MATHESON, Circuit Judge.
Petitioner-Appellant Western Watersheds Project (“WWP”) challenged a Bureau of Land Management (“BLM”) decision to grant a 10-year grazing permit to LHS Split Rock Ranch, LLC (“Split Rock”) for four federal public land allotments in central Wyoming (“the Split Rock allotments”). WWP asserted that BLM’s decision to grant the grazing permit was arbitrary and capricious because BLM had previously concluded that past grazing was a substantial cause of serious environmental degradation on the Split Rock allotments. The district court granted summary judgment to BLM. WWP appeals. Split Rock responds as Intervenor-Appel-lee.
Exercising jurisdiction under 28 U.S.C. § 1291, we affirm.
I. BACKGROUND
We begin by describing several relevant statutes and regulations. We then summarize the factual and procedural history of the case before turning to our analysis of the issues.
A.
Relevant Statutes and Regulations
The Taylor Grazing Act of 1934, 43 U.S.C. § 315
et seq.,
authorizes BLM to establish livestock grazing allotments on federal public lands. In carrying out this mission, BLM also must comply with the Federal Land Policy and Management Act of 1976 (“FLPMA”), 43 U.S.C. § 1711
et seq.,
which requires BLM,
inter alia,
to develop land use plans for the public lands it administers with a goal of improving resource conditions and avoiding “undue degradation” of the land.
Id.
§§ 1712(a), 1732(b). FLPMA also requires BLM to manage public lands according to a “[m]ul-tiple use management.”
See State of New Mexico ex rel. Bill Richardson v. BLM,
565 F.3d 683, 690 n. 3 (10th Cir.2009). To fulfill its multiple use mission, BLM must design its land use plans to strike “a balance among the many competing uses to which land can be put.”
Id.
(quotations omitted);
see also
43 C.F.R. § 1601.0-5®.
Federal regulations similarly direct BLM to ensure that livestock grazing does not exceed the “carrying capacity” of the land. 43 C.F.R. § 4130.3-1. “Carrying Capacity means the maximum stocking rate possible without inducing damage to vegetation or related resources.”
Id.
§ 4100.0-5.
In 1987, pursuant to these authorities, the BLM developed a land use plan for the Lander Field Office, which oversees the Split Rock allotments. Once a land use plan is implemented “all future resource management authorizations and actions ... shall conform to the approved plan.” 43 C.F.R. § 1610.5-3(a). This means BLM’s actions must be “clearly consistent with the terms, conditions, and decisions of the approved plan.”
Id.
at § 1601.0-5(b). The 1987 Lander Resource Management Plan (“RMP”) is the current land use plan and therefore guides BLM’s decisions concerning the land.
The Lander RMP provided for continued grazing on public lands consistent with multiple use planning. Aplt. Appx. at 141, 143. The RMP outlined three objectives for the rangeland use: (1) “[p]rovide enough forage on a sustained-yield basis to satisfy at least the present demands of livestock, wild horses, and wildlife”; (2) “[mjaintain range condition at a level that would provide for sustained yield of forage production”; and (3) “[mjaintain and improve the terrestrial, aquatic, and ripari-an
ecosystems to provide wildlife with
adequate amounts of
forage and
habitat to maintain planned population levels.”
Id.
at 144.
In carrying out these objectives under the FLPMA and the RMP, BLM considers six categories of ecological health for rangeland, which are referred to as the Rangeland Health Standards (“RHS”).
Id.
at 201-03;
see also
43 C.F.R. § 4180.2. They include: (1) Soils, (2) Riparian, (3) Upland Vegetation, (4) Diverse Species Habitat, (5) Water Quality, and (6) Air Quality. Aplt. Appx. at 201-03. As we later discuss, whenever RHS are not met, federal regulations require BLM to investigate and determine the cause.
See
43 C.F.R. § 4180.2(c)(1). If grazing is a culprit, BLM must take appropriate corrective action.
See id.
§ 4180.2(c)(3).
Finally, the National Environmental Policy Act of 1969 (“NEPA”) requires federal agencies to “assess potential environmental consequences of a proposed [agency] action.”
Utah Envtl. Congress v. Russell,
518 F.3d 817, 820 (10th Cir.2008);
see also
42 U.S.C. § 4331
et seq.
“NEPA dictates the process by which federal agencies must examine environmental impacts, but does not impose substantive limits on agency conduct.”
Russell,
518 F.3d at 821. Rather, it serves to promote informed agency decision making, government transparency, and public access to information.
See State of New Mexico,
565 F.3d at 703.
Regulations implementing NEPA establish a two-part process for an agency considering a proposed action. First, the agency must determine whether the proposed action will significantly affect the environment. If the answer is not immediately apparent, the agency must prepare an Environmental Assessment (“EA”), which is “a concise public document that briefly provides sufficient evidence and analysis for determining” the appropriate next step.
Russell,
518 F.3d at 821 (quotations omitted);
see also
40 C.F.R. § 1508.9.
If the EA concludes that the proposed action will have no significant effect on the environment, the agency may issue a Finding of No Significant Impact (“FONSI”) and move forward with the proposed action.
Russell,
518 F.3d at 821;
see also
§§ 1501.4(e), 1508.13. Otherwise, it must move on to the more extensive process of developing an Environmental Impact Statement (“EIS”) to “ ‘rigorously explore and objectively evaluate all reasonable alternatives’ ” and consider the comparative merits and environmental effects of the alternatives against the proposed action.
Wyoming v. U.S. Dep’t of Agriculture,
661 F.3d 1209, 1243 (10th Cir.2011) (quoting 40 C.F.R. § 1502.14(a)).
B.
Factual History
Petitioner-Appellant WWP is a nonprofit public interest organization committed to promoting species and habitat protection in the western states, including Wyoming rangelands. Permittee LHS Split Rock is a Delaware limited liability company that operates a ranch in central Wyoming. In 2003, BLM granted Split Rock a 10-year permit allowing livestock grazing on four allotments in two central Wyoming counties.
The four Split Rock allotments — Diamond Springs, North Do-bie Flat, South Dobie Flat, and Black Jack
Ranch — together encompass more than 90,000 acres of public land.
Grazing permits authorize limited grazing on designated land, within set limits defined by stocking level and animal-unit-months (“AUM”). “Stocking level” refers to the volume of grazing on the land.
“AUM” refers to the amount of forage needed to sustain one cow or its equivalent for one month.
See
43 C.F.R. § 4100.0-5. Split Rock’s grazing permit authorized a stocking level of 216 grazing days each year and up to 9,400 AUM. Since 1993, stocking levels on the Split Rock allotments have averaged 8,054 AUM.
1. 2005 Rangeland Health Standards Assessment
In 2005, BLM completed a comprehensive assessment of rangeland health standards (“RHS”) on public lands in the Lander area. The RHS assessment revealed significant rangeland health issues on all four Split Rock allotments. The allotments all clearly failed in four of the six RHS standards — Soils, Riparian, Upland Vegetation, and Diverse Species Habitat. One standard, Water Quality, was “unknown.” Aplt. Appx. at 202. The allotments passed only one of the six standards, Air Quality.
Failure under the standard of Riparian health was especially pronounced, as no BLM-administered acres in the area met the standard. All but .7 miles of riparian land was found to be “Functional-at-Risk with a downward trend.”
Id.
at 233;
also see
Aplt. Br. at 19. Split Rock’s failure to comply with the herding requirements under its initial permit allowed cattle to overuse riparian areas and contributed to this deficiency. Failure under the Soils standard also was notable, with only 24 to 36 percent of the total acreage in each allotment meeting soil productivity standards. The RHS assessment described the soils as having bare ground and sparse vegetative cover, with the entire topsoil layer absent in many areas. Although the water quality had not been recently tested, the RHS assessment noted the decline or depletion of fisheries that had thrived several decades earlier.
The results from this RHS assessment indicated that the stocking levels, i.e., grazing use, exceeded the carrying capacity of the Split Rock allotments. As noted previously, “carrying capacity” refers to “the maximum stocking rate possible without inducing damage to vegetation or related resources. It may vary from year to year on the same area due to fluctuating forage production.” 43 C.F.R. § 4100.0-5. Federal regulations require the BLM to “reduce permitted grazing use or otherwise modify management practices” whenever carrying capacity is exceeded.
Id.
§ 4110.3-2(b).
In early 2006, the Lander Field Manager (“Lander FM” or “FM”) issued RHS determinations that livestock overgrazing was a significant cause of these problems. Following any RHS failure, federal regulations require BLM to identify and implement “appropriate action” to make “significant progress toward fulfillment of the standard” by the start of the next grazing year.
Id.
§ 4180.2(c). Pursuant to this requirement, the FM outlined Appropriate Actions to stall or reverse the land degradation described in the 2005 RHS assessment. These Appropriate Actions called for nearly two-dozen aggressive changes to terms and conditions of Split Rock’s grazing permit, including three-day pasture limits; maximum of 20-days’ hot-season grazing on some pastures; and a rest-rotation system on six pastures, which would remove livestock entirely from one pasture per year.
2. “Second Look” at the RHS assessment
Before the FM could implement these changes, he was replaced by a new Lander FM, who rescinded the RHS findings and “directed the BLM staff to take a second look at the ... assessment [and] determination of cause.” Aplt. Appx. at 199. The new FM directed the staff to consider “an independent opinion of rangeland conditions,” which was to be commissioned by Split Rock.
Id.
He also ordered a peer review of the 2005 RHS Assessment. This peer review, conducted by experts outside the Lander Field Office, concluded that the 2005 RHS Assessment was based on “scientifically accepted and well established procedures” and that its conclusions were supported by “[professional experience and hard data.”
Id.
In June 2007, the Lander Field Office staff submitted its determinations, which affirmed its previous findings of severe rangeland health problems and again identified overgrazing as a substantial cause. NEPA therefore required BLM to undertake an EA before deciding whether to reissue Split Rock’s grazing permits.
See
43 C.F.R. § 4180.2(c)(3).
3. The Environmental Assessment
The 102-page EA was issued in 2009. Relying largely on the 2005 RHS Assessment, it acknowledged serious ecological problems on the rangeland. The EA considered five alternatives to address these problems, but only three were analyzed in detail. Two alternatives, referred to as “No Action” and “No Grazing,” were briefly considered but rejected without detailed analysis. Aplt. Appx. at 209-10. The EA explained that BLM had chosen not to analyze the No Action alternative because No Action would have meant maintaining the same 2003 permit terms that had been found detrimental. The EA did not analyze the No Grazing alternative because the 1987 Lander RMP had “concluded that eliminating livestock grazing from all public lands would not be a viable or necessary option,” and this document guided BLM. Aplt. Appx. at 210.
The EA'analyzed the remaining three alternatives in detail. No single alternative incorporated all of the nearly two-dozen Appropriate Actions developed in response to the RHS.
Alternative One would have accomplished rangeland recovery primarily through a robust rest-rotation grazing system, with only a slight decrease in volume of grazing. The rest-rotation system would have required one pasture to be fully rested each year, i.e., no grazing on that pasture for the year. It would have restricted hot-season grazing on two pastures, but otherwise require only a small decrease in the number of cattle, AUM, or annual grazing days. The EA found this alternative to be the most effective of the three in improving rangeland health.
Alternative Two would have accomplished rangeland recovery primarily through a large decrease in stocking rate, with only deferred rotation rather than full rest rotation. Deferred rotation would have meant delaying the start of the grazing season by several weeks on most pastures. This alternative would have required nearly a one-third decrease in the number of cattle and AUM and more intensive herding during the hot-season period. The EA found this alternative would have improved rangeland health in an “acceptable timeframe” but not as quickly as Alternative One.
Alternative Three was submitted by Split Rock. It called for an
increase
in stocking rate (more cattle, higher AUM), an expanded grazing season, and some deferred-rotation but no rest rotation. The EA found this alternative would have accelerated degradation in several areas.
4. The FONSI and the Proposed Decision
In April 2009, several months after the EA was issued, the Lander FM issued a FONSI — i.e., a finding that renewal of the Split Rock grazing permit would not significantly affect the environment. As a result of the FONSI, no EIS was conducted. Instead, BLM issued a Notice of Proposed Decision. The Proposed Decision did not match any of the alternatives described in the EA. Rather, it combined Alternatives One and Two by eliminating the most environmentally protective features of each: it required neither rest rotation nor a significant decrease in stocking levels. But the Proposed Decision did incorporate other protective features, such as fencing, deferred rotation, and a shorter grazing season.
Split Rock and WWP both protested. In response, BLM agreed to allow an additional 100 cattle (for a total of 1,200 head) in exchange for Split Rock’s commitment to provide a full-time employee to assist with herding compliance, which would help to even out grazing across the acreage and reduce overgrazing of some at-risk areas, especially riparian areas. It issued a final Notice of Decision in October 2009, granting Split Rock a 10-year grazing permit under the terms of the Proposed Decision with these negotiated adjustments.
The following chart summarizes the historic use of the allotments, the first two alternatives considered in the EA, and the final Proposed Decision:
Stocking level Change in AUM Rest rotation Fencing & or deferred herding rotation
Historic use 8,054 AUM (1,366 cattle; 216 days) Neither • No fencing • Herding required but with poor compliance
Alternative One 6,717 AUM (1,250 cattle; 195 days) 17% decrease Full rest rotation • Extensive fencing • No herding required
Alternative Two 5,358 AUM (1,000 cattle; 204 days) 34% decrease Deferred rotation only • Some fencing • Herding required
Final Proposed 6,669 AUM Decision (1,200 cattle; 204 days) 17% decrease Deferred rotation only • Some fencing • Herding required and more strictly enforced
C.
Procedural History
WWP filed a Petition for Review of Agency Action in the Wyoming federal district court pursuant to 5 U.S.C. §§ 701-706 of the Administrative Procedures Act (“APA”). The petition challenged the new proposed grazing permit. WWP argued that the permit decision was arbitrary and capricious because,
inter alia,
the EA failed to comply with NEPA requirements
to evaluate a reasonable range of alternatives and failed to take a “hard look” at potential environmental consequences of the proposed action.
See Balt. Gas & Elec. Co. v. Natural Res. Def. Council,
462 U.S. 87, 97, 103 S.Ct. 2246, 76 L.Ed.2d 437 (1983).
The district court granted summary judgment to BLM. It concluded that BLM’s permit decision was not arbitrary and capricious, finding that the EA had considered a reasonable range of alternatives by establishing an adequate baseline from current conditions and including environmentally protective features. The court also determined the EA met the hard look requirement.
II. DISCUSSION
We review de novo the district court’s grant of summary judgment for BLM.
State of New Mexico,
565 F.3d at 704-05 (10th Cir.2009). Although the district court’s decision is not afforded deference, BLM’s decision must be: “Our inquiry under the APA must be thorough, but the standard of review is very deferential to the agency.”
Hillsdale Envtl. Loss Prevention, Inc. v. U.S. Army Corps of Engineers,
702 F.3d 1156, 1165 (10th Cir.2012) (quotations omitted). “A presumption of validity attaches to the agency action and the burden of proof rests with” WWP.
Morris v. U.S. Nuclear Regulatory Comm’n,
598 F.3d 677, 691 (10th Cir.2010) (quotations omitted). Our deference is most pronounced in cases where, as here, the challenged decision involves “technical or scientific matters within the agency’s area of expertise.”
Utah Envtl. Congress v. Bosworth,
443 F.3d 732, 739 (10th Cir.2006).
This deference means we may set aside an agency action only if it is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). An agency decision is arbitrary and capricious only if the agency
(1) entirely failed to consider an important aspect of the problem, (2) offered an explanation for its decision 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, (3) failed to base its decision on consideration of the relevant factors, or (4) made a clear error of judgment.
State of New Mexico,
565
F.Sd
at 704 (quotations omitted).
WWP raises two issues on appeal. First, it argues that the EA failed to evaluate a reasonable range of alternatives as required under NEPA. Second, it argues that BLM failed to take the required “hard look” at the potential environmental consequences of its actions. We discuss each of these issues below and conclude thát BLM’s decision was not arbitrary and capricious.
We note, however, that WWP has raised serious questions about BLM’s decisions regarding the Split Rock grazing permit that make this case difficult even under our highly deferential review standard. The record reveals troubling problems with the rangeland health of the Split Rock allotments, and even BLM implicitly acknowledges that its Proposed Decision is unlikely to remedy these problems quickly. Nevertheless, NEPA “merely prohibits uninformed — rather than unwise — agency action.”
Id.
It is not within our authority to resolve whether BLM selected the best or wisest option. The agency considered a reasonable range of alternatives, and its analysis met the minimum threshold necessary to constitute a “hard look” at the consequences of its actions. We may “not substitute our judgment for that of’ BLM.
Davis v. Mineta,
302 F.3d 1104, 1114 (10th Cir.2002).
A.
Did the EA Fail to Evaluate a “Reasonable Range of Alternatives”?
WWP argues that the EA failed to evaluate a reasonable range of alternatives as required by NEPA because it analyzed only three alternatives that were on the permissive end of the spectrum. WWP points to two problems in particular: failure to analyze the No Action alternative and failure to consider an aggressive, i.e., environmentally protective, alternative.
1. No analysis of the No Action alternative
The EA expressly considered a “No Action” alternative, which would have maintained the terms of the 2003 permits. Aplt. Appx. at 209. But it rejected this alternative without analysis, explaining that No Action was not feasible because BLM had already determined that the terms of the 2003 permits were detrimental to land conditions. WWP does not advocate for
implementation
of the No Action alternative. But it insists the EA should have analyzed this alternative to create a more effective baseline for understanding the merits of Alternatives One, Two, and Three.
WWP argues that BLM was required to include a detailed analysis of the No Action alternative under 40 C.F.R. § 1502.14, which requires inclusion of a no action alternative in every EIS to provide “a clear basis for choice among options.”
Id.
But the problem for WWP is that § 1502.14 does not apply. That provision governs the EIS, which is more comprehensive than the preliminary EA document at issue here.
Regulations require both documents to incorporate a range of reasonable alternatives, but the
depth
of discussion and analysis required is different depending on whether the document is an EIS or an EA. For example, section 1502.14 provides that an EIS should “[rigorously explore ... all reasonable alternatives,” and “[d]evote substantial treatment to each alternative” with “detail.”
Id.
at (a)-(b). In contrast, § 1508.9 describes the EA as “a concise public document” that “[bjriefly provide[s] sufficient evidence and analysis for determining whether to prepare an [EIS],” with “brief discussions of the need for the proposal, of alternatives ... [and] of the environmental impacts.”
Id.
at (a)-(b).
In
Davis v. Mineta,
we said that a court of appeals reviews an EA’s FONSI conclusion to determine whether an EIS should have been prepared. 302 F.3d at 1112. This involves asking “whether the agency acted arbitrarily and capriciously in concluding that the proposed action will not have a significant effect on the human environment.”
Id.
(quotations omitted).
Given the different standards for an EIS and an EA, the absence of a detailed No Action analysis by itself does not render this FONSI arbitrary and capricious.
WWP nevertheless argues that the EA was defective without a detailed No Action analysis because there was no adequate baseline for evaluating the three analyzed alternatives. This argument fails because Section III of the EA devoted 19 pages to describing “the baseline conditions within the Split Rock Ranch Allotments.” Aplt. Appx. at 219. The EA used this discussion “as a comparison for determining the effects of the alternatives on the critical elements of the human environment.”
Id.; see id.
at 219-238. Section IV then described the anticipated environmental consequences of each of the three analyzed alternatives. Both sections addressed the same resource categories — soil and water, vegetation, riparian/wetland, wildlife/fisheries, special status species, and cultural/socioeconomics — to “allow[ ] the reader to compare existing resource conditions to potential impacts for the same resources.”
Id.
at 238;
compare id.
at 219-238,
with id.
at 238-279.
WWP argues that the baseline was not detailed or robust enough, but these arguments are again premised on EIS requirements, not the more lenient EA requirements that actually govern. Under our highly deferential review, we cannot set aside the agency’s decision merely because the EA could have been more thorough than it was. WWP must show that the absence of a No Action analysis compromised the EA so severely as to render the FONSI arbitrary and capricious. 5 U.S.C. § 706(2)(A).
After carefully reviewing the EA, we conclude that the absence of a No Action analysis did not render the EA or the FONSI arbitrary and capricious.
2. No aggressive alternative
The EA expressly considered a “No Grazing” alternative, which would end grazing altogether on the Split Rock allotments. But as with No Action, it rejected this alternative without analyzing it. The EA explained that it followed the 1987 Lander RMP, which had determined that eliminating grazing from the lands was “not ... a viable or necessary option.” Aplt. Appx. at 210.
WWP argues that the EA should nevertheless have considered and analyzed a more aggressive, environmentally protective alternative that would remedy overgrazing problems more rapidly. It cites to
State of New Mexico,
where we rejected a BLM EIS because it failed to consider reasonable alternatives that were more environmentally protective than those analyzed. 565 F.3d at 709. WWP argues that, as in
State of New Mexico,
BLM failed to evaluate more protective alternatives. It notes that none of the analyzed alternatives incorporated all of the recommended Appropriate Actions from the RHS Assessment and that Alternatives One and Two were predicted to have similar outcomes — static or slight improvement or stable to increasing trends in most of the six standard areas. This, WWP argues, does not meet NEPA’s required “range” of alternatives.
But
State of New Mexico
also explained that an agency is not obligated to analyze options that it reasonably determines are outside its statutory mandate or are impractical or ineffective as judged against the agency’s objectives for a particular action or project.
See id.
at 708-09;
see also Bering Strait Citizens for Responsible Res. Dev. v. United States Army Corps of Eng’rs.,
524 F.3d 938, 955 (9th Cir.2008) (holding that an agency may reject alternatives that it determines to be “infeasible, ineffective, or inconsistent with the basic policy objectives”).
And an
agency has wide discretion in defining its objectives and in determining which alternatives meet those objectives.
See Wyoming v. United States Dep’t of Agric.,
661 F.3d 1209, 1244 (10th Cir.2011); see
also
43 C.F.R. § 1610.4-5 (“[T]he decision to designate alternatives for further development and analysis remains the exclusive responsibility of the BLM”). A court may not reject BLM’s stated objectives unless they are defined so narrowly as to foreclose reasonable options.
See State of New Mexico,
565 F.3d at 709.
WWP’s arguments concern one objective: improvement of rangeland health. This is an important objective, but it was not BLM’s sole concern. The agency also sought to fulfill its multiple use mandate, which is reflected in the Lander RMP, the FLPMA, and the Taylor Grazing Act — all of which contemplate livestock grazing on the land. The EA therefore crafted two alternatives that incorporated many of the Appropriate Actions, with adjustments to balance the competing goal of facilitating continued grazing.
BLM’s effort to balance these competing objectives is sufficient to explain its failure to pursue aggressive environmentally protective alternatives. We therefore cannot conclude that the range of alternatives selected rendered the EA arbitrary and capricious.
B.
Did BLM Take a “Hard Look” at the Environmental Effects of Its Action?
WWP argues that the EA failed to take a “hard look” at the environmental consequences of its Proposed Decision as required by NEPA. Specifically, WWP argues that the EA failed to take a hard look at carrying capacity and at the effects of its own Proposed Decision. In reviewing BLM’s decision, we ask whether it “was based on a consideration of the relevant factors and whether [BLM] made a clear error of judgment.”
See Davis,
302 F.3d at 1114. We do not substitute our judgment for BLM’s.
Id.
1. Carrying capacity
Average stocking levels on the Split Rock allotments during the initial permit were 8,054 AUM. Given the deterioration of rangeland health during this timeframe, BLM acknowledged that these stocking levels “exceeded] the carrying capacity of the
utilized
portions of the allotments.” Aplt. Appx. at 311. The Proposed Decision allowed grazing to continue at a reduced stocking level of 6,669 AUM and concluded that this reduction would be sufficient to remediate the rangeland deterioration. Neither the EA nor the Proposed Decision included a specific calculation of the current carrying capacity of the land.
Instead, the EA applied a qualitative approach based on detailed analyses of various individual components that affect carrying capacity, such as soils, upland vegetation, and riparian health. Section VI analyzed the impacts expected from each of the three alternatives, with separate consideration of how each RHS standard would fare under each alternative. The EA also discussed the effects of specific strategies within each alternative on the relevant RHS standards.
Five months after the EA was written and only a few days before the final decision was signed, BLM conducted a brief quantitative analysis of carrying capacity in a memo dated September 24, 2009. The memo provided mathematical calculations and relied on standards from scientific lit
erature to conclude that neither historical stocking levels nor the proposed stocking level exceed carrying capacity of the allotments.
WWP attacks both the EA’s qualitative approach and the September 24, 2009 quantitative analysis. As to the EA, WWP’s arguments presume that NEPA’s hard look requirement calls for a quantitative analysis of whether the reduced stocking levels contemplated in the Proposed Decision would be within carrying capacity. But we have found nothing in the record or the regulations suggesting that a quantitative calculation of carrying capacity is the only reasonable method for determining appropriate grazing limits. We agree with WWP that the EA could have provided a more rigorous quantitative evaluation, but it does not follow that the EA’s qualitative analysis was arbitrary and capricious.
WWP also attacks the quantitative analysis in the September 24, 2009 memo, insisting its conclusions that historic grazing and the proposed new permit terms were within carrying capacity are contrary to “the twin facts that (1) overgrazing has caused such damage here and (2) grazing pressure will not be significantly diminished by the BLM’s preferred plan.” Aplt. Br. at 49. But as Split Rock has argued, the qualitative analysis in the EA does not indicate that
uniform
overgrazing of the Split Rock allotments as a whole caused historical rangeland degradation. Poor range management and excessive use and overgrazing of
specific areas
contributed significantly to rangeland health issues. For example, cattle had been allowed to spend too much time in riparian areas; salt supplements and feed had been left near the water, exacerbating overuse of riparian resources; and Split Rock had apparently failed to comply with herding requirements to ensure more uniform grazing across the allotment. The Proposed Decision addressed these issues with a modest decrease in the AUM limit combined with improved range management techniques such as increased fencing and better herding compliance.
WWP has shown that the estimates in the September 24, 2009 memo are debatable and that BLM’s predictions may ultimately prove too optimistic. And WWP makes a solid case for more quantitative analysis in the EA. But this is not enough to meet their burden to show that the Proposed Decision is arbitrary and capricious. Mindful that our deference is most pronounced with respect to “technical or scientific matters within the agency’s area of expertise,”
Bosworth,
443 F.3d at 739, we cannot say that it was arbitrary and capricious for BLM to conclude that the stocking level in the Proposed Decision would not exceed carrying capacity.
2. Effects of the Proposed Decision
The BLM crafted its Proposed Decision by combining two of the three alternatives analyzed in the EA. Alternative One would have achieved rangeland health improvement primarily through rest rotation. Alternative Two would have achieved range-land health improvement primarily through a dramatic decrease in stocking level, with only deferred rotation rather than the more protective practice of rest rotation. The Proposed Decision combined the plans by eliminating the most protective feature of each. It included neither the dramatic decrease in stocking level nor rest rotation. But it did include other protective measures, such as fencing and herding requirements.
When an agency adopts a modified alternative that was not specifically analyzed in an EA or EIS, this raises a question of whether a supplemental assessment is needed. “An agency must prepare a supplemental assessment if the agency
makes substantial changes in the proposed action that are relevant to environmental concerns.”
State of New Mexico,
565 F.3d at 705 (quotations omitted). But if the modified alternative is “qualitatively within the spectrum of alternatives ... discussed” in the EA, no supplemental assessment is required.
Id.
(quotations omitted). The relevant question here is whether the impact of the Proposed Decision can be reasonably predicted from the analysis conducted in the EA.
See id.
at 707.
Although the EA did not analyze the Proposed Decision itself, it did provide a detailed analysis of many of the features ultimately included in the Proposed Decision, including the impacts of various stocking levels, deferred rotation, length of grazing season, herding compliance, fencing, and other features. As Split Rock notes, the Proposed Decision is most similar to Alternative Two. Both provide a deferred rotation system, a 204-day grazing season, a similar grazing schedule, frequent herding, and some additional fencing. The EA had concluded that Alternative Two would lead to some range-land health improvement but not as much improvement as Alternative One. The Proposed Decision is less environmentally protective than Alternative Two because it allows more cattle and allows grazing to begin earlier in the season. On the other hand, the Proposed Decision demands stricter compliance with herding requirements than did Alternative Two.
WWP illustrates its critique of BLM’s Proposed Decision with this analogy:
I want to increase my savings so I formulate two plans. In Plan AI will forgo a planned vacation, but continue eating lunch every day at Cafe Milano in downtown Tucson. In Plan B I will forgo eating at Cafe Milano, but will go on vacation. I decide to adopt a Hybrid Plan that partakes of both: I will keep eating at Cafe Milano every day, and also go on vacation.
Aplt. Br. at 53. Under this analogy, two alternative plans to save money are combined to create a hybrid plan that lacks the most effective features of either alternative and is therefore likely to be less successful in advancing the goal of saving money.
But this analogy demonstrates a critical problem with WWP’s argument: It calls into question the
wisdom
of BLM’s Proposed Decision, but not whether BLM could predict its effects. As we explain above, the relevant question is whether the impact of the Proposed Decision can be reasonably predicted from the EA’s analysis, not whether it is the best possible decision.
See State of New Mexico,
565 F.3d at 707. NEPA “merely prohibits uninformed — rather than unwise — agency action.”
Id.
at 704 (quotations omitted). Moreover, even though the Proposed Decision omits environmentally protective features from Alternatives One and Two, it nevertheless adds other features that are more environmentally protective than historical practice — features that were analyzed in the EA, such as fencing, herding, rest rotation, and fewer grazing days.
Our review of the EA and the Proposed Decision indicates that BLM analyzed the various components of the plan sufficiently to meet NEPA’s hard look requirement and did not act arbitrarily or capriciously.
III. CONCLUSION
For the foregoing reasons, we affirm the district court’s summary judgment ruling for BLM.