Ctr. for Biological Diversity v. Ryan Zinke

FOR PUBLICATION

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

CENTER FOR BIOLOGICAL DIVERSITY; No. 16-35866
WESTERN WATERSHEDS PROJECT;
GEORGE WUERTHNER; PAT D.C. No.
MUNDAY, 2:15-cv-00004-
Plaintiffs-Appellants, SHE

v.
OPINION
RYAN K. ZINKE, Secretary, U.S.
Department of the Interior, in his
official capacity; DAN ASHE,
Director, U.S. Fish and Wildlife
Service, in his official capacity; U.S.
FISH & WILDLIFE SERVICE,
Defendants-Appellees,

and

STATE OF MONTANA; MONTANA
DEPARTMENT OF FISH, WILDLIFE
AND PARKS,
Intervenor-Defendants-Appellees.

Appeal from the United States District Court
for the District of Montana
Sam E. Haddon, Senior District Judge, Presiding

Argued and Submitted February 8, 2018
Seattle, Washington
2 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

Filed August 17, 2018

Before: Raymond C. Fisher, Ronald M. Gould,
and Richard A. Paez, Circuit Judges.

Opinion by Judge Paez

SUMMARY *

Environmental Law

The panel affirmed in part and reversed in part the
district court’s summary judgment in favor of the U.S. Fish
and Wildlife Service, in an action challenging the Service’s
decision, based on a 2014 Finding, not to list the arctic
grayling as an endangered or threatened species under the
Endangered Species Act.

In 2014, the Service promulgated its “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 (July 1, 2014) (“SPR policy”).

The panel affirmed the district court’s holding that the
Service did not err in considering only the current range of
the arctic grayling when determining whether it was in
danger of extinction “in all or a significant portion of its
range.” 16 U.S.C. §§ 1532(6), (20). The panel rejected

*
This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 3

plaintiffs’ contention that it was bound by two prior
decisions, Defenders of Wildlife v. Norton, 258 F.3d 1136
(9th Cir. 2001), and Tucson Herpetological Society v.
Salazar, 566 F.3d 870 (9th Cir. 2009), in deciding whether
“range” meant historical rather than current range. Applying
Chevron analysis, the panel held that the meaning of “range”
in 16 U.S.C. § 1532(6) and (20) was ambiguous; and
concluded that the SPR policy’s interpretation of “range” as
“current range” for the purposes of 16 U.S.C. § 1532 was a
reasonable interpretation of the statute, and warranted
deference.

The panel held that the Service acted in an arbitrary and
capricious manner in finding that the fluvial arctic grayling
population was increasing because it ignored available
biological data showing that the arctic grayling population
in the Big Hole River in Montana was declining.

The panel held that the Service acted arbitrarily and
capriciously by dismissing threats of low stream flows and
high stream temperatures to the arctic grayling. Specifically,
the panel held that the 2014 Finding’s reliance on cold water
refugia in the Big Hole River was arbitrary and capricious
and the district court’s summary judgment in favor of the
Service of this issue was error, but the panel affirmed the
district court’s ruling on the cold water refugia issue in all
other aspects.

The panel held that the Service acted in an arbitrary and
capricious manner by failing to explain why the uncertainty
of climate change favors not listing the arctic grayling when
the 2014 Finding acknowledged the warming of water
temperatures and decreasing water flow because of global
warning.
4 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

The panel considered the Service’s dismissal of threats
of small population sizes. Affirming the district court, the
panel held that the Service’s determination that the arctic
grayling’s small population size did not pose a risk to genetic
viability of the arctic grayling was not arbitrary or
capricious. Reversing the district court, the panel held the
2014 Finding did arbitrarily rely on the Ruby River
population to provide redundancy of fluvial arctic grayling
beyond the Big Hole River and to minimize the risk from
random environmental events.

The panel concluded that the 2014 Finding’s decision
that listing arctic grayling was not “warranted” was arbitrary
and capricious. The panel remanded with directions to
remand to the Service to reassess the 2014 Finding in light
of this opinion.

COUNSEL

Jenny K. Harbine (argued), Aurora R. Janke, and Timothy J.
Preso, Earthjustice, Bozeman, Montana, for Plaintiffs-
Appellants.

Thekla Hansen-Young (argued), Nicole M. Smith, Allen M.
Brabender, and Andrew C. Mergen, Attorneys; Jeffrey H.
Wood, Acting Assistant Attorney General; Environment and
Natural Resources Division, United States Department of
Justice, Washington, D.C.; Kate Williams-Shuck, Office of
the Solicitor, United States Department of the Interior,
Denver, Colorado; for Defendants-Appellees.

William A. Schenk (argued) and Zachary C. Zipfel, Special
Assistant Attorneys General, Montana Department of Fish,
Wildlife and Parks, Helena, Montana; Jeremiah Weiner,
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 5

Assistant Attorney General, Office of the Attorney General,
Helena, Montana; for Intervenor-Defendants-Appellees.

Deborah A. Sivas, Alicia Thesing, and Isaac Cheng,
Environmental Law Clinic, Mills Legal Clinic, Stanford,
California, for Amici Curiae Law Professors and Scientists.

OPINION

PAEZ, Circuit Judge:

This Endangered Species Act (“ESA”) case concerns the
Upper Missouri River Valley Distinct Population Segment
of Arctic Grayling (“arctic grayling”), a cold-water fish in
the Salmonidae family. Before us is a challenge to the Fish
and Wildlife Service (“FWS”)’s decision not to list the arctic
grayling as an endangered or threatened species under the
ESA. Plaintiffs-Appellants Center for Biological Diversity,
Western Watersheds Project, George Wuerthner, and Pat
Mundy (collectively “CBD”) argue that FWS erred in using
an incorrect definition of “range” in determining whether the
arctic grayling is extinct or in threat of becoming extinct “in
a significant portion of its range.” Additionally, CBD
challenges several aspects of the listing decision as arbitrary
and capricious.

The district court granted summary judgment in favor of
FWS. We have jurisdiction under 28 U.S.C. § 1291.
Because we conclude that in certain respects FWS acted in
an arbitrary and capricious manner, we reverse the district
court’s order granting summary judgment with instructions
to remand the arctic grayling listing decision to FWS for
further consideration.
6 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

I.

We begin by describing characteristics of the arctic
grayling population, as relevant to the challenged listing
decision. The arctic grayling is a cold-water fish belonging
to the Salmonidae family. It has a trout-like body with a
deeply forked tail and a sail-like dorsal fin. There are two
types of arctic grayling: fluvial, which dwell in rivers and
streams, and adfluvial, which dwell in lakes and migrate to
streams to spawn. Historically, fluvial populations
predominated in the Upper Missouri River. The two types
of arctic grayling are genetically distinct, although
experiments have shown some plasticity in the
characteristics between adfluvial and fluvial populations.
Although fluvial arctic grayling have been shown to adapt to
lake environments, all attempts to introduce adfluvial arctic
grayling to streams have failed. Given its adaptability, the
fluvial population of arctic grayling is considered to be
especially important to the survival of the species.

Within the contiguous United States, arctic grayling
historically existed in Montana, Wyoming, and Michigan.
Today, it exists only in the Upper Missouri River Basin in
Montana. Due to a host of threats to the arctic grayling’s
habitat, it presently occupies only a small fraction of its
historical range. Fluvial arctic grayling, for example,
currently occupy less than ten percent of their historical
range in the Missouri River system. There are presently
twenty-six populations of arctic grayling in the Upper
Missouri River Basin. Six of these—the Big Hole River,
Ennis Reservoir/Madison River, Centennial Valley’s lakes
and tributaries, Mussigbrod Lake, Miner Lake, and Ruby
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 7

River populations—are native populations. 1 The other
twenty populations have been introduced into habitat that
was not part of the arctic grayling’s historical range. Six of
these introduced populations have no significant
conservation value, as they have not yet become fully
established. The other twenty populations—both native and
introduced—have conservation value. 2 Two of the
populations—the Big Hole River and Ennis
Reservoir/Madison River populations—are located
primarily on private land, whereas the remaining eighteen
are found either entirely or primarily on federal land.

Arctic grayling prefer cooler water temperatures;
temperatures over 20 degrees Celsius (approximately
70 degrees Fahrenheit) can cause physiological stress and
impair biological functions such as breeding. The upper
incipient lethal temperature (“UILT”) is 25 degrees Celsius
(77 degrees Fahrenheit). 3 The twenty-six arctic grayling
populations in the Upper Missouri River Basin are
“biogeographically important to the species” because they
have adapted to warmer water temperatures in contrast to
populations of arctic grayling outside of the Upper Missouri
River Basin. Despite this adaptation, climate change
threatens the arctic grayling. Less water in streams poses a

1
Of these six native populations, only the Big Hole River population
and Ruby River population—which contains just forty-two breeding
adults—are entirely fluvial.

2
Of these twenty populations, sixteen are adfluvial, two are fluvial,
and two more are mixed fluvial/adfluvial.

3
The UILT is the temperature at which there is a 50% survival for
over a week in a “test population.” Thus, if the water temperature were
to remain this high for a week, scientists would predict 50% of the fish
in the test population would perish.
8 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

threat to the arctic grayling. Droughts and warmer-than-
normal air temperatures can reduce water levels and,
consequently, raise water temperatures higher than the range
of temperatures that the arctic grayling can tolerate.

II.

FWS must follow certain ESA requirements when
deciding whether to list a species as endangered or
threatened. We briefly review those requirements. We next
discuss the decades-long history of FWS listing decisions
involving the arctic grayling. We then briefly review the
procedural history of this case.

A.

The ESA is “the most comprehensive legislation for the
preservation of endangered species ever enacted by any
nation.” Tenn. Valley Auth. v. Hill, 437 U.S. 153, 180
(1978). It represents a commitment “to halt and reverse the
trend toward species extinction, whatever the cost.” Id. at
184; see also 16 U.S.C. § 1531(a)(1) (2012).

Under the ESA, the Secretary of the Interior
(“Secretary”) is charged with determining whether particular
species should be listed as “threatened” or “endangered.”
16 U.S.C. § 1533. 4 An endangered species is “any species
which is in danger of extinction throughout all or a
significant portion of its range.” Id. § 1532(6). A threatened
species is “any species which is likely to become an
endangered species within the foreseeable future throughout
all or a significant portion of its range.” Id. § 1532(20). The
4
The Secretary has delegated his authority to implement the ESA—
including his authority to make listing decisions—to FWS. See
50 C.F.R. § 402.01(b) (2017).
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 9

term “species” includes subspecies and “any distinct
population segment of any species of vertebrate fish or
wildlife which interbreeds when mature.” Id. § 1532(16).
Under this definition, a distinct population segment of a
species can be listed as an endangered or threatened species.

FWS must base its listing decision on “the best scientific
and commercial data available.” Id. § 1533(b)(1)(A). To
comply with this requirement, FWS “cannot ignore available
biological information.” Conner v. Burford, 848 F.2d 1441,
1454 (9th Cir. 1988); see also San Luis & Delta-Mendota
Water Auth. v. Locke, 776 F.3d 971, 995 (9th Cir. 2014)
(“An agency complies with the best available science
standard so long as it does not ignore available studies, even
if it disagrees with or discredits them.”). In making a listing
decision, FWS must evaluate five factors: “(a) the present or
threatened destruction, modification, or curtailment of [a
species’] habitat or range; (b) overutilization for
commercial, recreational, scientific, or educational
purposes; (c) disease or predation; (d) the inadequacy of
existing regulatory mechanisms; or (e) other natural or
manmade factors affecting [the species’] continued
existence.” 16 U.S.C. § 1533(a)(1).

Anyone may petition FWS to list a species. Id.
§ 1533(b)(3)(A). “To the maximum extent practicable,”
within ninety days of the petition FWS must determine
whether the petition presents “substantial scientific or
commercial information indicating that the petitioned action
may be warranted.” Id. If it does, FWS reviews the status
of the species and makes a “12-month finding” that listing
the species is either (a) not warranted; (b) warranted; or
(c) warranted but precluded by higher priority pending
proposals. Id. § 1533(b)(3)(B); 50 C.F.R. § 424.14. Species
in the third category become “candidates” for listing, and
10 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

FWS continues to review their status until it makes a
“warranted” or “not warranted” finding. 16 U.S.C.
§ 1533(b)(3)(C)(i). Additionally, FWS assigns these species
a listing-priority number. See Endangered Species Listing
and Recovery Priority Guidelines, 48 Fed. Reg. 43,098,
43,098 (Sept. 21, 1983). A 12-month finding that listing is
“not warranted” is subject to judicial review. 16 U.S.C.
§ 1533(b)(3)(C)(ii).

B.

FWS initially considered whether to list the arctic
grayling as endangered or threatened in 1982. Endangered
and Threatened Wildlife and Plants; Review of Vertebrate
Wildlife for Listing as Endangered or Threatened Species,
47 Fed. Reg. 58,454 (Dec. 30, 1982). FWS determined that
listing the arctic grayling was “possibly appropriate” but
chose not to list it due to a lack of sufficient data. Id. at
58,454–55. Two of the plaintiffs in the current case—
George Wuerthner and the Center for Biological
Diversity 5—then petitioned FWS to list the fluvial arctic
grayling as an endangered species. In response to that
petition, FWS determined in 1994 that listing the arctic
grayling was “warranted but precluded” by other listing
obligations, as threats were of moderate-to-low magnitude
due to “ongoing cooperative conservation actions.”
Endangered and Threatened Wildlife and Plants; Finding on
a Petition to List the Fluvial Population of the Arctic
Grayling as Endangered, 59 Fed. Reg. 37,738, 37,740–41
(July 25, 1994). FWS therefore gave the arctic grayling a
listing priority of nine. Id. The arctic grayling maintained
this status until 2003, when the Center for Biological

5
The Center for Biological Diversity was at the time called the
Biodiversity Legal Foundation.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 11

Diversity and the Western Watersheds Project challenged
the 1994 “warranted but precluded” decision in a complaint
filed in the United States District Court for the District of
Columbia. 6 In response, FWS raised the listing priority of
the arctic grayling to three, the highest priority that could be
afforded to a distinct population segment. The plaintiffs
responded with an amended complaint, requesting that FWS
emergency list the arctic grayling as either endangered or
threatened. The parties settled, with FWS agreeing to issue
a revised listing determination by April 2007.

In April 2007, FWS concluded that the arctic grayling
did not warrant protection because it was not a distinct
population segment, and therefore could not be listed as an
endangered or threatened species under the ESA.
Endangered and Threatened Wildlife and Plants; Revised
12-Month Finding for Upper Missouri River Distinct
Population Segment of Fluvial Arctic Grayling, 72 Fed. Reg.
20,305, 20,305 (Apr. 24, 2007). The Center for Biological
Diversity, Federation of Fly Fishers, Western Watersheds
Project, George Werthner, and Pat Munday filed an action in
the District Court for District of Montana challenging the
2007 listing decision. 7 The parties ultimately settled, with
FWS stipulating that by August 30, 2010 it would determine
whether listing the arctic grayling was warranted. In 2010,
FWS published a revised listing decision, concluding that
the arctic grayling was a distinct population segment and that
listing was “warranted but precluded” by higher priority
actions (“2010 Finding”). Endangered and Threatened

6
Ctr. for Biological Diversity v. U.S. Fish & Wildlife Serv., No.
CIV.A. 03-1110(JDB) (D.D.C.).
7
Ctr. for Biological Diversity v. U.S. Fish & Wildlife Serv., 1:07-
cv-00152-RFC (D. Mont.).
12 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

Wildlife and Plants; Revised 12-Month Finding to List the
Upper Missouri River Distinct Population Segment of Arctic
Grayling as Endangered or Threatened, 75 Fed. Reg. 54,708,
54,708 (Sept. 8, 2010).

The 2010 Finding was based on a variety of threats
facing the arctic grayling. For example, the 2010 Finding
noted low abundance of arctic grayling and downward
population trends, especially in the Big Hole River. Id. at
54,723. These trends played a role in FWS’s evaluation of
the arctic grayling’s range and habitat. Id. The 2010 Finding
also determined that low stream flows and high stream
temperatures imperiled the arctic grayling, and showed
resulting present and threatened destruction of grayling
habitat or range. Id. at 54,726–30. Additionally, climate
change was evaluated as an “other natural or manmade
factor” that would “potentially intensify some of the
significant current threats to all Arctic grayling populations.”
Id. at 54,739–40. In evaluating other factors relevant to the
“warranted but precluded” decision, the 2010 Finding also
found that small population sizes threatened the survival of
the species. Id. at 54,740–41. The effective population was
below the number needed for long-term genetic viability and
stochastic events could threaten the survival of the arctic
grayling due to its small population size. Id.

Shortly after the 2010 Finding, FWS settled numerous
lawsuits in a multi-district litigation concerning the backlog
of ESA listing decisions. 8 As part of that settlement, FWS
stipulated that it would issue either a proposed listing rule or

8
These cases were known as In re Endangered Species Act Section
4 Deadline Litigation, Misc. Action No. 10-377 (EGS), MDL Docket
No. 2185 (D.D.C.).
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 13

a not-warranted finding for the arctic grayling by the end of
fiscal year 2014.

As required, in 2014 FWS released its decision finding
that listing the arctic grayling as endangered or threatened
was not warranted (“2014 Finding”). Endangered and
Threatened Wildlife and Plants; Revised 12-Month Finding
on a Petition to List the Upper Missouri River Distinct
Population Segment of Arctic Grayling as an Endangered or
Threatened Species, 79 Fed. Reg. 49,384 (Aug. 20, 2014).
The 2014 Finding was based on a number of conclusions,
including some that were directly at odds with conclusions
in the 2010 Finding. In contrast to the 2010 Finding, FWS
found that the population of arctic grayling was actually
increasing, including an increase in abundance of the two
fluvial populations. The 2014 Finding relied on the Big Hole
River Candidate Conservation Agreement with Assurances
(“CCAA”) and related Strategic Habitat Conservation Plan,
in evaluating the threatened destruction, modification, or
curtailment of arctic grayling habitat. 9 Id. at 49,400–02.
FWS found that the CCAA would mitigate threats from low
stream flows and high water temperatures. Id. Additionally,
FWS found that climate change was not a threat because the
arctic grayling’s increased abundance and distribution in
conjunction with riparian restoration efforts mitigated
climate-change effects. Id. at 49,407. The 2014 Finding also
dismissed the risks from inbreeding and environmental

9
A CCAA is an agreement between FWS and non-Federal property
owners. The property owners agree to manage their land in a manner
that will reduce threats to a species in exchange for assurances against
additional regulatory requirements if that species is later listed as
threatened or endangered under the ESA. These CCAAs work to
implement the broader Strategic Habitat Conservation Plan for the arctic
grayling, which provides a framework for achieving positive biological
outcomes for the arctic grayling.
14 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

disturbances because of increasing population numbers and
geographic separation between populations.

In analyzing the curtailment of the arctic grayling’s
range, habitat fragmentation, and the effect of man-made
disturbances on the arctic grayling, FWS considered the
arctic grayling’s historic range. In deciding whether the
arctic grayling was threatened or endangered in a
“significant portion of its range” as provided in 16 U.S.C.
§ 1532(6) and (20), however, FWS interpreted “range” as
the arctic grayling’s “current range” rather than the range it
had historically inhabited.

C.

In February 2015, CBD challenged FWS’s negative
listing decision by filing this action in the District Court for
the District of Montana. CBD alleged that the 2014 Finding
(1) arbitrarily relied on unsupported population increases to
conclude that the arctic grayling is not threatened by small
population size; (2) did not properly evaluate whether the
arctic grayling is threatened by lack of water in streams and
high water temperatures, which will only be exacerbated by
global warming; and (3) did not properly analyze whether
lost historical range constitutes a “significant portion of [the
arctic grayling’s] range.” The State of Montana and the
Montana Department of Fish, Wildlife and Parks
(collectively, “Montana”) intervened as defendants in the
action. The parties filed cross-motions for summary
judgment and the district court granted summary judgment
in favor of FWS and Montana, rejecting each of CBD’s
claims.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 15

III.

We turn to the merits and address first CBD’s claim that
FWS erred in considering only the current range of the arctic
grayling when determining whether it was in danger of
extinction “in all or a significant portion of its range.”
16 U.S.C. § 1532(6), (20). The district court concluded that
FWS did not err by doing so. Reviewing the district court’s
summary judgment ruling de novo, we agree and affirm its
ruling on this issue. Greater Yellowstone Coal., Inc. v.
Servheen, 665 F.3d 1015, 1023 (9th Cir. 2011).

In 2014, FWS promulgated its “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 (July 1, 2014) (“SPR policy”). This policy
defined “range” as follows:

The range of a species is considered to be the
general geographical area within which
that species can be found at the time [FWS]
makes any particular status determination.
This range includes those areas used
throughout all or part of the species’ life
cycle, even if they are not used regularly
(e.g., seasonal habitats). Lost historical range
is relevant to the analysis of the status of the
species, but it cannot constitute a significant
portion of a species’ range.

Id. at 37,609.

As the SPR policy was enacted through notice-and-
comment rulemaking procedures as required by 16 U.S.C.
§ 1533(h), we apply the deference framework established by
16 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

Chevron. See Nw. Ecosystem All. v. U.S. Fish & Wildlife
Serv., 475 F.3d 1136, 1141–42 (9th Cir. 2007). We ask if
the meaning of “range” in 16 U.S.C. § 1532(6) and (20) is
ambiguous. If the term “range” is ambiguous, then we must
defer to the SPR policy unless it is an unreasonable
interpretation of the statute. See Chevron U.S.A., Inc. v. Nat.
Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984).

A.

Before addressing the first step of Chevron, we consider
the significance of two of our prior decisions—Defenders of
Wildlife v. Norton, 258 F.3d 1136 (9th Cir. 2001), and
Tucson Herpetological Society v. Salazar, 566 F.3d 870 (9th
Cir. 2009). CBD argues that these two cases hold that
“range” means historical range rather than current range for
the purposes of 16 U.S.C. § 1532(6) and (20), and therefore
we are bound by our precedent to invalidate the SPR policy.
A prior appellate court decision construing a statute controls
over a later agency decision “only if the prior court decision
holds that its construction follows from the unambiguous
terms of the statute and thus leaves no room for agency
discretion.” Nat’l Cable & Telecomms. Ass’n v. Brand X
Internet Servs., 545 U.S. 967, 982–83 (2005). We conclude
that neither of our previous decisions held that “range”
unambiguously means “historical range.”

In Defenders of Wildlife, we held that it was error for the
Secretary to fail to list the flat-tailed horned lizard as
“endangered” or “threatened” based on a finding that
adequate habitat existed on public land for the lizard, despite
recognizing that the lizard faced threats on private land.
258 F.3d at 1140. We analyzed the phrase “in danger of
extinction throughout . . . a significant portion of its range”
to determine if it required considering the threats that the
lizard faced on private land. Id. at 1140–41. We concluded
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 17

that “significant portion of its range” is inherently
ambiguous, and therefore the Secretary has a “wide degree
of discretion in delineating [what] ‘a significant portion of
its range’” means. Id. at 1145. We noted, however, that
where “it is on the record apparent that the area in which the
[species] is expected to survive is much smaller than its
historical range, the Secretary must at least explain her
conclusion that the area in which the species can no longer
live is not a ‘significant portion of its range.’” Id. Because
the Secretary did not address whether the private land it
discounted was sufficient to render the lizard “extinct . . . in
a significant portion of its range,” we reversed and
remanded. Id. at 1146–47.

Tucson Herpetological Society, which also involved the
flat-tailed horned lizard, interpreted Defenders of Wildlife to
mean that the criteria for “significance” is undefined, but that
FWS must “develop some rational explanation for why the
lost and threatened portions of a species’ range are
insignificant before deciding not to designate the species for
protection.” 566 F.3d at 876–77. Additionally, we
concluded that Defenders of Wildlife required FWS to
analyze lost historical range. Id. We ultimately held that
FWS had properly analyzed the flat-tailed horned lizard’s
lost historical range. Id. at 878.

Although Defenders of Wildlife and Tucson
Herpetological Society held that FWS must at least explain
why the lost and threatened portions of a species’ range are
insignificant before disregarding historical range, it does not
follow from their holdings that the ESA’s use of “range” in
16 U.S.C. § 1532 unambiguously refers to the species’
current range. Rather, we have noted that the phrase “extinct
throughout . . . a significant portion of [a species’] range” is
ambiguous. See Defs. of Wildlife, 258 F.3d at 1141. Because
18 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

neither of these cases provide a clear statement that “range”
unambiguously means “historical range,” we proceed to
consider CBD’s other arguments in favor of such a reading.
See Brand X, 545 U.S. at 982–83.

B.

Under Chevron, first we must ask if the meaning of
“range” in 16 U.S.C. § 1532(6) and (20) is ambiguous. CBD
argues that the term “range” unambiguously means a
species’ historical range. Although the term “range” is not
defined in the ESA, CBD argues that using traditional tools
of statutory construction, we can ascertain that Congress
unambiguously intended “range” to mean “historical range”
and therefore that “intention is the law and must be given
effect.” Chevron, 467 U.S. at 843 n.9.

Whether FWS’s interpretation of range in the SPR policy
deserves deference under Chevron is a question of first
impression in this circuit. The D.C. Circuit recently
considered this question in Humane Society of the United
States v. Zinke, 865 F.3d 585 (D.C. Cir. 2017), and
concluded that deference was warranted. In doing so it
analyzed the dictionary definition of the word “range,” its
use in 16 U.S.C. § 1532(6) and (20) and the use of “range”
in three other parts of the ESA to determine that “range” was
ambiguous. Id. at 604. For the following reasons, we agree
that “traditional rules of statutory construction do not answer
the question of whether ‘range’ means current or historical
range.” Id.

Starting with the text of 16 U.S.C. § 1532(6) and (20),
the word “range” is ambiguous as to whether it means
current or historical range. “Range” is commonly defined as
“a geographical reference to the physical area in which a
species lives or occurs.” Id. (citing 8 The Oxford English
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 19

Dictionary 139 (def. 7) (1933)). This definition, however,
does not provide clarity to the meaning of “range.” One
could argue that the use of the present tense in the
definition—“lives or occurs”—may suggest that range
means “current range.” As the D.C. Circuit observed,
however, the use of the present tense is likely a function of
dictionary drafting. Id. We agree with the D.C. Circuit that
the dictionary definition of the word “range” itself does not
shed much light on Congress’s intent.

The text of 16 U.S.C. § 1532 as a whole tells us equally
little. An endangered species is one that “is in danger of
extinction throughout all or a significant portion of its
range.” 16 U.S.C. § 1532(6). Similarly, a “threatened
species” is “any species which is likely to become an
endangered species within the foreseeable future throughout
all or a significant portion of its range.” 16 U.S.C.
§ 1532(20). The SPR policy explains that because these
definitions are phrased in the present tense, “range” means
current range. “[T]o say a species ‘is in danger’ in an area
where it no longer exists—i.e., in its historical range where
it has been extirpated—is inconsistent with common usage.”
79 Fed. Reg. at 37,583.

Like the D.C. Circuit, we conclude that the use of the
present tense in §1532(6) and (20) does not mean that
“range” must mean “current range.” The placement of the
present-tense “is” seems to require that the species
“currently be endangered or threatened within its range, not
to dictate the temporal scope of geographical evidence
[FWS] is to consider.” Humane Soc’y, 865 F.3d at 604. As
we explained in Defenders of Wildlife, “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.” 258 F.3d at 1145. CBD similarly
20 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

notes that a species can be “in danger” even in an area where
it does not currently exist but could in the future, if habitat
degradation endangers the possibility of reintroduction to
that habitat. 10 The statutory text does not demonstrate that
Congress unambiguously intended “range” to mean either
“current” or “historical range.”

Because the text of 16 U.S.C. § 1532(6) and (20) does
not conclusively shed light on the scope of the word “range,”
we next consider the statutory framework of the ESA and the
other uses of the word “range” throughout the statute. See
Sullivan v. Everhart, 494 U.S. 83, 89 (1990) (“In
ascertaining the plain meaning of the statute, the court must
look to the particular statutory language at issue, as well as
the language and design of the statute as a whole.” (citation
omitted)). Outside of the definitions of “endangered” and
“threatened” species, the term “range” appears three times in
the ESA.

The first use of “range” is in section 4(a)(1) of the ESA,
which lists “the present or threatened destruction,
modification, or curtailment of [a species’] habitat or range”
as one factor that FWS considers in its listing decision.
16 U.S.C. § 1533(a)(1)(A). This reference to “range” is “as
textually indeterminate as the initial use of the term in
[16 U.S.C. §] 1532.” Humane Soc’y, 865 F.3d at 604.
Indeed, while “present” may modify “habitat or range,” it
more likely modifies “destruction, modification, or

10
This interpretation has some force in light of the ESA provisions
that seek to recover lost range once a species is listed. See, e.g.,
16 U.S.C. §§ 1533(a)(3), 1539(j)(2)(A). Indeed, in this case many of the
current arctic grayling populations were reintroduced into their current
habitats.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 21

curtailment.” The use of the word “range” in this section
does not shed any light on Congress’s intent.

The next use of “range” is in section 4(c)(1), which
requires FWS to specify “over what portion of its range [the
species] is endangered or threatened, and specify any critical
habitat within such range.” 16 U.S.C. § 1533(c)(1). We
acknowledge that legislative history indicates that in this
section of the ESA, Congress used “[t]he term ‘range’ . . . in
the general sense [to] refer[] to the historical range of the
species.” H.R. Rep. No. 95-1625, at 18 (1978), reprinted in
1978 U.S.C.C.A.N. 9453, 9468. Combined with the
“presumption that a given term is used to mean the same
thing throughout a statute,” see Brown v. Gardner, 513 U.S.
115, 118 (1994), this legislative history provides support for
interpreting “range” as meaning historical range in 16 U.S.C.
§ 1532(6) and (20) as well.

We are not convinced that this lone indicator of
legislative intent disposes of all ambiguity as to the scope of
“range” throughout the ESA. The SPR Policy reads section
4(c)(1) as “an informational rather than a substantive
provision,” that is, an instruction that the agency should
specify where a species is endangered or threatened and
identify the critical habitat in those areas. 79 Fed. Reg. at
37,583. Indeed, as FWS recognizes, reading section 4(c)(1)
as a substantive provision may actually limit the scope of
protection for species under the ESA, as in that case critical
habitat could only be designated in the species’ “range.” Id.
In sum, we agree with the D.C. Circuit that this section also
does not unambiguously shed light on Congress’s intended
meaning of “range.”

The third use of “range” in the ESA is found in section
10(j), which “authorize[s] the release . . . of any population
. . . of an endangered species or a threatened species outside
22 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

the current range of such species.” 16 U.S.C.
§ 1539(j)(2)(A). As the D.C. Circuit noted, this provision
cuts both ways. Humane Soc’y, 865 F.3d at 604. On the one
hand, the use of the word “current” as a qualifier to “range”
in section 10(j) could indicate that elsewhere in the ESA
“range” means “historical range.” See, e.g., Duncan v.
Walker, 533 U.S. 167, 174 (2001) (describing the rule
against surplusage). By using “current range” in section
10(j), Congress knew how to limit “range” to mean “current
range,” and therefore it could have intended the broader
historical meaning in 16 U.S.C. § 1532(6) and (20). See
Hamdan v. Rumsfeld, 548 U.S. 557, 578 (2006). On the
other hand, the use of “current range” in section 10(j) “could
also be read to corroborate [FWS’s] view, since ‘current
range’ . . . may refer to the listed range of the endangered or
threatened species.” Humane Soc’y, 865 F.3d at 604. The
use of “range” in section 10(j) does not compel the
conclusion that “range” should be read to unambiguously
mean “historical range.”

Considering the statutory framework as a whole, then,
the term “range” in 16 U.S.C. § 1532(6) and (20) is
ambiguous. Although traditional tools of statutory
construction provide some support for interpreting “range”
to mean “historical range,” we are not persuaded that the
“unambiguously expressed intent of Congress” was to define
“range” as “historical range.” Chevron, 467 U.S. at 843. We
therefore conclude that the term “range” is ambiguous and
proceed to evaluate whether the SPR policy is “based on a
permissible construction of the statute.” Id.

C.

Since “range” is ambiguous, we must next determine
whether the SPR policy’s interpretation of “range” as
“current range” for the purposes of 16 U.S.C. § 1532 is a
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 23

reasonable interpretation of the statute. Id. We join the D.C.
Circuit in holding that it is. Humane Soc’y, 865 F.3d at 605.
As we discussed above, the statutory framework of the ESA
provides at least some support for interpreting “range” as the
current range of a species, although it may not compel this
interpretation. Additionally, despite CBD’s arguments to
the contrary, the SPR policy does not run afoul of the
purposes of the ESA. The largest threat to potentially
endangered or threatened species is the loss of habitat that
the species currently occupies. See id. (noting that where a
species “currently lives often affect its continued survival the
most and thus bear influentially on whether it should be
listed”). It would therefore be reasonable for FWS to focus
on the area the species currently occupies when evaluating
whether the species is endangered through “a significant
portion of its range.” Id. at 604.

The SPR policy still requires that FWS consider the
historical range of a species in evaluating other aspects of
the agency’s listing decision, including habitat degradation.
Id. at 605–06. The SPR policy recognizes that loss of
historical range can lead to reduced abundance, inhibited
gene flow, and increased susceptibility to extinction. 79 Fed.
Reg. at 37,584. The SPR policy’s interpretation of “range”
is “consonant with the purposes of the [ESA],” as it provides
protections for species that have lost a substantial part of
their historical range. See Human Soc’y, 865 F.3d at 605.

In sum, we hold that FWS’s interpretation of “range” in
16 U.S.C. § 1532(6) and (20) as “current range” warrants
deference. Consistent with that interpretation, FWS did
consider the arctic grayling’s historical range in evaluating
the factors that contributed to its negative listing decision.
24 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

IV.

We next turn to CBD’s other arguments that the 2014
Finding was arbitrary and capricious. FWS’s decision not to
list a species under the ESA is reviewed under the
Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701–
706. Native Ecosystems Council v. Dombeck, 304 F.3d 886,
901 (9th Cir. 2002). We “shall” set aside agency actions,
findings, or conclusions under the APA that are “arbitrary,
capricious, an abuse of discretion, or otherwise not in
accordance with law.” Japanese Vill., LLC v. Fed. Transit
Admin., 843 F.3d 445, 453 (9th Cir. 2016) (quoting 5 U.S.C.
§ 706(2)(A)).

In reviewing whether an agency decision is arbitrary or
capricious, we “ensure that the agency considered the
relevant factors and articulated a rational connection
between the facts found and the choices made.” Greater
Yellowstone Coal., 665 F.3d at 1023 (citation omitted).
“[A]n agency rule would be arbitrary and capricious if the
agency has relied on factors which Congress has not
intended it to consider, entirely failed to consider an
important aspect of the problem, offered an explanation for
its decision that runs counter to the evidence before the
agency, or is so implausible that it could not be ascribed to a
difference in view or the product of agency expertise.” Id.
(quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto.
Ins. Co., 463 U.S. 29, 43 (1983)).

Agency decisions deserve the highest deference when
“the agency is making predictions, within its area of special
expertise.” Lands Council v. McNair, 537 F.3d 981, 993
(9th Cir. 2008) (en banc) (alteration omitted). Even when an
agency is acting within its area of expertise, however, we
“need not defer to the agency when the agency’s decision is
without substantial basis in fact.” Ariz. Cattle Growers’
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 25

Ass’n v. Salazar, 606 F.3d 1160, 1163 (9th Cir. 2010). When
an agency changes a policy based on factual findings that
contradict those on which the prior policy was based, an
agency must provide a “reasoned explanation . . . for
disregarding facts and circumstances that underlay or were
engendered by the prior policy.” FCC v. Fox Television
Stations, Inc., 556 U.S. 502, 515–16 (2009); see also
Organized Vill. of Kake v. U.S. Dep’t of Agric., 795 F.3d
956, 966 (9th Cir. 2015) (en banc).

We review for substantial evidence an agency’s factual
conclusions based on the administrative record. See
Dickinson v. Zurko, 527 U.S. 150, 164 (1999). Where
“evidence is susceptible of more than one rational
interpretation,” we uphold the agency’s finding if a
“reasonable mind might accept [it] as adequate to support a
conclusion.” San Luis & Delta-Mendota Water Auth. v.
Jewell, 747 F.3d 581, 601 (9th Cir. 2014).

A.

We first consider CBD’s argument that the 2014 Finding
arbitrarily found that fluvial arctic grayling population is
increasing. CBD argues that this determination was not
based on the “best scientific and commercial data available,”
as required by 16 U.S.C. § 1533(b)(1)(A). Although FWS
has broad discretion to choose which expert opinions to rely
on when making a listing decision, it cannot ignore available
biological data. See Conner, 848 F.2d at 1454. Here, FWS
acted in an arbitrary and capricious manner by ignoring
available biological data showing that the arctic grayling
population in the Big Hole River was declining.

FWS failed to account for a 2014 report (“DeHaan
study”) by four scientists at the FWS Abernathy Fish
Technology Center, which found that the number of
26 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

effective breeders in the Big Hole River was declining. 11
The DeHaan study arrived at this conclusion by measuring
the effective number of breeders in the Big Hole River
during four different time periods: 1987–88; 1995–96;
2005–06; and 2011–12. The DeHaan study examined
whether there was any change in the number of effective
breeders between each time period and found that the
number of effective breeders decreased in each time period,
although the largest decrease occurred between 1996 and
2005. 12

FWS cited to a portion of the DeHaan study in its 2014
Finding as indicating that a decrease in the number of
effective breeders continued through the mid-2000s but did
not mention that other aspects of the DeHaan study
contradicted the data on which FWS relied (the “Leary
study”). Although FWS is free to choose among experts, it
must acknowledge that it is doing so. See Conner, 848 F.2d
at 1454. FWS clearly stated in the 2014 Finding that the
number of breeding arctic grayling increased in the Big Hole
River, and omitted the DeHaan study’s evidence to the

11
The number of effective breeders in the population is determined
through genetic analysis, which is one way of measuring a species
population. FWS relied on the number of effective breeders to document
population increases.
12
The DeHaan study did find that several other population
indicators, including effective population size, genetic diversity, and the
number of individuals in the population, have remained relatively stable
or increased over time. The DeHaan study also noted that despite a
declining number of breeders “the number of offspring produced may
not have similarly declined.” FWS used the number of effective breeders
as an indication of population abundance, which leads us to focus on this
aspect of the DeHaan study as well.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 27

contrary. We conclude that in ignoring available data FWS
acted in an arbitrary and capricious manner. Id.

FWS’s arguments to the contrary are unavailing. FWS
points out that the 2014 Finding relied on yearly data that
was “more current” than the longitudinal DeHaan study.
Although that could be a reason ultimately to rely on the
Leary study rather than the DeHaan study, the listing
decision should have included “adequate explanation and
support for its determinations.” San Luis, 747 F.3d at 625.
FWS must “provide[] a reasonable explanation for adopting
its approach and disclose[] the limitations of that approach,”
which it fails to do by not discussing the DeHaan study’s
data that contradicts the Leary study. Alaska Oil & Gas
Ass’n v. Pritzker, 840 F.3d 671, 679 (9th Cir. 2016).
Because FWS did not provide a reason to credit the Leary
study over the DeHaan study, “we are precluded from
undertaking meaningful judicial review.” Humane Soc’y of
U.S. v. Locke, 626 F.3d 1040, 1049 (9th Cir. 2010).

Additionally, FWS cannot rely on its briefing in this case
to explain why the 2014 Finding relied on the Leary study
rather than the DeHaan study. The explanation must be
evidenced from the listing decision itself. See Greater
Yellowstone Coal., 665 F.3d at 1027 n.4 (“[A]n agency’s
action must be upheld, if at all, on the basis articulated by
the agency itself, not post-hoc rationalizations.” (internal
quotation marks omitted)). By failing to consider the
DeHaan study’s evidence of decreasing population, FWS
acted in an arbitrary and capricious manner. 13 Because the

13
CBD also argues that FWS acted in an arbitrary and capricious
manner by ignoring state population monitoring data. For the Big Hole
River, FWS provided a valid reason for discounting the fluctuating
population data. In the 2014 Finding, FWS explained that the state
28 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

2014 Finding based its analysis of two of the five listing
factors—the “present or threatened destruction,
modification, or curtailment of [the arctic grayling’s] habitat
and range” and “other natural or manmade factors affecting
[the arctic grayling’s] continued existence”—on the fact that
the arctic grayling’s population was increasing, we reverse
the district court’s grant of summary judgment, with
directions to remand the 2014 Finding to FWS for further
consideration in light of this opinion. 14

B.

Next we consider CBD’s argument that the 2014 Finding
arbitrarily dismissed threats of low stream flows and high
stream temperatures to the arctic grayling. As discussed
below, in rejecting these threats to the arctic grayling in the
Big Hole River FWS acted arbitrarily and capriciously. The
similar findings regarding the Centennial Valley’s lakes and

population monitoring data showed a decline of arctic grayling in the Big
Hole River in 2013 as resulting from unusually high flows that likely
decreased capture efficiency, which is a sufficient explanation that is
neither arbitrary nor capricious. See San Luis, 747 F.3d at 625.
Additionally, while the 2014 Finding did not explicitly address state
population data for the Ruby River, the data does not show the decrease
that CBD claims it does. Although the Ruby River data did show a
decrease in fish in 2013, the size of the area surveyed also decreased.
The number of fish per unit, however, stayed roughly the same. Under
the circumstances, we are not persuaded that FWS acted arbitrarily and
capriciously by failing to address state monitoring data in its 2014
Finding.

14
Given our disposition of this issue, we need not decide CBD’s
additional argument that the Leary study does not provide sufficient
support for FWS’s determination that the arctic grayling population is
increasing, and therefore even if FWS did not improperly disregard the
DeHaan study, its determination that the arctic grayling population is
increasing would be arbitrary and capricious.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 29

tributaries, however, were adequately supported. Any error
in the 2014 Finding regarding the Madison River and its
tributaries was harmless, as FWS expressly did not rely upon
the survival of arctic grayling in the Madison River Valley
in deciding that listing the arctic grayling was not warranted.

1.

In the 2014 Finding, FWS determined that the arctic
grayling’s ability to migrate to coldwater refugia minimizes
the threat it faces from low stream levels and high water
temperatures in the Big Hole River. FWS based this
determination largely on a study that found that the
tributaries of the Big Hole River provide important
coldwater refugia to arctic grayling (“Vatland study”).
Relying on the Vatland study, FWS determined that despite
the existence of water temperatures that exceeded ideal
temperatures for arctic grayling in many areas of the Big
Hole River, arctic grayling could migrate to cold water
refugia over the summer to survive. 15 Additionally, the 2014
Finding reasons that because fish ladders are included in the
CCAA’s conservation projects and the increased
connectivity of the Big Hole River, the arctic grayling can

15
CBD highlights that despite decreases in water temperatures since
implementing the CCAA, the water temperature still frequently exceeds
70 degrees Fahrenheit. As mentioned above, 70 degrees Fahrenheit is
the temperature above which arctic grayling experience “physiological
stress.” Montana contends that harm does not result from temperatures
of 70 degrees Fahrenheit and that stream temperatures in the Big Hole
River tributaries did not exceed 70 degrees in 2013. These arguments,
however, are directly contradicted by data in FWS’s 2014 Finding.
30 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

access cold water tributaries that they could not access in
2010. 16

CBD disputes the 2014 Finding’s conclusion that the
arctic grayling seek refuge in coldwater tributaries of the Big
Hole River when water temperatures rise. CBD also
disputes that the Vatland study shows that arctic grayling
migrate, as the study found “[l]imited movement” among
arctic grayling during the summer. Additionally, CBD
argues that evidence does not suggest that tributaries actually
provide cold water refugia, as the temperatures in these
tributaries frequently exceed 70 degrees. CBD’s arguments
are persuasive.

FWS’s reliance on the ability of the arctic grayling to
migrate to cold water refugia was arbitrary and capricious.
The sole evidence of arctic grayling migrating to cold water
refugia in the Big Hole River tributaries is the Vatland study.
Notably, in 2010, FWS determined that despite the Vatland
study’s findings that arctic grayling have the ability to
migrate to cold water refugia in tributaries, water
temperatures were sufficiently high to warrant listing the
arctic grayling. Water temperatures remained high enough
to cause physiological stress in 2014, but the 2014 Finding
stated that this could be overcome by the arctic grayling’s
ability to migrate to cold water refugia via CCAA fish
ladders, without providing any additional evidence or
scientific studies demonstrating that this would likely occur.

16
In the 2010 Finding, FWS also determined that CCAA
conservation measures would reduce but not eliminate threats of
dewatering. Since CCAA conservation measures took effect, the record
reflects that minimum flow targets have been achieved 78 percent of the
time, up from 50 percent of the time pre-CCAA. Although this is an
improvement, CBD notes that FWS previously stated that the flow target
represented minimum values to promote recovery of the arctic grayling.
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 31

Because the 2010 Finding indicated that listing the arctic
grayling was warranted irrespective of the Vatland study and
recognized the ability of arctic grayling to migrate to
tributaries, the 2014 Finding was required to provide a
reasoned explanation for FWS’s change in position. See
Organized Vill. of Kake, 795 F.3d at 966 (stating that an
“[u]nexplained inconsistency” between two agency actions
can be grounds for holding that agency action is arbitrary and
capricious (quoting Brand X, 545 U.S. at 981)).

Nor do lower water temperatures or the CCAA
conservation measures save the agency’s flawed 2014
Finding. As discussed above, temperatures are still higher
than the scientific benchmarks cited by FWS as tolerable
water temperatures for arctic grayling. Cf. Greater
Yellowstone Coal., 665 F.3d at 1028 (“Having determined
what is necessary, the [FWS] cannot reasonably rely on
something less to be enough.” (internal quotation marks
omitted)). Even the tributaries of the Big Hole River that
supposedly serve as cold water refugia are above the desired
temperature according to the scientific studies on which
FWS relies. FWS disregarded this scientific evidence, and
instead based its conclusion on a study finding “limited
movements” of arctic grayling in the Big Hole River during
summer months. Given that water temperatures—even in
tributaries—still exceed temperatures where arctic grayling
can live and breed, FWS did not sufficiently “articulate[] a
rational connection between the facts found and the choices
made.” Id. at 1023 (internal quotation marks omitted).
Similarly, FWS’s reliance on the CCAA’s fish ladders as
evidencing a change in the status quo without any studies
finding that these measures will aid migration is arbitrary
and capricious—even if the ladders aid the arctic grayling in
migrating to tributaries, as this would be of little value if the
water in the tributaries is still too warm. Id. (stating that
32 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

agency rulings are arbitrary and capricious if the agency
offers “an explanation for its decision that runs counter to
the evidence before the agency” (quoting State Farm,
463 U.S. at 43)).

Although there have been improvements in stream flow
and water temperature since 2010, the water temperatures
are still above those that are ideal for the arctic grayling both
in the main stem of the Big Hole River and its tributaries. In
sum, the 2014 Finding that thermal refugia in the Big Hole
River would aid survival of the arctic grayling was arbitrary
and capricious.

2.

The 2014 Finding relied solely on the existence of
thermal refugia to dismiss the threat of increased water
temperatures in the Centennial Valley lakes and tributaries.
CBD argues that because the evidence does not adequately
support the existence of thermal refugia, FWS’s decision as
to the Centennial Valley lakes and tributaries was arbitrary
and capricious.

FWS supported its determination that cold water refugia
exist in the Centennial Valley primarily by relying on a study
that observed two arctic graylings in a tributary in 1994
(“Mogen study”), and which stated that the tributary
“possibly provid[ed] thermal refugia.” The Mogen study
was discussed in the 2010 Finding, but FWS still concluded
that high water temperatures were a threat. Indeed, the 2010
Finding cited to the Mogen study’s observation of two fish
seeking refuge to support a finding of high water
temperatures in the Centennial Valley lakes and tributaries.

FWS did not act in an arbitrary and capricious manner
by reversing its 2010 Finding that cold water thermal refugia
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 33

were insufficient to mitigate high water temperatures in the
Centennial Valley. Although FWS relies primarily on the
same information in 2014 as it did in 2010, the 2014 Finding
also relied on an email by a scientist named Matt Jaeger
(“Jaeger email”). The Jaeger email stated that there was
evidence that cold water refugia existed in the Centennial
Valley, but noted uncertainty in terms of whether this would
fully mitigate warm water temperatures. Given increasing
population of arctic grayling in the Centennial Valley,
however, the Jaeger email concluded that increasing
temperatures likely are not a threat. The Jaeger email, and
the corresponding increase in population in the Centennial
Valley, provides a sufficient “reasoned explanation” for
FWS’s change in position. See Organized. Vill. of Kake,
795 F.3d at 968.

3.

CBD also challenges the findings of cold water refugia
in the Madison River. The 2014 Finding cites no evidence
to support a finding that cold water refugia exist in the
tributaries of this river. Rather, FWS notes in the 2014
Finding that there are high water temperatures in the
Madison River and that the arctic grayling population has
been decreasing. Without any evidentiary support in the
2014 Finding record, FWS’s finding that cold water refugia
exist is improper. 17 See State Farm, 463 U.S. at 43.

This error, however, was harmless. In the 2014 Finding,
which recognizes that the arctic grayling in the Madison

17
FWS provides some support in its brief for the existence of cold
water refugia, but this is an ex post rationalization, which is informative
but of no force in evaluating FWS’s listing decision. Greater
Yellowstone Coal., 665 F.3d at 1027 n.4.
34 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

River are endangered by high water temperatures and low
population, FWS concluded that even if the arctic grayling
were no longer able to survive in the Madison River and its
tributaries, the population in the upper Missouri River
Valley as a whole would not be compromised. Under these
circumstances, any error in finding that refugia exist is
harmless, as FWS did not rest its ultimate 2014 Finding on
the continued existence of arctic grayling in the Madison
River. Cf. Organized. Vill. of Kake, 795 F.3d at 969 (stating
that the burden is on “the opponent of the action to
demonstrate [that] an error is prejudicial”).

***

In sum, the 2014 Finding’s reliance on cold water refugia
in the Big Hole River was arbitrary and capricious. The
district court erred in granting summary judgment in favor
of FWS on this issue. We affirm, however, the district
court’s ruling on the cold water refugia issue in all other
respects.

C.

We turn to CBD’s contention that FWS disregarded the
additive effects of climate change in considering the effects
of low stream flows and high water temperatures.
Specifically, CBD argues that FWS’s 2014 assessment of the
cumulative impacts of climate change arbitrarily relied on
uncertainty to avoid making determinations about the threat
of climate change. We have held that it is “not enough for
[FWS] to simply invoke ‘scientific uncertainty’ to justify its
action.” Greater Yellowstone Coal., 665 F.3d at 1028
(discussing uncertainty caused by the effects of climate
change). Rather, FWS must explain why uncertainty
justifies its conclusion, “[o]therwise, we might as well be
deferring to a coin flip.” Id. In its 2014 Finding, FWS states
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 35

that “[u]ncertainty about how different temperature and
precipitation scenarios could affect water availability make
projecting possible synergistic effects of climate change on
the Arctic grayling too speculative at this time.” With this
statement, FWS expressly disclaimed making any projection
as to the synergistic effects of climate change, simply
because of the uncertainty.

Greater Yellowstone Coalition counsels that this
approach is unacceptable. See 665 F.3d at 1028. There, we
held that because FWS had data showing that the population
of whitebark pine was declining due to climate change, it
could not simply state that it was uncertain what impact this
would have on grizzly bears. Id. Rather, FWS had to explain
why this uncertainty favored delisting the grizzly bear rather
than, for example, undertaking further studies to minimize
the uncertainty. Id.

Similarly, in the 2014 Finding, FWS did not explain how
uncertainty about water availability justifies not listing the
arctic grayling as opposed to taking another course of action.
Pursuing another course of action may have been
particularly prudent given the ESA’s policy of
“institutionalized caution,” Ariz. Cattle Growers, 606 F.3d
at 1167, especially since the 2014 Finding expressly cites to
evidence that climate change will increase water
temperatures and threats of low water flow. According to
the 2014 Finding, “water temperatures will likely increase
with climate change in the future,” 79 Fed. Reg. at 49,405,
and dewatering threats will be exacerbated by “[i]ncreases
in temperature and changes in precipitation [that] are likely
to affect the availability of water in the West,” id. at 49,419.
By failing to explain why the uncertainty of climate change
favors not listing the arctic grayling when the 2014 Finding
acknowledges the warming of water temperatures and
36 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

decreasing water flow because of global warming, FWS
acted in an arbitrary and capricious manner. See Greater
Yellowstone Coal., 665 F.3d at 1028; see also State Farm,
463 U.S. at 43.

D.

CBD also argues that FWS acted arbitrarily in dismissing
threats of small population sizes, especially since the 2010
Finding was based in part on the finding that four of the five
native arctic grayling populations are at risk because of their
low population numbers. Specifically, CBD argues that
FWS (1) did not provide a basis for determining the impact
of low population numbers on long-term genetic viability
and (2) concluded irrationally that stochastic events would
not threaten the arctic grayling despite small populations.
We address each argument in turn.

1.

When considering whether to list a species, FWS must
determine whether the species “is likely to become an
endangered species within the foreseeable future.”
16 U.S.C. § 1532(20). Notably, FWS previously applied
§ 1532(20) to encompass long-term genetic effects. In the
2010 Finding, FWS defined “foreseeable future” as thirty
years on the basis of a population viability analysis. 75 Fed.
Reg. at 54,725. Additionally, in 2010, FWS found that while
population levels were large enough for inbreeding not to be
an immediate concern, they were still “below the level
presumed to provide the genetic variation necessary to
conserve long-term adaptive potential.” Id. at 54,741. In the
2014 Finding, FWS found that genetic diversity does not
pose a short-term threat to the arctic grayling. 79 Fed. Reg.
at 49,418. It then discussed scientific literature debating the
effective population size adequate to conserve genetic
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 37

diversity over the long term, and concluded that generally
genetic diversity does not drive species to extinction and that
other processes are more important. Id. at 49,418–19. The
2014 Finding acknowledged that loss of genetic diversity is
a threat, but concluded that there are a sufficient number of
breeding adults to minimize this threat. Id.

FWS’s determination that the arctic grayling’s small
population size does not pose a risk to genetic viability of the
arctic grayling is not arbitrary or capricious. CBD insists
that FWS did not consider long-term genetic viability, but
the record does not support this argument. FWS did consider
long-term genetic viability and simply concluded that, given
increased population and “[u]pdated genetic information
that was not available in 2010,” any concern about long-term
genetic viability did not merit listing the arctic grayling.
79 Fed. Reg. at 49,420. FWS provided a reasoned
explanation for why it did not view lack of genetic diversity
as a threat. And that determination was not arbitrary or
capricious; difference of opinion does not warrant a contrary
conclusion. See, e.g., Lands Council, 537 F.3d at 988. We
affirm the district court’s ruling on this issue.

2.

The 2014 Finding did, however, arbitrarily rely on the
Ruby River population to provide redundancy of fluvial
arctic grayling beyond the Big Hole River and to minimize
the risk from random environmental events. In its 2010
Finding, FWS recognized the importance of having multiple
populations as genetic reservoirs in case of unexpected
“stochastic” events or environmental catastrophes that may
wipe out one or more populations of a species. FWS
concluded that “the lack of additional fluvial populations
[beyond the Big Hole River] represents a current threat to
the upper Missouri River [arctic grayling].” 75 Fed. Reg. at
38 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

54,741. In its 2014 Finding, FWS found that this was no
longer a concern in part because the increase in the number
of breeding individuals in the Ruby River over the last three
years provided “a viable replicate of the fluvial ecotype.”
79 Fed. Reg. at 49,419.

The 2014 Finding’s reliance on the Ruby River’s
viability as a genetic reservoir contradicts FWS’s criteria for
judging viability, which requires “at least 10 years” of
monitoring data to confirm that a population is viable.
Additionally, the 2010 Finding noted that at least five to ten
more years of monitoring would be needed at Ruby River to
determine if it is a viable population. The 2014 Finding
relies on the Leary study, which shows population increases
in Ruby River, but does not provide a reasoned explanation
for disregarding FWS’s prior criteria for judging viability or
the statement in the 2010 Finding about needing five to ten
more years of monitoring. See, e.g., Organized Vill. of Kake,
795 F.3d at 968–69; see also Greater Yellowstone Coal.,
665 F.3d at 1028.

Instead, the 2014 Finding describes the Ruby River
population as viable despite only five years of monitoring
data. This is less than the viability criteria recommends. It
is also only four more years of data than that used to support
the 2010 Finding, which indicated at least five to ten more
years of data would be needed. This lack of data is a crucial
omission as the Ruby River population is one of only two
fluvial populations. The 2014 Finding’s determination that
the Ruby River population was viable and could provide
 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE 39

redundancy was arbitrary and capricious, and we therefore
reverse the district court’s ruling on this issue. 18

V.

For the foregoing reasons, we hold that the 2014
Finding’s decision that listing the arctic grayling was “not
warranted” was arbitrary and capricious because it
(1) ignored the DeHaan study’s evidence that shows
decreasing numbers of breeders and instead heavily relied on
a contrary finding showing increasing population; (2) did not
provide a reasoned explanation for relying on the existence
of cold water refugia in the Big Hole River; (3) failed to
consider the synergistic effects of climate change solely
because of “uncertainty”; and (4) concluded that the Ruby
River population is viable based on data collected over a
shorter period than that underlying the 2010 Finding and
FWS’s own established criteria for viability. We therefore
reverse the district court’s grant of summary judgment with

18
CBD raises an additional argument that FWS acted arbitrarily in
basing its 2014 Finding in part on the geographic separation between
populations without explaining why it changed its position from 2010.
See, e.g., Organized Vill. Of Kake, 795 F.3d at 968–69. But, as FWS
points out, the 2010 and 2014 Findings each discussed how separation
of populations reduces the risk of multiple populations being negatively
impacted by a single environmental catastrophe. In 2010, however, FWS
concluded that these populations were at risk from other environmental
factors, which in turn increased the risk of harm of a stochastic event. In
2014, FWS determined that these other factors no longer counseled in
favor of listing the arctic grayling. Therefore, its conclusion that
separation between populations would help prevent a stochastic event
was not arbitrary or capricious.
40 CENTER FOR BIOLOGICAL DIVERSITY V. ZINKE

directions to remand to FWS to reassess the 2014 Finding in
light of this opinion.

AFFIRMED in part; REVERSED in part; and
REMANDED.

The parties shall bear their own costs on appeal.