Full opinion text
GREGORY, Chief Judge: These consolidated cases present two challenges to agency actions that provided necessary approvals for the Atlantic Coast Pipeline (ACP). The challenges were brought by Defenders of Wildlife, the Sierra Club, and the Virginia Wilderness Committee (collectively, "Petitioners"). For the reasons that follow, we conclude that both agency decisions are arbitrary and capricious. The first petition, No. 18-1083, concerns the U.S. Fish and Wildlife Service (FWS). Pursuant to the Endangered Species Act, FWS issued an Incidental Take Statement (ITS) authorizing the pipeline to "take"-i.e., kill, harm, or harass-five species that are listed as threatened or endangered. Petitioners challenged the ITS as arbitrary and capricious under § 706 of the Administrative Procedure Act (APA) because the amount of take authorized in the ITS (known as the "take limit") cannot be enforced. Petitioners identify two flaws that make the take limits unenforceable: first, FWS failed to set numeric limits on take of the five threatened and endangered species, and second, FWS failed to comply with the requirements for using habitat as a surrogate for a numeric limit. Although FWS is not required to set a numeric limit, it can only use a habitat surrogate if it demonstrates a causal link between the species and the delineated habitat, shows that setting a numerical limit is not practical, and sets a clear standard for determining when incidental take is exceeded. Here, FWS failed some or all of these requirements for all five challenged species. As such, FWS's take limits are not enforceable and therefore arbitrary and capricious. The second petition, No. 18-1082, concerns the U.S. National Park Service (NPS). The pipeline's proposed route intersects the Blue Ridge Parkway, a unit of the National Park System managed by NPS. Invoking the Blue Ridge Parkway Organic Act, NPS issued a right-of-way permit allowing the pipeline to drill and pass underneath the Parkway surface. The pipeline will also carve a path through a nearby forest, affecting views from the Parkway's scenic overlooks. Petitioners Sierra Club and the Virginia Wilderness Committee argue that NPS lacked the authority to grant a right-of-way to a gas pipeline and that doing so violated the statutory mandate that agency decisions not be inconsistent with the Parkway's conservation purpose. As detailed below, we assume for purposes of this case that NPS has the requisite statutory authority but because NPS does not explain how the pipeline crossing is not inconsistent with the purposes of the Parkway and the overall National Park System, the permit decision is arbitrary and capricious. Part I of this opinion will provide a brief background and address a statute of limitations question common to both cases. Part II will provide the relevant background facts and legal analysis for No. 18-1083 (FWS), while Part III will provide the relevant background facts and legal analysis for No. 18-1082 (NPS). Finally, having concluded that the respective agencies erred, Part IV will address a question of remedy common to both cases. I. A. The ACP is a 600-mile pipeline designed to transport natural gas from Harrison County, West Virginia, to the eastern portions of Virginia and North Carolina. J.A. 234. Constructing the pipeline would generally require a 125-foot right-of-way for most of the distance, which will disturb 11,776 acres of land. J.A. 553. Once completed, ACP would generally maintain a 50-foot permanent right-of-way along the length of the pipeline. J.A. 325. Under the Natural Gas Act, the Federal Energy Regulatory Commission (FERC) is the agency responsible for giving final approval, in the form of a certificate of public convenience and necessity, for the construction of natural gas pipelines. 15 U.S.C. § 717f. The Natural Gas Act also requires applicants such as ACP to obtain "any permits, special use authorizations, certifications, opinions, or other approvals as may be required under Federal law." N.Y. Dep't of Envtl. Conservation v. FERC , 884 F.3d 450, 452-53 (2d Cir. 2018). FERC serves as the "lead agency" responsible for "coordinating all applicable Federal authorizations." 15 U.S.C. § 717n. On October 13, 2017, FERC issued ACP a certificate of public convenience and necessity that authorized the construction and operation of the pipeline. J.A. 234-389. As it has done in other cases, FERC conditioned its approval of the pipeline on ACP receiving all "state and other federal authorizations required for the proposed project." Del. Riverkeeper Network v. Sec'y Pa. Dep't of Envtl. Prot. , 833 F.3d 360, 368 & n.5 (3d Cir. 2016) ; see J.A. 362-84 (listing conditions ACP must satisfy). Two of these authorizations must come from FWS and NPS. On October 16, 2017, FWS issued a biological opinion and incidental take statement that authorized the pipeline to take several endangered and threatened species. On December 12, 2017, NPS issued a right-of-way permit authorizing the pipeline to cross the Blue Ridge Parkway. Petitioners challenged both agency actions in this Court on January 19, 2018. B. The first question we must resolve is whether the Petitioners filed their challenges within the applicable statute of limitations. Federal Rule of Appellate Procedure 15(a) provides that "[r]eview of an agency order is commenced by filing, within the time prescribed by law , a petition for review[.]" Fed. R. App. P. 15(a)(1) (emphasis added). The Petitioners filed their challenges within 95 days (No. 18-1083) and 38 days (No. 18-1082) of the agency decisions, respectively. In their opening brief, Petitioners invoked the six-year statute of limitation created for claims "against the United States," which applies generally to challenges brought under the APA. See 28 U.S.C. § 2401(a) ; Jersey Heights Neighborhood Ass'n v. Glendening , 174 F.3d 180, 186 (4th Cir. 1999) (applying § 2401(a) to APA challenges). ACP argues that these challenges were actually brought under the Natural Gas Act, 15 U.S.C. § 717r(d)(1), which lacks a statute of limitations. Thus, ACP urges us to adopt the most closely analogous state law statute of limitations. ACP Resp. Br. 18-27 (citing Reed v. United Transp. Union , 488 U.S. 319, 323-24, 109 S.Ct. 621, 102 L.Ed.2d 665 (1989) ; Agency Holding Corp. v. Malley-Duff & Assocs., Inc. , 483 U.S. 143, 147, 107 S.Ct. 2759, 97 L.Ed.2d 121 (1987) ; DelCostello v. Int'l Bhd. of Teamsters , 462 U.S. 151, 158-59, 103 S.Ct. 2281, 76 L.Ed.2d 476 (1983) ). ACP proposes the 30-day limitations period applicable to petitions for review of state agency actions in West Virginia, Virginia, and North Carolina (the states through which the pipeline will be built). Such a short statute of limitations period, ACP argues, would comply with Congress's intent to create an expedited review process for agency decisions related to natural gas pipelines. The problem with ACP's argument is that the cases on which it relies predate Congress's establishment of a four-year default statute of limitations for any "civil action arising under an Act of Congress enacted after" December 1, 1990. Judicial Improvements Act of 1990, Pub. L. No. 101-650, § 313, 104 Stat 5089, 5114-15 (codified at 28 U.S.C. § 1658(a) ). This four-year default provision applies to any claim "made possible by a post-1990 enactment." Jones v. R.R. Donnelley & Sons Co. , 541 U.S. 369, 382, 124 S.Ct. 1836, 158 L.Ed.2d 645 (2004). Congress specifically enacted § 1658(a) to "alleviat[e] the uncertainty inherent in the practice of borrowing state statutes of limitations." Id. As Petitioners point out, Congress added 15 U.S.C. § 717r(d) in 2005. Energy Policy Act of 2005, Pub. L. No. 109-58, § 313(b), 119 Stat 594, 689-90. If § 717r(d) is the source of Petitioners' cause of action, then § 1658(a) applies and the statute of limitations is four years. The only other federal court to consider this issue also looked to § 1658(a) in declining to apply a more restrictive state limitations period. Del. Riverkeeper Network v. Sec'y of Pa. Dep't of Envtl. Prot. , 870 F.3d 171, 179 (3d Cir. 2017) (concluding that petitioner's challenge was timely under either § 1658(a) or the equitable doctrine of laches). We need not decide whether the Natural Gas Act, which ACP cites as the cause of action, or the APA, which Petitioners cite as theirs, provides the applicable statute of limitations. Petitioners filed their challenges to FWS and NPS after 95 and 38 days, respectively. By either measure, the petitions are timely. II. We turn first to petition No. 18-1083, in which Petitioners challenge the take limits set by FWS for five species that will be negatively impacted by the pipeline. We begin with the Endangered Species Act and the exception created for "incidental" take of threatened and endangered species. We next discuss how the Endangered Species Act obligated FWS to analyze how the pipeline would affect threatened and endangered species. We then summarize the statutory and regulatory requirements FWS must satisfy to issue a proper incidental take statement. Finally, we review in detail the agency's determinations about the five species at issue. We conclude that for each species, FWS failed to satisfy the requirements for a habitat surrogate and therefore failed to create enforceable take limits. A. 1. Congress enacted the Endangered Species Act in 1973 "to protect and conserve endangered and threatened species and their habitats." Nat'l Ass'n of Home Builders v. Defs. of Wildlife , 551 U.S. 644, 651, 127 S.Ct. 2518, 168 L.Ed.2d 467 (2007). In doing so, Congress made a "conscious decision ... to give endangered species priority over the 'primary missions' of federal agencies." Tenn. Valley Auth. v. Hill , 437 U.S. 153, 185, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978). Pursuant to Endangered Species Act § 4, FWS determines what species are endangered or threatened and designates their critical habitats. 16 U.S.C. § 1533 ; see Nat'l Ass'n of Home Builders , 551 U.S. at 651, 127 S.Ct. 2518 ; 50 C.F.R. § 17.11 (listing endangered and threatened wildlife). To protect threatened and endangered species, Endangered Species Act § 9 and FWS regulations prohibit their "take." 16 U.S.C. § 1538(a)(1)(B) (endangered species); 50 C.F.R. § 17.21(c) (endangered species); 50 C.F.R. § 17.31 (threatened species). To "take" a species is "to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect, or to attempt to engage in any such conduct." 16 C.F.R. § 1532(19). Harm and harassment include the disruption of normal behavioral patterns and indirect injury caused by habitat modification. 50 C.F.R. § 17.3 ; Babbitt v. Sweet Home Chapter of Cmtys. for a Great Or. , 515 U.S. 687, 702-04, 115 S.Ct. 2407, 132 L.Ed.2d 597 (1995). "Any person" who knowingly takes an endangered or threatened species is "subject to substantial civil and criminal penalties, including imprisonment." Bennett v. Spear , 520 U.S. 154, 170, 117 S.Ct. 1154, 137 L.Ed.2d 281 (1997) ; see 16 U.S.C. § 1540(a), (b). In 1982, Congress created a narrow exception to the prohibition against take: when "such taking is incidental to, and not the purpose of, the carrying out of an otherwise lawful activity." 16 U.S.C. § 1539(a)(1)(B) ; see Endangered Species Act Amendments of 1982, Pub. L. No. 97-304, § 6, 96 Stat. 1411, 1422. To take a species under this exception, agencies such as FERC-or private entities taking species pursuant to agency authority, such as ACP-must receive a valid Incidental Take Statement from FWS. 16 U.S.C. § 1536(b)(4) ; 50 C.F.R. § 402.14(g)(7), (i). The amount of take set by the ITS creates a " 'trigger' that, when reached, results in an unacceptable level of incidental take." Ariz. Cattle Growers' Ass'n v. U.S. Fish & Wildlife, Bureau of Land Mgmt. , 273 F.3d 1229, 1249 (9th Cir. 2001). 2. Building a natural gas pipeline implicates a number of federal laws, including the Endangered Species Act. J.A. 545-52. Pursuant to Endangered Species Act § 7, FERC must ensure that "any action authorized, funded, or carried out by such agency ... is not likely to jeopardize the continued existence of any endangered species or threatened species or result in the destruction or adverse modification of habitat of such species." 16 U.S.C. § 1536(a)(2). To satisfy this requirement, FERC must formally consult with FWS whenever a pipeline "may affect listed species or critical habitat." 50 C.F.R. § 402.14(a). FWS then provides FERC with a written statement (called a Biological Opinion) explaining "how the proposed action will affect the species or its habitat." Bennett , 520 U.S. at 158, 117 S.Ct. 1154 ; see 16 U.S.C. § 1536(b)(3). If FWS concludes that the pipeline will adversely affect the species but "will not result in jeopardy or adverse habitat modification," then it must provide FERC with an ITS authorizing the anticipated incidental take and specifying the "impact of such incidental taking on the species." 16 U.S.C. § 1536(b)(4) ; see Bennett , 520 U.S. at 158, 117 S.Ct. 1154. In July 2017, FERC requested formal consultation with FWS about the ACP. J.A. 391. Three months later, FWS issued a Biological Opinion that addressed six threatened and endangered non-plant species: the Roanoke Logperch (a fish), the Clubshell (a mussel), the Rusty Patched Bumble Bee, the Madison Cave Isopod (a crustacean), the Indiana Bat, and the Northern Long-Eared Bat. J.A. 390-439. FWS concluded that the pipeline as a whole would not "jeopardize the continued existence of" the six species. J.A. 430-39. But because pipeline will adversely affect individuals from each species, FWS issued an ITS that set out the "amount or extent of take anticipated" by the ACP. J.A. 439-50. Petitioners' sole challenge in this lawsuit is to the take limit set for five of the six species: the Clubshell, the Rusty Patched Bumble Bee, the Madison Cave Isopod, the Indiana Bat, and the Northern Long-Eared Bat. B. We have jurisdiction under the Natural Gas Act. 15 U.S.C. § 717r(d)(1). Because the Endangered Species Act does not specify a standard of review, "we apply the general standard of review of agency action established by" the APA. Or. Nat. Res. Council v. Allen , 476 F.3d 1031, 1036 (9th Cir. 2007). Under the APA, we "shall hold unlawful and set aside agency action, findings, and conclusions found to be arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). Because "resolution of this dispute involves primarily issues of fact" and "requires a high level of technical expertise," we must defer to "the informed discretion of the responsible federal agencies." Marsh v. Or. Nat. Res. Council , 490 U.S. 360, 377, 109 S.Ct. 1851, 104 L.Ed.2d 377 (1989) (quoting Kleppe v. Sierra Club , 427 U.S. 390, 412, 96 S.Ct. 2718, 49 L.Ed.2d 576 (1976) ). Nevertheless, we must conduct a "searching and careful" review to determine whether the agency's decision "was based on a consideration of the relevant factors and whether there has been a clear error of judgment." Id. at 378, 109 S.Ct. 1851 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe , 401 U.S. 402, 416, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971) ). C. An ITS is a safe harbor: it allows an agency (here, FERC) to approve a project that takes threatened or endangered species without falling afoul of the Endangered Species Act. 50 C.F.R. § 402.14(i)(5) ; see Bennett , 520 U.S. at 170, 117 S.Ct. 1154 ("Thus, the Biological Opinion's Incidental Take Statement constitutes a permit authorizing the action agency to 'take' the endangered or threatened species so long as it respects [FWS]'s 'terms and conditions.' "). "The action agency is technically free to disregard the Biological Opinion and proceed with its proposed action, but it does so at its own peril (and that of its employees), for 'any person' who knowingly 'takes' an endangered or threatened species is subject to substantial civil and criminal penalties, including imprisonment." Bennett , 520 U.S. at 170, 117 S.Ct. 1154 (citing 16 U.S.C. § 1540(a), (b) ). For an ITS to function as a safe harbor, FWS must set an incidental take limit that can be monitored and enforced. FERC is required to "report the progress of the action and its impact on the species to [FWS] as specified in the incidental take statement." 50 C.F.R. § 402.14(i)(3). If the "amount or extent of incidental taking ... is exceeded," FERC "must reinitiate consultation immediately ." 50 C.F.R. § 402.14(i)(4) (emphasis added). Section 7 of the Endangered Species Act requires an ITS to "[s]pecif[y] the impact, i.e., the amount or extent, of such incidental taking on the species." 50 C.F.R. § 402.14(i)(1)(i) ; accord 16 U.S.C. § 1536(b)(4). This impact-known as the incidental take limit-must set a "trigger" that can be monitored and enforced, else it is arbitrary and capricious. Miccosukee Tribe of Indians of Fla. v. United States , 566 F.3d 1257, 1275 (11th Cir. 2009) ; accord Ariz. Cattle Growers' Ass'n , 273 F.3d at 1249 ; Wild Fish Conservancy v. Salazar , 628 F.3d 513, 531-32 (9th Cir. 2010) (stating that FWS must set "triggers that can be monitored"). Both FWS and our sister circuits have recognized that Congress intended for this trigger to be a specific number whenever possible. Interagency Cooperation-Endangered Species Act of 1973, as Amended; Incidental Take Statements , 80 Fed. Reg. 26,832, 26,834 (May 11, 2015) (codified at 50 C.F.R. § 402.14(i)(1)(i) ) (hereinafter " Incidental Take Statements ") ("[FWS] acknowledge[s] congressional preference for expressing the impacts of take in incidental take statements in terms of a numerical limitation with respect to individuals of the listed species."); Miccosukee Tribe of Indians of Fla. , 566 F.3d at 1274-75 (citing H.R. Rep. No. 97-567, at 27 (1982), as reprinted in 1982 U.S.C.C.A.N. 2807, 2827) ("Where possible, the impact should be specified in terms of a numerical limitation on the federal agency[.]"); accord Ariz. Cattle Growers' Ass'n , 273 F.3d at 1249-50. An ITS "that 'contains no numerical cap on take and fails to explain why it does not' normally violates" the Endangered Species Act. Ctr. for Biological Diversity v. U.S. Bureau of Land Mgmt. , 698 F.3d 1101, 1126-27 (9th Cir. 2012) (quoting Allen , 476 F.3d at 1037 ). Here, FWS declined to set numeric limits on five of the six non-plant species that will be adversely affected by the pipeline. Instead, it set take limits as a "small percent," "majority" or "all" of the species within set geographic areas. As Petitioners point out, a "small percent" or a "majority" of a species is not an enforceable limit. FWS and ACP do not disagree. Instead, they claim that FWS used habitat surrogates. A habitat surrogate is a way of defining take by the amount of adversely affected habitat rather than by the number of individuals harassed or killed. FWS has explained that habitat surrogates are "often more practical and meaningful to monitor project effects" because they can "provide a clear standard for determining when the amount or extent of anticipated take has been exceeded and consultation should be reinitiated." Incidental Take Statements , 80 Fed. Reg. at 26,839. FWS regulations list three elements necessary for a proper habitat surrogate. First, FWS must include a description of "the causal link between the surrogate and take of the listed species." 50 C.F.R. § 402.14(i)(1)(i). A "causal link" is an "articulated, rational connection" between the activity and the taking of species. See Ariz. Cattle Growers' Ass'n , 273 F.3d at 1250-51. FWS establishes a causal link by examining the habitat requirements and behavior of the listed species and determining the effect of the expected habitat modification. Incidental Take Statements , 80 Fed. Reg. at 26,834, 26,842. Examples of a causal link include "the number of burrows affected or a quantitative loss of cover, food, water quality, or symbionts." Id. at 26,834 (quoting U.S. Fish & Wildlife Serv., Procedures for Conducting Consultation and Conference Activities under § 7 of ESA 4-47-48 (Mar. 1998) (hereinafter " Endangered Species Consultation Handbook ") ). Second, FWS must explain "why it is not practical to express the amount or extent of anticipated take or to monitor take-related impacts in terms of individuals of the listed species." 50 C.F.R. § 402.14(i)(1)(i). There is no clear definition of what makes a numerical limit "not practical" (or not practicable), but FWS has indicated that the standard does not require impossibility. Instead, FWS has long encouraged the use of surrogates when the incidental take is "difficult to detect," which occurs "when the species is wide-ranging; has small body size; finding a dead or impaired specimen is unlikely; losses may be masked by seasonal fluctuations in numbers or other causes (e.g., oxygen depletions for aquatic species); or the species occurs in habitat (e.g., caves) that makes detection difficult." Endangered Species Consultation Handbook 4-52. FWS has also suggested that the cost of monitoring a species relative to the scope of the project can make a numeric limit impractical. Incidental Take Statements , 80 Fed. Reg. at 26,842. Nevertheless, our sister circuits have acknowledged Congress's preference for numerical limits where "possible" and found important FWS's ability to express take in numeric limits even when a species is difficult to detect. E.g. , Miccosukee Tribe of Indians of Fla. , 566 F.3d at 1274-75 (collecting cases); Ariz. Cattle Growers' Ass'n , 273 F.3d at 1249 (collecting cases). Finally, FWS must set "a clear standard for determining when the level of anticipated take has been exceeded." 50 C.F.R. § 402.14(i)(1)(i). A "clear standard" must be able to "adequately trigger reinitiation of consultation," Allen , 476 F.3d at 1038 ; it cannot be "vague and undetectable criteria," Ariz. Cattle Growers' Ass'n , 273 F.3d at 1250-51 (internal quotation marks omitted). Nor can the standard be left to "the unfettered discretion of the Fish and Wildlife Service, leaving no method by which the applicant or the action agency can gauge their performance." Id. at 1250. FWS has endorsed habitat surrogates that are coextensive with a project's scope-i.e., the project can take all of the species within the bounds of the affected habitat-provided that the authorized agency (here, FERC) "monitor[s] project impacts to the surrogate during the course of the action ... [to] determine whether these impacts are consistent with the analysis in the biological opinion." Incidental Take Statements , 80 Fed. Reg. at 26,834, 26,841 -42. Turning to petition No. 18-1083, the only issue presented is whether FWS properly employed a habitat surrogate for five of the endangered and threatened species that will be adversely affected by the pipeline. Specifically, the Petitioners argue that: FWS did not establish that a numeric take limit was impractical, particularly because FWS had previously adopted numeric limits for some of the same species; FWS did not establish a causal link between the pipeline and the habitat selected for some of the species; and the surrogate limits adopted are unenforceable because they set the vague take limits of a "small percent" or a "majority" of individuals. See 50 C.F.R. § 402.14(i)(1)(i). As explained in more detail below, we agree. Because our analysis differs for each of the five challenged species, we will discuss each one separately. But first, we dispose of three overarching points. First, FWS claims that some numeric limits were not possible because either it lacked current survey information about many of the species or ACP had not completed the necessary surveys. This argument is "circular and unavailing." Ctr. for Biological Diversity v. Bureau of Land Mgmt. , 422 F.Supp.2d 1115, 1138 (N.D. Cal. 2006). FWS cannot escape its statutory and regulatory obligations by not obtaining accurate scientific information. Id. ; Allen , 476 F.3d at 1037-38. Moreover, with one exception noted below, FWS "never states that it is not possible" to obtain or update the survey data and arrive at a numeric take limit. Allen , 476 F.3d at 1038. Instead, the agency only states that it and ACP have not actually done the surveys. Id. Thus, FWS's argument does not "establish the numerical measure's impracticality." Id. Second, the agency argues that there was insufficient time for the agency or ACP to develop reliable survey information regarding certain of the species because the agency "must complete formal consultation on a short, ninety-day time timetable based on the 'best scientific and commercial data' then available." Gov't Br. 21 (citing 16 U.S.C. § 1536(a)(2) ) (emphasis added). The agency incorrectly characterizes the law for at least two reasons. First, the statute does not require that the consultation be completed within 90 days. Rather, it states that the agency and FWS may extend the 90-day consultation period to and until 150 days without the consent of the applicant and even longer with the consent of the applicant. § 1536(b)(1)(A)-(B). Accordingly, Congress expressly contemplated that FWS might need more than 90 days to conduct a review and provided a mechanism for it to do so. Second, contrary to the agency characterization, FWS is not required to rely on " 'best scientific and commercial data' then available." Gov't Br. 21 (emphasis added). Rather, the statute requires that "each agency shall use the best scientific and commercial data available," § 1536(a)(2), regardless of whether the data is available at the time of the application. Accordingly, the statute does not foreclose FWS or an applicant from developing additional data. See Allen , 476 F.3d at 1038 (rejecting FWS determination that establishing numerical take limit was impractical due to lack of up-to-date surveys because agency never stated "that it is not possible to update the survey in order to estimate the number of takings, only that it has not actually done the surveys"). Tellingly, neither the statute, nor the agency's implementing regulation, nor the agency's Endangered Species Consultation Handbook , identify lack of time as a proper basis for concluding that setting a numerical limit is impractical. See H.R. Rep. No. 97-567, at 27 (1982) ("For example, it may not be possible to determine the number of eggs of an endangered or threatened fish which will be sucked into a power plant when water is used as a cooling mechanism."); Incidental Take Statements , 80 Fed. Reg. at 26,834 (stating that, "in many cases, the biology of the listed species or the nature of the proposed action makes it impractical to detect or monitor take of individuals of the listed species"); id. (providing as an example that calculating numerical take for pool fairy shrimp would be impractical because a single vernal pool "may contain thousands of individual shrimp as well as their eggs or cysts"); Endangered Species Consultation Handbook , 4-52 (providing form ITS, which states "[The Wildlife Service] anticipates incidental take of (species) will be difficult to detect for the following reason(s): [Incidental take of actual species numbers may be difficult to detect when the species is wide-ranging; has small body size; finding a dead or impaired specimen is unlikely; losses may be masked by seasonal fluctuations in number or other causes (e.g., oxygen depletions for aquatic species); or the species occurs in a habitat (e.g. caves) that makes detection difficult.]"). Accordingly, we reject the agency's contention that lack of time is a proper basis for concluding establishing a numerical limit is impractical. Third, although FWS and ACP both argue that the ITS passes muster, they cannot agree about what the ITS actually did. ACP claims that FWS set pure habitat surrogates; therefore, the "small percent" and "majority" language "was not meant to serve as a quantifiable limit on take" but instead "reflected FWS' position on the limited effect the authorized take was expected to have on the species." ACP Resp. Br. 43 (emphasis added); see id at 33-43, 47. FWS, by contrast, seems to interpret the ITS as setting both numeric and habitat take limits. Gov't Resp. Br. 29 ("Petitioners are correct that the ITS also estimates take for some of those species in terms of percentages of individuals[.]"); id. at 25-26 (defining take for Clubshell as one population); id. at 28 (stating that FWS allowed take of a "small percent" of Indiana Bats within a narrowly drawn area). FWS and ACP cannot rewrite the ITS on appeal. Motor Vehicle Mfrs. Ass'n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co. , 463 U.S. 29, 50, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) ("[T]he courts may not accept appellate counsel's post hoc rationalizations for agency action."). Within a section titled "Amount or Extent of Take Anticipated," FWS prepared a table for each species and stated that the reader should look to that table to locate the anticipated take limit. J.A. 440-43 (stating, for example, that the "anticipated take is described in Table 6 below" for the Clubshell). We review the plain language of the take limits as set forth by FWS in each table. See State Farm , 463 U.S. at 50, 103 S.Ct. 2856. We also note that the inability of FWS and ACP to agree on exactly what level of incidental take is allowed only reinforces our conclusion that the ITS set vague and unenforceable limits. 1. Roanoke Logperch (Percina rex ) Before turning to the five challenged species, we take a moment to discuss briefly the sixth: the Roanoke Logperch, an endangered freshwater fish. Petitioners do not challenge the logperch take limit, but FWS's approach to the logperch is a useful comparator when evaluating its much different approach to the other five species. The pipeline will cross four waterbodies known or with potential to support the logperch. J.A. 407. Neither ACP nor FWS conducted presence/absence surveys; instead, they assumed the presence of the logperch because the pipeline will affect potential habitat in areas known to support the logperch. J.A. 405. ACP threatens the logperch both indirectly due to increased sedimentation and directly where it crosses waterbodies with logperch present. J.A. 407, 418. FWS estimated that a total of 150 logperch are expected to occur in the area affected by the pipeline. J.A. 406. It reached this number by using recent survey information for two of the four affected waterbodies. In the first waterbody, a 2017 survey observed 12 logperch. FWS multiplied that number by ten because "mark-recapture data indicates that only about 10% of [logperch] are actually detected during surveys." This resulted in an estimated 120 total logperch. In the second waterbody, a 2012 survey observed one logperch. Adjusting by the same multiple of ten, FWS estimated there were ten total logperch there. The third and fourth waterbodies lacked recent survey information, but FWS believed the fish to be present at a density comparable to the second waterbody (i.e., 10). And 120 plus ten plus ten plus ten equals 150. FWS set a take limit of 5 individual Roanoke Logperch through injury or death and 145 individuals through harm or harassment. J.A. 440. To arrive at this take limit, FWS divided the area of logperch habitat needed to actually build the crossing structures by the total area of logperch habitat at each pipeline crossing and determined that the crossings themselves will comprise 3.3 percent of the total affected habitat. Because FWS estimates that there are 150 logperch total in the area, this means 3.3 percent (or 5 logperch) will be directly killed by the crossings, while the other 145 will be harmed or harassed by increased sedimentation. Id. We note that FWS set a numeric take limit for the Roanoke Logperch despite using surveys from several years ago that observed only a few fish and despite the detection difficulties caused by the logperch's small size (five inches) and river habitat. 2. Clubshell (Pleurobema clava ) The Clubshell is an endangered mussel that grows to be about three inches long. J.A. 692. In the Monongahela River system of West Virginia, which contains only one Clubshell population in Hackers Creek, the Clubshell is in severe decline and is not reproducing. J.A. 401, 432. A 2009 survey at a long-term monitoring site in Hackers Creek found 29 individuals; a 2014 survey found 19 individuals. J.A. 407. ACP will conduct construction in the upstream drainage area of Hackers Creek and will cross six of its tributaries. J.A. 407. The resulting sediment load will adversely impact the Clubshell population at Hackers Creek. J.A. 432. FWS set the take limit as: killing a "[s]mall percent of individuals present within 585 m," and harming or harassing the "[m]ajority of individuals present within 585 m." J.A. 440-41. The agency explained that it "anticipates incidental take of clubshell will be difficult to detect for the following reason: up to 70% of a population can be distributed below the substrate surface." J.A. 440. Therefore, FWS concluded that the "level of take of this species can be anticipated by loss of habitat from 130 m downstream to 455 m upstream of Life's Run Bridge (County Route 14) (total of 585 m) because this area contains suitable clubshell habitat." J.A. 440. As part of the terms and conditions, FERC must collect all Clubshell in the 585-meter stretch of Hackers Creek. J.A. 444-45. The Clubshell will then be held at an approved facility for reintroduction after the pipeline is completed. Id. The incidental take authorized here is not a proper habitat surrogate because it lacks the three necessary elements. First, there is no clear and enforceable standard of take. Instead, take is limited to a "small percent" and a "majority" of Clubshell within a fixed geographic area. Although the geographic bounds are fixed, FWS authorized the pipeline to take only a subset of individuals located within those bounds . But it is impossible to know the size of the subset-i.e., how many individuals constitute a "small percent" or a "majority." Second, FWS offered no explanation for why the habitat surrogate is limited to 130 meters downstream and 455 meters upstream of Life's Run Bridge. Although the Biological Opinion mentioned that the Hackers Creek Clubshell population is located at Life's Run Bridge, the geographic bounds of 130 meters downstream and 455 meters upstream are not explained anywhere else in the ITS or Biological Opinion. The arbitrariness of this range is particularly conspicuous considering that the pipeline will introduce sediment upstream of Life's Run Bridge, meaning that the sediment will flow through all of Hackers Creek, not just 585 meters of it. Finally, FWS did not adequately explain why a numeric limit is not practical. The only proffered explanation is that Clubshell are difficult to detect because most of them live below the substrate surface. But FWS has been able to adequately survey Clubshell in the past; indeed, it knew there were 29 Clubshell in 2009 and 19 in 2014. See Miccosukee Tribe of Indians of Fla. , 566 F.3d at 1275 (finding unpersuasive FWS's argument that a species was difficult to detect when the record showed the species being counted regularly). Moreover, the ITS terms and conditions require FERC to remove and store the Hackers Creek Clubshell during the pendency of construction, evincing a clear expectation that FERC can locate them. And in a prior ITS, FWS was able to set numeric take limits for the Clubshell. Biological Opinion on the Washington Crossing Bridge Project 22 (Sept. 2, 2014) (setting a total incidental take of 18 Clubshell). In sum, the take limit for the Clubshell fails all three habitat surrogate requirements: it lacks a causal link between the specific 585-meter stretch of Hackers Creek and the local Clubshell population, it lacks an explanation as to why a numerical limit is impractical, and it lacks a clear enforcement standard. See 50 C.F.R. § 402.14(i)(1)(i). Because the ITS's Clubshell take limit is not a meaningful trigger, it violates the Endangered Species Act. 3. Rusty Patched Bumble Bee (Bombus affinis ) The Rusty Patched Bumble Bee (RPBB) was listed as endangered in January 2017. J.A. 401, 809-11. A 2017 survey found one bee in Bath County, Virginia, about a mile from the pipeline route. J.A. 408. FWS concluded that the magnitude of the Rusty Patched Bumble Bee's population losses "has greatly reduced the likelihood that the species is present elsewhere" along the pipeline route. J.A. 408. As a result, "comprehensive RPBB surveys were not conducted throughout the action area" in Virginia. J.A. 408. Based on this single bee, FWS estimated that the Rusty Patched Bumble Bee is most likely to occur in a 653-hectare "high potential zone" near the pipeline. J.A. 408-10. The pipeline will directly affect approximately 7.3 hectares of this zone, which could "crush individuals, crush a colony, expose RPBBs to noise/vibration, and render habitat temporarily and permanently unsuitable." J.A. 421. After making a number of assumptions about bee and bee colony density, FWS estimated that one colony-i.e., a cohesive biological unit of bees established every spring by a solitary queen and made up of her offspring, J.A. 807-is statistically likely to be present in the 7.3 hectares directly affected by the pipeline. J.A. 433. FWS set the take limit as: killing "1 colony present within 7.3 [hectares]" and harming or harassing a "[s]mall percent of [queen bees] from 1 colony present within 7.3 [hectares]." J.A. 441. FWS explained that it "anticipates incidental take of RPBB will be difficult to detect for the following reasons: species has small body size, losses may be masked by seasonal fluctuations in numbers and other environmental factors, and species occurs in habitat (i.e., underground) that makes detection difficult." J.A. 441. The incidental take authorized here is not a proper habitat surrogate because it lacks two of the three necessary elements. FWS has demonstrated a causal link between the bee and the geographic boundaries of the take limit: the pipeline will affect 7.3 hectares of high potential habitat for the Rusty Patched Bumble Bee, which statistically will affect only one Rusty Patched Bumble Bee colony. But FWS's incidental take limit is not a clear standard because it is not actually defined by the 7.3 hectares; instead, take is limited to exactly one colony and a small percent of queen bees within the 7.3 hectares . Neither one colony nor a small percent is an enforceable standard: There may be multiple colonies within the 7.3 hectares, FERC cannot know if taken bees are from the same colony or different colonies, and it is not clear what constitutes a "small percent" of queen bees. Moreover, FWS has not shown that a numeric limit is impractical. It claimed that it has "no accurate way to assess the status of the local population," J.A. 409-but, only a few months earlier, FWS issued its own survey protocols for the Rusty Patched Bumble Bee, J.A. 854-89, and it had conducted a prior survey that identified the lone bee. See Miccosukee Tribe of Indians of Fla. , 566 F.3d at 1275. Indeed, FWS managed to set a numeric limit for the logperch in spite of a survey that located only one fish. In sum, the take limit for the Rusty Patched Bumble Bee fails two of the habitat surrogate requirements: it lacks an explanation as to why a numerical limit is impractical, and it lacks a clear enforcement standard. See 50 C.F.R. § 402.14(i)(1)(i). Because the Rusty Patched Bumble Bee take limit is not a meaningful trigger, it violates the Endangered Species Act. 4. Madison Cave Isopod (Antrolana lira ) The Madison Cave Isopod (MCI) is a threatened subterranean freshwater crustacean about a half-inch in size. "The species is endemic to underground karst aquifer habitats and is restricted to the Shenandoah Valley, from Lexington, Virginia to Harpers Ferry, West Virginia." J.A. 608. The pipeline right of way, additional temporary work space, and access roads will affect "approximately 1,974 surface acres (approximately 0.74%)" of the isopod's potential habitat in Augusta County, Virginia. J.A. 411. Included in this area are Cochran's Cave and five sinkholes, which FWS assumed are connected and which provide a conduit for sediment and contaminants to the isopod's habitat. J.A. 411-12. Because FWS lacked the ability to survey the presence or abundance of the isopods, it assumed that they will be found in the pipeline project area. J.A. 411. The pipeline will threaten the isopods by crushing them or introducing sediment that smothers them or their habitat. J.A. 422. Although there are no "localities" in the pipeline construction area where Madison Cave Isopods have been sampled, FWS arbitrarily decided that Cochran's Cave would serve as an undocumented isopod "locality" and then chose to use "localities as a surrogate for a population." J.A. 434. It concluded that a total of 896.7 surface acres of isopod potential habitat is within 0.5 miles of the construction activities that bisect Cochran's Cave. J.A. 434. Within the 896.7 acres, the pipeline will directly displace 11.2 surface acres. J.A. 434. Within the rest of the 896.7 acres, ground-disturbing activities could smother or crush the isopods. J.A. 422. FWS set the take limit as: killing a "[s]mall percent of individuals present within 11.2 acres" and harming or harassing "[a]ll individuals present within 896.7 acres." J.A. 441-42. The agency explained that "incidental take of the MCI will be difficult to detect for the following reasons: small body size, finding a dead or impaired specimen is unlikely, and species occurs in habitat (underground) that makes detection difficult." J.A. 441. FWS has shown that a numeric limit is not practical here: the isopod is a half-inch crustacean that lives in underground aquifers. But the take limit fails as a habitat surrogate because it lacks the other two elements. First, FWS stated that the pipeline will affect 1,974 surface acres of MCI potential habitat, all of which it assumes contains isopods. But without providing a reasoned explanation, FWS arbitrarily limited the habitat surrogate to the 896.7 acres near Cochran's Cave. Second, the proffered take limits are not real surrogates: the authorized take limit is (in part) a "small percent" of isopods within the 11.2 acres that will be directly affected by the pipeline . But there is no precise way of measuring what a "small percent" of isopods would be, and thus no clear standard for enforcement. In sum, the take limit for the Madison Cave Isopod fails to satisfy two of the habitat surrogate requirements: its causal link between the isopod and the geographic bounds of the take limit is arbitrary, and it lacks a clear enforcement standard. See 50 C.F.R. § 402.14(i)(1)(i). Because the Madison Cave Isopod take limit is not a meaningful trigger, it violates the Endangered Species Act. 5. Indiana Bat (Myolis sodalis ) The Indiana Bat (Ibat) is an endangered migratory bat. FWS estimated that, as of 2017, there are only 425 in Virginia and 1,076 in West Virginia. J.A. 412. The pipeline crosses the Indiana Bat Appalachian Mountain Recovery Unit, an area of protected Indiana Bat habitat that covers all of West Virginia and part of western Virginia. J.A. 412, 725. The pipeline will adversely affect Indiana Bats by temporarily or permanently removing 4,448 acres of suitable habitat in the Recovery Unit. J.A. 423-24. Based on survey data and seasonal movements, FWS defined four relevant categories of Indiana Bat habitat in the Recovery Unit: (1) suitable unoccupied summer habitat, i.e., habitat that is unoccupied during the summer; (2) known use summer habitat, i.e., habitat that is known to be used by the bats in the summer; (3) unknown use spring staging/fall swarming habitat, i.e., habitat near unsurveyed but potentially suitable winter hibernation quarters (known as "hibernacula"); and (4) known use spring staging/fall swarming habitat, i.e., habitat near known hibernacula. J.A. 412. The pipeline will affect these different habitats in the Recovery Unit as follows: Habitat Category Total (acres) Suitable unoccupied summer habitat 3,275.382 Known use summer habitat 144.1 Unknown use spring staging/fall swarming habitat 178.1 Known use spring staging/fall swarming habitat 850.4 Total Acres of Recovery Unit 4,447.982 J.A. 413. FWS set the take limit as: Amount of Take Anticipated Type of Take Small percent of individuals present within Harm, Harass, 1,637.69 acres of suitable unoccupied summer Injure, or Kill habitat Small percent of individuals within 144.1 acres Harass of known use summer habitat Small percent of individuals present within 89.05 Harm, Harass, acres of unknown use spring staging/fall or Kill swarming habitat Small percent of individuals present within 850.4 Harm, Harass, acres known use spring staging/fall swarming or Kill habitat J.A. 442. The agency explained that it "anticipates incidental take of the Ibat will be difficult to detect for the following reasons: species has small body size, finding a dead or impaired specimen is unlikely, and species occurs in habitat (forest and caves) that makes detection difficult." J.A. 442. FWS concluded that the "take of this species can be anticipated by loss of 4,447.982 acres because this area contains suitable Ibat habitat." J.A. 442. But FWS imposed a 0.5 multiplier on the affected acreage for two of the four habitat categories-suitable unoccupied summer habitat and unknown use spring staging/fall swarming habitat-purportedly "[t]o account for differences in Ibat use of the habitat categories." J.A. 442. The incidental take authorized here is not a proper habitat surrogate because it lacks the three necessary elements. First and most significantly, it is not a true habitat surrogate: take is limited to a "small percent" of Indiana Bats within each geographic area. Although the geographic bounds are fixed, the pipeline can only take a subset of individuals located within those bounds . But it is impossible to know what a "small percent" of bats is. Therefore, there is no clear and enforceable standard of take. Second, two of the chosen geographic bounds are arbitrary. FWS knew that the pipeline will directly affect 3,275.382 acres of suitable unoccupied summer habitat and 178.1 acres of unknown use spring staging/fall swarming habitat. Yet, without any explanation, the agency set the take limit for these two habitats at half of these acreages. In other words, FWS set the take limit at half the affected bat habitat that it knows the pipeline is going to affect. Even if FWS removed the words "small percent of individuals" from the take limit, the limit would still fail as a habitat surrogate because FWS knows that the pipeline will exceed the geographic bounds. Oddly, despite the ITS clearly showing the use of the 0.5 multiplier on two of the four habitats, both FWS and ACP claim that the habitat surrogate is actually the full 4,448 acres directly impacted by the pipeline. ACP Resp. Br. 40; Gov't Resp. Br. 27-28 & n.5. ACP actually seems unaware that FWS arbitrarily reduced the habitat surrogate to below the 4,448 affected acres. Compare ACP Resp. Br. 40 (stating that the habitat surrogate is 4,448 acres), with ACP Resp. Br. 46 (stating that the habitat surrogate applies to "individuals present within 1,637.69 acres of suitable unoccupied habitat, 144.1 acres of known use summer habitat, 89.05 acres of unknown use spring staging/fall swarming habitat and 850.4 acres of known use spring staging/fall swarming habitat"-even though these four numbers add up only to 2721.24 acres). Finally, FWS has not shown that a numeric limit is impractical. The bats may be small, but FWS has been able to survey them in the past. Indeed, FWS made precise estimates as recently as 2017, determining that there are 425 bats in Virginia and 1,076 in West Virginia. See Miccosukee Tribe of Indians of Fla. , 566 F.3d at 1275. Moreover, FWS has previously issued incidental take statements with numeric limits for the Indiana Bat, even while recognizing that the bat is difficult to detect. Update to the Biological Opinion on the 2014 Revision of the George Washington National Forest Land and Resources Management Plan 2 (April 21, 2014) (setting an "incidental take of up to 7 Indiana bats on an annual basis as the result of oil and gas leasing, prescribed fire, timber harvest, salvage activities, wildlife habitat management, and special use activities"); Biological Opinion on Enbridge Pipelines (FSP) LLC's Flanagan South Pipeline Project 64-65 (July 24, 2013) (setting incidental take of 19 Indiana bats and 120 reproductive female Indiana bats; noting that take will be measured by "observing mortality or injury" and by "the number of active maternity roost trees removed"); Biological Opinion on the 2003 Revision of the Jefferson National Forest Land and Resource Management Plan 33-34 (Jan. 13, 2004) (noting that the incidental take of the Indiana Bat in the Jefferson National Forest will be difficult to quantify and detect, but nevertheless estimating "there may be up to 10 Indiana bats on the [Forest] incidentally taken on an annual basis through actions that kill, harm, or harass"). In sum, the take limit for the Indiana Bat fails all three habitat surrogate requirements: its causal link between the Indiana bat and the geographic bounds of the take limit is arbitrary, it lacks an explanation as to why a numerical limit is impractical, and it lacks a clear enforcement standard. See 50 C.F.R. § 402.14(i)(1)(i). Because the Indiana Bat take limit is not a meaningful trigger, it violates the Endangered Species Act. 6. Northern Long-Eared Bat (Myolis septentrionalts ) The Northern Long-Eared Bat (NLEB) is a threatened migratory bat. Pipeline construction will remove 171 acres of trees within five miles of a Northern Long-Eared Bat hibernaculum identified as PH-S018. J.A. 425. In 2016, FWS promulgated a generally applicable final rule that governs most incidental take of the Northern Long-Eared Bat. Endangered and Threatened Wildlife and Plants; 4(d) Rule for the Northern Long-Eared Bat , 81 Fed. Reg. 1900 (Jan. 14, 2016) (codified at 50 C.F.R. § 17.40 ). Pursuant to this rule, ACP can conduct incidental take of the bat anywhere more than 0.25 miles away from hibernaculum PH-S018. J.A. 442. In the ITS at issue here, FWS addressed only the incidental take that will occur within a quarter mile of hibernaculum PH-S018. ACP will affect 0.4 acres of that quarter-mile radius, adversely impacting the Northern Long-Eared Bat's ability to forage for food and roost and rendering the habitat permanently unsuitable. J.A. 425, 443. FWS set the take limit as: harming or harassing a "[s]mall percent of individuals present within 0.4 acres." J.A. 443. The agency explained that "incidental take of NLEB will be difficult to detect for the following reasons: species has small body size, finding a dead or impaired specimen is unlikely, and species occurs in habitat (forest and caves) that makes detection difficult." J.A. 443. The incidental take authorized here is not a proper habitat surrogate because it lacks two of the three necessary elements. We find that FWS has demonstrated a causal link between the Northern Long-Eared Bat and the 0.4 acres listed in the take limit. But, as with the other species, the take limit is not a true habitat surrogate: take is limited to a "small percent" of Northern Long-Eared Bats within the 0.4 acres. Although the geographic bounds are fixed, the pipeline can only take a subset of individuals located within those bounds . And it is impossible to know how many bats constitute a "small percent." Therefore, there is no clear and enforceable standard of take. Moreover, FWS has not shown that a numeric take limit is impractical in such a small geographic area. In sum, the take limit for the Northern Long-Eared Bat fails two of the habitat surrogate requirements: it lacks an explanation as to why a numerical limit is impractical, and it lacks a clear enforcement standard. See 50 C.F.R. § 402.14(i)(1)(i). Because the Northern Long-Eared Bat take limit is not a meaningful trigger, it violates the Endangered Species Act. * * * We find that FWS has failed to create proper habitat surrogates, failed to explain why numeric limits are not practical, and failed to create enforceable take limits for the Clubshell, the Rusty Patched Bumble Bee, the Madison Cave Isopod, the Indiana Bat, and the Northern Long-Eared Bat. Because FWS's vague and unenforceable take limits are arbitrary and capricious, we vacated the ITS pending the issuance of this opinion. Sierra Club v. United States Dep't of the Interior , 722 F. App'x 321, 322 (4th Cir. 2018). III. We next address petition No. 18-1082, which challenges the National Park Service (NPS)'s decision to issue a permit allowing the ACP to cross the Blue Ridge Parkway. Petitioners Sierra Club and Virginia Wilderness Committee argue that granting a right-of-way to a gas pipeline exceeds NPS's statutory authority. They also argue that the ACP permit violates a statutory requirement that all agency authorizations be consistent with parkway purposes. NPS challenges Petitioners' standing to bring this petition, while both NPS and ACP contest the merits. We begin by briefly reciting the facts relevant to this petition before turning first to standing and next the merits. A. The Blue Ridge Parkway is a component of the National Park System, linking the Shenandoah National Park in Virginia to the Great Smoky Mountains National Park in North Carolina. 16 U.S.C. § 460a-2. Like the rest of the National Park System, the Parkway is managed by NPS. According to NPS's Management Plan, the Blue Ridge Parkway serves not only as a connector between destinations but also as a recreational and scenic site in itself. J.A. 623-24. The ACP's proposed pathway intersects with the Blue Ridge Parkway. Accordingly, FERC's final approval of the pipeline hinges on NPS granting a right-of-way to cross the Parkway. As proposed, the pipeline would drill and pass underneath the Parkway without breaching the Parkway's surface. However, the proposed route would require removing all of the trees from a portion of a nearby forest, leaving a vertical clearing that would be visible from the Parkway. J.A. 1018. During initial construction of the pipeline, that clearing would be 125 feet wide. Once construction is complete, the clearing would be reduced to a permanent 50-foot wide corridor, reserved for pipeline maintenance purposes. J.A. 325, 1035. A visual impact study conducted by ACP and overseen by NPS concluded that the corridor would be visible from at least one key observation point along the Parkway, thus significantly decreasing the park's scenic value. J.A. 1020. Specifically, the analysis concluded that "[v]iews of the ACP corridor from the Three Ridges overlook ... would likely be inconsistent with NPS management objectives, given the proximity to the viewer, the axial nature of the view, and the corridor's contrast with the surrounding forest." J.A. 1020. On December 12, 2017, NPS issued a revocable permit granting right-of-way to ACP, subject to a list of terms and conditions. The permit cites only 16 U.S.C. § 460a-8 for its statutory authority. J.A. 897. The permit decision does not reference any harm to the Parkway's scenic or conservation value or the effectiveness of any mitigation strategies. Petitioners now seek review of the right-of-way permit, arguing that NPS failed to comply with the Mineral Leasing Act and the Blue Ridge Parkway Organic Act. 30 U.S.C. § 185 ; 16 U.S.C. §§ 460a-3, 460a-3. B. We first address standing. Petitioners Sierra Club and the Virginia Wilderness Committee are organizational plaintiffs that have associational standing to sue "on behalf of [their] members when [their] members would otherwise have standing to sue in their own right." Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc. , 528 U.S. 167, 180-81, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000). For the members to establish standing, they, like all plaintiffs, "must show (1) [they have] suffered an 'injury in fact' that is (a) concrete and particularized and (b) actual or imminent, not conjectural or hypothetical; (2) the injury is fairly traceable to the challenged action of the defendant; and (3) it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision." Id. (citing Lujan v. Defs. of Wildlife , 504 U.S. 555, 560-61, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992) ). For the reasons below, we conclude that Petitioners (via their members) have met the requirements of Article III standing. Members of the petitioning environmental groups aver that they regularly use and enjoy the Blue Ridge Parkway and its scenic views. Petitioners Opening Brief Addendum (Pet. Add.) 115, 126, 130, 171. One member avers that she and her husband have enjoyed using the Three Ridges Overlook for the past thirty-five years. Pet. Add. 117. Other members similarly affirm in their affidavits that they have been to Three Ridges in the past and intend to visit it regularly as part of their hikes and drives in the future because it is a particularly beautiful and cherished section of the Parkway. Pet. Add. 131, 167, 173. One of the members also owns a home near the Parkway, near where the construction is expected to occur. He expresses concerns about not only the pipeline's impact on the scenery he enjoys but also the noise and pollution expected to be caused by the drilling operation, which may affect his home. Pet. Add. 161-62, 164. The affidavits provided by Petitioners' members sufficiently demonstrate injury in fact. "[E]nvironmental plaintiffs adequately allege injury in fact when they aver that they use the affected area and are persons for whom the aesthetic and recreational values of the area will be lessened by the challenged activity." Laidlaw Envtl. Servs. , 528 U.S. at 183, 120 S.Ct. 693 (internal quotation marks and citation omitted). Here, the affidavits establish the members' longstanding history of enjoying not just the Blue Ridge Parkway generally but the Three Ridges Overlook specifically. The pipeline's construction and maintenance corridor would be visible from the Overlook and lessen the aesthetic value of the Parkway. Because the pipeline would prevent the members from enjoying the full beauty of the Parkway and the Overlook, they have established injury in fact. Petitioners have also shown that their members' injuries are fairly traceable to, or caused by, NPS's right-of-way decision. To establish traceability, Petitioners' members must show that the challenged action is "in part responsible for frustrating" their ability to enjoy the Blue Ridge Parkway. See Libertarian Party of Va. v. Judd , 718 F.3d 308, 316 (4th Cir. 2013). Here, NPS authorized the pipeline to cross the Blue Ridge Parkway near the Three Ridges Overlook. That crossing then created the need for the nearby construction and maintenance corridor that would diminish the Parkway's scenic value. In other words, NPS enabled and virtually ensured the alleged harm to the Parkway's aesthetic value. To the contrary, NPS argues that the alleged injuries are not directly caused by the segment of the pipeline that crosses the Parkway. Specifically, NPS emphasizes that the pipeline proceeds underneath the Parkway and does not disturb the Parkway's surface as it crosses. And, although the pipeline's construction and maintenance corridor will scar a nearby forest visible from the Parkway, NPS disclaims all responsibility because that corridor resides on federal lands not managed by NPS and does not require NPS authorization. The problem with this argument is that the causation