This appeal brings to mind the maxim that you can please all of the people some of the time, and some of the people all of the time, but you can’t please all of the people all of the time. At issue are the regulations for motorized water craft adopted by the United States Forest Service (“Forest Service”) for portions of the Snake River within the diverse and spec- *1173 taeular area known as the Hells Canyon National Recreation Area. Balancing the competing and often conflicting interests of motorized water craft users, including jetboaters, and non-motorized water craft users, such as rafters and kayakers, is no easy task. The legislative framework contemplates not only that such craft are legitimate recreational uses in Hells Canyon but that the area should be preserved and conserved for the public benefit. In 1998, after a lengthy environmental impact process and extensive public comment, the Forest Service implemented a recreation management plan that included a “non-motorized window” — a three-day period every other week throughout the primary season during which motorized water craft would be barred from part of the “wild” section of the river. The Hells Canyon Preservation Council (the “Council”), 2 representing the non-motorized craft users or “floaters,” and the Hells Canyon Alliance (the “Alliance”), representing primarily the motorized boaters, challenged the plan under a variety of statutes. We affirm the district court’s grant of summary judgment upholding the plan.

BACKGROUND

Hells Canyon is the deepest river canyon in North America; through it runs the Snake River, which divides Idaho and Oregon as it flows northward. In 1975, Congress established the Hells Canyon National Recreation Area (the “Hells Canyon NRA”) to preserve this area, designating portions of the Snake River as “wild” and “scenic” and designating certain adjacent areas as wilderness. Hells Canyon National Recreation Area Act (the “Hells Canyon Act” or “the Act”), Pub.L. 94-199, 89 Stat. 1117 (1975) (codified at 16 U.S.C. §§ 460gg-460gg-13).

Visitor use has soared since the Hells Canyon NRA was established, increasing 147% during the primary (summer) season from 1979-1991. Motorboat use during the primary season has grown at over 400%, from about 300 launches in 1979 to more than 1500 in 1994. Use of non-motorized water craft has also grown, from less than 100 float trips in 1973 to over 450 in 1992. Use restrictions for non-motorized water craft have been in place since 1977, but motorized use remained essentially unregulated until the Forest Service in 1998 implemented the Wild and Scenic Snake River Recreation Management Plan (“Recreation Management Plan”) prepared in 1994. 3

The Recreation Management Plan, developed in response to litigation challenging the agency’s failure to regulate motorized water craft, 4 has a lengthy and complex history. In 1993, after issuing a notice of intent to prepare such a plan and an environmental impact statement for the river corridor, the agency released a draft environmental impact statement (“EIS”) identifying eight outstandingly remarkable values (“values,” or “ORVs”) in the corridor — -scenic, recreational, geologic, fish, wildlife, cultural, botanic/vegetative, and ecological values — and presenting six alternatives for consideration. Most of these alternatives restricted motorized use levels; two also contemplated the elimination of motorized water craft from the wild part of the river. In response to comments on the draft EIS, Forest Supervisor R.M. Richmond asked the agency to develop a seventh alternative. ‘Alternative G,” denominated the *1174 “preferred alternative” in the final environmental impact statement (“FEIS”) issued in July 1994, included not only motorized use-level restrictions - but also a non-motorized window in part of the “wild” river for three-day periods from Monday-Wednesday every other week in July and August for a total of 24 motor-free days.

In October 1994, Richmond issued a Record of Decision (“ROD”) selecting Alternative G, with certain modifications we need not enumerate here. Implementation of the Recreation Management Plan was stayed pending the resolution of numerous administrative appeals that followed. On July 19, 1995, Deputy Regional Forester Richard Ferraro partially affirmed and partially reversed the ROD. He affirmed “the programmatic decision to provide a non-motorized window to achieve more primitive conditions in the wild river segment” but delayed implementation of the Recreation Management Plan pending a new assessment of commercial use addressing, among other issues, the “specifics of timing and duration of a non-motorized window.” Ferraro also directed the agency to conduct a new analysis of access to private lands. 5

In December 1995, the Forest Service gave notice of its intent to analyze the issues on remand in an environmental assessment (“EA”) regarding the economic effects of the Recreation Management Plan’s use limitations on individual commercial river permits (the “Outfitter EA”). One month later, in January 1996, the Council filed suit in the District of Oregon, claiming that the agency had unduly delayed implementation of the Recreation Management Plan and had failed to adequately regulate motorized river craft in the Hells Canyon. See Hells Canyon Preservation Council v. Williams, No. CV 96-68-RE (D. Or. 1996). In April 1996, Judge Redden denied the Council’s request for a preliminary injunction ordering the agency to implement the Recreation Management Plan for the 1996 summer season but indicated that: 1) he would retain jurisdiction over the case; 2) the agency’s failure to issue the EA promptly could prove the Council’s case of delay; and 3) the agency should anticipate implementing the plan for the 1997 summer season.

By April 1996, agency personnel began to have second thoughts about the wisdom of the window and consequently evaluated options, including its elimination. The Outfitter EA, released in June 1996, contained three alternatives: (1) a no-action alternative, which was windowless, and had no restrictions on motorized water craft; (2) an alternative mirroring Alternative G; and (3) and a new, preferred Alternative C that modified the window by limiting it to 21 days. Following public comment, agency personnel, apparently concerned by opposition to the window, actively researched ways to remove it but ultimately did not alter the options in the EA. 6 Richmond issued a Decision Notice (“DN”) in September 1996, selecting Alternative C and declaring a Finding of No Significant Impact, which obviated any need for a full-blown EIS.

In transmitting the numerous administrative appeals that resulted from this decision, Richmond wrote Ferraro a letter, dated November 25, 1996, in which he stated that he “could not eliminate the non-motorized period for the commercial outfitters without going through another NEPA process ... because you upheld the non-motorized period ... in your July *1175 1995 decision.” Explaining that the time necessary to complete such a process “would preclude implementation of the Plan during the 1997 primary use season,” Richmond concluded that he was “unable to resolve the non-motorized issue and also meet Judge Redden’s expectations for implementing the Plan [during the 1997 season].”

A month later, on December 23, 1996, Judge Redden dismissed the Williams litigation, noting that the administrative appeals were still ongoing. Referencing Richmond’s November 25 letter, he also expressed concern “about the chilling effect of my earlier rulings.” Another two months later, but still during the administrative appeals process, Ferraro permitted Richmond to review his analysis of the non-motorized window issue. Richmond responded with a letter dated March 17, 1997, and an accompanying document titled “Non-Motorized Period Review Rationale;” in both documents, he concluded that his review revealed no new or significant information establishing a need to eliminate the window and that the window should remain a component of the Recreation Management Plan.

Richmond’s decision was affirmed. Restrictions on motorized use levels began in the 1997 summer season, but implementation of the window itself was delayed by Richmond pending completion of the Private Lands EA. This prompted the Council to file suit again, Hells Canyon Preservation Council v. U.S. Forest Serv., No. CV 97-481-RE (D.Or.1997), this time for an order implementing the window during the 1997 season. The district court denied the Council’s request for a preliminary injunction and later dismissed the suit without prejudice to give the agency time to finish the Private Lands EA.

Prior to release of the Private Lands EA in March 1998, the new Forest Supervisor, Karyn Wood, announced that full implementation of the Recreation Management Plan would occur in the summer of 1998, leading the Alliance, which opposes implementation of the window, to file suit in the District of Idaho. The Council also sued, again in the District of Oregon, objecting to the Recreation Management Plan on several grounds. The cases were consolidated in the District of Oregon under Judge Redden, the same judge who handled the Council’s 1996 and 1997 lawsuits. Faced with motions for summary judgment from all parties, Judge Redden entered judgment in favor of the Forest Service, concluding that the “voluminous record supports [the] contention that it collected sufficient data and adequately supported its plan.” It is this judgment that the Alliance and the Council both appeal, on different grounds.

STATUTORY FRAMEWORK

The Hells Canyon Act established the Hells Canyon NRA with the following purposes:

Hells Canyon Act § 1(a), 16 U.S.C. § 460gg(a). Included in the Hells Canyon NRA was the Hells Canyon Wilderness, as well as 71 miles of the Snake River. Hells Canyon Act §§ l(a)-(b) & 2(a), 16 U.S.C. § 460gg(a), §§ 460gg(b) & 460gg-1(a). Of these 71 miles, Congress designated some 31.5 miles as “wild” and 36 miles as “scenic” under the Wild and Scenic Rivers Act 7 (“WSRA”), Pub.L. 90-542, 82 Stat. 906 (1968), 16 U.S.C. §§ 1271-87 ; Hells Can *1176 yon Act § 3(a) (amending the WSRA, 16 U.S.C. § 1274 , to include these portions of the Snake River).

The Act directs the Forest Service to promulgate “such rules and regulations as [it] deems necessary to accomplish the purposes of this Act,” including provisions “for the control of the use and number of motorized and nonmotorized river craft: Provided, That the use of such craft is hereby recognized as a valid use of the Snake River within the recreation area.” Hells Canyon Act § 10 & 10(d), 16 U.S.C. § 460gg-7 & 7(d). Except as otherwise provided in §§ 2-3, and subject to § 10, the Act provides for administration of the Hells Canyon NRA “in accordance with the laws, rules, and regulations applicable to the national forests for public outdoor recreation in a manner compatible with ... objectives” such as:

Hells Canyon Act § 7, 16 U.S.C. § 460gg-4.

The WSRA instituted a “national wild and scenic rivers system” in order to implement a national policy, 16 U.S.C. § 1271 , that “certain selected rivers ... which, with their immediate environments, possess outstandingly remarkable scenic, recreational, geologic, fish and wildlife, historic, cultural, or other similar values ... be preserved in free-flowing condition, and ... be protected for the benefit and enjoyment of future generations.” 16 U.S.C. § 1271 . To achieve this end, the WSRA requires that a river designated as wild and scenic be managed “in such manner as to protect and enhance the values which caused it to be included in said system without, insofar as is consistent therewith, limiting other uses that do not substantially interfere with public use and enjoyment of these values.” 16 U.S.C. § 1281 (a). “[P]rimary emphasis” is to be given to “protecting ... esthetic, scenic, historic, archaeological, and scientific features.” Id.

Under the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321 et seq., an EIS is required for all “major Federal actions significantly affecting the quality of the human environment,” id. § 4332(2)(C), to ensure that agencies possess and consider “detailed information concerning significant environmental impacts” and to guarantee “that the relevant information will be made available to the larger audience that may also play a role in both the decisionmaking process and the implementation of that decision.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349 , 109 S.Ct. 1835 , 104 L.Ed.2d 351 (1989). NEPA’s implementing regulations therefore require that agencies “insure the professional integrity, including the scientific integrity, of the discussions and analyses” in EISs, identifying any methodologies used and sources relied on. 40 C.F.R. § 1502.24 . Agencies must “[Vigorously explore and objectively evaluate all reasonable alternatives.” Id. § 1502.14.

STANDARD OF REVIEW

We review a grant of summary judgment de novo, see Margolis v. Ryan, 140 F.3d 850, 852 (9th Cir.1998), and may affirm on any ground supported by the record, see Sicor Ltd. v. Cetus Corp., 51 F.3d 848 , 860 n. 17 (9th Cir.1995). Review of agency action under the Hells Canyon Act, the WSRA, and NEPA is governed by the Administrative Procedure Act (“APA”), 5 U.S.C. § 701 et seq. Under the APA, a decision may be set aside only if it was “arbitrary, capricious, an abuse of dis *1177 cretion, or otherwise not in accordance with the law.” Morongo Band of Mission Indians v. Federal Aviation Admin., 161 F.3d 569, 573 (9th Cir.1998) (quoting 5 U.S.C. § 706 (2)(A)).

In determining whether an agency’s decision is arbitrary and capricious, we “must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment. This inquiry must be searching and careful, but the ultimate standard of review is a narrow one.” Morongo Band, 161 F.3d at 573 (citation omitted). Accordingly, we “may not substitute [our] judgment for that of the agency” and “must simply ensure that the agency has adequately considered and disclosed the environmental impact of its actions,” id., bearing in mind that NEPA exists to ensure a process, not particular substantive results, see id. at 575 .

In assessing the adequacy of an EIS, we employ a “rule of reason” test 8 to determine whether the EIS contains a “reasonably thorough discussion of the significant aspects of probable environmental consequences.” Neighbors of Cuddy Mountain v. U.S. Forest Serv., 137 F.3d 1372 , 1376 (9th Cir.1998). Under this standard, our task is to ensure that the Forest Service took a “hard look” at these consequences. Id. We also review the sufficiency of an agency’s methodology under a rule of reason. See Association of Public Customers v. Bonneville Power Admin., 126 F.3d 1158 , 1188 (9th Cir.1997).

DISCUSSION

Although they seek different outcomes from the Forest Service process, both the Council and the Alliance attack the agency’s decision-making process from several, sometimes overlapping angles.

A. The Council’s Appeal

Each of the Council’s challenges pertain to the ROD and the FEIS. In short, the Council argues that the Forest Service has not gone far enough in regulating the areas and levels of motorized water craft use.

1. Substantive statutory challenges

The Council argues that the Forest Service’s motorized use restrictions are incompatible with the WSRA and the Hells Canyon Act. According to the Council, the agency has violated the WSRA by allowing motorized use in areas and at levels that have degraded the Snake River’s outstandingly remarkable values, and it has violated the Hells Canyon Act by selecting restrictions incompatible with that Act’s mandate to preserve and conserve the scenic and biological resources of the Area. As analysis of the relevant provisions requires cross-referencing these interrelated statutes, we discuss the Council’s substantive statutory challenges together in this section.

As the Council points out, the WSRA requires that a river designated as wild and scenic be managed “in such manner as to protect and enhance the values which caused it to be included in said system.” 16 U.S.C. § 1281 (a). Lacking a statutory definition of “protect” or “enhance,” we give these terms their common meaning. The mandate to “protect and enhance” does not, however, lead inexorably to the conclusion that permitting motorized use on both the wild and scenic portions of the river violates the statute.

Significantly, the WSRA’s “protect and enhance” language does not stand alone. Congress instructed that designated rivers be managed “without, insofar as is consistent [with the values motivating designation of the river], limiting other uses that do not substantially interfere with public use and enjoyment of these values.” 16 U.S.C. § 1281 (a). Congress thus recognized that other uses could “interfere with public use and enjoyment” but that not all such uses were to be prohibited. Only *1178 those uses that, consistent with the protection and enhancement of values, “substantially interfere” with enjoyment and use of the values at issue — the values identified in the FEIS — are to be limited.

The Forest Service’s decisions with respect to what uses are inconsistent with protection and enhancement and “substantially interfere” with the river corridor’s values must be accorded substantial deference given the standard of review as well as the Act’s mandate that the agency “promulgate ... such rules and regulations as [it] deems necessary to accomplish the purposes of this Act ... including] ... provision for the control of the use and number of motorized river craft.” Hells Canyon Act § 10 & 10(d). The purposes of the Act parallel the WSRA mandate to “protect” and “enhance” the values for which a river was chosen; the Act directs the agency to administer the recreation area “in a manner compatible with,” inter alia, “conservation of scenic, wilderness, ... and other values,” “preservation ... of biologically unique [features],” and “protection and maintenance of ... habitat.” Id. § 7(2-4).

Review of the FEIS indicates that the Forest Service took a “hard look” at the environmental impacts of motorized water craft on the various values of the Snake River. The agency devoted 145 pages of the FEIS to exploring the possible environmental consequences of seven alternatives on each of the Snake’s eight outstandingly remarkable values. Within the recreation value alone, the agency analyzed, over some 70 pages, numerous “recreation opportunity spectrum” (“ROS”) settings — categories developed by agency researchers to allocate recreation uses 9 — and then also analyzed each option’s ability “to meet a great variety of visitor expectations by providing many different users with the opportunity to achieve their desired recreation experience.” In page after page, the agency specifically identified and analyzed the possible effects of motorized water craft on each value and ROS setting. Although one may disagree with its conclusions, we cannot conclude that the agency failed in its duty to take the requisite “hard look.”

To take just one example, consider the manner in which the Forest Service addressed the fisheries value. With respect to salmon, acknowledging that jetboats might pose risks to fall ehinook redds, the Forest Service conducted its own study (which turned out to be inconclusive) on the effect of jetboat wakes on salmon eggs, requested input from other agencies, and discussed several other studies (including the one cited by the Council). It ultimately concluded that although potential for jetboats to disturb a Chinook redd site was “moderate,” there was only a “low probability” of direct and indirect effects on the fisheries value.

Although the Council has identified several examples from the FEIS raising the specter of interference with the values for which the Snake River was included in the wild and scenic river system, it has not shown that the agency’s limitations on motorized use are arbitrary and capricious, or even that the extent to which the agency allows motorized use of the river in fact substantially interferes with the river’s outstandingly remarkable values. For example, although the agency acknowledges that motorized river craft are partly responsible for a “reduction in the unique waterway sound of the Scenic ORV,” the mere existence of some decline in scenic value does not establish that motorized use substantially interferes with this value, nor does it show that the agency’s chosen limitations in striking a balance between the recreation value — which expressly recognizes the legitimacy of motorized boating — and the scenic value are arbitrary and capricious or fail to protect and enhance the river’s value.

Nor do the agency’s motorized use restrictions violate the directive in § 7 of the Act to administer the Hells Canyon NRA “in a manner compatible with” conservation of various values, preservation of biologically unique features, and protection and maintenance of habitat (collectively, *1179 the “conservation objectives”). 10 In assessing this claim, we bear in mind that § 7 is subject both to § 3 of the Act, which directs the agency to administer the wild and scenic Snake River in accordance with the WSRA, and to § 10 of the Act, which expressly recognizes that use of motorized water craft is a valid use of the Snake. We also bear in mind that the Act established a recreation area, not a pure refuge or wilderness area. Recreation is not only one of the values for which the Snake River was included in the wild and scenic river system, but it is also specifically identified in § 1(a) of the Act as a value to be enhanced.

To the extent the Council intimates that jetboats should be barred at least on the “wild” river, this argument is foreclosed by the plain language of the Act. Hells Canyon Act § 3(a), which amends the WSRA to add the portions of the Snake River at issue and directs that the wild and scenic parts be administered in accordance with the WSRA, see id. § 3(b), recognizes the use of motorized water craft as a “valid use” of the river. Id. § 10(d). Congress also vested broad discretion in the agency to provide “for the control of the use and number of motorized ... river craft” on the Snake. Id. Accordingly, the fact that most rivers designated under the WSRA are managed with severe restrictions on jetboat use — if such use is allowed at all— is neither dispositive nor particularly instructive, given the clear congressional mandate with respect to motorized boating on the Snake River.

Once again, review of the FEIS reveals that the Forest Service’s restrictions are neither arbitrary nor capricious. 11 The agency’s thorough discussion, as noted above, of the environmental impact of motorized water craft on the various values and ROS settings for the Snake River shows that it took the requisite “hard look.” The examples the Council cites as evidence that the agency failed to conserve the Hells Canyon NRA — reduction in the waterway sounds of the scenic ORV, reduction of the recreation experience for wilderness users, and displacement of wildlife in the event habitat is disturbed by recreation use — do not reveal arbitrary and capricious decision-making. Indeed, the FEIS identified each of these effects as probable consequences that could not be avoided under any of the alternatives, including those that restricted motorized use more significantly in number, duration, and area. Notably, none of these effects was tied exclusively or even primarily to motorized water craft — the ROD specifically recognized that the trend away from a primitive or semi-primitive environment was associated with both motorized and non-motorized river craft, and the FEIS identified aircraft and the sounds generated by the presence of other recreationists as other causes of probable adverse effects. Under the circumstances here, we conclude that the Forest Service did not act arbitrarily and. capriciously in determining that its restrictions were compatible with the Act’s conservation objectives. 12

*1180 2. Compliance with management directives in Land Resources Management Plan

We also reject the claim that the Recreation Management Plan does not comply with management directives in the Land Resources Management Plan for the Wallowa-Whitman National Forest (“Forest Plan”). For the Snake River area, that plan states that the “primary emphasis is on maintaining the recreation experiences available at the time the area was established.” The Council contends that the Recreation Management Plan violates this directive because it allows human activities that degrade the primitive and semi-primitive settings that existed in 1975, when the river was designated wild and scenic.

The Council interprets the phrase “maintaining the recreation experiences available at the time the area was estab lished” (emphasis added) to refer to use levels in 1975, but this phrase is also susceptible to a different, common sense interpretation — namely, to maintaining the types of recreational experiences available in 1975, which included motorized boating. As such, we cannot impose the Council’s restrictive interpretation of this language on the Forest Service. Nothing in the Forest Plan or the statute requires any particular numeric level or ratio of motorized and non-motorized uses. Because the plan language is susceptible to more than one reasonable interpretation, we defer to the agency’s interpretation. See Auer v. Robbins, 519 U.S. 452, 461 , 117 S.Ct. 905 , 137 L.Ed.2d 79 (1997) (Secretary of Labor’s interpretation of his own regulations is controlling unless “plainly erroneous or inconsistent with the regulation”) (citing Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 359 , 109 S.Ct. 1835 , 104 L.Ed.2d 351 (1989)).

3. Exclusion of alternative analyzing jetboat use at pre-1980 levels

Finally, we also reject the claim that the Forest Service violated NEPA by failing to include a reasonable. range of alternatives. The Council attacks the agency’s reliance on 1980s use levels and contends that it should have considered setting motorized use at the 1970s levels existing when the Hells Canyon Act was passed.

The argument that the agency improperly relied on 1980s use levels to limit the range of alternatives in the FEIS misses the mark. In this regard, the Council quotes Richmond’s statement that he “asked the planning team to focus on the period between the mid to late 1980s. I believe the use levels occurring during this period were generally acceptable to most users as reflected in the results of the University of Idaho’s Visitor Profile and Recreation Use Study which analyzed the perceptions of visitors during the summer of 1988.” This comment, however, was targeted to Alternative G, an option developed for the FEIS as a result of public comment on the draft EIS. The Forest Service did not rely on the study in selecting the range of alternatives to be considered; rather, it relied on the study only to set use levels for the new Alternative G and carried forward to the FEIS each of the original alternatives from the draft EIS (including Alternative C, an option that established use levels lower than those of Alternative G).

The Council’s claim that the agency erred in excluding a 1970s use level alternative presents a closer question but is one that ultimately fails in light of the standard of review. Not only did the Forest Service consider a reasonable range of alternatives but inclusion of the low levels for the 1970s would have been unrealistic in light of the statutorily mandated objectives.

*1181 The Forest Service had a duty to “[s]tudy, develop, and describe appropriate alternatives to recommended courses of action,” 42 U.S.C. § 4332 (E), to “[u]se the NEPA process to identify and assess the reasonable alternatives to proposed actions that will avoid or minimize adverse effects of these actions upon the quality of the human environment,” 40 C.F.R. § 1500.2 (e), and to “[Vigorously explore and objectively evaluate all reasonable alternatives.” 40 C.F.R. § 1502.14 (a).

The Council argues that a 1970s use-level alternative was needed to provide a “yardstick to measure impacts to the [Outstanding Remarkable Values] and conservation objectives resulting from the high levels of jetboat use in the 1980s and early 1990s.” Given the broad range of alternatives considered by the agency, we cannot, however, say that it violated the rule of reason. See Northwest Env’l Defense Ctr. v. Bonneville Power Admin., 117 F.3d 1520 , 1538 (9th Cir.1997) (“We review an agency’s range of alternatives under a ‘rule of reason’ standard that ‘requires an agency to set forth only those alternatives necessary to permit a reasoned choice.’ ”) (citation omitted). The Forest Service’s seven alternatives, as set forth in the FEIS after extensive public comment, provided a range of alternatives sufficient to permit a reasoned choice. See Resources Ltd., 35 F.3d at 1307 (“[a]n agency’s consideration of alternatives is adequate ‘if it considers an appropriate range of alternatives, even if it does not consider every available alternative’ ”). 13

A 1970s alternative would have set levels too low to satisfy the agency’s reasonable goal of striking an appropriate balance between recreational and ecological values; as such, the Forest Service had no obligation to consider this alternative in the FEIS. Indeed, Alternative C, the environmentally preferred option, was rejected because even its use levels were unacceptably low; the FEIS states that it did “not provide the balance between protection of ... [Outstanding Remarkable Values] and the desired recreation experience” given “significant displacement of both motorized and non-motorized users.” Because the 1970s alternative would have set even lower use levels, the agency satisfied the rule of reason in declining to include this option in the FEIS. Having analyzed and rejected a higher use level (Alternative C), the Forest Service had no obligation to consider an alternative unlikely to be implemented and inconsistent with basic policy objectives for managing the area, which included a level of use acceptable to all river users. See Akiak Native Community v. United States Postal Service, 213 F.3d 1140, 1148 (9th Cir.2000) (“[T]he [EA] considers a reasonable range of alternatives given the objectives of the Project. The Postal Service seeks to improve the reliability and efficiency of the mail delivery service to remote Alaskan villages. The Postal Service was not required to consider alternatives that would not serve this reasonable purpose. Therefore, it was permissible for the [EA] to reject the use of alternative transportation modes such as trucks.... It was the inefficiencies of these traditional alternatives that gave rise to the need for the experimental hovercraft Project in the first place.”); Resources Ltd., 35 F.3d at 1307 (“[a]lterna- *1182 tives that are unlikely to be implemented need not be considered, nor must an agency consider alternatives which are infeasible, ineffective, or inconsistent with the basic policy objectives for the management of the area”) (internal quotation marks and citation omitted).

B. The Alliance’s Appeal

The Alliance, which opposes the non-motorized window, attacks the 1994 FEIS, the ROD and the Forest Service’s development of the 1996 Outfitter EA. We reject each of these challenges. 14

1. Challenges to the 1994 FEIS and ROD

The Alliance mounts substantive and procedural challenges to the FEIS and ROD. Its substantive claim focuses on the narrative findings of the FEIS, which it contends are unscientific and undocumented and thus do not justify selection of the window. Its procedural claim arises under NEPA and targets the agency’s alleged failure to disclose its methodology, hard data, expert opinion, or other basis for the window. Because the arguments underpinning these claims overlap substantially, we address both claims in this section.

We turn first to the Alliance’s substantive claim that the Forest Service’s analysis of the window was unscientific and inadequate, resulting in a narrative FEIS that lacks the data and analysis necessary to support selection of the window. The Alliance contends that the FEIS and the ROD reveal no rational connection between the facts found and the conclusion reached.

This is a harsh indictment that is not supported by the record. Although the agency might have better supported its preferred solution, its analysis is sufficient for reasoned decision-making. 15 The FEIS and ROD show that the agency was well aware of conflict between motorized and non-motorized users, and made a reasoned and reasonably informed decision to institute the window to reduce that conflict and allow diverse recreation opportunities.

Although the Alliance is correct that a court should not supply a “reasoned basis for the agency’s action that the agency itself has not given,” Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 , 103 S.Ct. 2856 , 77 L.Ed.2d 443 (1983), there is no need for us to do so here. The Forest Service provided a reasoned basis for its decision to select a non-motorized window — namely, the need to balance various statutory considerations, conflicts between user groups and the expressed preferences of some users for a non-motorized option. The ROD explains that the University of Idaho’s Visitor Profile and Recreation Use Study (the “Idaho study”) revealed “different perceptions of exactly what kind of experience meets [visitors’] personal needs” and that these perceptions “sometimes result in conflicts ... perceived as intrusions on an individual’s personal recreation experience.” It also references a poll showing that “while motorized recreation was among the top compatible activities [compatible with ‘best stewardship’ of the area], it was also the most frequently mentioned incompatible activity.” In addition, given the preference for a non-motorized period in comments received on the draft and final EIS, 16 the ROD concludes *1183 that the window appropriately provided “a reasonable non-motorized opportunity to interested private and commercial float-boaters” with “minimal effect on private powerboaters” that also “minimize[d] potentially adverse economic impacts on commercial powerboaters.” The ROD then addresses specifics of the window’s timing and duration, noting that it was scheduled for “the historically low powerboat use period of Monday through Wednesday.” 17

The FEIS also references user conflicts and preferences. The Abstract states that Alternative G was developed “in response to public comments and emphasizes a balance of motorized and non-motorized use ... [with] periods of exclusive non-motorized use in a portion of the wild river.” The description of Alternative G reiterates this point:

In discussing the recreation value in particular, the FEIS explains that a

Aside from justifying the need for the window on the basis of user conflict, the FEIS provides detailed discussion of each alternative, analyzing effects based upon each alternative’s “capability to protect and enhance the recreation [Outstanding Remarkable Value] by achieving the desired ROS settings” and each alternative’s “capability to meet a great variety of visitor expectations by providing many different users with the opportunity to achieve their desired recreation experience.” Included in the FEIS is a detailed description of each category and how it was measured.

The FEIS also repeatedly references the Idaho study. 18 It ties the “need” for action to the finding that “the increase in recreation use in the river corridor was negatively impacting visitors’ recreation experiences” 19 and concludes that power *1184 boat use levels must be managed in part due to findings that “35% of the recre-ationists surveyed felt the river was crowded. Conflicts and complaints between the diverse user groups are common.” Given this evidence of user conflict and the agency’s documented interest in resolving it, we conclude that the agency made its decision based on a rational connection between the facts found and the conclusion reached.

In so holding, we reject the Alliance’s claim that the agency’s analysis of the window was inadequate. In the context of this case, which heavily implicates subjective considerations, we cannot say that the narrative style of the FEIS was inappropriate. 20 Nor can we agree that the FEIS fails adequately to set forth its methodology, as required by 40 C.F.R. § 1502.24 . 21 In analyzing the recreation value — the value most relevant to the window analysis — the FEIS incorporates ROS settings; adds categories specifically relevant to the Snake River corridor; outlines the “desired future conditions” for the corridor; and specifies how it rates each option’s potential to meet the desired settings. As we see it, the Alliance’s true quarrel is with the sufficiency of the methodology used.

Although the Alliance’s attack on the scientific underpinnings of the FEIS raises reasonable questions, the agency’s methodology does not fail the “rule of reason.” Association of Public Customers, 126 F.3d at 1188. It is implicit throughout the FEIS that the Forest Service relied heavily on its own expertise both in developing the method of analysis outlined above and in conducting that analysis. The agency is entitled to rely on its own expertise. See generally Marsh, 490 U.S. at 378 , 109 S.Ct. 1851 (“an agency must have discretion to rely on the reasonable opinion of its own qualified experts” where specialists express conflicting views). The Forest Service’s analysis of the recreation value using ROS settings — a tool it developed for allocating recreation uses but that has not, as the Alliance points out, been validated specifically in the Snake River context — is a decision well within the realm of its expertise. In addition, the agency relied on other sources where such reliance was appropriate, as, for example, in setting forth the need and purpose for the Recreation Management Plan and the window itself. We are not convinced that the agency acted arbitrarily and capriciously in applying this tool to resolve the thorny use issues in this case.

Although the agency could have conducted empirical studies, it by and large took a different, and we believe reasonable, approach to analyzing the recreation Outstanding Remarkable Value. 22 Signifi *1185 cantly, this is not a case in which the agency’s analysis is simply missing; on the contrary, its analysis goes on for more than 70 pages. Even if the agency’s tools are “primitive,” as the Alliance claims, under the unique circumstances, the decision not to employ a different methodology or particular empirical studies does not suffice to show arbitrary or capricious action. See Laguna Greenbelt, Inc. v. United States Dep’t of Transp., 42 F.3d 517 , 526 (9th Cir.1994) (upholding EIS under “rule of reason” despite “weaknesses” in its analysis of growth-inducing impacts and explaining that “NEPA does not require us to decide whether an EIS is based on the best scientific methodology available”).

2. 1996 Outfitter EA

Finally, the Alliance charges that the Outfitter EA — conducted in 1996 as the agency had second thoughts about the window — “illegally excluded” the “Authorized Officer’s preferred course.” That course “would have eliminated the non-motorized window while imposing restrictions on numbers of private and commercial powerboaters.” The Alliance argues that the Outfitter EA thus failed to present a reasonable range of alternatives. The EA, as noted above, considered three alternatives: (1) a no-action alternative, which was windowless and had no restrictions on motorized water craft use; (2) an alternative mirroring Alternative G; and (3) the new preferred Alternative C that limited the window to a total of 21 days. This argument obscures a critical fact: Richmond, the Authorized Officer, ultimately concluded that a non-motorized window should remain part of the plan. The Alliance’s claim, however, fails on both the law and the facts.

First, at this stage of proceedings, the Forest Service had no legal obligation to include this option. Significantly, the concept of the window had already been approved by Ferraro in his July 1995 decision on the administrative appeals of the 1994 ROD and FEIS. Nowhere did Ferraro direct the Forest Service to reconsider the very concept of the window. Rather, in ordering that the “specifics of timing and duration of a non-motorized window ... be considered as part of the analysis of operational limitations for commercial uses,” he simply directed the agency to reevaluate the timing and duration of the window in conjunction with a new analysis of economic impacts to commercial outfitters.

Thus, the “range of alternatives” jurisprudence is not applicable in this case. Had the agency failed to include a reasonable range in the FEIS, it might have run afoul of the law. But this did not occur, for the agency included an appropriate range of alternatives in the FEIS. Absent significant new circumstances or information, 23 the Forest Service did not need to repeat this task once it reached the stage of conducting, on remand, the Outfitter EA, which was to focus on economic impacts to commercial outfitters and propose changes to the window only with regard to specifics of timing and duration. That is precisely what the Outfitter EA did.

Even assuming that the agency had a duty to include the specific alternative at issue in light of relevant new information, no such information surfaced during the EA process, including on administrative appeal. The Alliance identifies none, and Richmond himself ultimately concluded— after the November 25, 1996, letter that led Judge Redden to fear a chilling effect — that no new or significant information-had surfaced. See Letter of March 17, 1997 (review of the record and recent correspondence revealed “no significant information that establishes a compelling argument to eliminate the non-motorized period. Consequently there is no need for *1186 further analysis specific to the non-motorized period.”)- 24

Notably, given the opportunity to expand on the comments he made in his November 25 letter, Richmond endorsed the concept of the window in his March 17 letter and accompanying document. As he explained, having had more time to review the issue — time to “complete” his review— he not only concluded that no new or significant information justified revisiting inclusion of the window but affirmatively concluded that it should be part of the Recreation Management Plan. See Letter of March 17, 1997 (“The non-motorized period should remain as a component of the ... Plan.”); “Non-Motorized Period Review Rationale” (“I find the non-motorized window to be an appropriate mechanism for providing desired recreation experiences on the wild portion of the Snake River.”). Apparently to allay concerns raised by his November 25 letter, Richmond even stated that he “did not intend to imply that the non-motorized period should be eliminated even though I considered that option during the EA process.” The March 17 correspondence cuts the heart out of the Alliance’s argument that the EA omitted the very alternative that Richmond wanted to select.

CONCLUSION

For the foregoing reasons, the judgment of the district court is

AFFIRMED

. Several organizations join the Council in attacking the Forest Service’s plan. For ease of reference, we refer simply to the Council.

. In 1982, the Forest Service adopted a Land Resource Management Plan for the Wallowa-Whitman National Forest that included management objectives for the river corridor, but it did not adopt regulations specific to motorized water craft.

. See Oregon Natural Resources Council v. Lyng, 882 F.2d 1417, 1426 (9th Cir.1989) (duty to promulgate regulations to control the use and number of jetboats), amended on other grounds, 899 F.2d 1565 (9th Cir.1990); Hells Canyon Preservation Council v. Richmond, 841 F.Supp. 1039 (D.Or.1993) (Forest Service unreasonably delayed issuance of final regulations).

. This analysis was ultimately conducted separately from the commercial analysis in a document we refer to as the "Private Lands EA.”

. The agency concluded, allegedly as a result of legal advice regarding the pending Council lawsuit, that only options including the window were "viable given our position in court.” Despite the Alliance's assertion that the agency believed only windowed options were viable, there is no doubt that the Outfitter EA on which Richmond based his decision contained a windowless option, by virtue of the no-action alternative.

. The remaining few miles are part of the Hells Canyon NRA but do not fall under the WSRA.

. Review under the "rule of reason” test is essentially the same as review for an abuse of discretion. See Marsh v. Oregon Natural Resources Council, 490 U.S. 360 , 377 n. 23, 109 S.Ct. 1851 , 104 L.Ed.2d 377 (1989).

. Examples of such categories include access, sense of remoteness, visitor management, and provision of an appropriate level of social encounters.

. These conservation objectives resemble the river corridor's ORVs, except that they apply to the entire recreation area; lacking a statutory definition of terms such as '‘conservation” and "preservation,” we give them their ordinary meaning.

. The Council sets up the straw-man argument that motorized use is not a conservation objective nor a factor that may be favored at the expense of conservation objectives. Recognizing that any recreation use necessarily impacts conservation objectives, however, the question we must decide is whether the agency acted arbitrarily and capriciously in imposing its chosen limitations in an effort to implement the Act’s multiple objectives.

. The Wilderness Act, Pub.L. 88-577, 78 Stat. 890 (1964), does not enter into our analysis. Although the Hells Canyon NRA includes an area designated as the Hells Canyon Wilderness, Hells Canyon Act § 1(b), 16 U.S.C. § 460gg(b), which is to be administered in accordance with the Wilderness Act or the Hells Canyon Act, whichever is more restrictive, see Hells Canyon Act § 2(b), 16 U.S.C. § 460gg-1(b), the Hells Canyon Act provides for application of the Wilderness Act only in connection with the wilderness area, and the record indicates that the river corridor is adjacent to, not part of, the wilderness area. Accordingly, we do not address the Council’s contention that the Act’s § 2 man *1180 date to conserve wilderness is the more restrictive and therefore applicable provision as between the Act and the Wilderness Act.

. We also note in light of the competing objectives at issue here, that exclusion of a 1970s use-level alternative is not the type of egregious omission requiring reversal. Cf. Muckleshoot Indian Tribe v. U.S. Forest Serv., 177 F.3d 800, 813-14 (9th Cir.1999) (failure to consider forest swap involving modifications to the acreage involved; range of alternatives — a no-action alternative and two nearly identical action alternatives — was inadequate, especially given that agency failed to consider alternative more consistent with its basic policy objectives); City of Tenakee Springs v. Clough, 915 F.2d 1308, 1311-12 (9th Cir.1990) (failure to consider terminating, modifying, or suspending timber harvest contract contemplating environmentally destructive amount of timber despite subsequent legislation designed to prevent environmental damage to subsistence resources); Bob Marshall Alliance v. Hodel, 852 F.2d 1223 (9th Cir.1988) (failure adequately to consider abandoned no-leasing option in case where issuance of oil and gas leases could have significant impact on environment).

. We also reject the Alliance’s discovery challenge. We are not persuaded by the claim that the Alliance was unable to develop its arguments because the district court denied "essential” discovery.

. See Akiak Native Community, 213 F.3d at 1146 ("[T]he agency’s decision-making process is accorded a ‘presumption of regularity.' We consider only whether the Postal Service’s decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment. Mindful of the limited scope of our review, we conclude that the [EA, considered as a whole,] is sufficiently well-documented and explained.”) (citation omitted). "[S]atisfied that [the] agency’s discretion is truly informed,” we defer to that informed discretion. Greenpeace Action v. Franklin, 14 F.3d 1324, 1332 (9th Cir.1992).

.In discussing the non-motorized window in the ROD, Richmond explained that he instructed the agency “to include a period of *1183 motor-free time in the preferred alternative [G] in response to comments received on the [draft] EIS” and that comments on both the draft EIS and the FEIS showed that "a distinct group of respondents [existed] that favored some level of non-motorized opportunity.” In fact, that was the "predominant response from regional and national interests on the [draft] EIS.”

.Notably, on administrative appeal, a portion of the window analysis — specifics of timing and duration — was remanded for consideration in conjunction with the Outfitter EA, at which point extensive consideration was given to these issues.

. In discussing environmental consequences to the recreation ORV specifically, the FEIS cites to the Idaho study, in general terms, relying on it to conclude that:

. It bears noting that the Idaho study is referenced at the beginning of the agency’s *1184 analysis of the recreation value in discussing differing user expectations. Based on the many citations to the study throughout the FEIS — including on the first page setting forth the purpose and need for the Recreation Management Plan — it requires no leap to conclude that the agency relied (sometimes implicitly) on the study’s findings. Moreover, that the agency relied on the study primarily to show the existence of user conflicts and differing expectations — and thus the need for a non-motorized option' — is appropriate in light of the study’s limited scope.

.This is not to say that scientific methods of analysis cannot be applied to the window. Applied recreation research has long employed expert and technical analysis. Where we part ways with the Alliance is in our conclusion that the agency was entitled, in considering the window issue, to rely on the methodology it developed for allocating recreation uses.

. Our discussion here applies both to the Alliance’s substantive and procedural claims. We note that the Alliance also contends that the FEIS fails to identify its authors or the experts on which it relies. Although the agency could have been more specific, we reject this claim because the FEIS does contain a list of preparers, along with their qualifications and duties, and a list of consultants, along with their areas of expertise.

. Notably, the agency does provide hard data — its own as well as that of other researchers — in other areas of the FEIS that are ■ more readily quantifiable, such as with regard to the effect of the alternatives on the fisheries ORV.

. "NEPA imposes on federal agencies a continuing duty to supplement existing EAs and EISs in response to 'significant new circumstances or information relevant to environmental concerns and bearing on the proposed action or its impacts.' ” Idaho Sporting Congress, Inc. v. Alexander, 222 F.3d 562 , 569 n. 2 (9th Cir.2000) (quoting 40 C.F.R. § 1509 (c)(l)(ii)).

. See also "Non-Motorized Period Review Rationale” (”[N]o new information was revealed in my review that had not been addressed in the previous two NEPA processes and decisions. Nor was new significant information found that established a need to revisit the^ non-motorized period through a new NEPA process. I conclude that there is no substantive reason that requires further analysis to eliminate the non-motorized window.”).