Bighorn Sheep in Arizona at 164. "[G]ood distribution of water” helps disperse the sheep, reducing stress and disease transmission. Id.

. Once the Service concluded that additional water structures were necessary for the recovery of the bighorn, the Service took great care to preserve the wilderness character of the land. The Service’s respect for wilderness can be seen both from the locations chosen for the two water structures and from the manner in which the Service redeveloped the Yaqui and McPherson tanks. First, most of the redeveloped Yaqui tank is located just outside wilderness — indeed, only two or three "substantially unnoticeable” water diversion weirs are located on wilderness land — while the redeveloped McPherson tank is located inside wilderness, but within 0.1 miles of a designated road. Kofa Categorical Exclusion at 5. These strategic locations show that the Service sought to minimize intrusions on wilderness land. Second, the Service completed the work in three days, removed tracks left by workers and vehicles, covered visible portions of the water structures to blend them into the natural habitat, and replanted nearby plants. These efforts show the Service restored, as much as possible, the wilderness character of the land.

. For example, the Service found that "[s]evere drought conditions in 1996 and 2002 were likely causes for the bighorn population dips observed on the 1997 and 2003 aerial population surveys.” Kofa Bighorn Investigative Report at 7. This finding was unsurprising because "extreme ambient temperatures, reduced moisture content of forage, and mating activities necessitate additional water intake, and thus [create] a dependence on reliable water sources.” Id. at 8. The Service explained that, in addition to being "a critical habitat variable for desert bighorn, especially during summer months when temperatures can reach 120 [degrees],” water distribution affects other mortality factors for the bighorn sheep: "[d]isease in bighorn sheep is most prevalent when animals are stressed and during severe drought.” Id. at 8, 15. The report concluded that, because of water's critical role in preventing mortality, "a better distribution of permanent water supplies [was] needed.” Id. at 9.

. The Wilderness Act generally prohibits commercial enterprises in the wilderness area, but authorizes commercial services within wilderness areas “to the extent necessary for activities which are proper for realizing the recreational or other wilderness purposes of the areas.” 16 U.S.C. § 1133(c), (d)(5).

. The practical effect of the majority’s decision — requiring the Service to show that nothing but these additional water structures will help the sheep recover — subverts the majority's disclaimer that it “in no way hold[s] that the Service must make a finding of absolute necessity.” Maj. Op. at 1037 n. 8.

. Most obviously, the Supreme Court in McCulloch v. Maryland, 17 U.S. (14 Wheat.) 316, 4 L.Ed. 579 (1819), rejected interpreting the word "necessary” in the Necessary and Proper Clause to mean "always importing] an absolute physical necessity, so strong, that one thing to which another may be termed necessary, cannot exist without that other” and instead found that the word “import[ed] no more than that one thing is convenient, or useful, or essential to another. To employ the means necessary to an end, is generally understood as employing any means calculated to produce the end, and not as being confined to those single means, without which the end would be entirely unattainable.” Id. at 413—14; see also Stenbergv. Carhart, 530 U.S. 914, 937, 120 S.Ct. 2597, 147 L.Ed.2d 743 (2000) (explaining that the word "necessary” in Planned Parenthood of Southeastern Pennsylvania v. Casey's "phrase 'necessary, in appropriate medical judgment, for the preservation of the life or health of the mother' ... cannot refer to an absolute necessity or to absolute proof”).

Other courts have followed suit. In Cellular Telecomm. & Internet Ass'n v. F.C.C. 330 F.3d 502, 510 (D.C.Cir.2003), the court deferred to the FCC Commissioner's interpretation of "necessary” in 47 U.S.C. § 160(a), a statute allowing the FCC to forbear enforcing a regulation or a statutory provision if certain conditions are met. The court defined necessary as "that which is required to achieve a desired goal” and found that the term "does not foreclose a particular means to an end merely because other means are hypothetically available to achieve the desired end.” Id. Similarly, in Inbesa America, Inc. v. M/V Anglia, 134 F.3d 1035, 1036 (11th Cir.1998), the court stated that in order for a federal court to have admiralty jurisdiction over a contract, the elements of the contract must "pertain directly to and be necessary for commerce or navigation upon navigable waters.... The test we apply in deciding whether the subject matter of a contract is necessary to the operation, navigation, or management of a ship is a test of reasonableness, not of absolute necessity.” Id. (omission in original). In In re Mile Hi Metal Sys., Inc., 899 F.2d 887, 892-93 (10th Cir.1990), the court held that the term "necessary modifications” in the context of a collective bargaining agreement under 11 U.S.C. § 1113(b)(1)(A), "does not mean absolutely necessary” to the debtor’s reorganization but must be "more than potentially helpful.” And in Southland Royalty Co. v. United States, 217 Ct.Cl. 431, 582 F.2d 604, 605 (Cl.Ct.1978), the court interpreted "necessary” under 26 U.S.C. § 162(a), which allows a deduction for "all the ordinary and necessary expenses paid or incurred during the taxable year in carrying on any trade or business.” The court held, "[a]n expense satisfies the requirement of being 'necessary' within the statute if it is 'appropriate and helpful' 'to the development of the taxpayer’s business.' ” Id.