NELSON, Circuit Judge:
The appellants, Northwest Coalition for Alternatives to Pesticides, the Oregon Environmental Council, and the Audubon Society of Portland (collectively referred to as Northwest), challenge the adequacy of the Final Northwest Area Noxious Weed Control Program Environmental Impact Statement (FEIS) and its supplement (SEIS). The district court partially dissolved an injunction prohibiting the Bureau of Land Management (BLM) from using herbicides to control and eradicate noxious weeds on public lands in Oregon. Northwest Coalition for Alternatives to Pesticides v. Lyng, 673 F.Supp. 1019 (D.Or.1987). The district court had jurisdiction pursuant to 18 U.S.C. § 3231 and we have jurisdiction pursuant to 28 U.S.C. § 1292(a)(1). We affirm.
[590]*590FACTUAL AND PROCEDURAL BACKGROUND
On March 1, 1984, the district court enjoined the BLM from using herbicides to control noxious weeds until the BLM completed a Worst Case Analysis pursuant to 40 C.F.R. § 1502.22 (1985). Northwest, 673 F.Supp. at 1021. See also Southern Oregon Citizens Against Toxic Sprays v. Clark, 720 F.2d 1475 (9th Cir.1983), cert. denied, 469 U.S. 1028, 105 S.Ct. 446, 83 L.Ed.2d 372 (1984) (SOCATS) (requiring worst case analysis for BLM’s herbicide spraying because of the scientific uncertainty as to safe levels of human exposure to herbicides); Save Our Ecosystems v. Clark, 747 F.2d 1240 (9th Cir.1984) (worst case analysis for BLM and United States Forest Service (USFS) herbicide spraying program in Oregon held inadequate).
The BLM is required to control and eradicate noxious weeds on public lands by the Carlson-Foley Act of 1968, 43 U.S.C. § 1241 et seq. (1982), and the Federal Noxious Weed Control Act of 1974, 7 U.S.C. § 2801 et seq. (1982). Immediate elimination of the weeds is necessary because of the dangers posed by the weeds.
Noxious weeds, which have become established on public lands administered by the BLM, reduce and eliminate desirable vegetation from public lands by competing with native plants for water, sunlight, and soil nutrients. The secondary effects of this competition include lower soil quality, increased erosion, and reduced livestock and wildlife yields through the presence of less desirable foliage.
Northwest, 673 F.Supp. at 1021. The BLM’s proposed action is aimed at fourteen species of weeds, some poisonous to livestock, which are found in the Pacific Northwest. Each state covered by the proposal (Idaho, Montana, Oregon, Washington and Wyoming) suffers millions of dollars in losses annually due to these weeds.
By December 1985, the BLM had prepared the FEIS and in April 1986, the BLM completed a Record of Decision (ROD) choosing one of the FEIS alternatives to implement the program. On April 10, 1986, the BLM moved to dissolve the portion of the injunction precluding herbicide use in Oregon. Northwest, 673 F.Supp. at 1021. However, the BLM withdrew that motion on June 30, 1986 in order to supplement the FEIS. Id. at 1022. After completing the SEIS concerning the possible impacts on the human environment from the proposed herbicide use and a supplemental ROD, the BLM again moved to dissolve in part the injunction in 1987.
The SEIS proposes the use of 6 herbicide formulations to combat the noxious weeds: Banvel, Rodeo, Tordon 22K, Tordon 2K, Esteron 99 and DMA-4. The active ingredients of these formulations are dicamba, glyphosate, pieloram and 2,4-D. “Each of these herbicides may also contain emulsifiers, solvents, preservatives, anti-volatility agents, and other substances commonly referred to as inerts.” Id.
On November 24, 1987, the district court ruled that the FEIS and SEIS met the requirements of the National Environmental Policy Act of 1969 (NEPA), 42 U.S.C. § 4332(C) (1982), and its implementing Council on Environmental Quality (CEQ) regulations, 40 C.F.R. §§ 1500-1508 (1986). The injunction was dissolved to allow the use of products containing dicamba, gly-phosate, pieloram and 2,4-D by the BLM in Oregon.
Northwest timely appealed. Northwest seeks attorneys’ fees on appeal pursuant to the Equal Access to Justice Act (EAJA), 28 U.S.C. § 2412(d)(1)(A) (Supp. Ill 1985).
STANDARD OF REVIEW
The Administrative Procedure Act, 5 U.S.C. § 706(2)(D) (1982), providing that agency action undertaken “without observance of procedure required by law” may be set aside, governs judicial review of an agency’s preparation of an EIS. Oregon Environmental Council v. Kunzman, 817 F.2d 484, 492 (9th Cir.1987); see also Animal Defense Council v. Hodel, 740 F.2d 1432, 1435 (9th Cir.1988). “NEPA is essentially a procedural statute.” Kunzman, 817 F.2d at 492. “The reviewing court may not substitute its judgment for [591]*591that of the agency concerning the wisdom or prudence of a proposed action.” Id. Under this circuit’s rule of reason, the district court (1) “must make ‘a pragmatic judgment whether the EIS’s form, content and preparation foster both informed decision-making and informed public participation,’ ” and (2) determine “whether an EIS contains ‘a reasonably thorough discussion of the significant aspects of the probable environmental consequences.’ ” Id. (quoting California v. Block, 690 F.2d 753, 761 (9th Cir.1982)).
Thus, this court reviews the district court’s factual findings for clear error, and its ultimate conclusion that the FEIS and the SEIS are legally adequate under NEPA and the CEQ regulations de novo. Rum-man, 817 F.2d at 493. “Put another way, whether a particular deficiency, or combination of deficiencies, in an EIS is sufficient to warrant holding it legally inadequate, or constitutes merely a ‘fly speck,’ is essentially a legal question, reviewable de novo.” Id.
DISCUSSION
1. The Range of the FEIS/SEIS and Discussion of Alternatives
Northwest challenges the FEIS and SEIS as inadequate because they do not contain an alternative “addressing the causes, not just the symptoms, of noxious weeds in an EIS designed to control and eradicate noxious weeds.” The district court accurately summarized the challenge: “Plaintiffs want the BLM to resolve the noxious weed problem with an integrated pest management (IPM) process that considers the use of herbicides only as the last feasible alternative ... or through altered grazing patterns.” 673 F.Supp. at 1024 (citations omitted).
The district court determined that the BLM and Northwest simply disagree on the best management policy and “[s]o long as the BLM’s decisions are not irrational or contrary to law, it may manage the public lands as it sees fit.” Id. (citing Natural Resources Defense Counsel v. Hodel, 819 F.2d 927, 930 (9th Cir.1987) (NRDC)). In NRDC, this court upheld a decision that the range of alternatives presented in an EIS concerning livestock grazing was adequate. This court characterized NRDC’s challenge to the EIS as “a challenge to the BLM’s policy decision to postpone livestock grazing adjustments until reliable data was available.” 819 F.2d at 930.
Here, the district court correctly viewed Northwest’s challenge as a similar attempt to debate BLM policy, which is not the proper subject of a NEPA action. Northwest details one IPM alternative to the BLM’s adopted alternative, and argues that the FEIS and SEIS should have discussed this alternative or one like it. That alternative, used by the BLM in Prineville, Oregon, “examines the causes of the weeds and manipulates grazing practices, and other naturally occurring events, to control those weeds.” Northwest claims that this alternative is viable and must therefore be examined. Northwest points to the testimony of two BLM employees and other witnesses who said it was reasonable to consider the causes of the weeds when trying to control and eradicate them.
NEPA provides that an EIS must include alternatives to the proposed action. 42 U.S.C. §§ 4332(C), 4332(2)(E). The CEQ regulations provide guidelines for examining alternatives. 40 C.F.R. §§ 1502.14, 1502.16. The regulations call the alternative section “the heart of the environmental impact statement.” 40 C.F.R. § 1502.14. “An EIS aids the agency’s own decision-making process by ensuring that the agency has before it ‘all possible approaches to a particular project ... which would alter the environmental impact and the cost-benefit balance.’ ” Methow Valley Citizens Council v. Regional Forester, 833 F.2d 810, 814 (9th Cir.1987) (quoting State of Alaska v. Andrus, 580 F.2d 465, 474 (D.C.Cir.), vacated in part sub nom. Western Oil & Gas Ass’n v. Alaska, 439 U.S. 922, 99 S.Ct. 303, 58 L.Ed.2d 315 (1978)).
To be adequate, an environmental impact statement must consider every reasonable alternative. An EIS is rendered inadequate by the existence of a viable but unexamined alternative. Furthermore, [592]*592even if an alternative requires ‘legislative action’, this fact ‘does not automatically justify excluding it from an EIS.’ Thus, the range of alternatives considered must be sufficient to permit a reasoned choice.
Methow Valley, 833 F.2d at 815 (citations omitted). However, this requirement is tempered by the recognition that “[t]he detail that NEPA requires in an EIS depends upon the nature and scope of the proposed action.” California v. Block, 690 F.2d at 761.
The dilemma here is that Northwest and the BLM disagree about the scope of the program, and it is the scope of the program that influences any determination of what alternatives are viable and reasonable. We must examine whether it was reasonable for the BLM to exclude an alternative involving altered grazing patterns or the use of chemicals only as a last resort from this EIS.
The BLM contends that it considered and rejected the IPM suggested by Northwest. Only a brief discussion of the reasons for rejecting an alternative is mandated under CEQ regulations. 40 C.F.R. § 1502.14(a). The BLM essentially argues that (1) an alternative similar to the Prineville alternative using less herbicides was considered and (2) Northwest’s grazing concerns were addressed and rejected in the EIS.
The program adopted by the BLM (Alternative 1) is an IPM involving manual, mechanical and biological methods as well as herbicide use to control the noxious weeds. Alternative 2 resembled the chosen program except that aerial spraying of herbicides was excluded. Alternative 3 employed manual, mechanical and biological methods without any herbicide use. Alternative 4 was the required “no action” plan.
The ROD justified the choice of Alternative 1 as environmentally preferable, emphasizing the flexibility and effectiveness of herbicide spraying. The herbicides can suppress 85-94% of the weeds’ growth.
The continued spread of noxious weeds would contribute to a decline in ecological condition. Noxious weeds would out-compete desirable plant species, resulting in less forage for both wildlife and livestock. Weeds would also spread to private land, resulting in a decline in agricultural productivity. Alternative 1 provides the most effective means of dealing with this.
Adverse impacts to non-target vegetation and animals would be temporary and localized. The adverse environmental consequences of allowing extremely competitive noxious weeds to spread far offsets any of these potential short-term impacts. In addition, the analysis shows that no unreasonable human health risks are associated with implementation of the decision.
Alternative 2 was not chosen because it excludes the most effective method of herbicide application: aerial spraying. The BLM reasoned that this would inhibit their ability to treat sizeable weed infestations in rugged terrain. Alternative 3, which uses no herbicides, was not chosen for three reasons. The decision is explained in the ROD:
First, the alternative excludes the use of the most effective of the integrated weed management tools for treatment that is otherwise accepted and advocated by weed scientists and professionals who manage weed control programs. Second, it does not serve the Bureau’s mission to prevent undue and unnecessary degradation caused by noxious weeds. Third, the Bureau’s program costs increase under this alternative because a greater emphasis must be placed on mechanical and manual means to offset the limited availability of effective biological agents, resulting in fewer acres being treated under this alternative.
Alternative 4 was not chosen because no action would be contrary to the BLM's statutory duty to control and eradicate the weeds.
The environmental consequences of the various alternatives are detailed in the FEIS. The BLM discussed the impact on air resources, soil, water quality, vegetation, animals, fish, wilderness areas, visual resources, cultural and social impacts, human health and economic welfare. Thus, [593]*593the BLM argues that it considered an alternative using less herbicides and adopted an integrated treatment program using other methods along with herbicide use to combat the weeds. The program chosen uses three other methods of weed control on over half the acres involved.
The BLM also argues that grazing patterns are discussed in the FEIS and SEIS, but that serious consideration of an alternative involving grazing management is beyond the scope of the noxious weed program.
Finally, several commentors suggested that BLM forego proceeding with any alternative for controlling and eradicating noxious weeds until it evaluates another choice. This is reducing livestock grazing and pursuing other forms of intensive livestock grazing management. The rationale of this suggested alternative is that overgrazing by livestock contribute [sic] to, or is a cause of, noxious weeds. This suggested alternative ignores that, given the magnitude of noxious weed infestations, BLM defined the program as primarily performing control and eradication .. .The suggested alternative is beyond the scope of the program being considered by BLM. Livestock grazing management is a question addressed in other BLM planning efforts. It is publicly known and acknowledged that the BLM has an ongoing program to produce plans and environmental impact statements on how to best manage livestock grazing on public lands. To consider an alternative here focusing on livestock grazing would be duplicative. The question of heavy grazing as a factor contributing to noxious weeds is properly addressed elsewhere.
The BLM repeatedly stressed the need for immediate and effective action through this weed control program.
Although the BLM contends that grazing management is beyond the scope of this noxious weed program, the BLM did address the causes of noxious weeds and the attendant public concerns. After noting Northwest’s concerns, the BLM reasoned that heavy livestock grazing is one cause of noxious weeds, but not the primary cause. Noxious weeds also infest ungrazed areas of the Pacific Northwest.
The Supreme Court has noted that the range of alternatives must be limited “by some notion of feasibility.” Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Counsel, 435 U.S. 519, 551, 98 S.Ct. 1197, 1215, 55 L.Ed.2d 460 (1978). The Court termed alternatives which are “ ‘not meaningfully compatible with the time-frame of needs to which the underlying proposal is addressed’ ” remote and speculative. Id. (quoting Natural Resources Defense Counsel v. Morton, 458 F.2d 827, 837-38 (D.C.Cir.1972)). The range of alternatives considered “ ‘need not extend beyond those [alternatives] reasonably related to the purposes of the project.’ ” Methow Valley, 833 F.2d at 816 (quoting Trout Unlimited v. Morton, 509 F.2d 1276, 1268 (9th Cir.1974)).
In Methow Valley, this court recently held the Forest Service’s range of alternatives in granting a permit for a ski resort on national forest lands inadequate. 833 F.2d 810 (9th Cir.1987). Although the Forest Service listed four alternatives in attempting to provide the public with a “winter sports opportunity,” it had considered only one parcel of land. Id. at 815-16 & n. 7. This court held that “it should have appeared obvious that investigation was warranted” to explore alternatives such as the expansion of existing ski areas before constructing a new one. Id. at 815-16. We determined that such exploration was reasonably related to the project’s purpose of providing recreational facilities. Id. at 816. Similarly, in California v. Block, the EIS did not consider a “no development” alternative in a region designated as a special wilderness protection area. Because no alternative devoting more than one-third of the land to wilderness use, the range of alternatives was held inadequate. 690 F.2d at 765-69. Here, the BLM considered an alternative using no herbicides.
In Animal Defense Council v. Hodel, this court recently rejected a claim that the range of alternatives was inadequate in a Bureau of Reclamation EIS concerning a [594]*594Tuscon aqueduct. The EIS did not address a method of water storage because it was not part of the federal program. Water storage was the city’s responsibility. Because it was a rejected alternative, this court required only a brief explanation that the alternative was eliminated because it was less cost effective and not supported by some organizations. Animal Defense, 840 F.2d 1432, 1441 (9th Cir.1988). If an alternative is remote and speculative, less consideration is required. See, e.g., City of Angoon v. Hodel, 803 F.2d 1016, 1020-22 (9th Cir.1986), cert. denied, — U.S. —, 108 S.Ct. 197, 98 L.Ed.2d 148 (1987).
Here, Northwest’s alternative including alteration of grazing management techniques and use of chemicals as a last resort was sufficiently considered by the BLM. The alternative was not reasonably related to the purpose of the project in light of the immediate need to eradicate the noxious weeds. The BLM sufficiently considered less use of herbicides in the alternatives discussed and the aspect of Northwest’s alternative urging alteration of grazing patterns was too remote for consideration in this EIS. Although it may be wise to consider the causes of noxious weeds in long-term planning, the aim of the current program is also to eradicate the weeds in the short term. Grazing management is currently being studied in a separate EIS. The scope of the program has been limited reasonably by the BLM. The district court correctly concluded that a reasonably thorough discussion of the alternatives took place.
2. The Alleged Scoping Violation
Northwest contends that the BLM violated CEQ scoping regulations which require that the public be given opportunities to participate in the EIS process. The BLM asserts that it made an inadvertent mistake in not personally inviting Northwest to participate in scoping, but that Northwest has not demonstrated prejudice sufficiently. The district court found that the BLM’s neglect to inform Northwest personally was “inexcusable” in light of the prior litigation between the parties in which Northwest successfully sought the injunction. 673 F.Supp. at 1023. However, the district court ruled that although the BLM’s conduct violated the spirit of NEPA, it did not violate the letter of the law. Id. Further, the court found that Northwest was not prejudiced by the BLM’s failure to notify because “the BLM has considered and rejected plaintiffs’ suggestions regarding the scope of the environmental statement” and because Northwest was notified of and participated in the SEIS process. Id. at 1023-24.
A. The BLM’s Statutory Duties
We review the district court’s interpretation of CEQ regulations de novo. Kunzman, 817 F.2d at 493.
The district court noted that “scoping will be effective only if people who are, or may become, interested in the proposed action are involved.” 673 F.Supp. at 1023. To ensure this, the court summarized the regulations as requiring “the agency to publish a notice in the Federal Register (40 C.F.R. § 1508.22), invite them to participate in the scoping process (40 C.F.R. § 1501.7(a)(1)), and encourage meetings with the public about the impact statement’s scope (40 C.F.R. § 1501.7(b)(4)).” Id. Although the BLM published the FEIS notice in the Federal Register, issued a press release and mailed it to some interested persons, the BLM failed to notify any of the three plaintiffs (Northwest) who had successfully gained the injunction against herbicide use in the weed control program in 1984.
The district court rejected Northwest’s claim that the BLM violated 40 C.F.R. §§ 1501.7 or 1506.6, which require direct notice to interested parties and national organizations.1 The court found that [595]*595“plaintiffs did not say either that they requested notice of this action pursuant to 40 C.F.R. § 1506.6(b)(1) or that they are ‘national organizations reasonably expected to be interested in the matter’ under 40 C.F.R. § 1506.6(b)(2).” The district court determined that the plaintiffs were not national organizations, but local.
Appellants correctly argue that the additional provisions for notice in § 1506.6 do not undercut the mandatory notice .under § 1501.7. Northwest, as a litigant earlier in this action, was clearly an interested person. Appellants also convincingly argue that the situation at issue here does not fit neatly into the national/local dichotomy of § 1506.6. Appellants are in part regional branches of national organizations interested in a regional action. Northwest was clearly entitled to notice under § 1501.7 in these circumstances and arguably entitled to it under § 1506.6 in such regional actions.
In light of the purposes of NEPA and the scoping regulations and the appellants’ role in this action, we conclude that the BLM violated both the spirit and the letter of the law by failing to notify Northwest. However, the district court’s ruling must still be upheld because Northwest failed to demonstrate prejudice.
B. Prejudice
It is settled that prejudice must be demonstrated before administrative decision-making can be set aside. County of Del Norte v. United States, 732 F.2d 1462, 1467 (9th Cir.1984), cert. denied sub nom. Association of California Water Agencies v. United States, 469 U.S. 1189, 105 S.Ct. 958, 83 L.Ed.2d 964 (1985). The Fourth Circuit has held that persons not given notice of public hearings on a proposed wastewater treatment plant did not suffer prejudice. Providence Road Community Ass’n v. EPA, 683 F.2d 80 (4th Cir.1982). Although the persons could not “participate until the eleventh hour, their comments were received and considered by the EPA before the agency reached its final decision not to prepare an EIS.” Id. at 82.
The district court, using the same rationale, noted that Northwest’s attorney was notified of the BLM’s intent to prepare the SEIS around or after June 30, 1986, when the BLM withdrew its first motion to dissolve the injunction. Further, the district court found that Northwest had written several letters to the BLM in January and February of 1986 expressing its concern that the EIS consider an alternative discussing the causes as well as the symptoms of noxious weeds. 673 F.Supp. at 1023-24. The FEIS was completed in December [596]*5961985, and the ROD in April 1986. The district court concluded that Northwest has not shown it was unable to offer such an alternative throughout the proceedings.
Northwest argues that it suffered prejudice by being given a chance to participate too late in the process. Northwest contends that the BLM did not consider seriously some of its suggestions at a November 1986 meeting (during the SEIS process) because a draft had already been prepared and “no meaningful changes in the SEIS would be possible.” Northwest’s argument that participation in the SEIS cannot cure its lack of participation in the FEIS is persuasive. The SEIS concerned primarily the effect of herbicides on humans for cancer risks. Northwest argues that if it could have presented its comments earlier, the scope of the program might have been enlarged or the BLM might have been more receptive to its alternative.
However, Northwest did participate early in the EIS process. Northwest raised its grazing concerns and its view of a better IPM in its comments on the draft EIS on July 31, 1985. These comments are incorporated in the FEIS:
First, some general comments can be made about the DEIS [draft]. Nowhere is there an [IPM] alternative for this program ... An IPM alternative would be an alternative that looks to pesticides as a last resort and depends heavily on pest monitoring, evaluation of impact thresholds, various non-chemical control techniques and public education of the users and visitors to BLM lands as the tools to control noxious weeds ... Educational efforts to ensure sound grazing management could reduce acreage that needs to be treated. Many noxious weeds get a foothold on disturbed lands, and the spread of noxious weeds could be confined in some instances by management techniques.
Since Northwest clearly participated in the project and communicated these views, its argument must fail. Although the BLM’s violation of the scoping provisions is regrettable, Northwest was not prejudiced as a result of the violation, and the district court must be affirmed.
3. BLM's Independent Assessment
Northwest contends that the BLM adopted the EPA’s conclusions about the safety of the chemicals it proposed to use without conducting an independent analysis. The district court correctly noted that the EPA’s registration of an herbicide under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) “is inadequate to address NEPA environmental concerns and therefore reliance alone on that registration process is improper.” 673 F.Supp. at 1025. An agency must conduct independent research on the safety of an herbicide it proposes to use. SO CATS, 720 F.2d at 1480.
The district court distinguished the Forest Service’s failure to conduct an independent investigation in Save Our Ecosystems, 747 F.2d 1240 (9th Cir.1984) (SOE), from the BLM’s assessment here. The district court noted that in SOE this court held that an agency can satisfy this requirement by “ ‘appropriately considering] EPA’s data on the herbicides in the specific context of the area in which it proposes to spray.’ ” 673 F.Supp. at 1025 (quoting SOE, 747 F.2d at 1247). In SOE, the Forest Service had failed to examine any further chemical research and did not “address the health effects of using the herbicides in the area to be sprayed.” 747 F.2d at 1243. Although the BLM in this EIS proposes that some of the site-specific analysis will be completed by state and district levels, the BLM did acknowledge in the ROD that the EPA and FIFRA data alone was insufficient to assess the environmental impact of the herbicides. To remedy this insufficiency, the BLM considered numerous studies and the SEIS applied the EPA data by examining the effects on humans, wildlife, and the environment expected from use of the herbicide formulations in the noxious weed program. The BLM then determined in the ROD that the level of risk was acceptable. Thus, Northwest’s argument that no independent analysis was undertaken is not persuasive.
[597]*5974. The Worst Case Analysis
The district court enjoined the BLM from herbicide use in 1984 until the BLM completed a worst case analysis pursuant to 40 C.F.R. § 1502.22.2 This regulation requires an agency to “make a reasoned assessment of the environmental impacts of a course of action” rather than ignore the effect of inadequate or unavailable information. Methow Valley, 833 F.2d at 817. The regulation was amended to eliminate the “worst case analysis” requirement in May 1986. However, since the SEIS was already in progress, the BLM could choose which version of § 1502.22 it would comply with. 40 C.F.R. § 1502.22(c) (1986). The BLM chose to comply with the older version and the district court analyzed its compliance under that standard. 673 F.Supp. at 1022.3
The district court held that the BLM had complied with § 1502.22. The district court determined that the BLM had engaged in a reasonably thorough discussion of the significant aspects of the probable consequences. The court noted that “[t]he testimony and witness statements from the experts for plaintiffs and defendants display an internecine toxicological battle.” 673 F.Supp. at 1025. Northwest’s challenges to the worst case analysis can be grouped into three categories: (1) the impact of the inert, or secret, ingredients is not discussed adequately; (2) the SEIS does not address the contaminants of the herbicides; and (3) the SEIS ignores data about the more serious impacts of the herbicides.
The SEIS at issue primarily addresses the human health effects of herbicide use. The SEIS contains a chemical hazard assessment and cancer potential assessment, comparing potential doses to humans from applying or being near an application of the herbicides to safe doses based on animal studies. The SEIS analyzes the four herbicide formulations and their risk to the general public in performing the cancer analysis. The supplemental ROD then adopts Alternative 1, which uses herbicides, although it acknowledges that herbicide use is not without risks. After summarizing the underlying data, the ROD concludes that the level of risk posed by the herbicides to humans is acceptable.
1. Inerts
The BLM discusses the impact of the inerts, acknowledging their presence in the six formulations proposed. Only one inert, Esteron 99, was on the more harmful EPA lists, and a study was conducted for it. The SEIS acknowledges the data gaps for inert ingredients resulting from the EPA’s “confidential business information” policy, which prohibits the BLM from knowing all the ingredients of the formulations. The BLM acknowledges that only the active ingredients, not the formulations, have been extensively tested. However, the BLM reasons that the active ingredients rather than the inerts are the problems in the herbicide formulations. It points to available toxicity data which casts doubt on the possibility that the herbicide formula[598]*598tions are more toxic than the active ingredients alone. Taking the uncertainties into account, the BLM overstated the risk of harm posed by the active ingredients and proceeded to undertake a reasoned analysis.
2. Contaminants
Northwest claims that the BLM should have conducted a worst case analysis for 2,4-D because it might be contaminated with TCDD. This concern was raised in the SEIS but rejected. The BLM states that although some 2,4-D was contaminated in the early 1980s, that particular version of 2,4-D is no longer being produced. The EPA has monitored the situation and no contaminants have been found recently. The SEIS notes the possibility that 2.4-D might be contaminated but concludes that no human health hazards are posed. The BLM acknowledges the uncertainty of the data, and then reasons that even if contaminants are found, the hazards are reduced because of the low dosage humans are likely to be exposed to through the herbicide spraying and the low toxicity of 2.4-D.
3. Ignored Data
Northwest protests that no mammalian studies such as those offered by Dr. Ruth Shearer were included in the worst case analysis. However, the BLM presented a representative range of studies and the district court correctly concluded that it “cannot say that one set of experts is clearly more correct than the other.” 673 F.Supp. at 1025. Again, in summarizing the studies, the BLM acknowledges the existing uncertainties. The reports include studies with adverse as well as favorable toxicity results.
Although Northwest raises some legitimate environmental concerns, the record supports the district court’s conclusion that the BLM’s decisionmaking was not arbitrary or capricious. The district court correctly observed that the BLM “is entitled to be ‘arguably wrong.’ ” Id. at 1026 & n. 6 (quoting Kunzman, 817 F.2d at 496). The district court’s partial dissolution of the injunction is upheld.
Thus, Northwest’s request for attorneys’ fees on appeal pursuant to the Equal Access to Justice Act (EAJA), 28 U.S.C. § 2412(d)(1)(A) (Supp. Ill 1985) is denied. See SOCATS, 720 F.2d at 1481.
AFFIRMED.