Millett, Circuit Judge:

The gray wolf once roamed in large numbers across the contiguous forty-eight States. But by the 1960s, hunting, depredation, and habitat loss drove the gray wolf to the brink of extinction, and the federal government declared the gray wolf an endangered species. After a portion of the gray wolf population rebounded, the government promulgated the rule at issue here, which removes from federal protection a sub-population of gray wolves inhabiting all or portions of nine states in the Western Great Lakes region of the United States. The Humane Society of the United States challenges that rule as a violation of the Endangered Species Act of 1973 ("Act"), 16 U.S.C. § 1531 et seq., and the Administrative Procedure Act ("APA"), 5 U.S.C. § 551 et seq. Because the government failed to reasonably analyze or consider two significant aspects of the rule — the impacts of partial delisting and of historical range loss on the already-listed species — we affirm the judgment of the district court vacating the 2011 Rule.

I

A

Congress enacted the Endangered Species Act "to halt and reverse the trend toward species extinction," and to do so "whatever the cost." Tennessee Valley Auth. v. Hill, 437 U.S. 153, 184, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978). As relevant here, a species is "endangered" if it "is in danger of extinction throughout all or a significant portion of its range[.]" 16 U.S.C. § 1532(6). A species is "threatened" if it "is likely to become an endangered species within the foreseeable future throughout all or a significant portion of its range." Id. § 1532(20).

The Endangered Species Act directs the Secretary of the Interior to apply five factors in determining whether a "species" is endangered or threatened: (i) "the present or threatened destruction, modification, or curtailment of [the species'] habitat or range"; (ii) "overutilization [of the species] for commercial, recreational, scientific, or educational purposes"; (iii) "disease or predation"; (iv) "the inadequacy of existing regulatory mechanisms"; and (v) "other natural or manmade factors affecting [the species'] continued existence." 16 U.S.C. § 1533(a)(1). In making that determination, the Secretary must rely on "the best scientific and commercial data available[.]" Id. § 1533(b)(1)(A). The Secretary of the Interior has delegated the authority to determine whether a species is "endangered" or "threatened" to the Fish and Wildlife Service ("Service"). 50 C.F.R. § 402.01(b).

The "species" that the Endangered Species Act protects are defined to include "any subspecies of fish or wildlife or plants, and," of most relevance here, "any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature." 16 U.S.C. § 1532(16).1 The Act does not define "distinct population segment." Nor do agency regulations. The Service, however, has issued policy guidance stating that the existence of a "distinct population segment" turns upon the discreteness and significance of a sub-population as compared to the larger species population. Policy Regarding the Recognition of Distinct Vertebrate Population Segments Under the Endangered Species Act, 61 Fed. Reg. 4,722, 4,725 (Feb. 7, 1996) ("Segment Policy"). The Segment Policy emphasizes that the Service's authority to recognize distinct population segments should be "exercised sparingly." Id. at 4,724.

To qualify as "discrete" under the Segment Policy, a domestic animal population must be "markedly separated from other populations of the same taxon as a consequence of physical, physiological, ecological, or behavioral factors[.]" Segment Policy, 61 Fed. Reg. at 4,725. The "significance" of a potential segment turns on such factors as: (i) the "[p]ersistence of the discrete population segment in an ecological setting unusual or unique for the taxon"; (ii) "[e]vidence that loss of the discrete population segment would result in a significant gap in the range of a taxon"; (iii) "[e]vidence that the discrete population segment represents the only surviving natural occurrence of a taxon that may be more abundant elsewhere as an introduced population out-side its historic range"; or (iv) "[e]vidence that the discrete population segment differs markedly from other populations of the species in its genetic characteristics." Id.

Another key term in analyzing a species' need for protection — "range" — is also left undefined by the Act. In 2014, the Service adopted a policy statement defining "range" as a species' "`current range,' not [its] `historical range.'" Final Policy on Interpretation of the Phrase "Significant Portion of its Range" in the Endangered Species Act's Definitions of "Endangered Species" and "Threatened Species," 79 Fed. Reg. 37,578, 37,583 (July 1, 2014) ("Range Policy"). The Range Policy further explains that a portion of a species' range will be considered "significant" if the species would be in danger of extinction or likely to become so in the foreseeable future without that portion of its range. Id. at 37,581.

Once the Service determines that a species is endangered or threatened, it must add the species to a list of protected species in the Federal Register. 16 U.S.C. § 1533(c)(1). A listed species receives robust federal protections, including prohibitions on possessing, killing, selling, importing, or exporting its members. Id. § 1538(a). Any person that knowingly violates those prohibitions faces criminal sanctions, including fines of up to $50,000 or a year of imprisonment. Id. § 1540(b)(1).

The Act further requires the Service "from time to time [to] revise" its lists of endangered and threatened species "to reflect recent determinations, designations, and revisions." 16 U.S.C. § 1533(c)(1). Every five years, the Service must "review * * * and determine * * * whether any such species should * * * (i) be removed from such list; (ii) be changed in status from an endangered species to a threatened species; or (iii) be changed in status from a threatened species to an endangered species." Id. § 1533(c)(2)(A), (B).

B

1

Regional subspecies of the taxonomic species "gray wolf" (Canis lupis) were declared endangered by the federal government between 1966 and 1976. The timber wolf (Canis lupus lycaon) was first designated as endangered and afforded protection in 1967, 32 Fed. Reg. 4,001, 4,001 (March 11, 1967), followed by the Northern Rocky Mountain wolf (Canis lupus irremotus) in 1973, 38 Fed. Reg. 14,678, 14,678 (June 4, 1973). Both the Mexican wolf (Canis lupus baileyi) and the Texas wolf (Canis lupus monstrabilis) were added to the list in 1976. 41 Fed. Reg. 17,736, 17,737 (April 28, 1976); 41 Fed. Reg. 24,062, 24,066 (June 14, 1976).

With the wolves' numbers rebounding in certain areas, the federal government in 1978 reclassified the gray wolf from its regional listings into a single species listing divided into two groups: Minnesota gray wolves, which the Service determined had recovered to the point of only being threatened, and the gray wolf in the remaining forty-seven States, which remained endangered. See Reclassification of the Gray Wolf in the United States and Mexico, with Determination of Critical Habitat in Michigan and Minnesota, 43 Fed. Reg. 9,607, 9,608, 9612 (March 9, 1978) ("1978 Rule").2

In 2003, the Service subdivided the gray wolf listing into three "distinct population segments": an Eastern segment, a Western segment, and a Southwestern segment. Final Rule to Reclassify and Remove the Gray Wolf From the List of Endangered and Threatened Wildlife in Portions of the Conterminous United States; Establishment of Two Special Regulations for Threatened Gray Wolves, 68 Fed. Reg. 15,804, 15,818 (April 1, 2003) ("2003 Rule"). Included in the Eastern segment were the Minnesota gray wolf population and any gray wolf population that existed in the Northeast region of the United States. Id. at 15,859. The Service then designated the wolves in the Eastern and Western segments as threatened rather than endangered. Id. at 15,857-15,858, 15,862. The wolves within the Southwestern segment continued to be listed as endangered. Id.

Two district courts struck down the 2003 Rule's attempted designation of those three distinct population segments. First, a district court in Oregon ruled that, by downlisting the species based solely on the viability of a small population within that segment, the Service was effectively ignoring the species' status in its full range, as the Endangered Species Act requires. See Defenders of Wildlife v. Secretary, U.S. Dep't of the Interior, 354 F.Supp.2d 1156, 1168-1169 (D. Or. 2005). The 2003 Rule thus had the "effect of rendering the phrase [significant portion of its range] superfluous." Id. at 1168 (alteration in original; internal quotation marks and citation omitted).

Second, a district court in Vermont concluded that the Service impermissibly designated and downlisted the Eastern segment of gray wolves. National Wildlife Fed'n v. Norton, 386 F.Supp.2d 553, 564-565 (D. Vt. 2005). Fatal to the Service's determination, the court concluded, was the Service's decision to "lump" into the Eastern segment any gray wolves in the Northeast region of the United States, without ascertaining whether a gray wolf population even existed in the Northeast. See id. In so holding, the court rejected the Service's argument that the Endangered Species Act required it to include any Northeast region wolves in the segment to avoid turning them into an impermissible "non-DPS remnant" of gray wolves that neither fell within a recognized segment nor had independent species or subspecies status of its own. Id. at 564-565. In the district court's view, the Service instead could have continued the remnant's endangered-species status. See id. at 565.

2

The government did not appeal either the Oregon or the Vermont decision. Instead, in what turned out to be the first round in successive attempts to delist the gray wolves in the Western Great Lakes area, the Service promulgated a new rule in 2007 that created a "Western Great Lakes gray wolf distinct population segment" and simultaneously delisted that segment, removing it completely from the Endangered Species Act's protections. See Final Rule Designating the Western Great Lakes Populations of Gray Wolves as a Distinct Population Segment; Removing the Western Great Lakes Distinct Population Segment of the Gray Wolf From the List of Endangered and Threatened Wildlife, 72 Fed. Reg. 6,052, 6,052 (Feb. 8, 2007) ("2007 Rule"). That rule soon met the same fate as its two predecessors. A district court in this circuit vacated it for "fail[ing] to acknowledge and address crucial statutory ambiguities" concerning the creation of distinct population segments for the purpose of delisting. Humane Society of the U.S. v. Kempthorne, 579 F.Supp.2d 7, 9 (D.D.C. 2008); id. at 15. Again, the government did not appeal.

In December 2008, the Solicitor of the Department of the Interior issued a memorandum analyzing the statutory authority for designating distinct population segments for the specific purpose of delisting them. See U.S. Fish and Wildlife Service Authority under Section 4(c)(1) of the Endangered Species Act to Revise Lists of Endangered Species and Threatened Species to "Reflect Recent Determinations," Office of the Solicitor, U.S. Dep't of Interior (Dec. 12, 2008) ("Solicitor's Opinion"). The Solicitor concluded that the Act unambiguously allows the Service to identify a segment and then delist it. Id. at 3-5.

The Solicitor started by noting that, once the Service lists a species as threatened or endangered, it is obligated to periodically revise its list of endangered or threatened species in light of any changes in the conservation status of a species. Solicitor's Opinion 3 (citing 16 U.S.C. § 1533(a), (c)(1)). The Solicitor then reasoned that the Endangered Species Act imposes no textual limit on the Service's authority to revise its list of endangered or threatened species based on intervening information and determinations. Id. at 4. On that basis, the Solicitor determined that the Act unambiguously permits the Service to designate a segment within a listed species, determine that the segment is no longer endangered or threatened, and delist it. Id. at 3-5.

The Solicitor further opined that, even were the statutory text ambiguous, his interpretation was a reasonable construction of the statute and its purposes. Solicitor's Opinion 5-6. The Solicitor reasoned that, because subspecies and segments are parts of taxonomic species, any listing of a taxonomic species necessarily includes a listing of its constituent segments or subspecies. Id. at 7. On that basis, the Solicitor concluded that, even if the Service could only delist an already-listed segment, that requirement would be satisfied by the listing of the species that encompassed the segment. Id. The Solicitor also reasoned that delisting a recovered segment is consistent with the express statutory policies of the Act, including fostering federal-state cooperation and focusing resources where they are most needed. Id. at 13-19.

3

Based on the Solicitor's Opinion, the Service in 2009 republished the 2007 rule without notice and comment, adding a discussion of "Issues on Remand." Final Rule To Identify the Western Great Lakes Populations of Gray Wolves as a Distinct Population Segment and to Revise the List of Endangered and Threatened Wildlife, 74 Fed. Reg. 15,070, 15,075 (April 2, 2009) ("2009 Rule"). The Service relied on the Solicitor's determination that the Endangered Species Act permits the Service to "remove an already-listed entity from the appropriate list in its entirety, or to reduce the geographic or taxonomic scope of a listing to exclude a group of organisms previously included as part of an already-listed entity." Id. at 15,083 (quoting Solicitor's Opinion 5 n.8).

Round Two of the Service's attempt to delist the gray wolves in the Western Great Lakes area ended in the same manner as Round One: the 2009 Rule was challenged and vacated after the Service acknowledged that it had impermissibly promulgated the rule without notice and comment, and agreed to settle the case. Humane Society of the U.S. v. Salazar, No. 09-1092, Docket Entry No. 27 (D.D.C. July 2, 2009).

As a consequence of all those regulatory missteps, the status of gray wolves remained in 2009 what it had been in 1978: Gray wolves in Minnesota were listed as "threatened," while the wolves in the forty-seven other contiguous States were listed as "endangered."

C

1

This case is Round Three in the Service's effort to divide and delist gray wolves in the broader Western Great Lakes region. In 2011, the Service issued a final rule that, in reliance on the Solicitor's Opinion, purported to "revise the boundaries of the Minnesota" gray wolf population to include the wolves in all or portions of eight other states. Revising the Listing of the Gray Wolf (Canis lupus) in the Western Great Lakes, 76 Fed. Reg. 81,666, 81,666, 81,670 (Dec. 28, 2011) ("2011 Rule"). Specifically, the 2011 Rule designated the gray wolf population in Minnesota, Wisconsin, and Michigan, as well as portions of North Dakota, South Dakota, Iowa, Illinois, Indiana, and Ohio, as the Western Great Lakes Distinct Population Segment. Id. at 81,666, 81,670. In its next breath, the Service delisted that segment. Id. at 81,723.

In doing so, the Service again expressly adopted the legal analysis in the Solicitor's Opinion regarding its authority to delist a segment. See 2011 Rule, 76 Fed. Reg. at 81,670, 81,683. The Service then reasoned that, because more than 400 miles existed between the gray wolf population in the Western Great Lakes region and other gray wolf packs, the population qualified as "discrete." Id. at 81,671. The Service further found that the loss of the Western Great Lakes population of gray wolves, which contained "70 percent of North American gray wolves known to occur south of Canada," would constitute a "significant gap in the range" of the "gray wolves in the United States[.]" Id. at 81,672. The Service thus concluded that the population qualified as "significant." Id. Accordingly, the Service determined that the Western Great Lakes population of gray wolves constituted a "distinct population segment." Id.

The Service next considered whether the segment was endangered or threatened throughout all or a significant portion of its range. 2011 Rule, 76 Fed. Reg. at 81,721-81,723. In making that determination, the Service explained that it would interpret "range" to mean "current range." Id. at 81,722. The Service also clarified that it would consider a portion of a species' range to be "significant" if that portion is "important to the conservation of the species because it contributes meaningfully to the representation, resiliency, or redundancy of the species." Id.

Finally, the Service concluded, after analyzing the five statutory endangerment factors, that the Western Great Lakes segment was neither endangered nor threatened throughout all or a significant portion of its range. 2011 Rule, 76 Fed. Reg. at 81,721-81,723. The Service explained that existing rates of mortality from disease and human causes had been insufficient to prevent growth of the population, and that state plans provided adequate monitoring of and protection for the wolf segment. See, e.g., id. at 81,694; id. at 81,700.

2

The Humane Society filed suit alleging that the 2011 Rule violated both the Endangered Species Act and the APA. The district court agreed with the Humane Society and vacated the 2011 Rule, concluding that the Endangered Species Act does not permit the Service to designate a segment only to immediately delist it. See Humane Society of the U.S. v. Jewell, 76 F.Supp.3d 69, 110 (D.D.C. 2014). While the district court agreed that the statutory text was ambiguous, the court concluded that the Service's interpretation was unreasonable given the structure, history, and purpose of the Act. Id. at 110-113. In the district court's view, the distinct population segment designation could only function as a "one-way ratchet," allowing the Service to provide more, but not less, protection for a species. Id. at 112.

The district court also rejected the Service's argument that it was simply revising the prior Minnesota wolf listing when it created the Western Great Lakes segment. The court explained that the Minnesota wolves had never been listed as a segment, and that the newly created segment altered the original geographic boundaries of the Minnesota wolf population. Humane Society, 76 F.Supp.3d at 114-115.

The district court further concluded that the rule was arbitrary and capricious because the Service failed to address how large losses in the gray wolf's historical range affected the determination that the Western Great Lakes segment was not endangered or threatened. Humane Society, 76 F.Supp.3d at 128-132.

Finally, the district court held that the rule was invalid because the Service failed to adequately consider the threat to wolves from disease and human-caused mortality and the insufficiency of state regulatory measures to protect the wolves after delisting. Humane Society, 76 F.Supp.3d at 132-134. The district court accordingly vacated the rule. Id. at 136-137.

II

The Service's listing determinations are subject to review under Section 706 of the APA, 5 U.S.C. § 706. See American Wildlands v. Kempthorne, 530 F.3d 991, 997 (D.C. Cir. 2008). Under that standard, we must overturn an agency decision if it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A).

We review the Service's interpretation of the Endangered Species Act under the familiar two-step Chevron framework. See Chevron U.S.A., Inc. v. Natural Resources Def. Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). First, we apply the "traditional tools of statutory construction" to determine whether Congress has directly spoken to the question at issue. Id. at 842-843 & n.9, 104 S.Ct. 2778; Central United Life Ins. v. Burwell, 827 F.3d 70, 73 (D.C. Cir. 2016). If the statute's meaning is clear, the inquiry ends and "we must give effect to the unambiguously expressed intent of Congress." Secretary of Labor, Mine Safety & Health Admin. v. National Cement Co. of Cal., 494 F.3d 1066, 1073 (D.C. Cir. 2007) (internal quotation marks and citation omitted). If, however, "the statute is silent or ambiguous with respect to the specific issue," then we will defer to the agency's considered interpretation of the statute if it is "reasonable." Abington Crest Nursing & Rehab. Ctr. v. Sebelius, 575 F.3d 717, 719 (D.C. Cir. 2009) (internal quotation marks and citation omitted); see also United States v. Mead Corp., 533 U.S. 218, 227-229, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001).

The central dispute in this case is whether the Endangered Species Act permits the Service to carve out of an already-listed species a "distinct population segment" for the purpose of delisting that segment and withdrawing it from the Act's aegis. We hold that the Act permits such a designation, but only when the Service first makes the proper findings.

A

This question of statutory interpretation starts with the Endangered Species Act's plain text. The Act extends its protections to an endangered or threatened "species," and then defines such protected species to include "any distinct population segment of any species of * * * wildlife[.]" 16 U.S.C. § 1532(16). The identification and application of the Act to "distinct population segment[s]" thus falls straightforwardly within the Service's wheelhouse. That much cannot be textually disputed.

The tougher question is whether that distinct population segmentation process is, as the district court ruled, a one-way ratchet that only allows the Service to extend the Act's protections to newly recognized groupings. See Humane Society, 76 F.Supp.3d at 112. Or can the Service, even after a species as a whole has been identified as endangered or threatened, cleave out a subset of that already-listed species for delisting based on the segment's recovery (or uplisting if it has become distinctly imperiled)? Said another way, once a species has been listed, must any changes in its listing status occur species-wide or can the species and its status be dissevered? As to that question, the statutory text is murkier.

The Endangered Species Act quite plainly allows — actually, requires — the Service to periodically revisit and, as warranted, revise the status of a listed species. Section 1533(c)(1) directs the Service to, "from time to time revise each list * * * to reflect recent determinations, designations, and revisions." 16 U.S.C. § 1533(c)(1); see id. § 1533(a)(1), (b)(1)(A). And Section 1533(c)(2) separately directs the Service to "conduct, at least once every five years, a review of all [then-listed] species," and to "determine on the basis of such review whether any such species should — (i) be removed from such list; (ii) be changed in status from an endangered species to a threatened species; or (iii) be changed in status from a threatened species to an endangered species." Id. § 1533(c)(2)(A), (B). Each decision made under those two statutory provisions must rely upon the five statutorily designated criteria for listing and the best available scientific and commercial data. Id. § 1533(c)(1), (2); see id. § 1533(a)(1), (b)(1)(A).

Nothing in that statutory language forbids the recognition of recovered distinct population segments within a listed species. The Secretary's authority to revise a listing under Section 1533(c)(1) is generally unconditioned, as long as the underlying determination on which the revision is based (here, the finding of a distinct population segment) is grounded in the five statutory listing factors and the best available scientific and commercial data. See 16 U.S.C. § 1533(c)(1). To be sure, subsection (c)(1) cross-references the best-evidence requirement of Section 1533(b)(1)(A), which in turn directs the Service to "make determinations * * * after conducting a review of the status of the species," id. § 1533(b)(1)(A) (emphasis added). But that leaves textually unanswered the central question of whether a review of "the species" can itself result in the identification of a distinct population segment.

Section 1533(c)(2)'s quinquennial review provision offers some textual support for the Humane Society's argument that a change in status must be made for the species as listed. Subsection (c)(2) requires reconsideration of whether "such species" as "included in a list" that is "in effect at the time of such review" should be changed from endangered to threatened, changed from threatened to endangered, or removed from the list entirely. 16 U.S.C. § 1533(c)(2) (emphasis added). "[S]uch species" would seem to require review of the species' status as listed.

But while that reading of the statute would be reasonable, it is not ineluctable. To begin with, while subsection (c)(2) prescribes what decisions the Service must make, its text does not foreclose the Service from making additional status decisions. In other words, nothing in the statutory language indicates that the mandated decisions are a ceiling capping the Service's authority rather than just a minimum-requirements floor. In addition, subsection (c)(2) does not detract from the Service's more open-ended revision authority under subsection (c)(1).

The long and the short of all this is that the text of the Endangered Species Act does not itself answer the question whether the Service can designate a distinct population segment from within an already-listed species.

B

Because the statute is "silent or ambiguous with respect to the specific issue" at hand, the question before this court becomes whether the Service's interpretation is "based on a permissible construction of the statute." United States Sugar Corp. v. EPA, 830 F.3d 579, 605 (D.C. Cir. 2016) (quoting Chevron, 467 U.S. at 842-843, 104 S.Ct. 2778). Generally, a "`reasonable' explanation of how an agency's interpretation serves the statute's objectives is the stuff of which a `permissible' construction is made[.]" Northpoint Tech., Ltd. v. FCC, 412 F.3d 145, 151 (D.C. Cir. 2005) (citation omitted). "[A]n explanation that is `arbitrary, capricious, or manifestly contrary to the statute,' however, is not." Id. (quoting Chevron, 467 U.S. at 844, 104 S.Ct. 2778). Accordingly, this court must determine whether the Service "has advanced a reasonable explanation for its conclusion that the regulations serve * * * [the Act's] objectives," Chevron, 467 U.S. at 863, 104 S.Ct. 2778, and whether that "interpretation * * * is at least reasonable in light of any ambiguities in the statute[,]" District of Columbia v. Department of Labor, 819 F.3d 444, 449 (D.C. Cir. 2016). See United States Sugar Corp., 830 F.3d at 608 (regulation must reflect a "permissible reading" of the statute).

The Service's interpretation of Section 1533(c)(1) as allowing for the designation of a distinct population segment within a listed species is a reasonable reading of statutory text and — when properly undertaken, see Part II.C, infra — does not contravene the purposes of the Endangered Species Act.

With respect to the statutory language, the Service's position makes textual sense because subsection (c)(1) itself expressly contemplates that new "designations" and "determinations" could intervene that would require "revis[ing]" an extant listing. 16 U.S.C. § 1533(c)(1). Furthermore, the listing of an animal at the species-wide level can reasonably be understood to include within it a listing of all subspecies and segments within that species. That is to say, the Service's initial listing of all gray wolves in North America necessarily listed all possible segments and subspecies within that grouping. See Solicitor's Opinion 7; 2011 Rule, 76 Fed. Reg. at 81,670, 81,683. So understood, the Service's interpretation comports with Section 1533(c)(2)'s textual requirement that listing changes apply only to species that are already "included in a list * * * [that] is in effect," at the time of the Service's five-year status review, id. § 1533(c)(2).

In addition, the statute requires the Service to attend to both parts of the listing process — the initial listing, and the revision or delisting — with equal care. After all, both subsections (c)(1) and (c)(2) require the Service to rely on the same factors and best evidence that were first used to list a species when downlisting, delisting, uplisting, or otherwise revising that species' status. Nothing in the statutory text compels the Service to put a thumb on the scale in favor of listing, nor does the text require the Service to temporize when the best evidence indicates that a revision is warranted.

Indeed, the statutory text leaves room for the Service, at the initial stage, to list most of a species as threatened, while dividing out a distinct population segment for listing as endangered based on its unique circumstances and conditions. That same language would also permit the Service at the outset to list a segment as threatened even if the remainder of the taxon is endangered. See 16 U.S.C. § 1532(16) (defining "species" in non-exhaustive terms); cf. Modesto Irrigation Dist. v. Gutierrez, 619 F.3d 1024, 1031-1032 (9th Cir. 2010) (Service need not place interbreeding populations of a species in the same distinct population segment); Trout Unlimited v. Lohn, 559 F.3d 946, 959-961 (9th Cir. 2009) (upholding the Service's practice of distinguishing between geographically co-located hatchery and natural populations of the same taxonomic species when listing).

Because the statutory text and purposes can be read to permit such a divided listing on the front end of the listing process, the Service likewise can reasonably read the statute to permit similar determinations at the revision stage. The statutory text does not have to be treated like a one-way street leading only to uplisting.

The Service's position is also consonant with the purposes of the Endangered Species Act, which is to devote needed resources to the protection of endangered and threatened species, while abating the Act's comprehensive protections when a species — defined to include a distinct population segment — is recovered. See, e.g., 16 U.S.C. § 1531 (statutory purpose is to provide for the conservation of species when they are "endangered" or "threatened"); id. § 1533(b)(2) (directing the Service to take into account the "economic impact, the impact on national security, and any other relevant impact" when designating "any particular area as critical habitat"); Defenders of Wildlife v. Norton, 258 F.3d 1136, 1144 (9th Cir. 2001) (noting Congress's desire to allow the Service "more flexibility in [its] approach to wildlife management"). The Service's interpretation ensures that the most resources can be brought to bear where a species continues to be threatened or endangered. See Solicitor's Opinion 18-19; 2011 Rule, 76 Fed. Reg. at 81,670, 81,683. In that regard, the Act's direction to the Service to evaluate the status of not just species, but subspecies and segments, 16 U.S.C. § 1532(16), signals Congress's intent to target the Act's provisions where needed, rather than to require the woodenly undifferentiated treatment of all members of a taxonomic species regardless of how their actual status and condition might change over time.3

Another purpose of the Endangered Species Act is to foster state cooperation in the conservation of threatened or endangered species. See 16 U.S.C. § 1531(a)(5) (One purpose of the Act is "encouraging the States * * * to develop and maintain conservation programs which meet national and international standards[.]"); see also Solicitor's Opinion 16-17 ("Removing [Endangered Species Act] protections for recovered [Distinct Population Segments] of listed species reinforces the strong public policy goal of federal-State cooperation[.]"); 2011 Rule, 76 Fed. Reg. at 81,670, 81,683 (incorporating the views expressed in the Solicitor's Opinion).4 Because the locations of distinct population segments not uncommonly correspond with geographical lines, empowering the Service to alter the listing status of segments rewards those States that most actively encourage and promote species recovery within their jurisdictions. On the other hand, continuing to rigidly enforce the Act's stringent protections in the face of such success just because recovery has lagged elsewhere would discourage robust cooperation. The Service's interpretation thus reasonably "encourage[s] the States * * * through * * * a system of incentives." 16 U.S.C. § 1531(a)(5).

The Humane Society argues that Service action under the Act "must be, first and foremost, to provide protections to endangered or threatened species." Humane Society's Br. 33. True enough. But that premise does nothing to answer the specific question of whether the Service is permitted to tailor its protections to where they are most needed.

Relatedly, the Humane Society argues that designating a segment to delist it violates the Service's formal Segment Policy, 61 Fed. Reg. at 4,724-4,725. Aspects of that Policy certainly underscore its protective purpose. The Service has explained that the identification of distinct population segments should "be aimed at carrying out the purposes of the Act," including "to provide a means whereby the ecosystems upon which endangered species and threatened species depend may be conserved, [and] to provide a program for the conservation of such endangered species and threatened species." Id. at 4,722 (internal quotation mark and citation omitted). Indeed, some of the Policy's criteria for "significance" — which a population must possess to qualify as a segment — would seem to most often be relevant to enhancing protections for animals. For example, it would not make much sense to downlist or delist a population that "represents the only surviving natural occurrence of a taxon that may be more abundant elsewhere as an introduced population outside its historic range." Id. at 4,725.

On the other hand, other aspects of the Segment Policy plainly work in both the listing and delisting directions. See Segment Policy, 61 Fed. Reg. at 4,722 (describing segments as being used "for the purposes of listing, delisting, and reclassifying species") (emphasis added); id. at 4,725 (referring to the enumerated segment factors as "apply[ing] similarly for addition to the lists[,] * * * reclassification, and removal from the lists") (emphasis added); id. (identifying the Segment Policy as guiding "the evaluation of distinct vertebrate population segments for the purposes of listing, delisting, and reclassifying under the Act") (emphasis added).

The Solicitor's Opinion, formally adopted by the Service, has now explicitly interpreted the Act to allow the segment tool for delisting. See Solicitor's Opinion 3-5; 2011 Rule, 76 Fed. Reg. at 81,670, 81,683. See also Encino Motorcars, LLC v. Navarro, ___ U.S. ___, 136 S.Ct. 2117, 2125, 195 L.Ed.2d 382 (2016) ("Agencies are free to change their existing policies so long as they provide a reasoned explanation for the change.").

The Humane Society also argues that designating a segment to delist it violates the specific process set out in the Segment Policy, 61 Fed. Reg. at 4,725. That Policy identifies three factors to be considered in designating and listing a segment: the "discreteness," "significance," and "conservation status" of a population. Id. The Humane Society reads that provision to require findings that a population is discrete, significant, and has a conservation status of threatened or endangered before the Service can designate it.

But the far more natural reading of the Policy is that it only requires a determination that the potential segment is threatened or endangered prior to listing, not prior to the designation of the segment itself. See Segment Policy, 61 Fed. Reg. at 4,725. The Policy specifically sets out identification of the segment's conservation status as an independent step that follows after a segment has been identified as a distinct population segment because of its discreteness and significance. See id. ("Status: If a population segment is discrete and significant (i.e., it is a distinct population segment) its evaluation for endangered or threatened status will be based on the Act's definitions of those terms and a review of the factors enumerated in [Section 1533(a)].") (second emphasis added). The Service's decision in this case thus did not contradict its Segment Policy.5

* * * * *

"When it enacted the [Endangered Species Act], Congress delegated broad administrative and interpretive power to the [Service]." Babbitt v. Sweet Home Chapter of Communities for a Great Oregon, 515 U.S. 687, 708, 115 S.Ct. 2407, 132 L.Ed.2d 597 (1995). And "[t]he task of defining and listing endangered and threatened species requires an expertise and attention to detail that exceeds the normal province of Congress." Id. Given the ambiguity of the statutory text, the Humane Society's proffered interpretation of the Act as favoring the use of segments in a protective manner may very well be reasonable. But our task under Chevron is not to pick from amongst reasonable options. Our task is simply to determine whether the Service's interpretation of the ambiguous language is reasonable. We hold that the Service permissibly concluded that the Endangered Species Act allows the identification of a distinct population segment within an already-listed species, and further allows the assignment of a different conservation status to that segment if the statutory criteria for uplisting, downlisting, or delisting are met.

C

Holding that the Service has the legal authority to identify a distinct population segment from within an already-listed species does not mean it did so properly here. In fact, it did not. The fundamental error in the Service's decision is that, in evaluating whether gray wolves in the Western Great Lakes region are a "distinct" population segment, the Service failed to address the impact that extraction of the segment would have on the legal status of the remaining wolves in the already-listed species. More specifically, the Service cannot find that a population segment is distinct — in the Service's words, that it is severable because it is "discrete" and "significant" — without determining whether the remnant itself remains a species so that its own status under the Act will continue as needed.

1

The Endangered Species Act's text requires the Service, when reviewing and redetermining the status of a species, to look at the whole picture of the listed species, not just a segment of it. Section 1533(c)(2)(A) requires that the review cover the "species included in a list." 16 U.S.C. § 1533(c)(2)(A); see also id. § 1533(c)(1), (b)(1)(A) (directing the Service, when revising the status of a species, to "make [its] determinations * * * after conducting a review of the status of the species" as listed) (emphasis added); see also id. § 1533(c)(2)(B). As the Service itself argues, that review can reasonably be read to include any and all of the composite segments or subspecies that might be included within a taxonomically listed species. See Solicitor's Opinion 7-8 & n.10; 2011 Rule, 76 Fed. Reg. at 81,670, 81,683. Thus, when a species is already listed, the Service cannot review a single segment with blinders on, ignoring the continuing status of the species' remnant. The statute requires a comprehensive review of the entire listed species and its continuing status. Having started the process, the Service cannot call it quits upon finding a single distinct population segment.

The Service's definition of a "distinct population segment" confirms the point. See Segment Policy, 61 Fed. Reg. at 4,725. The Service's two critical criteria for such a segment are "discreteness" and "significance." Id. Both of those factors must be met before a segment can be recognized.

To start, the Service looks at the "[d]iscreteness of the population segment in relation to the remainder of the species to which it belongs." Segment Policy, 61 Fed. Reg. at 4,725. More specifically, to be distinct, a segment must be "markedly separated" out "from other populations of the same taxon as a consequence of physical, physiological, ecological, or behavioral factors." Id. Those two factors clearly envision a comparative analysis of a potential segment to the remnant and consideration of the segment's independent severability.6

With respect to the "significance" part of the test, the Service looks to such factors as whether the segment: (i) persists in an "ecological setting" that is "unusual or unique for the taxon"; (ii) is the "only surviving natural occurrence of a taxon"; (iii) has genetic characteristics that are "markedly" different from the rest of the taxon; or (iv) would cause a "significant gap" in the taxon's range if lost. Segment Policy, 61 Fed. Reg. at 4,725. Each of those factors measures a potential segment's "significance" in relation to the "taxon." That means that an evaluation of "significance" presupposes that there is a still-existing taxon against which to compare the potential segment. Indeed, without an evaluation of the taxon (both pre- and post-designation of the proposed segment), the Service could not in any meaningful way evaluate the proposed segment's significance to that taxon.

Requiring the Service to look at the implications for both the segment and the remnant during the delisting, uplisting, or downlisting process thus flows naturally from the Endangered Species Act's text and the Service's own Segment Policy.

Even more importantly, omitting analysis of the effect of designation on the already-listed species would divest the extant listing of legal force. The segment-designation process is meant to be "sparing[]" in its use. Segment Policy, 61 Fed. Reg. at 4,724. Yet the Service's disregard of the remnant's status would turn that sparing segment process into a backdoor route to the de facto delisting of already-listed species, in open defiance of the Endangered Species Act's specifically enumerated requirements for delisting. See 16 U.S.C. § 1533(a)(1) (listing five mandatory criteria for altering a listing). Accordingly, as a matter of plain statutory design, the act of designating a segment cannot in one fell swoop make an already-listed species an unlisted and unlistable non-species, "sidestep[ping]" the process "Congress has plainly" prescribed for delisting. See Natural Resources Def. Council v. EPA, 489 F.3d 1364, 1372 (D.C. Cir. 2007); see also Whitman v. American Trucking Ass'ns, 531 U.S. 457, 485, 121 S.Ct. 903, 149 L.Ed.2d 1 (2001) ("[An agency] may not construe the statute in a way that completely nullifies textually applicable provisions meant to limit its discretion."). Instead, the Service must make it part and parcel of its segment analysis to ensure that the remnant, if still endangered or threatened, remains protectable under the Endangered Species Act.

2

In designating the Western Great Lakes wolves as a distinct population segment, the Service looked only at the characteristics of the Western Great Lakes segment in a vacuum, ignoring the second step of determining whether both the segment and the remainder of the already-listed wolves would have mutually independent statuses as species.

Prior to the 2011 Rule designating the Western Great Lakes segment, the Service had made two listings of gray wolves: those in Minnesota that were found to be threatened, and those wolves in the lower forty-eight states outside of Minnesota that were determined to be an endangered species. When the Service attempted to carve the Western Great Lakes segment out of the latter, it left the remnant of that already-statutorily-protected group in legal limbo without any determination that the gray wolves in the continental United States outside of the Western Great Lakes segment were themselves a species, subspecies, or segment that could continue to be protected under the Endangered Species Act. Certainly "gray wolves outside the Western Great Lakes segment" have never been recognized as a taxonomic species. The Service also failed to analyze whether "gray wolves outside the Western Great Lakes segment" are either a subspecies or a segment.7 Absent such a determination, the Service has left entirely unexplained how the remaining wolves' existing endangered status would continue. Nor did the Service make any finding that the remnant was no longer endangered under the statutory listing criteria.

Worse still, the Service has announced that, with the Western Great Lakes segment carved out, the remnant is no longer a protectable "species" and has proposed its delisting for that reason alone. See Removing the Gray Wolf (Canis lupus) From the List of Endangered and Threatened Wildlife and Maintaining Protections for the Mexican Wolf (Canis lupus baileyi) by Listing It as Endangered, 78 Fed. Reg. 35,664, 35,668 (June 13, 2013) ("We conclude that the current C. lupus [gray wolf] entity is not a valid species under the Act and now propose to remove this entity from the List[.]"). The Service did not deny that position at oral argument. See Oral Arg. Tr. 85-90.

The Service's power is to designate genuinely discrete population segments; it is not to delist an already-protected species by balkanization. The Service cannot circumvent the Endangered Species Act's explicit delisting standards by riving an existing listing into a recovered sub-group and a leftover group that becomes an orphan to the law. Such a statutory dodge is the essence of arbitrary-and-capricious and ill-reasoned agency action. See Ragsdale v. Wolverine World Wide, Inc., 535 U.S. 81, 91-92, 94, 122 S.Ct. 1155, 152 L.Ed.2d 167 (2002) (invalidating agency action that constituted an "end run around important [statutory] limitations" and thus "contravene[d] Congress' will" and "subvert[ed] the careful balance" of the statute).

The Service argues in the alternative that it did not designate a new segment, but rather only revised the 1978 Minnesota gray wolf segment and then delisted it. We are doubtful that the Service's action can be fairly characterized as a revision, especially given the fact the purported revision roped-in wolf populations outside of the original Minnesota population. See 2011 Rule, 76 Fed. Reg. at 81,678-81,679, 81,716; 1978 Rule, 43 Fed. Reg. at 9,608 ("There is also a group [of wolves] on Isle Royale in Lake Superior, and possibly a few scattered individuals in northern Michigan and Wisconsin."). By bringing within the scope of the segment those outside-Minnesota wolves, the Service's "revision" of the Minnesota gray wolf segment is in fact a revamping of the entire gray wolf listing, including the non-Minnesota population listing. But we need not decide that question. Whether labeled a revision or a segment-designation, the flaw is the same: the failure to address the status of the remnant is fatal.

III

Under the Endangered Species Act, the determination of a species' endangered or threatened status turns on the threats that the species faces "throughout all or a significant portion of its range." 16 U.S.C. § 1532(6), (20). The Service concluded that "range" refers to the species' current range at the time its status is evaluated or reevaluated for listing. See 2011 Rule, 76 Fed. Reg. at 81,721-81,722. The district court held that the Service's decision to delist the Western Great Lakes segment failed to adequately address the wolves' loss of historical range. Humane Society, 76 F.Supp.3d at 128-132. Because the Service's interpretation of "range" as focusing on "current range" is reasonable, we uphold it. But because the Service categorically excluded the effects of loss of historical range from its analysis, we hold that the Service's conclusion about the ongoing threat to the Western Great Lakes segment within its current range was insufficiently reasoned, and therefore arbitrary and capricious.

A

Under Chevron, we ask first whether the Endangered Species Act speaks directly to the meaning of "range" and, if it does not, we must evaluate the reasonableness of the Service's interpretation. See 467 U.S. at 842-843, 104 S.Ct. 2778.

The Endangered Species Act does not itself define "range." See 16 U.S.C. § 1532. The definitions of "endangered" and "threatened" species, however, do use the present tense "is" to refer to the status of the species within its range. Id. § 1532(6), (20). That seems to accord with the Service's position that "range" refers to "current range." See Sherley v. Sebelius, 644 F.3d 388, 394 (D.C. Cir. 2011) ("The use of the present tense in a statute strongly suggests it does not extend to past actions."). But see McNeill v. United States, 563 U.S. 816, 821-822, 131 S.Ct. 2218, 180 L.Ed.2d 35 (2011) (Congress has "used the present tense to refer to past convictions").

Still, focusing on verb tense does not get the Service very far. That is because the placement of "is" in the definitions seems most naturally to require that the species currently be endangered or threatened within its range, not to dictate the temporal scope of geographical evidence the Service is to consider. A species can be found to be endangered now — "is in danger of extinction," 16 U.S.C. § 1532(6) (emphasis added) — based just as easily on threats to the species throughout its historical range as on threats throughout its contemporary range. Cf. Defenders of Wildlife v. Norton, 258 F.3d 1136, 1145 (9th Cir. 2001) ("[A] species can be extinct `throughout * * * a significant portion of its range' if there are major geographical areas in which it is no longer viable but once was.") (second alteration in original).

Outside of its use in the definitions of "endangered" and "threatened" species, "range" appears only three times in the Act. The term first appears in Section 1533(a), which lists "the present or threatened destruction, modification, or curtailment of [a species'] habitat or range," as one of several factors for the Service to consider in determining "endangered" or "threatened" status. 16 U.S.C. § 1533(a)(1)(A). The term appears a few subsections later in Section 1533(c), which mandates that the listing of a species as endangered or threatened shall "specify with respect to each * * * species over what portion of its range it is endangered or threatened, and specify any critical habitat within such range." Id. § 1533(c)(1). Neither of those usages casts material light on the meaning of "range" because the two references are as textually indeterminate as the initial use of the term in Section 1532.

The third place that "range" appears is in Section 1539(j), which provides that the Secretary may authorize the release of any population of an endangered or threatened species "outside the current range of such species if the Secretary determines that such release will further the conservation of such species." 16 U.S.C. § 1539(j)(2)(A). That provision cuts both ways. On the one hand, it could be argued that, if "range" already means current range, then the adjectival addition of "current" in Section 1539(j) would be redundant. On the other hand, the use of "current range" in Section 1539(j) could also be read to corroborate the Service's view, since "current range" in Section 1539(j) may refer to the listed range of the endangered or threatened species. See id. § 1539(j)(2)(A).

Dictionary definitions do not illuminate the meaning of "range" either. As a biological and zoological term of art, "range" is commonly defined as a geographical reference to the physical area in which a species lives or occurs. See, e.g., 8 THE OXFORD ENGLISH DICTIONARY 139 (def. 7) (1933) (defining "range" as: "[t]he geographical area over which a certain plant or animal is distributed"). The most that can be said is that such dictionary definitions employ the present tense. But that may reflect the nature of dictionary definitions generally, rather than suggest any contextual meaning within the Endangered Species Act.8

Accordingly, traditional rules of statutory construction do not answer the question of whether "range" means current or historical range. Indeed, the Service and the Humane Society both acknowledge that the Act leaves open the possibility that "range" may refer to either current or historical range. See Service's Br. 68-69; Humane Society's Br. 47-48.9 The question then becomes whether the Service's interpretation of "range" as "current range" "is based on a permissible construction of the statute." Mayo Found. for Med. Educ. & Research v. United States, 562 U.S. 44, 54, 131 S.Ct. 704, 178 L.Ed.2d 588 (2011) (quoting Chevron, 467 U.S. at 843, 104 S.Ct. 2778). It is.

Although the statute itself does not indicate the meaning of "range," the Service's interpretation is at least consistent with the Endangered Species Act's use of the present tense in provisions discussing the species' range, 16 U.S.C. § 1532(6), (20). And it also accords with Section 1539(j)(2)(A)'s use of "current range" in reference to a species' listed range. In addition, focusing on the species' survival in the range it currently occupies is consonant with the purposes of the Endangered Species Act, because the threats that a species confronts where it currently lives often affect its continued survival the most and thus bear influentially on whether it should be listed. See 2011 Rule, 76 Fed. Reg. at 81,722-81,723 (discussing threats in areas where the Western Great Lakes wolf population currently lives).

For those reasons, we conclude that the Service's interpretation of "range" to focus on a species' current range is a reasonable interpretation of the Act.

B

As with the Service's designation of distinct population segments, the rub in this case is not with the Service's interpretation of the statute, but with its application of the statute to the record at hand. While analysis of the reasonableness of agency action "under Chevron Step Two and arbitrary and capricious review is often the same," Pharmaceutical Research & Mfrs. of America v. Federal Trade Comm'n, 790 F.3d 198, 209 (D.C. Cir. 2015) (internal quotation marks and citation omitted), the Venn diagram of the two inquiries is not a circle. The question thus remains whether the agency arbitrarily and capriciously "`failed to consider an important aspect of the problem' it faces." SecurityPoint Holdings, Inc. v. Transportation Sec. Admin., 769 F.3d 1184, 1187 (D.C. Cir. 2014) (quoting Motor Vehicle Mfrs. Ass'n of the U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)).

We hold that the Service's analysis of the status of the Western Great Lakes segment within its current range wrongly omitted all consideration of lost historical range. Just because the Endangered Species Act does not compel the Service to interpret "range" to mean historical range, that does not mean that the Service can brush off a substantial loss of historical range as irrelevant to the species' endangered or threatened status. So says the Service itself: The Service's Range Policy is explicit that a species may be "endangered or threatened throughout all or a significant portion of its current range because [a] loss of historical range is so substantial that it undermines the viability of the species as it exists today." Range Policy, 79 Fed. Reg. at 37,584 (emphasis added).

That is an eminently sensible approach. Range loss can "result[] in a species for which distribution and abundance is restricted, gene flow is inhibited, or population redundancy is reduced to such a level that the entity is now vulnerable to extinction or likely to become so within the foreseeable future throughout all or a significant portion of its current range." Range Policy, 79 Fed. Reg. at 37,584. In addition, "a species with a reduced range is at greater risk of all or most of its populations being affected by a catastrophic event such as a hurricane or fire." Id.

In other words, an adequate evaluation of the threats confronting the survival of a species within its current range requires looking at more than just the current moment in time. The Service, consistent with its own Range Policy, also needs to consider the scope of the species' historical range, and the impact that material contraction or relocation might indicate for survival within a currently constricted or confined range.

There is, moreover, no question in this case that "gray wolves have been extirpated from most of the southern portions of their historical North American range." 2011 Rule, 76 Fed. Reg. at 81,672. The Humane Society estimates that 95% of the gray wolf's historical range has disappeared. Humane Society's Br. 48, 50. The Service does not dispute that figure.

Despite immense losses in the gray wolves' historical range — including the historical range of those wolves now occupying the Western Great Lakes area — the Service nowhere analyzed the impact of that loss on the survival of the gray wolves as a whole, the gray wolves remnant, or the Western Great Lakes segment. Such a failure to address "an important aspect of the problem" that is factually substantiated in the record is unreasoned, arbitrary, and capricious decisionmaking. State Farm, 463 U.S. at 43, 103 S.Ct. 2856; see also AEP Tex. N. Co. v. Surface Transp. Bd., 609 F.3d 432, 441 (D.C. Cir. 2010) (agency acted arbitrarily and capriciously when it "failed to address the unique circumstances of" an issue).

The Service does not deny the gap in its analysis. Instead, the Service points to its determination that the Western Great Lakes segment would remain viable in key portions of the Western Great Lakes area. See 2011 Rule, 76 Fed. Reg. at 81,722-81,723. That is a non sequitur. As the Range Policy explained, consideration of material changes in a species' historical range is critical to a reliable assessment of sustainability within the current range. So whatever the Service prognosticates about future viability in certain portions of the current range cannot be reliably reasoned if it was made in a historical vacuum. An important factor — the possible enduring consequences of significant loss of historical range — was left out of the analysis all together.

The Service also argues that the Act does not require the restoration of a species to its entire historical range. See Service's Reply Br. 27-29. Okay. But giving adequate consideration to the effects of large losses of historical range on a species' survival going forward has nothing to do with where geographically a species must be restored. The only obligation at issue here is for the Service to contend with the implications of massive range loss for the species' endangered or threatened status within its current environment.

Finally, we note that, in undertaking that omitted analysis on remand, the Service will have to grapple with predicate questions that the Service has evaded thus far, such as:

In sum, because the undisputedly vast loss of historical range is a salient factor in determining the endangered or threatened status of the Western Great Lakes segment and the remnant population within their current ranges, the Service's wholesale failure to address that factor renders the Service's decision unreasoned, arbitrary, and capricious.

IV

The Humane Society also argues that the final rule failed adequately (i) to explain why the wolf population's combined mortality from humans and disease is not a continuing threat to the species' existence, and (ii) to address the lack of adequate plan provisions or other protections for the gray wolves in the seven States that make up the Western Great Lakes area. Neither argument succeeds.

A

With respect to the combined threat to the gray wolves' survival from disease and human-caused death, the record reflects that the Service adequately wrestled with both problems and grounded its decision in substantial evidence.

1

With respect to disease mortality, the 2011 Rule addressed five diseases that afflict the Western Great Lakes gray wolves: canine parovirus, sarcoptic mange, lyme, dog louse, and canine distemper virus. The Rule also looked at the possibility of other diseases entering the wolf population. 2011 Rule, 76 Fed. Reg. at 81,694-81,698. The Service's analysis was grounded in scientific literature specifically analyzing the history and impact of disease on the gray wolf population.

For instance, the Service explained that there was "no evidence" that canine parovirus "caused a population decline or has had a significant impact on the recovery of the Minnesota wolf population." 2011 Rule, 76 Fed. Reg. at 81,694. While some studies suggested that canine parovirus had reduced pup survival from 1984 to 2004, the Service found that there had actually been an increase in pup survival since 1995. Id. Notably, data in Wisconsin showed only a single pup death attributable to canine parovirus since 2001, id., and no such deaths in Michigan, id. at 81,695.

With respect to sarcoptic mange, the Service determined that mange mortality was "stabilizing or perhaps declining in Wisconsin." 2011 Rule, 76 Fed. Reg. at 81,695. The Service pointed to studies indicating that mange infection had not increased in Minnesota since 2003, and had in fact declined from 17% in 2006 to 10% in 2008. Id. at 81,696. As for lyme disease and dog louse, the Service reported that there had been no confirmed deaths in the Minnesota wolf population from dog louse and no. reports of clinical symptoms of lyme disease. Id. Finally, with respect to canine distemper virus, the Service relied on scientific studies that "predict periodic short-term declines * * * but no long-term threat to the wolf population" from the disease. Id.

The Service also acknowledged the possibility that new diseases might arise, but concluded there was no sufficiently concrete risk to threaten the gray wolf's survival. 2011 Rule, 76 Fed. Reg. at 81,696. In that regard, the Service explained that Minnesota's, Michigan's, and Wisconsin's state plans all provided for the continued monitoring of dead wolves and testing of live-captured wolves and wolf feces to detect any new diseases that might require intervention. Id. at 81,697. Looking at the collective threat from disease, the Service concluded that "the overall trend for wolf populations in the [Western Great Lakes segment] continues to be upward." Id. at 81,698. The Service added that delisting the wolves "will not significantly change the incidence or impacts of disease and parasites on these wolves." Id.

The Service's judgment was corroborated through peer review by a veterinary pathologist specializing in wolf disease and mortality. The pathologist concluded that the proposed rule "definitely contain[ed] an accurate, comprehensive synthesis of published and unpublished data on disease and predation threats to the Western Great Lakes * * * wolf populations." J.A. 881. She further concurred that "wolf populations have grown despite introductions of new diseases," and that "all evidence indicate[d] that [diseases] are not likely to endanger the [Western Great Lake] wolf populations if delisted." Id.

2

The Service's analysis of human-caused mortality too was satisfactory. Human-caused mortality for wolves comes in three forms: fatal accidents (often involving vehicles), legal depredation programs, and intentional illegal killings.

The Service's 2011 Rule studied human-caused mortality in detail. The Service reviewed existing mortality data and acknowledged that humans were responsible for 56% of diagnosed deaths of radio-collared wolves in Wisconsin from 1979 to 2009, 69% of such deaths in Minnesota from 1994 to 2005, and 75% of such deaths in the Upper Peninsula of Michigan from 1960 to 1997. 2011 Rule, 76 Fed. Reg. at 81,698-81,699.

The Service nevertheless concluded that human-caused mortality was not a significant threat to the wolf's survival, as shown by the resilient growth of the gray wolf population despite the human-caused deaths. "As long as other mortality factors do not increase significantly," the Service concluded, and "monitoring is adequate to document, and if necessary counteract, the effects of excessive human-caused mortality should that occur, * * * [the] wolf population will not decline to nonviable levels" because of human-caused mortality. 2011 Rule, 76 Fed. Reg. at 81,700 (citation omitted).

The Service further reasoned that delisting the segment of gray wolves would not affect the rate of human-caused mortality. That is because accidental and depredation deaths are most directly tied to the wolves' proximity to areas densely populated by humans. See 2011 Rule, 76 Fed. Reg. at 81,700 ("[A] continuing increase in wolf mortalities from vehicle collisions, both in actual numbers and as a percent of total diagnosed mortalities, is expected as wolves continue their colonization of areas with more human developments and a denser network of roads and vehicle traffic."). Additionally, the Service looked to Minnesota depredation data from 2007 and 2008 (the time when the Western Great Lakes wolf population was temporarily delisted under the now-vacated 2007 Rule), and found that the rates of depredation deaths did not change materially from before the wolves were delisted. Id. at 81,704 (noting that 133 Minnesota wolves were killed in 2007 and 143 wolves in 2008, compared to 134 in 2005 and 122 in 2006). The Service further found that the Western Great Lakes wolf population continued to grow despite the increase in depredation deaths in 2007 and 2008 in Wisconsin and Michigan. See id. at 81,708 (noting that the Wisconsin wolf population increased 12% between 2008 and 2009); id. at 81,712 (noting that the Michigan wolf population grew 11% between 2008 and 2009).

The Service also addressed the problem of illegal killings. The Service candidly acknowledged the limited data available since such killings "generally occur in remote locations and the evidence is easily concealed[.]" Final Rule, 76 Fed. Reg. at 81,698. Nevertheless, the data that the Service was able to collect indicated that the number and proportion of wolves killed illegally in Wisconsin declined while the gray wolf was delisted under the later-vacated 2007 Rule. See id. at 81,696 (showing 17 of 72 wolves found dead in 2006 had been killed illegally and 20 of 72 in 2009, compared to 10 of 90 in 2007 and 14 of 94 in 2008); id. (illegal killings were 67% of all mortality in 2006, 62% in 2006, and 44% in 2010 compared to 19% in 2007 and 23% in 2008). Other than that, the Service concluded that it was "not possible at this time to determine if human-caused mortality (apart from mortalities from depredation control) has significantly changed over the nearly 35-year period that the gray wolf has been listed as threatened or endangered." Id. at 81,700.

What ultimately proved most relevant to the Service was that, over all of the studied time periods, "all sources of wolf mortality, including legal (for example, depredation control) and illegal human-caused mortality, have not been of sufficient magnitude to stop the continuing growth of the wolf population[.]" 2011 Rule, 76 Fed. Reg. at 81,700; see also id. ("Despite human-caused mortalities of wolves in Minnesota, Wisconsin, and Michigan, these wolf populations have continued to increase in both numbers and range.").

In short, the record supports the Service's conclusion that disease- and human-caused mortality have not materially threatened the expansion of the gray wolf population in the Western Great Lakes region, and thus the Service reasonably concluded that those factors do not counsel against delisting. See Rossello ex rel. Rossello v. Astrue, 529 F.3d 1181, 1185 (D.C. Cir. 2008) ("Substantial-evidence review is highly deferential to the agency fact-finder, requiring only `such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.'") (quoting Pierce v. Underwood, 487 U.S. 552, 565, 108 S.Ct. 2541, 101 L.Ed.2d 490 (1988)).

The Service further noted that state plans in Minnesota, Michigan, and Wisconsin would monitor wolf disease and mortality, providing an important backstop should new threats emerge. See 2011 Rule, 76 Fed. Reg. at 81,697. The veterinary pathologist peer reviewer agreed that the state plans would adequately monitor and protect against new disease threats. Finally, the data evidenced that mortality rates due to disease and depredation would not be affected in any significant measure by delisting. Id. at 81,700-81,701.

The Service's analysis also answers the Humane Society's concern about the combined impact of disease and human-caused mortality. The simple reality is that both disease- and human-caused deaths have been simultaneously afflicting the wolves, and yet the population has continued to grow nonetheless.

B

The district court also held that the 2011 Rule is arbitrary and capricious because the Service failed to adequately address what it described as an "unregulated killing zone" in Minnesota and the lack of state regulatory plans to monitor and protect the wolves in six of the Western Great Lakes States. Humane Society, 76 F.Supp.3d at 134-136. We again conclude that the Service's decision was reasonable and grounded in substantial evidence.

1

With respect to the Minnesota state plan, what the district court labeled an "unregulated killing zone" refers to depredation controls in one particular area of the State. The Minnesota Plan divides the State into two zones. Within Zone A, an area covering Northeastern Minnesota and the core of the wolves' territory, wolves can only be legally killed in defense of a human life, in situations of "immediate threat" to the life of a guard animal or domestic pet, or immediately after a verified loss of livestock, domestic animals, or pets. See MINN. STAT. § 97B.645, subdivs. 3, 5, 6. An "immediate threat" is defined as a wolf stalking, attacking, or killing. See id. § 97B.645, subdiv. 12(c). Within this zone, a "control area" can be opened immediately following the government's verification of the loss of livestock, domestic animals, or pets. See id. § 97B.671, subdiv. 4(c). However, the control area can only be opened for sixty days, and it must be within a one-mile radius of the site of the loss. See id. § 97B.671, subdiv. 4(c), (d). Trained and certified predator controllers will be paid $150 for each wolf they kill within the control area. In addition, owners of livestock and domesticated animals within Zone A may shoot or destroy a wolf that poses "an immediate threat" to such animals "on property owned, leased or occupied by the owner" of the animal. Id. § 97B.645, subdiv. 5.

Likewise, in Zone B, a controlled killing area can be opened following a verified loss of livestock, domestic animals, or pets. See MINN. STAT. § 97B.671, subdiv. 4(b). The control area may be opened for 30 to 214 days, depending on the time of year, and may be opened anytime within five years of a verified loss. See id.; J.A. 439. Only trained and certified predator controllers, operating with the permission of the landowners within the control area, may legally take a wolf. See MINN. STAT. § 97B.671, subdivs. 1, 4(b). In addition, a person may shoot a wolf on land owned, leased, or managed by that person at any time to protect that person's livestock, domestic animals, or pets, regardless of whether the wolf is an immediate threat. See id. § 97B.645, subdiv. 8.

Given those limitations, we disagree with the district court's and the Humane Society's characterization of Zone B as an "unregulated killing zone." See Humane Society, 76 F.Supp.3d at 134; Humane Society's Br. 77-78. Minnesota only permits the killing of a wolf in Zone B by a limited class of persons, and only after there has been some threat to livestock or domestic animals. And at the time the 2011 Rule issued, killing a wolf outside of the plan's specific authorizations or without a license was a gross misdemeanor. See MINN. STAT. § 97A.331, subdiv. 7 (2011), repealed by Act of May 3, 2012, ch. 277, art. 1, § 91, 2012 Minn. Laws 1169.

In other words, the Minnesota Plan does not authorize the unregulated killing of any wolf at any time by any person. It provides for regulated killing. To be sure, killings are allowed at higher numbers than the Humane Society wants. That does not make the Service's decision unreasoned or arbitrary and capricious, however. To the contrary, the Service adequately explained why Minnesota's new depredation scheme was unlikely to threaten the wolves' survival. In coming to that conclusion, the Service did not ignore that the wolves residing in Zone B "could be subject to substantial reduction in numbers, and at the extreme, wolves could be eliminated from Zone B." 2011 Rule, 76 Fed. Reg. at 81,704. Rather, the Service explained that "there is no way to reasonably evaluate in advance the extent to which residents of Zone B will use this new authority, nor how vulnerable Zone B wolves will be" in actuality. Id. What the Service did know and found relevant to its decision was that statistical data from 2007 to 2008, when the Minnesota gray wolves were not federally listed, revealed that only six wolves were shot in Zone B during that time period, and in 2009 only one additional wolf was taken. Id. Overall, the total number of wolves taken through depredation controls from 2007 to 2008 (133 in 2007, 143 in 2008) was comparable to the number that had been taken under a prior federal regulation. See id. (under pre-2007 federal regulation, 105 wolves were killed in 2004, 134 in 2005, and 122 in 2006).

Looking to data from 2007 to 2008 to predict the consequences of delisting was entirely reasonable because there was an absence of federal regulation and a presence of state depredation authorizations nearly identical to the regime that would operate after delisting. In addition, the Service is obligated to continue monitoring the gray wolf for five years following any successful delisting and to make "prompt use" of its emergency powers "to prevent a significant risk to the well being of any such recovered species." 16 U.S.C. § 1533(g)(2). Concerned entities also remain free to petition the Service to relist the gray wolf should it be threatened once more. Id. § 1533(b)(3). For those reasons, the Service's consideration of the Minnesota plan in its delisting decision was not arbitrary or capricious and was reasonably grounded in substantial evidence.

2

The Service's decision to delist notwithstanding the lack of state plans in North and South Dakota, Illinois, Iowa, Ohio, and Indiana also did not rise to the level of arbitrary-and-capricious decisionmaking, given the near non-existence of gray wolves within those jurisdictions. The boundaries of the Western Great Lakes segment includes only portions, at times very small ones, of those six States, and few if any gray wolves are found there. See, e.g., 2011 Rule, 76 Fed. Reg. at 81,671 (map of Great Lakes segment area); id. at 81,700 (sixteen total reports of wolf sightings or deaths in North Dakota from 1999 to 2003); id. at 81,713 (no gray wolves found in Ohio and Indiana). Scientific literature cited by the Service indicates that the only gray wolves found in North and South Dakota are primarily loners, not packs or mates with pups. Id. at 81,679; see also id. at 81,700 (detailing every report of wolves in North and South Dakota). Since 1993, there have been only five verified wolf deaths from depredation and eight total deaths from all causes of mortality in North and South Dakota. Id. at 81,713 (depredation deaths); id. at 81,700 (total deaths). Both Ohio and Indiana list the gray wolf as "extirpated" from their States. Id. at 81,713. Indiana, Illinois, Ohio, and Iowa, "do not contain suitable habitat [for gray wolves] or currently established packs[.]" Id.

In addition, the absence of formal state wolf plans does not mean that the few wolves in those States lack legal protection. The Service explained that, in North Dakota and Iowa, there is no open season on wolves because they are "furbearers." 2011 Rule, 76 Fed. Reg. at 81,713. Illinois includes gray wolves on its state endangered species list, making it illegal to possess or kill a wolf in the state. Id.; see 520 ILL. COMP. STAT. 10/3 (detailing state law protections for endangered species). In South Dakota, wolves are not listed as a game animal open for hunting. 2011 Rule, 76 Fed. Reg. at 81,713. In short, in North Dakota, South Dakota, Iowa, and Illinois, gray wolves are protected from indiscriminate hunting. Only in Indiana and Ohio, where wolves no longer exist, are there no formal protections in place. Id. But it is not arbitrary or capricious to overlook a State's failure to protect an animal that does not exist within its borders.

Further, for all six of the States with no wolf plans, the Service reasonably concluded that the deaths of any wolves that might enter those States would be so minimal as to pose no threat to the segment's survival. 2011 Rule, 76 Fed. Reg. at 81,713. For example, the deaths since 1993 of a total of eight wolves in North and South Dakota is quite unlikely to pose a survival threat to the several thousand wolves protected within Minnesota's borders. Similarly, the death of a lone wolf that might roam into Indiana or Ohio would be highly unlikely to affect the health or sustainability of the Western Great Lakes segment.

The district court faulted the Service's decision on the ground that the limited number of wolves "does not foreclose the possibility of an increased presence there[.]" Humane Society, 76 F.Supp.3d at 134. True. But the Endangered Species Act tasks the Service with determining whether the species is endangered or threatened, not whether the species could reach still higher population levels if given more protection. Challenges to expanding a species' territory do not by themselves undermine survival in existing territory.

We accordingly hold that the absence of conservation plans for the gray wolves in North Dakota, South Dakota, Illinois, Iowa, Ohio, and Indiana does not render the Service's decision to delist the Western Great Lakes gray wolves arbitrary and capricious on this record.

C

Finally, the Humane Society challenges the 2011 Rule as violating the Service's statutory obligation to ground the delisting decision in the best available science, 16 U.S.C. § 1533(b)(1)(A). The Humane Society argues first that science does not support the 2011 Rule because the Service admits that the Minnesota gray wolf is not a separate species of wolf under the Endangered Species Act. The Humane Society argues secondly that the 2011 Rule responded to political pressure, not science. Neither argument succeeds.

1

With respect to the status of the gray wolf in Minnesota, the Service initially proposed classifying it as a separate species (the "eastern wolf" or Canis lycaon). See Proposed Rule to Revise the List of Endangered and Threatened Wildlife for the Gray Wolf (Canis lupus), 76 Fed. Reg. 26,086, 20,086 (May 5, 2011) ("Proposed 2011 Rule"). But in the final 2011 Rule, the Service admitted that "neither a scientific consensus nor the majority opinion of researchers on the taxonomy of wolves" supported designating the wolves in Minnesota as a distinct species. 2011 Rule, 76 Fed. Reg. at 81,669. In the Humane Society's view, once the Service determined that the Minnesota gray wolf was not a separate species, the Service should have withdrawn the rule, rather than promulgate a final rule "without knowing the species that it is both listing and delisting." Humane Society's Br. 61 (emphasis in original).

But the Service did know that the Minnesota wolves that it added to the Western Great Lakes segment were part of the gray wolf species. See 2011 Rule, 76 Fed. Reg. at 81,669 ("The wolves that occupy the [Western Great Lakes segment] have long been accepted as gray wolves, C. lupus[.]"). In that respect then, the Service's treatment of the wolves in Minnesota as part of the gray wolf species aligns with the best science available, which was inconclusive as to whether the Minnesota wolves were a separate species or just a subspecies of the gray wolf. See id. at 81,668-81,669 (discussing the scientific literature on wolf taxonomy).

In any event, the Service's recognition of the Western Great Lakes segment did not depend on a finding that the Minnesota wolves were a separate species. From the outset, the Service proposed recognizing a distinct population segment in the region. See Proposed 2011 Rule, 76 Fed. Reg. at 26,094 ("Within this rule we are proposing changes to the listing for C. lupus and are initiating a status review for C. lycaon. These two actions combined will address all wolves in the [Western Great Lakes] region."); id. ("Our proposed action here is to establish the existence of a [Western Great Lakes] distinct population segment of C. lupus and to determine that the [segment] is neither endangered nor threatened, despite its proximity to a closely related species, C. lycaon — a species whose status we will evaluate for possible protection under the Act in the near future."). When the science did not support according a distinct species status to wolves in Minnesota, the Service followed where the science led by treating the Minnesota wolves as non-distinct and grouping them with all of the other wolves in the Western Great Lakes region.

2

The record likewise does not support the Humane Society's charge that the delisting decision was driven by politics rather than science. The Society's argument relies primarily on a single letter from United States Senator Amy Klobuchar of Minnesota supporting delisting, which prompted several emails within the Service discussing the Senator's letter. But a single communication from an elected representative conveying the views of constituents could not by itself politically poison an agency's decisionmaking. That is especially true when, as here, the Society does not point to any science that the Service ignored, misused, or manipulated, or to any material switch in the Service's position after receiving the letter.

On December 7, 2010, Senator Klobuchar sent a letter to Ken Salazar, the then-Secretary of the Interior, "urg[ing]" the Service "to expedite the delisting of the gray wolf in the Great Lakes" and communicating her intent to introduce legislation "to help speed-up this process." J.A. 771. Assistant Secretary Thomas Strickland's response to Senator Klobuchar referenced the 2009 Rule delisting the gray wolf in that region, which at that time had been withdrawn due to litigation, and further informed the Senator that the Service intended to publish a new proposed rule delisting the wolf by April 2011. Senator Klobuchar then made a public announcement about the Service's forthcoming proposed rule.

Such a commonplace senatorial inquiry, standing alone, cannot taint an agency decision that is otherwise adequately reasoned and grounded in the factual record. And certainly not here where the Service's threefold effort to delist the region's gray wolves preceded the Senator's letter by multiple years. See 2003 Rule, 68 Fed. Reg. at 15,857 (final rule delisting gray wolves); 2007 Rule, 72 Fed. Reg. at 6,101 (same); 2009 Rule, 74 Fed. Reg. at 15,120 (same). In addition, the Service had received multiple petitions to delist the Western Great Lakes wolves and had decided that those petitions presented substantial evidence meriting delisting several months before Senator Klobuchar's letter. See 90-Day Finding on Petitions to Delist the Gray Wolf in Minnesota, Wisconsin, Michigan and the Western Great Lakes, 75 Fed. Reg. 55,730, 55,735 (Sept. 14, 2010).

The Humane Society's reliance on Save Our Springs v. Babbitt, 27 F.Supp.2d 739 (W.D. Tex. 1997), actually proves the opposite. Unlike the case at hand, the Service in Save Our Springs succumbed to political pressure and withdrew a proposed listing of salamanders even though it had earlier stated that the salamander was the "top priority in the region for listing." Id. at 745. The level of political pressure, moreover, bears no resemblance to the single letter in this case. In Save Our Springs, the pressure came through a letter from the Governor of Texas, emails from employees worrying that listing was a "hot" issue and noting "intense opposition" to the proposal from all levels of Texas government, and "inferences that political lobbyists for the development community worked with political appointees." Id. The record here reflects no such concerted pressure, no manipulation or disregard of material evidence, and no change in the Service's course of action before and after the single communication.10

We accordingly find no merit to the charge of improper political influence in this case.

V

A common remedy when we find a rule is invalid is to vacate. See Sugar Cane Growers Co-op. of Fla. v. Veneman, 289 F.3d 89, 97 (D.C. Cir. 2002) ("Normally when an Agency * * * clearly violates the APA we would vacate its action[.]"). But we may remand without vacatur depending upon "the seriousness of the order's deficiencies (and thus the extent of doubt whether the agency chose correctly) and the disruptive consequences of [vacating the Rule,] an interim change that may itself be changed." Id. at 98 (citation omitted); Allina Health Servs. v. Sebelius, 746 F.3d 1102, 1110 (D.C. Cir. 2014); see also United States Sugar Corp. v. EPA, 844 F.3d 268, 270 (D.C. Cir. 2016) (per curiam) ("Although remand without vacatur may in some circumstances invite prejudicial agency delay, in other circumstances vacatur itself carries more-harmful consequences. We have therefore frequently remanded without vacating when a rule's defects are curable and where vacatur would at least temporarily defeat * * * the enhanced protection of the [rule].") (first alteration in original; internal quotation marks and citations omitted).

In this case, the agency's analysis (i) wholly failed to address the effect on the remnant population of carving out the Western Great Lakes segment, and in doing so (ii) misapplied the Service's own discreteness and significance tests, and also (iii) turned its back on the implications of historical range loss. Those are major shortcomings that go to the heart of the Service's delisting decision. Given the serious and pervading role those deficiencies played in the agency's decisionmaking, there is substantial "doubt whether the [Service] chose correctly" in promulgating the 2011 Rule, Sugar Cane Growers, 289 F.3d at 98 (citation omitted). That makes vacatur appropriate. See id. at 97-98; Fox Television Stations, Inc. v. FCC, 280 F.3d 1027, 1049, 1052-1053 (D.C. Cir. 2002).

In addition, vacatur would not trigger disruptive consequences. The agency has failed repeatedly over the last sixteen years to make a delisting decision that complies with the APA, and it has not shown that vacatur here would be any more disruptive than it was on the Service's last three failed occasions. With respect to protecting domestic animals in the interim, federal regulations already permit depredation control in Minnesota, which is where most of the gray wolves in the Western Great Lakes segment live. See 50 C.F.R. § 17.40(d).

Because of the "seriousness of the [Rule's] deficiencies" and the absence of materially "disruptive consequences," we affirm the judgment of the district court vacating the Service's 2011 Rule.

So ordered.