Friends of Animals v. Usfws

FOR PUBLICATION

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

FRIENDS OF ANIMALS, No. 21-35062
Plaintiff-Appellant,
D.C. No.
v. 6:17-cv-00860-
AA
UNITED STATES FISH AND WILDLIFE
SERVICE, an agency of the United
States; MARTHA WILLIAMS, in her OPINION
official capacity as the Director of
the United States Fish and Wildlife
Service,
Defendants-Appellees.

Appeal from the United States District Court
for the District of Oregon
Ann L. Aiken, District Judge, Presiding

Argued and Submitted November 17, 2021
Pasadena, California

Filed March 4, 2022

Before: Johnnie B. Rawlinson and Kenneth K. Lee, Circuit
Judges, and Matthew F. Kennelly, * District Judge.

Opinion by Judge Lee

*
The Honorable Matthew F. Kennelly, United States District Judge
for the Northern District of Illinois, sitting by designation.
2 FRIENDS OF ANIMALS V. USFWS

SUMMARY **

Environmental Law

The panel affirmed the district court’s summary
judgment in favor of the U.S. Fish and Wildlife Service in
an action brought by an environmental group challenging the
Service’s “barred owl removal experiment,” which was
designed to protect the northern spotted owl, a threatened
species under the Endangered Species Act (“ESA”).

Barred owls have encroached on the spotted owl’s
habitat. The barred owl removal experiment is a proposed
lethal removal of barred owls from certain areas to measure
their environmental and demographic effect on northern
spotted owls. To complete the experiment, the Service
issued permits and entered into Safe Harbor Agreements
with four non-federal landowners within the Oregon coast
ranges study area.

Plaintiff alleged that the ESA prohibited the government
from taking action that may incidentally harm spotted owls
or their habitat unless it provided a “net conservation
benefit,” and the barred owl removal experiment will not
yield a net conservation benefit because it does not likely
lead to the recovery of the spotted owl population or its
habitat.

Affirming the district court, the panel held that the barred
owl removal experiment will produce a “net conservation
benefit” under the ESA’s implementing regulations because
**
This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.
 FRIENDS OF ANIMALS V. USFWS 3

it allowed the agency to obtain critical information to craft a
policy to protect threatened or endangered species. The
panel also held that the Service reasonably described
baseline conditions/“resident” owl survey data. In addition,
the panel held that the Service adequately analyzed the small
portion of critical habitat affected by the Oregon permit/Safe
Harbor Agreements.

The panel held that the Service complied with the
National Environmental Policy Act (“NEPA”), and did not
have to conduct a supplemental environmental impact
statement under NEPA because it had adequately
contemplated this experiment in its earlier analysis. In
addition, the permits and the experiment need not be
analyzed in a single environmental impact statement because
they were not “connected actions.”

COUNSEL

Jennifer Best (argued) and Michael Ray Harris, Friends of
Animals, Centennial, Colorado, for Plaintiff-Appellant.

Andrew M. Bernie (argued), Andrew C. Mergen, Ellen J.
Durkee, and Coby Howell, Attorneys; Jean E. Williams,
Acting Assistant Attorney General; Environment and
Natural Resources Division, United States Department of
Justice, Washington, D.C.; Lydia Grimm, Office of the
Regional Solicitor, Department of the Interior, Washington,
D.C.; for Defendants-Appellees.
4 FRIENDS OF ANIMALS V. USFWS

OPINION
LEE, Circuit Judge:

This case is a tale of two owls. For the northern spotted
owl, it has been the worst of times: It remains a threatened
species, and its population continues to dwindle in the
Pacific Northwest and Northern California. But it has been
the best of times for the barred owl: Its abundant population
burgeoning, the barred owl has expanded westward and
encroached on the spotted owl’s habitat. And barred owls
have even been spotted attacking their brethren bird.

Trying to usher in a spring of hope for the northern
spotted owls, the United States Fish and Wildlife Service
(FWS) introduced a “barred owl removal experiment.” It
proposed lethally removing barred owls from certain areas
to measure their environmental and demographic effect on
the northern spotted owls. Friends of Animals (“Friends”),
an environmental group, sued the FWS, claiming that this
experiment violates the Endangered Species Act (ESA) and
the National Environmental Policy Act (NEPA). Friends
mainly argues that the ESA prohibits the government from
taking action that may incidentally harm spotted owls or
their habitat unless it provided a “net conservation benefit.”
On top of harming some spotted owls as scientists enter their
habitat, the experiment will not yield a “net conservation
benefit” because it does not directly lead to the recovery of
the spotted owl population or its habitat, according to
Friends.

We affirm the district court’s grant of summary
judgment for FWS. We hold that this experiment will
produce a “net conservation benefit” under the plain
language of the ESA’s implementing regulations because it
allows the agency to obtain critical information to craft a
policy to protect threatened or endangered species. We also
 FRIENDS OF ANIMALS V. USFWS 5

hold that FWS did not have to conduct a supplemental
environmental impact statement under NEPA because it had
adequately contemplated this experiment in its earlier
analysis.

BACKGROUND

I. Background

a. The barred owl threatens the northern spotted
owl.

The northern spotted owl is one of three subspecies of
spotted owls (Northern, California, and Mexican). The
species generally lives in mature and old-growth forests
around southwest British Columbia through the Cascade
Mountains and coastal ranges in Washington, Oregon, and
Northern California. Since 1990, FWS has listed the
northern spotted owl as a threatened species under the ESA.

The barred owl, in contrast, is an abundant species native
to eastern North America. Over the past century, the barred
owl has moved west, expanding its range across the
continent to the west coast. And in the process, the barred
owl has encroached on the northern spotted owl’s range,
becoming an invasive species. The barred owl’s and spotted
owl’s ranges now completely overlap, and the two species’
food needs and habitats share significant similarities as well.

In its most recent 2011 Recovery Plan for the northern
spotted owl, FWS stated that “[s]trong evidence indicates
that barred owls negatively affect spotted owls and their
populations” by displacing spotted owls from their habitat
and reducing spotted owl survival and reproduction. Barred
owls have also sometimes attacked the northern spotted owl.
6 FRIENDS OF ANIMALS V. USFWS

b. FWS designs a barred owl removal experiment.

FWS identified “barred owl management” as one of the
four basic steps to protect the northern spotted owl. The
2011 Recovery Plan noted that while evidence suggests that
“barred owls compete with spotted owls for nesting sites,
roosting sites, and food, and possibly predate spotted owls,”
there are still “substantial information gaps.” To fill these
informational gaps, the Recovery Plan called for FWS to
“[d]esign and implement large-scale control experiments to
assess the effects of barred owl removal on spotted owl site
occupancy, reproduction, and survival.”

FWS then issued a Record of Decision in 2013
authorizing the lethal barred owl removal experiment. FWS
expected the experiment to provide “needed information”
such as “the effects of barred owls on spotted owl vital rates
of occupancy, survival, reproduction, and population trend”;
the “feasibility of removing barred owls from an area”; the
“level of effort required to maintain reduced barred owl
population levels”; the “cost of barred owl removal”; and the
overall utility of barred owl removal for “management of
barred owls.”

FWS’ experiment designated four study areas spread
across the northern spotted owl’s range. Relevant here is the
Oregon Coast Ranges study area, which covers just over
500,000 acres. Within the study area, FWS designated
“treatment areas” from which about 3,600 barred owls
would be removed. FWS expected removal to be completed
in four years, though the overall experiment, including
conducting surveys and gathering data, could last up to ten
years.

FWS also issued an Environmental Impact Statement
(EIS) in 2013 for the experiment, as required by NEPA.
 FRIENDS OF ANIMALS V. USFWS 7

NEPA is “a procedural statute intended to ensure Federal
agencies consider the environmental impacts of their actions
in the decision-making process.” 40 C.F.R. § 1500.1. 1 For
major actions that will significantly affect the environment,
NEPA requires federal agencies to inform the public of their
analysis in an EIS. See 42 U.S.C. § 4332(C); 40 C.F.R.
§ 1502.4(a). Additionally, an agency may need to issue a
supplemental environmental impact statement
(Supplemental EIS) to augment a prior EIS if the agency
makes “substantial changes” to the action, or “significant
new circumstances or information” arises. 40 C.F.R.
§1502.9(d)(1). 2

The EIS for the barred owl removal experiment
concluded that it would have a negligible effect on the barred
owl population, given the species’ abundance. The EIS also
acknowledged that the northern spotted owl could
experience “minor and short-term negative effects” because
of the intrusions by scientists into its habitat. But the main
anticipated effect would be a potential “positive change in

1
In the time between the filing of the lawsuit and our decision today,
NEPA’s regulations were updated. See Update to the Regulations
Implementing the Procedural Provisions of the National Environmental
Policy Act, 85 Fed. Reg. 43,304 (July 16, 2020) (to be codified at 40
C.F.R. §§ 1500–1508, 1515–1518). We cite the current version of the
regulations throughout our decision. The old and new versions of the
cited regulations are substantively the same.
2
If the proposed action on its face does not appear to have a
significant impact, an agency may prepare a less-intensive
environmental assessment (EA) to determine whether the action’s effects
would be significant. See 40 C.F.R. § 1501.5(a)–(c). If the agency
concludes in its EA that the proposed action will not significantly affect
the human environment, the agency may issue a “finding of no
significant impact” (FONSI) rather than produce a full-scale EIS. Id.
§ 1501.6(a).
8 FRIENDS OF ANIMALS V. USFWS

spotted owl demographic performance” because of
decreased competition from the barred owl. More broadly,
the major benefit of the experiment would be obtaining data
necessary to craft long-term recovery strategies for the
northern spotted owl.

c. Endangered Species Act bars actions that harm
threatened or endangered species and their
habitats unless they provide a “net conservation
benefit.”

Before FWS could begin this experiment, it had to
comply with the ESA’s many requirements. Section 7(a)(2)
of the ESA requires each federal agency to consult an expert
wildlife agency to ensure “that any action authorized,
funded, or carried out . . . is not likely to jeopardize the
continued existence of any endangered species or threatened
species or result in the destruction or adverse modification”
of the critical habitat. 16 U.S.C. § 1536(a)(2).

Section 7 and its implementing regulations detail the
consultation process for determining the biological impacts
of a proposed action, leading to a Biological Opinion
(BiOp). Id. § 1536(b)(3)(A); 50 C.F.R. § 402.14(h). In the
Biological Opinion, the expert agency (here, FWS) renders
its opinion whether the proposed action will likely
“jeopardize the continued existence of a listed species or
result in the destruction or adverse modification of critical
habitat.” 50 C.F.R. § 402.14(g)(4).

Along with requiring cross-agency consultation, the
ESA broadly prohibits the “take” of any endangered or
threatened species in the United States. 16 U.S.C. § 1538.
“Take” means “to harass, harm, pursue, hunt, shoot, wound,
kill, trap, capture, or collect, or to attempt to engage in any
such conduct.” Id. § 1532(19). The ESA’s implementing
 FRIENDS OF ANIMALS V. USFWS 9

regulations also define “harm” in the definition of “take” to
include “significant habitat modification or degradation
where it actually kills or injures wildlife by significantly
impairing essential behavioral patterns, including breeding,
feeding or sheltering.” 50 C.F.R. § 17.3(c)(3).

Despite this general prohibition on take, the ESA
provides a few exceptions. Relevant here, FWS may issue
“Enhancement of Survival Permits” that authorize take “for
scientific purposes or to enhance the propagation or survival
of the affected species . . . .” 16 U.S.C. § 1539(a)(1)(A).
FWS may issue these permits and enter into corresponding
Safe Harbor Agreements with non-federal landowners
whose lands that the agency seeks to use for conservation
efforts. See 50 C.F.R. § 17.32(c)(1). The Safe Harbor
Agreements set the terms of the permits and try to
incentivize non-federal property owners to voluntarily
undertake conservation activities on their property. Id. In
exchange, FWS assures property owners that it will “not
require additional or different management activities” and
will allow the property to return to its previous condition at
the agreement’s conclusion. Id. § 17.32(c)(5).

To issue a permit, FWS must find that:

“The implementation of the terms of the Safe
Harbor Agreement is reasonably expected to
provide a net conservation benefit to the
affected listed species by contributing to the
recovery of [the] listed species included in
the permit, and the Safe Harbor Agreement
otherwise complies with the Safe Harbor
policy available from the Service.”

Id. § 17.32(c)(2) (emphasis added). In 1999, FWS issued its
Safe Harbor Policy that fleshed out the requirements for
10 FRIENDS OF ANIMALS V. USFWS

issuing permits. See Announcement of Final Safe Harbor
Policy, 64 Fed. Reg. 32,717 (June 17, 1999) (the Policy).

d. FWS issues Enhancement of Survival Permits
and enters into Safe Harbor Agreements with
four non-federal landowners.

The Oregon Coast Ranges study area is a checkerboard
of federal, state, and private land. To create contiguous areas
for the experiment, FWS sought cooperation from non-
federal landowners to gain access to their lands. Though not
strictly necessary to complete the experiment, such access
would help FWS complete the experiment in the most
efficient and complete manner. Failure to gain such access
could reduce the ability to detect changes in the spotted owl
population caused by barred owl removal.

To that end, FWS issued permits and entered into Safe
Harbor Agreements with four non-federal landowners
within the Oregon Coast Ranges study area: Roseburg
Resource Company (Roseburg), Weyerhaeuser Company
(Weyerhaeuser), Oxbow Timber I, LLC (Oxbow), and the
Oregon Department of Forestry (Oregon). 3

Each permittee agreed to allow FWS to access their
property and roads to remove barred owls and agreed to
conduct or support spotted owl surveys on their lands. In
exchange, the permittees may keep harvesting timber on
their property in areas where no northern spotted owls
resided when the parties entered into the Safe Harbor
Agreements (“non-baseline sites”) without incurring
3
After FWS issued the permits, Roseburg acquired Oxbow. Because
of this acquisition, there are now only three permits for the Oregon Coast
Ranges study area that are being challenged (Roseburg-Oxbow,
Weyerhaeuser, and Oregon).
 FRIENDS OF ANIMALS V. USFWS 11

liability for incidentally taking any spotted owls that later
reoccupy those locations. The permittees, however, receive
no liability protection for any incidental take in areas where
the owls already resided (“baseline sites”).

FWS used survey data to designate any site in which a
“resident” northern spotted owl had been detected in the
previous three to five years as a baseline site. Thus, the
permits authorized incidental take only in non-baseline sites
(i.e., areas where no “resident” northern spotted owls have
been observed in the past three to five years). And even in
those non-baseline sites, the permits restricted the land
during nesting and rearing season.

As required by the ESA’s implementing regulations,
FWS determined that each permit was “reasonably expected
to provide a net conservation benefit” to the northern spotted
owl. See 50 C.F.R. § 17.32(c)(2). Although FWS
acknowledged that there was potential for “take of spotted
owls on the temporarily reoccupied” non-baseline sites,
FWS concluded such take would be “more than offset by the
value of the information gained from the Experiment and
potential contribution to long-term barred owl management
strategy.”

e. FWS prepares Biological Opinions and EAs to
comply with the ESA and NEPA, respectively.

As required by the ESA, FWS issued a series of
Biological Opinions simultaneously with the permits and
Safe Harbor Agreements. The Biological Opinions
determined that the Roseburg, Oxbow, and Weyerhaeuser
permits would neither jeopardize the northern spotted owl
nor adversely modify its critical habitat. Given each
permit’s small effect on spotted owl habitat, FWS concluded
that the potential harm caused by the experiment would
12 FRIENDS OF ANIMALS V. USFWS

likely be offset by the information gained if the experiment
succeeded.

FWS’ analysis differed slightly for the Oregon permit
because of its potential effect on a small portion of critical
habitat. In total, the Oregon permit authorized up to 3,345
acres of critical habitat loss. This represented less than 0.04
percent of total range-wide spotted owl critical habitat. FWS
still believed that such habitat destruction was justified
because it would aid the barred owl removal experiment.
FWS thus concluded that the Oregon permit was unlikely to
jeopardize the northern spotted owl or its critical habitat.

And as required by NEPA, FWS also prepared an
Environmental Assessment for each permit. Each
Environmental Assessment made a Finding of No
Significant Impact, concluding that the permit is “not likely
to have a significant impact on the spotted owl.” The permits
only authorized incidental take in non-baseline sites that do
not currently have spotted owls and are unlikely to be
recolonized without barred owl removal. Thus, northern
spotted owls would be taken only if the experiment
facilitated spotted owl recolonization in previously
unoccupied areas.

f. Friends sues FWS, alleging violations of the ESA
and NEPA.

In June 2017, Friends sued FWS, challenging the
issuance of the Enhancement of Survival permits and Safe
Harbor Agreements in the Oregon Coast Ranges and
Klamath study areas. The district court held that Friends
lacked standing to bring either claim. See Friends of
Animals v. Sheehan, 2018 WL 6531676, *5 (D. Or. Dec. 11,
2018). This court reversed in part, holding that Friends had
standing to challenge permits issued in the Oregon Coast
 FRIENDS OF ANIMALS V. USFWS 13

Ranges, but not the Klamath, study area. Friends of Animals
v. U.S. Fish & Wildlife Serv., 789 F. App’x 599, 600–01 (9th
Cir. 2020).

On remand, the district court considered the merits and
granted summary judgment for FWS. Friends of Animals v.
Sheehan, 2021 WL 150011, *1 (D. Or. Jan. 15, 2021).
Friends alleged FWS violated the ESA by “(1) issuing a
permit that fails to achieve a ‘net conservation benefit,’ (2)
failing to use the best biological and habitat information in
forming baseline conditions, and (3) failing to analyze the
SHA’s effect on critical habitat.” Id. at *3. Additionally,
Friends claimed FWS violated NEPA because it (1) failed to
conduct a Supplemental EIS after issuing the permits, and
(2) failed to discuss the experiment and permits in a single
EIS as required for “connected actions.” Id. at *13.

For the first ESA claim, the district court determined that
the regulation specifying that a Safe Harbor Agreement
should be “reasonably expected to provide a net
conservation benefit,” see 50 C.F.R. § 17.32(c)(2), was
ambiguous. Id. at *3. Because the district court found that
the Safe Harbor Policy referenced by § 17.32 did not carry
the force of law, it held that the Safe Harbor Policy—which
elaborates on the meaning of “net conservation benefit”—
could not be used to resolve the ambiguity. Id. at *5.
Because § 17.32 was ambiguous, the district court applied
Auer 4 deference to FWS’ interpretation that “information”
may constitute a “net conservation benefit” and found the
agency’s interpretation reasonable. Id. at *3, *6. The district
court also noted that the Policy supported FWS’
interpretation. See id. at *6–7.

4
See Auer v. Robbins, 519 U.S. 452 (1997).
14 FRIENDS OF ANIMALS V. USFWS

The district court next held that FWS reasonably defined
baseline conditions as suggested in the Safe Harbor Policy
because the agency used the best techniques and information
available. Id. at *8. The district court held that FWS’
reliance on “resident” owl populations, rather than “floater”
(i.e., young and displaced) owls, tracked the Safe Harbor
Policy. Id. at *10. The district court also rejected Friends’
argument that FWS’ survey data could not establish that
non-baseline sites were “abandoned” because neither the
Safe Harbor Policy nor other agency guidance requires
baseline conditions to be designated based on
“abandonment.” Id. at *11.

Lastly, the district court found that FWS adequately
considered impacts to critical habitat affected by the Oregon
permit in the issued Biological Opinions. Id. at *12. The
court rejected Friends’ argument that the Biological
Opinions insufficiently analyzed important subsets of
critical habitat because the Biological Opinions did
“consider and assess[] the foraging, transience, and
colonization value of the affected critical habitat . . . .” Id.

The district court then turned to Friends’ NEPA claims.
The district court held that FWS did not have to conduct a
Supplemental EIS because the “2013 EIS accounted for the
possibility that nonfederal lands could be included in the
experiment” and the authorization of incidental take in non-
baseline sites did “not constitute a substantial change
relevant to environmental concerns.” Id. at *13. The district
court also held that FWS did not have to analyze the
experiment and permits in a single EIS. Id. at *14. Because
“each action could exist without the other,” the experiment
and the permits were not “connected actions.” Id.

On appeal, Friends presses the same ESA and NEPA
claims.
 FRIENDS OF ANIMALS V. USFWS 15

STANDARD OF REVIEW

We review a district court’s grant of summary judgment
de novo. Native Ecosystems Council v. Marten, 883 F.3d
783, 789 (9th Cir. 2018). “We review an agency’s
compliance with the ESA . . . and NEPA under the ‘arbitrary
and capricious’ standard of the APA.” Id. at 788; see also
5 U.S.C. § 706(2)(A). Under this standard, we “must
determine whether the agency considered the relevant
factors and articulated a rational connection between the
facts found and the choices made.” Ranchers Cattlemen
Action Legal Fund v. U.S. Dep’t of Agric., 499 F.3d 1108,
1115 (9th Cir. 2007) (internal quotation marks omitted).
Arbitrary and capricious review is “highly deferential” and
presumes that the agency action is valid if “a reasonable
basis exists” for the agency’s decision. Id. (citation omitted).

ANALYSIS

I. FWS complied with the ESA.

Friends renews its same ESA arguments rejected by the
district court. We reject them and conclude that FWS
complied with the ESA.

a. The “informational benefit” from the experiment
may constitute a “net conservation benefit” under
ESA regulations.

As noted above, the ESA generally bars anyone—
including the federal government—from “taking”
endangered or threatened species or making significant
modifications to their habitat that kill or injure wildlife. 16
U.S.C. § 1538 (prohibiting the “take” of threatened or
endangered species); 50 C.F.R. § 17.3(c)(3) (defining
16 FRIENDS OF ANIMALS V. USFWS

“harm” under the “take” definition to include significant
modification of habitat).

The ESA, however, carves out a few exceptions.
Relevant here, FWS can issue a permit allowing someone to
“take” an endangered or threatened species if it is “for
scientific purposes or to enhance the propagation or survival
of the affected species . . . .” 16 U.S.C. § 1539(a)(1)(A).
And ESA regulations permit the agency to enter into Safe
Harbor Agreements with non-federal landowners whose
lands the agency wants to use for conservation efforts if the
proposed actions are “reasonably expected to provide a net
conservation benefit to the affected listed species” and
“otherwise compl[y] with the Safe Harbor policy.” See 50
C.F.R. § 17.32(c)(1); (c)(2)(ii) (emphasis added).

FWS designed its barred owl removal experiment
relying on its authority to issue these permits and to enter
into Safe Harbor Agreements. But Friends argues FWS
cannot authorize this experiment because it does not provide
a “net conservation benefit.” Their argument proceeds as
follows: Because the ESA regulation does not define “net
conservation benefit,” that term is ambiguous. The Safe
Harbor Policy, however, defines “net conservation benefit.”
64 Fed. Reg. at 32,722. And according to Friends, that
definition of “net conservation benefit” in the Safe Harbor
Policy supposedly requires direct recovery of the species,
and thus does not include the “informational benefit” that
FWS expects from its experiment.

We disagree with Friends’ reading of “net conservation
benefit.” For starters, “net conservation benefit” is not
ambiguous—at least on whether that term includes
informational benefit. The ESA defines “conservation” as
“all methods and procedures” necessary for the recovery of
the species, which “include, but are not limited to, all
 FRIENDS OF ANIMALS V. USFWS 17

activities associated with scientific resources management
such as research.” 16 U.S.C. § 1532(3) (emphasis added).
“Research” means “the collecting of information about a
particular subject.” Merriam-Webster Dictionary,
https://www.merriam-webster.com/dictionary/research.
And we generally assume that a word in an implementing
regulation tracks the meaning of that same word in the
authorizing statute. See Decker v. Nw. Envtl. Def. Ctr., 568
U.S. 597, 609 (2013) (regulations must be interpreted
consistently with statute they implement).

So the definition of “conservation” in the ESA—and, by
extension, in the ESA regulation at issue—includes
activities aimed at collecting information (such as the
efficacy of barred owl removal as a conservation strategy).
And thus “net conservation benefit” includes informational
and research benefit contemplated by the barred owl removal
experiment. Whether this informational benefit outweighs
the harm done from any incidental take is an expert judgment
that we generally defer to the agency. San Luis & Delta-
Mendota Water Auth. v. Locke, 776 F.3d 971, 994 (9th Cir.
2014).

Ignoring the definition of “conservation” in the ESA,
Friends insists that we should look at the definition of “net
conservation benefit” in the agency’s Safe Harbor Policy:

“the cumulative benefits of the management
activities identified in a Safe Harbor
Agreement that provide for an increase in a
species’ population and/or the enhancement,
restoration, or maintenance of the covered
species’ suitable habitat within the enrolled
property, taking into account the length of the
Agreement and any off-setting adverse
effects attributable to the incidental taking
18 FRIENDS OF ANIMALS V. USFWS

allowed by the enhancement of survival
permit. Net conservation benefits must be
sufficient to contribute, either directly or
indirectly, to the recovery of the covered
species.”

64 Fed. Reg. at 32,722. According to Friends, the Policy’s
definition requires that the permits directly cause an
“increase” in either the northern spotted owl’s population or
its suitable habitat. And because the information from the
experiment does neither on its own, Friends argues that the
experiment cannot qualify as a “net conservation benefit.”

But even if we assume that the Safe Harbor Policy has
the force of law, Friends’ reading of “net conservation
benefit” remains unconvincing. 5 See W. Radio Servs. Co.,
Inc., v. Espy, 79 F.3d 896, 900 (9th Cir. 1996) (“[W]e will
review an agency’s alleged noncompliance with an agency
pronouncement only if that pronouncement actually has the
force and effect of law.”). To begin, the last sentence of the
Policy’s definition states that “net conservation benefit”
must “directly or indirectly” contribute to the recovery of the
species. Id. at 32,722 (emphasis added). The experiment
here “indirectly” aids the recovery of the northern spotted
owl.

Other parts of the Safe Harbor Policy also undermine
Friends’ cramped reading of “net conservation benefit.”
Much like we review an entire statute to determine a specific
provision’s meaning under traditional canons of statutory
interpretation, we must examine the entire Policy to discern
the meaning of “net conservation benefit.” Cf. U.S. Nat’l

5
While the Safer Harbor Policy underwent a notice-and-comment
process, it was not codified in the Code of Federal Regulations.
 FRIENDS OF ANIMALS V. USFWS 19

Bank of Or. v. Indep. Ins. Agents of America, Inc., 508 U.S.
439, 455 (1993) (When “expounding a statute, we must not
be guided by a single sentence or member of a sentence, but
look to the provisions of the whole law, and to its object and
policy.”) (internal quotation marks omitted). And in
responding to public comments asking to clarify the meaning
of “net conservation benefit,” FWS stated three times that
“net conservation benefits may result from . . . creating areas
for testing and implementing new conservation strategies.”
64 Fed. Reg. at 32,719 (Response 5), 32,720 (Response 11),
32,722 (Purpose of the Policy) (emphasis added). FWS here
is testing new conservation strategies by conducting the
barred owl removal experiment.

In short, we hold that “net conservation benefit” as used
in the ESA regulation includes informational benefit
provided by the barred owl removal experiment. 6

b. FWS reasonably described baseline conditions
using “resident” owl survey data.

Friends also claims that FWS improperly defined
baseline sites where it cannot remove the barred owls. 7 The

6
It does not necessarily mean that the agency can justify an
incidental take of a threatened or endangered species based on
speculative or questionable research. That issue goes to the “net
conservation benefit”—i.e., whether the research data outweighs the
harm caused by the take. Friends, however, does not appear to challenge
FWS’ determination that the experiment will provide useful data.

7
As with its “net conservation benefit” argument, Friends’ “baseline
conditions” claim presumes non-compliance with the Safe Harbor Policy
and other agency guidance. We do not review claims of non-compliance
with an agency’s own pronouncement unless that pronouncement carries
the force of law. See W. Radio, 79 F.3d at 900. Here, we need not decide
20 FRIENDS OF ANIMALS V. USFWS

Safe Harbor Policy requires that each Safe Harbor
Agreement fully describe “the agreed upon baseline
conditions” for the covered species within the property. 64
Fed. Reg. at 32,723. “Baseline conditions” are the
“population estimates and distribution and/or habitat
characteristics and determined area of the enrolled property
that sustain seasonal or permanent use by the covered
species . . . .” Id. at 32,722.

For the Roseburg, Oxbow, and Weyerhaeuser Safe
Harbor Agreements, FWS designated a site as “baseline”—
and thus not subject to the permits’ incidental take
authorizations—whenever a single resident spotted owl had
been spotted on that site in annual surveys from 2013 to
2015. For the Oregon Safe Harbor Agreement, FWS used
the same methodology but relied on surveys from 2011 to
2015.

Friends argues that FWS’ methodology suffered two
flaws. First, Friends claims that FWS determined that
baseline sites were “effectively abandoned” but that the
agency’s own policy statements show three to five years of
survey data cannot establish “abandonment.” This
“abandonment” argument is a red herring. There is simply
no requirement—in either the Safe Harbor Policy or the
agency’s other guidance—that FWS designate baseline
conditions based on “abandonment.” The Safe Harbor
Policy does not mention “abandonment” in its discussion of
baseline conditions. See 64 Fed. Reg. at 32,722–24. And
the Safe Harbor Agreements make no such finding either.
Each Safe Harbor Agreement determined that baseline sites
were “unoccupied,” not “abandoned.” The guidance relied

whether the Policy and other guidance carry the force of law because
FWS in any event complied with both.
 FRIENDS OF ANIMALS V. USFWS 21

on by Friends explicitly distinguishes between “unoccupied”
and “abandoned” sites. The guidance states that
“[o]ccupancy is an annual rate and is not equivalent to
‘abandoned,’ which is a permanent status.” And the
guidance says unoccupied sites should be determined by
using at least “3 years of survey.” FWS complied with this
methodology by using three to five years of surveys in
designating the Safe Harbor Agreements’ baseline
conditions.

Second, because the Safe Harbor Policy defines
“baseline conditions” as including areas that sustain
“seasonal” use, Friends alleges FWS erred in limiting
baseline sites to only those areas where a “resident” spotted
owl was detected and not considering “floaters” (young and
displaced spotted owls). The Safe Harbor Policy instructs
that determination of baseline conditions should be flexible
and based on agreement between FWS and the landowner.
64 Fed. Reg. at 32,719. Given the flexibility granted to the
parties, we cannot say exclusion of “floater” owls violates
the Safe Harbor Policy. As FWS explained, there is “no
evidence that floaters (young and displaced territorial
spotted owls) successfully breed unless they first become
established on a territory” and are thus unlikely to contribute
to the recovery of the species. It was reasonable for FWS
and the parties to set baseline sites based on “resident” owls
that are of primary conservation importance. Moreover, the
Safe Harbor Agreements include special protections during
nesting and roosting seasons in non-baseline areas, requiring
permittees to “refrain from removal or alteration of habitat”
within the core area containing nesting trees.
22 FRIENDS OF ANIMALS V. USFWS

c. FWS adequately analyzed the small portion of
critical habitat affected by the Oregon permit.

The ESA requires federal agencies to ensure that their
actions will not destroy or adversely modify a species’
designated critical habitat. 16 U.S.C. § 1536(a)(2). To this
end, FWS issued a Biological Opinion for each permit. Each
Biological Opinion concluded that the permit was unlikely
to destroy or adversely modify the northern spotted owl’s
critical habitat. 50 C.F.R. § 402.14(h)(iv)(A)–(B).

Friends objects to the Biological Opinions for two
reasons. First, Friends claims that the permits at issue
“overlap with critical habitat on state lands” and that FWS
failed to analyze this fact in the Biological Opinions. But
Friends cannot point to anything in the administrative record
showing that FWS failed to analyze affected critical habitat.
Only the Oregon permit/Safe Harbor Agreement overlaps
with critical habitat. Because the actual amount of critical
habitat that would be destroyed by the experiment was
unknown, FWS took a conservative approach, assuming the
entire 3,345 acres of critical habitat in the Oregon lands
would be removed. Even under this conservative
assumption, less than 0.04 percent of the spotted owl’s total
critical habitat would be destroyed. FWS concluded such a
“low level of potential loss would not impair the overall
recovery of the spotted owl . . . .” We cannot say such a
conclusion was erroneous.

Second, Friends argues that the Biological Opinions
were arbitrary and capricious because they analyze only one
subset of designated critical habitat—nesting/roosting
habitat—and ignore impacts to other subsets such as “habitat
for foraging and/or transience or colonization stages of
dispersal.” We agree with FWS that the Biological Opinions
sufficiently analyzed the relevant subsets of critical habitat.
 FRIENDS OF ANIMALS V. USFWS 23

Contrary to Friends’ claim, the Oregon Biological Opinions
did analyze the permit’s effect on foraging, transience, and
colonization habitat, but still concluded that the effect would
not appreciably reduce such subsets because of their
“scattered nature” throughout the Oregon lands.

But even without such an analysis, we would not
consider the agency’s focus on nesting/roosting habitat to be
arbitrary and capricious. FWS said in the Biological
Opinion that nesting/roosting habitat is “likely the most
important habitat in determining whether spotted owls can
support themselves within a specific area.” Given its
importance to spotted owl survival, we cannot say that the
agency acted improperly by focusing its analysis on a vital
habitat subset. See San Luis & Delta-Mendota, 776 F.3d at
994 (agency’s scientific judgments owed great deference).
We thus reject Friends’ critical habitat claims.

II. FWS complied with NEPA.

In 2013, FWS issued an Environmental Impact
Statement (EIS) analyzing the experiment’s environmental
impacts. Later, when FWS issued the permits and Safe
Harbor Agreements, it conducted a less-intensive
Environmental Assessment (EA) for each permit. All EAs
concluded that the spotted owl would not be significantly
affected. Friends contends FWS’ environmental analyses
did not meet NEPA requirements in two ways. First,
although FWS issued an initial EIS, Friends claims FWS had
to issue a Supplemental Environmental Impact Statement
(Supplemental EIS), instead of the lesser EAs, when it later
issued the permits. Second, Friends argues FWS should
have considered environmental effects of each permit with
those from other permits and the broader experiment. We
find FWS properly complied with its NEPA obligations.
24 FRIENDS OF ANIMALS V. USFWS

a. FWS did not have to conduct a Supplemental EIS
when it issued the permits.

NEPA does not expressly address when an agency must
prepare a Supplemental EIS. Marsh v. Or. Nat. Res.
Council, 490 U.S. 360, 370 (1989). But “NEPA cases have
generally required agencies to file [a Supplemental EIS]
when the remaining governmental action would be
environmentally ‘significant.’” Id. at 372 (quoting Tenn.
Valley Auth. v. Hill, 437 U.S. 153, 188 n.34 (1978)).
Whether an action, such as issuing the permits at issue, has
environmental significance is a “classic example of a factual
dispute” that “implicates substantial agency expertise” to
which “we must defer.” Id. at 376–77.

While NEPA does not squarely address Supplemental
EIS obligations, the Council on Environmental Quality
(CEQ) has issued regulations that “impose a duty on all
federal agencies to prepare” a Supplemental EIS if “(i) the
agency makes substantial changes to the proposed action that
are relevant to environmental concerns; or (ii) there are
significant new circumstances or information relevant to
environmental concerns and bearing on the proposed action
or its impacts.” Id. at 372; 40 C.F.R. § 1502.9(d)(1).

CEQ has published more guidance, which our circuit has
adopted as the proper framework for applying
§ 1502.9(d)(1). Russell Country Sportsmen v. U.S. Forest
Serv., 668 F.3d 1037, 1045 (9th Cir. 2011) (citing Forty
Most Asked Questions Concerning CEQ’s National
Environmental Policy Act Regulations, 46 Fed. Reg. 18,026,
18,035 (Mar. 23, 1981)). Under this framework, a
Supplemental EIS is not required if: “(1) the new alternative
is a ‘minor variation of one of the alternatives discussed in
the [original] EIS,’ and (2) the new alternative is
 FRIENDS OF ANIMALS V. USFWS 25

‘qualitatively within the spectrum of alternatives that were
discussed in the [original EIS].’” Id.

Friends argues that FWS had to issue a Supplemental
EIS under either prong of § 1502.9(d)(1). First, Friends
maintains that FWS made “substantial changes” to the
“heart” of the barred owl removal experiment because the
goal of the experiment was to conserve the northern spotted
owl but the permits authorized the take of spotted owls.
Second, Friends contends that the specifics of each permit
and Safe Harbor Agreement constitute “significant new
information” that was not considered in the initial EIS. We
disagree on both points.

First, the incidental take of northern spotted owls
authorized by the permits is only a “minor variation” of the
broader barred owl removal experiment analyzed in the
original EIS. The central component of the action was and
still is the removal of barred owls from treatment areas. The
permits help the removal of barred owls. Though FWS
stated that the permits would allow the experiment to
proceed in the most efficient and complete manner, the
experiment would still be possible without access to any
non-federal lands. We thus agree with FWS that the permits
were an ancillary aspect of the experiment and constitute a
“minor variation.” See Russell, 668 F.3d at 1048–49
(holding supplementation not required where variation is a
“secondary rather than primary” aspect of the action).

Additionally, the permits and Safe Harbor Agreements
were clearly “within the spectrum of alternatives” discussed
in the 2013 EIS. Id. at 1048. In the original EIS, FWS stated
that “[w]here possible, we would seek cooperation from
nonfederal landowners,” although “nonfederal lands would
be included in the active experiment only if the landowners
are willing.” Thus, the EIS adequately contemplated FWS’
26 FRIENDS OF ANIMALS V. USFWS

later issuance of the permits. And because issuance of the
permits ultimately depended on the cooperation of non-
federal parties, it “would be incongruous” with NEPA to
conclude FWS was without power to proceed with the
experiment until such specifics of the Safe Harbor
Agreements were fully fleshed out and assessed in a
Supplemental EIS. Cf. Robertson v. Methow Valley Citizens
Council, 490 U.S. 332, 352–53 (1989).

We are also satisfied that FWS conducted the required
“hard look” review in determining that the permits were not
environmentally significant. See Cal. ex rel. Imperial Cty.
Air Pollution Control Dist. v. U.S. Dep’t of the Interior, 767
F.3d 781, 792 (9th Cir. 2014). FWS conducted an EA for
each permit. Each EA determined the authorized incidental
take of northern spotted owls was likely to be small because
it would only occur if spotted owls repopulated non-baseline
sites after barred owls were removed. In other words,
spotted owls would be taken only if the experiment managed
to increase the spotted owl’s population and range. And
FWS concluded that such gains would be temporary, as
barred owls would resume displacing the spotted owls after
the experiment. In FWS’ opinion, the environmental effects
of the experiment were the same with or without the permits.
A Supplemental EIS is not required.

b. The permits and the experiment need not be
analyzed in a single EIS because they are not
“connected actions.”

An agency must discuss “connected actions” in a single
EIS. 40 C.F.R. § 1501.9(e)(1). Friends claims that the
broader experiment and the permits were “connected
actions.” As Friends sees it, each permit and SHA depends
on the experiment’s informational benefit to satisfy the “net
conservation benefit” requirement. Friends thus claims that
 FRIENDS OF ANIMALS V. USFWS 27

FWS erred in analyzing the experiment separately from the
permits and addressing each permit in isolation from the
other permits.

Actions are connected if they “[c]annot or will not
proceed unless other actions are taken previously or
simultaneously” or are “interdependent parts of a larger
action and depend on the larger action for their justification.”
Id. In applying § 1501.9(e)(1), we employ an “independent
utility” test. Great Basin Mine Watch v. Hankins, 456 F.3d
955, 969 (9th Cir. 2006). “When one of the projects might
reasonably have been completed without the existence of the
other, the two projects have independent utility and are not
‘connected’ for NEPA’s purposes.” Id. (citation omitted).

The permits are not “connected” to the broader
experiment because the experiment would proceed without
the permits. Friends claims that each permit’s legality
depends on the experiment. But access to any of the non-
federal lands (let alone all of them) was not considered
necessary by FWS to complete the experiment. While
failure to gain access to non-federal lands could delay the
result of the experiment, it would not altogether inhibit it.
Put another way, “one of the projects”—the barred owl
removal experiment—would be completed without the
other, meaning the experiment and the permits have
“independent utility” and are not “connected.” Id. (emphasis
added).

The permits are also not “connected” to each other. Each
permit has “independent utility” because the issuance of one
permit did not depend on the issuance of any other permit.
The EIS stated that each permit depended on “cooperation
from nonfederal landowners” and “nonfederal lands would
be included in the active experiment only if the landowners
are willing.” FWS issued the permits individually to each
28 FRIENDS OF ANIMALS V. USFWS

landowner, and irrespective of whether the other permits
would issue, so the permits are not “connected.” Id.
Because the permits and the experiment were not “connected
actions,” FWS did not have to assess their environmental
impacts together in a single document. 8

CONCLUSION

We hold that FWS complied with the ESA and NEPA in
issuing the permits and Safe Harbor Agreements. We
AFFIRM the district court’s grant of summary judgment for
FWS.

8
In the Roseburg-Oxbow final EA, FWS analyzed the cumulative
effects of the Roseburg-Oxbow, Weyerhaeuser, and Oregon SHAs to
“ensure a robust NEPA analysis.”