League of Wilderness Defenders/Blue Mountains Biodiversity Project v. United States Forest Service 883 F. Supp. 2d 979

AMENDED OPINION AND ORDER

MICHAEL H. SIMON, District Judge.

The League of Wilderness Defenders/Blue Mountains Biodiversity Project (“LOWD” or “Plaintiff’) challenges the decision of the U.S. Forest Service and its Regional Forester for the Pacific Northwest Region (collectively, “Forest Service” or “Defendant”) to increase the use of herbicides in controlling invasive plant species in the Wallowa-Whitman National Forest.1 The Wallowa-Whitman National Forest comprises 2.3 million acres in the northeast corner of Oregon and the western edge of Idaho — an area larger than the states of Delaware and Rhode Island combined. LOWD argues that the Forest Service, in approving an Invasive Plants Treatment Project for the Wallowa-Whitman National Forest (the “Project”), did not comply with three federal statutes: the National Forest Management Act, 16 U.S.C. § 1600 et seq. (“NFMA”); the National Environmental Policy Act, 42 U.S.C. § 4321 et seq. (“NEPA”); and the Clean Water Act, 33 U.S.C. § 1251 et seq. It seeks declaratory and injunctive relief. The parties have cross-moved for summary judgment. For the reasons stated below, the court GRANTS IN PART each party’s motion for summary judgment and DENIES IN PART each party’s motion.

In 2005, the Regional Forester for the Pacific Northwest Region of the U.S. Forest Service (also known as Region Six) approved a new management direction on preventing and managing invasive plants. Notably, the new regional management direction approved an updated list of ten herbicides for use within Region Six. To reflect this new regional direction, the Wallowa-Whitman National Forest amended its local management plan. As required by NEPA, the Forest Service first prepared an Environmental Impact Statement (“EIS”). After revising the EIS based on comments from stakeholders such as LOWD, the Forest Service issued a Final Environmental Impact Statement (the “Project FEIS”) in March 2010, which recommended increasing the use of herbicides to control invasive plants and allowing the use of all ten of the herbicides approved at the regional level. The Forest Supervisor of the Wallowa-Whitman National Forest approved this recommended approach in a Record of Decision (the “Project ROD”) in April 2010.

LOWD appealed that decision within the agency without success. LOWD now seeks judicial review before this court. The court’s review of an agency’s compliance with NFMA, NEPA, and the Clean Water Act is limited by the Administrative Procedure Act (“APA”), 5 U.S.C. § 551 et seq. The Lands Council v. McNair, 537 F.3d 981, 987 (9th Cir.2008) (en banc); Great Basin Mine Watch v. Hankins, 456 F.3d 955, 961 (9th Cir.2006). Under the relevant APA standard, this court may set aside the agency’s decision only if it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). “Review un*983der the arbitrary and capricious standard ‘is narrow, and [we do] not substitute [our] judgment for that of the agency.’ ” McNair, 537 F.3d at 987 (quoting Earth Island Inst. v. U.S. Forest Serv., 442 F.3d 1147, 1156 (9th Cir.2006)) (alterations in original). As directed by the Ninth Circuit in McNair, this court must defer to the agency’s technical expertise, as long as there is no clear error of judgment. See id. at 993-94.

LOWD makes three principal arguments. First, LOWD argues that the Project does not comply with the local forest management plan and that the Forest Service did not adequately analyze the Project’s compliance with NFMA. I conclude that the Forest Service’s explanation of its modeling data was reasonable and that the modeling data did not contradict the Forest Service’s conclusion that the Project will comply with the local forest management plan. I also conclude that the Forest Service adequately analyzed the Project’s compliance with NFMA and that it did so at an appropriate scale.

Second, LOWD argues that the Project FEIS and Project ROD do not satisfy the requirements of NEPA. I conclude that the Forest Service’s statement of purpose and need for the Project was not arbitrary or capricious. In particular, I accept the Forest Supervisor’s decision that the Project should focus primarily on the treatment of existing infestations of invasive plants rather than on prevention of future infestations. I also conclude that the Forest Service evaluated a reasonable range of alternative actions. I agree with LOWD, however, that the Forest Service did not adequately evaluate the cumulative impacts that the Project might have when considered in conjunction with other actions. As a result, I hold that the Project ROD was arbitrary and capricious under the APA.

Third and finally, LOWD argues that the Forest Service should have discussed the significant possibility that it would need to obtain permits in order for the Project to comply with the Clean Water Act. LOWD acknowledges that the Forest Service did not need permits in order to comply with the Clean Water Act at the time the Project ROD and FEIS were adopted. Instead, LOWD argues that NEPA requires the Forest Service to discuss a reasonably foreseeable need to obtain permits. I conclude that it was not arbitrary or capricious under NEPA for the Forest Service not to discuss a likely change in law that would require it to obtain permits.

BACKGROUND

I. National Environmental Policy Act

The National Environmental Policy Act (“NEPA”) “declares a broad national commitment to protecting and promoting environmental quality.” Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 348, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989). It achieves this purpose by forcing agency deliberation. Id. at 350, 109 S.Ct. 1835. NEPA does not “mandate particular results” for agency decisionmaking, but “simply prescribes the necessary process.” Id.; see also McNair, 537 F.3d at 1000.

The heart of NEPA is the requirement that, before any “major Federal action[] significantly affecting the quality of the human environment,” the responsible official must prepare “a detailed statement” that includes:

(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,

*984(iv) the relationship between local short-term uses of man’s environment and the maintenance and enhancement of long-term productivity, and

(v) any irreversible and irretrievable commitments of resources which would be involved in the proposed action should it be implemented.

42 U.S.C. § 4332(2)(C). This environmental impact statement (“EIS”) serves two important purposes: “[i]t ensures that the agency, in reaching its decision, will have available, and will carefully consider, detailed information concerning significant environmental impacts,” and it “guarantees that the relevant information will be made available to the larger [public] audience.” Robertson, 490 U.S. at 349, 109 S.Ct. 1835. Under Ninth Circuit precedent, an agency has taken the requisite “hard look” at environmental consequences when its EIS contains all the items listed in 42 U.S.C. § 4332(2)(C) and includes “a full and fair discussion of environmental impacts.” McNair, 537 F.3d at 1001. After the agency has taken this “hard look,” however, it is free to decide that other values outweigh any environmental costs it has identified. See Robertson, 490 U.S. at 350-51, 109 S.Ct. 1835 (“NEPA merely prohibits uninformed — rather than unwise — agency action.”). Courts will not “substitute [their] judgment for that of the agency concerning the wisdom or prudence of a proposed action.” City of Carmel-by-the-Sea v. U.S. Dep’t. of Transp., 123 F.3d 1142, 1150 (9th Cir.1997) (quoting Or. Envtl. Council v. Kunzman, 817 F.2d 484, 492 (9th Cir.1987)) (quotation marks omitted).

Private individuals who can establish standing may challenge the adequacy of an agency’s NEPA process under the APA.2 An agency’s decision is arbitrary or capricious in violation of the APA “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.’” McNair, 537 F.3d at 987 (quoting Earth Island Inst., 442 F.3d at 1156).

II. National Forest Management Act

The National Forest Management Act (“NFMA”) requires the Forest Service to develop “land and resource management plans” for units of the National Forest System. 16 U.S.C. § 1604(a). Among other considerations, these plans must “provide for multiple use and sustained yield of the products and services obtained [from National Forest System lands] in accordance with the Multiple-Use Sustained-Yield Act of 1960, and, in particular, include coordination of outdoor recreation, range, timber, watershed, wildlife and fish, and wilderness.” Id. § 1604(e)(1).3 As the Ninth Circuit has explained, “NFMA is explicit that wildlife viability is not the Forest Service’s only consideration when developing site-specific plans.” McNair, 537 F.3d at 990. After a *985land and resource management plan is adopted for a particular forest, all subsequent actions (e.g., “[r]esource plans and permits, contracts, and other instruments for the use and occupancy of National Forest System lands”) must be consistent with that plan. 16 U.S.C. § 1604(i).

As with NEPA, courts review the Forest Service’s compliance with NFMA under the APA’s arbitrary or capricious standard. McNair, 537 F.3d at 987. The Ninth Circuit has cautioned that courts must be particularly deferential when the Forest Service is considering highly technical issues within its area of special expertise. Id. at 993. As “non-scientists,” courts must not “impose bright-line rules on the Forest Service regarding particular means that it must take in every case to show us that it has met the NFMA’s requirements.” Id. at 993-94. Instead, a court may conclude “that the Forest Service acts arbitrarily or capriciously only when the record plainly demonstrates that the Forest Service made a clear error in judgment in concluding that a project meets the requirements of the NFMA and relevant Forest Plan.” Id. at 994.

III. The Wallowa-Whitman National Forest Land and Resource Management Plan

The Wallowa-Whitman National Forest’s current land and resource management plan (the “Forest Plan”) was adopted in 1990 and has since been amended a number of times. AR21933. Of particular relevance, it was amended in 1995 by the Pacific Fish Strategy (“PACFISH”) and the Inland Native Fish Strategy (“IN-FISH”). The Forest Plan also incorporates national policies on invasive plant management, but it had not been updated to reflect developments in these policies over the last twenty years.

A. PACFISH and INFISH

PACFISH and INFISH are very similar management directions, except that PACFISH protects the habitat of anadromous fish like Pacific salmon, steelhead and sea-run cutthroat trout, while IN-FISH protects the habitat of inland native fish like bull trout. See AR03520, 03540 (PACFISH); AR03821 (INFISH). PACFISH and INFISH are discussed largely interchangeably throughout this opinion.

PACFISH and INFISH instituted a series of Riparian Management Objectives (“RMOs”), which are benchmarks for achieving good habitat conditions for protected fish. AR03839. For example, the RMOs describe the appropriate water temperature and the optimal frequency of pools in streams. AR03841. PACFISH and INFISH also require the demarcation in each watershed of Riparian Habitat Conservation Areas (“RHCAs”). Id. For activities occurring in the RHCAs or that may potentially degrade them, PACFISH and INFISH set out specific standards and guidelines. See AR03843. These standards and guidelines include restrictions on timber harvesting, road construction and maintenance, grazing, recreation, mining, fire control, hydroelectric projects, and general land management. AR0384403850. Of particular relevance in the present litigation, PACFISH/INFISH standard RA-3 requires the Forest Service to “[a]pply herbicides ... in a manner that does not retard or prevent attainment of Riparian Management Objectives and avoids adverse effects on” listed anadromous or inland native fish. AR03849 (IN-FISH); AR03656 (PACFISH).

B. Invasive Species and Noxious Weed Management

The Forest Plan also implements the Forest Service’s national strategy, man*986dated by statute and Executive Order, to combat the spread of invasive species and noxious weeds. See AR21934. Before the Project FEIS and ROD, the Forest Plan was aligned with Region Six’s 1988 Record of Decision for Managing Competing and Unwanted Vegetation. AR21934-21935. This programmatic direction for managing invasive plants relied primarily on manual and mechanical treatment methods — e.g., hand-pulling weeds or using hand or power tools to root out or debilitate unwanted plants. See AR21950. Use of herbicides was only allowed as a last resort, and only on sites identified during environmental assessments conducted in the WallowaWhitman National Forest in 1992 and 1994. Id. That is, herbicides could not be used on new infestations of invasive plants. In addition, the herbicides approved for use under this programmatic direction were all developed before 1980. AR21935.

Federal policy on the prevention and eradication of invasive species has developed substantially since 1988. The Federal Noxious Weed Act of 1974 was amended in 1990 to reflect a more integrated and coordinated weed management approach. See 7 U.S.C. § 2814. By “integrated,” Congress meant the controlling of invasive plants through an interdisciplinary approach, including education, preventive measures, physical treatment methods,4 biological agents,5 herbicide use, cultural treatment methods,6 and “general land management practices such as manipulation of livestock or wildlife grazing strategies or improving wildlife or livestock habitat.” Id. § 2814(e)(4).

The Forest Service implemented this statutory direction through Forest Service Manual Chapter 2080, which was recently renumbered as Chapter 2900.7 The Manual currently requires all Forest Service invasive species management activities to include five components: prevention of the introduction and spread of invasive plants; early detection and rapid response (“EDRR”) to new infestations; control and management activities consistent with guidance from the National Invasive Species Council; restoration of native ecosystems; and collaboration with other state, local, and tribal governments and other interested parties. Efforts to prevent, control, and eliminate infestations should use an “integrated pest management approach,” which the Forest Service defines as involving biological, cultural, physical, and chemical (meaning herbicidal) techniques. Among other action items, Chapter 2900 instructs that the management of invasive plants should be incorporated into regional programmatic directions, as well as into local forest land and resource management plans.

In 1999, President Clinton issued Executive Order 13112, which directed federal *987agencies to prevent, detect, monitor, and control invasive plants and to restore native species. Exec. Order No. 13,112, 64 Fed.Reg. 6,183 (Feb. 3, 1999). It also created the inter-agency Invasive Species Council, the guidance of which is incorporated through Forest Service Manual Chapter 2900. In 2001, the Forest Service adopted a nationwide Guide to Noxious Weed Prevention Practices.8 The Guide provides a “toolbox” of weed prevention strategies that local forest rangers can incorporate into resource plans, permits, and contracts. In 2004, the Forest Service adopted a National Strategy and Implementation Plan for Invasive Plant Management,9 which reiterated the integrated management approach to preventing, controlling, and eradicating invasive plant species.

To account for these policy developments, as well as scientific and technological advances in treatment methods, the Regional Forester for Region Six adopted in 2005 a final environmental impact statement and record of decision regarding the prevention and management of invasive plants (the “Region Six 2005 FEIS” and “Region Six 2005 ROD,” respectively). These documents amended all forest plans in the region, including the Forest Plan of the Wallowa-Whitman National Forest, to include goals and objectives for invasive plant management, as well as specific prevention and planning standards. Further implementation of the updated policy was left, however, to be performed in the context of the individual forests. Significantly, the Region Six 2005 ROD approved a list of ten herbicides that could be used by forests in the region to control and eradicate invasive species. See AR17276. As the Regional Forester noted in the Region Six 2005 ROD, the “need for an updated list of herbicides for use on National Forest System lands in Region Six” was “one of the primary components of the need for” an updated invasive plants management direction. AR17277.

C. The Project

To align the Forest Plan with the Region Six 2005 ROD, the Wallowa-Whitman National Forest undertook an Invasive Plants Treatment Project, resulting in the Project FEIS and ROD. The Project FEIS recommends, and the Project ROD adopts, a plan to treat 22,842 acres of sites infested with invasive species, which is 0.9 percent of the Wallowa-Whitman National Forest’s land base. AR22625. Herbicides would be used in conjunction with other treatments, and the selection of particular herbicides and other treatment methods would be dependent on the species of invasive plant and its location in the forest. More than 90 percent of known sites with invasive species would be treated with herbicides. See id. All ten herbicides approved by the Region Six 2005 ROD would be available for use. Id. An EDRR policy was also adopted to allow the use of herbicides on new infestations of invasive plants. See AR22668-22669.

LOWD administratively appealed the Region Six 2005 ROD and subsequently the Project ROD. The Forest Service rejected both appeals. LOWD then brought this action under the APA, challenging the Project FEIS and Project ROD’s compliance with NFMA, NEPA, and the Clean *988Water Act. Because there are no disputed issues of material fact, LOWD moved for summary judgment on all counts. The Forest Service cross-moved for summary judgment, arguing among other grounds that LOWD had failed to raise its NFMA arguments in its administrative appeal before the agency. For the reasons that follow, the Forest Service’s exhaustion argument fails, as do LOWD’s NFMA and Clean Water Act claims. The court, however, grants LOWD’s motion for summary judgment on its NEPA claim concerning cumulative impacts.

DISCUSSION

I. Compliance with the National Forest Management Act

LOWD raises three arguments related to NFMA. First, it argues that the Project violates NFMA because it is not compliant with PACFISH/INFISH standard RA-3, which has been incorporated into the Forest Plan. Second, LOWD argues that the Forest Service did not adequately analyze the Project’s compliance with NFMA, specifically its compliance with the PACFISH/INFISH component of the Forest Plan. Third, LOWD argues that the Forest Service evaluated the PACFISH/INFISH standards at too broad a scale, thereby masking harmful effects of increased herbicide use at the stream-specific level.

As a preliminary matter, the Forest Service responds that LOWD did not exhaust its administrative remedies for its NFMA claims. For the reasons that follow, I hold that Plaintiff did adequately raise these concerns before the agency. I then consider LOWD’s three NFMA arguments in turn and find them to be without merit.

A. Exhaustion

The Forest Service argues that LOWD did not raise its NFMA compliance arguments in its administrative appeal of the Project ROD and FEIS and therefore failed to exhaust its administrative remedies.

The Ninth Circuit interprets the exhaustion requirement broadly. See, e.g., Natl. Parks & Conservation Ass’n. v. Bureau of Land Mgmt., 606 F.3d 1058, 1065 (9th Cir.2010). The purpose of the requirement is “to avoid premature claims and to ensure that the agency possessed of the most expertise in an area be given first shot at resolving a claimant’s difficulties.” Ida. Sporting Cong., Inc. v. Rittenhouse, 305 F.3d 957, 965 (9th Cir.2002). Thus, “[t]he plaintiffs have exhausted their administrative appeals if the appeal, taken as a whole, provided sufficient notice to the Forest Service to afford it the opportunity to rectify the violations that the plaintiffs alleged.” Native Ecosystems Council v. Dombeck, 304 F.3d 886, 899 (9th Cir.2002). In particular, “a claimant need not raise an issue using precise legal formulations, as long as enough clarity is provided that the decision maker understands the issue raised.” The Lands Council v. McNair, 629 F.3d 1070, 1076 (9th Cir.2010). An argument is not preserved, however, if “the connection between [the law at issue] and the concerns raised is too attenuated.” Great Basin Mine Watch, 456 F.3d at 967. Particularly where different laws might be implicated by the plaintiffs “general comments,” some identification of the law allegedly violated may be necessary. See id. Ultimately, “there is no bright-line standard as to when this requirement has been met[,] and we must consider exhaustion arguments on a case-by-case basis.” Rittenhouse, 305 F.3d at 965.

LOWD notes that in its administrative appeal of the Project ROD and FEIS, it explicitly incorporated all of the arguments it raised in its administrative appeal of the Region Six 2005 ROD. See AR22718-22719. The Forest Service, when it previously denied LOWD’s appeal *989of the Region Six 2005 ROD, dearly understood that appeal to raise NFMA compliance concerns.10 In addition, in a passage of LOWD’s administrative appeal of the Project ROD and FEIS entitled “NFMA Violations,” LOWD argued that: The ROD and FEIS violate the NFMA,

... applicable planning regulations, ... and the [Wallowa-Whitman National Forest] Forest Plan.... These statutes, regulations and plan provisions require the USFS to make site specific decisions consistent with the applicable land management plan, to maintain diversity if [stet] plant and animal communities, to maintain viable populations of native species, to evaluate impacts to and to protect endangered threatened and sensitive species, [and] to protect and monitor streams using field surveys.... The USFS’s failure to ... cite adequate science in support of its numerous unsupported assertions and conclusions regarding the safety of the proposed herbicide applications and the efficacy of the [project design features], is also a violation of 36 CFR Section 219.35(a)’s requirement to use the best available science when implementing forest plans.

AR22753. Thus, LOWD argued before the agency that the Project ROD and FEIS lacked the data and analysis to establish compliance with NFMA and the Forest Plan. This is sufficient to preserve LOWD’s procedural arguments that the Forest Service did not ensure NFMA compliance. These concerns are not by themselves, however, sufficiently specific to preserve LOWD’s argument that the subsequent Project ROD and FEIS violate PACFISH/INFISH standard RA-3 in particular.

LOWD does not appear to have mentioned standard RA-3 explicitly in its administrative appeal. Instead, LOWD appears to only have referenced PACFISH/INFISH once, in passing. See AR22746. It was not, however, necessarily required to do so in order to preserve the full scope of its NFMA claims. See, e.g., Great Basin Mine Watch, 456 F.3d at 965. The court finds that LOWD provided the Forest Service with adequate notice of its substantive concern that the Project would not comply with standard RA-3. In addition to explicitly appealing the Forest Service’s compliance with NFMA and the Forest Plan, LOWD repeatedly expressed concern that the use of a specific herbicide, picloram, could harm riparian habitats. For example, LOWD was worried that “[s]everal streams within the Wallowa-Whitman National Forest are already impaired, and the Proposed Action plans to use herbicides near these water bodies, both in spot and broadcast treatments (6,345 acres within PACFISH/INFISH Riparian Habitat Conservation Areas).” AR22746; see also id. (“Appellant is very concerned about effects from spraying in riparian areas, especially regarding the possible effects to Snake River Sockeye Salmon[.]”); AR22738. Collectively, these comments alerted the Forest Service to LOWD’s concern that the Project did not comply with the Forest *990Plan and that the use of herbicides near streams would harm habitats protected by PACFISH/INFISH. This is sufficient to preserve LOWD’s substantive NFMA argument.

In sum, while more precision and detail would have been preferable,11 LOWD did exhaust its administrative remedies for its NFMA compliance claims and may now bring those arguments before this court.

B. PACFISH/INFISH Standard RA-3

LOWD first argues that the Project does not comply with the Wallowa-Whitman National Forest’s Forest Plan, in particular standard RA-3 of PACFISH/INFISH. RA-3 requires the Forest Service to “[ajpply herbicides, pesticides, and other toxicants, and other chemicals in a manner that does not retard or prevent attainment of Riparian Management Objectives and avoids adverse effects on” listed anadromous or inland native fish. AR03849 (INFISH); AR03656 (PACFISH). Before addressing the merits of LOWD’s argument, I resolve the parties’ disagreement over the meaning of standard RA-3.

1. Interpretation of Standard RA-3

The Forest Service’s interpretation of a Forest Plan is entitled to substantial deference. See League of Wilderness Defenders Blue Mountains Biodiversity Project v. U.S. Forest Serv., 549 F.3d 1211, 1223 (9th Cir.2008) (citing Auer v. Robbins, 519 U.S. 452, 461-62, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997)). “[W]e have effectively treated forest plan directives as equivalent to federal regulations adopted under the APA, deferring to the Forest Service’s interpretation of plan directives that are susceptible to more than one meaning unless the interpretation is plainly erroneous or inconsistent with the directive.” Siskiyou Reg’l. Educ. Project v. U.S. Forest Serv., 565 F.3d 545, 555 (9th Cir.2009); see also Ecology Ctr. v. Castaneda, 574 F.3d 652, 661 (9th Cir.2009) (“[W]e defer to the Forest Service’s reasonable interpretation of the Forest Plan’s requirements.”).12

RA-3 includes two distinct requirements: that herbicides be applied in a manner that “does not retard or prevent the attainment of Riparian Management Objectives,” and that they be applied in a manner that “avoids adverse effects” on protected fish. The parties do not dispute the meaning of the first requirement. The RMOs are listed elsewhere in PACFISH/INFISH, and PACFISH and IN-FISH both define “retard.” See AR03840 (INFISH) (“[T]o slow the rate of recovery below the near natural rate of recovery if no additional human caused disturbance was placed on the system.”); AR03617 (PACFISH) (“Measurably slow recovery of any identified RMO feature ... that is worse than the objective level.”).

The parties dispute, however, what it means to “avoid adverse effects.” Each starts with the definition of “adverse effects” included in PACFISH and INFISH:

*991Adverse effects include short- or long-term, direct or indirect management-related impacts of an individual or cumulative nature, such as mortality, reduced growth or other adverse physiological changes, harassment of fish, physical disturbance of redds, reduced reproductive success, delayed or premature migration, or other adverse behavioral changes to listed anadromous fish at any life stage.

AR03612 (PACFISH); see also AR03949 (INFISH). The parties also both point to the following definition of “avoid” included in PACFISH: “Apply pre-project planning, best available technology, management practices, and scientific knowledge to eliminate known management induced impacts to the greatest extent practicable and minimize the risk of other potential impacts.” AR03612 (emphasis added).13 The Forest Service further notes that PACFISH defines “minimize” as “[ajpply pre-project planning, best available technology, management practices, and scientific knowledge to limit, to the greatest extent practicable, the magnitude, extent, and/or duration of an activity and/or effect.” AR03615 (emphasis added).

Based on these definitions, the Forest Service interprets “avoid adverse effects” as requiring it to use pre-project planning, best available technology, management practices, and scientific knowledge to limit or eliminate, to the greatest extent practicable, both known and unknown impacts that adversely affect listed anadromous and native inland fish and their critical habitat. This is a reasonable interpretation of RA-3 by the Forest Service and warrants deference.

LOWD would instead interpret “avoid” as prohibiting any action that may result in harm to protected fish. Pointing to the Fourth Circuit’s decision in City of Columbia v. Costle, 710 F.2d 1009 (4th Cir.1983), LOWD argues that the phrase “to the greatest extent practicable” should be understood “as requiring an agency to comply with legal requirements to the fullest extent to which it is legally capable of complying under the law.” Pl.’s Sur-Reply Opp’n Defs.’ Cross-Mot. Summ. J. 7 n. 7. The Fourth Circuit’s reasoning in Costle, however, sheds no light on the interpretation of standard RA-3. The Costle court interpreted 42 U.S.C. § 4655, which is part of the Uniform Relocation and Real Property Acquisitions Policies Act of 1970 (“URRPAPA”). For a state agency to receive federal funds for land acquisition projects, URRPAPA requires the state to apply URRPAPA’s land acquisition policies “to the greatest extent practicable under State law.” 42 U.S.C. § 4655(a)(1) (emphasis added). The Fourth Circuit concluded that “practicable,” in light of the adjoining phrase “under State law,” meant “to the fullest extent to which [the state agency] is legally capable of complying under state law.” Costle, 710 F.2d at 1013. In contrast, standard RA-3 and the related PACFISH/TNFISH definitions make no reference to legal limitations. Here, the phrase “to the greatest extent practicable” qualifies a reference to pre-project planning, best available technology, management practices, and scientific knowledge— all of which are tools, not mandates.

More significantly, LOWD’s proposed interpretation of RA-3 would conflict with the context of NFMA, which requires the Forest Service to manage the National Forest System lands not only to protect wildlife and wilderness, but also to enable sustainable use of the forests’ resources. See McNair, 537 F.3d at 990. Other uses of the Wallowa-Whitman National Forest, such as logging and grazing, will necessarily have some impact on fish habitat. If standard RA-3 were interpreted as pro*992hibiting all possible adverse effects, it would effectively foreclose other uses of the forest, which is contrary to the purpose of NFMA. Indeed, contrary to LOWD’s characterization,14 PACFISH and INFISH do not prioritize habitat conservation and restoration at the expense of all other interests. As the INFISH ROD noted, “we recognize the selection of this alternative will concern many people who felt this alternative provided either too much or not enough protection.” AR03826. The adopted strategy “represents that agencies’ judgment of the best balance among competing interests: it responds to the need to provide a high level of protection for anadromous fish habitat, without unnecessarily restricting existing contracts, permits and other authorizations, management flexibility, or the flow of goods and services.” AR03524-03525 (PACFISH).15

The Forest Service’s interpretation of RA-3 aligns with the context and purpose of NFMA, the Forest Plan, and PACFISH/INFISH. The court therefore defers to the Forest Service’s interpretation of RA-3’s “avoid adverse effects” standard as requiring the Forest Service to use management practices, scientific knowledge, and best available technology to limit or eliminate harms to protected fish to the greatest extent practicable.

2. Application of Standard RA-3

This brings us to the question of whether the Forest Service acted arbitrarily or capriciously when it concluded that the Project would comply with the Forest Plan, including standard RA-3. As the Ninth Circuit explained in McNair:

[T]he Forest Service must support its conclusions that a project meets the requirements of the NFMA and relevant Forest Plan with studies that the agency, in its expertise, deems reliable. The Forest Service must explain the conclusions it has drawn from its chosen methodology, and the reasons it considers the underlying evidence to be reliable. We will conclude that the Forest Service acts arbitrarily and capriciously only when the record plainly demonstrates that the Forest Service made a clear error in judgment in concluding that a project meets the requirements of the NFMA and relevant Forest Plan.

537 F.3d at 994. Applying this standard to the Project FEIS and ROD, I hold that the Forest Service made no clear error in judgment in concluding that the Project would comply with NFMA.

LOWD argues that the Forest Service’s own data demonstrates that the Project will violate standard RA-3. In particular, LOWD points to two site-specific models of the aerial application of the herbicide picloram that are discussed in the Project FEIS. See AR22277. The modeling suggested a “hazard quotient” greater than one, which means that the expected exposure concentration for the herbicide was more than negligible. See AR22268, 22277.16 Based on this modeling, the Pro*993ject FEIS states that the aerial application of picloram “may have adverse effects to steelhead, salmon and bull trout populations directly downstream of the treatment site.” AR22278. LOWD relies on this and related statements to argue that the Project will adversely affect anadromous and inland native fish and therefore does not comply with standard RA-3.

I do not agree with this argument for three reasons. First, the statements that the Project “may have adverse effects” on fish were related to a “likely to adversely affect” finding under the Endangered Species Act (“ESA”). A “likely to adversely affect” finding under the ESA does not equate with a violation of the “avoid adverse effects” standard of RA-3. Under the ESA, a federal agency must consider whether a proposed action “may affect listed species or critical habitat.” 50 C.F.R. § 402.14(a). If it concludes that the action may affect protected species, it must initiate a consultation process with other agencies. Id. The triggering level for this ESA consultation process is very low. Because it is illegal for a federal agency to “take” any species protected under the ESA, 16 U.S.C. § 1538(a)(1)(B), any risk of a taking requires the agency to comply with the ESA’s consultation and permitting requirements.17 Thus, if the Forest Service could not say with certainty that no harm would result to protected species, it was required by the ESA to initiate the consultation process. This was the import and the effect of the Project FEIS’s finding that the Project “may have adverse effects” on protected fish.

LOWD, however, argues that such statements represent concessions that the Project would violate RA-3. This incorrectly assumes RA-3 directs a zero-tolerance standard. The court has accepted the Forest Service’s interpretation of RA-3, under which the bare possibility of harm is not a per se violation of RA-3.

Second, the Forest Service gave multiple reasons for why it determined that the picloram models overstated the risk of harm. The two sites modeled were selected because they have higher rainfall and thus “represent a worst case of all known sites.” AR22277. The model also assumed significant levels of precipitation to account for the worst case amount of rainfall and thus the worst case amount of runoff into streams. Id. Further, the sites modeled do not have protected fish directly adjacent to the treatment area. Id. That there might be some run-off in these worst-case scenarios thus does not mean that fish would necessarily come into contact with that run-off, particularly given that the “[tjransitory water quality impact, if any, would be limited to the point of contact with water and not an entire stream reach.” AR22279.

Third, LOWD discounts the management practices the Forest Service adopted in order to limit any adverse impacts from the Project. The Project includes mandatory “project design features” (“PDFs”) and buffer zones around treatment sites. See AR22651-22652, AR22666-22668. The PDFs require pre-project planning and restrict where, when, and how herbicides are applied. See AR22652-22665. For example, the PDFs require a 300-foot buffer zone between aerial applications of herbicides and any lakes, wetlands, and perennial wet intermittent streams. See AR22666-22668.18 This buffer zone was not accounted for in the site-specific mod*994eling. See AR22279. The Forest Service concluded that these PDFs and buffer zones would further reduce the risk identified by the picloram models to a negligible amount. That conclusion is not unreasonable. See Hapner v. Tidwell, 621 F.3d 1239, 1248-49 (9th Cir.2010) (Forest Service could rely on proposed mitigation measures for concluding that the project would not violate Forest Plan standards despite anticipated short-term harms).

In sum, the Forest Service explained both why it did not think the identified risk was significant and why it thought PDFs, buffers, and other measures would reduce that risk to a negligible amount, thereby limiting or eliminating adverse impacts to the greatest extent practicable.19 Given this record and the Forest Service’s explanation, it was not a clear error of judgment for the Forest Service to conclude that the Project complied with standard RA-3.

C. Ensuring Consistency with the Forest Plan

LOWD next argues that the Forest Service did not ensure that the Project was consistent with the Forest Plan or adequately analyze whether it was consistent. The Project FEIS states, “[t]his project would be consistent with PACFISH/INFISH standards and guidelines, and not retard or prevent attainment of riparian management objectives.” AR21942. LOWD notes that the Project FEIS only lists the INFISH conservation strategies in an appendix and does not otherwise specify how the Project complies with PACFISH/INFISH. See AR21934. It argues that something more is required.

LOWD’s argument “seeks too much from the [EIS].” City of Garmel-by-the-Sea, 123 F.3d at 1151. NFMA does not require the Forest Service to analyze explicitly each and every standard of the Forest Plan when approving a new project. LOWD suggests that NEPA does, but that appears to be a novel interpretation of the NEPA regulations. NEPA requires an EIS to contain the specific items outlined in 42 U.S.C. § 4332(2)(C), as well as a “full and fair discussion of environmental impacts.” McNair, 537 F.3d at 1001. These enumerated requirements do not include an analysis of the proposed action’s compliance with other laws. NEPA’s implementing regulations do require an EIS to “state how alternatives considered in it and decisions based on it will or will not achieve the requirements of sections 101 and 102(1) of [NEPA] and other environmental laws and policies.” 40 C.F.R. § 1502.2(d). LOWD, however, has not identified, and this court has not been able to discover, any ease interpreting this regulation as requiring an EIS to discuss fully every applicable environmental law.20 Indeed, *995such a reading would directly conflict with the repeated admonitions that “[ejnvironmental impact statements shall be kept concise and shall be no longer than absolutely necessary to comply with NEPA and with these regulations.” 40 C.F.R. § 1502.2(c).

LOWD points to the definition of “significantly’ in the regulations, which requires the agency to consider “[w]hether the action threatens a violation of Federal, State, or local law or requirements imposed for the protection of the environment.” 40 C.F.R. § 1508.27(b)(10). This regulation, however, relates to the preliminary question of whether an EIS is required. NEPA directs federal agencies to prepare an EIS for all “major Federal actions significantly affecting the quality of the human environment.” 42 U.S.C. § 4382(2)(C) (emphasis added). The regulations specify that the meaning of “significantly” in this context is defined in 40 C.F.R. § 1508.27. See 40 C.F.R. § 1502.3.21 In other words, the provision to which LOWD points requires agencies to consider potential conflicts with environmental laws when determining whether a proposed action will “significantly” affect the quality of the human environment, which would trigger the requirement to prepare an EIS under 42 U.S.C. § 4332(2)(C). It does not describe what that EIS must then contain.22

The court does not agree with Plaintiff that NEPA requires something more than the Forest Service’s conclusion in the Project FEIS that the Project would comply with PACFISH/INFISH. To hold otherwise would impose a substantial requirement on agencies that is not clearly set out in statute or regulation, and the court is reluctant to do so in the absence of such direction. Instead, this court will consider LOWD’s argument that the Forest Service did not adequately ensure compliance with *996the Forest Plan as an alleged NFMA violation. I must sustain the Forest Service’s conclusion that the Project complies with the Forest Plan unless the agency entirely failed to consider an important aspect of the problem or its conclusion was contrary to the evidence before the agency. See McNair, 537 F.3d at 987. Neither condition occurs here.

First, it appears that the Forest Service did consider the substance of the PACFISH/INFISH RMOs, even if it did not explicitly analyze them. In the ESA discussion, the Project FEIS describes and evaluates two lists of habitat criteria: the “primary constituent elements” of designated critical habitat and the National Marine Fisheries Service’s “matrix of pathways and indicators.” See AR22280. The “primary constituent elements” and the “matrix of pathways and indicators” are, like the PACFISH/INFISH RMOs, lists of habitat conditions critical for the survival of protected fish. All three sets of habitat conditions substantively overlap with one another. For example, the “primary constituent elements,” the “matrix of pathways and indicators,” and the RMOs all require consideration of water quality and temperature; the prevalence of “large woody debris” or other natural cover to streams; and the frequency of pools in streams. Compare AR22280 with AR03841. In explicitly analyzing the Project’s compliance with the “primary constituent elements” and the “matrix of pathways and indicators,” see AR22281-22287, the Forest Service analyzed the impact of the Project on critical fish habitat, which is the substantive concern behind the PACFISH/INFISH RMOs. Thus, this court cannot say that the Forest Service entirely failed to consider an important aspect of the problem.

Nor has LOWD pointed to any conclusion regarding NFMA compliance that this court could conclude was contrary to the evidence before the agency. As explained above, the picloram modeling data does not necessarily establish a violation of standard RA-3. LOWD argues that the Forest Service did not adequately address the water temperature RMO. The Forest Service did consider the impact of the Project on water temperature and concluded that “[cjhanges in water temperature resulting from herbicide use to control invasive plants would be negligible to nonexistent.” AR22285. LOWD would further require the Forest Service to demonstrate that the Project will not “retard” improvements in water temperature in order to establish the Project’s compliance with RA-3. Neither NFMA nor NEPA require the Forest Service to provide such a detailed analysis of individual Forest Plan standards in its NEPA documents.

In sum, the Forest Service has not failed to consider an. important aspect of the problem, and its conclusion that the Project complies with NFMA and the Forest Plan does not run counter to the evidence in the administrative record. There are, therefore, no grounds for concluding that the Forest Service acted arbitrarily or capriciously in asserting NFMA compliance.

D. The Question of Scale

LOWD next argues that the Forest Service should have analyzed the Project’s consistency with PACFISH/INFISH at a “site-specific level,” by which it appears to mean each of the 1,740 infested sites identified for treatment. Hr’g Tr. 39, Jan. 23, 2012. Instead, the Forest Service analyzed the effects of the Project at a “watershed” scale, concluding that treatments within each watershed would be sufficiently scattered and isolated so as to dilute any adverse impacts. See, e.g., AR22118. LOWD argues that this broader watershed-scale analysis could mask significant degradations of particular streams. The Forest Service justifies its choice of scale by pointing to the language of PACF*997ISH/INFISH, which describes the RMOs as “landscape[ ]scale” and as applicable “to 3rd to 6th order watersheds.” AR0383903840 (INFISH).23 Given the language of PACFISH/INFISH, it was not a clear error of judgment for the Forest Service to evaluate PACFISH/INFISH standards primarily on a fifth-field watershed scale.24

Both PACFISH and INFISH explain that “[a]Il of the described [RMO] features may not occur in a specific segment of stream within a watershed, but all generally should occur at the watershed scale for stream systems of moderate to large size (3rd to 7th order streams).” AR03644 (PACFISH); see also AR03839 (INFISH, with the modification that all RMOs should occur in “3rd to 6th order” watersheds). The application of RMOs is not a simple checklist of requirements, but involves “thorough analysis”:

That is, if the objective for an important feature such as pool frequency is met or exceeded, there may be some latitude in assessing the importance of the objectives for other features that contribute to good habitat conditions.... The goal is to achieve a high level of habitat diversity and complexity through a combination of habitat features, to meet the life-history requirements of the fish community inhabiting a watershed. AR03840 (INFISH); see also A03644 (PACFISH). In light of this language, the Forest Service’s interpretation of PACFISH/INFISH as allowing analysis at the watershed scale was not unreasonable.

In arguing that the Forest Service was nonetheless required to analyze RMO attainment on a smaller scale, LOWD relies heavily on Pacific Coast Federation of Fishermen’s Association v. National Marine Fisheries Service, 71 F.Supp.2d 1063 (W.D.Wash.1999) (“PCFFA I ”), vacated in part by Pac. Coast Fed’n. of Fishermen’s Ass’n., Inc. v. Nat’l. Marine Fisheries Serv., 265 F.3d 1028 (9th Cir.2001) (“PCFFA II”). In the PCFFA cases, the plaintiffs sued the National Marine Fisheries Service (“NMFS”) under the ESA, challenging the NMFS's conclusion that continued logging in the Umpqua River Basin would not jeopardize the survival of the Umpqua cutthroat trout. In particular, the . plaintiffs contended that the NMFS’s biological opinion did not adequately ensure that the action agencies (the Forest Service and the Bureau of Land Management) were complying with the Aquatic Conservation Strategy (“ACS”), a management plan that is very similar to PACFISH/INFISH.25 PCFFA I, 71 F.Supp.2d at 1065-66. According to the district court, the NMFS was required by the Northwest Forest Plan and its own *998prior biological opinion to ensure ACS compliance on four spatial scales: “regional, province (river basin), watershed, and site (or project).” Id. at 1069, 1073.26 Thus, the court held that the NMFS acted arbitrarily and capriciously when it measured ACS compliance only at the watershed scale. Id. at 1073. The Ninth Circuit agreed that “[appropriate analysis of ACS compliance is undertaken at both the watershed and project levels.” PCFFA II, 265 F.3d at 1036. It clarified, however, that the Northwest Forest Plan and the ACS do not dictate the proper scale for evaluation; they only “explain that spatial levels should be considered and that watershed consistency is a primary goal.” Id.

The two PCFFA cases do not require the Forest Service to analyze PACFISH/INFISH compliance for each of the 1,740 sites proposed for treatment, as LOWD seems to suggest. First, the PCFFA cases are not clearly analogous to the present litigation. In PCFFA the ACS was a proxy for ESA compliance, which imposes more stringent requirements than NFMA. There was also a concern in PCFFA that the NMFS had enlarged the scale of its analysis purposefully to gloss over the habitat degradations anticipated from the proposed projects. PCFFA I, 71 F.Supp.2d at 1067-68.27 No such evasive tactics are apparent here. Given these and other differences, the applicability of the PCFFA cases is not as clear as LOWD suggests.

Second, the Ninth Circuit in PCFFA II did not hold that the agency had to evaluate ACS compliance on each of the four spatial scales identified. Rather, it explained that ACS compliance should be analyzed at both the watershed and the project level, and that the NMFS erred because it had not considered the cumulative effects of the discrete projects at the watershed level. See 265 F.3d at 1036-37. PCFFA involved twenty-three proposed timber sales, each considered a separate project. It appears that the agencies in PCFFA did not consider the cumulative impact of all twenty-three projects at the watershed level, only the impacts of each individual timber sale at the watershed level. Here, on the other hand, there is only one project, and it appears that the watershed analysis did consider the cumulative effects of treating all infested sites within each watershed. See, e.g., AR21392 (FWS BiOp); AR21568 (NMFS BiOp).28

*999Particularly in light of the language of PACFISH/INFISH, it was not a clear error for the Forest Service to choose to analyze- the impact of the Project primarily at a watershed scale. In sum, the Forest Service’s analysis and conclusion that the Project will comply with NFMA, including the requirements of the Forest Plan and PACFISH/INFISH, was not arbitrary or capricious.

II. Compliance with the National Environmental Policy Act

LOWD argues that the Project FEIS did not comply with the requirements of NEPA because it: (1) was based on an unreasonably narrow statement of purpose and need; (2) failed to consider a reasonable range of alternatives; and (8) did not adequately analyze cumulative impacts. This court reviews the Forest Service’s compliance with NEPA under the APA’s arbitrary or capricious standard. McNair, 537 F.3d at 987.

A. Statement of Purpose and Need

NEPA requires the environmental impact statement to “briefly specify the underlying purpose and need to which the agency is responding in proposing the alternatives including the proposed action.” 40 C.F.R. § 1502.13. The Ninth Circuit “has afforded agencies considerable discretion to define the purpose and need of a project.” Friends of Se. Future v. Morrison, 153 F.3d 1059, 1066 (9th Cir.1998).

The identified need in the Project FEIS is “to arrest and reverse the spread of invasive plants to help restore ecological integrity on infested National Forest System land.” AR21931. More specifically, the Project FEIS explains that:

Invasive plant control is needed to maintain or improve the diversity, function, and sustainability of desired native plant communities and other natural resources that can be adversely impacted by invasive plant species. Specifically, there is an underlying need on the Forest to: (1) implement treatment actions and site restoration to contain, control and eradicate the extent of invasive plants at existing inventoried sites, and (2) rapidly respond to new or expanded invasive plant sites as they are detected in the future. Without action, invasive plant populations will become increasingly difficult and costly to control and will further degrade forest and grassland ecosystems. Untreated infested areas will also contribute to the spread of invasive plants onto neighboring lands.

AR21932. The Forest Service has mapped 40 invasive plant species on 1,740 sites in the Wallowa-Whitman National Forest, encompassing approximately 22,840 acres. AR22623. These plant species are displacing native plants, degrading fish habitats, increasing fire hazards, and adversely affecting scenic beauty and recreational opportunities. AR21924. The identified purpose in the Project FEIS is therefore “to bring the treatment program into compliance with the new standards and allow for effective treatments on all sites currently known, and those that may be detected during the life of the project.” AR21931. The Project FEIS was prepared specifically to allow the WallowaWhitman National Forest to use the new methods, including the new herbicides, approved by the Region Six 2005 ROD. Id.

The purpose and need statements are reasonable. The Project FEIS describes the harm posed by invasive plants to the *1000Wallowa-Whitman National Forest’s ecosystem, as well as the importance of eliminating existing invasive plants and responding rapidly to new infestations. AR21924, 21931-21932. It explains how the existing management strategy, which focused on prevention and limited the use of herbicides, had proven ineffective at controlling invasive plants. AR21930. It also notes the need to update the existing management strategy in light of the new regional direction. See id. In particular, the Project FEIS acknowledges that the new herbicides approved in the Region Six 2005 ROD “offer many advantages over the more limited set of herbicides allowed previously, including greater selectivity, less harm to desired vegetation, reduced application rates, and lower toxicity to animals and people.” Id.

The Forest Service specifically explained its decision to focus this project on treatment rather than prevention. It noted that weed prevention strategy was addressed by the Wallowar-Whitman Nar tional Forest Weed Prevention Practices and Analysis Guidelines; that the Region Six 2005 ROD had already incorporated additional, concrete weed prevention measures into the local Forest Plan; and that prevention efforts are also incorporated into other management plans, such as the Travel Management Plan, the Timber Management Plan, and management of grazing allotments. AR21932. In light of the Forest Service’s explanations, it was not arbitrary or capricious for the Forest Service to focus the Project’s purpose and need statement on treatment rather than prevention. Cf. Nw. Coal. for Alts. to Pesticides v. Lyng, 844 F.2d 588, 591-94 (9th Cir.1988) (agency could define scope of noxious weed management project to focus on the use of herbicides to control and eradicate weeds instead of on prevention measures).

LOWD argues that the purpose is unreasonably narrow because it assumes herbicides will be used as the primary means for controlling and eradicating invasive plants. LOWD points to a caveat, which follows the statement of purpose in the Project FEIS, that “[i]nitial treatments will rely heavily on herbicides, but the goal of this project is to eventually, as invasive plant objectives are met, reduce the use of herbicides.” AR 21932. LOWD interprets this statement as modifying the Project’s purpose to require the “heavy” use of herbicides, at least initially. Rather, this statement appears to refer to a specific goal incorporated into the Forest Plan by the Region Six 2005 ROD: to “[rjeduee reliance on herbicide use over time.” AR17295. The statement does not restrict the purpose itself to the heavy use of herbicides, but instead acknowledges that the heavy use of herbicides is not a viable long-term solution.

Further, the remainder of the Project FEIS demonstrates that the Project’s purpose is not restricted to near-exclusive reliance on herbicides. Most notably, all three action alternatives considered in the Project FEIS incorporate a range of treatment methods. These are summarized in a table of “common control measures,” which specifies for each invasive plant species the most effective combination of manual, mechanical, chemical, and/or biological control methods of treatment. The proposed treatments in the table of common control measures do incorporate herbicides, but they also recommend other treatment methods and describe the conditions under which those methods would be effective.29 For example, the following is the prescribed treatment of Scotch thistle:

*1001Cutting and mowing can be effective when combined with revegetation of native species. Repeated mowing, in combination with other management methods, often is necessary for long-term control. Manual removal is effective when entire aboveground plant growth is removed. Herbicide is the most effective control.

The table then suggests using, in order of preference, the herbicides picloram or clopyralid, chlorsulfuron, or metsulfuron. AR21978. Another illustrative example of proposed treatment is that for yellow star thistle, a particularly challenging invasive species:

Hand-pull small patches or maintenance programs where plants are sporadically located. Otherwise, mechanical treatment to contain and herbicides in combination with other methods to control or eradicate.

• Biocontrol available (see Appendix E).

• Revegetate high priority sites if needed with desirable species.

Aerial [application of herbicides] proposed for large, remote sites.

1. Clopyralid or Picloram.

2. Glyphosate.

AR21980. These common control measures align with the Forest Service’s “integrated weed management” approach to invasive species, which includes the use of herbicides in conjunction with other treatment methods. Particularly in light of the Region Six 2005 ROD, it was not unreasonable for the Project’s purpose and need to reference the use of herbicides in invasive plant management. Cf City of Carmelby-the-Sea, 123 F.3d at 1155-57 (approving purpose and need statement that specified motor vehicle traffic considerations where agency explained the need for the traffic improvement and the traffic concerns were not the only factors weighed in the agency’s analysis).

The Ninth Circuit’s reasoning in Westlands Water District v. U.S. Department of the Interior, 376 F.3d 853 (9th Cir.2004), is instructive. In Westlands, the stated purpose for the disputed project was “to restore and maintain the natural production of anadromous fish on the Trinity River mainstem downstream of Lewiston Dam.” Id. at 866. Plaintiffs objected that this purpose was biased towards mandating increased water flow (e.g., water spills over the dam) to the exclusion of other non-flow measures. Id. at 867. The Ninth Circuit rejected this argument for three reasons. First, the language of the purpose statement did not itself limit consideration of non-flow measures. Id. Likewise here, the statement of purpose and need does not require exclusive or even primary reliance on herbicides.

Second, it was not arbitrary or capricious for the agency in Westlands to consider that habitat restoration (enabled by increased water flow) would most effectively restore the salmon and steelhead runs. Id. Similarly here, given the analysis in both the Region Six 2005 FEIS and the Project FEIS that the prior invasive plant management policy had not worked, it was not arbitrary or capricious for the Forest Service to conclude that herbicide use would be an integral part of an effective treatment regime.

Third, the Westlands court noted that all the alternatives considered by the agency did include some non-flow measures. Id. at 867-68. The same is true here, where all action alternatives incorporate non-herbicide treatment methods, even though the focus of the Project FEIS is on weighing the risks and benefits of different methods of applying herbicides. As the Ninth Circuit concluded in Westlands, “[t]he Statement of Purpose and Need reasonably defined the objectives of the project; the preparers did not arbitrarily or *1002capriciously narrow the scope of the Statement.” Id. at 868.

B. Reasonable Range of Alternatives

1. Standard

An environmental impact statement must “Vigorously explore and objectively evaluate all reasonable alternatives” to the proposed action, including “reasonable alternatives not within the jurisdiction of the lead agency.” 40 C.F.R. § 1502.14. This analysis of the comparative environmental impacts of alternative courses of action “is the heart of the environmental impact statement.” Id. An agency need not, however, “consider an infinite range of alternatives, only reasonable or feasible ones.” City of Carmel-by-the-Sea, 123 F.3d at 1155. Nor need it consider “alternatives which are not significantly distinguishable from alternatives actually considered, or which have substantially similar consequences.” Headwaters, Inc. v. Bureau of Land Mgmt., 914 F.2d 1174, 1181 (9th Cir.1990). The agency does not need to consider remote or speculative alternatives or alternatives not reasonably related to the project’s purpose. Westlands, 376 F.3d at 868; see also Headwaters, 914 F.2d at 1180 (agency need not “consider alternatives which are infeasible, ineffective, or inconsistent with the basic policy objectives for the management of the area”). Finally, there is no minimum number of alternatives that must be considered, as long as the EIS “discusses in detail all the alternatives that were feasible and briefly discusses the reasons others were eliminated.” Laguna Greenbelt, Inc. v. U.S. Dep’t. of Transp., 42 F.3d 517, 524 (9th Cir.1994) (NEPA analysis adequate where agency considered only no action, proposed action, and one slight variation of proposed action).30

This court reviews the selection of alternatives considered by the agency under the APA’s arbitrary and capricious standard. McNair, 537 F.3d at 987. “The touchstone for our inquiry is whether an EIS’s selection and discussion of alternatives fosters informed decision making and informed public participation.” Westlands, 376 F.3d at 868 (quoting California v. Block, 690 F.2d 753, 767 (9th Cir.1982)) (quotation marks omitted).

2. The Project FEIS

The Project FEIS evaluated four alternatives in detail. Alternative A, the “no action” alternative, represents the previous management direction, which allowed the use of herbicides to treat invasive species *1003only after manual and mechanical treatment methods had failed. New infestations of invasive species could not be treated with herbicides absent a new NEPA analysis. See AR21950.

Alternative B, the “proposed action,” would use herbicides in conjunction with manual, mechanical, and biological methods to treat invasive species. AR2195121952. Depending on the site location, herbicides would be applied through selective means (ie., wicking, wiping, or stem injection); through broadcast or spot spraying; or through aerial spraying. AR21952. New infestations would be treated under an Early Detection Rapid Response (“EDRR”) policy. Id. With EDRR, no new NEPA analysis would be required as long as a treatment method approved in the FEIS could be used (with the exception of the aerial application of herbicides, which would not be allowed without a separate NEPA review). AR22005-22006. PDFs restrict how different treatment methods (particularly herbicides) can be used so as to minimize adverse impacts. AR2198022004. Alternative B also establishes planning requirements, including the development of site prescriptions, compliance monitoring, and post-treatment monitoring. AR22007-22008.

Alternative C is similar to Alternative B, except that it does not allow broadcast application of herbicides in riparian regions. This alternative was considered due to concerns that the use of herbicides in sensitive riparian zones would pose a greater risk to wildlife. AR22009. Alternative D is also similar to Alternative B, except that it would not allow the aerial application of herbicides. This alternative was considered to address concerns about the greater environmental and health risks posed by aerial spraying. AR22010.

The Forest Service acknowledged but declined to pursue six additional action alternatives. First, it considered using herbicides only on high priority sites, which it defined as sites with a high potential for spread or that hosted high-priority invasive plant species. The Forest Service rejected this alternative because it would only exclude 300 acres (out of nearly 21,-000) currently proposed for herbicide treatment. AR22011. In other words, this alternative would not have significantly differed from the alternatives considered in depth.

Second, the Forest Service considered restricting other uses of the WallowaWhitman National Forest to prevent the spread of invasive plants. The Forest Service rejected this alternative because it cannot exclude other uses like grazing and recreation, given the multiple-use policy of the National Forest System. It also noted that the purpose and need of the project is to treat existing invasive plants, which prevention alone cannot accomplish. Id.

Third, the Forest Service considered and rejected a treatment regime that would not use herbicides, based on its past experience that invasive plant species cannot be successfully contained and eradicated without the use of herbicides. This alternative would therefore not have met the identified purpose and need. AR22011-22012.

Fourth, the Forest Service considered applying the current management direction forest-wide, lifting the restrictions on which sites could be treated with herbicides. It would have retained the use of herbicides as a last resort, and it would not have incorporated the newly approved herbicides. The Forest Service rejected this alternative as having been ineffective in the past. AR22012.

Fifth, the Forest Service considered banning the use of herbicides in riparian or other sensitive areas. This alternative would have allowed herbicide treatment *1004for less than half of the infected acres. The Forest Service estimated that this approach would be significantly more expensive and less effective, based on an assumption that herbicide treatments have an 80 percent effectiveness rate, while manual and mechanical treatments only have a 25 percent effectiveness rate. The Forest Service also was concerned that this limitation would scatter treatment ineffectively, and that some of the more lightly treated areas would become “safe harbors” for invasive species. AR2201222015.

Finally, the Forest Service considered and rejected evaluating herbicides in addition to those approved in the Region Six 2005 ROD. This process would have been too costly and time-consuming and was therefore not feasible. AR22016.

3. Proposed Alternatives

LOWD proposes five specific alternatives that it argues the Forest Service should have considered before deciding on the proposed course of action. LOWD also challenges the Forest Service’s assumption that herbicides have an 80 percent effectiveness rate in eradicating invasive plants, while manual and mechanical methods have only a 25 percent effectiveness rate. Because the 80/25 figure influenced the Forest Service’s evaluation of alternatives, I address it first before turning to LOWD’s proposed alternatives.

(a) The Relative Rate of Effectiveness

In its comments to the draft EIS, LOWD challenged the Forest Service’s assumption in its cost-benefit analyses that herbicide treatments were 80 percent effective while non-herbicide treatments were only 25 percent effective. The Forest Service responded in the FEIS as follows:

There is no way to precisely estimate exactly how invasive plant populations will respond to each treatment.... Practitioners report that the same treatment on a similar site may be more or less effective. The 80 percent estimate is based on anecdotal evidence consistent with other invasive plant projects across the Pacific Northwest Region (Olympic, Gifford Pinchot, Mt. Hood). The 80 percent effectiveness assumption illustrates that some repeated treatment will be necessary to accommodate skips and regrowth. In contrast, nonherbicide treatments usually require a greater number of repeated visits and a longer time before invasives are controlled; and in many cases, eradication is impossible. The estimates used in the [draft EIS] are intended to demonstrate these concepts and provide [a] way to compare the effects of the most ambitious treatments under each alternative..... Erickson (personal communication 2006) thought that 25 [percent] effectiveness was probably close if even a little high with relation to treatment of newly established invasive species on the Forest using manual control methods only.

AR22467-22468. The telephone conversation with Vicky Erickson is documented in the record.31 AR21775. Erickson spoke anecdotally from her experience with the Umatilla National Forest, and she estimated that manual methods alone proved to be about 25 percent effective on newly established invasive species in the forest. Id. The Forest Service also points to the explanation provided at the Region Six level for the 80 percent effectiveness figure for herbicide treatments. AR00337-00338. That explanation acknowledges that “[t]he 80% estimate is not precise and the range *1005of results varies widely,” but notes that “[rjeports of effectiveness in the field generally approach or exceed 80%.” AR00337. It describes three such field reports from around the Pacific Northwest. Id.

LOWD argues that anecdotal evidence alone is insufficient to support a figure that drives the Forest Service’s cost-effectiveness analysis. The Forest Service counters that it provided an explanation for these figures and that it was entitled to rely on the expertise of its forest rangers. The court agrees with the Forest Service. The basis for the Forest Service’s figure is thin, but courts must defer to agencies in their fields of expertise, particularly regarding their selection of methodology. See McNair, 537 F.3d at 993; Hells Canyon Alliance v. U.S. Forest Serv., 227 F.3d 1170, 1184-85 (9th Cir.2000) (deferring to the Forest Service’s reliance on its own expertise to develop and apply method of analysis, even though plaintiff considered the approach to be “primitive”); Moronga Band of Mission Indians v. Fed. Aviation Admin., 161 F.3d 569, 577-78 (9th Cir. 1998) (deferring to agency’s choice to use noise measurements gathered at different but analogous location); Inland Empire Pub. Lands Council v. Schultz, 992 F.2d 977, 981 (9th Cir.1993) (“We defer to agency expertise on questions of methodology .... ”); see also League of Wilderness Defenders Blue Mountains Biodiversity Project v. Allen, 615 F.3d 1122, 1131 (9th Cir.2010) (“Our highest deference is owed to the Forest Service’s technical analyses and judgments within its area of expertise. ... We went en banc [in McNair] to foreclose precisely this type of second-guessing of the Forest Service.”). The Forest Service is not limited to methodologies that are peer reviewed or published; rather, it may rely on information “derived from field testing and practical experience.” The Lands Council v. Martin, 529 F.3d 1219, 1226 (9th Cir.2008). Further, LOWD has not identified any contrary evidence or literature that would call into doubt the approximations used by the Forest Service in its attempt to quantify the different costs of these two treatment methods.

I turn now to the alternatives that LOWD argues should have been considered in the Project FEIS.

(b) The “Revised No Action” Alternative

LOWD suggests an alternative modeled closely after the prior management direction, which limited the use of herbicides to a tool of “last resort.” LOWD argues the Forest Service should have considered combining the old direction’s de-emphasis on herbicide treatment with the new proposal’s prevention measures, PDFs, newly approved herbicides, and EDRR program.

The use of herbicides only as a tool of last resort was rejected at the programmatic level by the Region Six 2005 ROD. See AR17259, 17280; see also AR21944 (Project FEIS).32 As the Project FEIS explains, a focus on prevention and on manual and mechanical control methods could slow the rate of spread of invasive plants, but it would not effectively treat those already in the forests, which is the stated purpose of this Project. AR2201122012. An agency need not consider either ineffective alternatives or alternatives that would not be compatible with the stated purpose of a project or with overarching management policy. See Nw. Coal. for *1006Alts. to Pesticides, 844 F.2d at 594. The Forest Service’s omission of an alternative that would continue the policy of using herbicides only as a last resort was therefore not arbitrary or capricious.

(c)The “Balanced Treatment” Alternative

LOWD argues that the Forest Service should have considered an alternative that focused on a more balanced combination of treatment methods, in which herbicides would be neither a first nor a last resort. This proposal is not clearly differentiated, however, from what the Forest Service did consider. All three action alternatives evaluated in depth in the Project FEIS assumed that the Forest Service would use a range of treatment methods (not always including herbicides) tailored to the species of invasive plant and its geographic location within the forest. The table of common control methods, described above, illustrates that herbicides are neither the first nor the last line of defense. Rather, the common control methods embody the Forest Service’s integrated weed management approach, which treats herbicides as a necessary but not independently sufficient component of an effective invasive plants management program. An agency need not include alternatives that are not significantly different from alternatives actually considered. See Laguna Greenbelt, 42 F.3d at 525. It was therefore not arbitrary or capricious for the Forest Service to omit an alternative that balanced the use of herbicide and non-herbicide treatment methods in an undifferentiated way from the alternatives in fact considered in the Project FEIS.

(d)The “More Analysis” Alternative

LOWD proposes an alternative that would require additional analysis of environmental impacts before herbicides may be used to treat newly identified sites of invasive plants. LOWD does not describe what such analysis would entail or how this proposal would differ from the EDRR policy adopted. See Morongo Band of Mission Indians, 161 F.3d at 575-77 (burden is on the plaintiff to specify a detailed counterproposal, at least where the agency considered similar alternatives). Further, it is not clear how such a requirement for additional analysis would be compatible with the Project’s purpose and need to respond rapidly to new infestations. Given that an agency need not consider speculative alternatives or alternatives that would not meet the purpose and need of a project, it was not arbitrary or capricious for the Forest Service to omit such an alternative.

(e)The “Prevention” Alternative

LOWD argues that the Forest Service should have considered an alternative that focused on prevention measures, particularly measures directed at forest activities that facilitate the introduction and spread of invasive plants. The Project FEIS explained why the Forest Service rejected such an alternative: it would not meet the stated purpose and need to treat existing infestations of invasive plants, and the prevention measures urged by LOWD, such as restrictions or exclusions on grazing, would run counter to the Forest Service’s overarching multiple-use policy.33 AR22011. Further, as the Project FEIS notes in its discussion of purpose and need, the Region Six 2005 ROD already incorporated concrete prevention measures into *1007the local Forest Plan. AR21932. For example, the Region Six 2005 ROD adopted a requirement that only pelletized or certified weed-free feed could be used on all Region Six Wilderness areas as of January 1, 2007. AR17296. It also requires the revision of grazing permits to incorporate weed prevention practices. AR17297. The Regional Forester adopted this requirement even though it may result “in changes in grazing locations, timing, intensity, and outputs,” to the detriment of permit holders. AR17269. It was not unreasonable for the Forest Service to rely on these and other prevention measures already in place and focus instead on how to implement the treatment portion of the programmatic direction at the local level.

LOWD argues, however, that the Forest Service should at least have analyzed whether these prevention measures would be sufficient when applied locally. LOWD emphasizes that preventing the introduction of new weeds is connected to the success of efforts to eradicate those weeds already in the forest: without the former, the latter will never be achieved. This interrelationship between prevention and treatment, however, is better addressed through the cumulative impacts analysis, as discussed below. In the context of alternatives, it was not arbitrary or capricious for the Forest Service to conclude that: (i) the role of prevention had been adequately analyzed at the regional level and incorporated into the Forest Plan through the addition of concrete requirements; (ii) the purpose and need for the Project FEIS was to treat and eradicate the existing infestations of invasive plants; and (iii) further prevention measures along the lines proposed by LOWD would impermissibly limit other uses on the forest. See Nw. Coal, for Alts, to Pesticides, 844 F.2d at 593-94 (agency did not need to consider alteration of grazing permits in lieu of herbicide use, where purpose of project was to eradicate weeds already on the land).34

(f) The “Least Toxic Subset” Alternative

Finally, LOWD argues that the Forest Service should have considered an alternative that uses a smaller subset of the ten herbicides approved at the regional level. As an example, LOWD asserts that picloram and triclopyr have greater toxicity to non-target plants and wildlife than the other approved herbicides. Although the Region Six 2005 FEIS determined that all ten herbicides were necessary to treat invasive species effectively throughout Region Six, LOWD argues that the Forest Service should have evaluated whether all ten were reasonably necessary within the confínes of the Wallowa-Whitman National Forest.

In response, the Forest Service argues that all ten herbicides are in fact necessary. This analysis is explicit in the Region Six 2005 FEIS and implicit in the Project FEIS. The Region Six 2005 FEIS explains how the ten herbicides differ in their effectiveness in treating different *1008plant species; how the choice of herbicide would depend on the terrain of the infested site; and the agency’s concern that a limited set of herbicides may increase herbicide resistance. See AR15339-15340, 15350-15354, 15357-15358. In comparing the effectiveness of using all ten herbicides versus using a subset thereof, the Region Six 2005 FEIS identifies ten invasive plants that would not be effectively treated under the reduced set of herbicides. AR15340. It appears that all ten of these invasive plant species are already present in the Wallowa-Whitman National Forest. See AR22642-22648. Further, as Plaintiff conceded at oral argument, the Project FEIS contains the information necessary to establish why all ten herbicides may be needed to control invasive species effectively, even if the Forest Service did not explicitly analyze the alternative of using fewer than all ten approved herbicides. See Hr’g Tr. 65. The Forest Service’s failure to include this alternative in its Project FEIS does not rise to the level of arbitrariness or capriciousness.

The court therefore holds that the Forest Service’s discussion of reasonable alternatives, including its failure to address alternatives identified by Plaintiff, was not arbitrary or capricious.

C. Cumulative Impacts

Environmental impact statements must discuss cumulative impacts. See 40 C.F.R. §§ 1502.16, 1508.8, 1508.25; City of Carmel-by-the-Sea, 123 F.3d at 1160. A cumulative impact “is the impact on the environment which results from the incremental impact of the [proposed] action when added to other past, present, and reasonably foreseeable future actions regardless of what agency (Federal or non-Federal) or person undertakes such other actions.” 40 C.F.R. § 1508.7.

The court reviews the agency’s cumulative effects analysis under the APA’s arbitrary and capricious standard, and it must be particularly deferential “to an agency’s determination in an area involving a ‘high level of technical expertise.’ ” McNair, 537 F.3d at 987, 993. Given this deference, “the Forest Service is free to consider cumulative effects in the aggregate or to use any other procedure it deems appropriate. It is not for this court to tell the Forest Service what specific evidence to include, nor how specifically to present it.” League of Wilderness Defenders, 549 F.3d at 1218 (emphasis in original). The Forest Service may conclude that the effects of a related action would be of a different type and thus would not overlap with the anticipated effects of the proposed action, and it may reason that any identified impact would be off-set by mitigation and management measures. See id. at 1220.

On the other hand, a merely perfunctory cumulative impacts analysis is insufficient. Klamath-Siskiyou Wildlands Ctr. v. Bureau of Land Mgmt. 387 F.3d 989, 994 (9th Cir.2004). An analysis falls short if it only considers the impacts of the proposed action or only the beneficial impacts of cumulative actions. See Te-Moak Tribe of W. Shoshone v. U.S. Dep’t. of the Interior, 608 F.3d 592, 603-04 (9th Cir.2010); Klamath-Siskiyou Wildlands Ctr., 387 F.3d at 994-96; Muckleshoot Indian Tribe v. U.S. Forest Serv., 177 F.3d 800, 811 (9th Cir.1999) (per curiam). “General statements about ‘possible’ effects and ‘some risk’ do not constitute a ‘hard look’ absent a justification regarding why more definitive information could not be provided.” Neighbors of Cuddy Mountain v. U.S. Forest Serv., 137 F.3d 1372, 1380 (9th Cir.1998); see also Muckleshoot Indian Tribe, 177 F.3d at 811 (rejecting cumulative impacts analysis sections that “merely provide very broad and general statements devoid of specific, reasoned conclusions”). In particular, “some quantified or detailed information is re*1009quired. Without such information, neither the courts nor the public, in reviewing the Forest Service’s decisions, can be assured that the Forest Service provided the hard look that it is required to provide.” Neighbors of Cuddy Mountain, 137 F.3d at 1379; see also Klamath-Siskiyou Wildlands Ctr., 387 F.3d at 994 (cumulative impacts analysis inadequate where “there is no quantified assessment of [the] combined environmental impacts”); Muckleshoot Indian Tribe, 177 F.3d at 810 (requiring detailed analysis of cumulative impacts).

The cumulative impacts analysis in the Project FEIS is insufficient. The Project FEIS contains a general discussion titled “Basis for Cumulative Effects Analysis,” as well as resource-specific discussions scattered throughout Chapter 3. The general cumulative impacts section begins with an acknowledgment that other land uses will continue to introduce and spread invasive species, and that the prevention measures currently in place will only reduce the rate of spread, not end it entirely. See AR22022. Among other “vectors” that spread invasive plants, the Project FEIS lists recreation, grazing, vegetation management, wildfire and fire prevention, logging, road use and maintenance, and agriculture on adjacent farmlands. AR22022-22023. The Project FEIS does not, however, consider the impact of this continued introduction and spread of invasive species. For example, continued introduction implies a need for continued treatment, which will presumably include continued applications of herbicides, yet the Project FEIS does not address how the cyclical use of herbicides could further affect forest lands that are already highly impacted by the activities that are introducing the invasive species in the first place.35

The general cumulative impacts section next acknowledges that all forms of invasive plant treatments will have some adverse impacts on non-target plants, but it dismisses consideration of all but chemical treatments because “the potential for non-herbicide treatments to result in effects of concern to the public is very low.” AR22023.36 This misunderstands the purpose of the cumulative impacts analysis. The Forest Service cannot start with the *1010assertion that direct impacts will be minimal and conclude that a thorough cumulative impacts analysis is therefore not needed. The very point of a cumulative impacts analysis is to draw attention to combined impacts that might otherwise be overlooked when considered separately. See 40 C.F.R. § 1508.7 (cumulative impacts “can result from individually minor but collectively significant actions taking place over a period of time”); Klamath-Siskiyou Wildlands Ctr., 387 F.3d at 997 (rejecting analysis that did not consider how “individual impacts might combine or synergistically interact with each other to affect” the forest environment).

The general cumulative impacts section then considers at length the use of herbicides and pesticides throughout the country and in Oregon, as well as the background levels of chemicals found in the Clackamas River (which does not run near the Wallowa-Whitman National Forest). It concludes that the use of herbicides under the Project would not add to the background level of chemicals present in the water supply, but would only have temporary, localized impacts due to PDFs, buffers, and other mitigation measures. See AR22024-22029. This analysis falls short on several levels.

First, the analysis primarily focuses on the direct effects of the use of herbicides to treat invasive species. As the Ninth Circuit held in Te-Moak Tribe, the Forest Service cannot simply describe the direct impacts of the proposed action and then conclude in general terms that there are no cumulative impacts. See 608 F.3d at 603-04.

Second, the analysis focuses exclusively on herbicide use and the impacts of chemicals; it does not consider the possible non-chemical impacts of the proposed treatment regime, such as ground disturbance, or similar impacts caused by other forest uses. For example, the Project FEIS does not consider how the presence of herbicides in streams, even if fleeting and localized, might nonetheless exacerbate stream quality already impaired by other activities like logging, grazing, road maintenance, fire management, or recreation. The Forest Service, in rejecting LOWD’s appeal of the Region Six 2005 FEIS, recognized that such other activities should be accounted for in project-specific cumulative impact analyses. See Supp. AR17 (“The cumulative effects analyses conducted for [future] project proposals would provide more site-specific analyses related to specific invasive plant control activities, including disclosure of potential impacts of other activities such as roads management, logging, fuels management, livestock grazing, and recreational use.”). Further, if logging, grazing, recreation, road building, and fire management are all vectors for the introduction and spread of invasive plants, many of the infested sites are presumably impacted by these other uses already. Those impacts, even if qualitatively different from the chemical impacts of herbicides, nonetheless add stress to the native ecosystems and should be accounted for in the cumulative impacts analysis.37

*1011Third, the analysis relies on mitigation measures and project management to assume away any cumulative impacts without first identifying what those cumulative impacts might be. The Forest Service may rely on project design and mitigation to conclude that any cumulative effects will be minimal, but only after it has identified and quantified the scope of the potential problem. See Te-Moak Tribe, 608 F.3d at 604.

The subsequent cumulative impacts discussions regarding particular forest resources do not remedy the deficiencies of the general analysis. Many of these sections discuss only the direct impacts of the proposed use of herbicides.38 Other sections do refer to the impacts caused by other uses, but only in a cursory fashion devoid of any real analysis.39 Cf. Muckleshoot Indian Tribe, 177 F.3d at 811 (resource-specific cumulative effects sections do not satisfy NEPA where they “merely provide very broad and general statements devoid of specific, reasoned conclusions”).

In short, “the potential for ... serious cumulative impacts is apparent here, such that the subject requires more discussion” than the Project FEIS provides. Klamath-Siskiyou Wildlands Ctr., 387 F.3d at 996. Because the Project FEIS does not present a “full and fair discussion of environmental impacts,” McNair, 537 F.3d at 1001, it is arbitrary and capricious in violation of the APA.

III. Compliance with the Clean Water Act

The Clean Water Act, 33 U.S.C. §§ 1251-1387, requires government agen*1012cies to obtain a National Pollution Discharge Elimination System permit (“NPDES permit”) before discharging any pollutants from a “point source” into navigable waters of the United States. League of Wilderness Defenders Blue Mountains Biodiversity Project v. Forsgren, 309 F.3d 1181, 1183 (9th Cir.2002). A “point source” is “any discernible, confined and discrete conveyance, including but not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, or vessel or other floating craft, from which pollutants are or may be discharged.” 33 U.S.C. § 1362(14). What constitutes a nonpoint source is not statutorily defined, but is understood to mean “the type of pollution that arises from many dispersed activities over large areas, and is not traceable to any single discrete source.” Forsgren, 309 F.3d at 1184.

The Project ROD states that the Project “will meet and conform to the Clean Water Act.” AR22635. It does not, however, mention any particular permitting requirement. LOWD concedes that at the time the Project ROD was adopted, no permit would have been required for the Forest Service to use herbicides in accordance with the Project. PL’s Sur-Reply Opp’n Defs.’ Cross-Mot. Summ. J. 32. Thus, LOWD does not challenge the Project’s compliance with the Clean Water Act per se. Rather, LOWD argues that the Forest Service violated NEPA by not including in the Project FEIS a discussion of a pending change in law that would require the Forest Service to obtain permits before using herbicides under the Project. This court reviews an agency’s compliance with the Clean Water Act and with NEPA under the APA’s arbitrary and capricious standard. Great Basin Mine Watch, 456 F.3d at 961.

In 2007, “the EPA issued a Final Rule concluding that pesticides applied in accordance with the Federal Insecticide, Fungicide, and Rodenticide Act (the ‘FIFRA’) are exempt from the Clean Water Act’s permitting requirements.” Natl Cotton Council of Am. v. U.S. Envtl. Prot. Agency, 553 F.3d 927, 929 (6th Cir.2009). The parties agree that this rule would have applied to the Project and that it exempted the proposed herbicide treatments from the NPDES permitting requirement. The Sixth Circuit, however, invalidated that rule in 2009, id. at 940, but it also stayed the issuance of its mandate until October 31, 2011, to allow the Environmental Protection Agency and the states time to develop new permitting procedures. The parties thus agree that in April 2011, when the Project ROD was adopted, a NPDES permit was not required. The question here is whether NEPA obliges the Forest Service to discuss in its FEIS a likely future permit requirement.

LOWD has pointed the court to a recent opinion from the District of Montana. See Native Ecosystems Council v. Weldon, 848 F.Supp.2d 1207 (D.Mont.2012). In Weldon, the court agreed with the plaintiffs that the Forest Service erred by not discussing the possible need for NPDES permits in a draft environmental impact statement (“DEIS”). 848 F.Supp.2d at 1223-25. Similarly to the present situation, a federal court had made clear before the issuance of the DEIS that a permit would be required; while a federal statute had created a permit exemption to circumvent that court ruling, that exemption was only temporary and the need for a permit in the future was still likely. See id. The district court, however, based its holding on the regulatory requirement that a DEIS “shall list all Federal permits ... which must be obtained in implementing the proposal. If it is uncertain whether a Federal permit ... is necessary, the [DEIS] shall *1013so indicate.” Id. at 1223 (quoting 40 C.F.R. § 1502.25(b)). As the Weldon court noted, that requirement is specific to the DEIS because it allows the public an opportunity to comment on an important issue before the FEIS is issued and the project is approved. Id. at 1224.

LOWD has not argued before this court that the Forest Service erred by not discussing the permitting requirement in its DEIS.40 Weldon is thus of limited application in this case. The court generally agrees with LOWD that the Forest Service should have at least stated in the Project ROD or FEIS its intention to comply with all applicable permitting requirements of the Clean Water Act, particularly given the impending change in law. Instead, the Forest Service only stated that the Project would comply with the Clean Water Act and noted, in the annual implementation plan for the Project, that prior to treating any individual infested site, the Forest Service would “[ajpply for any herbicide application permits when needed for treatments in Riparian Areas.” AR22635, 22670. Clearer reference to the Clean Water Act’s permitting requirements might have improved NEPA’s goal of ensuring public access to relevant information.

This court, however, cannot conclude that the absence of such discussion violates NEPA. LOWD’s reliance on 40 C.F.R. § 1508.27(b)(10) is misplaced for the reasons discussed above. Section 1502.2(d) does apply, but it only requires an EIS to “state how alternatives in it and decisions based on it will or will not achieve the requirements of ... other environmental laws and policies.” Again, the court has not found any precedent suggesting this regulation requires more than the Forest Service actually provided here. The court declines to adopt a novel interpretation of this regulation that would require the Forest Service to discuss future changes in the law, however likely, particularly where as here the precise requirements of the future rules are still unknown. Thus, the Forest Service’s failure to discuss contingent, future permitting requirements under the Clean Water Act was not arbitrary or capricious.

CONCLUSION

For the foregoing reasons, the Forest Service violated NEPA because its Project FEIS arbitrarily and capriciously did not analyze the cumulative impacts of the proposed action. Plaintiffs Motion for Summary Judgment (Doc. No. 20) is GRANTED IN PART and DENIED IN PART. Defendants’ Cross-Motion for Summary Judgment (Doc. No. 25) is GRANTED IN PART and DENIED IN PART.