ORDER
The opinion filed on May 26, 2016, and reported at '823 F.3d 1268 is hereby amended. The amended opinion will be filed concurrently with this order.
. The opinion is amended as follows:
1. On 823 F.3d at 1270, after < For the foregoing reasons, we reverse the district court’s entry of summary judgment in part. > add < We remand this action to the district court with instructions to vacate the Secretary of the Interior’s Record of Decision unless the district court determines that this is one of the “rare circumstances, when [it is] advisable that the agency action remain in force until the action can be reconsidered or replaced....” Humane Soc’y of U.S. v. Locke, 626 F.3d 1040, 1053 n.7 (9th Cir. 2010). Whether, and when, construction of the transmission line and wind turbine complex will go forward absent vacatur may be relevant to that determination. >
2. On 823 F.3d at 1270, the final line of the opinion is changed to: < REVERSED and REMANDED. >
The Petition for Panel Rehearing is DENIED. No further petitions for rehearing or petitions for rehearing en banc will be entertained.
OPINION
BERZON, Circuit Judge:Renewable energy projects, although critical to the effort to combat climate change, can have significant adverse environmental impacts, just as other large-scale developments do. Here, the Oregon Natural Desert Association and the Audubon Society of Portland (collectively, “ONDA”) challenge a wind-energy development on the ground that the U.S. Bureau of Land Management’s (“BLM”) environmental review of the project did not adequately address impacts to the greater sage grouse, a relatively large ground-dwelling bird once abundant in the western United States. Greater sage grouse depend on sagebrush habitat for their survival. The challenged project entails the construction of wind turbines and a right-of-way across a sagebrush landscape in southeastern Oregon’s Harney County.-
We conclude that the BLM’s review did not adequately assess baseline sage grouse numbers during winter at the Echante site, where the wind turbines-are to be installed. As to that point, we reverse the district court’s entry of summary judgment in favor of the BLM, Harney County, and Columbia Energy Partners, the project developer. We also conclude, however, that ONDA did not exhaust its argument regarding genetic connectivity, and so we affirm as to that issue.
*565I.
A.The Project
The Echanis Wind Energy Project “is a 104-megawatt (MW) wind energy facility that would be constructed on a 10,500-aere privately-owned tract” on Steens Mountain in Harney County, Oregon. BLM, North Steens 230-kV Transmission Line Project Final Environmental Impact Statement (Oct. 2011) (“FEIS”) ES-1. Between 40 and 69 wind turbines would be built on the Echanis site. FEIS ES-11, 2-21, 3.1-2; see FEIS 2-22-23. The North Steens 230-kV Transmission Line, which involves “the construction, operation, and maintenance of a new [230-kilovolt] overhead electric transmission line and associated facilities on BLM-administered land,” would transport energy from the turbines to the electrical grid. FEIS ES-1-2. The entire undertaking—that is, both the transmission line and the turbine complex—(“the Project”), was approved in the BLM’s FEIS and Record of Decision (“ROD”) here challenged.
Columbia Energy Partners received a conditional use permit from Harney County to develop the Project, commissioned several studies of the Project, and secured a 20-year agreement to sell energy generated by the wind facility.1 FEIS ES-1. Because the right-of-way for the transmission line crosses public lands administered by the BLM, and the construction of the turbines is a “connected action,” 40 C.F.R. § 1508.25(a)(1), the entire Project is subject to environmental review under the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321, et seq. See FEIS 1-1.
The Echanis site was chosen because “[i]nitial site reconnaissance revealed wind-swept areas well exposed to prevailing west winds and—where present—significant ‘flagging’ of vegetation, indicating a robust westerly wind resource.” FEIS app. F at 6. This preliminary assessment was confirmed after a meteorological tower was erected at the site. Id. After considering three alternatives, the BLM chose a route for the transmission line and associated infrastructure that would cut across, in-part, the Steens Mountain Cooperative Management and Protection Area (“Steens Protection Area”). See, e.g., FEIS ES-3, ES-11,1-4-5.
B. Steens Mountain
Steens Mountain is many miles long and nearly 10,000 feet in elevation at its highest point. In 2000, Congress enacted the Steens Mountain Cooperative Management and Protection Act (“Steens Act”), which, among other things, established the Steens Protection Area and the Steens Mountain Wilderness Area. FEIS 1-19; see 16 U.S.C. § 460nnn, et seq. “The purpose of the [Steens Protection Area] is to conserve, protect, and manage the long-term ecological integrity of Steens Mountain for future and present generations.” 16 U.S.C. § 460nnn-12(a). Under the Steens Act, the “ecological integrity” that must be conserved, protected, and managed includes “the maintenance of ... genetic interchange.” 16 U.S.C. § 460nnn(5)(B). Steens Mountain, home to many sagebrush communities, lies near the center of one of the last remaining “strongholds of contiguous sagebrush habitat essential for the long-term persistence of greater sage-grouse.”
C. Greater Sage Grouse
The greater sage grouse is a sagebrush-obligate bird, meaning that it relies on *566sagebrush for its survival year-round. FEIS 3.5-22; Oregon Department of Fish & Wildlife, Greater Sage-Grouse Conservation Assessment and Strategy for Oregon: A Plan to Maintain and Enhance Populations and Habitat, Draft, March 1, 2011 (“Sage Grouse Strategy”) at 8. Sage grouse use different aspects of sagebrush habitats for various purposes. FEIS 3.5-22. For instance, at leks, “open areas surrounded by sagebrush,” male sage grouse strut and compete for female mates, displaying their elaborate plumage. FEIS 3.5-22; Sage Grouse Strategy at 8. In addition, sage grouse use sagebrush habitats for nesting and brood rearing. FEIS 3.5-22.
Sagebrush habitat is also essential for winter survival of sage grouse. “During the winter months, [the] greater sage-grouse’s diet consists almost entirely of sagebrush leaves and buds.” FEIS 3.5-23. To facilitate sagebrush consumption, sage grouse in the winter months “tend toward areas with high canopy and taller sagebrush plants_ Sagebrush must be exposed at least 9.8 to 11.8 inches (25 to 30 cm) above the snow level to provide adequate forage and cover.” FEIS 3.5-23. “[I]f sagebrush is covered with snow, greater sage-grouse will move to areas where the sagebrush is exposed.... The availability of sagebrush above the snow-pack is critical to the survival of greater sage-grouse through the winter.” FEIS 3.5-23-24; see also Sage Grouse Strategy at 10.
Once abundant across much of the western United States and Canada, the greater sage grouse now lives in “continually declining” and “increasingly separate” populations. FEIS 3.5-22. Since the 1950s, the overall population of sage grouse has declined by somewhere between 45% and 80%. Id. “Habitat loss and fragmentation are the primary cause[s] for long-term changes in [sage grouse] population abundance and distribution.” Sage Grouse Strategy at 1. As a consequence, “[m]ainte-nance of connectivity and reduction of fragmentation of sagebrush habitats is key to the long-term welfare of all ... sagebrush associated species.” Id. at 4. Oregon is unique in that, “[c]ompared to other states within the range of sage-grouse, [it] has large expanses of contiguous habitat with minimal threats of fossil fuel exploration or development.” Id. at x. “Oregon sage-grouse populations and sagebrush habitats likely comprise nearly 20% of the North American range wide distribution.” Id. at 2.
D. Environmental Review
The impacts of the Project on sage grouse were by far the most significant concern during the environmental review process at issue here. In the draft environmental impact statement (“DEIS”), FEIS, and ROD, the BLM adopted information, guidance, and mitigation measures concerning the sage grouse from the Oregon Department of Fish & Wildlife’s Mitigation Framework and Sage Grouse Strategy documents. FEIS ES-19, 3.5-21, 3.5-25-26; see C.A. Hagen, Mitigation Framework for Sage-Grouse Habitats, Aug. 23, 2011 (“Mitigation Framework”).
In response to the DEIS, ONDA submitted to the BLM numerous comments on a variety of issues, supporting the comments with scientific studies, wildlife management materials, and other documents. After the comment period ended, the BLM issued the FEIS and ROD, selecting the North Route transmission line as the preferred alternative to be implemented.2 *567FEIS ES-3. The North Route line would be approximately 46 miles long, connecting an electric substation at the Echanis site with an interconnection station near Crane, Oregon. FEIS ES-3.
Of relevance here, the FEIS acknowledged the “potential conflict between wind energy development and greater sage-grouse winter foraging habitats, because the windswept ridges that keep sagebrush exposed during winter months could also be ideal locations for wind energy development.” FEIS 3.5-25. Despite this concern, no surveys were conducted to determine if sage grouse are present at the Echanis site during .the winter months of November through April, Instead, the BLM assumed, based on surveys done at the nearby East Ridge and West Ridge sites, that no grouse use the Echanis site during winter.3 FEIS 3.5-25, 3.19-4. The FEIS stated that “no greater sage-grouse were found” at the East and West Ridge sites between late December and April, during the period of snow accumulation. FEIS 3.5-25. It explained:
The East Ridge and West Ridge projects are similar but potentially at even lower elevations [than is the Echanis site].... Because the Echanis Project area is generally covered with snow earlier and later in the season because of it’s [sic] relatively higher elevation, it is reasonable' to extrapolate winter use from the surveys at the East Ridge and West Ridge sites. Therefore, based upon these dáta, greater sage-grouse are assumed not to utilize the Echanis Project Area for winter habitat from the time that the vegetation is covered with snow until snowmelt, roughly December through April.
FEIS 3.5-25. '
As to connectivity concerns, the transmission lines and associated access and maintenance roads would physically divide sage grouse habitat, and the lines would provide perches for predatory raptors (such as hawks and eagles) and .corvids (such as ravens, chows, and jays). See Sage Grotise Strategy at 47-48. Accordingly, the FEIS assumed that grouse would avoid and be displaced from areas near , transmission lines arid poles. FEIS 3.5-43-44. This displacement, combined with possible avoidance of some areas due to noise or other project-related disturbances, would result in habitat fragmentation, as the lines transect otherwise contiguous grouse habitat. FEIS 3.5-43, 3.5-80.
E. This Litigation
After the FEIS and ROD issued, ONDA filed a complaint in the U.S. District Court for the District of Oregon, challenging environmental review of the Project under NEPA. Harney County and Columbia Energy Partners intervened. The parties then filed cross-motions for summary judgment.4 Ultimately, the court granted the defendants’ motions for summary judgment and denied ONDA’s.
We. reverse the district court’s ruling.
*568II.
ONDA asserts that the BLM’s review of the Project did not comply with NEPA, “our basic national charter for protection of the environment.” 40 C.F.R. § 1500.1(a). The centerpiece of environmental review under NEPA is the environmental impact statement (“EIS”), in which the responsible federal agency describes the proposed project and its impacts, alternatives to the project, and possible mitigation for any impacts. See 40 C.F.R. §§ 1500.1, 1502.15. NEPA imposes procedural requirements on federal agencies undertaking review; it does not mandate outcomes. WildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d 920, 924 (9th Cir. 2015).
NEPA challenges are reviewed under the Administrative Procedure Act (“APA”). Id. Under the APA, we ask whether the agency’s action was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). An agency’s action can be set aside
as arbitrary and capricious only if the agency relied on factors Congress did not intend it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation that runs counter to the evidence before the agency or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.
Lands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008) (en banc) (citations omitted), overruled on other grounds by Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008).
A. Baseline Winter Conditions
ONDA first contends that the BLM erred in failing directly to assess baseline conditions at the Echanis site, instead relying on an extrapolation from nearby sites to conclude that there is no greater sage grouse winter habitat at Echanis.
The establishment of a “baseline is not an independent legal requirement, but rather, a practical requirement in environmental analysis often employed to identify the environmental consequences of a proposed agency action.” Am. Rivers v. FERC, 201 F.3d 1186, 1195 n.15 (9th Cir. 1999). An EIS must “succinctly describe the environment of the area(s) to be affected ... by the alternatives under consideration,” 40 C.F.R. § 1502.15, and “insure that environmental information is available to public officials and citizens before decisions are made and before actions are taken,” id. § 1500.1(b) (emphases added). “Accurate scientific analysis ... [is] essential to implementing NEPA.” Id.
Applying these principles, several cases have found environmental analyses insufficient for failing to establish an environmental baseline. Indeed, as to another project planned for sage grouse territory, the BLM submitted comments to the U.S. Department of Transportation (“DOT”) urging DOT to assess baseline winter sage-grouse populations along a rail line. See N. Plains Res. Council v. Surface Transp. Bd., 668 F.3d 1067, 1084 (9th Cir. 2011). Specifically, with regard to DOT’S EIS, the BLM commented that “[w]ith the increasing importance of sage grouse, more discussion on sage grouse is needed, including discussion on wintering areas.... Sage grouse inventories need to be conducted at least two miles from any proposed disturbance.” Id. Northern Plains Resource Council agreed with the BLM that DOT did not adequately assess baseline conditions for the challenged project. Id. at 1083-85. Similarly, Half Moon Bay ruled that analysis in an EIS was inadequate because it failed to assess baseline under*569water conditions at a site where it was proposed dredged materials would be dumped. Half Moon Bay Fisherman’s Marketing Ass’n v. Carlucci, 857 F.2d 505, 510 (9th Cir. 1988).
Under Northern Plains Resource Council and Half Moon Bay, the BLM had a duty to assess, in some reasonable way, the actual baseline conditions at the Echanis site. Baseline conditions were particularly important here because impacts to sage grouse were by far the most significant concern during environmental review, and the unique features of winter habitat are essential to sage-grouse survival. See, e.g., Sage Grouse Strategy at 47, 83. Baseline conditions at the Echanis site thus warranted comprehensive study in the FEIS.
The FEIS did not report on any observations of the Echanis site surveying winter sage grouse use of the area. Instead, the FEIS assumed that sage grouse are absent from the site during winter. FEIS 3.5-25. To justify this assumption, the FEIS relied on data from the East and West Ridge sites, located near the Echanis site but at generally lower elevations. Id. In doing so, the FEIS asserted that, although 36 sage grouse were found at the East Ridge site on December 17, 2008 and nine birds were found on the West Ridge site on December 11, “no greater sage-grouse were found later in December, or in January, February, March, or April, during the time that snow had accumulated.” Id. The FEIS then explained that its extrapolation from surveys at these sites was reasonable because the “potentially” lower elevation of the sites, as compared to the Echanis site, indicated that it is more likely that snow would accumulate at Echanis earlier and dissipate later in winter. It is less likely, the FEIS asserted, that sage grouse use the Echanis site than the East and West Ridge sites in winter. Id. A fundamental flaw infects this reasoning.
Contrary to what the FEIS stated, four sage grouse were found at the East Ridge site—the surveyed site closer to Echanis— during February, indicating that some sage grouse do spend the winter there. The FEIS thus did not comply with the requirement to provide “[accurate scientific analysis,” which is “essential to implementing NEPA,” 40 C.F.R. § 1500.1(b), or with the agency’s obligation to “insure the professional integrity, including scientific integrity, of the discussions and analyses in [EISs],” id. § 1502.24.
Further, that some grouse were found at the East Ridge site in mid-winter greatly undermines the validity of the BLM’s assumed absence of sage grouse at the Echanis site. Given that grouse do use the East Ridge site during the winter, the BLM’s own extrapolation method should have resulted in assuming the birds’ presence, not their absence.
The record as a whole confirms the validity of this contrary assumption. Christian Hagen, the Oregon Department of Fish & Wildlife sciéntist who prepared the Mitigation Framework, suggested that, if grouse were still present at the East and West Ridge sites in December, the conditions were probably right to spend the entire winter there; in fact, as noted, grouse were present in February. Further, several sources on which the FEIS relied, and the FEIS itself, acknowledge that the wind-swept character of the Echanis site— the aspect of the site that makes it ideal for wind-energy generation—suggests it could be good winter habitat for sage grouse, despite its “potentially” higher elevation, as snow there may be blown off sagebrush and exposed for grouse to eat. See, e.g., FEIS 3.5-25, app. F at 6; Sage Grouse Strategy at 47. And scientists and cooperating agencies recommended to the *570BLM either that actual winter surveys of sage grouse be conducted or, if not, that the BLM assume sage grouse were present at the Echanis site during the entire winter. See, e.g., Sage Grouse Strategy at 86.
In short, the FEIS’s inaccurate data concerning the closer East Ridge site that was surveyed rendered its assumption concerning the winter presence of sage grouse at the Echanis site arbitrary and capricious. See Lands Council, 537 F.3d at 987.
The defendants maintain that the BLM is owed special deference when undertaking scientific or technical analysis within its purview, which it is. See Lands Council, 537 F.3d at 993. But deference does not excuse the BLM from ensuring the accuracy and scientific integrity of its analysis, a NEPA requirement. See 40 C.F.R. §§ 1500.1(b), 1502.24, The defendants also posit that invalidating the BLM’s assessment of winter conditions at Echanis, and therefore requiring the BLM to gather better information, would essentially impose a procedural requirement not derived from NEPA. But we do not hold that habitat extrapolations from one site to another are impermissible. Instead, our holding is that any such extrapolation must be based on accurate information and defensible reasoning.
The errors in the BLM’s analysis were not harmless. See Tucson Herpetological Soc’y v. Salazar, 566 F.3d 870, 880 (9th Cir. 2009). The inaccurate information and unsupported assumption materially impeded informed decisionmaking and public participation. See id.; cf. Montana Snowmobile, 790 F.3d at 926. Without appropriate data regarding sage grouse use of the Echanis site during the winter, whether direct or via a supportable extrapolation, it was not possible to begin to assess whether sage grouse would be impacted with regard to access to viable sagebrush habitat in the winter months.
In addition, had the BLM assumed the Echanis site, provides winter sage grouse habitat, rather than that it does not, the site would be deemed “Category-1 Habitat” pursuant to the Sage Grouse Strategy and Mitigation Framework. Under that designation and the mitigation measures adopted in the FEIS and ROD, the Project would not go forward there.5 FEIS 3.5-53; ROD 14-15; see Sage Grouse Strategy at 83, 86; Mitigation Framework at 1-2. In that respect, the BLM’s analysis materially affected the outcome of environmental review. See Idaho Wool Growers Ass’n v. Vilsack, 816 F.3d 1095, 1104 (9th Cir. 2016).
The defendants urge that the mitigation measures adopted in the FEIS cured any potential prejudice resulting from a faulty baseline analysis.6 Mitigation measures, *571however, while relevant to the adequacy of an environmental analysis, see City of Sausalito v. O’Neill, 386 F.3d 1186, 1212-13 (9th Cir. 2004), are not a panacea for inadequate data collection and- analysis. More specifically, they do not address the-concerns relevant to the prejudice analysis: the error’s effect on informed decisionmak-ing and public participation, and on the outcome of the decision. See, e.g., Cal. Wilderness Coal. v. U.S. Dep’t of Energy, 631 F.3d 1072, 1093 (9th Cir. 2011). Here, with baseline conditions inadequately established, the public was not able to tailor its comments to address concerns regarding the potential winter presence of sage grouse at the Eehanis site. Nor was the BLM’s explanation of the impacts to winter grouse habitat adequately informed. Having no reasonable assessment as to whether sage grouse are present at the Eehanis site in winter, the BLM could not assess the Project’s impacts to them, qualitatively or quantitatively. And with the impacts on sage grouse not properly established, the BLM did not know what impacts to mitigate, or whether the mitigation proposed' would be adequate to offset damage to wintering sage grouse. Most importantly, had the BLM assumed the presence of sage grouse, rather than their absence, the Eehanis site would be deemed Category-1 Habitat, and the mitigation measures adopted in the FEIS and ROD would not allow development to go forward there.
Given the BLM’s prejudicial error, the district court’s entry of summary judgment in favor of the defendants must be reversed.
B. Genetic Connectivity
ONDA next asserts that the FEIS erroneously failed to address genetic connectivity between sage grouse populations. “Genetic connectivity” means the extent to which separate populations of a speciés are able to share genes and thereby to maintain a healthy genetic diversity within each population. The defendants argue that ONDA failed to exhaust this argument during environmental review. They also point out that the FElS and ROD adequately addressed the more general issue of habitat connectivity and fragmentation. We agree with the defendants.
Judicial review is available for NEPA challenges under the APA only if the NEPA plaintiffs exhaust their administrative remedies. Great Basin Mine Watch v. Hankins, 456 F.3d 955, 965 (9th Cir. 2006); see 5 U.S.C. § 704. ONDA did not use the term “genetic connectivity” in its comments on the draft EIS, nor did it make specific arguments about that issue, separately from the more general issues of habitat connectivity and fragmentation.
The closest ONDA came to raising the genetic connectivity argument as a distinct issue was the following comment to the BLM:
According to [the U.S. Fish & Wildlife Service’s] candidate species listing decision, leks within 18 km (11.2 miles) of each other have common features, such as genetic characteristic^] (genetic evi*572dence proves that exchange between different leks by individual birds has not been restricted), compared to leks farther away. [The U.S. Fish & Wildlife Service] used this distance to determine connectivity between lek sites. Because Echanis and the transmission line alternatives fall within the parameters of connectivity, there will likely be impacts on the' ability for sage-grouse’ to move across the landscape to lek sites for breeding and courtship.
Comment Letter from ONDA to BLM (September 17, 2010), at 35-36 (“ONDA Comment Letter”) (footnote omitted). Notably, the foregoing statement mentions the issue of genetic exchange only as a premise to assert that the Project will affect connectivity in general. This statement, although related to genetic connectivity concerns, is not sufficient— particularly in light of ONDA’s otherwise extremely comprehensive comments—to alert the BLM that ONDA was asking for a genetic connectivity analysis • regarding separate sage grouse populations.
In one of its comment letters ONDA, quoted a BLM instruction memorandum, which articulated BLM policy regarding the designation of priority habitat for various wildlife species, as stating that “[p]ri-ority habitat will be areas of high quality habitat supporting important sage-grouse populations, including those populations that are vulnerable to localized extirpation but necessary to maintain range-wide connectivity and genetic diversity.” ONDA Comment Letter at 25 (emphasis added) (quoting BLM, IM 2010-071, Gunnison and Greater Sage-grouse Management Considerations for Energy Development (2010)). In context, this statement was insufficient to put the BLM on notice that ONDA sought discussion of the genetic connectivity issue as such. The overall comment of which the statement was a part asserted that the DEIS did not demonstrate that the BLM properly followed its own statutory, regulatory, and guidance mandates. The comment referred to “range-wide connectivity” and “genetic diversity” only as a subpremise of its insistence that the BLM should follow mandates regarding the designation of priority habitat for many species, including sage grouse. This opaque comment was not adequately specific to alert the BLM that it should analyze the substantively distinct issue of cross-population genetic connectivity.
Elsewhere, ONDA’s comments address connectivity and fragmentation at length, but only in a general sense, not specifically with regard to cross-population genetic connectivity. See ONDA Comment Letter at 22 (noting that the Project will “riddle[]” core sage grouse habitat in the Steens Protection Area with turbines, transmission lines, roads, and other infrastructure), 26 (noting the BLM policy that “it is imperative that fragmentation and degradation of ... greater sage-grouse habitat not continue to the point that sustainable sage-grouse populations can no longer be supported”), 33 (“Habitat fragmentation and disturbance across much of the sage-grouse’s range has contributed to significant population declines over the past century. If current trends persist, many local populations may disappear in the next several decades, with the remaining fragmented population vulnerable to extinction.”), 34-35 (referring to habitat fragmentation), 36 (referring to “connectivity corridors”), 37 (referring to “new habitat loss and fragmentation”), 38 (noting “habitat loss, fragmentation, or degradation” concerns), 39 (referring to “the need for excluding habitat-fragmenting activities from sage-grouse core habitat”), 80-82 (referring to habitat fragmentation), 86 (referring to habitat fragmentation and *573degradation); Letter from ONDA to Secretary, Dep’t of Interior (November 15, 2011), at 1 (referring to habitat fragmentation and a “fragmented population”); Letter from' ONDA to Secretary, Dep’t of Interior (July 21, 2011), at 2, (“fragment vital habitat”), 3 (“loss of irreplaceable habitat ... due to fragmentation”), 5 (“wildlife habitat connéctivity”); Supplemental Comment Letter from ONDA to BLM (January 26, 2011), at 2 (“[H]arm to sage-grouse on Steens would be exacerbated by habitat fragmentation.... ”). Because ONDA’s comments refer only to overall habitat connectivity and fragmentation, they were too vague to raise the specific genetic connectivity issue regarding separate populations as a distinct concern. Great Basin Mine Watch, 456 F.3d at 967.
Barnes v. U.S. Department of Transportation confirms that ONDA’s comments were inadequate to exhaust the genetic connectivity issue. See 655 F.3d 1124, 1133, 1135 (9th Cir. 2011). Barnes ruled that where a commenter could have been “more expansive or more detailed with’his comments,” he nonetheless adequately raised his concerns regarding increased air traffic by specifically noting the “increased air traffic” the project would cause. Id. at 1133. By contrast, Barnes held that com-menters did not adequately raise the issue of impacts caused by a potential new control tower, because the comments “did not include one single reference to a new control tower.” Id. at 1135. Here, ONDA did not use the phrase “genetic connectivity” anywhere in its comments, nor did it‘raise any distinct concern regarding genetic interchange between otherwise separate sage grouse populations. Under Barnes, therefore, ONDA did not put the BLM sufficiently on notice that it should address genetic connectivity in the FEIS.
Contrary to ONDA’s assertion, genetic connectivity and the distinction between genetic connectivity and habitat connectivity are not such obvious issues that ONDA had no obligation to raise them to the agency. ONDA cites Department of Transportation v. Public Citizen for the proposition that “an EIS’s flaws might be so obvious that there is no need for a commentator to point them out specifically in order to preserve its ability to challenge a proposed action.” See 541 U.S. 752, 765, 124 S.Ct. 2204, 159 L.Ed.2d 60 (2004). But Public Citizen decided that the issue of alternatives to the project under review was not so obvious that it did not have to be exhausted. See id. If the analysis of alternatives in Public Citizen was not an obvious issue, then -the much more specific issue of genetic connectivity, as well as the fairly nuanced distinction between genetic connectivity and habitat connectivity, are also not that “obvious.”
ONDA also cites the Steens Act’s reference to “genetic interchange,” 16 U.S.C. § 460nnn(5)(B); several passages in the Sage Grouse Strategy, Sage Grouse Strategy at 10, 57; discussion by the U.S. Fish & Wildlife Service in its March 2010 determination not to list the sage grouse as an endangered species; and mention in the proposed Resource Management Plan and Final Environmental Impact Statement for the Andrews Management Area, which covers some of the land that the proposed right of way would cross, to suggest that genetic connectivity is an important issue and therefore that the FEIS should have discussed it. But where no other exceptions to the exhaustion requirement are met, the importance of genetic connectivity to the health and wellbeing of the greater sage grouse does not excuse ONDA from raising the issue to the BLM so that the agency could “give the issue meaningful consideration.” Great Basin Mine Watch, *574456 F.3d at 965 (citations omitted).7
ONDA cites Oregon Natural Desert Association v. Bureau of Land Management, 625 F.3d 1092, 1112 (9th Cir. 2008), for the proposition that the Steens Act’s mention of the BLM’s duty to manage the long-term ecological integrity of the Steens Protection Area, including the maintenance of genetic interchange, indicates that the BLM had to discuss genetic connectivity in the FEIS. But the statutes at issue in Oregon Natural Desert Association specifically required the BLM to undertake review of certain issues, not, as is the case with the Steens Act, to undertake management. Because the Steens Act does not impose a requirement to review the issue of genetic interchange, Oregon Natural Desert Association is not pertinent.
Although ONDA never raised the genetic connectivity issue, as such, to the BLM, it suggests that it did so by making a distinction between mfer-population connectivity (essentially the issue of genetic connectivity) and mira-population connectivity. Not so. Although ONDA now clearly distinguishes between inter- and intra-pop-ulation connectivity, its comments to the BLM did not mention that distinction. For that reason as well, the comments were simply not “structure[d] ... [to] alert[ ] the agency to [ONDA’s] positions and contentions.” Barnes, 655 F.3d at 1132 (citations and alterations omitted).
ONDA also maintains that it raised the inter-population connectivity issue by pointing to what it calls the Steens Mountain connectivity corridor. But ONDA did not specifically identify the Steens Corridor in its comments; it only referred generally to “connectivity corridors.” Again, ONDA did not point to the inter-population connectivity issue as a separate, more specific concern that the BLM should address in the FEIS.
In short, ONDA never raised the issue of cross-population genetic connectivity, specifically, to the BLM, either by using the term “genetic connectivity” or by making the inter- versus intra-population connectivity distinction, or by referring to the Steens Corridor. The BLM responded to comments regarding habitat connectivity and fragmentation issues at the level of detail at which those comments were presented. See FEIS 3.5-22-23, 3.5-25-26. Because ONDA never brought the issue of inter-population or genetic connectivity to the BLM’s attention during the environmental review process, we conclude that the issue was not exhausted and is not now subject to review.8’9
*575III.
For the foregoing reasons, we reverse the district court’s entry of summary judgment in part, We remand this action to the district court with instructions to vacate the Secretary of the Interior’s Record of Decision unless the district court determines that this is one of the “rare circumstances, when [it is] advisable that the agency action remain in force until the action can be reconsidered or replaced— ” Humane Soc’y of U.S. v. Locke, 626 F.3d 1040, 1053 n.7 (9th Cir. 2010). Whether, and when, construction of the transmission line and wind turbine complex will go forward absent vacatur may be relevant to that determination.
REVERSED and REMANDED.
. Although the entire Project includes both the transmission line and the wind turbines, the proposed action alternatives varied only as to the route the transmission line would *567take. Harney County issued the permit to build the wind turbines at the Echanis site, which is private land, so the proposed turbines would be built there and nowhere else. For that reason, other than the no-action alternative, the FEIS and ROD did not consider alternative sites for the turbines.
. The East and West Ridge sites were originally also slated for wind-energy development, but applications for those projects were withdrawn. FEIS 3.19-2-3.
. In support of its motion, the district court permitted ONDA to supplement the record with expert declarations. The defendants object to ONDA’s reliance on these declarations. The declarations do not affect our decision, so we do not address the objection.
. The Sage Grouse Strategy, which the BLM relied on for its sage grouse conservation approach, FEIS ES-19, 3.5-25-26, identifies winter habitat as " Category-1,” meaning that . it is "essential for greater sage-grouse populations and is limited by the inability to mitigate for habitat loss ... in [a] reasonable time frame, and is irreplaceable.” Sage Grouse Strategy at 86. Pursuant to that designation, the "mitigation goal ... is no loss of either habitat quality or quantity,” so impacts are to be "avoid[ed] through [the use of] alternatives to the proposed development action,” or, if impacts cannot be avoided through the use of alternatives, there should be “[n]o authorization of the proposed development action.” Id. In addition, the Mitigation Framework recommends ■ that impacts to Category-1 Habitat "with documented sage-grouse presence” be avoided, as such habitat is "both essential and irreplaceable." Mitigation Framework at 1-2.
. The FEIS and ROD incorporated the sage grouse mitigation measures recommended in the Mitigation Framework. See FEIS ES-18-19. These measures include the no-impact recommendation for Category-1 Habitat, described supra n.5, as well as a "no net loss *571with a net benefit” strategy, leading to the rehabilitation of approximately 9,000 acres of non-Project-area sagebrush habitat to make up for the habitat potentially lost within the Project-area, FEIS ES-18-19; see Mitigation Framework at 1-2, 5-6. To effect these measures, the BLM,’ Harney County, and Columbia Energy Partners could take measures to improve the quality of grouse habitat, including could securing conservation easements for or purchasing particular mitigation sites. Mitigation Framework at 6. In addition, modeling and monitoring strategies would be used to assess the effectiveness of proposed and implemented rehabilitation of given areas, respectively. See, e.g., id. at 2, 5.
.ONDA has not argued that the BLM had "independent knowledge” of a genetic connectivity impact of the Project, so that exhaustion was not required. See Barnes, 655 F.3d at 1132. Nor has ONDA argued that other com-menters raised the issue of genetic connectivity, and therefore that it did not have to exhaust the issue. See, e.g., Shasta Res. Council v. U.S. Dep’t of Interior, 629 F.Supp.2d 1045, 1052 (E.D. Cal. 2009). At most, ONDA cites sections of the U.S. Fish & Wildlife Service's March 2010 decision not to list the sage grouse as endangered to suggest that genetic connectivity was a live issue in this administrative process. The discussion in that decision, however, is not a comment on this Project's impact on genetic interchange.
. ONDA made the inter-population connectivity argument clearly enough to the district court that it ruled on the issue. It thus was not waived. See. State of Ariz. v. Components Inc., 66 F.3d 213, 217 (9th Cir. 1995).
. We note that the exhaustion analysis in this case is unusual, as the issue of genetic connectivity is a technical, specific issue that in this context required clear differentiation from the general habitat connectivity issue. The same mode of analysis might not apply to cases involving the exhaustion of more generic issues.