Native Village of Point Hope v. Sally Jewell

FOR PUBLICATION

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

NATIVE VILLAGE OF POINT HOPE; No. 12-35287
INUPIAT COMMUNITY OF THE ARCTIC
SLOPE; ALASKA WILDERNESS D.C. No.
LEAGUE; CENTER FOR BIOLOGICAL 1:08-cv-00004-
DIVERSITY; NATIONAL AUDUBON RRB
SOCIETY; NATURAL RESOURCES
DEFENSE COUNCIL; NORTHERN
ALASKA ENVIRONMENTAL CENTER; OPINION
OCEANA; PACIFIC ENVIRONMENT;
RESISTING ENVIRONMENTAL
DESTRUCTION ON INDIGENOUS
LANDS, A PROJECT OF THE
INDIGENOUS ENVIRONMENTAL
NETWORK (REDOIL); SIERRA CLUB;
THE WILDERNESS SOCIETY; WORLD
WILDLIFE FUND; DEFENDERS OF
WILDLIFE,
Plaintiffs-Appellants,

v.

SALLY JEWELL, Secretary of the
Interior; BUREAU OF OCEAN ENERGY
MANAGEMENT; TOMMY
BEAUDREAU, Director of the Bureau
of Ocean Energy Management,
Defendants-Appellees,
2 NATIVE VILLAGE OF POINT HOPE V. JEWELL

SHELL GULF OF MEXICO, INC.;
CONOCOPHILLIPS COMPANY; STATE
OF ALASKA; STATOIL USA E&P,
INC.,
Intervenor-Defendants-
Appellees.

Appeal from the United States District Court
for the District of Alaska
Ralph R. Beistline, Chief District Judge, Presiding

Argued and Submitted
March 5, 2013—Seattle, Washington

Filed January 22, 2014

Before: Ferdinand F. Fernandez, William A. Fletcher,
and Johnnie B. Rawlinson, Circuit Judges.

Opinion by Judge W. Fletcher;
Partial Concurrence and Partial Dissent by Judge
Rawlinson
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 3

SUMMARY*

Environmental Law

The panel reversed the district court’s summary judgment
entered in favor of federal defendants in an action challenging
the government’s environmental impact statements analyzing
the environmental effects of proposed leases for oil and gas
development in the Chukchi Sea of the northwest coast of
Alaska.

The panel held that the Final Environmental Impact
Statement and Supplemental Environmental Impact
Statement prepared by the federal defendants properly took
account of incomplete or unavailable information. The panel
held, however, that the reliance in the Final Environmental
Impact Statement on a one million barrel estimate of total
economically recoverable oil was arbitrary and capricious.
The panel remanded for further proceedings.

Judge Rawlinson concurred in part and dissented in part.
Judge Rawlinson agreed with most of the majority opinion,
but she did not agree that the federal Bureau of Ocean Energy
Management, Regulation and Enforcement acted arbitrarily
in selecting one billion barrels of oil as the benchmark for
analyzing the environmental affects of the proposed leases.

*
This summary constitutes no part of the opinion of the court. It has
been prepared by court staff for the convenience of the reader.
4 NATIVE VILLAGE OF POINT HOPE V. JEWELL

COUNSEL

Erik Clifford Grafe (argued), Earthjustice, Anchorage,
Alaska; Eric Paul Jorgensen, Earthjustice, Juneau, Alaska, for
Plaintiffs-Appellants.

David C. Shilton (argued), United States Department of
Justice, Washington, D.C., for Defendants-Appellees.

Kyle W. Parker (argued), Crowell & Moring LLP,
Anchorage, Alaska; Jeffrey Wayne Leppo and Ryan P. Steen,
Stoel Rives, LLP, Seattle, Washington; Ken Diemer and
Rebecca Kruse, Office of Alaska Attorney General,
Anchorage, Alaska; and James Leik, Perkins Coie LLP,
Anchorage, Alaska, for Intervenor-Defendants-Appellees.

OPINION

W. FLETCHER, Circuit Judge:

The Bureau of Ocean Energy Management (“BOEM”)1 of
the Department of the Interior has sought to lease “excellent
prospects” for oil and gas development in the Chukchi Sea off
the northwest coast of Alaska. The parcels available for lease
are cumulatively known as Lease Sale 193. Pursuant to the
National Environmental Policy Act (“NEPA”), BOEM
prepared a Final Environmental Impact Statement (“FEIS”)
analyzing the environmental effects of the proposed leases.

1
Earlier incarnations of the agency have been the Minerals Management
Service and the Bureau of Ocean Energy Management, Regulation and
Enforcement. For simplicity, we refer to the agency, including its earlier
incarnations, as BOEM.
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 5

BOEM based its environmental analysis on the assumption
that if oil development actually occurs, one billion barrels of
oil will be economically recoverable.

Plaintiffs argued in the district court that BOEM abused
its discretion by failing to account for essential missing
information in the FEIS. Plaintiffs also argued that BOEM’s
estimate of one billion barrels is arbitrary and capricious.
They contended that the potential economically recoverable
oil from the lease sale is far higher than one billion barrels,
and that BOEM had not given an adequate explanation for
using its lower estimate. The district court initially rejected
the FEIS for failing to account for the missing information.
After remand, BOEM prepared a Supplemental EIS (“SEIS”)
addressing the missing information. Based on the FEIS and
SEIS, the district court granted summary judgment to
defendants.

We largely agree with the district court that the agency
did not abuse its discretion in its analysis of the missing
information. However, we agree with plaintiffs that the
agency’s estimate of one billion barrels was chosen
arbitrarily, and that this arbitrary decision meant that the
agency based its decision on inadequate information about the
amount of oil to be produced pursuant to the lease sale.

I. Background

The Chukchi Sea is a southern arm of the Arctic Ocean
between Alaska and Russia. The Sea contains a wide variety
of animals, including bowhead whales, polar bears, pacific
walrus, seals, fish, and birds. Some of these animals provide
subsistence for native Inupiat communities along the Alaskan
coast. Some of the animals are listed under the Endangered
6 NATIVE VILLAGE OF POINT HOPE V. JEWELL

Species Act (“ESA”) as endangered or threatened. Five
exploratory wells were drilled in the Sea between 1989 and
1991. They had “positive shows” but did not lead to
commercial production.

The Outer Continental Shelf Lands Act (“OCSLA”)
prescribes four steps the federal government must take in
order to pursue offshore oil and gas development: “‘(1)
formulation of a five year leasing plan by the Department of
the Interior; (2) lease sales; (3) exploration by the lessees;
[and] (4) development and production.’” Edwardsen v. U.S.
Dep’t of the Interior, 268 F.3d 781, 784 (9th Cir. 2001)
(quoting Sec’y of the Interior v. California, 464 U.S. 312, 337
(1984)). At the “lease sale” stage, the Secretary of the
Interior selects the parcels that will be offered for lease,
accepts bids from parties, and collects funds from parties with
winning bids. The Department of the Interior must review
and approve specific exploration and development plans
before winning bidders can “proceed with full exploration,
development, or production” of oil or gas. Sec’y of the
Interior, 464 U.S. at 339. However, successful bidders have
the right to undertake “ancillary activities” in the field such
as geological and geophysical surveys and studies that
“model potential oil and hazardous substance spills.” 30
C.F.R. § 550.207.

A. NEPA

NEPA “protect[s] the environment by requiring that
federal agencies carefully weigh environmental
considerations and consider potential alternatives to the
proposed action before the government launches any major
federal action.” Barnes v. U.S. Dep’t of Transp., 655 F.3d
1124, 1131 (9th Cir. 2011) (internal quotation marks
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 7

omitted). “‘NEPA imposes procedural requirements designed
to force agencies to take a “hard look” at environmental
consequences’” of major federal action. Id. (quoting Earth
Island Inst. v. U.S. Forest Serv., 351 F.3d 1291, 1300 (9th
Cir. 2003)). The statute requires federal agencies to
“consider every significant aspect of the environmental
impact of a proposed action” and to “inform the public that
[they] ha[ve] indeed considered environmental concerns in
[their] decisionmaking process.” Balt. Gas & Elec. Co. v.
Natural Res. Def. Council, Inc., 462 U.S. 87, 97 (1983)
(internal quotation marks omitted).

NEPA requires that federal agencies prepare an EIS for
any “major Federal actions significantly affecting the quality
of the human environment.” 42 U.S.C. § 4332(2)(C). An
agency must consider:

(i) the environmental impact of the proposed
action,

(ii) any adverse environmental effects which
cannot be avoided should the proposal be
implemented,

(iii) alternatives to the proposed action,

(iv) the relationship between local short-term
uses of man’s environment and the
maintenance and enhancement of long-term
productivity, and

(v) any irreversible and irretrievable
commitments of resources which would be
8 NATIVE VILLAGE OF POINT HOPE V. JEWELL

involved in the proposed action should it be
implemented.

Id. An agency must take into account all “reasonably
foreseeable significant adverse effects” of the proposed action
in its analysis of environmental effects. 40 C.F.R. § 1502.22;
see also id. § 1508.7. NEPA also requires an agency to
analyze missing and incomplete information. As we explain
in greater detail below, an agency must either obtain
information that is “essential to a reasoned choice among
alternatives” or explain why such information was too costly
or difficult to obtain. Id. § 1502.22.

An agency is required to comply with NEPA at various
stages of the oil and gas development process. An agency is
not required at the lease sale stage to analyze potential
environmental effects on a site-specific level of detail. N.
Alaska Envtl. Ctr. v. Kempthorne, 457 F.3d 969, 975–76 (9th
Cir. 2006). To some degree, lease sale analyses may be based
on information that is uncertain or missing at the time of the
sale when that information can be obtained at a “later stage[]
of the exploration process.” Tribal Vill. of Akutan v. Hodel,
869 F.2d 1185, 1192 (9th Cir. 1988). At the same time, the
agency cannot shirk its responsibility to “consider[] all
foreseeable direct and indirect impacts” of the proposed
action in its EIS. N. Alaska Envtl. Ctr., 457 F.3d at 975
(internal quotation marks omitted). The agency also must
“discuss[] . . . adverse impacts” without “improperly
minimiz[ing] negative side effects.” Id.

B. Lease Sale 193

After completing a five-year leasing plan for the Chukchi
Sea, BOEM decided to offer a large portion of the Sea for oil
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 9

and gas leasing. The FEIS analyzed four alternatives for the
lease sale: (1) a 34-million acre proposed lease option
covering 6,156 blocks of the Chukchi Sea; (2) a no-lease
option; (3) a proposed lease option excluding 1,765 blocks
extending along a corridor about 60 miles from the Alaskan
coast; and (4) a proposed lease option excluding 795 blocks
extending along a corridor between 25 and 50 miles from the
Alaskan coast.

The National Marine Fisheries Service recommended that
the Secretary of the Interior select the third alternative, under
which development would be farther from the coast, based on
its conclusion that numerous endangered and threatened
species living close to shore would be adversely affected by
oil development. The Secretary of the Interior accepted
BOEM’s recommendation and selected the fourth alternative,
under which development would be closer to the coast.

The lease sale occurred on February 6, 2008. The federal
government collected over $2.6 billion from the winning
bidders. At the time of the lease sale, there were no active
leases in the Sea.

C. Procedural History

Plaintiffs filed suit, alleging seven deficiencies in the
FEIS:

1. [The FEIS] does not adequately analyze
and present the impacts of Lease Sale 193 on
the environment and human communities;

2. [It] fails to include essential missing
information about the Chukchi Sea and the
10 NATIVE VILLAGE OF POINT HOPE V. JEWELL

potential impacts of the lease sale, or explain
why excluding this information is justified;

3. [It] fails to adequately analyze the impact
of the lease sale in the context of a warming
climate;

4. [It] understates the potential impacts of oil
and gas development pursuant to the leases by
analyzing a limited development scenario;

5. [It] understates the risks of an oil spill;

6. [It] fails to fully analyze the cumulative
impacts to threatened eiders of the lease sale
and other oil and gas development in
threatened eiders’ Arctic habitat; and

7. [It] provides a misleading analysis of the
effects of seismic surveying.

The parties cross-moved for summary judgment.

The district court agreed with defendants that much of the
FEIS complied with NEPA, including the FEIS’s assumption
that there would be one billion barrels of economically
recoverable oil. However, the court concluded that the
FEIS’s analysis was flawed in three respects: it “failed to
analyze the environmental impact of natural gas development,
despite industry interest and specific lease incentives for such
development”; it “failed to determine whether missing
information identified by the agency was relevant or essential
under 40 C.F.R § 1502.22”; and it “failed to determine
whether the cost of obtaining the missing information was
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 11

exorbitant, or the means of doing so unknown.” The district
court granted in part plaintiffs’ motion for summary
judgment, issued a limited injunction, and remanded to
BOEM for further proceedings.

After remand from the district court, BOEM prepared an
SEIS. The SEIS analyzed the consequences of natural gas
exploration and production. In the aftermath of the
Deepwater Horizon oil spill in the Gulf Coast, it also
analyzed the environmental impacts of a very large oil spill.
Finally, BOEM prepared an appendix analyzing “whether the
information gaps that were identified in the Sale 193 FEIS
were relevant and necessary to evaluate reasonably
foreseeable significant adverse effects.” Based on the FEIS,
now supplemented by the SEIS, the Secretary of Interior
again chose the fourth alternative for oil and gas leasing.

Based on the FEIS and SEIS, the district court granted
BOEM’s motion for summary judgment. The court found
that “BOEM has identified missing or incomplete information
and has adequately evaluated it in a manner that is clearly
sufficient at this stage of the development process to satisfy
the requirements of 40 C.F.R. § 1502.22.” The court gave
“considerable deference . . . to BOEM’s expertise.” Plaintiffs
timely appealed.

II. Standard of Review

Our review of an EIS is governed by the Administrative
Procedure Act (“APA”). “Under the APA, we may set aside
an agency decision if it is ‘arbitrary, capricious, an abuse of
discretion, or otherwise not in accordance with law.’” Native
Ecosystems Council v. U.S. Forest Serv., 428 F.3d 1233, 1238
(9th Cir. 2005) (quoting 5 U.S.C. § 706(2)(A)). “Review
12 NATIVE VILLAGE OF POINT HOPE V. JEWELL

under the arbitrary and capricious standard is narrow, and we
do not substitute our judgment for that of the agency.” Lands
Council v. McNair (Lands Council II), 537 F.3d 981, 987 (9th
Cir. 2008) (en banc) (alterations and internal quotation marks
omitted). However, an agency’s decision can be set aside if:

the agency relied on factors Congress did not
intend it to consider, entirely failed to
consider an important aspect of the problem,
or offered an explanation that runs counter to
the evidence before the agency or is so
implausible that it could not be ascribed to a
difference in view or the product of agency
expertise.

Id. (internal quotation marks omitted). Such actions would be
“clear error[s] of judgment that would render [the agency’s]
action arbitrary and capricious.” Id. at 993 (internal quotation
marks omitted).

We “may affirm a summary judgment only if, viewing the
evidence in the light most favorable to the party against
whom it is granted, we find no genuine issue of material fact,
and we find that the prevailing party is clearly entitled to
judgment as a matter of law.” California v. Watt, 683 F.2d
1253, 1258 (9th Cir. 1982), rev’d on other grounds sub nom.
Sec’y of the Interior v. California, 464 U.S. 312 (1984).

III. Discussion

On appeal, plaintiffs argue that BOEM abused its
discretion in two respects. First, they argue that “essential”
information is missing from the FEIS and SEIS, in violation
of 40 C.F.R. § 1502.22. Second, they argue that the FEIS and
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 13

SEIS underestimate the adverse environmental impact of the
lease sale because they use an unrealistically low estimate of
the economically recoverable oil. We disagree with
plaintiffs’ first argument, but agree with their second
argument.

A. Essential Information

An agency’s obligation with respect to incomplete or
unavailable information is spelled out in 40 C.F.R. § 1502.22.
The agency “shall always make clear that . . . information is
lacking.” Id. If the missing information is “relevant to
reasonably foreseeable significant adverse impacts” and is
“essential to a reasoned choice among alternatives and the
overall costs of obtaining it are not exorbitant,” the agency
must include that information in the EIS. Id. § 1502.22(a).
If the missing information “cannot be obtained because the
overall costs of obtaining it are exorbitant or the means to
obtain it are not known,” the agency must include the
following in the EIS: (1) a statement that such information is
“incomplete or unavailable”; (2) a statement of the “relevance
of the incomplete or unavailable information to evaluating
reasonably foreseeable significant adverse impacts on the
human environment”; (3) a “summary of existing credible
scientific evidence . . . relevant to evaluating the reasonably
foreseeable adverse impacts”; and (4) the agency’s
“evaluation of such impacts based upon theoretical
approaches or research methods generally accepted in the
scientific community.” Id. § 1502.22(b). Section 1502.22(b)
clarifies that reasonably foreseeable effects “include[]
impacts which have catastrophic consequences, even if their
probability of occurrence is low.”
14 NATIVE VILLAGE OF POINT HOPE V. JEWELL

Much of the information missing from the EIS concerns
animal populations potentially affected by oil exploration and
production under the leases. The missing information
concerns such things as population levels of various species
of animals in the Chukchi Sea, including endangered or
threatened animals; the locations of various animal
populations during the year; the feeding and breeding habits
of various animal populations; and the vulnerability of
various animal populations to drilling and other exploration
and production-related activities.

With respect to the potential environmental harm from a
large oil spill, BOEM concluded that the missing information
was not essential to a reasoned choice among the alternatives.
It wrote in the SEIS, “[I]n the unlikely event of a large oil
spill, it is well-understood that environmental impacts could
be severe. The severity of potential impacts would be nearly
identical under any action alternative; therefore, very specific
types of information relevant to species, particular life history
traits, or behavior do not help substantially in distinguishing
among alternatives.” With respect to other activities or
events with potential adverse impacts on the animal
populations in the Chukchi Sea, BOEM concluded that
sufficient protections would be provided by the requirements
of other environmental statutes, such as the Clean Air Act, the
Marine Mammal Protection Act (“MMPA”), and the ESA,
and by the requirement under NEPA to provide site-specific
analyses at later stages of development.

Based on these conclusions, BOEM stated in the SEIS
that it did not consider any of the incomplete or unavailable
information at issue to be “essential to a reasoned choice
among alternatives” at this stage of the development process.
40 C.F.R. § 1502.22(a). BOEM therefore did not determine
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 15

whether the information was unobtainable “because the
overall costs of obtaining it are exorbitant or the means to
obtain it are not known.” Id. § 1502.22(b). Nor did BOEM
go through the steps required by § 1502.22(b) if it had found
“essential” information to be unobtainable. Instead, BOEM
specifically relied in the SEIS on what it characterized as its
“understanding that certain items of presently missing or
incomplete information will be known (and utilized to avoid
or minimize adverse impacts) at a later stage of OCS Lands
Act environmental review.” BOEM promised in the SEIS
that it “would thoroughly review specific development &
production plans at Step 4 [‘development and production’],
if and when a project proponent actually submits a plan.
Thus, while certain information may, in fact, be essential at
a later stage of OCS Lands Act [review], such information
may not be essential to a reasoned choice among alternatives
at this lease sale stage.”

In Village of False Pass v. Clark, 733 F.2d 605 (9th Cir.
1984), we reviewed an EIS of an oil and gas lease sale under
OCSLA. The plaintiff had argued that the commitment made
by the government when entering into leases under OCSLA
is so substantial that a fully exhaustive environmental
analysis under NEPA had to be performed at the lease sale
stage. We disagreed, writing:

NEPA may require an environmental impact
statement at each stage: leasing, exploration,
and production and development.
Furthermore, each stage remains separate.
The completion of one stage does not entitle
a lessee to begin the next.
16 NATIVE VILLAGE OF POINT HOPE V. JEWELL

Id. at 614. We wrote to the same effect in Northern Alaska
Environmental Center:

[P]rojects [for the development of oil and gas
natural resources] generally entail separate
stages of leasing, exploration and
development. At the earliest stage, the leasing
stage we have before us, there is no way of
knowing what plans for development, if any,
may eventually materialize.

457 F.3d at 977.

A lease sale under OCSLA is analogous to a
“programmatic” plan. The required level of analysis in an
EIS is different for programmatic and site-specific plans. We
wrote in Friends of Yosemite Valley v. Norton, 348 F.3d 789
(9th Cir. 2003):

An agency’s planning and management
decisions may occur at two distinct
administrative levels:

(1) the “programmatic level” at which
the [agency] develops alternative
management scenarios responsive to
public concerns, analyzes the costs,
benefits and consequences of each
alternative in an [EIS], and adopts an
amendable [management] plan to guide
management of multiple use resources;
and (2) the implementation stage during
which individual site specific projects,
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 17

consistent with the [management] plan,
are proposed and assessed.

Ecology Ctr., Inc. v. United States Forest
Serv., 192 F.3d 922, 923, [n.2] (9th Cir.
1999). An EIS for a programmatic plan . . .
must provide ‘sufficient detail to foster
informed decision-making,’ but ‘site-specific
impacts need not be fully evaluated until a
critical decision has been made to act on site
development.’ N. Alaska Envtl. Ctr. v. Lujan,
961 F.2d 886, 890–91 (9th Cir. 1992).

Id. at 800 (alterations in original).

Regardless of whether a programmatic or site-specific
plan is at issue, NEPA requires that an EIS analyze
environmental consequences of a proposed plan as soon as it
is “reasonably possible” to do so. We wrote in Kern v. U.S.
Bureau of Land Management, 284 F.3d 1062 (9th Cir. 2002),
with respect to a programmatic plan:

Once an agency has an obligation to
prepare an EIS, the scope of its analysis of
environmental consequences in that EIS must
be appropriate to the action in question.
NEPA is not designed to postpone analysis of
an environmental consequence to the last
possible moment. Rather, it is designed to
require such analysis as soon as it can
reasonably be done. If it is reasonably
possible to analyze the environmental
consequences in an EIS for [a Resource
18 NATIVE VILLAGE OF POINT HOPE V. JEWELL

Management Plan], the agency is required to
perform that analysis.

Id. at 1072 (citation omitted); see also Pac. Rivers Council v.
U.S. Forest Serv., 689 F.3d 1012, 1025–27, 1029–30 (9th Cir.
2012), dismissed as moot, 133 S. Ct. 2843 (2013). This is not
to say that an agency must provide the most extensive
environmental analysis possible at the earliest possible
moment, for an agency has some flexibility in deciding the
level of analysis to be performed at a particular stage. We
will defer to the agency’s judgment about the appropriate
level of analysis so long as the EIS provides as much
environmental analysis as is reasonably possible under the
circumstances, thereby “provid[ing] sufficient detail to foster
informed decision-making” at the stage in question. Friends
of Yosemite Valley, 348 F.3d at 800 (internal quotation marks
omitted).

In the case before us, we conclude that BOEM has
reasonably concluded that the missing information from the
FEIS and SEIS is not “essential” to informed decisionmaking
at the lease sale stage. We agree with BOEM that compliance
with statutes such as the MMPA and the ESA will provide
protection for animals covered by those statutes. The MMPA
generally prohibits the “take” of marine mammals. 16 U.S.C.
§ 1371(a). A “take” encompasses any act of “torment” or
“annoyance” that “has the potential to injure . . . or . . . disturb
a marine mammal or marine mammal stock in the wild by
causing disruption of natural behavioral patterns, including,
but not limited to, migration, surfacing, nursing, breeding,
feeding, or sheltering.” Id. § 1362(13), (18)(A)(i)–(ii).
Unlawful “takes” trigger civil and criminal penalties. Id.
§ 1375(a)(1), (b). Further, under the ESA § 7(a)(2),
16 U.S.C. § 1536(a)(2), BOEM must consult with the
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 19

National Marine Fisheries Service and the U.S. Fish and
Wildlife Service to “insure that any action authorized, funded,
or carried out by such agency . . . is not likely to jeopardize
the continued existence of any endangered species or
threatened species.” If an action is likely to jeopardize a
species, the acting agency must determine whether any
“reasonable and prudent alternatives” exist that will avoid
jeopardizing that species. 16 U.S.C. § 1536(b)(3)(A). We
recognize that BOEM has already consulted with these
agencies at the lease sale stage. It may well have to consult
with them again at the development and production stage
when specific plans have been proposed and site-specific
activities are contemplated. (We note that it may also have to
consult again at the lease sale stage, once it has performed a
proper analysis of the estimated overall oil production.)
Because these statutes provide additional protections for
animals in the Chukchi Sea, they support BOEM’s conclusion
that missing information about these animals was not
“essential” at this stage.

We also agree with BOEM that further environmental
analysis will be appropriate at a later stage. In BOEM’s
words, “certain items of presently missing or incomplete
information will be known (and utilized to avoid or minimize
adverse impacts) at a later stage of OCS Lands Act
environmental review.” That is, “when a project proponent
actually submits a plan,” BOEM will be required under
NEPA to perform a plan- or site-specific environmental
analysis of that proposed plan. At that stage, missing or
incomplete information that has not been “essential to a
reasoned choice among alternatives” at the lease sale stage
may later become essential. If there is “essential”
information at the plan- or site-specific development and
production stage, BOEM will be required to perform the
20 NATIVE VILLAGE OF POINT HOPE V. JEWELL

analysis under § 1502.22(b) that it has not performed in the
FEIS and SEIS now before us.

Of course, we recognize that our discussion and decision
in the next Section regarding BOEM’s one billion barrel
estimate may have some effects upon the remainder of the
FEIS. But we will not at this time speculate about the extent
of those effects, if any. The Defendants are in the best
position to analyze those effects, if any, and have the duty to
analyze them in the first instance.

B. One Billion Barrel Estimate

Plaintiffs contend that BOEM chose an arbitrary number
for the total barrels of economically recoverable oil from
Lease Sale 193. The FEIS estimated the amount of
recoverable oil by estimating production from the “first
offshore oil field” that would be developed within the area of
the leases. BOEM did not make any estimate of recoverable
oil from additional fields that might be developed. The FEIS
specified that the “recoverable oil resources from this field
are assumed to be 1 billion barrels (Bbbl).” The FEIS then
used the one billion barrel estimate as the basis for its
environmental analysis.

We must determine whether BOEM has articulated a
rational basis for its decision to use the one billion barrel
estimate. Mora-Meraz v. Thomas, 601 F.3d 933, 939 (9th
Cir. 2010). We must reverse a decision as arbitrary and
capricious if

the agency relied on factors Congress did not
intend it to consider, entirely failed to
consider an important aspect of the problem,
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 21

or offered an explanation that runs counter to
the evidence before the agency or is so
implausible that it could not be ascribed to a
difference in view or the product of agency
expertise.

Lands Council II, 537 F.3d at 987 (internal quotation marks
omitted). For the reasons that follow, we conclude that
BOEM’s one billion barrel estimate is arbitrary and
capricious.

1. The Selection of One Billion Barrels

BOEM first announced it was developing an EIS in
preparation for Lease Sale 193 in July 2005. According to
internal BOEM emails, BOEM analyst Jim Craig was
assigned to provide “resource estimates and a scenario”
which other BOEM scientists would use to analyze
environmental effects. Craig emailed his supervisor, Deborah
Cranswick, stating that he believed that “[t]he reasonably
foreseeable scenario” should include “oil production from the
first field only, not the full economic potential.” Craig’s
reason for focusing on the first field production was practical;
he would have to wait for about two months to have
information that would allow him to develop a scenario for
the entire area covered by the lease sale. Craig stated in his
email, “You realize that we won’t have the 2005 resource
assessment numbers until Sept, so we must base the scenario
on the ‘first development’ not the total economic potential.”
Craig also indicated that this emphasis on the first oil field
was a “departure from previous work.”

Craig asked Cranswick whether the scenario should
employ a single estimate of oil production from that first
22 NATIVE VILLAGE OF POINT HOPE V. JEWELL

field, or whether it should employ a range. Cranswick
responded by email that she preferred a range. Craig then
suggested, in a July 29 email, a range from 500 million
barrels to 1.5 billion barrels. Craig emphasized in his email
that the scenario should assume “equal probability for any
volume within the range” so that one billion barrels “does not
become the de facto most-likely” outcome. Craig’s draft
scenario also noted, with respect to recoverable oil in the
Chukchi Sea, that “[o]ur current petroleum assessment
indicates that recoverable oil resources could range from 3.6
to 11.8 billion barrels.”

There is a gap in the email chain in our record, so we do
not know Cranswick’s next response to Craig. But we do
know that in a subsequent email from Craig to Cranswick on
August 2, Craig proposed a single one billion barrel estimate
as an alternative to using a range that was “too broad”:

Attached is a table with E&D data. If this
represents too broad of a range, then I think
we should fall back to a single volume (1.0
Bbbl) for the EIS analysis with a
corresponding set of single E&D numbers.
It’s hard to have it both ways (very narrow
range) when these figures are entirely
speculative.

There are two clear options:

1) 500-1500 MMbbl, as a uniform
distribution (every point in range is
equally likely). This will require a low and
high case analysis.
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 23

2) 1.0 Bbbl as a single point estimate with
no confidence interval. This will require
a mostly likely case analysis only.

Although it would be nice to propose a
recoverable oil volume of 932 MMbbl +/- 134
MMbbl in a 90% confidence interval, we
don’t have any data to support it.

Pick (1) or (2), but not (1) and (2).

On August 3, Cranswick emailed Craig a data chart reflecting
Craig’s second option. It contained only a single one billion
barrel estimate.

On that same day, Craig emailed to Cranswick a draft
scenario relying on the one billion barrel estimate of oil
production. This draft explained:

The scenario assumed for environmental
analysis involves the discovery, development,
and production of the first oil field in the
Chukchi sale area. Ultimately recoverable oil
resources from this field are assumed to be 1
billion barrels (Bbbl). Smaller oil volumes are
not likely to be economic to produce and
single pools containing larger volumes are
increasingly rare. If oil prices drop below
$30.00 per barrel (they are above $50.00
when this scenario was written), exploration
in the Chukchi OCS is expected to be minimal
and oil discoveries may not be developed.
24 NATIVE VILLAGE OF POINT HOPE V. JEWELL

The draft also pointed out that “the mean recoverable oil
resource [in the Chukchi Sea] is 12 Bbbl with a 5%
probability of 29 Bbbl.” Craig also prepared a chart for
Cranswick comparing the numbers Craig had selected for the
Lease Sale 193 EIS to an EIS prepared for the Chukchi Sea
and Hope Sea Basin for the 2002–2007 Five Year Oil and
Gas Leasing Program. That previous EIS had estimated a
range for economical oil production from 0.96 billion barrels
to 2.42 billion barrels.

On August 10, Cranswick circulated the Lease Sale 193
EIS scenario to other BOEM scientists who would be
working on the EIS. The scenario contained the one billion
barrel estimate. Cranswick explained in an email that

[t]he scenario is based on a one mid-range
economic resource number (note - this is not
necessarily the most likely. A lower volume is
more likely to occur but less likely to be
developed from an economic standpoint; a
higher volume is less likely to occur but more
likely to be developed).

Several BOEM employees expressed concern with the
agency’s proposed scenario. For example, one NEPA analyst
employed by BOEM, Dee Williams, wrote, “I don’t
understand why [the agency] doesn’t use their sophisticated
assessment indices to impose a more definitive likely
scenario. Clearly, it is impossible to predict ‘with certainty’,
but the narrative needs to inspire greater public confidence by
explaining the parameters of reasonable expectations.”
Williams further stated:
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 25

If it becomes economical to build one
platform to produce an estimated 1 billion
barrels, and there is between 12 and 29 billion
barrels that are recoverable, why is the
scenario not compelled to imagine more than
one platform (i.e. is a single platform always
the initial scenario, in which case maybe we
should just explain that)?

Cranswick responded that “the initial scenario is one platform
because we can’t have only a partial platform if that is all that
the resource estimate support[s].” At the same time,
Cranswick suggested that smaller oil developments would be
associated with the first oil platform. “Once the first platform
goes in, it is likely that additional satellite subsea completions
would be developed before another host platform would be
considered.”

Once the draft EIS was completed, BOEM sought
comments from other agencies and from the public.
Numerous outside commentators expressed concern about the
scenario BOEM had developed. For example, the
Environmental Protection Agency wrote that

the hypothetical development scenario that is
used in the document add[s] additional layers
of uncertainty regarding the probabilities of
exploration, production and development
activities and the risks associated with those
activities. . . . EPA is concerned that, overall,
the depth and diversity of uncertainties
presented in the document resulted in the lack
of adequate support for many of the
document’s conclusions.
26 NATIVE VILLAGE OF POINT HOPE V. JEWELL

The Division of Migratory Bird Management at the U.S. Fish
and Wildlife Service (“FWS”) similarly challenged the one
billion barrel estimate as inaccurate:

The basic assumptions used in the analysis of
effects are flawed with regards to the size of
development scenarios. The [Draft EIS
(“DEIS”)] states that the current petroleum
assessment indicates a mean recoverable oil
resource of 12 billion barrels; yet all
environmental analyses reported in the DEIS
are based on a development of 1 billion
barrels, thereby significantly underestimating
likely scenarios.

The Division recommended that BOEM not proceed with the
lease sale until problems with the EIS were corrected. Public
commentators similarly pointed to flaws in employing a one
billion barrel production estimate, including that such an
estimate was “based on a price of oil at half the current
market value,” that the estimate “severely understates the true
cumulative impacts” of oil production, and that it was
“nowhere . . . justified with scientific analysis.” Despite these
criticisms, BOEM continued to rely on its one billion barrel
estimate. The one billion barrel estimate was the basis for the
entire FEIS, including its analysis of the risk of a large oil
spill. For example, BOEM instructed the FWS to rely on that
estimate in that agency’s analysis of whether the lease sale
would jeopardize listed threatened species such as the
spectacled and Steller’s eiders. Had FWS made a jeopardy
finding, BOEM either would not have been able to proceed
with the Lease Sale under the ESA or would have had to
obtain an exemption from the “no jeopardy” rule. 16 U.S.C.
§ 1536(a)(2).
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 27

2. Arbitrary and Capricious

Plaintiffs contend that the one billion barrel estimate was
chosen arbitrarily, and that BOEM did not provide an
adequate explanation for its selection. We agree for three
reasons.

First, BOEM has not justified its choice of the lowest
possible amount of oil that was economical to produce as the
basis for its analysis. The draft EIS scenario stated that the
agency chose to focus on a one billion barrel estimate in part
because any volume lower than one billion barrels would not
be economical to produce. At the same time, BOEM was
well aware that if any oil was produced from Lease Sale 193,
the economically recoverable oil was very likely to exceed
one billion barrels. In an August 18, 2005, email
commenting on the in-progress draft EIS, Jim Craig wrote,
“We assume 1 billion bbl for the first field, but there is
another 11 Bbbl that is economic at $70.” Craig attached a
table to a December 2005 email, listing “Estimates for
Speculative Oil and Gas Reserves,” specifying a range
between 1.0 and 6.1 billion barrels for the “Chukchi Shelf.”
Finally, in a May 2006 email Craig wrote, “The ‘1-billion
barrel, first field’ assumption is subjective (‘for purposes of
analysis’) and represents only a fraction of the full economic
resource potential in the Chukchi (which was recently
published).”

The mean estimate of economical oil production, at the
center of the distribution curve, is by definition a more likely
occurrence than is the lowest estimate of viable oil
production. Previous EISes in the Chukchi Sea had used the
mean estimate of oil production as the basis for their
analyses, and those EISes had also included low and high
28 NATIVE VILLAGE OF POINT HOPE V. JEWELL

estimates. For example, BOEM previously leased portions of
the Chukchi Sea in now-expired Lease Sale 109. The parcels
leased in Lease Sale 109 overlap substantially with the
parcels leased in Lease Sale 193. Documents prepared in
advance of Lease Sale 109 stated that “[t]he mean resource
estimate . . . is 2.68 billion barrels of oil with a 20 percent
chance of a discovery of commercially recoverable oil.” In
estimating the effects of oil spills from Lease Sale 109,
BOEM “assume[d] the full development of the resource
estimate of 2.68 billion barrels.” In contrast, while estimates
in the record about the economically recoverable amount of
oil from Lease Sale 193 vary, nowhere is the mean amount of
economical production calculated to be less than 2.37 billion
barrels. But the FEIS for Lease Sale 193 uses one billion
rather than 2.37 billion barrels as the basis for its analysis of
environmental consequences.

BOEM’s primary explanation for using its low-end
estimate for oil production is that this scenario overestimates
the likely amount of production. BOEM emphasizes that
because of the remoteness of the area and the risk of
economic failure, any oil production activity is an unlikely
result of the lease sale. More specifically, BOEM estimates
that there is a less than 10 percent likelihood that oil
development in the region will occur. Defendants argue that
since the most likely foreseeable outcome is no oil
development at all, one billion barrels of oil production is
actually a generous estimate.

This analysis is flawed. The assumption that there is a 10
percent chance of commercial oil development is itself
without a rational basis in the record. Jim Craig first
developed the estimate “off the top of [his] head” in an email
exchange. That calculation contradicts estimates used earlier
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 29

in the EIS, as well as estimates used in past EISes for the
Chukchi Sea. Further, BOEM conflates the likelihood of oil
and gas production with the likelihood of environmental
effects if such production occurs. Based on its responsibility
to “‘consider[] all foreseeable direct and indirect impacts’” of
the proposed action, N. Alaska Envtl. Ctr., 457 F.3d at 975
(citation omitted), BOEM concluded that oil production was
“reasonably foreseeable.” There is a substantial basis for this
in the record because, as noted by BOEM, “the area has high
oil resource potential and there is existing transportation
infrastructure to move oil from northern Alaska to distant
markets.” Once BOEM made the determination that
production is reasonably foreseeable, it was required to
consider the full cumulative impact of that production. See
40 C.F.R. § 1508.7. Put differently, BOEM might well be
right that the most likely outcome is that there will be no oil
development in the Chukchi Sea. But that fact should have
made no difference to BOEM’s analysis of the reasonably
foreseeable environmental effects of oil development, if such
development does occur.

Second, the FEIS did not take into account variation in oil
prices in arriving at the estimate that one billion barrels of oil
are economically recoverable. An assumption of stable prices
ignores the fact that the amount of economically recoverable
oil varies substantially depending on oil prices. This may be
seen, for example, in a 2006 report of the Minerals
Management Service (a prior incarnation of BOEM), which
estimated economically recoverable oil from the Chukchi
Shelf at different prices. At $30 per barrel, the mean estimate
was 0 barrels; at $46 per barrel, the mean estimate was 2.37
billion barrels; at $60 per barrel, the mean estimate was 8.38
billion barrels; at $80 dollars per barrel, the mean estimate
was 12 billion barrels.
30 NATIVE VILLAGE OF POINT HOPE V. JEWELL

Third, BOEM has not provided an adequate explanation
for its decision to base its EIS only on the amount of oil
expected to be produced from the first field in the leased area
of the Chukchi Sea. It is unclear from the record how BOEM
initially estimated that the first field would produce one
billion barrels of oil. Jim Craig himself suggested that his
calculations regarding that first development were “entirely
speculative.” But even assuming that one billion barrels is an
accurate estimate of the amount of oil to be produced from
the first field, it is unclear why BOEM assumed that only one
oil field would be developed in the lease area. The FEIS
itself acknowledges that “[w]hen the first project overcomes
the cost, logistical, and regulatory hurdles, more projects are
. . . likely to follow.”

The FEIS explains that it is unlikely that “all
economically viable resources will be developed” in the Sea
due to the difficulties in operating in a frontier area of oil
production. But the FEIS does not explain why production
would be expected to stop if the first oil field is developed.
The primary explanation for that assumption suggested by the
record is that data to analyze “the full economic potential” of
the lease sale would not be available until about two months
after Jim Craig initially proposed an estimate based on the
first field. Previous evaluations of Chukchi Sea oil
development had assumed that multiple oil fields would
develop once commercial development was viable. In a
technology assessment of Chukchi Sea petroleum
development performed in 1983 for BOEM, the Bureau of
Land Management had used a 1.5 billion barrel estimate to
measure prospects in “the central Chukchi shelf.” That
assessment assumed that two oil fields would be developed:
one of one billion barrels and one of 0.5 billion barrels. On
the record before us, it remains unclear why BOEM chose to
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 31

analyze the lowest amount of oil that could be produced in
the Chukchi Sea from the smallest number of oil fields that
could be developed.

Defendants contend that any error resulting from using
the one billion barrel estimate can be corrected through site-
specific EISes later in the development process. We disagree.
An agency is required to analyze the environmental effects in
an EIS as soon as it is “reasonably possible” to do so. Kern,
284 F.3d at 1072. An appropriate time to estimate the total
oil production from the lease sale is the time of the lease sale
itself. Under NEPA, BOEM is required to take into account
the full environmental effects of its actions when deciding
whether and in what manner to pursue the lease sale.
42 U.S.C. § 4332(2)(C). A later project or site-specific
environmental analysis is an inadequate substitute for an
estimate of total production from the lease sale as a whole.
It is only at the lease sale stage that the agency can adequately
consider cumulative effects of the lease sale on the
environment, including the overall risk of oil spills and the
effects of the sale on climate change. It is also only at the
lease sale stage that the agency can take into account the
effects of oil production in deciding which parcels to offer for
lease.

We also disagree with defendants that our decisions in
Akutan, 869 F.2d at 1191–92, False Pass, 733 F.2d at 617,
and Watt, 683 F.2d at 1267–68, compel a contrary result. In
False Pass, plaintiffs challenged the agency for failing to
consider the worst case scenario for oil development.
773 F.2d at 614. In the circumstances presented there, we
held that there was a rational explanation for not considering
the worst case at the lease sale stage. Here, in contrast, the
BOEM considered only the best case scenario for
32 NATIVE VILLAGE OF POINT HOPE V. JEWELL

environmental harm, assuming oil development. A best case
scenario “skew[s]” the data toward fewer environmental
impacts, and thus impedes a “full and fair discussion of the
potential effects of the project.” Native Ecosystems Council
v. U.S. Forest Serv., 418 F.3d 953, 965 (9th Cir. 2005)
(citation and internal quotation marks omitted).

Unlike in Akutan, BOEM’s estimate did not merely
inform an assessment of the likelihood of an oil spill.
869 F.2d at 1192. Among other things, its estimate informed
an assessment of seismic effects, habitat effects, oil
production, and the cumulative effects of the sale on global
warming. BOEM’s estimate also informed FWS’s
determination that Lease Sale 193 would not jeopardize listed
species. The record suggests that FWS was close to finding,
even under the one billion barrel assumption, that the lease
sale would jeopardize the spectacled and Steller’s eiders.
Had BOEM not selected the least amount of oil necessary for
production, FWS may well have concluded that the listed
species were in jeopardy. See 16 U.S.C. § 1536(a)(2).

Finally, the degree of error in the estimation of total oil
production is greater here than in our earlier cases. In Watt,
the agency was ready to publish its EIS when newly available
figures suggested that oil reserves were “roughly twice those
originally estimated.” 683 F.2d at 1267. We held in Watt
that the agency had acted reasonably when it decided not to
supplement its EIS with last-minute analysis of the risk of an
oil spill based on the new figures. Id. at 1267–68. In the case
before us, BOEM was fully aware from the very beginning
that if one billion barrels could be economically produced,
many more barrels could also be economically produced.
Indeed, at current oil prices, it would be economical to
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 33

recover twelve times the one billion barrel estimate used by
BOEM. This is a far more dramatic difference than in Watt.

We do not criticize BOEM’s decision to estimate the total
amount of economically recoverable oil from the lease sale.
Given the uncertainties involved in the Chukchi Sea, BOEM
had no choice but to make an estimate. But having decided
that oil production was reasonably foreseeable, NEPA
required BOEM to base its analysis on the full range of likely
production if oil production were to occur. It did not do so
here.

Conclusion

We conclude that the FEIS and SEIS properly took
account of incomplete or unavailable information. However,
we conclude that reliance in the FEIS on a one billion barrel
estimate of total economically recoverable oil was arbitrary
and capricious.

We REVERSE and REMAND to the district court for
further proceedings consistent with this opinion.

RAWLINSON, Circuit Judge, concurring in part and
dissenting in part:

I agree with most of the majority opinion, including that
the missing information from the final environmental impact
statement (FEIS) and the Supplemental Impact Statement
(SEIS) is not essential to informed decisionmaking at the
lease sale stage, and that further environmental analysis will
be more appropriate at a later stage. However, I do not agree
34 NATIVE VILLAGE OF POINT HOPE V. JEWELL

that the Bureau of Ocean Energy Management, Regulation
and Enforcement (BOEM) acted arbitrarily in selecting one
billion barrels of oil as the benchmark for analyzing the
environmental affects of the proposed leases.

I begin with a reminder that our review of the agency’s
analysis of technical data is extremely limited. See Lands
Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008) (en
banc), overruled on other grounds as recognized by Am.
Trucking Ass’ns v. City of Los Angeles, 559 F.3d 1046, 1052
(9th Cir. 2009) (“Review under the arbitrary and capricious
standard is narrow, and we do not substitute our judgment for
that of the agency. . . .”) (citation, alteration and internal
quotation marks omitted). We should also keep in mind that
the National Environmental Policy Act (NEPA) “does not
mandate particular results.” Dep’t of Transp. v. Pub. Citizen,
541 U.S. 752, 756 (2004). Rather, the statute “imposes only
procedural requirements on federal agencies . . .” Id. Under
NEPA, “[w]e review an [Environmental Impact Statement]
under a rule of reason to determine whether it contains a
reasonably thorough discussion” of the potential
environmental effects of a planned federal action. Edwardsen
v. Department of the Interior, 268 F.3d 781, 784 (9th Cir.
2001) (citation omitted).

The majority opinion takes issue with the agency’s
selection of one billion barrels of oil as the benchmark
amount for assessing potential environmental effects of the
oil leases. However, our review is at its most deferential
when we consider a predictive estimate such as BOEM’s
estimate of the amount of oil recovery that should be included
in the environmental effects analysis. See Lands Council,
537 F.3d at 993 (“[W]e are to conduct a particularly
deferential review of an agency’s predictive judgments about
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 35

areas that are within the agency’s field of discretion and
expertise . . .) (citations and internal quotation marks
omitted). Our task is only to ensure that the agency has not:

relied on factors which Congress has not
intended it to consider, entirely failed to
consider an aspect of the problem, offered an
explanation for its decision that runs counter
to the evidence before the agency, or an
explanation that is so implausible that it could
not be ascribed to a difference in view or the
product of agency expertise.

Id. (citations, alteration, and internal quotation marks
omitted).

The majority does not intimate that BOEM relied on
factors Congress did not intend it to consider, or that BOEM
entirely failed to consider an aspect of the problem. The
majority opinion also cannot be fairly read to describe
BOEM’s benchmark choice as so implausible that it could not
be ascribed to a difference in view or the product of agency
expertise. In fact, the majority opinion discusses the different
view and agency expertise brought to bear on this issue. See
Majority Opinion, pp. 21–26 (discussing the differing views
from within and without the agency). So it appears that the
basis for the majority’s ruling is that BOEM’s benchmark
estimate runs counter to the evidence before the agency. But
it doesn’t.

The potential size of commercially extractable oil
deposits in the Chukchi Sea is a quintessential example of a
predictive judgment uniquely within BOEM’s area of
expertise. Indeed, we have previously recognized that
36 NATIVE VILLAGE OF POINT HOPE V. JEWELL

“[p]rior to exploration, it is difficult to make so much as an
educated guess as to the volume of oil likely to be produced
. . . Without this information, an oil spill risk analysis can
never be more than speculative, regardless of what
methodology is used. . . .” Tribal Village of Akutan v. Hodel,
869 F.2d 1185, 1192 (9th Cir. 1989), as amended.

It is beyond dispute that the Chukchi Sea contains oil
deposits well in excess of one billion barrels. But that is not
the point. The point is whether the selection of one billion
barrels as the benchmark was the product of agency expertise.
See Lands Council, 537 F.3d at 993. After considering the
available evidence, BOEM concluded that substantial
obstacles to oil development in the region made it unlikely
that future production would “ever reach the full economic
potential” in the Chukchi Sea. Five explorations had already
tested some of the largest prospective sites without
discovering a “commercial-size” oil source. With these
circumstances in mind, BOEM ultimately selected one billion
barrels as the benchmark estimate because lower oil volumes
were not likely to be economically feasible. Rather than
relying on general resource assessments as was done
previously, BOEM opted for the more “realistic” benchmark
tied to the discovery/development of the initial commercially
viable offshore oil field. BOEM explained that the unique,
remote, and previously unexplored nature of the Chukchi
region required analysis of the “statistically most-likely
development activity associated with a reasonable range of
resources . . . given the uncertainties of geology, engineering,
and economics that exist now” and the “streamlined”
environmental impact statement (EIS) undertaken at the
leasing stage of the process. See Akutan, 869 F.2d at 1192
(“We are the least troubled by what may seem to be
incomplete or speculative data at the lease sale stage. . . .”).
 NATIVE VILLAGE OF POINT HOPE V. JEWELL 37

The majority is of the view that BOEM’s “analysis is
flawed.” Majority Opinion, p. 28. But we do not sit as a
panel of super scientists to dissect the agency’s analysis.
Rather, we only review the process for reasonable
thoroughness. See Edwardsen, 268 F.3d at 784 (establishing
the role of the reviewing court to determine whether the
agency’s environmental impact statement “contains a
reasonably thorough discussion”). Not only was BOEM’s
discussion of the selected benchmark “reasonably thorough,”
id., its selection of the benchmark was within the range of
alternatives contained in the record. As the majority opinion
acknowledges, a previous EIS had estimated a range of
economical oil production from 0.96 billion barrels to 2.42
billion barrels. See Majority Opinion, p. 24. One billion
barrels is certainly within that range. The same is true for the
“range between 1.0 and 6.1 billion barrels” referenced at page
27 of the Majority Opinion and for the 10 percent estimate of
the chance of commercial oil development. See Majority
Opinion, p. 28 (noting the EIS’s earlier reference to a 10–20
percent estimate and past estimates of 20 percent likely).

I readily acknowledge that there was disagreement in the
scientific community concerning the selected benchmark.
But disagreement does not render the chosen estimate
irrational. Rather, it typifies the “difference in view” that we
have established as a safe harbor against successful attack
under NEPA. See Lands Council, 537 F.3d at 993. There is
no such thing as a perfect estimate and BOEM was not
required to adopt a different benchmark in the face of its
critics. See Environmental Defense Center, Inc. v. EPA,
344 F.3d 832, 872 (9th Cir. 2003) (“We defer to an agency
decision not to invest the resources necessary to conduct the
perfect study . . .”). BOEM explained its reasons for
selecting its benchmark estimate, and we are uniquely
38 NATIVE VILLAGE OF POINT HOPE V. JEWELL

unqualified to second-guess that selection. As the D.C.
Circuit recognized in City of L.A. v. Dep’t of Transp.,
165 F.3d 972, 977 (D.C. Cir. 1999): “[That some or many
[experts] would disapprove of [BOEM’s] approach does not
answer the question presented to us. In reviewing [BOEM’s
EIS], we do not sit as a panel of referees on a professional
[scientific] journal, but as a panel of generalist judges obliged
to defer to a reasonable judgment by an agency acting
pursuant to congressionally delegated authority. . . .”)
(citations omitted).

Because of the deference due the agency, and because
BOEM’s chosen benchmark was reasonably selected and
adequately explained, our work here is done. We should
afford BOEM’s EIS the deference due and affirm the district
court’s order of dismissal.

I respectfully dissent.