The League of Wilderness Defenders-Blue Mountains Biodiversity Project, Cascadia Wildlands Project, and the Sierra Club (collectively, the Conservation Groups, or Groups) brought suit against John Allen, Forest Supervisor of the Deschutes National Forest, and the U.S. Forest Service (collectively, the Forest Service), alleging that the Five Buttes Project (Project) violates the National Forest Management Act (NFMA) and the National Environmental Policy Act (NEPA). The district court agreed, and granted summary judgment and an injunction in favor of the Conservation Groups. The Forest Service appeals.
We reverse, vacate the injunction, and remand with directions to the district court to grant summary judgment in favor of the Forest Service.
FACTS AND PRIOR PROCEEDINGS
1. The Northwest Forest Plan
The NFMA, 16 U.S.C. §§ 1600-1614, describes the statutory framework and requirements under which the Forest Service must manage National Forest System lands. The NFMA requires the Forest Service to develop a forest plan for each unit of the forest system, id. § 1604(a), and all subsequent agency actions must be consistent with the governing plan, id. § 1604(i). As required by the NFMA, the Forest Service developed the Northwest Forest Plan (NWFP) to protect and enhance old-growth forest ecosystems in the Pacific Northwest and Northern California that serve as habitats for numerous species. See Record of Decision for Amendments to Forest Service and Bureau of Land Management Planning Documents Within the Range of the Northern Spotted Owl, Summary, April 13, 1994, available at http://www.reo.gov/library/reports/ newroda.pdf (FS ROD). The endangered northern spotted owl (spotted owl) is an indicator species for the Deschutes National Forest, which lies within the NWFP area. As an indicator species, the spotted owl is a “bellwether ... for the other species that have the same special habitat or population characteristics.” Inland Empire Pub. Lands Council v. U.S. Forest Service, 88 F.3d 754, 762 n. 11 (9th Cir.1996).
2. The Davis Late Successional Reserve
In order to balance environmental and economic needs, the NWFP designates certain forest areas for logging and reserves other areas, called late successional reserves (LSRs), for conservation. Specifically, the NWFP created the Davis LSR to “protect and enhance conditions of latesuccessional and old-growth forest ecosystems, which serve as habitat for late-successional and old-growth related species including the northern spotted owl.” Except as otherwise permitted by law, commercial logging activities are prohibited in LSRs.
Wildfire and other disturbances occur frequently within the Davis LSR. Most notably, in 2003, a major fire in the Davis LSR (the Davis fire) burned approximately 21,000 acres of forest, including 3,736 acres of spotted owl nesting, roosting, and foraging (NRF) habitat, approximately 16,000 of which suffered near complete tree mortality. In all, the Davis fire burned twenty-four percent of the Davis LSR. In response to the Davis fire, the Forest Ser*1126vice revised its Davis LSR assessment to reflect the “immediate need” to “reduc[e] the risk of large-scale loss in a portion of the existing late and old-structure stands that are susceptible to insect attack and/or wildfire.” The objective of the Forest Service’s Project is to reduce that risk, in part, by thinning some of the trees in the Davis LSR. Objection to this logging component of the Project is the gravamen of the Conservation Groups’ complaint.
3. The Five Buttes Project
a. Purpose and Scope
The Forest Service is tasked with developing area-specific projects to fulfill the NWFP’s goals. The projects generally describe planned management and treatment activities in the relevant areas of the National Forest System lands. Treatment activities, or silviculture, include commercial thinning, regeneration cuts, salvage harvesting, and other activities intended to improve forest health.
The Project was designed in part to address the need identified in the updated Davis LSR assessment to reduce risks to the LSR from fire and disease. The Project covers approximately 160,000 acres (including the 48,900-acre Davis LSR) and authorizes management treatments, in-eluding commercial logging, across approximately 5,522 acres. It authorizes commercial logging in 618 acres of NRF habitat in the Davis LSR.1
The Project’s prescribed treatments differ depending on whether they are to occur within spotted owl habitat or home ranges. Within spotted owl home ranges, treatments would be “less intense or not done at all,” while NRF habitat outside the home range would be treated depending on vegetation type and crown fire potential. The Project is structured so that no spotted owls will be directly harmed. Five Buttes Project Environmental Impact Statement 391(EIS)(“There is no commercial thinning of NRF habitat proposed within an occupied spotted owl home range.”). The stated goal of the Project is to accelerate the development of large trees and NRF habitat to promote the objectives of the Davis LSR.
b. Environmental Impact Statement and Record of Decision
As required by NEPA, 42 U.S.C. § 4321 et seq., the Forest Service prepared a draft Environmental Impact Statement (EIS) for the Project that describes its expected environmental impacts. Id. § 4332(2)(C). The Forest Service circulated a draft EIS, received comments from the public, and *1127eventually issued a final EIS and a record of decision (ROD). In creating the EIS, the Forest Service conducted computerized simulations to determine the effects of wildfires on the Project area under three different treatment scenarios (A, B, and C) and to evaluate what level of treatment, if any, was needed to protect and preserve the Davis LSR.
Alternative A, or “No Action,” described the expected effects of fire and disease on the Project area if no additional thinning or fuels treatments are implemented. In other words, it modeled the effects of “allowing] current processes to continue, along with associated risks and benefits, in the [Project] area.”
Alternative B involved the most intensive treatments. It called for management activities across 5,522 acres and commercial harvesting of around 18.9 million board feet of lumber. This alternative proposed the greatest amount of commercial thinning within spotted owl NRF habitat (2,822 acres).
The Forest Service ultimately adopted Alternative C, which includes some treatments within the Davis LSR. Alternative C was developed to address landscape-scale fire prevention and retention of spotted owl habitat. It would “strategically place fuels treatments on the landscape to coordinate with past treatments to create and maintain fuel modifications around identified habitats.” This alternative calls for management of 7,798 acres and would harvest around 14.4 million board feet of lumber. It proposes commercial thinning on 2,023 acres of NRF habitat and would not involve treatment within any occupied spotted owl territory, but “habit modification would occur within unoccupied home ranges and could affect the ability of new owls to locate and establish a territory.” Under this alternative, “thinning would generally only remove trees less than 21 inches in diameter and less than 5 percent of all trees removed would exceed 21 inches in diameter.” Harvesting of trees larger than twenty-one inches “would only occur to meet basal area objectives or to lessen disease spread.”
The EIS states that Alternative C would reduce average burn probability by 40 percent over Alternative A, and that it “protects owl home ranges the best.” The Forest Service estimates that, if no preventative action is taken, the risk of large-scale loss of late-structure forest is extremely high, and the risk of a Davis-like problem fire is moderate to high. EIS 84 (“Due to current fuel loadings ... much of the landscape is classified as moderate to high risk of experiencing a Problem Fire similar to the Davis Fire.”); see also EIS 359(describing the risk of large-scale loss of large trees and late structure forest as “extremely high”).2 On appeal, the Conservation Groups contest this characterization, stating that “when the Forest Service considered all relevant factors, including ignition sources and location, it concluded that another large fire in the Five Buttes Project area was highly unlikely.” They *1128cite the Ager study, see infra pages 1133-34, for evidence that “conditions for a large wildfire event are rare within the study area.”3
In response to public concerns regarding the long-term effects on spotted owl habitat, the Forest Service stated that “[biased on modeling, the return to NRF conditions will take 2-5 decades depending on the thinning intensity prescribed. ... There is no commercial thinning of NRF habitat proposed within an occupied spotted owl home range; all other thinning within these areas would be small-diameter (3 inches or less) and stands would remain in NRF condition following activities.”
In response to comments opposing the harvest of larger trees, the Forest Service stated that the modeling had “determined that small-diameter thinning alone ... in most places, would not change the vegetation and fuels structure enough to reduce fire risk,” and would not solve the problem of overcompetition among larger trees. Also, limited harvest of larger trees would only take place outside NRF habitat, so that these areas would remain in NRF condition.
The EIS also contains a table and description of past, present, and reasonably foreseeable future actions in an effort to analyze the cumulative impact of the Project. Additionally, cumulative impacts are discussed in detail throughout the EIS in relation to each affected area, type of vegetation, or animal. See, e.g., EIS 66-70(discussing the existing conditions of vegetation to create baseline); EIS 90 (discussing the cumulative effects of past fires and past and future fuels treatments); EIS 117 (discussing cumulative impacts on spotted owls).
On June 8, 2007, the Forest Supervisor for the Deschutes National Forest signed the ROD, which adopted Alternative C with minor modifications and incorporated the EIS analysis. The ROD incorporates and responds to public concerns, and outlines the Forest Service’s final decision on the Project. It also reports the results of the computerized simulations and explains the Forest Service’s decision to adopt Alternative C.
4. Independent Review and Approval by Outside Agencies
The NWFP requires the Regional Ecosystem Office (REO)4 to review Forest Service projects to ensure consistency with the NWFP. Here, the REO concluded that Alternative C was consistent with the NWFP, stating that the “proposed treatment in the LSR meets the objectives for managing LSRs,” especially because “[treatments are placed on the landscape to alter fuel profiles in strategic locations such that it minimizes the likelihood of loss of valuable late- and old-structure habitat.” The REO also found that the proposed treatments “will result in a balanced mix of multi-story stands that are more ... con*1129ducive to late-suceessional species, with single story stands that are more resilient and less susceptible to loss,” and that “[t]reatments will focus on retaining the late and old structure components that are both desirable for late-successional species and adapted to the local fire regimes.”
Although not required to do so, out of an abundance of caution, the Forest Service also requested formal review of the Project by the United States Fish and Wildlife Service (FWS). The FWS conducted an extensive empirical analysis and published an eighteen-page opinion stating that “the[Project] is not likely to jeopardize the continued existence of the spotted owl” because the NWFP “provides a well distributed set of reserves which protect suitable habitat across the range of the spotted owl,” and because “no suitable habitat will be removed within spotted owl home ranges or core areas.” It concluded that spotted owl “home ranges will be maintained through fire risk reduction and stand density treatments” and that the short-term loss of some spotted owl habitat was justified by the “long-term benefits to owl habitat across the landscape by reducing fuel loads, strategic placement of treatments that reduce the risk of [fires], and reducing tree stocking densities to promote development of large trees in the future.”
5. Prior Proceedings
The Conservation Groups brought suit against the Forest Service, alleging that the Project calls for logging within the Davis LSR that does not comply with the NWFP. The Conservation Groups also claimed that the Forest Service’s EIS violates NEPA because the EIS does not adequately consider cumulative effects and does not respond to opposing views regarding logging and the prevention of catastrophic fires.
Regarding the NFMA challenge, the district court recognized that “[r]isk reduction activities are permitted in LSRs under limited circumstances.” However, in light of statements in the ROD recognizing possible negative effects on the LSR and spotted owls, the court found that “[t]he findings in the ROD are not strong enough to meet” the NWFP requirement that commercial thinning projects focused on older stands must “clearly result in greater assurance of long-term maintenance of habitat.”
With respect to the NEPA claim, the district court found that the cumulative impact discussion in the EIS was deficient for lack of detailed, quantitative information about past projects. League of Wilderness Defenders-Blue Mountains Biodiversity Project v. Weldon, No. 07-6238-HO, 2008 WL 4279807, at *4 (D.Or. Sept.11, 2008) (District Court Order) (“[T]hese [past] projects have not been quantified with time, place, and scale data.”). The court also found that the Forest Service’s use of an “aggregate effects” approach was “in contradiction to current Ninth Circuit law.” Id. at *5. The court declined to reach the Conservation Groups’ second NEPA claim regarding disclosure of opposing scientific opinions. Id.
The district court granted summary judgment in favor of the Conservation Groups, enjoined the Forest Service from any additional logging, and remanded to the Forest Service for preparation of a new ROD that complies with the NFMA and NEPA.5 The Forest Service appeals the district court’s final judgment.
*1130STANDARDS OF REVIEW AND JURISDICTION
Under the Administrative Procedure Act (APA), a reviewing court can only reverse an agency decision if that decision was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). To address what had been a gradual divergence from this highly deferential standard by our court, we recently went en banc and unanimously held that:
Review under the arbitrary and capricious standard is narrow, and [we do] not substitute [our] judgment for that of the agency. Rather, we will reverse a decision as arbitrary and capricious only if the agency relied on factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.
Lands Council v. McNair, 587 F.3d 981, 987 (9th Cir.2008) (en banc) (internal quotation marks and citations omitted) (alterations in original). This deference is highest when reviewing an agency’s technical analyses and judgments involving the evaluation of complex scientific data within the agency’s technical expertise. Id. at 993.
For NEPA claims, “[w]e employ a rule of reason [standard] to determine whether the [EIS] contains a reasonably thorough discussion of the significant aspects of the probable environmental consequences.” Ctr. for Biological Diversity v. U.S. Forest Serv., 349 F.3d 1157, 1166 (9th Cir.2003) (internal quotation marks omitted) (second alteration in original). We must ensure “that the agency has taken a ‘hard look’ at the environmental consequences of the proposed action,” id. (quoting Churchill County v. Norton, 276 F.3d 1060, 1072 (9th Cir.2001)), and we must uphold the agency decision as long as the agency has “considered the relevant factors and articulated a rational connection between the facts found and the choice made,” Selkirk Conservation Alliance v. Forsgren, 336 F.3d 944, 953-54 (9th Cir.2003) (quoting Wash. Crab Producers, Inc. v. Mosbacher, 924 F.2d 1438, 1441(9th Cir.1990)).
We review the district court’s grant of summary judgment de novo, Feldman v. Allstate Ins. Co., 322 F.3d 660, 665(9th Cir.2003), and its grant of injunctive relief for abuse of discretion, Rolex Watch, U.S.A., Inc. v. Michel Co., 179 F.3d 704, 708 (9th Cir.1999). We have jurisdiction under 28 U.S.C. § 1291.
DISCUSSION
I. Compliance with the NFMA and the NWFP
The NWFP requires LSRs to be managed to “protect and enhance conditions of late-successional and old-growth forest ecosystems.” NWFP 8. The NWFP has “[guidelines to reduce risks of large-scale disturbance,” which provide that “[s]ilvicultural activities aimed at reducing risk shall focus on younger stands in [LSRs],” and “the scale of ... treatments should not generally result in degeneration of currently suitable owl habitat or other late-successional conditions.” To avoid degeneration, “logging and other ground-disturbing activities are generally prohibited” in LSRs. Or. Natural Res. Council Fund v. Brong, 492 F.3d 1120, 1126 (9th Cir. 2007) (quoting Seattle Audubon Soc’y, 871 F.Supp. at 1304-05).
However, the NWFP permits logging activities in LSRs if: “(1) the proposed management activities will clearly result in *1131greater assurance of long-term maintenance of habitat, (2) the activities are clearly needed to reduce risks, and (3) the activities will not prevent the [LSRs] from playing an effective role in the objectives for which they were established.” Standards and Guidelines for Management of Late-Successional and Old-Growth Forest Related Species within the Range of the Northern Spotted Owl C-13 (April 13, 1994) [hereinafter NWFP S. & G.]. The NWFP acknowledges that some logging “may reduce the quality of habitat for latesuccessional organisms” and that “managers need to seek a balanced approach that reduces the risk of fire while protecting large areas of fire-prone late-successional forest.” NWFP S. & G. B-7.
Our highest deference is owed to the Forest Service’s technical analyses and judgments within its area of expertise, Lands Council, 537 F.3d at 993; nonetheless, our dissenting colleague would have us halt the Forest Service’s Project because he does not like the Forest Service’s approach to solving the problems addressed. We went en banc to foreclose precisely this type of second-guessing of the Forest Service. See id. at 988(noting that “in recent years, our environmental jurisprudence has, at times, shifted away from the appropriate standard of review and could be read to suggest that this court should” “act as a panel of scientists that instructs the Forest Service” how to perform its expert duties). The Forest Service thoroughly considered various reasonable approaches to “protect and enhance conditions” of the LRSs, NWFP S. & G. at C — 11, and offered a plan that does not “run[ ] counter to the evidence before the agency or is so implausible that it could be not ascribed to a difference in view or the product of agency expertise,” Earth Island Inst. v. U.S. Forest Serv., 442 F.3d 1147, 1156 (9th Cir.2006), abrogated on other grounds by Winter v. Natural Re. Def. Council, Inc., — U.S.-, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008). Far from conflicting with the protection of LSRs, carefully controlled logging is a tool expressly authorized by the NWFP for long-term LSR maintenance.
Our dissenting colleague simply articulates a different point of view, and an extreme one at that. He begins by implying that the law provides that under no circumstances can any old growth timber be cut from an LSR. Dissent at 1140-41. Eventually, however, our colleague reluctantly admits that is simply not the case. Dissent at 1141 (“[T]he NWFP ... recognize[s] that in some LRSs [including the Davis LSR], stand management that includes older trees may be considered.”). Thus, it is clear that the limited logging of LSR trees of all types and ages is permissible if such logging complies with the three NWFP requirements previously described. There is no dispute that the Forest Service correctly identified that standard. The question, then, is not whether such logging within the LSR is generally permitted, but whether the Forest Service’s determination that the NWFP requirements for cutting some trees in an LSR were met was arbitrary or capricious. We now turn to that analysis.
After the Davis fire, the Forest Service determined that logging was necessary to protect the Davis LSR from future fires and that the Project met the three NWFP criteria for logging within LSRs. It noted that in the previous five years, “approximately 16,654 acres of NRF habitat ha[d] been lost mostly due to wildfires.” The Project, in comparison, calls for thinning only 618 acres of NRF habitat. This would leave 93 percent of existing NRF habitat and 96 percent of spotted owl critical habitat in the Davis LSR. The Project would reduce fire risks by 40 percent over Alternative A(the no action alternative). *1132Wildfire accounted for 75 percent of the disturbance-caused loss of spotted owl habitat between 1994 and 2008.
After a lengthy and thorough discussion of the alternatives considered, the Forest Service concluded that:
Alternative C best responds to the issue of impacts to the northern spotted owl because [1] Alternative C would alter fewer acres of NRF habitat than Alternative B; [2] Alternative C would maintain at least three areas of habitat that are available for immediate occupancy by dispersing or relocating spotted owls; and [3] Alternative C provides the best strategy for risk reduction and long-term maintenance of spotted owl habitat on the landscape.
According to the Forest Service’s modeling, “in Alternative C, fires appear to have the least travel times and protect [spotted] owl home ranges the best.”6 After comparing the alternatives, the Forest Service determined that this plan would “clearly result in greater assurance of long-term maintenance of habitat,” given the high threat of another catastrophic wildfire. In fact, it found that the “[proposed activities would not only reduce risk of large-scale disturbance, but would accelerate the ability of the Davis LSR to play a role for which late-successional reserves were established.” The Conservation Groups challenge the Forest Service’s conclusions on three grounds.
First, the Conservation Groups allege that the Forest Service knew that the Project would not clearly result in greater assurance of long-term maintenance of habitat. There are two related problems with this position. The Conservation Groups rely heavily on a section of the EIS labeled “Issues” to suggest that the Forest Service knew that the treatments would have long-term detrimental effects on spotted owl habitat. This section states: “These activities may reduce the quality, effectiveness, and the distribution of habitat available to the northern spotted owl in the planning area for the short- and long-term as well as directly, indirectly and/or cumulatively.” However, this excerpt is simply a republication of potential issues identified by the public and the Forest Service; it is not & finding made by the Forest Service and is not, as the Conservation Groups allege, evidence that the Forest Service “specifically determined” that such damage would actually occur.
More to the point, however, there is no dispute that the Forest Service knew of (and acknowledged) possible detrimental effects to the spotted owl. Specifically, it stated that “[i]n those units proposed for commercial timber harvest the conversion of existing NRF habitat to a foraging and dispersal condition is expected to be at least a short-term effect.” The Forest Service also estimated that “[b]ased on modeling, the return to NRF conditions will take 2-5 decades depending on the thinning intensity prescribed.”7' The Con*1133servation Groups’ statement of harms obfuscates the issue, which is not how long the harms might last but whether the benefits will outweigh the costs in the long-term.
Thus, the Forest Service’s alleged admissions about possible harms actually describe the balancing of risks that the Forest Service was required to undertake. Such balancing is entirely appropriate under the NWFP, which states that treatments in LSRs “may reduce the quality of habitat for late-suceessional organisms,” and that “managers need to seek a balanced approach that reduces risk of fire while protecting large areas of fire-prone late-successional forest.” NWFP B-8 (emphasis added). After detailing possible detrimental effects, the Forest Service made clear that those effects were limited and relatively short-term and, most importantly, would be outweighed by the overall improvement and continued viability of spotted owl NRF habitat. Indeed, it seems that avoiding detrimental effects altogether was impossible, because the Forest Service found that “[a]ll alternatives, including passive management, [m]ay affect, and are likely to adversely affectf,] the northern spotted owl.” Five Buttes ROD 13.8
Second, the Conservation Groups argue that the Forest Service failed to consider a study published by Forest Service scientists examining the Forest Service’s forest treatment and fire models. See Alan A. Ager, et al., Modeling Wildfire Risk to Northern Spotted Owl (Strix Oceidentalis Caurina) Habitat in Central Oregon, USA, 246 Forest Ecology & Mgmt. 45 (2007) (the Ager study). Unlike past studies, the Ager study examined the possibility of risk reduction without treatments within spotted owl territory. Id. at. 54.
As an initial matter, the Ager study was published after the administrative record was closed and the Forest Service had made its final decision, and well after the Forest Service developed the Project. Its relevance to what the Forest Service knew about the viability of risk-reduction exclusively outside of spotted owl territory is tenuous at best. Second, the Ager study was intended as a general examination of modeling capabilities, not as a means of determining an appropriate course of action for the Project. Ager Study 46 (“In this paper, we describe a wildfire risk analysis system for quantifying potential wildfire impacts on spotted owl habitat and measuring the efficacy of landscape fuel treatment on reducing risk.”). The study estimated that, by logging 20 percent of the forest area outside spotted owl habitat, the Forest Service could reduce the average burn probability by 44 percent. Ager Study 50, Table 1. Alternative C, which includes logging in some unoccupied spotted owl habitat, would reduce burn probability by 40 percent. Thus, the Conservation Groups use the Ager study as a strawman because the study did not determine if it was possible, let alone desirable, to log 20 percent of the Project Area. Alternative C, in comparison, treats less than 6 percent of the Project Area *1134with almost the same reduction of burn probability.
In fact, the Ager study itself acknowledges that “allowing treatments within spotted owl habitat in the present study would have substantially decreased the expected habitat loss at a given treatment intensity.” Ager Study 55(emphases added). Because the Forest Service is tasked with developing a viable plan that will clearly result in the long-term maintenance of spotted owl habitat, we fail to see how the Ager study, which deals with general modeling capabilities and experimental risk-reduction approaches, contradicts the Forest Service’s practical determination that some logging in the Davis LSR was clearly necessary to retain long-term forest and species health.
Finally, the Conservation Groups claim that it was not clearly necessary to cut larger diameter trees. The Groups raised this issue with the Forest Service after reviewing the draft EIS, and the Forest Service actively considered the alternative of thinning only small diameter trees by conducting additional modeling. The Forest Service found that “modeling of fire behavior and vegetation indicated that small diameter thinning by itself would not considerably change the expected fire behavior on a landscape scale.” In the ROD, the Forest Service also explained that “[r]eduction of competition between trees in overstocked sites through commercial thinning is a hedge against epidemic loss of the larger trees to insect and disease.” Thus, the Forest Service determined that cutting some9 larger diameter trees was clearly necessary and met the NWFP standard.
In sum, the Conservation Groups have not demonstrated that the Forest Service made a clear error of judgment in determining that the proposed treatments conformed with the NWFP, or that the decision to implement Alternative C was arbitrary or capricious.10 See Lands Council, 537 F.3d at 993 (holding that the proper role for a reviewing court “is simply to ensure that the Forest Service made no clear error of judgment that would render its action arbitrary and capricious” (internal quotation marks omitted)).
We agree with the dissent’s sentiment regarding the “importance of a[n] LSR in fulfilling the objectives of the NWFP.” Dissent at 1139. We differ from the dissent, however, in our view that no member of the panel is better equipped than the experts at the Forest Service to determine how best to fulfill those objectives. That task goes to the very heart of the Forest Service’s expertise. Indeed, where the Forest Service has determined that stand treatments are clearly needed to reduce risks, will clearly result in greater assurance of long term maintenance of habitat, and will not prevent the LSR from playing the role for which they were established, NWFP S. & G. at C-13, we should be loathe to second guess their efforts absent some glaring error, oversight, or arbitrary action, lest we be the ones who upset the LSR’s “ancient and intricate process ... because we do not fully understand [its] inner workings,” Dissent at 1138.
As the Conservation Groups conceded at oral argument, the district court’s holding that “[t]he findings in the ROD are not strong enough to meet” the NWFP stan*1135dard does not accord the Forest Service proper deference. Because we find that the Project does not reflect any clear error in judgment by the Forest Service, we reverse the district court and hold that the Project does not violate the NFMA.
II. Compliance with NEPA
The Conservation Groups also allege that the Forest Service violated NEPA because, first, the EIS’s cumulative impact methodology was deficient and, second, the EIS failed to adequately deal with opposing views on the impact and efficacy of thinning larger trees. We disagree.
We note first that NEPA does not impose substantive requirements — it “exists to ensure a process,” Inland Empire Pub. Lands Council, 88 F.3d at 758, and “aims to make certain that ‘the agency ... will have available, and will carefully consider, detailed information concerning significant environmental impacts,’ and ‘that the relevant information will be made available to the [public],’ ” Lands Council, 537 F.3d at 1000 (quoting Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 349, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989)).
A. Cumulative Impact
NEPA requires federal agencies proposing “major Federal action[] significantly affecting the quality of the human environment” to prepare a detailed EIS on the proposed action, including an analysis of alternatives and a discussion of the significant environmental impacts. 42 U.S.C. § 4332(2)(C). An EIS must include a detailed statement on:
(i) the environmental impact of the proposed action, (ii) any adverse environmental effects which cannot be avoided should the proposal be implemented, (iii) alternatives to the proposed action, (iv) the relationship between local short-term uses ... and enhancement of long-term productivity, and (v) any irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented.
Id.
The Council on Environmental Quality (CEQ), a body created by NEPA, “promulgated regulations to promote compliance with the ‘action-forcing’ requirements of NEPA----Although initially advisory in nature, the regulations were made binding on the administrative agencies by Executive Order No. 11991, 3 C.F.R. § 124.” Churchill County, 276 F.3d at 1072 n. 7. The CEQ regulations require an EIS to assess the “cumulative impact” of a proposed action, defined as the “impact on the environment which results from the incremental impact of the action when added to other past, present, and reasonably foreseeable future actions.” 40 C.F.R. § 1508.7. Our role in reviewing an EIS is “to ensure that the agency has taken a ‘hard look’ at the potential environmental consequences of the proposed action.” Klamath-Siskiyou Wildlands Ctr. v. Bureau of Land Mgmt., 387 F.3d 989, 993 (9th Cir.2004) (quoting Churchill County, 276 F.3d at 1072).
In 2005, the CEQ issued a memorandum advising agencies that they “are not required to list or analyze the effects of individual past actions unless such information is necessary to describe the cumulative effect of all past actions combined.” James L. Connaughton, Chairman, CEQ, Guidance on the Consideration of Past Actions on Cumulative Effects Analysis 2 (June 24, 2005), available at http://gc. energy.gov/NEPA/ nepa_documents/TOOLS/GUIDAN C E/Volumel/411.2_cumulative_effects_june24-05.pdf (CEQ Memorandum) (last visited July 30, 2010). This interpretation of the cumulative impact requirement is referred to as *1136the “aggregate effects” approach. CEQ Memorandum 2. The “CEQ’s interpretation of NEPA is entitled to substantial deference.” Andrus v. Sierra Club, 442 U.S. 347, 358, 99 S.Ct. 2335, 60 L.Ed.2d 943 (1979).
The district court determined that the EIS’s cumulative impact analysis was deficient for lack of detailed, quantitative information about past projects. District Court Order, 2008 WL 4279807, at *4 (“[T]he [past] projects have not been quantified with time, place, and scale data.”). This “time, place, and scale” language comes from Lands Council v. Powell, 395 F.3d 1019, 1028 (9th Cir.2005), which held that a cumulative effects analysis violated NEPA because it failed to provide “adequate data of the time, type, place, and scale of past timber harvests.” See also Brong, 492 F.3d at 1133. However, soon after the district court issued its decision, and well after Powell and Brong were decided, we held that the CEQ’s aggregate effects approach — which does not necessarily require specific time, place, and scale data — is not plainly erroneous or inconsistent "with NEPA. See League of Wilderness Defenders-Blue Mountains Biodiversity Project v. U.S. Forest Serv., 549 F.3d 1211, 1217-18 (9th Cir.2008); accord Ecology Ctr. v. Castaneda, 574 F.3d 652, 666 (9th Cir.2009) (confirming that “the Forest Service may aggregate its cumulative effects analysis pursuant to 40 C.F.R. § 1508.7” (internal quotation marks omitted)).
The Forest Service explicitly relied on the CEQ Memorandum in creating its cumulative impact analysis. Specifically, the EIS noted that it was permissible to “focus[] on the current aggregate effects of past actions without delving into the details of individual past actions.” It then provided a table listing past, present, and foreseeable future projects, followed by a detailed description of the relevant activities and the status of the projects.
The Conservation Groups assert that, apart from not meeting the formalistic requirements that were erroneously imposed by the district court, the table does not contain enough detailed information about cumulative impact. However, this ignores the lengthy discussion of cumulative impact that follows in the EIS. With regard to the spotted owl, the EIS contains a thorough twenty-three-page discussion of previous declines, trends, and threats to the spotted owl population and habitat. It discusses each of the activities listed in the table to “assess whether, in combination with the Five Buttes Project, there would be overlap of time and space.” The EIS then describes possible overlapping effects from other projects and natural disasters such as wildfires, mushroom harvesting, and planned vegetation projects, and conducts a similar analysis with regard to soil quality, fires, fuels, and other species. This analysis fully complies with the requirement that the EIS consider aggregate effects of past, present, and future actions, and the Conservation Groups have not pointed to any past action or effect that was.not considered. In fact, the inclusion of additional, unnecessary data in the EIS would run contrary to the purpose of NEPA, as “NEPA documents must concentrate on the issues that are truly significant to the action in question, rather than amassing needless detail.” 40 C.F.R. § 1500.1(b).
The aggregate effects approach is not, as the district court held, “in contradiction to current Ninth Circuit law.” See League of Wilderness Defenders, 549 F.3d at 1217-18; Castaneda, 574 F.3d at 666. The EIS’s cumulative impact analysis was consistent with the CEQ guidelines and adequately considered cumulative effects of *1137past, present, and reasonably foreseeable future projects in compliance with NEPA.
B. Opposing Views
The Conservation Groups also assert that the Project’s EIS failed to consider opposing scientific views on the impact of thinning. They claim that the Project record “contains volumes of scientific information that counsel against logging large trees from older forests and that establish that there is, at a minimum, uncertainty as to whether the proposed treatment of older forests will have the desired effect of reducing the risk of natural disturbances.”
An EIS must contain “a reasonably thorough discussion of the significant aspects of the probable environmental consequences.” Ctr. for Biological Diversity, 349 F.3d at 1166 (internal quotation marks omitted). A proper NEPA analysis “foster[s] both informed decision-making and informed public participation.” Churchill County, 276 F.3d at 1071 (quoting California v. Block, 690 F.2d 753, 761(9th Cir.1982)). Significantly, the Forest Service is only required to “acknowledge and respond to comments by outside parties that raise significant scientific uncertainties and reasonably support that such uncertainties exist.” Lands Council, 537 F.3d at 1001 (emphasis added).
Here, the Conservation Groups incorrectly claim that the Forest Service failed to consider the view that priority should be given to cutting smaller trees. In fact, the Forest Service subscribes to that view, and the majority of tree removal is aimed at smaller trees. EIS 410(“[T]he greatest density of trees that will be removed is in the smaller size classes, contributing to the greatest risk reduction.”). The EIS also explains that cutting small trees alone is not sufficient because “reduction of the smaller size classes, in general, does not contribute to reduction of crown bulk density,” which is important in preventing crown fires. The Forest Service also concluded that the removal of larger trees is required to reduce susceptibility to insect infestation and disease. In the EIS and ROD, the Forest Service addressed each of the Conservation Groups’ suggestions and concerns, even conducting additional modeling to determine if thinning only smaller trees might be feasible.
The Conservation Groups also describe two additional opposing views they feel were not taken into consideration. First, they claim that logging older trees would increase the risk of fire because of debris. This issue was adequately addressed and rebutted. The Forest Service stated that debris generated from the cutting would be removed, and the EIS established a maintenance schedule (burning every eight-to-twelve years, hand piling and disposal every fifteen-to-twenty years) for thinned areas to prevent dangerous buildup of ladder fuels. Finally, the Conservation Groups raised concerns about roads that would be built, but the Forest Service explained that the roads would be temporary, barricaded when not in use, and sub-soiled to facilitate their return to vegetative productivity.
We also note that the REO and the FWS conducted extensive independent reviews of the Project and found it to be consistent with NEPA. See supra Section 4; cf. Ctr. for Biological Diversity, 349 F.3d at 1161, 1169 (holding that Forest Service failed to consider opposing views where the FWS challenged the Forest Service’s scientific conclusions). We hold that the Forest Service adequately considered and responded to alternative views about the Project’s potential environmental consequences.
The Forest Service adequately considered the cumulative impact of past, pres*1138ent, and foreseeable future actions and sufficiently considered and responded to opposing scientific views. Therefore, we hold that the Forest Service complied with NEPA and took the requisite “hard look” at the environmental impacts of the Project.
CONCLUSION
For the foregoing reasons, we REVERSE, VACATE the injunction, and REMAND to the district court for entry of summary judgment in favor of the Forest Service.
. The Five Buttes Record of Decision clarifies that the commercial benefit that the sale of harvested timber will bring to the local economy is merely "attendant” to the primary goal of lessening "the risk that disturbance events ... will lead to large-scale loss of forest.” Five Buttes Record of Decision, 9, 11, June 2007, available at https://scholarsbank. uoregon.edu/xmlui/handle/1794/7069/ (Five Buttes ROD) (last visited July 30, 2010). Cf. Siskiyou Reg'l Educ. Project v. Goodman, 219 Fed.Appx. 692, 695 (9th Cir.2007) ("The fact that there may be some incidental economic benefit from the recovery project’s sale of burned trees is not contrary to and does not overshadow the NWFP’s primary goals of forest protection and restoration.”). If commercial logging was the sole purpose of the project, we would agree with Judge Paez that the NWFP had been flouted. See Seattle Audubon Soc'y v. Evans, 771 F.Supp. 1081, 1096(W.D.Wash.l991) (granting an injunction against sale of purely commercial logging rights in spotted owl habitat because the Forest Service had acted under the mistaken belief that it did not have to comply with the NFMA at all). The district court in Seattle Audubon Soc'y distinguished that case as "not the usual situation in which the court reviews an administrative decision and, in doing so, gives deference to agency expertise.” Id. at 1096. Here, on the other hand, the Forest Service's primary objective is preservation of the LSR, and it spent approximately three years doing the analysis necessary to comply with the NFMA.
. The dissent misreads these two assessments as "inconsistent.” Dissent at 1144-45. The Forest Service’s assessment is more nuanced than the dissent cares to admit: the risk of a large-scale loss of late-structure forest is extremely high, and the risk of a wholly devastating Davis-like fire is moderate to high. There is no inconsistency. Regardless, as described below, the NWFP does not require a specified level of risk for logging to take place. Our colleague's lengthy complaint about the syntax of risk is a distraction: the Forest Service determined, after extensive review and modeling, that the risk of disturbance was great enough to meet the three NWFP requirements for stand treatments. The question is whether that review and modeling met the APA’s extremely deferential standard; whether that risk is labeled "high” or "extremely high” is ultimately irrelevant and, as Judge Paez himself notes, “meaningless.” Dissent at 1143-44.
. As discussed more fully below, the Ager study was not conducted to assess the Project, see infra pages 1133-34, and was published after the administrative record was closed and the Forest Service had made its decision. Accordingly, we find the Ager study to be of little use in evaluating the risk of fire within the Project area, and of no use in determining what the Forest Service knew about that risk at the time the Forest Service made its decision.
. The REO provides staff and support to the Regional Interagency Executive Committee, which is "composed of Regional Directors from the various land management, regulatory, research, and other relevant agencies in the Federal government located [in the Northwest]” and "serves as the senior regional entity to assure the prompt, coordinated, and successful implementation of the [NWFP] at the regional level.” Regional Interagency Executive Committee, http://www.reo.gov/riec/ index.htm# who (last visited July 28, 2010).
. The district court permitted Interior Pacific and American Forest Resource Council (collectively, Intervenors) to intervene in the remedial phase. The court granted Intervenors' motion to modify and limit the injunction to permit removal of remaining downed logs *1130and other limited actions to protect the area. Intervenors are parties in this appeal.
. The dissent is mistaken that ”[t]he three alternatives (A, B, and C) are not compared with either a decision not to log in the LSR or to thin only small trees.” Dissent at 11596. Alternative A is the no logging alternative. See FS ROD at 13("Alternative A — No Action[:] The purpose of this alternative is to allow current processes to continue, along with associated risks and benefits!.] • • • Under the No Action alternative, ... [n]o additional thinning or fuels treatments would be implemented to accomplish project goals.... In Alternative A (no action), no risk reduction activities would occur; therefore, the potential remains for large-scale loss of northern spotted owl habitat, similar to the scale seen in the Davis Fire of 2003. These disturbance events are expected to increase the potential to become more frequent and larger in scope than at present.”). The Forest Service also considered the option to thin only small trees. See infra p. 1134.
. The Conservation Groups vigorously argued that two to five decades is an intolerably long time, but ultimately conceded at oral argument that twenty to fifty years is a compara*1133tively short span of time for forests that have been in existence for thousands of years.
. The dissent argues that the Forest Service's assessment of risk is "skewed by its focus on one lone variable — reduction in fire risk.” Dissent at 1145. But that is clearly not accurate. We agree with the dissent that the absurd result of bulldozing the LSR could be justified by an approach that values only fire risk reduction. But the Forest Service did not choose such a "nonsensical” scenario precisely because it was concerned with long-term forest preservation, not simply reduced fire risk. In fact, the Forest Service chose Alternative C, which cut more LSR trees than Alternative A (the no-cutting alternative) but fewer LSR trees than Alternative B. This choice was made after a careful balancing of the risks and benefits of the treatments.
. A majority of the trees to be cut under the Project are smaller diameter trees.
. Importantly, the Conservation Groups do not challenge the modeling techniques or quantitative results used by the Forest Service, and our independent review of those matters gives us no reason to doubt their validity.